Minimize Landsat-8

Landsat-8 / LDCM (Landsat Data Continuity Mission)

Spacecraft     Launch    Mission Status     Sensor Complement    Ground Segment    References

The Landsat spacecraft series of NASA represents the longest continuous Earth imaging program in history, starting with the launch of Landsat-1 in 1972 through Landsat-7 with the ETM+ imager (launch April 15, 1999). With the evolution of the program has come an increased emphasis on the scientific utility of the data accompanied by more stringent requirements for instrument and data characterization, calibration and validation. This trend continues with LDCM, the next mission in the Landsat sequence. The enhancements of the Landsat-7 system, e.g., more on-board calibration hardware and an image assessment system and personnel, have been retained and improved, where required, for LDCM. Aspects of the calibration requirements are spread throughout the mission, including the instrument and its characterization, the spacecraft, operations and the ground system. 1) 2)

The following are the major mission objectives: 3)

• Collect and archive moderate-resolution, reflective multispectral image data affording seasonal coverage of the global land mass for a period of no less than five years.

• Collect and archive moderate-resolution, thermal multispectral image data affording seasonal coverage of the global land mass for a period of no less than three years.

• Ensure that LDCM data are sufficiently consistent with data from the earlier Landsat missions, in terms of acquisition geometry, calibration, coverage characteristics, spectral and spatial characteristics, output product quality, and data availability to permit studies of land cover and land use change over multi-decadal periods.

• Distribute standard LDCM data products to users on a nondiscriminatory basis and at no cost to the users.

Background: In 2002, the Landsat program had its 30th anniversary of providing satellite remote sensing information to the world; indeed a record history of service with the longest continuous spaceborne optical medium-resolution imaging dataset available anywhere. The imagery has been and is being used for a multitude of land surface monitoring tasks covering a broad spectrum of resource management and global change issues and applications.

In 1992 the US Congress noted that Landsat commercialization had not worked and brought Landsat back into the government resulting in the launches of Landsat 6 (which failed on launch) and Landsat 7. However there was still much conflict within the government over how to continue the program.

In view of the outstanding value of the data to the user community as a whole, NASA and USGS (United States Geological Survey) were working together (planning, rule definition, forum of ideas and discussion among all parties involved, coordination) on the next generation of the Landsat series satellites, referred to as LDCM (Landsat Data Continuity Mission). The overall timeline foresaw a formulation phase until early 2003, followed by an implementation phase until 2006. The goal was to acquire the first LDCM imagery in 2007 - to ensure the continuity of the Landsat dataset [185 km swath width, 15 m resolution (Pan) and a new set of spectral bands]. 4) 5) 6) 7) 8) 9) 10) 11)

The LDCM project suffered some setbacks on its way to realization resulting in considerable delays:

• An initial major programmatic objective of LDCM was to explore the use of imagery purchases from a commercial satellite system in the next phase of the Landsat program. In March 2002, NASA awarded two study contracts to: a) Resource21 LLC. of Englewood, CO, and b) DigitalGlobe Inc. of Longmont, CO. The aim was to formulate a proper requirements set and an implementation scenario (options) for LDCM. NASA envisioned a PPP (Public Private Partnership) program in which the satellite system was going to be owned and operated commercially. A contract was to be awarded in the spring of 2003. - However, it turned out that DigitalGlobe lost interest and dropped out of the race. And the bid of Resource21 turned out to be too high for NASA to be considered.

• In 2004, NASA was directed by the OSTP (Office of Science and Technology Policy) to fly a Landsat instrument on the new NPOESS satellite series of NOAA.

• In Dec. 2005, a memorandum with the tittle “Landsat Data Continuity Strategy Adjustment” was released by the OSTP which directed NASA to acquire a free-flyer spacecraft for LDCM - thus, superseding the previous direction to fly a Landsat sensor on NPOESS. 12)

However, the matter was not resolved until 2007 when it was determined that NASA would procure the next mission, the LDCM, and that the USGS would operate it as well as determine all future Earth observation missions. This decision means that Earth observation has found a home in an operating agency whose mission is directly concerned with the mapping and analysis of the Earth’s surface allowing NASA to focus on advancing space technologies and the future of man in space.

Overall science objectives of the LDCM imager observations are:

• To permit change detection analysis and to ensure consistency of the LDCM data with the Landsat series data

• To provide global coverage of the Earth's land surfaces on a seasonal basis

• To acquire imagery at spatial, spectral and temporal resolutions sufficient to characterize and understand the causes and consequences of change

• To make the data available to the user community.

The procurement approach for the LDCM project represents a departure from a conventional NASA mission. NASA traditionally specifies the design of the spacecraft, instruments, and ground systems acquiring data for its Earth science missions. For LDCM, NASA and USGS (the science and technology agency of the Department of the Interior, DOI) have instead specified the content, quantity, and characteristics of data to be delivered.


Figure 1: History of the Landsat program (image credit: NASA) 13)

Legend to Figure 1: The small white arrow within the Landsat-7 arrow on this timeline indicates the collection of data without the Scan Line Corrector.

“The Landsat series of satellites is a cornerstone of our Earth observing capability. The world relies on Landsat data to detect and measure land cover/land use change, the health of ecosystems, and water availability,” NASA Administrator Charles Bolden told the Subcommittee on Space Committee on Science, Space and Technology U.S House of Representatives in April 2015.

“With a launch in 2023, Landsat-9 would propel the program past 50 years of collecting global land cover data,” said Jeffrey Masek, Landsat-9 Project Scientist at Goddard. “That’s the hallmark of Landsat: the longer the satellites view the Earth, the more phenomena you can observe and understand. We see changing areas of irrigated agriculture worldwide, systemic conversion of forest to pasture – activities where either human pressures or natural environmental pressures are causing the shifts in land use over decades.”

Landsat-8 successfully launched on Feb. 11, 2013 and the Landsat data archive continues to expand. — Landsat-9 was announced on April 16, 2015. The launch is planned for 2023. 14)

Dec. 31, 2015: NASA has awarded a sole source letter contract to BACT (Ball Aerospace & Technologies Corporation), Boulder, Colo., to build the OLI-2 (Operational Land Imager-2) instrument for the Landsat-9 project. 15)


In April 2008, NASA selected GDAIS (General Dynamics Advanced Information Systems), Inc., Gilbert, AZ, to build the LDCM spacecraft on a fixed price contract. An option provides for the inclusion of a second payload instrument. LDCM is a NASA/USGS partnership mission with the following responsibilities: 16) 17) 18) 19)

• NASA is providing the LDCM spacecraft, the instruments, the launch vehicle, and the mission operations element of the ground system. NASA will also manage the space segment early on-orbit evaluation phase -from launch to acceptance.

• USGS is providing the mission operations center and ground processing systems (including archive and data networks), as well as the flight operations team. USGS will also co-chair and fund the Landsat science team.

In April 2010, OSC (Orbital Sciences Corporation) of Dulles VA acquired GDAIS. Hence, OSC will continue to manufacture and integrate the LDCM program as outlined by GDAIS. Already in Dec. 2009, GDAIS successfully completed the CDR (Critical Design Review) of LDCM for NASA/GSFC. 20) 21)


Figure 2: Artist's rendition of the LDCM spacecraft in orbit (image credit: NASA, OSC)

The LDCM spacecraft uses a nadir-pointing three-axis stabilized platform (zero momentum biased), a modular architecture referred to as SA-200HP. The SA-200HP (High Performance) bus is of DS1 (Deep Space 1) and Coriolis mission heritage. The spacecraft consists of an aluminum frame and panel prime structure.

The spacecraft is 3-axis stabilized (zero momentum biased). The ADCS (Attitude Determination and Control Subsystem) employs six reaction wheels, three torque rods and thrusters as actuators. Attitude is sensed with three precision star trackers (2 of 3 star trackers are active), a redundant SIRU (Scalable Inertial Reference Unit), twelve coarse sun sensors, redundant GPS receivers (Viceroy), and two TAMs (Three Axis Magnetometers).

- Attitude control error (3σ): ≤ 30 µrad

- Attitude knowledge error (3σ): ≤ 46 µrad

- Attitude knowledge stability (3σ): ≤ 0.12 µrad in 2.5 seconds; ≤ 1.45 µrad in 30 seconds

- Slew time: 180º any axis: ≤ 14 minutes, including settling; 15º roll: ≤ 4.5 minutes, including settling.

Key aspects of the satellite performance related to imager calibration and validation are pointing, stability and maneuverability. Pointing and stability affect geometric performance; maneuverability allows data acquisitions for calibration using the sun, moon and stars. For LDCM, an off nadir acquisition capability is included (up to 1 path off nadir) for imaging high priority targets (event monitoring capability).
Also, the spacecraft pointing capability will allow the calibration of the OLI using the sun (roughly weekly), the moon (monthly), stars (during commissioning) and the Earth (at 90° from normal orientation, a.k.a., side slither) quarterly. The solar calibration will be used for OLI absolute and relative calibration, the moon for trending the stability of the OLI response, the stars will be used for Line of Sight determination and the side slither will be an alternate OLI and relative gain determination methodology. 22) 23)

C&DH (Command & Data Handling) subsystem: The C&DH subsystem uses a standard cPCI backplane RAD750 CPU. The MIL-STD-1553B data bus is used for onboard ADCS, C&DH functions and instrument communications. The SSR (Solid State Recorder) provides a storage capacity of 4 Tbit @ BOL and 3.1 Tbit @ EOL.

The C&DH subsystem provides the mission data interfaces between instruments, the SSR, and the X-band transmitter. The C&DH subsystem consists of an IEM (Integrated Electronics Module), a PIE (Payload Interface Electronics), the SSR, and two OCXO (Oven Controlled Crystal Oscillators).


Figure 3: Photo of the EM SSR (Solid State Recorder), image credit: NASA


Figure 4: Block diagram of the C&DH subsystem (image credit: NASA, USGS, Ref. 125)

- The IEM subsystem provides the command and data handling function for the observatory, including mission data management between the PIE and SSR using FSW on the Rad750 processor. The IEM is block redundant with cross strapped interfaces for command and telemetry management, attitude control, SOH (State of Health) data and ancillary data processing, and for controlling image collection and file downlinks to the ground.

- The SSR subsystem provides for mission data and spacecraft SOH storage during all mission operations. The OCXO provides a stable, accurate time base for ADCS fine pointing.

- The C&DH accepts encrypted ground commands for immediate execution or for storage in the FSW file system using the relative time and absolute time command sequences (RTS, ATS respectfully). The commanding interface is connected to the uplink of each S-band transceiver, providing for cross-strapped redundancy to the C&DH. All commands are verified onboard prior to execution. Real-time commands are executed upon reception, while stored commands are placed in the FSW file system and executed under control of the FSW. Command counters and execution history are maintained by the C&DH FSW and reported in SOH telemetry.

- The IEM provides the command and housekeeping telemetry interfaces between the payload instruments and the ADCS components using a MIL-STD-1553B serial data bus and discrete control and monitoring interfaces. The C&DH provides the command and housekeeping interfaces between the CCU (Charge Control Unit), LCU (Load Control Unit) , and the PIE boxes.

- The PIE is the one of the key electrical system interfaces and mission data processing systems between the instruments, the spacecraft C&DH, SSR, and RF communications to the ground. The PIE contains the PIB (Payload Interface Boards ) for OLI (PIB-O) and TIRS (PIB-T).

Each PIB contains an assortment of specialized FPGAs (Field Programmable Gate Arrays) and ASICs, and each accepts instrument image data across the HSSDB for C&DH processing. A RS-485 communication bus collects SOH and ACS ancillary data for interleaving with the image data.


Figure 5: Block diagram of PIB (image credit: USGS, NASA)

- Data compression: Only the OLI data, sent through the PIB-O interface, implements lossless compression, by utilizing a pre-processor and entropy encoder in the USES ASIC. The compression can be enabled or bypassed on an image-by-image basis. When compression is enabled the first image line of each 1 GB file is uncompressed to provide a reference line to start that file. A reference line is generated every 1,024 lines (about every 4 seconds) to support real-time ground contacts to begin receiving data in the middle of a file and decompressing the image with the reception of a reference line.

- XIB (X-band Interface Board): The XIB is the C&DH interface between the PIE, SSR, and X-band transmitter, with the functional data path shown in Figure 6.

The XIB receives real-time data from the PIE PIB-O and PIB-T and receives stored data from the SSR via the 2 playback ports. The XIB sends mission data to the X-band transmitter via a parallel LVDS interface. The XIB receives a clock from the X-band transmitter to determine the data transfer rates between the XIB and the transmitter to maintain a 384 Mbit/s downlink. The XIB receives OLI realtime data from the PIB-O board, and TIRS real-time data from the PIB-T board across the backplane. The SSR data from the PIB-O and PIB-T interfaces are multiplexed and sent to the X-Band transmitter through parallel LVDS byte-wide interfaces.


Figure 6: X-band mission data flow (image credit: USGS, NASA)

- SSR (Solid Ste Recorder): The SSR is designed with radiation hard ASIC controllers, and up-screened commercial grade 4GB SDRAM (Synchronous Dynamic Random Access Memory) memory devices. Protection against on-orbit radiation induced errors is provided by a Reed-Solomon EDAC (Error Detection and Correction) algorithm. The SSR provides the primary means for storing all image, ancillary, and state of health data using a file management architecture. Manufactured in a single mechanical chassis, containing a total of 14 memory boards, the system provides fully redundant sides and interfaces to the spacecraft C&DH.

The spacecraft FSW (Flight Software) plays an integral role in the management of the file directory system for recording and file playback. FSW creates file attributes for identifier, size, priority, protection based upon instructions from the ground defining the length of imaging in the interval request, and its associated priority. FSW also maintains the file directory, and creates the ordered lists for autonomous playback based upon image priority. FSW automatically updates and maintains the spacecraft directory while recording or performing playback, and it periodically updates the SSR FSW directory when no recording is occurring to synchronize the two directories (Ref. 125).

TCS (Thermal Control Subsystem): The TCS uses standard Kapton etched-foil strip heaters. In general, a passive, cold-biased system is used for the spacecraft. Multi-layer insulation on spacecraft and payload as required. A deep space view is provided for the instrument radiators.

EPS (Electric Power Subsystem): The EPS consists of a single deployable solar array with single-axis articulation capability and with a stepping gimbal. Triple-junction solar cells are being used providing a power of 4300 W @ EOL. The NiH2 battery has a capacity of 125 Ah. Use of unregulated 22-36 V power bus.

The onboard propulsion subsystem provides a total velocity change of ΔV = 334 m/s using eight 22 N thrusters for insertion error correction, altitude adjustments, attitude recovery, EOL disposal, and other operational maintenance as necessary.

The spacecraft has a launch mass of 2780 kg (1512 kg dry mass). The mission design life is 5 years; the onboard consumable supply (386 kg of hydrazine) will last for 10 years of operations.

Spacecraft platform

SA-200HP (High Performance) bus

Spacecraft mass

Launch mass of 2780 kg; dry mass of 1512 kg

Spacecraft design life

5 years; the onboard consumable supply (386 kg of hydrazine) will last for 10 years of operations

EPS (Electric Power Subsystem)

- Power: 4.3 kW @ EOL (End of Life)
- Single deployable solar array with single-axis articulation capability
- Triple-junction solar cells
- NiH2 battery with 125 Ah capacity
- Unregulated 22 V - 36 V power bus
- Two power distribution boxes

ADCS (Attitude Determination &
Control Subsystem)

- Actuation: 6 reaction wheels and 3 torque rods
- Attitude is sensed with 3 precision star trackers, a redundant SIRU (Scalable Inertial Reference Unit),
12 coarse sun sensors, redundant GPS receivers (Viceroy), and 2 TAMs (Three Axis Magnetometers)
- Attitude control error (3σ): ≤ 30 µrad
- Attitude knowledge error (3σ): ≤ 29 µrad
- Attitude knowledge stability (3σ): ≤ 0.12 µrad in 2.5 seconds
- Attitude jitter: ≤ 0.28 µrad, 0.1-1.0 Hz
- Slew time, 180º pitch: ≤ 14 minutes, inclusive settling
- Slew time, 15º roll: ≤ 4.5 minutes, inclusive settling

C&DH (Command & Data Handling)

- Standard cPCI backplane RAD750 CPU
- MIL-STD-1553B data bus
- Solid state recorder provides a storage capacity of 4 TB @ BOL and 3.1 TB @ EOL

Propulsion subsystem

- Total velocity change of ΔV = 334 m/s using eight 22 N thrusters
- Hydrazine blow-down propulsion module

Table 1: Overview of spacecraft parameters


Figure 7: Two views of the LDCM spacecraft (without solar arrays) and major components (image credit: NASA, USGS)

RF communications: Earth coverage antennas are being used for all data links. The X-band downlink uses lossless compression and spectral filtering. The payload data rate is 440 Mbit/s. The X-band RF system consists of the X-band transmitter, TWTA (Travelling Wave Tube Amplifier), DSN (Deep Space Network) filter, and an ECA (Earth Coverage Antenna). The serial data output is set at 440.825 Mbit/s and is up-converted to 8200.5 MHz. The TWTA amplifies the signal such that the output of the DSN filter is 62 W. The DSN filter maintains the signal’s spectral compliance. An ECA provides nadir full simultaneous coverage, utilizing 120º half-power beamwidth, for all in view ground sites below the spacecraft's current position with no gimbal or actuation system. The system is designed to handle up to 35 separate ground contacts per day as forecasted by the DRC-16 (Design Reference Case-16).

The X-band transmitter is a single customized unit, including the LDPC FEC algorithms, the modulator, and up converter circuits. The transmitter uses a local TXCO (Thermally Controlled Crystal Oscillator) as a clock source for tight spectral quality and minimum data jitter. This clock is provided to the PIE XIB to clock mission data up to a 384Mbit/s data rate to the transmitter. The X-band transmitter includes an on-board synthesized clock operating at 441.625 Mbit/s coded data rate using the local 48 MHz clock as a reference. Using the on-board FIFO buffer, this architecture provides a continuous data flow through the transmitter (Ref. 125).

The S-band is used for all TT&C functions. The S-band uplink is encrypted providing data rates of 1, 32, and 64 kbit/s. The S-band downlink offers data rates of 2, 16, 32, RTSOH; 1 Mbit/s SSOH/RTSOH GN; 1 kbit/s RTSOH SN. Redundant pairs of S-band omni’s provide transmit/receive coverage in any orientation. The S-band is provided through a typical S-band transceiver, with TDRSS (Tracking and Data Relay Satellite System) capability for use during launch and early orbit and in case of spacecraft emergencies.

Onboard data transmission from an earth-coverage antenna:

• Real-time data received from PIE (Payload Interface Electronics) equipment

• Play-back data from SSR (Solid State Recorder)

• To three LGN (LDCM Ground Network) stations

- NOAA Interagency Agreement (IA) to use Gilmore Creek Station (GLC) near Fairbanks, AK

- Landsat Ground Station (LGS) at USGS/EROS near Sioux Falls, SD

- NASA contract with KSAT for Svalbard; options for operational use by USGS (provides ≥ 200 minutes of contact time)

• To International Cooperator ground stations (partnerships of existing stations currently supporting Landsat).


Figure 8: Photo of the EM X-band transponder (left) and AMT S-band transponder (right), image credit: NASA


Figure 9: Alternate view of the deployed LDCM spacecraft showing the calibration ports of the instruments TIRS and OLI (image credit: NASA/GSFC)


Figure 10: The LDCM spacecraft with both instruments onboard, OLI and TIRS (image credit: USGS) 24)

Launch: The LDCM mission was launched on February 11, 2013 from VAFB, CA. The launch provider was ULA (United Launch Alliance), a joint venture of Lockheed Martin and Boeing; use of the Atlas-V-401 the launch vehicle with a Centaur upper stage. 25) 26)

Note: Initially, the LDCM launch was set for July 2011. However, since this launch date was considered as too optimistic, NASA changed the launch date to the end of 2012. This new launch delay buys some time for an extra sensor with TIR (Thermal Infrared) imaging capabilities.

Orbit: Sun-synchronous near-circular orbit, altitude = 705 km, inclination = 98.2º, period = 99 minutes, repeat coverage = 16 days (233 orbits), the nominal LTDN (Local Time on Descending Node) equator crossing time is at 10:00 hours. The ground tracks will be maintained along heritage WRS-2 paths. At the end of the commissioning period, LDCM is required to be phased about half a period ahead of Landsat 7. 27)

Figure 11: Anatomy of Landsat 8. Have you ever wondered what all the parts of a satellite do? This video identifies a few of the main components onboard Landsat 8 and tells you about their role in flying the satellite and capturing images of the Earth's surface below (video credit: USGS, NASA) 28)

Figure 12: The Landsat Data Continuity Mission (LDCM), a collaboration between NASA and the USGS (U.S. Geological Survey), will provide moderate-resolution measurements of Earth's terrestrial and polar regions in the visible, near-infrared, short wave infrared, and thermal infrared. There are two instruments on the spacecraft, the Thermal InfraRed Sensor (TIRS) and the Operational Land Imager (OLI). LDCM will provide continuity with the nearly 40-year long Landsat land imaging data set, enabling people to study many aspects of our planet and to evaluate the dynamic changes caused by both natural processes and human practices (video credit: NASA, USGS) 29)

Note: As of February 2020, the previously single large Landsat-8 file has been split into four files, to make the file handling manageable for all parties concerned, in particular for the user community.

This article covers the Landsat-8 mission and its imagery in the period 2021, in addition to some of the mission milestones.

Landsat-8 imagery in the period 2020

LandSat-8 imagery in the period 2019

Landsat-8 imagery in the period 2018

Landsat-8 imagery in the period 2017 to June 2013

Mission status and some imagery of 2021

• October 5, 2021: If it seems like enormous wildfires have been constantly raging in California in recent summers, it's because they have. Eight of the state’s ten largest fires on record—and twelve of the top twenty—have happened within the past five years, according to the California Department of Forestry and Fire Protection (Cal Fire). Together, those twelve fires have burned about 4 percent of California’s total area—a Connecticut-sized amount of land. 30)


Figure 13: Two recent incidents—the Dixie fire (2021) and the August fire complex (2020)—stand out for their size. Each of these burned nearly 1 million acres—an area larger than Rhode Island—as they raged for months in forests in Northern California. Several other large fires, as well as many smaller ones in densely populated areas, have proven catastrophic in terms of structures destroyed and lives lost. Thirteen of California’s twenty most destructive wildfires have occurred in the past five years; they collectively destroyed 40,000 homes, businesses, and pieces of infrastructure. The effects of all these fires are dramatic from the ground and from space. This false-color image, captured by the Operational Land Imager (OLI) on Landsat-8, shows the burn scar left by the Dixie fire. The blaze destroyed 1,329 structures and cost hundreds of millions of dollars to fight (image credit: NASA Earth Observatory, images by Joshua Stevens and Lauren Dauphin, using Landsat data from the U.S. Geological Survey, fire perimeters from the National Interagency Fire Center, and drought conditions from the U.S. Drought Monitor/University of Nebraska-Lincoln. Photograph courtesy of InciWeb. Story by Adam Voiland)


Figure 14: California's Wildfires are Growing, Simply put, the fires of the recent years dwarf those of previous decades (image credit: NASA Earth Observatory)

- The total area burned by fires each year and the average size of fires is up as well, according to Keith Weber, a remote sensing ecologist at Idaho State University and the principal investigator of the Historic Fires Database, a project of NASA’s Earth Science Applied Sciences program. The database shows that about 3 percent of the state’s land surfaces burned between 1970-1980; from 2010-2020 it was 11 percent. The shift toward larger fires is clear in the decadal maps (above) of fire perimeter data from the National Interagency Fire Center.

- “The numbers are really worrisome, but they are not at all surprising to fire scientists,” said Jon Keeley, a U.S. Geological Survey scientist based in Sequoia National Park. He is among several experts who say a confluence of factors has driven the surge of large, destructive fires in California: unusual drought and heat exacerbated by climate change, overgrown forests caused by decades of fire suppression, and rapid population growth along the edges of forests.


Figure 15: This photograph shows charred forests in Plumas National Forest in the wake of the Dixie fire (image credit: NASA Earth Observatory)

- “The current drought is unprecedented,” said Keeley. “Each of the past three decades has had substantially worse drought than any decade over the last 150 years.” In the short-term, drought exacerbates fires by sapping trees and plants of moisture and making them easier to burn. Over the long-term, it adds vast amounts of dead wood to the landscape and makes intense fires more likely.

- The 2020-2021 drought has been especially extreme. “The last two years in California have brought compound drought conditions—effectively, very dry winters followed by relentless summer heat and atmospheric aridity,” explained John Abatzoglou, a climate scientist at the University of California, Merced. “This has left soil and vegetation parched across much of California, so the landscape is capable of carrying fire that resists suppression.”

- Data from the Western Regional Climate Center indicates that the northern two-thirds of the state received only half of normal rainfall over the past few years. The U.S. Drought Monitor has categorized about 85 to 90 percent of California as experiencing “exceptional” or “extreme” drought for all of summer 2021. And the period between September 2019 and August 2021 ranked as the second-driest on record for the state, according to data from the National Centers for Environmental Information.


Figure 16: More and More of California is in Drought. Percent of California in Drought by U.S. Drought Monitor Category (image credit: NASA Earth Observatory)

- Daniel Swain, a climatologist at the University of California, Los Angeles, added that one of the most direct ways that climate change is influencing California fires is by dialing up the temperature. “Heat essentially turns the atmosphere into a giant sponge that draws moisture from plants and makes it possible for fires to burn hotter and longer,” he said. Meteorological data shows that the two-year period from September 2019 through August 2021 ranks as the third-warmest on record in California, with temperatures that were roughly 2.9° (1.6°C) degrees warmer than average. Air can absorb about 7 percent more water for every degree Celsius it warms.

- Abatzoglou noted that some of the harrowing scenes across Northern California in 2020 were due to an extreme and unusual dry lightning siege in mid-August that ignited thousands of fires in one night. “But in 2021 I am less convinced of bad luck,” he said. “Climate change is aiding in the warming and the more rapid drying of fuels that predispose the land to large fires.”

• September 24, 2021: As North America approaches the end of the 2021 water year, the two largest reservoirs in the United States stand at their lowest levels since they were first filled. After two years of intense drought and two decades of long-term drought in the American Southwest, government water managers have been forced to reconsider how supplies will be portioned out in the 2022 water year. 31)

- Straddling the border of southeastern Utah and northeastern Arizona, Lake Powell is the second largest reservoir by capacity in the United States. In July 2021, water levels on the lake fell to the lowest point since 1969 and have continued dropping. As of September 20, 2021, the water elevation at Glen Canyon Dam was 3,546.93 feet, more than 153 feet below “full pool” (elevation 3,700 feet). The lake held just 30 percent of its capacity. To compensate, federal managers started releasing water from upstream reservoirs to help keep Lake Powell from dropping below a threshold that threatens hydropower equipment at the dam.


Figure 17: The natural-color images (Figures 17 Sept. 1, 2017, and 18 Aug. 27, 2021) show Lake Powell in the late summer of 2017 and 2021, as observed by the Operational Land Imager on Landsat-8. The September 2017 image was chosen because it represents the highest water level (3,630.76 feet) from the past decade (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and lake elevation data from the Bureau of Reclamation. Story by Michael Carlowicz)


Figure 18: Water levels at Glen Canyon Dam have fallen to their lowest level since 1969 and are still dropping. Landsat-8 image of 27 August 2021 (image credit: NASA Earth Observatory)


Figure 19: This line plot shows water levels since 1999, when Lake Powell approached 94 percent capacity (image credit: NASA Earth Observatory)

- Downstream in the Colorado River water management system, Lake Mead is filled to just 35 percent of capacity. More than 94 percent of the land area across nine western states is now affected by some level of drought, according to the September 23 report from the U.S. Drought Monitor.

- In an announcement on September 22, the U.S. Bureau of Reclamation (USBR) explained that updated hydrological models for the next five years “show continued elevated risk of Lake Powell and Lake Mead reaching critically-low elevations as a result of the historic drought and low-runoff conditions in the Colorado River Basin. At Lake Powell, the projections indicate the potential of falling below minimum power pool as early as July 2022 should extremely dry hydrology continue into next year.” Minimum power pool refers to an elevation—3,490 feet—that water levels must remain above to keep the dam’s hydropower turbines working properly.

- With the entire Lower Colorado River water storage system at 39 percent of capacity, the Bureau of Reclamation recently announced that water allocations in the U.S. Southwest would be cut over the next year. ”Given ongoing historic drought and low runoff conditions in the Colorado River Basin, downstream releases from Glen Canyon Dam and Hoover Dam will be reduced in 2022 due to declining reservoir levels,” the USBR statement said. “In the Lower Basin the reductions represent the first “shortage” declaration—demonstrating the severity of the drought and low reservoir conditions.”


Figure 20: These natural-color images were acquired in March 1999, April 2005, May 2011, and April 2021 by the Landsat 5, 7, and 8 satellites, respectively. Springtime typically marks the lowest water levels before mountaintop snow starts to melt and run down into the watershed. The images capture years with the two highest and lowest levels over the past 22 years. -For a year-by-year view, see the Earth Observatory feature World of Change: Water Level in Lake Powell (image credit: NASA Earth Observatory)

- The Colorado River basin is managed to provide water to millions of people—most notably the cities of San Diego, Las Vegas, Phoenix, and Los Angeles—and 4 to 5 million acres of farmland in the U.S. and Mexico. Water is allotted through laws like the 1922 Colorado River Compact and by a recent drought contingency plan announced in 2019.

- In a report and op-ed released on September 22, members of a NOAA Drought Task Force offered some context for the low water levels across the region. “Successive dry winter seasons in 2019-2020 and 2020-2021, together with a failed 2020 summer southwestern monsoon, led precipitation totals since January 2020 to be the lowest on record since at least 1895 over the entirety of the Southwest. At the same time, temperatures across the six states considered in the report (Arizona, California, Colorado, Nevada, New Mexico and Utah) were at their third highest on record. Together, the exceptionally low precipitation and warm temperatures reduced snowpack and increased evaporation of soil moisture, leading to a persistent and widespread drought over most of the American West.“

• September 20, 2021: Numerous craters on Earth are exceptionally compelling when viewed from space, displaying clearly visible rims and well-defined bowls. Not Sudbury Basin. It can take a moment looking at images to discern the shape of this impact structure amid the modern landscape. But few craters are as large or as old. 32)

- The object responsible for creating Sudbury Basin crashed into Earth about 1.8 billion years ago. That makes this crater in Canada fifty times older than Popigai—one of the world’s most well-preserved craters—which was created a mere 36 million years ago. Much of Sudbury’s original crater, thought to have measured at least 200 km (120 miles) across, has been deformed and eroded. Despite this, the crater has had a lasting impact on the region.


Figure 21: The OLI instrument on Landsat-8 acquired this image of Sudbury Basin in southeastern Ontario on September 11, 2020. Notice the many mines located around the basin, particularly along the rim. This is due to the abundance of ore deposits rich in nickel and copper, which were discovered here long before people were aware of the basin’s cosmic origin (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and topographic data from the Shuttle Radar Topography Mission (SRTM). Story by Kathryn Hansen)

- This region of Canada owes its unique geology to that powerful collision—initially thought to be an asteroid and later interpreted as a comet. The collision punctured Earth’s crust, allowing material from the mantle to well up from below and fill the basin with melted rock. Then after a shockwave shattered the surrounding rocks, minerals from the melted rock below infiltrated the cracks.


Figure 22: The relief of the basin is apparent in this map. Data for the map comes from a digital elevation model acquired by the Shuttle Radar Topography Mission (SRTM). Few craters are as large, or as old, as this impact structure in southeastern Ontario, Canada. (image credit: NASA Earth Observatory)

- People have been making use of the minerals in Sudbury Basin for thousands of years. Large-scale mining operations started with the Murray Mine (now defunct) in the late 1800s. The mining took a toll on the landscape, polluting the region with sulfur dioxide and metals released during smelting processes. In recent decades, efforts have been made to capture emissions and restore the health of the basin’s land and water.

• September 17, 2021: In the midst of another brutal fire season that has threatened many lives, homes, and businesses, several of California’s natural treasures have also been threatened. Still burning after nearly nine weeks, the Caldor fire has encroached on Lake Tahoe. Now some of the world’s oldest and largest trees are being threatened by fires at the southern end of the Sierra Nevada range. 33)


Figure 23: On September 15, 2021, the OLI instrument on Landsat-8 acquired imagery of the KNP fire complex (Kings-Canyon National Park), the Windy fire, and the thick smoke plumes both have released (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Michael Carlowicz)

- According to InciWeb and other sources, the KNP complex was ignited by a significant lightning storm on September 9–10. The Paradise fire and the Colony fire started separately near Sequoia National Park and have been marching across the drought-ravaged landscape toward a merger. By the morning of September 16, the KNP complex had burned 8,940 acres (36 km2). (A complex includes two or more separate fires that burn in very close proximity, have the potential to merge, and are managed by a unified firefighting group.)

- The KNP complex has led to the closure of Sequoia National Park and the evacuation of parts of the nearby community of Three Rivers. Fire officials told The Los Angeles Times on September 15 that the blazes were about a mile from the “Giant Forest,” the largest concentration of giant sequoias in the park and home to the 275-foot (83 m) General Sherman tree. Nearby Kings Canyon National Park remains open, but air quality is poor.

- The KNP fires are raging in very steep, dangerous terrain, so most of the firefighting has been done by aircraft so far. The National Park Service wrote in an update: “In the case of the Paradise Fire, extremely steep topography and a total lack of access has prevented any ground crew operations, and in the case of the Colony Fire, only a limited amount of ground crew access has been possible. Both fires are utilizing extensive aerial resources performing water and retardant drops.”

- Due south of the KNP complex, the Windy fire is burning in Sierra National Forest. It started in the Tule River Reservation during the September 9–10 lightning storm. About 2,800 acres have burned so far in an area not far from Giant Sequoia National Monument.

- According to CalFire, 1.97 million acres (nearly 3,100 square miles) have burned in California so far this year, and the fire season still has several months to go. The total is about half of the 2020 fire season—the worst on record—and roughly equal to the total burned in all of 2018. Near the end of the last severe drought in the state (2012-16), fire totals were 30 to 40 percent of the 2021 count.


Figure 24: In the midst of another brutal fire season, several of California’s natural treasures have also been threatened. This image includes infrared data with the thermal signature of some fire fronts beneath the plumes. NASA’s Terra satellite acquired broad-area images of the same region from September 10-16 (image credit: NASA Earth Observatory)

• September 12, 2021: Hurricane Ida left an extensive trail of damaged homes, infrastructure, and lives from Louisiana to New England. It also has left a stain on the sea. Two weeks after the storm, several federal and state agencies and some private companies are working to find and contain oil leaks in the Gulf of Mexico. 34)

- The U.S. Coast Guard has assessed more than 1,500 reports of pollution in the Gulf and in Louisiana, and it “is prioritizing nearly 350 reported incidents for further investigation by state, local, and federal authorities in the aftermath of Hurricane Ida.” The Coast Guard is working with the Environmental Protection Agency, the state of Louisiana, the National Ocean Service, and other agencies to chronicle and monitor the state of coastal waters and infrastructure.


Figure 25: Federal and state agencies and private companies are working to find and contain oil leaks in the Gulf of Mexico. On September 3, 2021, the Operational Land Imager (OLI) on Landsat-8 acquired this natural-color image of apparent oil slicks off the southeastern Louisiana coast near Port Fourchon, a major hub of the oil and gas industry (image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Michael Carlowicz)

- Hurricane Ida caused the disruption of 90 to 95 percent of the region’s crude oil and gas production, while also damaging current and abandoned pipelines and structures. According to many news reports, the surface oil slicks near Port Fourchon (shown above) are likely related to as many as three damaged or ruptured submarine pipelines. It is unclear how much oil has spilled into the Gulf of Mexico.

- NOAA (National Oceanic and Atmospheric Administration) has conducted aerial surveys of some offshore waters and has released the photos online. The NASA-sponsored Delta-X research team has also been working in the area and was called upon to make some observations of the slicks and other coastal changes with synthetic aperture radar.

- Beyond active oil and gas extraction platforms, the seafloor of the Gulf of Mexico is covered in a maze of pipelines, capped wellheads, and other infrastructure that can be vulnerable to storm events. In a report issued earlier this year, the U.S. Government Accountability Office stated: “Since the 1960s, the Bureau of Safety and Environmental Enforcement has allowed the offshore oil and gas industry to leave 97 percent of pipelines (18,000 miles) on the seafloor when no longer in use. Pipelines can contain oil or gas if not properly cleaned in decommissioning.”

• August 30, 2021: Lake Mead is the largest reservoir in the United States and part of a system that supplies water to at least 40 million people across seven states and northern Mexico. It stands today at its lowest level since Franklin Delano Roosevelt was president. This means less water will be portioned out to some states in the 2022 water year. 35)


Figure 26: As of August 22, 2021, Lake Mead was filled to just 35% of its capacity. The low water level comes at a time when 95 percent of the land in nine Western states is affected by some level of drought (64% is extreme or worse). It continues a 22-year megadrought that may be the region’s worst dry spell in twelve centuries. This natural color image was acquired by Landsat-8. The tan fringes along the shoreline are areas of the lakebed that would be underwater when the reservoir is filled closer to capacity. The phenomenon is often referred to as a “bathtub ring.”(image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and lake elevation data from the Bureau of Reclamation. Story by Michael Carlowicz and Kathryn Hansen)

- The lake elevation data below come from the U.S. Bureau of Reclamation, which manages Lake Mead, Lake Powell, and other portions of the Colorado River watershed. At the end of July 2021, the water elevation at the Hoover Dam was 1067.65 feet (325 meters) above sea level, the lowest since April 1937, when the lake was still being filled. The elevation at the end of July 2000—around the time of the Landsat 7 images above and below—was 1199.97 feet (341 meters).


Figure 27: Lake Meade elevation levels at Hoover Dam (image credit: NASA Earth Observatory)

- At maximum capacity, Lake Mead reaches an elevation 1,220 feet (372 meters) near the dam and would hold 9.3 trillion gallons (36 trillion liters, corresponding to 36,000 km3) of water. The lake last approached full capacity in the summers of 1983 and 1999. It has been dropping ever since.

- In most years, about 10% of the water in the lake comes from local precipitation and groundwater, with the rest coming from snowmelt in the Rocky Mountains that melts and flows down to rivers, traveling through Lake Powell, Glen Canyon, and the Grand Canyon on the way. The Colorado River basin is managed to provide water to millions of people—most notably the cities of San Diego, Las Vegas, Phoenix, and Los Angeles—and 4-5 million acres of farmland in the Southwest. The river is allotted to states and to Mexico through laws like the 1922 Colorado River Compact and by a recent drought contingency plan announced in 2019.

- With the Lake Mead reservoir at 35 percent of capacity, Lake Powell at 31 percent, and the entire Lower Colorado system at 40 percent, the Bureau of Reclamation announced on August 16 that water allocations would be cut over the next year. “The Upper [Colorado] Basin experienced an exceptionally dry spring in 2021, with April to July runoff into Lake Powell totaling just 26 percent of average despite near-average snowfall last winter,” the USBR statement said. ”Given ongoing historic drought and low runoff conditions in the Colorado River Basin, downstream releases from Glen Canyon Dam and Hoover Dam will be reduced in 2022 due to declining reservoir levels. In the Lower Basin the reductions represent the first “shortage” declaration—demonstrating the severity of the drought and low reservoir conditions.”


Figure 28: The Overton Arm of Lake Meade on August 7, 2000 (left) and on August 9, 2021 (image credit: NASA Earth Observatory)

- For the 2022 water year, which begins October 1, Mexico will receive 80,000 fewer acre-feet, approximately 5% of the country’s annual allotment and Nevada’s take will be cut by: 21,000 acre-feet (about 7% of the state’s annual apportionment). The biggest cuts will come to Arizona, which will receive 512,000 fewer acre-feet, approximately 18 % of the state’s annual apportionment and 8 % of the state’s total water use (for agriculture and human consumption). An acre-foot is enough water to supply one to two households a year.

• August 26, 2021: Since 2017, August 26 has been known as Katherine Johnson Day in West Virginia. The celebration commemorates the birthday of the ground-breaking NASA mathematician who was born on August 26, 1918, in White Sulphur Springs. 36)

- Katherine Johnson contributed her mathematical expertise to the first human space travel missions in the United States. In 1953, in a time of racial segregation, she started a job as a human “computer” with the National Advisory Committee for Aeronautics (NACA), the predecessor to NASA. She worked in the West Area Computing section at Langley Research Center on a team of Black women headed by fellow West Virginian Dorothy Vaughan.

- In 1961, Johnson did trajectory analysis for Alan Shepard’s Freedom 7 mission, America’s first human spaceflight. Her work was also instrumental in John Glenn’s successful orbit around Earth in 1962.

- Glenn became a household name in the United States, but it wasn’t until recently that Katherine Johnson’s name became well-known. Her story came to light in 2016 through the book and film Hidden Figures by Margot Lee Shetterly. In a pivotal scene in the movie, John Glenn hesitated over trusting his fate in space to a network of new IBM computers. He asked Johnson to check the equations for his orbit against the computer output. “If she says they’re good, I am ready to go,” said Glenn.


Figure 29: Johnson’s birthplace, White Sulphur Springs, is shown in this image, acquired by the Landsat-8 satellite in 2019. The small city sits in the Allegheny Mountains, one of the smaller ranges running through the Appalachians. The city was settled in the 18th century around a natural freshwater spring, which is now on the grounds of the Greenbrier Hotel. Katherine’s father, Joshua Johnson, worked at that resort as a bellman, but he was determined to get his talented daughter an education that allowed her to excel (image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Photograph by NASA. Story by Emily Cassidy, NASA Earthdata)

- Because of the segregated school system at the time, Black Americans could not attend high school in White Sulphur Springs. Her father moved the family to Institute, West Virginia, so that Katherine and her siblings could attend a high school that was on the West Virginia State University campus. Johnson graduated high school at 14, and then graduated summa cum laude from West Virginia State when she was 18, earning a double major in mathematics and French.


Figure 30: Katherine Johnson at work in her office at NASA in 1966 (photo credit: NASA)

- Johnson’s fingerprints are on some of NASA’s greatest achievements. She precisely calculated trajectories for the 1969 Apollo 11 flight to the Moon, and she worked on the Space Shuttle and the Earth Resources Technology Satellite (later renamed Landsat 1). Across three decades at Langley, she authored or co-authored more than two dozen research reports before retiring in 1986.

- In 2015, President Barack Obama awarded Johnson the Presidential Medal of Freedom, citing her as a pioneering example of African-American women in science, technology, engineering, and mathematics. “Katherine G. Johnson refused to be limited by society’s expectations of her gender and race, while expanding the boundaries of humanity’s reach,” said Charles Bolden, NASA's first Black administrator and a former astronaut.

- In 2019, NASA renamed its Independent Verification and Validation Facility in Fairmont, West Virginia, for Katherine Johnson. When she died on February 24, 2020, then-NASA Administrator James Bridenstine said: “She was an American hero and her pioneering legacy will never be forgotten.”

• August 17, 2021: Eleven years after an earthquake devastated the Haitian capital of Port-Au-Prince, another major earthquake has shaken the Caribbean nation. The epicenter of the magnitude 7.2 earthquake was centered about 100 km (60 miles) west of the 2010 quake, in a mountainous area between Petit-Trou-de-Nippes and Aquin. Like the previous event, this earthquake occurred along the Enriquillo-Plantain Garden fault, an area where two tectonic plates grind against each other. 37)


Figure 31: A break in the clouds allowed the Operational Land Imager (OLI) on Landsat-8 to acquire this natural-color view of landslides in and around Pic Macaya National Park in southwestern Haiti on August 14, 2021, the same day the earthquake hit (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- The earthquake exposed more than one million people to very strong to severe shaking, according to the U.S. Geological Survey. In preliminary estimates, news media and Haiti’s civil protection agency are reporting large numbers of deaths and extensive damage to buildings and infrastructure.

- Many of the landslides in this image appear to be in sparsely populated areas. It is possible that landslides also occurred in areas south and east of the park that experienced intense shaking, but cloud cover on August 14 prevented Landsat from acquiring a clear view. Additional imagery from Landsat and other satellites should eventually provide more clarity about the extent of the landslides.

- The situation could be exacerbated in the coming days by heavy rains from tropical depression Grace. Some forecasts call for the storm to drop between 13 to 25 cm (5 to 10 inches) of rain on the areas hit hardest by the earthquake.


Figure 32: For comparison, this second image shows the same area on a clear day on 2 January 2021 (image credit: NASA Earth Observatory)

- “Some hillslopes that have been destabilized by the earthquake but did not become landslides may be pushed past the limit of stability by the rain, leading to further landslides,” said Robert Emberson, a landslide expert with NASA’s Earth Applied Sciences Disasters Program. “Debris and rock already mobilized by the earthquake may be transported by flash flooding as devastating debris flows. The material is mostly at the base of hills currently, but rivers quickly filled by rain could push that downstream and cause severe impacts to communities living farther from the location of the landslides.”

- NASA’s disasters program is monitoring the situation and coordinating with the United States Agency for International Development and other partners to share relevant data about the event with emergency responders. Data and updates from the team will be shared here.

• August 12, 2021: In the first two weeks of August 2021, Greece has endured a series of wildland fires that have charred a large swath of the island of Evia and several areas of the Peloponnese region. The fires followed closely after one of the worst heatwaves in the country since the 1980s, which dried up scarce moisture and left forests primed to burn. Greek Prime Minister Kyriakos Mitsotakis told several news agencies that the fire outbreak has been a “disaster of unprecedented proportions.” 38)


Figure 33: Fires in the country have consumed five times as much land as they do in an average year. The OLI instrument on Landsat-8 acquired natural- and false-color views (Figure 34) of the north end of Evia on August 10, 2021 (image credit: NASA Earth Observatory image by Lauren Dauphin, using VIIRS data from NASA EOSDIS LANCE, GIBS/Worldview, and the Suomi National Polar-orbiting Partnership. Story by Michael Carlowicz)

- According to data from the European Forest Fire Information System (EFFIS), more than 110,000 hectares (424 square miles) have burned in Greece this year, more than five times the yearly average from 2008 to 2020 (21,000 hectares). EFFIS counted 58 fires (30 hectares or larger) in the country in 2021, already above the yearly average total of 46.


Figure 34: This false-color image combines shortwave infrared, near infrared, and red light (OLI bands 6-5-4). In this view, burned vegetation appears dark brown, and greens and yellows indicate a combination of unburned trees and scrub (image credit: NASA Earth Observatory)

- Some of the worst fires in the country have burned on Evia, the second largest island in Greece and a major hub for tourism. Much of the island has been in a state of high fire alert for a week. The Associated Press reported that an estimated 50,000 hectares (123,000 acres) have burned on Evia, as well as hundreds of homes.

- Significant fires also broke out near Athens, Olympia, and Arcadia, and 63 organized evacuations have been reported across Greece in the past nine days. Firefighters and equipment have been sent from at least 15 countries to help Greek authorities.

- As of August 11, EFFIS reported that more than 338,000 hectares (1,300 square miles) have already burned across Europe in 2021, more than the 2008-2020 average for an entire year (295,000). More than 109,000 hectares have burned so far in Italy, 2.5 times the annual average. Large fires have also been burning in Algeria and Turkey.


Figure 35: On August 8, the Visible Infrared Imaging Radiometer Suite (VIIRS) on Suomi NPP acquired a wider view of the fires and smoke in Greece. NASA Worldview imagery from August 3–11 shows the evolution of the smoke plumes with changing winds (image credit: NASA Earth Observatory)

- The heatwaves and fires fit with patterns described in the latest assessment report from the Intergovernmental Panel on Climate Change (IPCC), to which NASA-funded scientists contribute. In its summary of climate conditions in Europe, the IPCC noted: “The frequency and intensity of hot extremes ... have increased in recent decades and are projected to keep increasing regardless of the greenhouse gas emissions scenario. Despite strong internal variability, observed trends in European mean and extreme temperatures cannot be explained without accounting for anthropogenic factors.”

• August 11, 2021: Every year, scientists at the University of Maryland publish new data about the state of Earth’s forests based on observations from Landsat satellites. As has often been the case in recent years, the update for 2020 painted a bleak picture. In that one year, Earth lost nearly 26 million hectares of tree cover—an area larger than the United Kingdom. 39)


Figure 36: Using satellite data from the past two decades, scientists are starting to pinpoint which crops and farming styles have lasting impacts on forests. This map is based on an analysis of Landsat data by The Sustainability Consortium and WRI, highlights several key drivers of forest loss. Shifting agriculture (yellow) typically involves the clearing of small plots within forests in Africa, Central America, and parts of South America. The clearing is done by subsistence farmers, often families, who raise a mixture of vegetables, fruits, grains, and small livestock herds for a few years and then let fields go fallow and move on as soil loses its fertility. The practice is especially common in Africa, and has become more so since 2000 due to increasing human populations. (image credit: NASA Earth Observatory images by Lauren Dauphin, using data from Curtis, P.G., et al. (2018), data from Goldman, Elizabeth, et al. (2020), and Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- The raw numbers can tell us how much and where forests were lost, but they do not explain what was driving those losses. How much deforestation was due to wildfires? Food production? Forestry management? An ongoing effort by researchers from The Sustainability Consortium and the World Resources Institute (WRI) attempts to answer such questions with maps and datasets that categorize and quantify the major drivers of annual forest losses. In doing so, the researchers have put a spotlight on the impact that food production has on forests, particularly in the tropics.

- In 2020, for instance, Earth lost about 4.2 million hectares (16,000 square miles) of humid tropical primary forest—an area about the size of the Netherlands. Nearly half of that, their analysis shows, was due to food production, and half of that was due to commodity crops. In recent years, commodity crop production has pushed rates of forest loss to record levels.


Figure 37: In South America and Southeast Asia, commodity crops (tan on the map) have become the dominant driver of forest loss. Common commodity crops include beef, soybeans, palm oil, corn, and cotton. They are typically grown on an industrial scale and traded internationally. Unlike the temporary forest clearings associated with small-scale agriculture, commodity-scale production often involves clear-cutting and results in significant impacts on forests (like the Indonesian palm oil plantation in this image), image credit: NASA Earth Observatory

- “In many cases, commodity-driven deforestation is essentially a permanent change compared to shifting agriculture,” explained Christy Slay, a conservation ecologist and the senior director of science and research applications at The Sustainability Consortium. “These areas will likely never be forests again.”

- In contrast, forests cleared for forestry management or by wildfires generally grow back over time. In the U.S. Southeast, for instance, managers maintain certain ecosystems and animal habitats by periodically burning and planting forests to mimic natural cycles of burning and regrowth. Likewise, forests in the Pacific Northwest and Europe are often managed for timber in ways that cycle between periods of forest clearing and periods of regrowth.

- Note that food production was once a major driver of deforestation in North America and Europe, but much of the clearing happened a hundred or more years ago. Since many forests in these areas were already gone by 2000, their absence does not register as forest loss. Nor does the map capture the impact of large-scale conversion of natural grasslands to agriculture, a common practice in both North and South America.

Figure 38: The NASA/USGS Landsat satellite mission is helping scientists study how the Amazon rainforest has changed over decades. The Amazon is the largest tropical rainforest in the world, but every year, less of that forest is still standing. Today's deforestation across the Amazon (video credit: NASA Goddard)

- With tropical forest cover dwindling and the effect of climate change becoming more acute, some companies and consumers are trying to ensure that food production does not lead to new deforestation. In recent years, hundreds of companies have committed to eliminating or reducing products in their supply chains that cause deforestation. But ensuring that is often challenging.

- “Global supply chains can be complicated and opaque,” said Slay. “You often have companies buying commodities off the spot market, such that the source regions change frequently or even daily. Retailers and food manufacturers often don't know the source of their ingredients down to the individual farm and field scale.”

- By regularly collecting data on the health of forests, satellites are making it easier for scientists to untangle which commodities and regions are the biggest contributors to deforestation. Doug Morton, a forest ecologist at NASA’s Goddard Space Flight Center, has witnessed a shift in the dominant drivers of deforestation.

- “Forty years ago, we often saw small-scale deforestation creating roads that look like fishbone patterns,” said Morton, who monitors agricultural frontiers in the Amazon. At the time, many people were moving into the Amazon to escape drought and hunger in eastern Brazil. “By the middle of the Landsat record, we see large-scale commodity production taking hold. Today’s deforestation isn’t about individual families. It’s often tractors and bulldozers clearing large tracts of forest for industrial scale cattle ranching and crops.”

- For companies trying to keep their supply chains free of deforestation, knowing which commodity crops are being grown where is critical. “If we know where deforestation is common and what crops are involved, we can go to companies and say: ‘Be careful if you’re working with suppliers that are sourcing this particular product from this particular part of the world,’ ” said Slay. “Satellite data of forest change and loss is the first step in the process.”

- One recent WRI analysis combined Landsat imagery with economic and land-use data to parse the impact of seven different commodities on forests around the world. “One of the big things you notice in the data is the outsized role of cattle pastures in driving deforestation,” said Mikaela Weisse, one of the report’s authors. “Cattle pastures caused about five times more deforestation than any of the other commodities we analyzed.”


Figure 39: The map shows forests being cleared for cattle all over the world, but particularly in Brazil, where deforestation has been on the rise. Large tracts of forest have also been cleared in Paraguay, Bolivia, and Peru according to WRI data (image credit: NASA Earth Observatory)

- In Southeast Asia, where deforestation rates have dropped recently, most forest losses are associated with palm oil, which is used in many types of processed foods and various health and beauty products like deodorant, shampoo, toothpaste, soap, and lipstick. Deforestation for cocoa production had a sizable impact in certain countries—notably Ghana and Côte d'Ivoire—but only represented 3 percent of total forests losses. Other commodities with similarly modest effects on global forests included rubber, coffee, and wood fiber.

- While new tools are making it easier to understand where food production is intersecting with new deforestation, huge challenges remain. “Deforestation rates are going up instead of down,” said Elizabeth Goldman of WRI. “There’s a lot of work left to do.”

• August 4, 2021: In the midst of a severe heatwave and following months of dry weather, Turkey is facing some of its worst wildfires in years. Over the past seven days, more than 130 wildfires have been reported across 30 Turkish provinces. Most of the fires have ignited along the Mediterranean and Aegean Sea coasts, several in resort areas around Antalya, Mugla, and Marmaris. 40)


Figure 40: On 31 July 2021, the OLI instrument on Landsat-8 acquired natural-color imagery of fires near the coastal towns of Alanya and Manavgat (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and MODIS data from NASA EOSDIS LANCE and GIBS/Worldview. Story by Michael Carlowicz)


Figure 41: Detail image of Landsat-8. More land has been consumed already this year than usually burns in an entire year in the country (image credit: NASA Earth Observatory)

- The European Forest Fire Information Service reported more than 136,000 hectares (525 square miles) have burned in Turkey already this year, about three times the average for an entire year. The European Space Agency’s Sentinel-3 satellite also acquired a view of the fires on July 30.


Figure 42: As of August 3, at least nine wildfires were still burning across Turkey. The MODIS instrument on NASA’s Aqua satellite captured a wider natural-color image of several of them near Antalya and Marmaris (image credit: NASA Earth Observatory)

- Fires were still being fed by strong winds, air temperatures above 40º Celsius (104° Fahrenheit), and low humidity. Croatia, Iran, Spain, Russia, Ukraine, and Azerbaijan provided equipment and personnel to help Turkish firefighters bring the blazes under control.

- Much of southern Europe has been baking for weeks under extreme heat not seen since the 1980s. National temperature records were set in both Greece and Turkey in the past month. Air temperatures reached 45°C (113°F) in Greece and surrounding areas yesterday, and the heat is forecasted to continue for several days. Fires are also burning this week in Greece and Lebanon.

• July 21, 2021: Covered with lakes, forests, and mountains, Dalarna County has been called “Sweden in miniature.” But the same region that today draws people to its idyllic lakeside villages and midsummer celebrations was also the site of an ancient, catastrophic impact. 41)


Figure 43: The idyllic region of Dalarna County is the site of an ancient, powerful collision. The Siljan impact structure, or “Siljan Ring,” is visible in this image, acquired on June 24, 2020, with the Operational Land Imager (OLI) on Landsat-8. Measuring more than 50 kilometers (30 miles) across, Siljan is the largest-known impact structure in Europe and among the top-20 largest on Earth (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- Around 380 million years ago, in the Late Devonian period, an asteroid slammed into the land that is now south-central Sweden. The impact left quite a mark. Even after hundreds of millions of years of erosion, the scar is still recognizable. It is especially apparent when viewed from above.

- Surveys of the structure have shown that the ground is slightly raised up across parts of the crater’s center. It is surrounded by a ring-like graben, or depression, which today is partially filled with water. Lake Siljan, on the crater’s southwest side, is the largest lake; it connects to Lake Orsa via a small river.

- People have lived for millennia near the crater without knowing its cosmic origin. In the late 1960s, scientists used drill cores to uncover the complex and ancient geology deep below the ground.

- Research at Siljan is ongoing today. In a 2019 study, scientists described how they used drill cores to find that the deep, fractured rocks in the crater were suitable for ancient life. A subsequent paper in 2021 described the fossilized remains of fungi discovered at a depth of more than 500 meters.

• July 19, 2021: For several months, communities along the west coast of Florida have observed substantial blooms of the harmful algae Karenia brevis. The algae occur naturally in the waters around Florida, but the bloom in 2021 has been particularly bad near Tampa Bay, causing large-scale fish kills in what some people refer to as a ‘red tide’ event. The bloom is also unusual for how early it is occurring. 42)


Figure 44: The natural-color images of Figures 44 and 45 were acquired on July 14, 2021, by the OLI instrument on Landsat-8. The scene from Tampa Bay north to Horseshoe Beach shows dynamic coastal waters, with plumes of dissolved organic matter (dark brown to black) running off from the land; shallow seafloors and re-suspended sediment from the bottom (brighter greens and blues); and some hints of algae and phytoplankton (often diatoms) in green (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the USGS and chlorophyll data from the Harmful Algal Bloom Monitoring System from the National Centers for Coastal Ocean Science/NOAA using modified Copernicus Sentinel data (2021) processed by the European Space Agency. Story by Michael Carlowicz)

- Karenia brevis is a microscopic algae that, like other phytoplankton, can multiply into massive blooms when there are enough nutrients in the water—often in the autumn along the Gulf Coast. The algae produce neurotoxins that can kill fish and cause skin irritation and respiratory problems for humans, particularly those prone to asthma and other lung diseases. In extreme concentrations, K. brevis can turn water brown, red, black, or green; however, it is not always visible from space.


Figure 45: Tampa Bay is teeming with Karenia brevis months before it usually blooms (image credit: NASA Earth Observatory)

- “This Karenia brevis ‘red tide’ bloom is doubly unusual,” said Richard Stumpf, an oceanographer for the National Oceanic and Atmospheric Administration (NOAA). “It is summer, which is rare, and it is intense well into Tampa Bay, which is rare even during a ‘normal’ fall bloom.”

- “If a bloom is out on the continental shelf, it is more easily diluted,” said Chuanmin Hu, an optical oceanographer at the University of South Florida (USF). “The bloom this year is so troublesome because it is both inside Tampa Bay and around the Tampa Bay mouth.”


Figure 46: This map, based on data processed by the NOAA National Centers for Coastal Ocean Science, shows measurements of chlorophyll fluorescence on July 11, 2021. Scientists can use fluorescence and distinct wavelengths of light to detect signatures of algae and phytoplankton amid turbid, churning waters along the coast. The data are collected by the Copernicus Sentinel-3 satellite of the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). Similar observations from July 13 are available from the Optical Oceanography Laboratory at the University of South Florida (image credit: NASA Earth Observatory)

- “Although Karenia brevis blooms are common to the West Florida Shelf and have been observed in almost every coastal region of the Gulf of Mexico, I have never seen anything like that inside Tampa Bay,” said Inia Soto Ramos, an ocean color specialist at NASA’s Goddard Space Flight Center (GSFC) and former researcher at USF. “Massive blooms were observed back in the late 1990s, and even the Spanish conquistadors described them in their books. But the bloom this year inside the bay is worrisome. It could be a one-year thing, and hopefully it is. But if water quality in the bay continues to decline, residents should prepare for more blooms, and not only K. brevis.”

- Since early June 2021, Karenia brevis has been abundant along the Gulf Coast from just north of Clearwater to Sarasota. In a July 14 report, the Florida Fish and Wildlife Conservation Commission noted: “A bloom of the red tide organism, Karenia brevis, persists on the Florida Gulf Coast. Over the past week, K. brevis was detected in 107 samples.”

- According to the Sarasota Herald-Tribune, coastal work crews have collected more than 600 tons of dead fish and marine life killed by the bloom. On July 15, the city council of St Petersburg asked the governor to declare a state of emergency over the bloom. Officials are still trying to pinpoint the trigger for the event, but many scientists note that the area has been unusually rich with algae-sustaining nutrients in 2021.

- “Karenia brevis blooms, although studied for decades, do not follow a strict recipe. Some years, circulation and advection are the main drivers,” said Soto Ramos. “However, we know if there is an excess of nutrients, the algae will utilize them. I think the bloom right now is due to a combination of available nutrients, warm temperatures, and circulation patterns keeping the algae contained within the bay. Once the algae are there, they stay for a while.”

- NASA is currently developing the Plankton, Aerosol, Cloud, ocean-Ecosystem (PACE) satellite mission for launch around 2024. The satellite is being designed with sensors tuned to the signatures of blooms. “Whereas heritage ocean color instruments observe roughly six visible wavelengths, PACE will collect a continuum of colors that span the visible rainbow,” said Jeremy Werdell, project scientist for PACE at NASA GSFC. “Its ocean color instrument will be the first of its kind to collect hyperspectral radiometry on global scales, which will allow unique and highly advanced identification of aquatic phytoplankton communities, including potentially harmful algae such as these on the West Florida Shelf.”

• July 17, 2021: In recent decades, aquaculture has boomed in Andhra Pradesh. The state has become one of India’s largest producers of farmed fish and shrimp. Among the reasons for the boom:a major expansion a of inland aquaculture farms along rivers and canals where people once raised crops. 43)


Figure 47: The OLI instrument on Landsat-8 acquired this natural-color detail image of an area dense with inland aquaculture ponds along the Upputeru River on June 8, 2021. Aquaculture ponds appear dark green. Farmland is generally brown. Coastal areas with mangrove forests are lighter green (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- According to satellite imagery, aquaculture was scarce in this area in the mid-1980s. Now carp, catfish, and other types of finned fish are commonly raised in the area. There are numerous shrimp ponds, too, which tend to be the narrow according to one satellite survey of the area.

- The Indian government established the first aquaculture ponds in this area in the 1970s around Lake Kolleru. Since then, the initial success of those projects has made aquaculture an appealing and profitable choice for many farmers in the region who regularly dealt with crops being flooded, the intrusion of salt into water used for irrigation, and Bay of Bengal cyclones.


Figure 48: Inland areas along rivers and canals where people once raised crops are now dotted with fish and shrimp ponds (image credit: NASA Earth Observatory)

- Despite the expansion, India’s aquaculture sector has faced challenges recently. One recent study calculated that its shrimp farming sector may have lost as much as $1.5 billion in 2020-2021 due to disruptions related to the pandemic. The state of Andhra Pradesh accounts for about 70 percent of India’s shrimp production.

• July 11, 2021: In research published in 2017, scientists reported that summer pulses of freshwater from melting glaciers along Greenland’s southwest coast often coincide with phytoplankton blooms. The flow of fresh meltwater out to sea carries nutrients that can sustain and promote abundant growth of the floating, plant-like organisms that form the center of the ocean food web. 44)


Figure 49: Pulses of fresh glacial meltwater and nutrients provoke summertime phytoplankton blooms. That appears to be what was happening in the waters off of Nuuk, Greenland, when the Operational Land Imager (OLI) on Landsat-8 flew over on July 8, 2021. Close to the coast, the water in Ameralik Fjord and other inlets is stained chalky tan and gray by sediments and glacial flour—rock that has been ground to powder by the ice sheets. Offshore in the Labrador Sea and Davis Strait, light green swirls indicate the presence of phytoplankton in summer bloom. Chlorophyll measurements confirm this (image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Michael Carlowicz)

- The waters of the Labrador Sea, Davis Strait, and Baffin Bay—between Greenland and Nunavut, Canada—form a transitional zone between the Arctic and Atlantic oceans. Fresh meltwater from the ice sheets and strong regional tides (which promote nutrient mixing) help make these waters biologically rich, particularly in summertime. The abundant phytoplankton draw in copepods and other grazers that ultimately feed shrimp, cod, and other species up to the size of whales.

- In the 2017 paper, Stanford University ocean scientist Kevin Arrigo and colleagues noted that summer blooms tend to start in early July and can extend as far as 300 kilometers (200 miles) offshore from Greenland. Fed by sunlight and water rich in iron, silicate, and phosphorous, the blooms account for about 40 percent of annual net primary production for the region.

- Blooms in high-latitude and Arctic waters are happening more often and lasting longer, according to another study published in 2020 by Arrigo’s research group. Incorporating satellite data from NASA’s SeaWiFS and MODIS instruments, they found that the rate of growth of phytoplankton biomass across the Arctic Ocean increased by 57 percent between 1998 and 2018. The study contradicted an older idea that increasing glacier melting might lead to fewer nutrients and blooms.

• July 8, 2021: Toward the end of the last Ice Age, as mile-thick glaciers weighed down the land surface and then melted, parts of New England and eastern Canada became inundated by water. Some lowlands flooded and formed inland basins like the Champlain Sea. 45)

- Ten thousand years later, with seas now rising because of global warming, scientists are combing through an array of data and building increasingly detailed models to understand the processes that drive regional and local changes in sea level. The goal is to project when, where, and how much seas are likely to rise in the coming decades and centuries. It's an incredibly complicated set of interdependent calculations.


Figure 50: While scientists have grown more confident about projections of sea level rise for the next few decades, many competing factors make it hard to see far into the coastal future (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and sea level rise projections courtesy of Benjamin Hamlington/NASA/JPL-Caltech. Story by Adam Voiland)

- “People tend to think that sea level is like a bathtub with the water level simply rising and falling depending on how much water is coming out of the faucet,” said paleoclimatologist Anders Carlson of the Oregon Glaciers Institute. “In reality, it’s more like a spinning bathtub that’s changing shape, moving up and down, and has water pouring into and out of different drains and over the sides. Where the water will ultimately slosh over the edge of the tub is influenced by many things, making it difficult to say where the overtopping will occur.”

- Despite the complexities, the scientific understanding of the factors that control sea level has improved dramatically in recent decades, as have measurements of past sea level change and projections of future change.

- “We can tell you how much the ocean has warmed in recent decades, and how much more space the water takes up. We have satellites and other tools that have measured that,” said Ben Hamlington, the current lead of NASA’s sea level change team. “The same thing is true for several of other factors that influence sea level, such as the mass of the ocean, the salinity, and how much water is stored on land.”

- That growing knowledge base is why scientific organizations like the Intergovernmental Panel on Climate Change (IPCC) are publishing sea level rise projections with increasing levels of confidence. In its 2019 report, the IPCC projected (chart Figure 50) 0.6 to 1.1 meters (1 to 3 feet) of global sea level rise by 2100 (or about 15 mm per year) if greenhouse gas emissions remain at high rates (RCP8.5). By 2300, seas could stand as much as 5 meters higher under the worst-case scenario. If countries do cut their emissions significantly (RCP2.6), the IPCC expects 0.3 to 0.6 meters of sea level rise by 2100.

- A host of competing factors will influence how global sea changes translate to regional and local scales. Among them: the rising or falling of the land surface due to plate tectonics and human activity; gravity anomalies that can create regional bulges and dips in sea surface height; variations in the temperature and salinity of seawater; changes in the amount of water stored on land in reservoirs; isostatic adjustment due to the addition, loss, and movement of land ice; and changes in erosion and how much sediment rivers carry to coastal areas.


Figure 51: Sorting out how river deltas will respond is a particularly thorny and consequential issue. Tens of millions of people live on river deltas around the world (such as India’s Krishna Delta), and many of them are subsiding (sinking), often at twice the mean rate of sea level rise. The subsidence is due to a combination of factors like the natural settling of sediments, groundwater and oil extraction, and the extra weight of buildings. Inland dam construction and land management practices can also starve deltas of the raw material needed to replenish and build coastal land (image credit: NASA Earth Observatory, Landsat-8 image of 8 June 2021)

- But it is hard to predict human settlement patterns—and the subsidence it causes—decades or centuries from now. Many IPCC projections do not even attempt to incorporate estimates of this subsidence partly because of the uncertainties in future land use and human behavior and because there is a lack of readily available, large-scale data on vertical land motion to feed into models of sea level rise.


Figure 52: The current shortage of land motion data is poised to become an abundance with the launch of the NASA-ISRO Synthetic Aperture Radar (NISAR) mission in 2022. The radar will make daily, global measurements of land motion that sea level experts like Manoochehr Shirzaei of Virginia Tech say will lead to major improvements in regional sea level rise projections (image credit: NASA Earth Observatory)

- Likewise, teams of scientists have been surveying the fast-sinking Mississippi River Delta to get a better understanding of how changes in sediment and vegetation affect the delta. Scientists participating in NASA’s Delta-X campaign have collected several types of data to develop and calibrate a model of how the delta might respond to rising sea levels in the next century.

- “The combination of anthropogenic subsidence and increasing rates of sea level rise is a five-alarm fire for many delta cities,” said Shirzaei. “Places like New Orleans, Kolkata, Yangon, Bangkok, Ho Chi Min City, and Jakarta will undoubtedly face increasing pressures from flooding and saltwater intrusion.”

- Still, the long-term picture—hundreds of years into the future—is unlikely to be perfectly clear. “When you think about future impacts of sea level rise, you also have to consider what people might do in response," said Hamlington. Some countries—like The Netherlands and the United States—have already built elaborate sea walls and water-control systems that protect vulnerable deltas like the Rhine and the Sacramento-San Joaquin. They will likely continue reinforcing these systems as sea levels rise. In others deltas, like the Krishna (Figure 51) or Ganges in India, the Chao Phraya in Thailand, and the Mekong in Vietnam, coastal defenses are more limited so far.

Figure 53: It's hard to "see" sea level rise by just looking at the ocean, but its effects are very real. A new video covers some of the basics (video credit: NASA/JPL-Caltech)

- “The reason people within the scientific community are working so hard on regional sea level rise projections is that if we can get them right, it will give cities and nations a chance to prepare,” said Hamlington. “Even if some of the more distant projections are inexact, they still provide critical constraints that could end up being the difference between places that successfully adapt to rising seas and those that experience the most damaging consequences.”

Minimize Landsat-8 continued

• July 6, 2021: With heights ranging from 600 to 1800 meters (2,000 to 5,900 feet), the Barberton Makhonjwa Mountains in South Africa and Eswatini are not particularly tall. What distinguishes the belt of greenstone rock formations found here is their age. 46)


Figure 54: Rare igneous rock and early signs of life are found beneath the grassy hills of the mountain range in South Africa and Eswatini. The natural-color image shows part of the Komati River Valley in South Africa. Lava flows made of komatiites were first identified within this valley in 1969. The image was acquired by the Operational Land Imager (OLI) on Landsat-8 on March 10, 2021. The United Nations Educational, Scientific and Cultural Organization declared the mountains a World Heritage Site in 2018 (image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- Beneath the rolling, grassy uplands and forested valleys of the mountain range, lie some of the oldest, best-preserved, and diverse sequences of volcanic and sedimentary rock layers found anywhere on the planet. They hold evidence of some of Earth’s earliest forms of life, including microfossils, stromatolites, and other biologically derived material. Geological sampling indicates that some rock formations in these mountains are 3.2 to 3.6 billion years old.

- One type of rock in this area that especially intrigues geologists is komatiite. The rare igneous rock formed from magmas that were hotter, more liquid, and denser than any lavas found on Earth today. Geologists still debate what conditions allowed komatiite to form, but many think Earth’s mantle was likely hotter or wetter 3 billion years ago than today, and that likely played an important role.

• July 3, 2021: On March 19, 2021, the Fagradalsfjall volcano erupted after lying dormant for 800 years. Three months later, the volcano on Iceland’s Reykjanes peninsula is still spewing lava and expanding its flow field. 47)


Figure 55: Lava flows from the Icelandic volcano were estimated to cover a total area of 3 km2, three months after the eruption began. The natural-color images show the lava flow progression from March, May, and June 2021. Note the ground around the volcano was still covered in snow in March. The darkest areas in May and June show where lava has cooled and piled up across the valley floors. Fresh lava flows that are still hot appear orange. All of the images were acquired by the Operational Land Imager (OLI) on Landsat-8 (image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel)

- The Icelandic Met Office reported that by May 3, the lava flow was largely confined to one main crater, a fifth fissure that opened in April. In late May, lava flows broke through an artificial barrier built to contain it; the lava continued flowing south towards Nátthagi Valley. The lava flow has since cut off access to the most popular hiking trail to the eruption site. As of June 15, the lava flows were estimated to cover a total area of 3 square kilometers (about 1 square mile), with an estimated volume of 63 million cubic meters.

- Icelandic officials are concerned that a prolonged eruption will cause lava to flow south and cross the Suðurstrandarvegur, a road used to transport goods and connects Reykjanes peninsula to South Iceland. After crossing the road, the lava flow could continue toward the ocean.

• June 30, 2021: Skies were clear and the waters of Mistastin Lake were placid when the Operational Land Imager (OLI) on Landsat 8 captured this natural-color image of Labrador, Canada, on a fall day in 2017. 48)


Figure 56: The lake covers part of a crater where an asteroid once slammed into Labrador, Canada (image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- The scene would have looked quite different about 36 million years ago when an asteroid smashed into Earth and left an impact crater where the lake (called Kamestastin by the Innu people) now sits. While erosion has changed and obscured some of the features, a 50-meter (164-foot) wall still rings much of the crater. Geologists estimate the original crater had a diameter of about 28 kilometers (17 miles)—about twice the size of the current lake.

- Parts of the central peak are also visible in the lake as Horseshoe Island. These mound-like features are often found in the center of large craters as a product of the melting and rebound of subsurface rocks. Meanwhile, the elongated, elliptical appearance of the crater is a result of periods when glaciers slid across this area during several ice ages.

- Based on the presence of an unusual diamond-like mineral called cubic zirconia, the asteroid impact must have heated rocks at the site to at least 2370°C (4,300°F). That would be the hottest-known temperature recorded by a surface rock on Earth, according to one team of researchers.

• June 21, 2021: The Yukon-Kuskokswim Delta is one of the world’s largest deltas, and it stands as remarkable example of how water and ice can shape the land. These images show the delta’s northern lobe, where the Yukon River spills into the Bering Sea along the west coast of Alaska. 49)


Figure 57: One of the world’s largest deltas stands as remarkable example of how water and ice can shape the land. “The Yukon Delta is an exceptionally vivid landscape, whether viewed from the ground, from the air, or from low-Earth orbit,” said Gerald Frost, a scientist at ABR, Inc.—Environmental Research and Services in Fairbanks, Alaska. The vivid landscape is captured in these images acquired with the Operational Land Imager (OLI) on Landsat-8 on May 29, 2021. The images are composites, blending natural-color imagery of water with a false-color image of the land (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- While the image could be considered a work of art, there are some useful aspects to looking at the land this way. For example, you can easily distinguish areas of live vegetation (green) from land that is bare or contains dead vegetation (light brown) from the network of sediment-rich rivers and ponded flood water (dark brown). A sprinkling of thermokarst lakes are also part of the scene.


Figure 58: Detail image of the Yukon Delta (image credit: NASA Earth Observatory)

- In general, the green areas across the delta are tall willow shrublands. They are especially apparent on either side of the river channels in the detailed image above. The light-brown areas are primarily moist sedge meadows; they appear brown because much of it is the dead remains of last year’s growth. Away from the delta (right side of the image) the vegetation is shrub-tussock tundra.

- “To me, one of the interesting things about the delta is that it is a highly transitional area, with some elements of Arctic tundra and some of boreal forest,” Frost said.

- The delta also transitions with the seasons. At the time of this image, the signature of spring flooding is written across the delta. Melting snow and ice cause the rivers to spill over their banks and by late May, many of the marshes are filled with floodwater, which appears as dark-brown ponds.

- According to Lawrence Vulis, a graduate student at the University of California, Irvine, the delta would have appeared much more inundated immediately following the melting of snow and ice a few weeks prior to this image. Stream gauges and satellite images suggest that the bulk of the flooding had already subsided. Still, the flooding was recent enough that the plenty of ponding remained on May 29. As summer advances, the floodwater will continue to recede and the wetlands will continue to green up with vegetation.

- Also notice the colorful water where the delta meets the Bering Sea. This is a product of glacial runoff far upstream, which carries a large amount of sediment toward the coast. This sediment is also instrumental to the formation of tall “levees” on the sides of the channels, deposited there when floodwaters spill over their banks. These “levees” support stands of tall willows—important habitat for moose.

- “Interestingly, tall shrubs have expanded a lot on the delta in recent decades, and the moose have followed,” Frost said. “Today, the delta has one of the highest moose densities in the state of Alaska.”

- The delta did not always look this way. Studies have shown that the modern Yukon Delta is just a few thousand years old. It’s young age “is incredible to think about,” Vulis said. “We are used to thinking about relatively ancient landscapes, but modern river deltas have only formed in the last 10,000 to 8,000 years since global sea level has stabilized.”

- The delta could quite possibly look different in the future. “The Yukon and other Arctic deltas are thought to be particularly vulnerable to climate change,” Vulis noted, “due to the roles of permafrost and ice in shaping these deltas.”

• June 16, 2021: Over the past three decades, small-scale gold mining has led to more than 100,000 hectares (250,000 acres) of forest loss in the Peruvian Amazon. While government agencies and conservation groups have successfully curbed such activity in recent years, new mining hotspots still pop up in unauthorized zones. 50)


Figure 59: The natural-color Landsat-8 images show the spread of mining activity along the Pariamanu River between May 2020 and May 2021, in a popular gold mining district in Peru’s Madre de Dios department. According to the Monitoring of Andean Amazon Project, more than 200 hectares (500 acres) had been deforested in the Pariamanu area since 2017. The new mines are located outside the permitted mining corridor (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and topographic data from the Shuttle Radar Topography Mission (SRTM). Radar Mining Monitoring Tool data courtesy of the SERVIR-Amazonia Program. Story by Kasha Patel)

- Peruvian researchers from Conservación Amazónica (ACCA), working with NASA and the Peruvian government, have developed a satellite-based tool to locate emerging mining hotspots in the Amazon. The Radar Mining Monitoring Tool (RAMI) identifies areas that appear to have new mining activity and examines their proximity to protected buffer zones, indigenous lands, mining concessions, and already degraded lands. The information is shared with Peruvian authorities to help pinpoint new activity and stop illegal mining.

- “Authorities may have limited resources to find all of the mining hotspots in a region,” said Sidney Novoa, project manager for SERVIR-Amazonia and a researcher at ACCA. “There are many systems that monitor forests in Peru, but no one else is focused just on gold mining. Our main goal is to empower authorities and give them enough resources to prioritize and focus their efforts.”


Figure 60: These Landsat-8 images show examples of two different artisanal gold mining techniques. Highly mechanized excavation uses heavy machinery to dig into the ground. Minimally mechanized mining uses high-pressure water cannons and suction pumps to move sediments from the bottom of creeks and rivers (image credit: NASA Earth Observatory)

- The research team is particularly interested in detecting small-scale, artisanal gold mining, typically operated by independent miners known as garimperos. In recent decades, the number of artisanal mines in the Madre de Dios department of Peru has increased, carving out a larger environmental footprint than industrial mines. Artisanal miners use mercury, which can pollute water sources and lead to neurological disorders or kidney issues in humans who are exposed to it. Illegal mining in protected zones also can compromise the homes of indigenous people, as well as endemic plants and animals.


Figure 61: This map, which uses data from the RAMI tool, shows a snapshot of mining in southeastern Peru from January to May 2021. The red polygons show changes in land cover attributed to new mining activity. The data are derived from synthetic aperture radar (SAR) observations made by the European Space Agency’s Copernicus Sentinel-1 satellite. Unlike optical imagery from Landsat, SAR penetrates cloud cover and allows more frequent observations in the often-cloudy region (image credit: NASA Earth Observatory) .

- “I used to put aside looking at mining activity during the rainy season because authorities thought it would not occur intensely during those months,” said Novoa. “We finally have information to demonstrate the importance of monitoring mining activities during the rainy season.” He explained that activity typically intensifies once the peak of the rainy season passes in February and March, as garimperos use the accumulated water in forests and waterways for their mining.

- RAMI mining alerts are updated every 15 days depending on the availability of satellite data. Novoa’s team creates private reports about new hotspots for the Monitoring of the Andean Amazon Project (MAAP), an initiative from Amazon Conservation Association and ACCA. Summaries are also shared directly with Peruvian authorities. The RAMI tool is hosted on the Internet and open to the public.

- RAMI is supported by SERVIR-Amazonia, a joint program of NASA and the U.S. Agency for International Development (USAID) that uses remote sensing to provide support for sustainable development. It is led locally by the Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT). The RAMI tool was co-developed with and incorporated data from ACCA, Peru’s Ministry of Environment, and Peru’s National Forest Conservation Program.

• June 15, 2021: People have been mining for gold in Ghana for centuries. Long before European colonists set foot in the area in the 1400s, Ghanaians looked for gold with pickaxes, shovels, and pans. They washed or “panned” for gold along river banks or dug holes on the surface to find deposits of gold dust and nuggets. Serious indigenous miners dug deep tunnels—records indicate some up to 80 feet deep. 51)


Figure 62: The natural-color image shows a large-scale mine and several artisanal mines in the Central Region of Ghana. The image was captured on March 29, 2020, by the Operational Land Imager (OLI) on Landsat-8. The mines lie within the Ashanti gold belt, one of the richest gold regions in West Africa (image credit: NASA Earth Observatory images by Lauren Dauphin, using data from Barenblitt, Abigail, et al. (2021), Landsat data from the U.S. Geological Survey, and topographic data from the Shuttle Radar Topography Mission (SRTM). Story by Kasha Patel)

- A version of this small-scale mining persists—known today as artisanal mining—but new research shows it is having a growing and sometimes devastating effect on the environment. Researchers from NASA, U.S. universities, and government agencies in Ghana recently used satellite data to estimate the extent of vegetation lost to artisanal mining in the southwestern portion of the country, where the majority of gold mining takes place. They found artisanal mining accounted for 25 percent of vegetation loss in the region from 2005-2019.

- “The accumulation of the small-scale mines across the landscape is startling. The deforestation impact is huge compared to industrial mines,” said Abigail Barenblitt, the main author of the study and data analyst in the biosphere sciences lab at NASA’s Goddard Space Flight Center. Although it only accounts for about one-third of the country’s gold production today, artisanal mining caused seven times more deforestation than industrial efforts between 2007-2017.

- The differing impact of artisanal and industrial mines is related to the process of gold extraction, said study co-author Amanda Payton of East Carolina University. Large-scale industrial mines use heavy machinery to dig deep in a concentrated area. Industrial miners are also sometimes required to fill the holes in the landscape after extraction in order to help with remediation.

- Because they typically do not have heavy-duty equipment, artisanal miners tend to dig many shallow holes across large swaths of land. They extract and process gold at the site and then move onto the next area. They usually do not refill holes after extractions. And they often use mercury to remove gold from sediments, which can lead to serious health problems and long-term water and soil contamination. Unregulated artisanal mining is locally known as galamsey, derived from the Ghanaian words “gather” and “sell.”

- “Artisanal mining has a quicker turnaround time on the landscape, with operations excavating a shallower area and then moving on to another section of the river. Some of the artisanal mines stretch for great distances along rivers,” said Payton. “With industrial mining, more research into the gold deposits is done and more resources are committed to a single area of land to excavate deeper.”

- Barenblitt, Payton, and colleagues worked with the Ghana Space Science and Technology Institute and Ghana Statistical Service to determine the total footprint of vegetation loss to artisanal gold mining. They analyzed decades of Landsat data, creating a machine-learning algorithm to classify any vegetation loss in one of four categories: mining, urban development, water, and other (agriculture, bare soil, etc.). The team found more than 160,000 hectares (400,000 acres) of vegetation were lost from 2005-2019. About 28 percent was lost to both industrial and artisanal gold mining, while 29 percent was lost to urban development. About 17 percent was converted to water, mainly due to the formation of a lagoon complex. The remaining 25 percent was attributed to the “other” category of land losses.

- The team further classified mining as large-scale industrial or small-scale artisanal by looking at elevation data and the texture of the landscape. Industrial mines have larger elevation changes since they dig deeper into the surface. Highly textured landscapes tend to indicate artisanal mining due to the small holes compared to wider, smoother industrial areas. Artisanal mines accounted for 85.7 percent of vegetation loss, while industrial mines accounted for 14.3 percent from 2005-2019.


Figure 63: In the past decade, unregulated artisanal mines accounted for more deforestation than industrial mines. The team also calculated artisanal mining activity by year to pinpoint its rise in popularity over the past decade. The map above shows new mining activity over a subset of the study period from 2007 to 2017 near Kumasi, Ghana. (The researchers chose 2007 to 2017 to focus on years with sufficient cloud-free imagery to identify annual changes.) Darker orange and red represent more recent activity. At least 700 hectares (1,700 acres) of loss occurred in protected zones (image credit: NASA Earth Observatory)


Figure 64: This graph shows the amount of artisanal and industrial mining by year (image credit: NASA Earth Observatory)

- “There has definitely been an expansion of small-scale mining by more people over time because of the price of gold,” said Lola Fatoyinbo, a forest ecologist at NASA’s Goddard Space Flight Center and a contributor to the study. “The high gold prices probably made artisanal mining worth the labor.” The team calculated a correlation between gold prices and annual mining conversion after a two-year lag. Over the study period, the value of gold jumped from about $700 per ounce to as high as $1,700 in 2012; it is now nearly $1,800 per ounce.

- The team’s next step is to automate their image analysis process so new mining can be detected quickly by African and international organizations addressing the issue. This research is part of larger efforts across NASA to detect unregulated gold mining in Ghana. The team has collaborated and compared methods with researchers of SERVIR-West Africa, a program between NASA and the U.S. Agency for International Development (USAID). SERVIR-West Africa uses similar mining analyses and shares the data to government officials to help curb artisanal activities and reform past mining sites.

• June 4, 2021: Off the coast of Sonoma and Mendocino counties, changing climate and a marine epidemic have combined to decimate one of California’s most productive ecosystems. In the span of a single year, the region’s renowned kelp forests almost completely collapsed, and they are still struggling. Floating forests that once harbored and fed many marine species have turned into barrens devoid of biodiversity. 52)


Figure 65: Changing climate and a marine epidemic have combined to decimate one of Northern California’s most productive ecosystems. This map,based on data from McPherson and colleagues, shows the location of bull kelp forests in 2008 and 2019 (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and topographic data from the USGS 3D Elevation Program (3DEP). Historical sea surface temperature image by Jesse Allen, using microwave and infrared multi-sensor SST data from Remote Sensing Systems. Photo by Steve Lonhart, NOAA Monterey Bay National Marine Sanctuary. Story by Laura Rocchio, Landsat Communication and Public Engagement Team, with Mike Carlowicz)

- Using 34 years of Landsat imagery, a team of researchers led by Meredith McPherson of the University of California, Santa Cruz, documented the fast and catastrophic collapse of the once hardy kelp forest, as well as its struggle to regenerate. The research team found that the Northern California kelp canopy declined more than 95 percent in 2014-15, and the effects persisted for five years.


Figure 66: These images show the same areas as observed in shortwave infrared, near-infrared, and red light by Landsat-5 (bands 7,5,3) in 2008 and Landsat-8 (bands 7,5,4) in 2019. The combination helps make some kelp forests visible from space (image credit: NASA Earth Observatory)

- Bull kelp is a canopy-forming macroalgae that flourishes in nutrient-rich, cool water and grows as much as 60 cm (nearly 2 feet) per day. The kelp is considered an “ecosystem engineer”—the foundational species of a nearshore ecosystem that feeds and shelters other marine life. It is the dominant kelp species north of Monterey Bay, California, with underwater forests thriving along 160 km (100 miles) of rocky reefs from Fort Bragg to Jenner.

- Unlike the giant kelp more common to the south, bull kelp is an annual species that grows vigorously from June through August. It then disperses its spores before fall and winter storms dislodge the mature plants from their rocky perches. While the exact location and extent of the bull kelp can change from year to year (based on spore dispersal and environmental factors), the underwater forest has historically regenerated regularly.

- Looking across several decades of Landsat observations, McPherson and colleagues found that the geographic distribution of bull kelp contracted, first receding in 2008 in the sandier regions north of Fort Bragg, and then in 2012 in sandier sections south of Jenner. (These areas are just north and south of the map area shown.) But along the rocky substrate in the middle, the bull kelp held strong.


Figure 67: Then came “the blob.” In 2013, a marine heatwave started warming the Bering Sea, and by 2014 the warm waters reached the California coast. Water temperatures rose as much as 2.5ºC (4ºF) above normal off the U.S. and Canadian coast and stayed high for 226 days—the longest marine heatwave ever recorded. (Sea surface temperatures from July 2015 are shown below.) “The blob” eventually merged with warm waters from the “Godzilla El Niño” of 2015–2016 (image credit: NASA Earth Observatory)

- The nutrient-poor waters associated with marine heatwaves hinder kelp growth, leading to smaller canopies. Historically kelp have been resilient, though, coming back in force once waters have cooled down. But this time, a cascading series of environmental and biological events—exacerbated by climate change—combined to decimate the forests.

- The delicate interplay of species that safeguards kelp forest biodiversity was shifted in 2013 when more than 20 sea star species from Alaska to Mexico started wasting away. In particular, sunflower sea stars, the primary predator of kelp-devouring purple sea urchins, were ravaged by a mysterious wasting syndrome. Renowned regenerators known to grow back entire limbs, the sea stars (starfish) looked as if they had melted to goo.


Figure 68: With this pivotal predator functionally extinct, and bull kelp growing poorly due to the warm water, the balance of predators and feeders was thrown off. Purple sea urchins that had previously occupied shallow tidal pools and ate kelp leaf litter were suddenly eating growing kelp stalks, or stipes. Urchins climbed down the stipes all the way to the seafloor, eating until there was nothing left (image credit: NASA Earth Observatory)

- By 2015, the kelp forests were mostly gone, replaced by urchin barrens. Divers described the conversion of once-rich kelp forests into spiky purple carpets. With no kelp left to eat, the purple sea urchins now mostly subsist in a starvation state, rousing occasionally to eat any nascent kelp that tries to establish itself. These zombie urchins are effectively killing the chances of kelp recovery.

- The loss of bull kelp forests has meant the loss of the ecosystem services they rendered. California’s recreational abalone fishery—the world’s largest, with over 35,000 fishers—was closed in 2018 after more than 80 percent of these edible sea snails died for lack of kelp sustenance. Kelp harvesting and recreational diving have been clobbered, too. The ecosystem also lost capacity to sequester carbon—kelp are 20 times more efficient than their terrestrial counterparts—and to temper the destructive power of waves.

- Restoring the kelp forests is a priority for marine managers, but it is a massive challenge. The purple urchins are of little nutritional interest to most predators or fishermen in their diminished state, yet they have still been observed spawning. A group of citizen scientists known as Reef Check has taken to diving to remove the urchins manually in an effort to create small urchin-free oases where new kelp can grow. In 2020, they scooped, hauled, and composted 20,000 pounds of urchins. Some innovative conservationists also have been removing emaciated urchins to onshore tanks to fatten them up for humans to eat.

- The dire kelp situation is an expression of catastrophic tipping points and ecosystem shifts that climate change can bring. The collapse of Northern California’s kelp forests was quick and nearly total. Meanwhile, marine heatwaves are increasing in intensity and frequency, making the long-term recovery of kelp forests uncertain.

- Yet there are some hopeful signs. Closer to Alaska, sunflower sea stars are starting to recover. Near Monterey Bay, urchin-eating sea otters have been able to protect local kelp forests. And in spring 2021, Reef Check reported new bull kelp growing at one of the surviving patches off the Mendocino coast.

- Freely availability satellite data can provide insights about the environmental drivers influencing kelp productivity, potentially helping managers time their restoration efforts for years when conditions will best support kelp growth, McPherson explained. “Landsat has allowed managers to observe regional trends in kelp canopy area and biomass across more than 30 years,” she said. “This is very valuable.”

• June 2, 2021: Ghana is one of the leading producers of gold in Africa and the seventh leading producer in the world. Large commercial companies mine the majority of it using heavy machinery. But about 35 percent is extracted through small-scale mines, many of which operate informally or without a valid license. 53)

- This unregulated small-scale and artisanal gold mining is known locally as galamsey, a slang word derived from the Ghanaian words “gather” and “sell.” About one million Ghanaians engage in the practice, supporting about 4.5 million people in the country. Many of the galamseyers live in poverty, and their activities often come at a cost to both human health and the environment.

- Although individual galamsey sites cover less area than an industrial mine, their cumulative effect on the landscape outweighs those of larger mines. In the southwestern forests of Ghana, for instance, the footprint of small-scale mines is nearly seven times greater than that of industrial mines. The mercury and heavy metals used in galamsey can contaminate drinking water for entire communities. It also causes major health issues, such as kidney problems and neurological disorders, to those continually exposed to the metals.


Figure 69: The Ghanaian government has been increasing law enforcement in recent years related to galamsey activities, but locating the small gold mines is tricky. Many are tucked away in densely forested areas, and some only span a few acres. Unlike larger sites, these mines are usually operated by a few people and sometimes with handheld tools. Unless you knew it was there, the odds of bumping into an artisanal mine are small. Ground photo by Ruth McDowall.


Figure 70: Researchers are using satellite data to locate small mines that can cause long-term damage to forest communities and human health [image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey, protected area data from the World Database on Protected Areas (WDPA), and mining data from Center for Remote Sensing and Geographic Information Services (CERSGIS) and SERVIR West Africa. Story by Kasha Patel]

- “Local authorities may have knowledge about a specific area, but if the mines are scattered all over the place, then they are difficult to find,” said Foster Mensah, executive director at the Center for Remote Sensing and Geographic Information Services (CERSGIS) in Ghana. “The maps and products we can generate through satellite imagery help them see areas that need attention and intervention.”

- Mensah and colleagues at CERSGIS have been working with the SERVIR-West Africa program to identify and quantify mining activities in highly forested areas, which are mostly located in southern Ghana. SERVIR-West Africa is a program between NASA and the U.S. Agency for International Development (USAID) that uses remote sensing to provide support for protection of food and water resources and sustainable development.

- Mensah’s team uses radar data from the European Space Agency’s Sentinel satellites, which can penetrate clouds to see the ground activities below. The team also uses Landsat data to decipher long-term changes in forest coverage and degradation. The visualization at the top of this page shows mining activities from 2015-2020 in southern Ghana. As of 2018, galamsey had led to about 29,000 hectares (72,000 acres) of deforestation, with 1,000 hectares (2,600 acres) occurring in protected areas of the country.

- “It can be hard to distinguish between illegal mining and legal mining in an area,” said Mary Amponsah, also a researcher with CERSGIS and SERVIR-West Africa. “When you look at the maps, most illegal activity sits close to legal mining concessions.”

- When the government grants legal mining concessions to large companies, Amponsah noted, galamseyers explore the surrounding areas for other places to mine. However, they may not have a license or they may be mining in unauthorized or protected areas. Some license holders also mine more area than allowed.

- Together with the non-governmental organization A Rocha Ghana, the CERSGIS and SERVIR teams have met with community leaders and showed how galmasey is affecting the landscape and resources. They demonstrated a mobile app that allows anyone to report illegal mining that they see. The satellite data and the crowdsourced information are stored on a web-based portal that the public can access.

- The team has also been working with Ghana’s Environmental Protection Agency and its Forestry Commission to highlight areas where mining is affecting forest coverage and degradation. For closed or abandoned mines, the team is also using the satellite data to help inform reclamation projects. Knowing the location and extent of degraded forests can help land managers better project the labor and expense to reclaim an area (by planting tree seedlings or adding plants that could detoxify the area, for instance).

- “It boils down to providing authorities information and data they did not have before, especially over a wide area,” said Mensah. “The satellite data is cost effective and gives them a head start on how to pinpoint mining hot spots that need immediate attention.”


Figure 71: This image shows gold mining encroachment in the Upper Wassaw Forest Reserve, a habitat for the green-tailed bristlebill and Tai Forest treefrog, which are classified as species of conservation concern. The image was captured on April 30, 2020, by the Operational Land Imager (OLI) on Landsat 8. Out of 28 protected areas in southwestern Ghana, Upper Wassaw had the most mining. As of 2019, about 3.4 percent of the area had been converted for mining activities (image credit: NASA Earth Observatory)

• May 31, 2021: Most asteroids that survive an encounter with Earth’s atmosphere ultimately plummet into water, simply because oceans cover 70 percent of the planet. But massive space rocks occasionally hit land. That was the case 50,000 years ago when an iron asteroid smashed into North America and left a gaping hole in what is today northern Arizona. 54)


Figure 72: Meteor Crater (also called Barringer Meteor Crater) is located between Flagstaff and Winslow on the Colorado Plateau. The Operational Land Imager (OLI) on Landsat-8 acquired this image of the area on May 16, 2021 (image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- Even at 50,000 years old, the crater is relatively young and remarkably well-preserved compared to other craters. Because of this, scientists have studied the site extensively to learn about cratering processes—how they work on Earth and elsewhere in the solar system—and about the modern hazards posed by asteroid impacts.

- “A similar-size impact event today could destroy a city the size of Kansas City,” said David Kring, an impact cratering expert at the Lunar and Planetary Institute. Meteor Crater measures 0.75 miles (1.2 km) across and about 600 feet (180 m) deep. The size of the asteroid that produced the impact is uncertain—likely in the range of 100 to 170 feet (30 to 50 meters) across—but it had to be large enough to excavate 175 million metric tons of rock.

- The wide perspective pictured above gives a sense of the crater in context with the surrounding area. This part of the Colorado Plateau drains from Anderson Mesa (lower left) and across a surface that dips toward the Little Colorado River near Winslow. The red blotchy areas near the crater are Moenkopi red siltstone amid light-brown Kaibab limestone. Volcanic landforms dot part of the wider landscape, including Anderson Mesa and the West and East Sunset Mountains.

- Note how the crater’s rim and areas just outside it are much lighter tan. This is the debris that was ejected from the crater, consisting primarily of Kaibab limestone and Coconino sandstone. Also notice how the crater is not exactly circular, exhibiting almost a square shape. According to Kring, this is because pre-existing flaws in the rock caused it to peel back farther in four directions upon impact. These cracks, oriented northwest-southeast and northeast-southwest, formed when the Colorado Plateau was uplifted from below sea level to its current mile-high elevation.

- The landscape has not always looked like this. When the asteroid hit, humans had not yet reached North America. The terrain of forested rolling hills was likely inhabited by mammoths, mastodons, and giant ground sloths. Now the crater stands amid shrub-covered desert.

- Kring continues to host a NASA-sponsored field training and research program at Meteor Crater, in which graduate students train to study impact craters on Earth, the Moon, Mars, and other planets. He also trains astronauts “so they are familiar with impact-cratered planetary surfaces,“ Kring said. “NASA’s Artemis astronauts will, for example, be landing in an impact-cratered terrain around the lunar south pole.”

• May 29, 2021: Strong northwesterly winds routinely blow down the eastern side of the Andes Mountains and whip across the central Patagonia Desert. In the process, they lift abundant dust from Argentina’s Lake Colhué Huapi, making it the largest and most active source of dust storms in the region. 55)


Figure 73: The OLI instrument on Landsat-8 acquired this natural-color image of dust streaming from the silty lakebed on May 24, 2021. Colhué Huapi is a particularly abundant source of dust because the shallow lake regularly grows and shrinks in sync with variations in the flow of the Senguer River and the pace of evaporation. When lake water levels are low, as they were when Landsat-8 captured this image, fine-grained, light particles are easily transported by the wind (image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- While dust storms are common here, scientists are only beginning to track them rigorously and research the role they play in the regional environment. Colhué Huapi dust likely affects the region in several ways, explained NASA remote sensing scientist Santiago Gassó. Field research and ice cores suggest that winds may transport Colhué Huapi dust as far as East Antarctica, where it could have consequences for how quickly snow and ice melts. The dust storms also may be a significant fertilizer for the South Atlantic Ocean, providing key minerals that may trigger blooms of phytoplankton.

- To better understand the role of Colhué Huapi dust storms, Gassó recently analyzed satellite and surface weather data from the past five decades, assessing the year-to-year variability in dust storms and identifying periods of high activity. Dust storms peak during the summer (December through March), though wintertime events (May through August) are also common, Gassó found. Most years brought 15 to 30 moderate to large events. There has also been steady increase in the number of dusty days observed since the 1970s, according to his analysis.

- “Events like these are a reminder that dust activity is not just a warm weather phenomenon,” said Gassó. “It can happen in cold places, too. You just need loose soil, limited moisture, and winds.”

• May 28, 2021: Point Roberts, Washington, is like many small coastal towns in the Pacific Northwest, with access to epic places to fish, hike on the beach, and watch whales. But unlike other coastal towns, getting to Point Roberts is a bit more complicated. To drive there from mainland Washington, you must cross an international border twice. 56)


Figure 74: Point Roberts is what’s known as an exclave—part of a territory that is geographically separated from its main part by another territory. In this case, the 5-square-miles (13-square-kilometers) of U.S. territory that constitutes Point Roberts is separated from the rest of Washington by British Columbia, Canada. This geopolitical curiosity is the focus of these images acquired on 29 July 2020, by the Operational Land Imager (OLI) on Landsat-8 (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- The wide view shows Point Roberts dangling below the 49th parallel—the line of latitude that was established in 1846 as the political boundary between the northwestern United States and Canada. Point Roberts is further isolated by the Strait of Georgia to the west and south and Boundary Bay to the east. The image of Figure 75 shows a detailed view of Tsawwassen peninsula and Point Roberts.


Figure 75: The tiny town is a geopolitical curiosity, located closer to Canada than it is to mainland Washington (image credit: NASA Earth Observatory)

- Political boundaries can sometimes affect built landscapes in ways that are visible from space. In this case, the green expanse of the Bald Eagle Golf Club abruptly ends south of the border; just north of the border in Canada, the geometric patterns of the suburban community of Tsawwassen take shape.

- Point Roberts is a popular vacation spot for Canadians, which helps drive the town’s economy. Since the closing of the U.S.-Canada border in mid-March 2020 due to COVID-19, news reports have likened it to a ghost town in the making. The closure put a halt to routine border crossings to mainland Washington—just 25 miles away—temporarily ending previously routine trips by the town’s residents to schools and medical care.

- The ferry terminal visible in these images connects two Canadian points—the town of Tsawwassen and Vancouver Island. For now, emergency ferry service is available as necessary from the Point Roberts marina to Bellingham, Washington. At the time of this story, the border was expected to remain closed until at least June 21, 2021.

- Some natural features are unaffected by political borders. Notice, for example, the striking plume streaming from the mouth of the Fraser, the longest river in British Columbia. The river carries about 20 million tons of silt each year, much of it into the Strait of Georgia. Moved around by winds, currents, and tides, the silt provides nutrients that fertilize the region’s waters and support its salmon populations, which in turn make Point Roberts a great place to view the local pods of orcas.

• May 21, 2021: Zombie fires, holdover fires, hibernating fires, or overwintering fires: Whatever you choose to call them, you’re probably going to hear a lot more about them in the coming years. New research shows that this type of wildfire—which can survive the snow and rain of winter to re-emerge in spring—is becoming more common in high northern latitudes as the climate warms. 57)

- “Smoldering fires are flaming fires that have entered ‘energy-saver mode,’” said Rebecca Scholten of Vrije Universiteit Amsterdam. The fires start above ground, then continue to smolder in the soil or under tree roots through winter. “These fires are only just surviving based on the resources they have—oxygen and fuel—and can transition back into flaming fires once conditions are more favorable.”


Figure 76: These images, acquired with the Operational Land Imager (OLI) on Landsat-8, highlight the progression of a particularly potent overwintering fire in Alaska in 2015-16. The images are false color (OLI bands 7-6-2), which emphasizes hot spots and actively burning fires while distinguishing burned vegetation (brown) from unburned vegetation (green). The first image (top-left), acquired in September 2015, shows the burn scar from the Soda Creek Fire, which scorched nearly 17,000 acres in southwest Alaska near the Kuskokwim River. The fire was never completely extinguished before winter set in. In April 2016 (top-right), the fire continued to smolder in the soil under a layer of snow. - When the snow finally melted in late May (bottom-left), the additional heat and oxygen caused flames to re-emerge quickly spread. The June 2016 image (bottom-right) outlines new burned area from these overwintered fires, which added nearly 10,000 acres to the previously burned area (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and fire perimeter data from the Alaska Interagency Coordination Center (AICC). Story by Kathryn Hansen)


Figure 77: New research shows that this type of wildfire—which can survive the winter to re-emerge in spring—is becoming more common in high northern latitudes as the climate warms. This image shows a natural-color version of Figure 76 (bottom left), overlaid with the shortwave-infrared signature of active fire fronts (image credit: NASA Earth Observatory)

- The incident was not an isolated case. The study points to numerous fires that overwintered after Alaska’s large fire years of 2009 and 2015, although they can happen after other hot and active fire years, too.

- “Although our satellite record of these fires in itself is too short to look at long-term trends, we found that the number of fires that overwinter is strongly linked to summer temperatures and large fire seasons,” Scholten said. “And for these we do see a pronounced upward trend—hotter summers and more burned area—with continued climate warming.”

• May 19, 2021: This is the first section in a three-part story about the history and regional aspects of sea level rise. The other sections will be published later this month. 58)

- In August 1849, a farmer named George Thorp noticed some odd, grooved bones poking up from a pile of dirt unearthed by railroad workers building a new line through Charlotte, Vermont. The bone came from a large animal, but not something familiar like a horse or cow. Thorp boxed up the mysterious bone and some others he found in the pile and sent them by wagon to University of Vermont naturalist Zadock Thompson.


Figure 78: The unexpected discovery of a whale skeleton hundreds of miles from the sea and more than 200 feet above sea level in 1849 is a reminder of how much sea level can change (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey, topographic data from the Shuttle Radar Topography Mission (SRTM), bathymetric data from the General Bathymetric Chart of the Oceans (GEBCO), ice sheet records courtesy of Dyke, A. S. (2004), and sea extent data from the Vermont Agency of Natural Resources. Whale drawing courtesy of NOAA. Story by Adam Voiland)

- After scrutinizing the bones and consulting with leading American and European scientists, Thompson offered an answer: they were whale bones. Specifically, a beluga whale. “How do you get a whale in Vermont?” Thompson wondered. The bones were excavated from a landlocked central part of the state, about 200 feet (60 meters) above sea level and 200 miles (300 kilometers) from the ocean.

- It was a question that would occupy some of the greatest scientific minds of the day, recounts Jeff Howe, author of a book about the “Charlotte Whale.” Discovered at a time when little was understood about how or why Earth had ice ages, the whale bones eventually became a key piece of evidence that a huge sheet of glacial ice had once covered much of eastern Canada and New England. The bones also served as a hint of something that wasn’t initially obvious. It was not just higher sea levels put this part of Vermont underwater about 13,000 years ago; the land itself had sunk.

- The Laurentide ice sheet covered almost all of Canada and New England at the peak of the last glacial maximum. Like the ice sheets on Antarctica and Greenland today, much of the Laurentide ice sheet was at least one mile thick. Since Earth’s crust sits on a layer of flexible rock in the upper mantle, the immense weight of so much ice would have pushed the Earth’s surface down by hundreds of feet.


Figure 79: “One way to understand what ice sheets do to land masses is to think about what would happen if you put a bag of ice on an inflatable mattress floating in a pool,” explained Jet Propulsion Laboratory geophysicist Erik Ivins. “The mattress—the land—would sag. And the more ice you piled on, the more it would sag.” (image credit: NASA Earth Observatory)

- After the peak of the glacial maximum, as the climate warmed significantly, the height of the land and sea changed. “A great deal of ice was lost from the global ice sheets during that period—equivalent to about 130 feet (40 meters) of global mean sea level rise,” explained Ivins, who studies past and current sea level rise. As the ice sheet retreated north, ocean water and meltwater inundated the vast depression in the land surface that had been created by the weight of the ice— an area that included the St. Lawrence Valley, southern Quebec, eastern Ontario, and parts of New York. The Champlain Sea was formed.

- At its greatest extent, the sea likely covered an area as large as modern Lake Michigan. Its northern shores were flanked by cliffs of towering ice that dropped a steady supply of icebergs into the sea; its southern shores transitioned into marshy tundra and forests. Based on the diversity of fossils found in the fine-grained sediments below it, the Champlain Sea must have teemed with sea life ranging from barnacles and clams to seals and walruses—much like Hudson Bay today.

- Subtle shifts in Earth’s orbit called Milankovitch cycles have played a key role in triggering and ending ice ages for millions of years. By about 12,000 years ago, orbital conditions had grown less favorable to ice, pushing Earth into our current warmer, interglacial period known as the Holocene.

- “Despite continued melting of glacial ice during the Holocene, sea level rise could not keep up with a competing effect — the rising of the land,” said Ivins. After being pressed down and compressed by so much ice, land surfaces slowly bounced back after the icy weight was lifted. The process—known as glacial isostatic adjustment—occurs slowly because Earth’s crust “floats” on a layer of slow-flowing, partially molten rock called the asthenosphere.

- “Eastern Canada was rising about 5 to 8 times faster than sea level between 12,000 and 8,000 years ago. Within a few thousand years, this rising cut the young Champlain Sea off from the Atlantic Ocean, and it slowly began to disappear,” explained Ivins. As the land rose, the Champlain Sea turned first into a series of freshwater lakes. Over time, most of these lakes dried up, though one large relic persists to this day as Lake Champlain.

- The uplift of land due to glacial isostatic adjustment continues, though at a slowing rate. Most scientists think the land in New England will take several tens of thousands of years to rebound completely.

• May 13, 2021: The eruption at Fagradalsfjall volcano in southwestern Iceland has put on quite a show this year, lighting up the night sky and even appearing to influence the clouds above it. This natural-color satellite image shows the volcano by daytime, with a rare clear view of the eruption and the geologic features of the landscape. 59)


Figure 80: The landscape around the volcano in Iceland changes daily, as lava piles up and spreads across valley floors. The OLI instrument on Landsat-8 acquired this image around midday on May 9, 2021. Dark brown areas indicate where cooling lava has piled up and spread across valley floors. Notice the lava (red) actively pouring from one of the vent systems (image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- A volcano activity update from the Icelandic Met Office on May 12 noted that the vents associated with this eruption have spilled nearly 30 million cubic meters of lava since the start of the eruption in late March. Measurements on May 10 indicated that the lava discharge rate was increasing, reaching 13 cubic meters per second.

- According to a video by Reykjavík Newscast, the nearby town of Grindavík has voted to name the lava field Fagradalshraun: beautiful valley lava.

• May 12, 2021: With global temperatures rising and ice sheets melting, plenty of coastal cities face a growing risk of flooding due to sea level rise. Few places, however, face challenges like those in front of the Jakarta metropolitan area, a conglomeration of 32 million people on the Indonesian island of Java. 60)

- Since the city’s early days, flooding has been a problem because Jakarta is situated along several low-lying rivers that swell during the monsoon season. In recent decades, the flooding problems have grown even worse, driven partly by widespread pumping of groundwater that has caused the land to sink, or subside, at rapid rates. By some estimates, as much as 40 percent of the city now sits below sea level.

- With mean global sea levels rising by 3.3 mm per year, and amid signs that rainstorms are getting more intense as the atmosphere heats up, damaging floods have become commonplace. Since 1990, major floods have happened every few years in Jakarta, with tens of thousands of people often displaced. The monsoon in 2007 brought especially damaging floods, with more than 70 percent of the city submerged.

- Rapid urbanization, land use change, and population growth have exacerbated the problem. The Landsat images above show the evolution of the city over the past three decades. The widespread replacement of forests and other vegetation with impervious surfaces in inland areas along the Ciliwung and Cisadane rivers has reduced how much water the landscape can absorb, contributing to runoff and flash floods. With the population of the metropolitan area more than doubling between 1990 and 2020, more people have crowded into high-risk floodplains. Also, many river channels and canals have narrowed or become periodically clogged with sediment and trash, making them especially prone to overflowing.


Figure 81: Sinking land, rising seas, and rainfall-driven floods pose big problems for Indonesia’s largest city. This image of Jakarta was captured by the TM instrument of Landsat -5 on 9 July 1990 (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- Since the image of Figure 81 was captured in 1990, artificial land and new development has spread into the shallow waters of Jakarta Bay. According to one analysis of Landsat data, people have built at least 1185 hectares (5 square miles) of new land along the coast. Much of the land has been used for high-end residential developments and a golf course, explained Dhritiraj Sengupta, a remote sensing scientist at East China Normal University. Such developments come with risks because they sit at the front lines of Jakarta’s inevitable battle against sea level rise and storm surges, cautioned Sengupta.

- Artificial islands are often among the fastest types of land to subside because their sand and soils settle and become compacted over time. Satellites and ground-based sensors have recorded parts of North Jakarta subsiding by dozens of mm per year. On new artificial islands, that rate has soared as high as 80 mm per year, Sengupta said.

- Some of the new islands were built as part of Jakarta’s National Capital Integrated Coastal Development master plan—an effort to protect the city from flooding and to foster economic development. A key initiative was the construction of a giant seawall and 17 new artificial islands around Jakarta Bay. Though work on the project began in 2015, a range of environmental, economic, and technical concerns have slowed construction and reduced the scope.

- The plan to construct a huge seawall is still in place, but it may not be enough to preserve the status quo in Jakarta. With environmental pressures mounting, Indonesian politicians hope to move the seat of government from Jakarta to a new location on the island of Borneo.


Figure 82: This image of Jakarta was captured by the OLI instrument on Landsat-8 on 11 September 2019 (image credit: NASA Earth Observatory)

• May 10, 2021: Floods have long plagued Saint Petersburg, Russia’s canal-filled “Venice of the North.” Spread across 42 marshy islands of the Neva River Delta, the historical core of the city rises just 1 to 2 meters (3 to 7 feet) above sea level. 61)

- In 1703, construction had barely begun on Saint Petersburg’s first building—the star-shaped Peter and Paul Fortress—when floodwaters washed away construction materials at the site. Since then, more than 300 floods have hit the city, including three catastrophic events where water levels rose more than 3 meters and swamped thousands of buildings.

- The largest floods are typically triggered when cyclones in the Baltic Sea push water east into the Gulf of Finland and Neva Bay. The narrow, shallow gulf can set up powerful seiche waves that are especially dangerous if they coincide with high tides or seasonal floods on the Neva River.


Figure 83: Russia’s answer to this flood prone-geography is the Saint Petersburg Flood Prevention Facility—a colossal complex that includes 11 dams, 6 locks, 30 water purification stations, and 2 navigation channels. As seen in this image from the Operational Land Imager (OLI) on Landsat-8, the structure spans 25 km across the Gulf of Finland, from Lomonosov northward to Kotlin Island, and then east toward Gorskaya. A six-lane highway runs across the structure’s wide top (image credit:NASA Earth Observatory, images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)


Figure 84: A city full of cultural, historical, and architectural riches has gone to great lengths to protect itself from floods. Work began on the project in 1979, but it was not until 2011 that officials declared it operational. The $3.85 billion structure is designed to withstand storm surges of 5 meters. Most of the time the floodgates are left open to allow water and marine life to pass. However, the flow can be cut within 45 minutes if a flood is imminent, as has been done more than a dozen times in the past decade. Vulnerable areas in the historic core of the city—which is a UNESCO World Heritage Site—have not experienced damaging flooding since the dam opened (image credit: NASA Earth Observatory)

• May 07, 2021: When parallel rows of clouds lined up over southwestern Iceland on April 30, 2021, they appeared to be strikingly pretty, but relatively common, wave clouds. But this instance becomes more compelling when you consider what lies below the blanket of white. 62)

- Wave clouds are a visible component of waves in the atmosphere, which form for a variety of reasons. Sometimes they are caused by land topography, such as when an air mass is forced over an obstacle like a mountain ridge, an iceberg, an island, or a volcano. According to cloud researcher Bastiaan Van Diedenhoven of SRON Netherlands Institute for Space Research, this is a reasonable explanation for the cloud pattern visible on April 30. He pointed to similar patterns in images from 2020.


Figure 85: This natural-color image, acquired by the Operational Land Imager (OLI) on Landsat-8—shows the clouds as they appeared around midday on April 30, 2021. Waves in the atmosphere can form for a variety of reasons, from rugged topography to the collision of air masses (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- Some scientists think that the clouds might have been influenced by the eruption at Fagradalsfjall, a shield volcano on Iceland’s Reykjanes peninsula. If this is the case, the waves in the atmosphere were formed by the collision of different air masses, not by the topography.

- “The difference in density between air heated by the volcano—even if not explosive—and the surrounding environment is very likely responsible for creating turbulence through Kelvin waves that propagate downwind,” said Jean-Paul Vernier, a NASA atmospheric scientist.

- Though it is not erupting explosively, the volcanic system has spewed plenty of hot lava since the start of the eruption in late March 2021. Activity from one of the cones intensified in late April, with fountains of lava reaching hundreds of meters into the air.


Figure 86: Later that same night, OLI acquired a false-color image of the same area, showing the infrared light emissions from the Fagradalsfjall eruption (image credit: NASA Earth Observatory)

- Throstur Thorsteinsson, a scientist at the University of Iceland, also thinks the cloud waves were probably influenced by the eruption. He noted that in early May, the eruption displayed even greater activity and began to pulsate—starting and stopping in intervals of minutes. The plume from those fiery spasms produced its own unique pattern in the atmosphere.

• May 4, 2021: Unlike the sea ice that caps the Arctic Ocean—some of which can survive the summer—the ice on the northern Baltic Sea will completely melt away before summer starts. These images offer a late-season look at some icy features before they are wiped away by spring melting. 63)


Figure 87: Acquired on April 19, 2021, by the Operational Land Imager (OLI) on Landsat-8, these natural-color images show the northwestern side of Bothnian Bay. Located in the northernmost part of the Baltic Sea, the bay is bounded by Sweden (west) and Finland (east of this image), image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen, with image interpretation by Renée Mie Fredensborg Hansen/FMI; Eero Rinne/FMI; and Sinead Farrell/UMD

- The wide view shows plenty of ice still clinging to the coast of Sweden. This “land-fast ice” is anchored to the shore and does not drift. Farther out in the bay, drift ice moves freely with the winds or currents.


Figure 88: The detailed view shows a closer view of land-fast ice in the Luleå Municipality, along the coast of northern Sweden. Notice the ice’s rusty color in places, most notably near Måttsund: This is due to sediment-laden water that flooded the surface of the ice at some point. This can happen when the water level rises, but the ice—anchored to the land—cannot rise with it (image credit: NASA Earth Observatory)

- When this image was acquired on April 19, the fast ice was still mostly intact. By May 1, ice charts from the Finnish Meteorological Institute indicate that much of this fast ice was in an advanced state of disintegration, or “rotten.” This is typical for the bay’s fast ice, which usually starts to decline by mid-April and disappears completely by mid-May.

- Though seasonal, the presence of ice in Bothnian Bay each year is hugely important for the region’s wildlife. Seals, for example, use the icy habitat for giving birth to their pups. People also find utility in the seasonal ice, using it to easily access the bay’s archipelagos. Thousands of islands are clustered off the shores of Sweden and Finland; some are populated, others have seasonal fishing villages, and many are uninhabited. Some of the linear features on the ice close to shore are likely tracks made by people during these offshore excursions.

- Other patterns in the ice, especially those farther offshore, are caused by natural processes. The bright white spaghetti-like features on the ice just west of the island of Germandön (detailed image) are ridges—areas where ice floes have collided, causing broken pieces to pile up on the sea ice surface. Ridges can stand many meters high and become quite dense across the sea ice, making winter navigation for ships especially challenging and slow. Observations from a Finnish icebreaker from April 18-20 indicate that areas of sea ice east of this image were still heavily ridged.

- In a new research paper accepted for publication in The Cryosphere, researchers described how they could use satellite data to enhance the safety of navigation in ice-covered waters. The research, led by Renée Mie Fredensborg Hansen of the Finnish Meteorological Institute, used high-resolution topographic measurements from NASA’s Ice, Cloud, and land Elevation Satellite 2 (ICESat-2) to estimate the degree of ridging in Bothnian Bay sea ice.

- According to study co-author Sinead Farrell of the University of Maryland, the study makes a case for the rapid, near-real-time release of ICESat-2 data for similar uses in the Arctic and other ice-covered seas.

• May 1, 2021: With its rocky terrain, mountain caves, and beautiful beaches, Hingol National Park is one of the natural wonders of Pakistan. It is also has significant cultural importance. 64)


Figure 89: The images on this page show sections of Hingol National Park as observed on February 13, 2021, by the OLI instrument on Landsat-8. Hingol spans around 6,200 km2 (2,400 square miles) across three districts of the Balochistan Province: Lasbela, Awaran, and Gwadar (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel)

- The park is named for the Hingol River, which flows through this dry region year round and is the longest in Balochistan. Before emptying into the Arabian Sea, the Hingol flows into an estuary that supports threatened fish, birds, and crocodiles. It is part of the largest national park for the protection of endangered species in the country. The park is also home to wild Sindh Ibex, Balochistan Urial, and Chinkara Gazelle.

- Located approximately 200 km (120 miles) northwest of Karachi, Hingol National Park features several distinct ecosystems. In the north, it includes an arid subtropical forest, while dry, mountainous terrain covers the western portion. In the east, the park is renowned for a group of mud volcanoes that spew methane and mud instead of lava. Along the coast, Hingol includes caves, beaches, and a marine ecological zone that is home to dolphins, sea turtles, and mangroves. The water body in the image above is an ephemeral lake near Sapat Beach.


Figure 90: Detail image of the Hingol National Park observed by Landsat-8 on 13 February 2021 (image credit: NASA Earth Observatory)

- Many Hindus travel to the park to visit the Hinglaj Mata Mandar, a famous religious site located in a mountain cave on the banks of the Hingol River. On their pilgrimage, worshippers walk on rock outcrops and between steep cliffs, while also performing ritual bathing in the river.

- The park is also well known for its unique rock statues (just out of the image to the west). One formation, called the Princess of Hope, resembles a woman looking into the distance. The Balochistan Sphinx is a natural rock formation that looks like the Great Sphinx of Giza. A portion of the national Makran Coastal Highway runs through the park and provides drivers with a front row seat to the many of these rock formations and landscapes.

• April 29, 2021: Beginning on April 9, 2021, intermittent explosive eruptions from La Soufriére volcano have hurled plumes of ash and gas high into the air above the Caribbean island of Saint Vincent. Although winds have carried some ash plumes great distances, clouds of the tiny pulverized rock and glass shards have also rained down on the island and the Atlantic Ocean. 65)


Figure 91: Following several explosive eruptions on the Caribbean island of Saint Vincent, volcanic ash poses myriad hazards in the air and on the ground. The fallout has coated large parts of Saint Vincent. The images, acquired by the OLI on Landsat-8, show the northwestern part of the island before and after two weeks of powerful eruptions and ashfalls. The brown scar in the vegetation in the image on the left was caused by damage from gases leaked by the volcano before it erupted explosively (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- Volcanic ash is quite different than the soft, fluffy material you might find in a fireplace, and the sharp edges and other properties of volcanic particles make them especially problematic. Ash plumes pose a threat to aircraft because the particles can damage jet engines, propellers, and other aircraft systems in ways that can cause them to fail. Roughly ten times denser than snow, ash also can accumulate into heavy layers that can smother crops, collapse roofs, and taint water supplies. When soaked by rain, it can form slurries of muddy debris called lahars that rush down slopes and into valleys. Wet volcanic ash can even conduct electricity, meaning it can trigger short circuits and the failure of some electronic equipment.

- The layers of ash that fell on Saint Vincent in April 2021—along with several pyroclastic flows of hot debris rushing down La Soufriére’s slopes—have caused widespread destruction. Most island residents and tourists evacuated the most affected areas in time, but large numbers of buildings were flattened and farms and infrastructure have sustained extensive damage.


Figure 92: The data visualization above offers a view of the vertical distribution of ash in the atmosphere over the Atlantic Ocean about 300 km downwind from La Soufriére. The data were collected on April 12, 2021, by the Advanced Topographic Laser Altimeter System (ATLAS) on NASA’s ICESat-2. Note that much of the ash lingered at heights ranging from 4 to 10 km (image credit: NASA Earth Observatory, the ICESat-2 data are from the National Snow & Ice Center courtesy of Stephen Palm/NASA GSFC)

- The ATLAS instrument was designed to measure changes—on the scale of centimeters—in ice and land surfaces. In fact, volcanologists at the University of Buffalo are using ICESat-2 data to identify small bulges in volcanic domes that can precede explosive eruptions. They hope such observations might someday aid warnings about imminent eruptions.

- ATLAS can make observations of the atmosphere up to a height of 14 km. Though the ICESat-2 mission is focused on measurements of icy surfaces, it collects data relevant to atmospheric features like wildfire smoke, dust, clouds, blowing snow, and the height of the planetary boundary layer. Since real-time data showing the height of volcanic plumes is often scarce, data like this can serve as an important tool for atmospheric scientists developing ash dispersion models.

- A few other satellite sensors can also measure plume height, but having multiple sensors tracking an eruption increases the chances that one will make a measurement in near-real time, which is useful for aviation safety and air quality warnings. “One of the most important things about this type data is that it shows the vertical distribution of the plume,” said Stephen Palm, a research meteorologist based at NASA’s Goddard Space Flight Center. “That’s key to getting warnings to aircraft pilots.”

- “I don’t think the volcanology community is well aware of ICESat-2 atmospheric data,” said Michigan Tech volcanologist Simon Carn. “However, it certainly provides useful atmospheric observations, especially when ash is dense and at night.”

• April 28, 2021: In northwestern China’s Gansu province, at the northernmost extent of the Tibetan Plateau, the landscape offers layer upon layer of spectacularly colorful rocks. The formations have a compelling geological history that dates back tens of millions of years and involves a continental collision more than 2,000 kilometers away. 66)

- The image of Figure 93 shows Zhangye National Geological Park, which spans 322 km2 (124 square miles) of the prefecture of Zhangye. The widespread rusty color is sandstone, which was colored deep red during its formation by iron oxide. Other oxides imparted browns, yellows, and even greens to the various layers of rocks.

- It is a geologic marvel that the park’s colorful layers—deposited tens of millions of years ago during the Cretaceous Period—are visible at all. Folding and faulting processes have since lifted and deformed the rock, exposing layers that would otherwise have remained out of sight. Much of this crumpling and disruption of the stratigraphy is thought to have resulted from the “recent” collision of the Indian and Eurasian plates about 50 million years ago during the Cenozoic Era. Recent research suggests, however, that some of the deformation is even older.

- “The implications are that there was somewhat rugged, pre-existing topography prior to the India-Asia collision,” said Andrew Zuza, a scientist at the University of Nevada. “The Qinghai-Gansu province areas of the northern Tibetan Plateau may have already had some topography before development of the Tibetan Plateau.”

- Erosion from wind and water have continued to shape the rock, sculpting natural pillars, towers, and ravines. In July 2020 the site was designated as a UNESCO Global Geopark due to its geological significance.

- Colorful rocks are not confined within the park boundaries. The second image shows an area about 150 km (100 miles) northwest of the geological park. Like the park’s rusty rocks, these sandstones are from the Cretaceous and appear strikingly red. But even just a hundred miles away, different geological and erosional processes have played out. “That region just doesn’t have the same tilted colorful beds and small-scale rugged topography that the park area has.”


Figure 93: The landscape of the Quilian Mountains exhibits layer upon layer of spectacularly colorful rocks. OLI on Landsat-8 acquired these images on September 17, 2020. The reds and browns of exposed sandstones and other sedimentary rocks poke out from the range’s northern foothills, where the mountains meet a flat basin to the north known as the Hexi Corridor (image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)


Figure 94: Colorful rocks are not confined within the park boundaries. This image shows an area about 150 km (100 miles) northwest of the geological park. Like the park’s rusty rocks, these sandstones are from the Cretaceous and appear strikingly red. But even just a hundred miles away, different geological and erosional processes have played out. “That region just doesn’t have the same tilted colorful beds and small-scale rugged topography that the park area has.” (image credit: NASA Earth Observatory)

• April 20, 2021: Fertilizers used in farming contain high amounts of nutrients, such as phosphorous, to help crops grow. But these same nutrients can cause unwanted plant growth and potentially harm ecosystems miles away if agricultural runoff flows into nearby rivers, lakes, or coastal waters. 67)

- These effects represent one of the many ways that the different parts of the Earth system are connected. Waterways like rivers and streams are natural highways that connect areas hundreds to thousands of miles apart. They are also essential ecosystems for fish and other aquatic life, as well as sources of drinking water and recreational areas for people. Earth-observing satellites from NASA and its partners have a unique perspective from which to study the links between water and other parts of the Earth system – and are uniquely poised to help researchers address the consequences of those links, namely water quality.

- Most bodies of water contain microscopic, photosynthesizing organisms called cyanobacteria, which are harmless at normal levels. Individual cyanobacteria are tiny, visible only under a microscope. But certain conditions – lots of sunlight, stagnant water, and high amounts of nutrients like those in fertilizers – allow cyanobacteria populations to grow exponentially. The result is scummy green water that can be seen from space. These events, called harmful algal blooms, can lead to economic losses and poor water quality, and pose a health risk to humans and animals.

Figure 95: From space, satellites such as the NASA and USGS Landsat 8 can help scientists identify where an algal bloom has formed in lakes or rivers. It’s a complicated data analysis process, but one that researchers are automating so resource managers around the country can use the satellite data to identify potential problems (video credit: NASA's Goddard Space Flight Center)

Green Lakes and Satellites

- Cyanobacteria have a photosynthesizing pigment called chlorophyll-a, which gives algal blooms a green hue when seen from space. Satellites can measure the concentration of chlorophyll-a in a body of water, allowing scientists to estimate the amount of cyanobacteria in the water. Several Earth-observing satellites are used to monitor algal blooms from space: NASA’s Terra and Aqua satellites, the joint NASA/USGS Landsat satellites, and the European Space Agency’s Copernicus Sentinel-2 and Copernicus Sentinel-3 satellites. Which one is used often depends on the resolution of the satellite instrument and which satellite passes over the algal bloom at the right time to capture a cloud-free image.

- Harmful algal blooms are often hard to predict. NASA is part of a multi-agency Cyanobacteria Assessment Network (CyAN project), which includes the Environmental Protection Agency (EPA), National Oceanic and Atmospheric Administration (NOAA) and U.S. Geological Survey (USGS), to monitor harmful algal blooms and other water quality issues. In 2021, NASA will launch Landsat 9, gaining another Earth-observing satellite to help track algal blooms from space.

The Cost of Harmful Algal Blooms

- During an algal bloom, the water becomes covered with clumpy green scum that gives off a musty smell. Aquatic recreation like swimming and water sports are often suspended until the levels of cyanobacteria return to safe levels. Harmful algal blooms also lead to economic losses in a less obvious sector: healthcare.

- A NASA-funded study found that detecting harmful algal blooms early led to significant savings on healthcare, lost work hours and other economic losses totaling approximately $370,000. The study, published in the journal GeoHealth, focused on a 2017 algal bloom in Utah Lake. The team compared the economic losses from two scenarios: the real-world scenario in which satellites detected the bloom, and a hypothetical scenario in which the decision was based on human observers and on-site testing.

- Satellite data showed the beginnings of an algal bloom in time for Utah public health officials to put up warning posters by June 29, 2017 to alert visitors to use caution while boating, not to swim or water ski, and how to fish safely. In the hypothetical scenario, scientists calculated what would’ve happened if officials waited for human observers to report the bloom and confirm with on-site testing, then posted signs on July 6. The week-long delay would have cost $370,000 according to health economics models, showing how detecting harmful algal blooms early can result in significant savings on healthcare and other economic costs.

- Harmful algal blooms pose a health risk to fish and other wildlife as well as humans. During an algal bloom, cyanobacteria grow exponentially. That algae uses up oxygen in the water as it decomposes, which decreases the amount of oxygen dissolved in the water and can asphyxiate fish and other aquatic animals. The most severe cases lead to massive fish die offs. In 2016, some lagoons in Florida became obscured by the upturned white bellies of thousands of dead fish after an algal bloom.

- Instances like these are a reminder that ecosystems on Earth are interconnected, and actions in one part of the planet have downstream impacts on other ecosystems and humans.


Figure 96: Seen from space, large numbers of cyanobacteria look like swaths or patches of green in a body of water – like this algal bloom in Lake Erie captured by Landsat-8 on Sept. 26, 2017 (image credit: Joshua Stevens / NASA Earth Observatory using Landsat-8 data from the U.S. Geological Survey)

• April 14, 2021: Eruptions at La Soufrière volcano have propelled ash and gas high into the air over the Caribbean islands of Saint Vincent and Barbados. The eruption—the volcano’s first explosive event since 1979—prompted thousands of people to evacuate. 68)


Figure 97: Explosive activity has propelled ash and gas high into the air over the Caribbean islands of Saint Vincent and Barbados. The recent bout of explosive activity began on April 9, 2021. At about 10:30 a.m. local time that day, the Operational Land Imager (OLI) on Landsat-8 acquired this image of volcanic ash billowing from La Soufrière. The plume obscures the volcano below, a peak that stands 1178 meters (3,864 feet) above sea level on the northern side of Saint Vincent (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and MODIS data from NASA EOSDIS LANCE and GIBS/Worldview. Story by Kathryn Hansen)

- According to Jean-Paul Vernier, an atmospheric scientist with NASA’s Earth Applied Sciences Disasters Program, activity was apparent months before the explosive eruptions. It started with an effusive eruption in which magma that reached the surface slowly built up a lava dome. In April, the dome “finally turned out a massive explosion without many precursor signs,” Vernier said. Explosive eruptions result from the rapid expansion of pressurized gasses trapped in the rock or magma; the pressure violently breaks rocks apart and produces a plume of rock, ash, and gas.


Figure 98: Winds carried much of the ash and gas east from Saint Vincent. On the afternoon of April 10, 2021, the MODIS instrument on NASA’s Aqua satellite acquired this image showing ash reaching Barbados, 190 km (120 miles) away. Clouds (white) are also abundant in this view (image credit: NASA Earth Observatory)

- These images show ash aloft in the atmosphere, but some of it fell back to the ground. According to news reports, ashfall has blanketed parts of Saint Vincent and Barbados. It also has threatened food and water supplies on Saint Vincent and has reduced visibility, which can complicate evacuation efforts. People displaced to the island’s southern side—away from the volcano and generally safer—still had to contend with falling ash and poor air quality.

- Scientists are investigating the extent and height reached of the ash and gas plume. They think some ash has risen all the way into the stratosphere, where strong winds can potentially carry it great distances. Other satellite instruments have detected sulfur dioxide reaching Cape Verde, an archipelago in the central Atlantic Ocean. Sulfur dioxide near ground level can irritate the human nose and throat; higher in the atmosphere it can make sulfuric acid aerosols and, in extreme cases, lead to a cooling effect.

- The NASA Disasters team is working with several science institutions and agencies to continue assessing the eruption and to make data available to emergency responders and aid groups. “Our program has been working with stakeholders in the region since the first signs of the eruption,” Vernier said.

• April 3, 2021: Located along the southwest coast of South Korea, Sinan County attracts people from many walks of life. Its world-renowned tidal flats host unique marine life as well a thriving salt production industry. Meanwhile, purple-painted islands draw tourists from around the country. 69)

- Sinan County includes more than 1,000 islands, about a quarter of all islands in the country. The majority are surrounded by shallow tidal flats that are alternately covered or exposed by the rise and fall of tides. Depending on the time of the year, the flats can be muddier, sandier, or a combination of both. Finer mud tends to build in the zones during the summer, then erodes in the winter. Monsoons and strong waves in the winter create sandier flats.


Figure 99: From expansive tidal flats to purple-painted towns, southwestern South Korea features unique ecosystems for salt production, wildlife, and tourists. The images show portions of Sinan County, or Shinan-gun, on October 15, 2020. The images were acquired by the OLI instrument on the Landsat-8 satellite (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel)

- Reclaimed mudflats are also used for commercial salt production. The region’s fresh air, clean seawater, and abundant sunshine create prime conditions for making salt. Salt production begins by storing sea water in reservoirs and moving it to evaporation ponds (appearing as checkered fields) to naturally increase the water’s salinity with the help of the Sun and wind. On crystallization ponds, the saline water turns into salt crystals, which are stored in silos for two to three years to remove the bitter-tasting solution and improve the taste.

- Shinean sea salt contains low concentrations of sodium chloride, but relatively high amounts of moisture, calcium, potassium, magnesium, and sulfuric acid ions that help bring out the flavor in traditional Korean foods. Jeungdo Island (Figure 100), which contains the most extensive mudflat in South Korea, is home to the country’s largest sun-dried salt producer. The island also contains a salt museum and sea salt ice cream shop.


Figure 100: Detail 1. The tidal flats, or getbol in Korean, are highly productive ecosystems. The mineral-rich sediments are full of microorganisms that attract marine animals such as clams and mud octopuses. The flats serve also as an important stopover for many migratory birds (image credit: NASA Earth Observatory)


Figure 101: Detail 2. The image shows another unique aspect of the region: the brightly-colored Banwol and Bakji Islands. Nicknamed the "the purple islands," they are known for displays of purple paint on their buildings, roofs, phone booths, and bridge. There is even a restaurant that serves purple food. The purple complements the native bellflowers called campanula, which cover the landscape in lilac. The Korean government launched the purple initiative to improve tourism on the two islands, which collectively have a population of around 250 people. Since 2018, more than 490,000 people have visited Banwol and Bakji (image credit: NASA Earth Observatory)

• March 25, 2021: Persistent, heavy rain fell for several days in late summer in New South Wales, Australia, leading to the region’s worst flooding in six decades. The Australian Bureau of Meteorology reported that areas around Sydney and in the Hunter and Mid North Coast regions were drenched with 400 to 600 mm (16 to 24 inches) of rain across four days, with the most extreme totals approaching one meter. 70)

- Water levels rose to major flood levels along the Clarence, Gwydir, Mehi, Lower Hunter, Manning, and Colo rivers, among others. The Hawkesbury-Nepean River system around Sydney saw its highest crests since 1961. At least 40,000 people were evacuated and several died across New South Wales (NSW) state, while farmers suffered significant crop and livestock losses.

- Upstream from Sydney, the Warragamba Dam has been overflowing since March 20 and is expected to continue doing so for a week. The BBC reported: “Warragamba Dam discharged 500 gigalitres on Sydney—equivalent to the volume of Sydney Harbour.” The downstream Hawkesbury-Nepean valley has several choke points that cause river water to pile up and rise onto floodplains west of Sydney in what emergency management authorities refer to as a bathtub effect.


Figure 102: Persistent heavy rain raised rivers to levels not seen since 1961. On March 23, 2021, the Operational Land Imager (OLI) on Landsat-8 acquired a natural-color image of flooding in the Hawkesbury-Nepean River system along the western edge of Sydney [image credit: NASA Earth Observatory, images by Lauren Dauphin, using modified Copernicus Sentinel data (2021), processed by ESA and analyzed by the National Central University of Taiwan in collaboration with NASA-JPL and Caltech. Landsat data from the U.S. Geological Survey and topographic data from the Shuttle Radar Topography Mission (SRTM). Story by Michael Carlowicz]


Figure 103: The flood proxy maps (Figures 103 and 104) highlight areas of the Mid North Coast region that were likely to be flooded (indicated in blue) on March 20, 2021. The maps were derived from synthetic aperture radar (SAR) data acquired by the Copernicus Sentinel-1 satellites, operated by the European Space Agency (ESA). The maps were created by the National Central University of Taiwan in collaboration with the Advanced Rapid Imaging and Analysis (ARIA) team at the Jet Propulsion Laboratory and Caltech. The ARIA team is supported by NASA’s Earth Science Disasters Program (image credit: NASA Earth Observatory)


Figure 104: Radar signals can penetrate cloud cover, allowing researchers to observe landscapes that are obscured from other satellite sensors. The team created the maps by comparing radar observations collected before and after the rainfall. Specifically, the researchers look for changes in brightness: if a normally rough ground surface is replaced with a smooth water surface, the brightness of those pixels will change (image credit: NASA Earth Observatory)

- Many of the areas affected by floods in March 2021 were afflicted with extreme drought and wildfire in the summer of 2020. Burn-scarred landscapes often produce more runoff during extreme rain events because the heat from fires reduces the capacity of the soil to absorb and hold on to water. Furthermore, fire strips away plants and trees that could intercept raindrops before they reach the ground.

Figure 105: The animation shows rainfall rates and accumulation across eastern Australia from March 16-23, 2021. Those data are overlaid on shades of white and gray from NOAA infrared satellite observations of cloudiness. The rainfall data are remotely-sensed estimates that come from the Integrated Multi-Satellite Retrievals for GPM (IMERG), a product of the Global Precipitation Measurement (GPM) mission. Rainfall rates are marked in blue, while accumulation is represented in green. Due to averaging of the satellite data, local rainfall amounts may be significantly higher when measured from the ground (video credit: NASA)

- Preliminary estimates from NASA’s IMERG analysis indicate that more than 600 mm (24 inches) of rain fell just off the coast across the week, with accumulations in coastal areas exceeding 400 mm (16 inches). The region usually sees 1000 to 1500 mm (40 to 60 inches) of rainfall in a year.

- La Niña patterns in the tropical Pacific have brought more rain than usual to eastern and southeastern Australia this summer. That extra rain likely left the soils and waterways with less capacity for absorbing new rainfall in March.

• March 24, 2021: Few rivers carry as much sediment as the Huang He (Yellow River) in China. The name itself comes from the muddy color of the water—a consequence of the river’s upper and middle reaches flowing through a region in northwestern China with unusually fine and powdery soil called loess. 71)

- All of the silt in the water supercharges the river’s ability to build new land at its delta, the area where it dumps its sediment into the shallows of the Bo Hai Sea. In this pair of Landsat images, note how much the easternmost lobe of the delta changed shape between 1989 and 2020 as the river delivered new sediment to some parts of the delta and erosion ate away at older coastlines. (Read our Yellow River Delta World of Change story to see more imagery of the delta.)

- One of the most noticeable changes resulted from a diversion project that Chinese engineers completed in 1996, blocking the main channel and steering water and sediment to the northeast. The project’s purpose was to create new land in an area with offshore oil and gas to make the resource easier to extract. Before completion, new land formed along a rounded peninsula oriented to the southeast; afterward, the abandoned channel narrowed and new land began forming to the northeast, even as erosion ate away at parts of the older peninsula.

- Other features in this area have seen equally dramatic changes. Aquaculture and salt evaporation ponds—the green and blue rectangular features along the coasts—have proliferated. So has oil drilling infrastructure (small rectangular features) due to the rapid expansion of Shengli Oil Field, now China’s largest. Several smooth-edged sea walls and dykes have been built along the coast in an attempt to protect the new oil, aquaculture, and other infrastructure from encroaching tides.

- On the youngest land, different types of vegetation&mdash:notably the cordgrass Spartina alterniflora—have spread widely, creating dense new pockets of green in the 2020 image. The invasive cordgrass first reached the Yellow River Delta in the late-1980s, and began to spread rapidly in the intertidal zone in the early 2000s. While the grass does stabilize the shoreline, it has crowded out a local reed species (Phragmites australis) and an annual plant (Suaeda salsa), significantly reducing how much carbon the delta ecosystem stores and increasing methane emissions. By replacing S. salsa, the cordgrass has also made the area less habitable for certain rare birds, including red-crowned cranes and black-billed gulls.

- Like many deltas around the world, the Yellow River Delta faces growing pressure from the sea for several reasons. By 2020, many of the coastlines shown here had retreated inland by a few kilometers as the sea overwhelmed tidal mud flats and marshes. This is partly because the delta itself is sinking. Freshly deposited mud naturally settles and compresses over time.

- Human activity—particularly the pumping of groundwater for aquaculture—has accelerated the process. Though less influential, the process of pumping oil from below the surface and bringing in heavy equipment may have contributed to the subsidence as well. For much of the area shown in this image, scientists have reported subsidence rates of 20 millimeters (0.8 inches) per year. Layered onto both phenomena is global warming and sea level rise. Warming ocean water and the addition of fresh water to the oceans from melting ice sheets and glaciers is thought to contribute about 3 millimeters of sea level rise per year in this area.

- Finally, the Yellow River now carries only a tenth of the sediment that it did during the 1960s and about half of what it did in the 1980s. Several dams, erosion-control projects, and reforestation projects in upstream farming areas now trap much of the water and sediment that would otherwise reach the delta naturally.

- Efforts to flush sediment from clogged reservoirs and to scour sediment from the river bed led to a spurt of accelerated land formation in the delta between 2002-2014. However, the volume of sediment reaching the delta began dwindling in 2014 as coarser sediments coated and “armored” the river channel in key areas, preventing additional scouring. Since 2014, the delta has once again begun to lose more land each year than it gains.


Figure 106: Landsat-4 TM image of the Yellow River delta acquired on February 13, 1989 (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)


Figure 107: Landsat-8 OLI image of the Yellow River delta acquired on 24 October 2020. Changes in sediment load, vegetation, and the river’s course have brought stark changes to this dynamic river delta (image credit: NASA Earth Observatory)

• March 22, 2021: Come summer, Utahns will flock to the state’s lakes and reservoirs to boat, swim and picnic along the shore. And every week, if not every day, scientists like Kate Fickas of Utah State University in Logan will use satellite images and other data to monitor recreation sites to check for rapid growth of algae into a bloom, and make sure the water is safe for people and pets. 72)


Figure 108: Data from the Landsat-8 satellite can help resource managers identify potentially harmful algal blooms in water bodies like Utah Lake, seen here (image credit: NASA/USGS)

- From the vantage point of space, satellites, including the NASA and U.S. Geological Survey’s (USGS) Landsat 8, can help scientists identify lakes where a bloom has formed. It’s a complicated data analysis process, but one that researchers are automating to assist resource managers in identifying potential problems.

- “I grew up swimming in the Willamette River in Oregon, and diving in lakes over the summer,” said Fickas. “So it means a lot to me that I’m able to not only help develop algorithms for monitoring cyanobacteria blooms, which is interesting in itself, but to be able to take that next step and keep the public safe, and allow them to safely recreate and enjoy the water the way that I do.”

Figure 109: From space, satellites such as the NASA and USGS Landsat 8 can help scientists identify where an algal bloom has formed in lakes or rivers. It’s a complicated data analysis process, but one that researchers are automating so resource managers around the country can use the satellite data to identify potential problems (video credit: NASA's Goddard Space Flight Center)

- Blooms are made up of naturally occurring algae, phytoplankton, and cyanobacteria that explode in number under the right conditions: warm temperatures, lots of nutrients, and calm waters. Many water bodies in Utah meet those conditions, Fickas said, especially with warming temperatures due to climate change, as well as nutrient-rich runoff from agricultural fields and other sources.

- Satellites including Landsat-8 and ESA's (the European Space Agency) Sentinel-3 can detect when a lake changes color due to the mats of greenish organisms – allowing scientists like Fickas to tell water managers where to test to see if the waters are harmful or not. The two satellites have different strengths: Sentinel-3 collects data on individual lakes more frequently and measures wavelengths of light that are more indicative of phytoplankton, but Landsat-8 has a higher spatial resolution, so it can observe smaller lakes and identify specific problem areas within a larger lake.

- Landsat satellite-based detection of a 2017 bloom in Lake Utah helped save an estimated $370,000 in healthcare and related costs for the area, according to a 2020 study published in the journal GeoHealth. The case study builds on a larger multi-agency project to track algal blooms.

- When Landsat-8 measures a bloom, it detects chlorophyll-a, a green pigment found in phytoplankton, said Nima Pahlevan, a scientist at NASA’s Goddard Space Flight Center in Greenbelt, Maryland. Water is a tricky thing for Landsat satellites to measure since it’s dark compared to brightly reflective trees, buildings, and other landscapes. But Landsat-8 is more sensitive than its predecessors, able to distinguish between many more intensities of light, essentially detecting more shades of green.

- “Any increased level of chlorophyll-a over the norm could be alarming, and that’s what we’re looking for from the satellite data,” he said.

- Nima and his team have developed an algorithm to take the data collected by Landsat-8 over lakes, analyze it, and create a product that tells local water or recreation managers where that increase in chlorophyll-a might be. To get from the raw data to the usable product involves multiple steps, including accounting for atmospheric particles and gases that might otherwise skew the results.

- “Not everyone has access to the computing power to be able to process satellite images, or the time or expertise,” Pahlevan said. “Having these products readily available to the community will significantly increase the number of people who can use the satellite data products.”

- These Landsat aquatic reflectance products are still provisional, he stressed, but they are newly available from the USGS, which provides all Landsat data as well as other data products for free.

- While this data product could help decision-makers spot potential problem areas for boaters and swimmers, other Landsat data products measure things like forested areas, burned areas, and snow cover.

- “Data products convert the complex observations made by the instrument to the kind of information people need,” said Jeff Masek of NASA Goddard, project scientist for the upcoming Landsat 9 satellite. “They allow allow people who aren’t as familiar with remote sensing complexities to make use of the data.”

- Landsat-9, which is scheduled to launch in September 2021, has all the attributes of Landsat-8 that allow it to quantify chlorophyll-a, and will have added capabilities to distinguish between even more intensities of light reflecting from water bodies and other surfaces. Scientists are looking forward to future satellites as well. Landsat Next, the satellite following Landsat-9, could have additional capabilities that allow it to better detect the specific organisms that cause harmful blooms, and not just benign phytoplankton growth that doesn’t release any concerning toxins.

- “We’re seeing more water quality issues around the world,” Masek said, “which is why we’re so interested in the capability to monitor them.”

• March 17, 2021: Humans have inhabited Egypt’s Sinai Peninsula since prehistoric times. As a land bridge between Asia and Africa, the Sinai has provided a path to countless travelers, conquerors, and settlers over the centuries. The southwestern region still has traces of some of the peninsula’s earliest inhabitants, from fragments of an ancient alphabet to remnants of turquoise mines. 73)


Figure 110: The mountains in the southwestern Sinai Peninsula hold ancient relics of temples and turquoise mining. This detail image shows the southwest Sinai on March 11, 2021, as captured by the OLI instrument on Landsat-8. Mountains dominate the region, making it difficult terrain to traverse (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Caption by Kasha Patel)

- The Sinai Peninsula has a dry desert climate, yet is also one of the colder provinces of Egypt due to its topography and relatively high elevation. Not many animals live in the area, but species of ibex, gazelles, wildcats, jackals, and sand foxes have been spotted there. Shrubs grow on steep slopes in the south, while succulents and salt-tolerant plants are found on coastal plains. The mountains of the Sinai have long been a destination for human hermits and mystics. Today, people make a living on the peninsula through the petroleum industry, agriculture, mining, fishing, and tourism.


Figure 111: The wider image shows the intersection of the mountains and El Ramla, the largest sand desert in the southern part of the Sinai Peninsula (image credit: NASA Earth Observatory)

- Archeologists estimate the earliest inhabitants in the southwestern Sinai were miners who excavated copper and turquoise deposits around 3,500 B.C.E. Two popular mining locations were Serabit el-Khadim and Wadi Maghareh (also known as the “Valley of Caves”). In many cases, the miners were slaves captured by Egyptians in war. They mined turquoise by hollowing out portions of the mountains, and then transported the mineral to the Egyptian mainland. The turquoise was used for jewelry and color pigments. Ancient Egyptians called the Sinai Mafkat, meaning “Country of Turquoise.”

- Serabit el-Khadim is well-known today for its ancient ruins. Excavators have found scattered relics of a temple, including a red sandstone sphinx. Dedicated to the goddess Hathor, the temple is one of the few known monuments to a pharaoh in the Sinai.

- The temple ruins also contain inscriptions believed to be precursors to an alphabet. The scripts were hieroglyphic signs—symbols were used to represent sounds. For example, linguists determined an inscription on the sphinx read “mahbalt,” meaning “beloved of the Lady.” An ox-head character is thought to be a forerunner of the letter a in the Latin alphabet. The script may also have been used to write the names of miners and keep track of their labors. There are also multiple engravings near the temple, including drawings of ships carrying turquoise.

• March 9, 2021: Though it covers just 1 percent of Earth’s land surfaces, Indonesia’s rainforest is believed to shelter 10 percent of the world’s known plant species, 12 percent of mammal species, and 17 percent of bird species. Spread across 18,000 islands, it covers an area large enough to make it the world’s third-largest rainforest, trailing only those in the Amazon and Congo basins. 74)


Figure 112: The data used in this earlier image was acquired by the Thematic Mapper (TM) on Landsat 5 in 2002 (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey and forest loss data from the University of Maryland. Story by Adam Voiland)

- While satellite data indicate that Indonesia has had high rates of forest loss in recent decades, the situation seems to be changing. Deforestation declined significantly between 2017-2019, according to data from Global Forest Watch. The forest change data used in the analysis was collected by Landsat satellites and processed by a team from the University of Maryland.

- But even as deforestation slows on major Indonesian islands such as Sumatra and Kalimantan, there are signs of a shift to other areas. One of those areas is Papua (also called Western New Guinea). Papua’s rugged terrain and scarcity of transportation infrastructure has led to less development and economic growth than in other parts of Indonesia. But in some parts of the island, there has been noticeable new activity in the past decade.


Figure 113: While the region has seen less deforestation than other parts of Indonesia, large-scale clearing is still evident. The image shows forest clearing along the Digul River near Banamepe, an area that was cleared between 2011 and 2016.This image was acquired by the Operational Land Imager (OLI) on Landsat-8 in 2020 (image credit: NASA Earth Observatory)


Figure 114: This map, based on forest change data from the University of Maryland, shows part of southern Papua where lowland rainforest and swamp forest have been cleared to establish several large plantations. While large-scale deforestation has been happening in this area for about two decades, several particularly large plots were cleared in the past few years, including some near the river town Tanahmerah.

- The smaller, more scattered clearings along rivers are likely associated with selective logging, natural shifts in water courses, and small-scale clearing by subsistence farmers, explained remote sensing scientist David Gaveau, the author of a new study about deforestation trends in Papua. In the lower third of the map, an area where forests transition into the Trans-Fly savanna and grasslands, some of the changes are likely associated with seasonal fires.

- “The slowdown in Sumatra and Kalimantan is due, at least in part, to the exhaustion of available suitable land for plantation agriculture and increasing land prices on these islands,” explained Kemen Austin, an analyst with the non-profit research organization RTI International and the author of a 2019 study about the drivers of deforestation in Indonesia. “Papua is seen as the next frontier, and recent investments in infrastructure have made plantation agriculture in the region more economically compelling.”

- According to Gaveau’s analysis of two decades of Landsat data, nearly 750,000 hectares of forest were cleared in Papua between 2001-2019—about 2 percent of the island’s forests. Of that total, the analysis found that about 28 percent was cleared for industrial plantations (oil palm and pulpwood), 23 percent for shifting cultivation, 16 percent for selective logging, 11 percent for rivers and lakes expanding or changing course, 15 percent for urban expansion and roads, 5 percent for fires, and 2 percent for mining. (Shifting cultivation is a type of farming where fields are only used temporarily and then left to regrow naturally for a number of years before being cleared again.)

- Biological surveys have been rare on the relatively undeveloped New Guinea, so the island’s immense biodiversity remains only partly catalogued and understood. Since the island was once connected to Australia, it is home to unusual marsupials, such as tree kangaroos and forest wallabies. Among the island’s more notable animals are two species of egg-laying mammals (monotremes) called echidna.

• March 8, 2021: Since late January 2021, blue-green algae have spread across Lake Burrinjuck in New South Wales, Australia. Authorities issued warnings to stay out of the lake, which usually attracts many people for waterskiing and fishing around this time of the year. 75)

- Blue-green algae, also known as cyanobacteria, typically appear as greenish clumps or scum on the surface of the water and have a strong musty odor. They occur naturally in modest numbers but can reproduce quickly under favorable circumstances—namely sufficient sunlight, stagnant water, and high amounts of dissolved nutrients, such as fertilizer runoff.

- Blue-green algae blooms can be fatal for pets and can cause stomach problems, rashes, and even vomiting for humans, if ingested. They could also harm the fish population at the lake, which is known for its golden perch, Murray cod, rainbow trout, and more. When the algae die, they sink to the bottom of the lake, where they are decomposed by bacteria. If the concentrations of algae and bacteria are high enough, the process can deplete oxygen concentrations in the water, causing fish to suffocate.

- Based on algal samples, the state-owned water supplier and river operator WaterNSW issued alerts in late January and February to stay out of the water and to stop recreational activities in the lake. As of March 2, it reported lower concentrations of algae but still advised people not to drink untreated lake water and to exercise caution if partaking in water activities.

- Longtime local residents told The Canberra Times that the algal outbreaks were the worst they have seen in more than a decade. According to WaterNSW, the blooms were somewhat unusual since the lake is located in a cooler part of the state and the Burrinjuck Dam recently received a large inflow of water. A spokesperson for WaterNSW said, however, the inflows may have brought in nutrients from other catchments.


Figure 115: On February 10, 2021, the Operational Land Imager (OLI) on Landsat-8 captured imagery of algae blooms in Lake Burrinjuck and the Murrumbidgee River (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel)

• March 3, 2021: Antarctica’s Brunt Ice Shelf finally calved a large iceberg in February 2021, two years after rifts opened rapidly across the ice and raised concerns about the shelf’s stability. 76)


Figure 116: The break was first detected by GPS equipment on February 26, 2021, and then confirmed the next day with radar images from the European Space Agency’s Sentinel-1A satellite. On March 1, clouds were sparse enough for the Operational Land Imager (OLI) on Landsat-8 to acquire this natural-color image of the new iceberg [image credit: NASA Earth Observatory image by Joshua Stevens, using Landsat data from the U.S. Geological Survey and data © OpenStreetMap contributors via CC BY-SA 2.0. Story by Kathryn Hansen with information from Christopher Shuman (NASA GSFC/UMBC JCET)]

- Named A-74, the berg spans about 1270 km2 (490 square miles), or about twice the size of Chicago. That’s a large piece of ice for the Brunt Ice Shelf, but Antarctica is known for churning out some enormous bergs. For comparison, Iceberg A-68A was almost five times that size when it calved from the Larsen C Ice Shelf in 2017.

- A-74 broke from the ice shelf northeast of the McDonald Ice Rumples—an area where the flow of ice is impeded by an underwater formation that causes pressure waves, crevasses, and rifts to form at the surface. The rift that spawned the new berg appeared near the rumples in satellite images in September 2019, and it advanced across the ice shelf with remarkable speed during the austral summer of 2020-2021.

- “I would not have thought that this rift could go zipping across the northeast side of the Brunt Ice Shelf and cause a significant calving—all in a tiny fraction of the time it has taken Chasm 1 to extend toward the ice rumples from the south,” said Christopher Shuman, a University of Maryland, Baltimore County, glaciologist based at NASA’s Goddard Space Flight Center.

- Chasm 1 is a separate rift located south of the ice rumples and the Halloween Crack. After decades of growth and then a rapid acceleration in 2019, that rift appeared poised to spawn its own iceberg, prompting safety concerns for researchers “upstream” at the British Antarctic Survey’s Halley VI Research Station. This section of the shelf is still holding on, but when it eventually breaks the berg will likely measure about 1700 km2 (660 square miles).

- Scientists are waiting to see how the complex structure responds to the recent calving. “The Halloween Crack may or may not be the first to respond,” Shuman said. “We’ll be closely watching that pinning point for changes to the larger Brunt Ice Shelf remnant.”

- It also remains to be seen what will become of the new iceberg. Most likely, it will eventually get caught up in the Weddell Gyre—similar to the fate of A-68. But first it needs to be pushed offshore, and to date it does not appear to have moved very far.

• February 25, 2021: Sheer, glacier-covered ridges separated by gorges soar over the Chamoli district in northern India. On the morning of February 7, 2021, this spectacular terrain in Uttarakhand turned deadly when a torrent of rock, ice, sediment, and water surged through the Rishiganga River valley past multiple villages and slammed into two hydropower stations. 77)


Figure 117: On February 21, 2021, the Operational Land Imager (OLI) on Landsat-8 captured a view of the landscape in the wake of the event. In the image, natural-color Landsat-8 data were overlaid on a digital elevation model from the Shuttle Radar Topography Mission (SRTM) to depict the rugged topography [image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and topographic data from the Shuttle Radar Topography Mission (SRTM). Story by Adam Voiland, with information from Dan Shugar (University of Calgary) and Christopher Shuman (NASA GSFC/UMBC JCET)]

- The scale of the damage in the Himalayan district was devastating. Hundreds of people were swept away by the chaotic rush of water and debris. Dozens of people, many of them workers at the power plants, lost their lives; others ended up trapped in tunnels, prompting dramatic rescue attempts. Numerous homes, bridges, and roads were ruined.


Figure 118: The torrent of debris from a mountain in the Himalaya devastated remote valleys in Uttarakhand. The image pair above shows a closeup of the same area before and after the debris flow. Note the dark scar near the origin of the landslide and the trail of dust and debris that blanketed the valley walls downstream (image credit: NASA Earth Observatory)

- Initially, there was some confusion about what caused the catastrophe, but a group of remote sensing scientists have mined satellite data for clues to fill in the sequence of events.

- Months before the landslide, satellite images showed a crack opening on an ice-covered flank of Ronti, a 6,029-meter (19,780-foot) mountain peak. On February 7, 2021, a huge chunk of a steep slope broke off from the peak, bringing down part of a hanging glacier perched on the ridge. After freefalling for roughly two kilometers, the rock and ice shattered as it slammed into the ground, producing an enormous landslide and dust cloud. As the accelerating rock and ice raced through Ronti Gad and then Rishiganga River valley, it picked up glacial sediments and melted snow. All the materials mixed into a fast-moving slurry that overwhelmed the river and churned wildly as it rushed through the river valley.

- What triggered the rock and hanging glacier to fall in Uttarakhand remains an open question. University of Calgary geomorphologist Dan Shugar is among a group of scientists trying to find an answer to that and other questions about the disaster. As part of the effort, they are analyzing several types of meteorological, geologic, and modeling data to supplement and contextualize the satellite imagery. They hope to determine what role weather conditions, the tectonic environment, and shifting climate conditions might have played in priming the rock and ice for collapse.

- “Unfortunately, there were no weather stations that we know of that were nearby, but we are looking at things like whether cycles of ongoing freezing and thawing may have weakened the rock,” said Shugar. “Climate change may have even helped destabilize the rock face through increased water infiltration over a period of years and by thawing permafrost. For now, we can hypothesize about these possibilities, but careful work is required to understand exactly what happened.”

• February 23, 2021: Earth science satellites are generally used to observe certain features of the planet—landforms, atmospheric chemistry, ocean patterns. But at the same time, they periodically show us things that few people have seen or even looked for. 78)

- In February 2020, our team noticed a twitter message with a peculiar and beautiful image from Russia near 66 degrees north latitude. It turned into a scientific detective story and an unresolved case.


Figure 119: In this OLI image on Landsat-8, acquired on 15 September 2016, stripe patterns twist and turn around the hills of the northern Central Siberian Plateau. On steeper hills, the stripes form tight loops that spiral from the top of the hill to the bottom. As they descend toward the riverbanks, they start to fade. Eventually, the stripes disappear at lower elevations and at latitudes. There are several possible causes for the distinctive striping pattern, and the answers vary by the season and by the expertise of the researcher (image credit: NASA Earth Observatory, images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and topographic information from the ArcticDEM Project at the Polar Geospatial Center, University of Minnesota. Story by Andi Brinn Thomas, with Mike Carlowicz)


Figure 120: OLI image on Landsat-8, acquired on 29 October 2020. Researchers are puzzling over a distinctive striping pattern in the Central Siberian Plateau (image credit: NASA Earth Observatory)

- There are several possible causes for the distinctive striping pattern, and the answers vary by the season and by the expertise of the researcher.

- This portion of the Central Siberian Plateau lies within the Arctic Circle, where air temperatures remain below freezing for most of the year. Much of the landscape is covered in permafrost that can stretch tens to hundreds of meters below the surface. There are different levels of intensity, but this area generally has permafrost coverage for 90 percent of the year.

- The land does occasionally thaw, and cycles of freezing and thawing are known to create polygon, circle, and stripe patterns on the surface (referred to as “patterned ground”). In the case of the images above, the stripes could be elongated circles stretched out on the slopes by such thawing cycles. Yet studies have shown that this type of striping usually occurs at a much smaller scale and tends to be oriented downslope.


Figure 121: OLI images on Landsat-8, acquired in the period 15 September 2016 and 29 October 2020 (image credit: NASA Earth Observatory)

- To geomorphologists, the nature of the soil offers another explanation for the stripes. In regions this cold, soils can turn into Gelisols—soils with permafrost in their top two meters and often with darker and lighter layers distinguished by more organic matter or more mineral and sediment content. As the ground freezes and thaws, the layers break up and mix vertically in a process called cryoturbation. The persistent freezing and thawing action through the seasons can cause layers to align in a striping pattern. Different tundra vegetation—lichens, low shrubs, and moss—might grow preferentially on these Gelisol layers, accentuating the stripes we see from above. But this hypothesis has not been proven at large scales.


Figure 122: Several rivers cut across the plateau, including the Markha, and as the stripe pattern moves closer to the river, it starts to fade. This could be a result of sediment buildup along the riverbanks from millions of years of erosion (image credit: NASA Earth Observatory)

- From a geologist’s perspective, the different stripes appear similar to sedimentary rock layers. Thomas Crafford of the U.S. Geological Survey called the pattern “layer cake geology,” where sedimentary rock layers have been exposed and dissected by erosion. As snowmelt or rain travels downhill, pieces of sedimentary rock are chipped away and sent down to the ravines below. Such erosion could cause a step-like pattern that appears as stripes from space similar to a slice of layer cake. This pattern is also referred to as “cliff and bench topography.”

- In the winter Landsat image of Figure 120, snow causes the striping pattern to stand out more than in other seasons. The benches would be the lighter stripes (covered in snow) and the cliffs would be darker stripes. The Arctic digital elevation map above, based on data from the ArcticDEM Project, gives a clearer perspective on the possible cliff and bench features.

- “It looks like small canyons, maybe like the Badlands of South Dakota. The horizontal striping appears to be different layers of sedimentary rock,” said Walt Meier, an ice specialist at the U.S. National Snow and Ice Data Center. “The shape of the erosion pattern looks a bit different than standard sedimentary erosion, but my guess is that is due to the permafrost. The rivers are eroding through frozen ground. There could also be some effect from frost heaves affecting the topography.”

- Louise Farquharson, an Arctic geologist at the University of Alaska-Fairbanks, pointed to a region in northern Alaska with a very similar stripe pattern that could be formed by a similar process.

• February 15, 2021: For much of the year, an efflorescent salt crust makes Lake Lefroy stand out as a bright, white spot in satellite images. But after heavy rains, the ephemeral lake in Western Australia takes on a different look. 79)


Figure 123: When the OLI instrument on Landsat-8 acquired this natural-color image on February 9, 2021, water had pooled in the playa’s lowest points. The rain fell as part of a tropical low that soaked the Eastern Goldfields region in early February. The water was discolored by some combination of suspended sediments from the region’s red soils, light reflecting off the rust-colored lake bed, or bacterial activity in the salty water (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Adam Voiland)

- The smaller pools of green water in the center of the lake are areas where a mine discharges groundwater, a process called mine dewatering. The mine, built along a causeway that bisects the lake, taps into rich deposits of gold and nickel. Mining pits, roads, tailing ponds, and other mining infrastructure are visible along the causeway.


Figure 124: Spectacular patterns emerged as stormwater pooled on the salt crust of this ephemeral lake in Western Australia (image credit: NASA Earth Observatory)

- Large volumes of water do not persist for long in Lake Lefroy because the region’s hot, dry climate encourages evaporation. While water pooled in early February in a pattern that resembles a tropical fish, it’s unlikely the pattern will last. Lake Lefroy is frequently reshaped by changes in the prevailing winds that transport water back and forth between different parts of the playa. Nor are fish often found in these waters. Aside from certain flies, small crustaceans, phytoplankton, and algae, not much thrives in the hypersaline and impermanent waters.

• February 12, 2021: Beneath Earth’s crust lies 2,900 km (1,800 miles) of viscous mineral and rock known as the mantle. Famous and fanciful literature aside, no human is likely to visit the mantle or deep interior of Earth. But at Gros Morne National Park, people can step on fragments of the mantle without having to dig an inch. 80)

- Gros Morne provides some of the world’s best exhibits of the process of plate tectonics. The park contains a portion of the Long Range Mountains, a subrange of the Canadian Appalachians that dates back to around 1.2 billion years ago, when present-day North America collided with another continent. Those mountains have since eroded and left behind the gneiss and granite peaks of the Long Range. The park contains some of the tallest peaks of the Long Range mountains, including Big Level and Gros Morne Mountain (French for “great somber”).

- The Tablelands, located on the south end of the park, are considered one of its most striking features. The flat-topped, rust-colored land is rich with peridotite rock from the upper part of Earth’s mantle. The rock was thrust towards the surface around 500 million years ago through a process known as subduction. When two plates on Earth’s crust collide, one is often pushed back (subducted) toward the mantle. Standing out amid the lush green park, the yellowish-red Tablelands played a crucial role in confirming the theory of plate tectonics.

- The Canadian Space Agency has also studied the area to aid in the search for life beyond Earth. Scientists study how microscopic life forms can survive in the iron-rich Tablelands to better understand how they might survive on the extreme environment on Mars.

- Gros Morne National Park also features some recent geologic history at the Western Brook Pond. The freshwater fjord was carved by advancing glaciers tens of thousands of years ago during the most recent ice age. After the glaciers melted and receded, the land rebounded and cut off the outlet from the sea. Saltwater was slowly and naturally flushed from the 30 km (20-mile) long pond. Today, the fjord is surrounded by steep rock walls up to 600 meters (2,000 feet) high and contains nearly pure fresh water. The setting is a favorite for photographers.

- Today, the park is protected by the Canada National Parks Act. One of the biggest natural threats to the park is a large moose population, which is five to 20 times higher here than in other parts of Canada. Introduced into the area about 100 years ago, the hungry population has eaten through large portions of the boreal forest and hindered regrowth.


Figure 125: A geologist’s dream, Gros Morne National Park is one of the few places where you can set foot on the Earth's mantle without digging an inch. On October 3, 2017, the OLI instrument on Landsat-8 acquired natural-color imagery of Gros Morne National Park. The UNESCO World Heritage site covers 1,800 km2 (690 square miles) in the Great Northern Peninsula of western Newfoundland (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel)


Figure 126: A detailed view of the Tablelands, in the southern portion of the Gros Morne National Park (image credit: NASA Earth Observatory)

• February 9, 2021: Snow is not as rare as you might think in the Hawaiian Islands. But it never stops being beautiful. 81)

- Starting with a moderate storm on January 18, 2021, snow has fallen three times on the highlands of Hawai'i in the past three weeks. The snow cover has persisted on Mauna Kea and Mauna Loa—the two tallest volcanoes in the island chain—since January 25. Some snow also briefly crowned Haleakalā volcano (elevation 10,000 feet/3000 meters) on the island of Maui.


Figure 127: Three storms in three weeks have left abundant snow atop Hawaii’s tallest volcanic mountains. On February 6, 2021, the OLI instrument on Landsat-8 acquired natural-color images of the “Big Island” of Hawai'i with abundant snow on its two tallest peaks. Nearly every year, Mauna Kea and Mauna Loa (elevation above 13,600 feet/4200 meters) receive at least a dusting that lasts a few days. Sometimes, like this year, it is more like a winter blanket of snow (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey and data from the National Snow and Ice Data Center. Story by Michael Carlowicz)


Figure 128: The bar chart below shows the Normalized Difference Snow Index (NDSI) for Hawai'i as observed by NASA’s Terra satellite. NDSI incorporates a blend of visible light and shortwave infrared to assess the amount of snow within a given geographic area. The chart shows the combined NDSI for Mauna Loa (teal) and Mauna Kea (blue) for the first week of February in each year from 2001 to 2021. The combined weekly NDSI in 2021 for the two volcanoes is the highest since 2014 and second-highest in the record (image credit: NASA Earth Observatory)

- According to news and social media accounts, Hawaiians have found their way up the volcanic mountains with snowboards and boogie boards to sled through the fluffy white blanket. Others have filled their pickup truck beds to bring snow down to friends. Hawaiian weather blogger Weatherboy posted several photos from the scene.

- Snowfall in Hawai'i is often associated with a weather phenomenon referred to as a Kona low. Winds that typically blow out of the northeast shift and blow from the southwest. The winds from the leeward or “Kona” side draw moisture from the tropical Pacific, turning it from rain to snow as the air rises up into the high elevations.

- With the recent snowfall in Hawai'i, Florida is now the only state that has not yet seen snow this winter, according to The Weather Channel.

• February 3, 2021: In late January 2021, Tropical Cyclone Eloise caused widespread damage and heavy flooding in central Mozambique. The storm displaced more than 16,000 people, damaged around 17,000 houses, and killed more than a dozen people across a few countries in southeast Africa. 82)


Figure 129: These images show flooding on January 30, 2021, seven days after Eloise made landfall near the coastal city of Beira. The images from December 2019 are provided to compare the area under non-flooded conditions in the same season. The false-color images, acquired by the Operational Land Imager (OLI) on Landsat-8, use a combination of visible and infrared light (bands 7-5-3) to help differentiate flood water (dark blue), bare land (brown), and vegetation (bright green), image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel.

- After crossing northern Madagascar and before making landfall on mainland Africa, Eloise slightly strengthened due to warm waters in the Mozambique Channel. Stations in Beira recorded 25 cm (10 inches) of rain in 24 hours. Several rivers burst their banks, and roads became impassable. Tens of thousands of hectares of farmland were submerged in brown water, which could affect harvest this April. The storm, which brought winds up to 160 kilometers (100 miles) per hour, also blew over trees, power lines, and signs.

- Most of the areas hit by Eloise are still recovering from cyclones Idai and Kenneth in 2019, which claimed hundreds of lives. When Eloise hit, some villages were already flooded. In Dec. 2020, Beira and other surrounding areas endured heavy rains and flooding from severe weather.


Figure 130: Landsat-8 image of Mozambique on 27 December 2019 (image credit: NASA Earth Observatory)


Figure 131: Landsat-8 image of Mozambique on 30 January 2021 (image credit: NASA Earth Observatory)

- After making landfall in Mozambique, Eloise continued across southern Africa, though in a weakened state. The storm caused damage and flooding to South Africa, Eswatini, and Zimbabwe.

• January 30, 2021: Gold has been found on every continent except Antarctica, but the lustrous yellow metal is not exactly ubiquitous. The element (Au on the periodic table) is actually quite rare, accounting for just one out of every billion atoms in Earth’s crust. But in places such as the Central Aldan ore district in the Russian Far East—where concentrations of the precious metal have been discovered — mining operations are large enough to be seen from space. 83)


Figure 132: On September 11, 2019, the OLI instrument on Landsat-8 acquired this natural-color image showing part of the ore district in the Republic of Sakha (Yakutia). The image is centered about 25 kilometers (15 miles) northwest of the gold-mining town of Aldan, and about 450 kilometers southwest of the regional capital city, Yakutsk (image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- Central Aldan is one of Russia’s largest gold ore districts, with the mineral occurring in numerous deposits, or “lodes,” in the fractured rock. One of the largest lodes lies in the Kuranakh deposit, a shallow, ribbon-like orebody (up to 50 meters thick and 25 kilometers long) sandwiched between Cambrian limestone below and Jurassic sandstone above. Mining sites developed to to extract this gold are visible in the detailed images of Figures 133 and 134).


Figure 133: In places where concentrations of the precious metal have been discovered, mining operations are large enough to be seen from space (image credit: NASA Earth Observatory)

- The Kuranakh gold deposit was discovered in 1947, and a moderate amount of gold was extracted by 1955. Ten years later, large-scale open-pit mining began and continues today. Open-cut, drilling, and blasting techniques are now used to access the ore, which is processed at an onsite mill. In 2019, the Kuranakh mine produced 224,700 ounces of refined gold.

- Not all of the region’s gold shows up as lode deposits. In areas where a lode has been eroded, pieces of gold can become concentrated by rivers and streams into placer deposits.

- To excavate the placer, bucket-lined dredges scoop up material in the front and dump the tailings behind in curved piles. The accumulation of arc-shaped piles forms the long, maze like-pattern, which is visible in the image above. From April to December in the 2019 mining season, three dredges extracted 18,600 ounces of gold from the Bolshoy Kuranakh placer deposit.


Figure 134: This detail image, centered west of the town of Nizhny Kuranakh, shows the excavation site of buried placer along a tributary of the Aldan River (image credit: NASA Earth Observatory)

• January 20, 2021: Two years after the Brunt Ice Shelf seemed poised to produce a berg twice the size of New York City, the ice is still hanging on. But the calving of one, maybe two, large icebergs is inevitable. The question is: when? Ice scientists are watching to see if a rapidly accelerating crack will cause the shelf to rip apart before the sunlit summer season ends. 84)


Figure 135: The OLI instrument on Landsat-8 acquired this image of the Brunt Ice Shelf on January 12, 2021. The ice flows away from the Antarctic mainland and floats on the eastern Weddell Sea. The main shelf area has long been home to the British Antarctic Survey’s Halley Research Station, from which scientists study Earth, atmospheric, and space weather processes (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen)

- The breaking, or “calving,” of icebergs from ice shelves is part of a natural, cyclical process of growth and decay at the limits of Earth’s ice sheets. As glacial ice flows from land and spreads out over the sea, shelf areas farthest from shore grow thinner. These areas are stretched thin, and can be melted from above or below, making them more prone to forming rifts and eventually breaking away. The Brunt Ice Shelf appears to be in a period of instability, with cracks spreading across its surface.

- The major rifts are visible in the wide view of Figure 135. In late October 2016, the “Halloween crack” appeared and rapidly extended eastward. In early 2019, Chasm 1 extended northward as fast as 4 km per year. Now, a new crack is zippering across the shelf north of the Halloween crack, far faster than the fissure to its south.

- “It is impossible to know exactly what caused this new rift to extend so quickly,” said Christopher Shuman, a University of Maryland, Baltimore County, glaciologist based at NASA’s Goddard Space Flight Center. “It’s likely that fracture dynamics near the McDonald Ice Rumples played a role, as they did in the quick propagation of the ‘Halloween Crack’ in 2016. The unusual mix of ice blocks and mélange in this part of the Brunt Ice Shelf ‘system’ is another factor.”

- The rumples are the result of ice that flows over an underwater formation, where the bedrock rises high enough to reach into the underside of the floating ice shelf. This rocky formation impedes the flow of ice and causes pressure waves, crevasses, and rifts to form at the surface.

- All of these cracks, combined with a recent speed up at the leading edge of the ice shelf (detected by ESA’s Sentinel-1), point to an instability that is likely to spawn a new iceberg or two. The exact timing is uncertain, but until the break occurs and the shelf has been reformed, Halley Research Station is being kept minimally staffed for safety reasons. In 2016-2017, the Halley VI station was relocated to a safer location (Halley VIa) upstream of the then-growing Chasm 1.

- “I think we are going to see big changes here,” Shuman said. And with more than two months left of sunlight, changes should be visible in natural-color satellite images for a while longer before the onset of winter darkness.


Figure 136: The detailed view shows the new rift growing away from an area known as the McDonald Ice Rumples. The rift shows up in satellite images as early as September 2019, when it had grown just over 2 kilometers longer during the austral winter. But the biggest growth just occurred recently. Between November 18 and December 22, 2020, the rift grew in length by about 20 kilometers. Then it jogged toward the north and grew an additional 8 kilometers by January 12, 2021 (image credit: NASA Earth Observatory)

• January 19, 2021: Smooth, stationary clouds are occasionally reported by the public as sightings of “unidentified flying objects.” But these clouds are not as mysterious as they might first seem. 85)


Figure 137: On December 29, 2020, the OLI instrument on Landsat-8 acquired these images of soft-edged clouds hovering over the Eisenhower Range of Antarctica’s Transantarctic Mountains. The range is bounded to the north by Priestley Glacier and to the south by Reeves Glacier, both of which feed into the Nansen Ice Shelf on Terra Nova Bay [image credit: NASA Earth Observatory images by Joshua Stevens, using Landsat data from the U.S. Geological Survey. Story by Kathryn Hansen with image interpretation by Bastiaan Van Diedenhoven (NASA GISS/Columbia) and Jan Lenaerts (CU Boulder)]

- The clouds have the hallmarks of lenticular clouds that can form along the crests of mountain waves. Mountain waves form when fast moving wind is disturbed by a topographic barrier—in this case, the Eisenhower Range. Air is forced to flow up and over the mountains, causing waves of rising and falling air downwind of the range. The rising air cools and water vapor condenses into clouds. Conversely, falling air leads to evaporation.

- Adding to their mystique, this cloud type appears to stay put—sometimes for hours—defying the strong horizontal winds. In reality, the clouds are constantly building around the crest of the wave and then dissipating just beyond.


Figure 138: Detail image of smooth, soft-edged clouds hovered over the Eisenhower Range in Victoria Land, Antarctica (image credit: NASA Earth Observatory)

- In the United States, lenticular clouds are particularly common around the Rocky Mountains. They have been known to occur over Antarctic mountains, too, but there are not many witnesses besides satellites. The white-on-white color of clouds over ice make the Antarctic versions harder to discern, even in satellite images. This natural-color image has been enhanced with infrared light to separate the white clouds from the white snow and ice below. The clouds also threw rounded shadows on the landscape.

- Still, a few people have witnessed lenticular clouds in Antarctica firsthand. Scientists working with NASA’s Operation IceBridge shot photos of the phenomenon near Mount Discovery in 2013 and over Penny Ice Cap in 2015.

• January 11, 2021: With its population rising three times faster than the national average, the Charleston metropolitan area in South Carolina is among the fastest growing places in the United States. 86)


Figure 139: Large tracts of coastal forests and farmland have been cleared and developed in recent decades to accommodate new residents to the area. The pair of natural-color Landsat images above—this image from 1985 (on Landsat-5, TM) and the image of Figure 140 from 27 December 2020—show some of the changes. Forests and marshes appear green; developed areas are gray. Places where widespread development has occurred include James Island, Johns Island, Daniel Island, West Ashley, and Mount Pleasant (image credit: NASA Earth Observatory images by Lauren Dauphin, using Landsat data from the U.S. Geological Survey)


Figure 140: Sea level rise and new development are on a collision course in South Carolina lowcountry.. Charleston metropolitan area observed by OLI on Landsat-8 on 27 December 2020 (image credit: NASA Earth Observatory)

- A similar story is playing out in cities all across the United States, but the Charleston area stands out in one critical way—much of the new development has happened on low-lying land that is especially vulnerable to sea level rise and flooding. Older, more established parts of Charleston—often on slightly higher land but surrounded by water on three sides—faces similar challenges. As one form of remediation, local and federal government officials are moving forward with plans to build a seawall to protect the city’s historic downtown from encroaching water.

- “Other southeastern coastal cities face similar problems but with one caveat: the lowcountry of South Carolina is low,” said Norman Levine, director of the Santee Cooper GIS Laboratory and Lowcountry Hazards Center at the College of Charleston. “Over one-third of all homes are built on land that sits below 10 feet (3 meters) of elevation.”

- However, hurricane storm surges up to 9 feet have been measured in the past, and climatologists expect surges to grow larger as global climate warms and storms become more intense.

- High tide, or “nuisance flooding,” is already far more common now than it was decades ago, according to Dale Morris, the coauthor of a 2019 report that assessed the region’s flood risks. On average, Charleston saw 10 to 25 tidal floods per year in the 1990s. There were 89 such events in 2019 and 69 in 2020, he said. In other words, the city now sees tidal flooding every 4 to 5 days.

- Both problems are amplified by sea level rise. Relative sea level in Charleston has risen by 10 inches (25 cm) since 1950, with an acceleration to 1 inch (3 cm) every 2 years since 2010.

- “If you look at a lot of the recent development, it impinges upon or is in low-lying floodplains and adjacent land,” said Morris. “These areas used to flood and no one really noticed. Now they flood and impact people’s lives, resources, and livelihoods.”

- The report offers some general principals and recommendations for future development. Development should respect the landscape's natural drainage patterns and soil qualities. Coastal forests—which sponge up water—should be preserved wherever possible. And according to the report authors, development on the lowest-lying areas should not happen.

- “We are not saying don’t develop at all,” said Morris. “We are saying to develop wisely, carefully, sensibly given the current and future flood risks. Those risks are not going to decrease.”

• January 7, 2021: Popocatépetl volcano—the name is Aztec for “smoking mountain”—is one of Mexico’s most active volcanoes. The glacier-clad stratovolcano has been erupting since January 2005, with daily low-intensity emissions of gas, steam, and ash. 87)


Figure 141: Ash and gas emissions continue from one of Mexico’s most active volcanoes. On January 2, 2021, the Operational Land Imager (OLI) on Landsat-8 captured this image of a plume rising from Popocatépetl (nicknamed El Popo), image credit: NASA Earth Observatory image by Lauren Dauphin, using Landsat data from the U.S. Geological Survey. Story by Kasha Patel

- On January 6, the Washington Volcanic Ash Advisory Center (VAAC) reported a volcanic ash plume that rose to around 6,400 meters (21,000 feet) above the volcano. Mexico’s National Center for Prevention of Disasters (CENAPRED), which continuously monitors Popo, warned people not to approach the volcano or its crater due to falling ash and rock fragments. Some ashfall was blown downwind to the city of Puebla, located about 45 kilometers (30 miles) away from the volcano.

- At 5,426 meters (17,802 feet) above sea level, Popocatépetl is the second tallest volcano in Mexico (after Citlaltépetl). It is comprised of alternating layers of volcanic ash, lava, and rocks from earlier eruptions. The volcano is located around 70 kilometers (40 miles) southeast of Mexico City and more than 20 million people live close enough to be affected by a major eruption. However, most of the eruptions in the past 600 years have been relatively mild.

Landsat-8 Initial imagery until May 2013 when Landsat-8 was declared operational

Landsat-8 is operational — LDCM was officially renamed to Landsat-8. On May 30, 2013, NASA transferred operational control of the Landsat-8 satellite to the USGS (U.S. Geological Survey ) in Sioux Falls, S.D. This marks the beginning of the operational phase of the Landsat-8. The USGS now manages the satellite flight operations team within the Mission Operations Center, which remains located at NASA’s Goddard Space Flight Center in Greenbelt, MD.

The mission carries on a long tradition of Landsat satellites that for more than 40 years have helped to study how Earth works, to understand how humans are affecting it and to make wiser decisions for the future. The USGS will collect at least 400 Landsat-8 scenes every day from around the world to be processed and archived at the USGS/EROS (Earth Resources Observation and Science Center) in Sioux Falls. 88)

• May 22, 2013: One of two new spectral bands identifies high-altitude, wispy cirrus clouds that are not apparent in the images from any of the other spectral bands. The March 24, 2013, natural color image of the Aral Sea, for example, appears to be from a relatively clear day. But when viewed in the cirrus-detecting band, bright white clouds appear. 89)

The SWIR band No 9 (1360-1390 nm) is the cirrus detection band of the OLI (Operational Land Imager) instrument. Cirrus clouds are composed of ice crystals. The radiation in this band bounces off of ice crystals of the high altitude clouds, but in the lower regions, the radiation is absorbed by the water vapor in the air closer to the ground. The information in the cirrus band is to alert scientists and other Landsat users to the presence of cirrus clouds, so they know the data in the pixels under the high-altitude clouds could be slightly askew. Scientists could instead use images taken on a cloud-free day, or correct data from the other spectral bands to account for any cirrus clouds detected in the new band.

Figures 142 and 143 are simultaneous OLI observations of the same area of the Aral Sea region in Central Asia which illustrate the power of interpretation of a scene. The cirrus clouds of Figure 143 are simply not visible in the natural color image of Figure 142. This new analysis feature will give scientists a better handle to study the changing environment.


Figure 142: Natural color image of the Aral Sea region observed on March 24, 2013 (image credit: NASA)


Figure 143: Cirrus cloud detection band image of the Aral Sea region observed on March 24, 2013 (image credit: NASA)

• May 9, 2013: Availability of free long-term Landsat imagery to the public. Today, Google released more than a quarter-century of images, provided free to the public, of Earth taken from space and compiled into an interactive time-lapse experience. Working with data from the Landsat Program managed by the USGS (U.S. Geological Survey), the images display a historical perspective on changes to Earth's surface over time. 90) 91) 92) 93)

The long-term archive of Landsat images of every spot on Earth is a treasure trove of scientific information that can form the basis for a myriad of useful applications by commercial enterprises, government scientists and managers, the academic community, and the public at large.

In 2009, Google started working with USGS to make this historic archive of Earth imagery available online. Using Google Earth Engine technology, the Google team sifted through 2,068,467 images—a total of 909 terabytes of data—to find the highest-quality pixels (e.g., those without clouds), for every year since 1984 and for every spot on Earth. The team then compiled these into enormous planetary images, 1.78 terapixels each, one for each year.

• May 6, 2013: As the LDCM satellite flew over Indonesia's Flores Sea on April 29, it captured an image of Paluweh volcano spewing ash into the air. The satellite's OLI instrument detected the white cloud of smoke and ash drifting northwest, over the green forests of the island and the blue waters of the tropical sea. The TIRS (Thermal Infrared Sensor) on LDCM picked up even more. 94) 95)


Figure 144: An ash plume drifts from Paluweh volcano in Indonesia in this image, taken April 29, 2013 with OLI (image credit: NASA)

By imaging the heat emanating from the 5-mile-wide volcanic island, TIRS revealed a hot spot at the top of the volcano where lava has been oozing in recent months (Figure 145).


Figure 145: This thermal image was taken by the TIRS instrument on April 29, 2013 (image credit: USGS, NASA)

Legend to Figure 145: A bright white hot spot, surrounded by cooler dark ash clouds, shows the volcanic activity at Paluweh volcano in the Flores Sea, Indonesia. The image of Paluweh also illuminates TIRS' abilities to capture the boundaries between the hot volcanic activity and the cooler volcanic ash without the signal from the hot spot bleeding over into pixels imaging the cooler surrounding areas.

• May 2, 2013: All spacecraft and instrument systems continue to perform normally. LDCM continues to collect more than 400 scenes per day and the U.S. Geological Survey Data Processing and Archive System continues to test its ability to process the data flow while waiting for the validation and delivery of on-orbit calibration, which convert raw data into reliable data products. 96)

• On April 12, 2013, LDCM (Landsat Data Continuity Mission) reached its final altitude of 705 km. One week later, the satellite’s natural-color imager (OLI) scanned a swath of land 185 km wide and 9,000 km long. 97) 98)

• Since April 4, 2013, LDCM is on WRS-2 (Worldwide Reference System-2),


Figure 146: These images show a portion of the Great Salt Lake, Utah as seen by LS-7 (left) and LS-8 (LDCM) satellites (right); both images were acquired on March 29, 2013 (image credit: USGS, Ref. 98)

Legend to Figure 146: On March 29-30, 2013, the LDCM was in position under the Landsat 7 satellite. This provided opportunities for near-coincident data collection from both satellites. The images below show a portion of the Great Salt Lake in Utah, and the Dolan Springs, Arizona area, the latter of which is used in Landsat calibration activities. 99)

• March 21, 2013: Since launch, LDCM has been going through on-orbit testing. The mission operations team has completed its review of all major spacecraft and instrument subsystems, and performed multiple spacecraft attitude maneuvers to verify the ability to accurately point the instruments. 100)

- As planned, LDCM currently is flying in an orbit slightly lower than its operational orbit of 705 km above Earth's surface. As the spacecraft's thrusters raise its orbit, the NASA-USGS team will take the opportunity to collect imagery while LDCM is flying under Landsat 7, also operating in orbit. Measurements collected simultaneously from both satellites will allow the team to cross-calibrate the LDCM sensors with Landsat 7's Enhanced Thematic Mapper-Plus instrument.

- After its checkout and commissioning phase is complete, LDCM will begin its normal operations in May. At that time, NASA will hand over control of the satellite to the USGS, which will operate it throughout its planned five-year mission life. The satellite will be renamed Landsat 8. USGS will process data from OLI and TIRS and add it to the Landsat Data Archive at the USGS Earth Resources Observation and Science Center, where it will be distributed for free via the Internet.


Figure 147: First image of LDCM released in March 2013 (image credit: NASA) 101)

Legend to Figure 147: The first image shows the meeting of the Great Plains with the Front Ranges of the Rocky Mountains in Wyoming and Colorado. The natural-color image shows the green coniferous forest of the mountains coming down to the dormant brown plains. The cities of Cheyenne, Fort Collins, Loveland, Longmont, Boulder and Denver string out from north to south. Popcorn clouds dot the plains while more complete cloud cover obscures the mountains.
The image was observed on March 18, 2013 using data from OLI (Operational Land Imager) bands 3 (green), 5 (near infrared), and 7 (short wave infrared 2) displayed as blue, green and red, respectively.

• March 18, 2013: First day of simultaneous OLI and TIRS Earth imaging (Ref. 98).

• Feb. 21, 2013: The LDCM mission operations team successfully completed the first phase of spacecraft activation. All spacecraft subsystems have been turned on, including propulsion, and power has been supplied to the OLI (Operational Land Imager) and TIRS (Thermal Infrared Sensor) instruments. 102)

• LDCM will go through a check-out phase for the next three months. Afterward, operational control will be transferred to NASA's mission partner, the USGS (U.S. Geological Survey), and the satellite will be renamed to Landsat-8. The data will be archived and distributed free over the Internet from the EROS (Earth Resources Observation and Science) center in Sioux Falls, S.D. Distribution of Landsat-8 data from the USGS archive is expected to begin within 100 days of launch.

• The LDCM spacecraft separated from the rocket 79 minutes after launch and the first signal was received 3 minutes later at the ground station in Svalbard, Norway. The solar arrays deployed 86 minutes after launch, and the spacecraft is generating power from them (Ref. 25).

Minimize Landsat 8 continued

Sensor complement: (OLI, TIRS)

Background: In 2008 the TIRS (Thermal Infrared Sensor) instrument was still regarded an option to the LDCM mission. However, in Dec. 2009, the US government confirmed that TIRS would be developed and would be on board the LDCM spacecraft. In the spring of 2010, TIRS passed the CDR (Critical Design Review). 103) 104)

The OLI and TIRS data are merged into a single data stream. Together the OLI and TIRS instruments on LDCM replace the ETM+ instrument on Landsat-7 with significant enhancements.


Figure 148: Photo of the EM PIE (Payload Interface Electronics) equipment, image credit: NASA

OLI (Operational Land Imager):

Already in July 2007, NASA had awarded a contract to BATC (Ball Aerospace Technology Corporation) of Boulder, CO, to develop the OLI (Operational Land Imager) key instrument for LDCM. The BATC contract terms call for the design, development, fabrication and integration of one OLI flight model. Furthermore, the company is also required to test, deliver and provide post-delivery support and five years of on-orbit support for the instrument.

The multispectral and moderate resolution OLI instrument has similar spectral bands to the ETM+ (Enhanced Thermal Mapper plus) sensor of Landsat-7. It includes new coastal aerosol (443 nm, band 1) and cirrus detection (1375 nm, band 9) bands, though it does not have a thermal infrared band.

The following list provides an overview of the most important observation requirements for the OLI instrument: 105)

• The specifications require delivery of data covering at least 400 Landsat scenes/day (185 km x 180 km) for the US archive. The data are to be acquired in a manner that affords seasonal coverage of the global land mass. Data are required for the heritage reflective Thematic Mapper (TM) spectral bands plus two new bands, a blue band for coastal zone observations and a short wave infrared band for cirrus cloud detection.

• 30 m GSD (Ground Sample Distance) for VIS/NIR/SWIR, 15m GSD for PAN data.

• The specifications do not require thermal data (TIR band), representing a departure from the TM (Thematic Mapper) heritage. The specification also requires data providing a 30 m GSD (Ground Sample Distance) for each of the multispectral bands. Note: The TIR band was deselected due to the extra cost of active cooling.

• An edge response slope is also specified for the image data from each spectral band. The edge response is defined as the normalized response of the image data to a sharp edge as expressed in a Level 1R VDP (Validation Data Product). An edge response slope of 0.027 is required for bands 1 through 7, a slope of 0.054 is required for the panchromatic band, band 8, and a slope of 0.006 for the cirrus band, band 9.

• All instrument source data will be quantized to 12 bit resolution.

Band Nr

Band Name

Spectral range (nm)

Use of data


Radiance (W/m2 sr μm), typical



New Deep Blue


Aerosol/coastal zone

30 m









30 m
(TM heritage bands)






























Minerals/litter/no scatter






Image sharpening

15 m






Cirrus cloud detection

30 m



Table 2: NASA/USGS requirements for LDCM imager spectral bands

• The WRS-2 (Worldwide Reference System-2) defines Landsat scenes as 185 km x 180 km rectangular areas on the Earth's surface designated by path and row coordinates. This heritage system is used to catalogue the data acquired by the Landsat 4, 5, and 7 satellites and will also be used for the LDCM.

• Provide “standard”, orthorectified data products within 24 hours of observation (products available via the web at no cost)

• Data calibration consistent with previous Landsat missions

• Continue IC (International Cooperator) downlinks

• Support priority imaging and a limited off-nadir collection capability (± 1 path/row).


ETM+ (Landsat-7)

Band Nr

Wavelength (µm)

GSD (m)

Band No.

Wavelength (µm)

GSD (m)

8 (PAN)

0.500 - 0.680


8 (PAN)

0.52 - 0.90



0.433 - 0.453






0.450 - 0.515



0.45 - 0.52



0.525 - 0.600



0.53 - 0.61



0.630 - 0.680



0.63 - 0.69






0.78 - 0.90



0.845 - 0.885






1.360 - 1.390






1.560 - 1.660



1.55 - 1.75



2.100 - 2.300



2.09 - 2.35


OLI does not include thermal imaging capabilities

6 (TIR)

10.40 - 12.50


Figure 149: Spectral parameter comparison of OLI and ETM+ instruments


Figure 150: OLI and ETM spectral bands (image credit: NASA)

OLI instrument:

The OLI design features a multispectral imager with a pushbroom architecture (Figure 151) of ALI (Advanced Land Imager) heritage, a technology demonstration instrument flown on the EO-1 spacecraft of NASA (launch Nov. 21, 2000). A pushbroom implementation is considered to be more geometrically stable than the whiskbroom scanner of the ETM+ instrument. As a tradeoff of this architecture selection, the imagery must be terrain corrected to ensure accurate band registration.. 106) 107) 108) 109) 110)


Figure 151: Schematic view of the OLI instrument design (image credit: BATC)

The FPA (Focal Plane Assembly) consists of 14 FPMs (Focal Plane Modules). This is a consequence of the pushbroom architecture selection for OLI leading to a different set of geometric challenges than a cross-track whiskbroom implementation. Instead of using a small focal plane and a scanning mirror, 14 FPMs are required to cover the full Landsat cross-track field of view. Each FPM contains nine spectral bands in along-track (Figure 152). The along-track spectral band separation leads to an approximately 0.96-second time delay between the leading and trailing bands. This time delay creates a small but significant terrain parallax effect between spectral bands, making band registration more challenging.

The along-track dimension of the OLI focal plane (see Figure 153) also makes it desirable to “yaw steer” the spacecraft. This means that the spacecraft flight axis is aligned with the ground (Earth fixed) velocity vector, rather than with the inertial velocity vector, in order to compensate for cross-track image motion due to Earth rotation.

Although the pushbroom architecture requires many more detectors and a correspondingly larger focal plane, it also allows for a much longer detector dwell time (~4 ms for OLI vs. 9.6 µs for ETM+), leading to much higher signal-to-noise ratios. The lack of moving parts in the pushbroom design also allows for a more stable imaging platform and good internal image geometry.


Figure 152: Schematic view of the FPM layout concept (image credit: BATC, USGS)


Figure 153: Orientation of the FPMs in the FPA (Focal Plane Assembly) of the OLI instrument (image credit: BATC)

Each FPM contains detectors for each spectral band, silicon for the VNIR bands and HgCdTe for the SWIR bands and a butcher-block filter assembly to provide the spectral bands.

OLI features about 6500 active detectors per multispectral band and 13000 detectors for the panchromatic band. These detectors are organized as blocks ~500 multispectral (1000 panchromatic) detectors wide within 14 focal plane modules (FPMs) that make up the focal plane assembly. Each module has its own butcher-block assembly spectral filter. This provides significantly improved signal to noise performance, but complicates the process of radiometrically matching the detectors responses. Similarly, the lack of a scan mirror removes the need for knowledge of its movement, but requires knowledge of the detectors locations across a much larger focal plane (Ref. 2).

Observation technique

Pushbroom imager

Spectral bands

9 bands in VNIR/SWIR covering a spectral range from 443 nm to 2300 nm


- Four-mirror off-axis telescope design with a front aperture stop
- Use of optical bench
- Telecentric design with excellent stray light rejection

FPA (Focal Plane Assembly)

- Consisting of 14 sensor chip assemblies mounted on a single plate
- FPA is passively cooled
- Hybrid silicon / HgCdTe detectors
- Butcher block filter assembly over each SCA (Sensor Chip Assembly)

Swath width (FOV=15º)

185 km

GSD (Ground Sample Distance)

15 m for PAN data; 30 m for VNIR/SWIR multispectral data

Data quantization

12 bit


- Solar calibrator (diffuser) used once/week
- Stimulation lamps used to check intra-orbit calibration
- Dark shutter for offset calibration (used twice per orbit)
- Dark detectors on focal plane to monitor offset drift

Instrument, mass, power, size


Table 3: Overview of OLI instrument parameters

The OLI will provide global coverage by acquiring ~400 scenes per day in six VNIR and three SWIR bands, all at 12 bit radiometric resolution. In addition to these bands, there will be a tenth band consisting of covered SWIR detectors, referred to as the ‘blind’ band, that will be used to estimate variation in detector bias during nominal Earth image acquisitions. The OLI bands are distributed over 14 SCAs (Sensor Chip Assemblies) or FPMs, each with 494 detectors per 30 m band and twice as many for the 15 m panchromatic band - totaling in over 75000 imaging detectors. 111)

OLI calibration:

The OLI calibration subsystem (Figures 154 and 155) consists of two solar diffusers (a working and a pristine), and a shutter. When positioned so that the sun enters the solar lightshade, the diffusers reflect light diffusely into the instruments aperture and provide a full system full aperture calibration. The shutter, when closed, provides a dark reference. In addition, two stim lamp assemblies are located at the front aperture stop. Each lamp assembly contains three lamps (per redundant configuration) that are operated at constant current and monitored by a silicon photodiode. The lamp signal goes through the full telescope system. Additionally, the OLI focal plane will include masked HgCdTe detectors, that is, detectors that will be blocked from seeing the Earth’s radiance (Ref. 2). 112) 113)

Solar diffusers:

- Full-aperture full system Spectralon diffuser, designed to be used at different frequencies to aid in tracking the system and diffuser changes. The pristine diffuser will be used to check degradation of main diffuser.

- The primary solar diffuser will nominally be deployed every 8 days to track the calibration of the OLI sensor and perform detector-to-detector normalization.

- The solar diffuser based calibration requires a spacecraft maneuver to point the OLI solar calibration aperture towards the sun. The pristine diffuser will be used on a less frequent basis, about every six months, as a check on the primary diffuser's degradation.

Stimulation lamps:

- Multi—bulbed tungsten lamp assemblies, that illuminate the OLI detectors through the full optical system, similarly designed to be used at different frequencies to separate lamp and system changes. The working lamp will be used daily for intra-orbit calibration/characterization; the reference lamp set approximately monthly, and the pristine lamp set approximately twice a year.

- The lamb assembly can also be compared to solar diffuser measurements to check stability.

Dark shutter:

- Used twice per orbit for offset calibration

• Dark detectors on focal plane to monitor offset drift

• Linearity checked by varying detector integration time.

The LDCM operational concept also calls for the spacecraft to be maneuvered every lunar cycle to view the moon, providing a "known" stable source for tracking stability over the mission. A side-slither maneuver, where the spacecraft is rotated 90º to align the detector rows with the velocity vector, is also planned. These data will provide an additional method to assess the detector-to-detector radiometric normalization.

Pre-launch spectro-radiometric characterization and calibration (Ref. 112):

The spectral characterization of the OLI instrument is being performed at the component, focal plane module and fill instrument levels. The components, which have all completed testing, include detector witness samples, spectral filters prior to dicing into flight filter sticks, the focal plane assembly window witness samples and telescope mirror witness samples.

The FPM (Focal Plane Module) level tests, which are also complete, are specifically designed to characterize the spectral out-of-band response. The FPM level tests measure the spectral response of all the detectors by illuminating the full focal plane at approximately the correct cone angle.

An integrating sphere is used in the pre-launch radiance calibration of the OLI. The traceability of the calibration of this sphere will start with the 11" OLI transfer sphere directly calibrated at the NIST Facility for Spectroradiometric Calibration (FASCAL). While still at NIST, this OLI transfer sphere is checked by independently NIST calibrated University of Arizona (UAR VNIR transfer radiometer), NASA and NIST (Government Transfer Radiometers) radiometers. Also, the Ball Standard Radiometer (BSR), that has filters matching the OLI bands, views the sphere.


Figure 154: OLI block diagram illustrating the calibration subsystem in front of the telescope (image credit: NASA, BATC)


Figure 155: Blow-up of the calibration subsystem illustrating the solar diffuser and shutter assemblies (image credit: NASA, BATC)


Figure 156: Illustration of the OLI instrument (image credit: NASA, BATC)

In Nov. 2008, the OLI instrument passed the ICDR (Instrument Critical Design Review). 114)


Figure 157: Photo of the completed OLI instrument with electronics (image credit: BATC, NASA, USGS)

Delivery of the OLI instrument in the summer of 2011 (Ref. 3).

TIRS (Thermal Infrared Sensor)

The TIRS instrument is providing continuity for two infrared bands not imaged by OLI. NASA/GSFC is building the TIRS instrument inhouse. TIRS is a late addition to the LDCM mission, the requirements call for a GSD (Ground Sample Distance of 120 m for the imagery; however, the actual GSD will be 100 m.

The LDCM ground system will merge the data from both sensors into a single multispectral image product. These data products will be available for free to the general public from the USGS enabling a broad scope of scientific research and land management applications. 115) 116)

TIRS is a QWIP (Quantum Well Infrared Photodetector) based instrument intended to supplement the observations of the OLI instrument. The TIRS instrument is a TIR (Thermal Infrared) imager operating in the pushbroom mode with two IR channels: 10.8 µm and 12 µm. The two spectral bands are achieved through interference filters that cover the FPA (Focal Plane Assembly). The pushbroom implementation increases the system sensitivity by allowing longer integration times than whiskbroom sensors. The two channels allow the use of the “split-window” technique to aid in atmospheric correction.


Figure 158: Functional block diagram of TIRS (image credit: NASA, Ref. 113)

The focal plane consists of three 640 x 512 QWIP GaAs arrays mounted on a silicon substrate that is mounted on an invar baseplate. The two spectral bands are defined by bandpass filters mounted in close proximity to the detector surfaces. The QWIP arrays are hybridized to ISC9803 readout integrated circuits (ROICs) of Indigo Corporation. The focal plane operating temperature will be maintained at 43 K (nominally). 117) 118) 119)

Instrument type

Pushbroom imager

Two channel thermal imaging instrument

10.8 and 12.0 µm band centers


10.3-11.3 µm,
11.5-12.5 µm

GSD (Ground Sample Distance)

100 m (nominal), 120 m (requirement)

Swath width

185 km, FOV = 15º

Operating cadence

70 frames/s

Instrument calibration

- Scene select mirror to select between 2 calibration sources
- Two full aperture calibration sources: onboard internal calibration and space view


- Three SCA (Sub-Chip Assembly) QWIP detectors built in-house at Goddard
- FPA consists of three 640 x 512 detector arrays
- Pixel size of 25 µm producing an IFOV of 142 µrad
- The FPA consists of an invar “spider” which is bonded to the silicon interface board
containing the QWIPs and on which the “daughter boards” are mounted.
- Actively cooled FPA operating at 43 K
- Two-stage cryocooler provided by BATC


- The telescope is a 4-element refractive lens system.
- Passively cooled telescope operating at 185 K

Telescope f number


Data quantization

12 bit

Instrument mass, size, power

236 kg, approx: 80 cm x 76 cm x 43 cm, 380 W

Table 4: TIRS instrument parameters

QWIP detector: The development of the QWIP detector technology has made great strides in the first decade of the 21st century. In 2008, NASA/GSFC revised the design of the infrared detector concept of the TIRS (Thermal Infrared Sensor) imager, under development for the LDCM (Landsat Data Continuity Mission). The initially considered HgCdTe-based detector design was changed to a QWIP design due to the emergence of broadband QWIP capabilities in the MWIR and TIR (LWIR) regions of the spectrum. The introduction of QWIP technology for an operational EO mission represents a breakthrough made possible through collaborative efforts of GSFC, the Army Research Lab and industry (Ref. 118).

An important advantage of GaAs QWIP technology is the ability to fabricate arrays in a fashion similar to and compatible with the silicon IC technology. The designer’s ability to easily select the spectral response of the material from 3 µm to beyond 15 µm is the result of the success of band-gap engineering. 120)

Advantages of QWIP technology:

- Large lattice-matched substrates

- Mature materials technology

- No unstable mid-gap traps

- Inherently, radiation hard.


Figure 159: QWIP quantum state diagram (image credit: NASA/JPL)


Figure 160: TIRS 10-13 µm QWIP spectral response requirement (image credit: NASA)


Figure 161: Overview of the TIRS focal plane layout (image credit: NASA, Ref. 113)

The three arrays are precisely aligned to each other in the horizontal and vertical directions (to within 2 µm). There is a requirement that the detection region within the QWIP array be within 10 µm of a common focal plane altitude. This specification is challenging since it includes surface non-uniformities of the baseplate, substrate, the QWIP/ROIC hybrid and the epoxy bond lines between these components. Nonetheless, since there are three discreet arrays they must all fall within a single focus position.

The filter bands are further confined to specific regions of the QWIP array. Although each array contains 512 rows, after all the operational requirements are satisfied (frame rate, windowing, co-registration, scene reconstruction, etc.) only 32 rows are available under each filter band separated by 76 rows of occluded pixels (for dark current subtraction). Once all these requirements are incorporated into the focal plane design, eligible rows on any given array are pre-determined. Of these eligible rows, there must be three that can be combined to make two perfect rows, or preferably, at least two perfect rows (that is, rows where all pixels meet every specification).

TRL (Technology Readiness Level) tests: An important and essential process for qualifying new or previously unused technology in a NASA space mission is the technology readiness level demonstration. There are nine levels with level 6 (TRL 6) being the level at which new hardware must be demonstrated. Typically, this means qualification in the environment which the instrument will be subjected through out the mission; radiation effects, vibration, thermal cycling and (in some cases) shock. Both the readout and QWIP hybrids were subjected to gamma, proton and heavy ion radiation equivalent to 35 krad or almost 10 times the expected mission dose. At these levels and at the operating temperature of 43 K minimal effects were observed and none were considered to be a mission risk.

A fully functioning focal plane assembly was subjected to 40 thermal cycles from 300 K to 77 K and back to 300 K. Every tenth cycle went to 43 K to collect the array performance data. After the completion of the 40 cycles there was essentially no change in any of the three QWIP arrays (2 grating QWIP hybrids and one C-QWIP hybrid). - The final environmental test performed was vibration to simulate the effect of the launch. Since this is a qualification test the vibration loads are specified 3db above the expected loads. The focal plane assembly was subjected to a series of vibration input loads including x, y and z-axis random vibration for 2 minutes/axis, a sine sweep and sine burst test (15 g at 20 Hz). No failures occurred and this assembly and the overall design was certified by an independent review panel as having met the requirements for TRL 6.


Figure 162: Schematic view of the FPA (Focal Plane Assembly), image credit: NASA


Figure 163: Photos of the FPA (image credit: NASA)

Legend to Figure 163: The left photo of the FPA is without filters showing the 3 QWIPs in the center. The daughter boards are the red and green assemblies to the left and right, respectively. The invar spider is the component with the 4 arms. - The right picture of the FPA comes with the filters attached. Note that there are two filters over each array with a thin dark strip between them.

Optical system: The imaging telescope is a 4-element refractive lens system. A scene select mechanism (SSM) rotates a scene mirror (SM) to change the field of regard from a nadir Earth view to either an on-board blackbody calibrator or a deep space view. The blackbody is a full aperture calibrator whose temperature may be varied from 270 to 330 K.

The optical system, consisting of a lens with three Ge elements and one ZnSe element, produces nearly diffraction-limited images at the focal plane. All but 2 of the surfaces are spherical, which simplifies fabrication. The optics are radiatively cooled to a nominal temperature of 185 K to reduce the contribution of background thermal emission to the measurement noise. Because of the fairly strong thermal dependence of the index of refraction of Ge, the focus position of the lens is a function of the optics temperature. This provides a method of adjusting focus so that, in the unlikely event that launch conditions or some other effect defocus the system, the temperature of the optics may be changed by ±5 K to refocus. That is, thermal control of the lens provides a non-mechanical focus mechanism. A +5 K change does not significantly degrade the noise performance.

A precision scene select mirror is an essential component of the TIRS instrument and it is driven by the scene select mechanism. It rotates around the optical axis on a 45º plane to provide the telescope with a view of Earth through the nadir baffle and two full aperture sources of calibration, onboard variable temperature blackbody (hot calibration target) and space view (cold calibration target). The onboard blackbody will be a NIST (National Institute of Standards and Technology) certified reference source (Figure 164).

TIRS is able to achieve a 185 km ground swath with a 15º FOV (Field of View) functioning in the pushbroom sample collection method. This method will have the benefit of being able to collect and record data without movement artifacts due to its wide instantaneous field of view. Frames will be collected at an operating cadence of 70 per second. The collected data will be stored temporarily stored on board and periodically sent to the USGS EROS facility for further storage. The instrument is designed to have an expected lifetime of at least a three years.


Figure 164: The TIRS optical sensor unit concept (image credit: NASA)


Figure 165: Schematic view of the TIRS instrument internal assembly (image credit: NASA, Ref. 113)

Legend to Figure 165: Model of the TIRS instrument showing the major components of the TIRS sensor. The scene select mechanism rotates the field of regard from the Earth view to either the space view or to the on-board calibrator. The right side provides some detail of optical system showing the 4-element lens, a cut-away view of the SM and the thermal strap connecting the FPA to the cryocooler cold tip. The MEB (Main Electronics Box) and the CCE (Cyrocooler and its associated Control Electronics), not shown, are mounted to the spacecraft.

TIRS instrument calibration:

Consistent with previous Landsat missions, LDCM TIRS will be fully calibrated prior to launch. Calibration measurements will be made at GSFC and will be done at the component, subsystem and instrument level. NIST-traceable instrument level calibration will be done using an in-chamber calibration system. 121) 122)

Among other uses, TIRS data will be used to measure evapotranspiration (evaporation from soil and transpiration from plants); to map urban heat fluxes, to monitor lake thermal plumes from power plants; to identify mosquito breeding areas and vector-borne illness potential; and to provide cloud measurements. The evapotranspiration data may be used to estimate consumptive water use on a field-by-field basis.

TIRS instrument calibration makes use of the following elements:

• Precision scene select mirror to select between calibration sources and nadir view

• Two full aperture calibration sources

- Onboard variable temperature blackbody

- Space view

- Calibration every 34 minutes

• NIST Traceable radiometric calibration


Figure 166: Schematic of inclusion of NIST standards (image credit: NASA)

TIRS calibration system:

• A 41 cm diameter source is covering full field and aperture of TIRS (Flood Source)

• Target Source Module (GeoRadSource)

- Blackbody point source w/ filter & chopper

- All reflective, off-axis parabola collimator

- Motorized target and filter wheels

- A square steering mirror system (33 cm side length) is permitting coverage of the full aperture and field

• Cooled enclosure over entire system

• External monochromator (spectral source)

• Components are mounted to common base plate.

The TIRS radiometric response is determined via the prelaunch characterization relative to the laboratory blackbody. This approach provides the highest accuracy calibration. The calibration philosophy is then to evaluate (or validate) the calibration parameters once TIRS is on orbit. If the calibration of TIRS is demonstrated to change significantly while on orbit using measurements during the checkout period, then the on-board blackbody (OBB) will be used as the primary pathway to NIST traceability.


Figure 167: Illustration of the TIRS calibration system (image credit: USGS)


Figure 168: Illustration of TIRS on the LDCM spacecraft (image credit: NASA, Ref. 3)

SSM (Scene Select Mechanism) of TIRS:

The SSM for the TI RS instrument, developed at NASA/GSFC, is a single axis, direct drive mechanism which rotates a 207 mm scene mirror from the nadir science position to the 2 calibration positions twice per orbit. It provides pointing knowledge and stability to ~10 µradians. The SSM can be driven in either direction for unlimited rotations. The rotating mirror is dynamically balanced over the spin axis, and does not require launch locking. 123)

The design of the SSM is straightforward; it is a single axis rotational mechanism. The operational cadence was to hold the scene mirror stationary for ~40 minutes staring at nadir, rotate 120º to the space view aperture and stare for 30 seconds, rotate 120º to the internal blackbody and stare for 30 seconds and then rotate the mirror to the back to nadir. Then the entire process would start again. The mechanism would be operating all of the time, or have a 100% duty cycle. Since LDCM/TI RS was to be in a highly-inclined polar orbit, the general idea was to calibrate twice per orbit while over the poles.


Figure 169: Cutaway view of the SSM (image credit: NASA)

Instrument mass, power

15 kg, 6 W average

Pointing knowledge, stability

±9.7 µradians over 34 minutes, ±9.7 µradians over 2.5 seconds

Duty cycle


Thermal operational

0 / +20ºC stable to ±1ºC

Thermal survival range

-50 / +40ºC


3.25 years on orbit


A/B side block redundancy

Operational cadence

Stare nadir for 30-40 minutes
Rotate 120º in < 2 minutes to space view
Stare for ~30 seconds,
Rotate 120º in < 2 minutes to blackbody view
Stare for ~30 seconds
Rotate to 120º in < 2 minutes to nadir view

Table 5: SSM driving requirements


Landsat ETM+


GMES/Sentinel-2 MSI

Spectral bands









1 (blue)


B1 (blue)


1 (blue)


2 (blue)


B2 (blue)


2 (green)


3 (green)


B3 (green)




4 (red)


B4 (red)






B5 (red edge)






B6 (red edge)






B7 (red edge)


4 (NIR)




B8 (NIR)




5 (NIR)


B8a (NIR)






B9 (water vapor)




9 (cirrus)


B10 (cirrus)


5 (SWIR1)


6 (SWIR1)


B11 (SWIR1)


7 (SWIR2)


7 (SWIR2)


B12 (SWRIR2)







6 (TIR)


10 (TIR1)






11 (TIR2)




GSD at nadir

30 m VNIR
15 m Pan
60 m TIR

30 m VNIR
15 m Pan
100 m TIR

10 m (B2, B3, B4, B8)
20 m (B5, B6, B7, B8a, B11, B12)
60 m (B1, B9, B10)


8 bit

12 bit

12 bit

Onboard Calibration




Resivit time

16 days

16 days

5 days (2 satellites)

Off-axis viewing

Up to 7.5º off nadir

Up to 7.5º off nadir

Up to 10.3º off nadir (w/o pointing)

Orbit altitude

705 km

705 km

786 km

Swath width

185 km

185 km

290 km


Cross-track scanner (Whiskbroom)



Table 6: Comparison of Landsat and GMES/Sentinel-2 imager specifications 124)

Collection of imagery onboard LDCM:

The co-aligned instruments are nominally nadir pointed and sweep the ground track land surface in contiguous image data collections, also known as image intervals. Each image interval may contain from a few WRS-2 scenes for an island or coastal area up to 77 contiguous WRS-2 scenes for an extended area of interest. For each image interval, the observatory executes a pre-defined imaging and ancillary data collection sequence as shown in Figure 170. 125)

Prior to the image interval, the spacecraft configures the onboard systems for the mission data collection session. A specific number of intervals are pre-defined on the ground based upon the number of WRS-2 scenes scheduled for collection, and allocated in the SSR (Solid State Recorder). Each instrument will transmit focal plane sensor data and instrument ancillary data (voltages, temperatures, etc.), which the spacecraft will interleave with the spacecraft ancillary data (attitude, ephemeris, etc), and record to files in the SSR.


Figure 170: Data collection sequence (image credit: USGS, NASA)

If the observatory is over an IC (International Cooperatoror) or LGN (Landsat Ground Network) station, it will simultaneously transmit data in real time to the ground. In addition, each instrument performs routine on-board calibrations (blackbody, lamps, etc) before and after each image interval, and during less frequent occasions utilizing the sun and moon as external calibration sources. A representation of the global image collection and calibration opportunities within the WRS-2 grid is shown in the 16-day repeating DRC-16 (Design Reference Case-16) in Figure 171.

The DRC-16 was developed to aid the mission architects in identification of all image and calibration activities and to verify that all are consistent with spacecraft power and mission data management capabilities. Instrument solar, lunar, and internal calibrations are required by ground system processing systems for image reconstruction, and to produce finished and distributable image products.


Figure 171: Illustration of DRC-16 collections (image credit: USGS, NASA)

End to end mission data flow is represented in Figure172 . Mission data originates as instrument sensor (or “image”) data, and are collected and processed by the instrument electronics. The instrument electronics transmits the image data to the spacecraft PIE (Payload Interface Electronics) over a HSSDB (High Speed Serial Data Bus), using a serializer-deserializer integrated circuit pair. OLI image data are compressed using the USES (Universal Source Encoder for Space) ASIC (Application-Specific Integrated Circuit), which implements the Rice algorithm for lossless compression. - TIRS data are not compressed due to the low data rate. Instrument image data are interleaved with spacecraft ancillary data to create a file, which is stored and/or provided to the transmitter for downlink.

Ancillary data are collected at rates up to 50 Hz, and is comprised of GPS (Global Positioning System) data, IMU (Inertial Measurement Unit) data, star tracker data, and select instrument engineering information required by ground system algorithms for image product generation. The ancillary data are multiplexed within the mission data files every second.

Mission data files are intentionally fixed in size at 1GB. A system architecture trade study was performed early in mission definition to establish the optimum file size given the implementation of Class 1 CCSDS (Consultative Committee for Space Data Systems) CFDP (File Delivery Protocol), and a required link BER (Bit Error Rate) of < 10-12. Utilizing the 440 Mbit/s available downlink capacity, each downlinked mission data file requires 22 seconds of continuous transmission for a completed delivery to the ground system. The low BER requirement on the communication link provides the confidence that only one file over several days would require retransmission, well within the available contact time with the ground stations.


Figure 172: Overview of the mission data flow (image credit: USGS, NASA)

Simultaneous real-time and playback mission data files are transmitted to the ground through virtual channels within a single physical channel. Five data streams on the transmitter interface board may be multiplexed on to the link via an arbiter, which interleaves the data streams according to a pre-established priority scheme. The data streams priorities are:

1) Real-time for the OLI instrument

2) Real-time for the TIRS instrument

3) SSR playback channel 1

4) SSR playback channel 2

5) A virtual channel for fill frames in case no image data are available for downlink.

The five virtual channels are arbitrated in priority order on a frame-by-frame basis. The OLI instrument has the highest priority followed by the TIRS instrument and, since the combined mission data rates are less than the total downlink bandwidth available, there is always residual bandwidth available for mission data playback. This enables maximum utilization of the downlink bandwidth. SSR playback 1 has priority over SSR playback 2. SSR playback 2 is controlled by an autonomous on-board spacecraft flight software task that queues files for playback by a predefined algorithm (i.e. oldest to newest priority files first, oldest to newest non-priority files next).

SSR playback 1 is specifically for ground system intervention, as required to supersede the onboard autonomous SSR playback 1, to downlink files which are a higher priority than originally categorized. SSR playback 2 will resume automatically upon the completion of SSR playback 1 ground commands, and there is no need to stop and restart the autonomous SSR playback 2 queue. As a frame finishes transmission, the priority arbiter selects the highest priority channel that has a frame buffer ready for transmission for the next frame.

To ensure the bandwidth of the space to ground data link is at least twice the bandwidth of the real-time mission data, the spacecraft compresses OLI image data in near real time using CCSDS compression. During early hardware development, using simulated data derived from the ALI (Advanced Land Imager) sensor aboard the EO-1 (Earth Observing-1) satellite (the precursor to the OLI instrument), data sets were constructed and flowed through the compression chip and achieved a nominal 1.55:1 compression. As compression varies on the OLI real-time virtual channel, the playback capacity also varies to use the bandwidth that is available.

The multiplexed virtual channels of mission data are provided to the RF X-band subsystem, where the transmitter adds CCSDS layers and LDPC (Low Parity Density Check) 7/8-rate forward error correction to the 384 Mbit/s data stream, resulting in a 440 Mbit/s stream from the X-band subsystem. The X-band stream is modulated, amplified and down-linked from the spacecraft antenna to the ground station antenna/receivers.

The ground station antenna system receives the 8200.5 MHz OQPSK (Offset-keyed Quadrature Phase Shift Keying) X-band signal from the observatory and forwards the down converted 1.2 GHz or 720 MHz intermediate frequency (IF) signal to a programmable telemetry receiver. The IF signals are routed through a matrix switch, providing signal distribution or routing to redundant equipment as needed.

Within the programmable telemetry receiver, the IF signal from the switching matrix is subject to low-pass filtering to prevent subsequent aliasing followed by an AGC (Automatic Gain Control) action. The AGC action is the last analog handling of the signal prior to the digitizer. The entire ground processing that remains is accomplished in the digital domain. The signal is immediately digitally demodulated within a specially designed modified Costas loop and the resultant baseband signal, now a softbit stream, is sent to the bit synchronizer. The bit stream has ambiguity resolved and is then frame synchronized. The frame synchronizer parses the data stream into equal length frames; queuing on a predefined frame synchronization pattern. The data are de-randomized using the conventional CCSDS algorithm and then stripped of parity and bit-corrected by the LDPC 7/8-rate FEC (Forward Error Correction) decoder.

The frame synchronization processor routes the framealigned data stream to the VCDU (Virtual Channel Data Unit) processor. The VCDU processor identifies the unique virtual channels within the frames and outputs these VCDU into individual data streams for packet processing.

The mission data stream from the VCDU processor is processed through the CCSDS packet processor to separate APID (Application Process Identifiers). The packet processor outputs the resulting mission stream to the CFDP processor.

Landsat-8 / LDCM ground system:

The LDCM ground system includes all of the ground-based assets needed to operate the LDCM observatory. The primary components of the ground segment are : 126) 127) 128)

- MOE (Mission Operations Element)

- CAPE (Collection Activity Planning Element)

- GNE (Ground Network Element)

- DPAS (Data Processing and Archive System).

The USGS (United States Geological Survey) -and their associated support and development contractors - will:

- Develop the Ground System (comprised of the Flight Operations and Data Processing and Archive Segments), excluding procurement of the MOE

- Provide ground system functional area expertise across all mission segments

- Lead, fund, and manage the Landsat Science Team

- Acquire the FOT (Flight Operations Team) and produce the FOT products 129)

- Lead the LDCM mission operations, after the completion of the on-orbit checkout period

- Accept and execute all responsibilities associated with the transfer of the LDCM OLI (Operational Land Imager) instrument, TIRS (Thermal Infrared Sensor) instrument, spacecraft bus and Mission Operations Element contracts from NASA following on-orbit acceptance of the LDCM system including assuming contract management”

- Provide system engineering for the USGS-managed segments and elements.

The MOE is being provided by the Hammer Corporation. The MOE contract was awarded in September 2008. The MOE provides capability for command and control, mission planning and scheduling, long-term trending and analysis, and flight dynamics analysis. The overall activity planning for the mission is divided between the MOE and CAPE. The MOE hardware and software systems reside in the LDCM MOC (Mission Operations Center).

The CAPE develops a set of image collection and imaging sensor(s) calibration activities to be performed by the observatory. The CAPE schedules activities on a path-row scene basis. The MOE converts CAPE-generated path-row scenes to observatory activities, schedules these and any other detailed observatory activities, and generates commands necessary to collect the identified scenes and operate the observatory.

The GNE is comprised of two nodes located at Fairbanks, Alaska and Sioux Falls, SD. Each node in the GNE includes a ground station that will be capable of receiving LDCM X-band data. Additionally, each station provides complete S-band uplink and downlink capabilities. The GNE will route mission data and observatory housekeeping telemetry to the DPAS.

The DPAS includes those functions related to ingesting, archiving, calibration, processing, and distribution of LDCM data and data products. It also includes the portal to the user community. The ground system, other than the MOE, is developed by USGS largely through their support service contract. The DAPS will be located at the USGS EROS (Earth Resources Observation and Science) Center in Sioux Falls, SD.

Data access policy: All Landsat data are freely available over the Internet.


Figure 173: Illustration of the Landsat-8 mission elements and communication architecture (image credit: NASA) 130) 131)

LGN (LDCM Ground Network) stations: The LGN is a collection of ground stations with state of the art electronics and sophisticated ground software, each providing similar mission services. The configuration of LGN uses the ground stations located at the EROS Center campus in Sioux Falls, South Dakota, the GLC (Gilmore Creek) ground station in Fairbanks, Alaska, and the SvalSat (Svalbard Satellite Station) ground station in Svalbard (Spitsbergen), Norway.

Each LGN ground station consists of a tracking antenna, S-band and X-band communication equipment, mission data storage and a file routing DCRS (Data Collection and Routing Subsystem). The LGN antenna receives X-band mission data files (autonomous playback or commanded) from the observatory, while simultaneously performing file management and subsequent image collection operations over S-band. The S-band and X-band systems of each LGN station interfaces with the MOE and DPAS in a closed loop fashion.

The USGS maintains agreements with several foreign governments referred to as the Landsat ICs (International Cooperators). The ICs are a special user community that has the ability to receive LDCM mission data from the observatory real-time X-band downlink stream. Real-time imaging sensor and ancillary data (including spacecraft and calibration data) required to process the science data are contained within the real-time stream downlink.

The ICs will be capable of receiving real-time X-band imaging sensor data downlinks and sending metadata to the DPAS. The ICs will submit imaging sensor data collection and downlink requests to the CAPE (via the DPAS user portal). ICs participate in a bilateral DV&E (Data Validation & Exchange) program with the DPAS. This program includes exchange of archive data upon request, and validation of IC processed level 1 data products by the USGS.


Figure 174: Overview of the LDCM system architecture (image credit: USGS)



Total size


Daily volume of 400 scenes

390 GByte


C&DH data rate

260.92 Mbit/s

Space to ground communication

Downlink data rate
LDPC ⅞ rate packet

384 Mbit/s data, 441 Msample/s symbol
8160 bit

Ground station

Minutes per day (14 contacts)

98 minutes

Science archive

5 year archive

~ 400 TB

Table 7: Overview of data volumes for processing and archiving functions

IC (International Cooperator) Ground Stations of the Landsat Program:

• In 41 years, 39 IC stations in 23 countries

• Most still collect and/or distribute Landsat products, reducing the load on U.S. Systems

• More than 215,000 products distributed in 2012

- Represents a nearly 10% off-loading of network bandwidth

- Enhanced regional exploitation of Landsat data


Figure 175: Overview of the IC (International Cooperator) network (image credit: USGS)

IAS (Image Assessment System):

Once the LDCM spacecraft is in orbit, the radiometric, geometric and spatial performance of OLI and TIRS sensors will be continually monitored, characterized and calibrated using the IAS (Ref. 111).

Background: The IAS was originally developed to monitor radiometric and geometric performance of the Landsat-7 ETM+ sensor and the quality of the image data in the Landsat-7 archive. The operational performance monitoring is achieved by processing a number of randomly selected Level 0R (raw reformatted) images to Level 1R (radiometrically corrected) and Level 1G (geometrically corrected) products. In that process, image statistics at different processing levels, calibration data, and telemetry data are extracted and stored in the IAS database for automatic and off-line assessment. The IAS also processes and analyzes the pre-selected geometric and radiometric calibration sites and special calibration acquisitions, e.g. solar diffuser or night data needed for radiometric calibration or noise and stability studies. The final and most important product of the IAS trending and processing is the CPF (Calibration Parameter File), the file that contains parameters needed for artifact corrections and radiometric and geometric processing of raw image data. To maintain the accuracy of the dynamic parameters, the CPF is updated at least once every three months.

The purpose of the LDCM IAS is to maintain accurate spectral, radiometric, spatial and geometric characterization and calibration of LDCM data products and sensors, ensuring compliance with the OLI and TIRS data quality requirements. The IAS will trend results of processing standard Earth images and nonstandard products, such as lunar, solar, dark Earth or stellar images, evaluate image statistics and calculate and store image characteristics for further analysis.


Figure 176: The LDCM ground system concept (image credit: USGS)

The IAS will automatically generate calibration parameters, which will be evaluated by the calibration analysts. In addition to standard operations within the IAS, the CVT (Calibration and Validation Team) will use a ‘toolkit’ module containing instrument vendor developed code and routines developed by the CVT, as a research and development environment for improvements of algorithm functionality and anomaly investigations.

Compared to the previous IAS versions, the LDCM IAS system will have to handle a significantly larger and more complex database that will include characterization data from all normally acquired images (~ 400 scenes per day, with special calibration acquisitions, e.g solar and lunar) processed through the product generation system. OLI’s pushbroom design (~ 75000 detectors), as opposed to an ETM+ whiskbroom design, requires characterization and calibration of about 550 times more detectors than in case of ETM+ (136 detectors) and represents a major challenge for the LDCM IAS. An additional challenge is that the LDCM IAS must handle data from two sensors, as the LDCM products will contain both the OLI and TIRS data.

For radiometric and geometric processing, see Ref. 111).

• Processing latency for real-time downlinks

• Average latency is ~ 5 hours from acquisition to product availability

• Closed loop between ground and space for data management

• The system requirement calls for 85% data availability to the user community through EROS Portal within 48 hours. The actual performance for Landsat-8 averages within 5 hours.

Table 8: Landsat 8 operational characteristics

Landsat-8 reprocessing:

• All Landsat 8 data is being reprocessed to make corrections based on first year data analysis.

• Corrections to both OLI (Operational Land Imager) and the TIRS (Thermal Infrared Sensor) data are being made including:

- all calibration parameter file updates since launch

- improved OLI reflectance conversion coefficients for the cirrus band

- improved OLI radiance conversion coefficients for all bands

- refined OLI detector linearization to decrease striping

- a radiometric offset correction for both TIRS bands

- a slight improvement to the geolocation of the TIRS data

• Approximately 90% of reprocessing is completed with estimated completion by March 30, 2014.

Landsat: Continuing the Legacy

April 1, 2021: Five decades ago, NASA and the U.S. Geological Society launched a satellite to monitor Earth’s landmasses. The Apollo era had given us our first look at Earth from space and inspired scientists to regularly collect images of our planet. The first Landsat — originally known as the Earth Resources Technology Satellite (ERTS) — rocketed into space in 1972. Today we are preparing to launch the ninth satellite in the series. 132)

Each Landsat has improved our view of Earth, while providing a continuous record of how our home has evolved. We decided to examine the legacy of the Landsat program in a four-part series of videos narrated by actor Marc Evan Jackson (who played a Landsat scientist in the movie Kong: Skull Island). The series moves from the birth of the program to preparations for launching Landsat 9 and even into the future of these satellites.

Figure 177: Episode 1: Getting Off the Ground (video credit: NASA Earth Observatory)

The soon-to-be-launched, Landsat-9 is the intellectual and technical successor to eight generations of Landsat missions. Episode 1 answers the “why?” questions. Why did space exploration between 1962 and 1972 lead to such a mission? Why did the leadership of several U.S. government agencies commit to it? Why did scientists come to see satellites as important to advancing earth science? In this episode, we are introduced to William Pecora and Stewart Udall, two men who propelled the project forward, as well as Virginia Norwood, who breathed life into new technology.

Figure 178: Episode 2: Designing for the Future (video credit: NASA Earth Observatory)

The early Landsat satellites carried a sensor that could “see” visible light, plus a little bit of near-infrared light. Newer Landsats, including the coming Landsat 9 mission, have two sensors: the Operational Land Imager (OLI) and the Thermal Infrared Sensor (TIRS). Together they observe in visible, near-infrared, shortwave-infrared, and thermal infrared wavelengths. By comparing observations of different wavelengths, scientists can identify algal blooms, storm damage, fire burn scars, the health of plants, and more.

Episode 2 takes us inside the spacecraft, showing how Landsat instruments collect carefully calibrated data. We are introduced to Matt Bromley, who studies water usage in the western United States, as well as Phil Dabney and Melody Djam, who have worked on designing and building Landsat-9. Together, they are making sure that Landsat continues to deliver data to help manage Earth’s precious resources.

Figure 179: Episode 3: More Than Just a Pretty Picture (video credit: NASA Earth Observatory)

The Landsat legacy includes five decades of observations, one of the longest continuous Earth data records in existence. The length of that record is crucial for studying change over time, from the growth of cities to the extension of irrigation in the desert, from insect damage to forests to plant regrowth after a volcanic eruption. Since 2008, that data has been free to the public. Anyone can download and use Landsat imagery for everything from scientific papers to crop maps to beautiful art.

Episode 3 explores the efforts of USGS to downlink and archive five decades of Landsat data. We introduce Mike O’Brien, who is on the receiving end of daily satellite downloads, as well as Kristi Kline, who works to make Landsat data available to users. Jeff Masek, the Landsat 9 project scientist at NASA, describes how free access to data has revolutionized what we are learning about our home planet.

Figure 180: Episode 4: Plays Well With Others (video credit: NASA Earth Observatory)

For the past 50 years, Landsat satellites have shown us Earth in unprecedented ways, but they haven’t operated in isolation. Landsat works in conjunction with other satellites from NASA, NOAA, and the European Space Agency, as well as private companies. It takes a combination of datasets to get a full picture of what’s happening on the surface of Earth.

In Episode 4, we are introduced to Danielle Rappaport, who combines audio recordings with Landsat data to measure biodiversity in rainforests. Jeff Masek also describes using Landsat and other data to understand depleted groundwater.

1) Bill Ochs, “Status of the Landsat Data Continuity Mission,” Landsat Science Team Meeting, Boise, Idaho, June 15-17, 2010, URL:

2) Brian L. Markham, Philip W. Dabney, James C. Storey, Ron Morfitt, Edward J. Knight, Geir Kvaran, Kenton Lee, “Landsat Data Continuity Mission Calibration and Validation,” Proceedings of the Pecora 17 Memorial Remote Sensing Symposium, Denver, CO, USA, Nov. 18-20, 2008

3) Brian Markham, “Landsat Data Continuity Mission: Overview and Status,” 10th Annual JACIE ( Joint Agency Commercial Imagery Evaluation) Workshop, March 29-31, 2011, Boulder CO, USA, URL:

4) J. R. Irons, J. G. Masek, “Requirements for a Landsat Data Continuity Mission,” PE&RS, Vol. 72, No 10, Oct. 2006, pp. 1102-1108

5) “Landsat Data Continuity Mission,” NASA/GSFC,


7) J. D. McCuistion, C. D. Wende, J. R. Irons, “Landsat Data Continuity Mission: Creating a Unique Government-Industry Partnership for Global Research,” Proceedings of IGARSS 2003, Toulouse, France, July 21-25, 2003

8) J. R. Irons, N. J. Speciale, et al., “Data Specifications for the Landsat Data Continuity Mission,” Proceedings of IGARSS 2003, Toulouse, France, July 21-25, 2003

9) Landsat Data Continuity Mission (LDCM) Implementation Phase - Data Specification,” Jan. 6, 2003, URL:

10) James Verdin, “The requirement for high quality data and information for Science,” 10th Annual JACIE ( Joint Agency Commercial Imagery Evaluation) Workshop, March 29-31, 2011, Boulder CO, USA, URL:

11) “Landsat Data Continuity Mission,” USGS, URL:

12) J. Irons, “New Landsat Data Continuity Mission (LDCM) Memorandum from OSTP,” The Earth Observer, NASA/GSFC, Vol. 18, Issue 1, January-February 2006, pp. 4-5

13) “A Landsat Timeline,” NASA/GSFC, URL:

14) “NASA, USGS Begin Work on Landsat 9 to Continue Land Imaging Legacy,” NASA, April 16, 2015, URL:

15) Cynthia M. O'Carroll, ”NASA Awards Letter Contract for Landsat 9 Imager-2,” NASA, Dec. 31, 2015, URL:

16) “NASA selects contractor for Landsat Data Continuity Mission spacecraft,” April 22, 2008, URL:

17) B. Berger, “Price the Deciding Factor in General Dynamics' LDCM Win,” Space News, May 5, 2008, p. 6

18) Bill Anselm, “LDCM Spacecraft Status,” June 23, 2009, URL:

19) Mike Wulder, Landsat Science Team, “Landsat Data Continuity Mission and Beyond,” 4th Global Vegetation Workshop, Missoula, MT, USA, June 16-19, 2009


21) James R. Irons, John L. Dwyer, “An overview of the Landsat Data Continuity Mission (LDCM),” Proceedings of the SPIE, Volume 7695, 2010, pp. 769508-769508-7, 'Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XVI,' edited by Sylvia S. Shen, Paul E. Lewis, Paul, April 5-9, 2010, Orlando, FLA, USA,

22) Brian Markham, James Irons, Philip Dabney, “Landsat Data Continuity Mission,” Proceedings of IGARSS (International Geoscience and Remote Sensing Symposium), Vancouver, Canada, July 24-29, 2011

23) Kate Ramsayer, “Landsat Looks to the Moon,” NASA, July 11, 2014, URL:

24) USGS, March 2, 2012, URL:

25) Steve Cole, George Diller, Rani Gran, “NASA Launches New Earth Observation Satellite to Continue 40-Year Legacy ,” NASA News Release 13-040, Feb. 11, 2013, URL:

26) NASA Selects Launch Services Provider for Earth Imagery Satellite,” Oct. 3, 2007, URL:

27) Laurie M Mann, Susan M. Good, Ann M. Nicholson, Mark A. Woodard, “Landsat Data Continuity Mission (LDCM) Safe Operations Ascent Design,” Proceedings of SpaceOps 2012, The 12th International Conference on Space Operations, Stockholm, Sweden, June 11-15, 2012

28) ”Anatomy of Landsat 8, USGS satellite,” USGS, NASA, 13 June 2017, URL:

29) ”Landsat Data Continuity Mission Overview,” NASA, 22 May 2013, URL:

30) ”What’s Behind California’s Surge of Large Fires?,” NASA Earth Observatory, Image of the Day for 5 October 2021, URL:

31) ”Lake Powell Reaches New Low,” NASA Earth Observatory, Image of the Day for 24 September 2021, URL:

32) ”Sudbury Impact Structure,” NASA Earth Observatory, Image of the Day for 20 September 2021, URL:

33) ”Fire Encroaching on Giant Sequoias,” NASA Earth Observatory, Image of the Day for 17 September 2021, URL:

34) ”Hurricane Ida Leaves a Trail of Oil,” NASA Earth Observatory, Image of the Day for 12 September 2021, URL:

35) ”Lake Mead Drops to a Record Low,” NASA Earth Observatory, Image of the Day for 30 August 2021, URL:

36) ”Birthplace of a Hidden Figure,” NASA Earth Observatory, Image of the Day for 26 August 2021, URL:

37) ”Earthquake in Haiti Triggers Landslides,” NASA Earth Observatory, Image of the Day for 17 August 2021, URL:

38) ”Fire Consumes Large Swaths of Greece,” NASA Earth Observatory, Image of the Day for 12 August 2021, URL:

39) ”Sizing Up How Agriculture Connects to Deforestation,” NASA Earth Observatory, Image of the Day for 11 August 2021, URL:

40) ”Fires Rage in Turkey,” NASA Earth Observatory, Image of the Day for 4 August 2021, URL:

41) ”Sweden’s Siljan Ring,” NASA Earth Observatory, Image of the Day for 21 July 2021, URL:

42) ”Algae Abound Along Florida Coast,” NASA Earth Observatory, Image of the Day for 19 July 2021, URL:

43) ”An Abundance of Aquaculture in Andhra Pradesh,” NASA Earth Observatory, 17 July 2021, URL:

44) ”Autotrophs Abound in Arctic Waters,” NASA Earth Observatory, Image of the Day for 11 July 2021, URL:

45) ”Anticipating Future Sea Levels,” NASA Earth Observatory, Image of the Day for 8 July 2021, URL:

46) ”The Ancient Barberton Makhonjwa Mountains,” NASA Earth Observatory, Image of the Day for 6 July 2021, URL:

47) ”Fagradalsfjall Continues to Erupt,” NASA Earth Observatory, Image of the Day for 3 July 2021, URL:

48) ”Violent Formation for Mistastin Lake,” NASA Earth Observatory, Image of the Day for 30 June 2021, URL:

49) ”Yukon-Kuskokswim in Colorful Transition,” NASA Earth Observatory, Image of the Day for 21 June 2021, URL:

50) ”Finding Gold Mining Hotspots in Peru,”NASA Earth Observatory, Image of the Day for 16 June 2021, URL:

51) ”The Large Footprint of Small-Scale Mining in Ghana,” NASA Earth Observatory, Image of the Day for 15 June 2021, URL:

52) ”Monitoring the Collapse of Kelp Forests,” NASA Earth Observatory, Image of the Day for 4 June 2021, URL:

53) ”Detecting Gold Mining in Ghana,” NASA Earth Observatory, 2 June 2021, URL:

54) ”Arizona’s Meteor Crater,” NASA Earth Observatory, Image of the Day for 31 May 2021, URL:

55) ”A Dusty Day in Patagonia,” NASA Earth Observatory, Image of the Day for 29 May 2021, URL:

56) ”Point Roberts,” NASA Earth Observatory, Image of the Day for 28 May 2021, URL:

57) Overwintering Fires on the Rise,” NASA Earth Observatory, Image of the Day for 21 May 2021, URL:

58) ”Discovering the Charlotte Whale,” NASA Earth Observatory, Image of the Day for 19 May 2021, URL:

59) ”Lava Flows From Fagradalsfjall,” NASA Earth Observatory, Image of the Day for 13 May 2021, URL:

60) ”As Jakarta Grows, So Do the Water Issues,” NASA Earth Observatory, Image of the Day for 12 May 2021, URL:

61) Saint Petersburg Keeps the Sea at Bay,” NASA Earth Observatory, Image of the Day for 10 May 2021, URL:

62) ”A Curious Case of Clouds in Iceland,” NASA Earth Observatory, Image of the Day for 7 May 2021, URL:

63) ”Bothnian Bay Before the Breakup,” NASA Earth Observatory, Image of the Day for 4 May 2021, URL:

64) ”Hingol National Park,” NASA Earth Observatory, Image of the Day for 01 May 2021, URL:

65) ”Rock and Glass Shards Blanket La Soufriére,” NASA Earth Observatory, 29 April 2021, URL:

66) ”China’s Red Rocks and Rainbow Ridges,” NASA Earth Observatory, Image of the Day for 28 April 2021, URL:

67) Sofie Bates, ”Downstream Consequences: How NASA Satellites Track Harmful Algal Blooms,” NASA Feature, 20 April 2021, URL:

68) ”Eruption at La Soufrière,” NASA Earth Observatory, Image of the Day for 14 April 2021, URL:

69) ”A Thousand Islands in South Korea,” NASA Earth Observatory, Image of the Day for 3 April 2021, URL:

70) ”Historic Floods in New South Wales,” NASA Earth Observatory, Image of the Day for 25 March 2021, URL:

71) ”A Fast-Changing Delta in China,” NASA Earth Observatory, Image of the Day for 24 March 2021, URL:

72) Kate Ramsayer, ”Landsat Satellite Data Warns of Harmful Algal Blooms,” NASA Feature, 22 March 2021, URL:

73) ”Dry Country of Turquoise,” NASA Earth Observatory, Image of the Day for 17 March 2021, URL:

74) ”Deforestation in Papua,” NASA Earth Observatory, Image of the Day for9 March 2021, URL:

75) ”Blue-green Algae at Lake Burrinjuck,” NASA Earth Observatory, Image of the Day for 8 March 2021, URL:

76) ”Breakup at Brunt,” NASA Earth Observatory, Image of the Day for 3 March 2021, URL:

77) ”A Deadly Debris Flow in India,” NASA Earth Observatory, Image of the Day for 25 February 2021, URL:

78) ”From Russia with Questions,” NASA Earth Observatory, Image of the day for 23 February 2021, URL:

79) ”A Watery Day for Lake Lefroy,” NASA Earth Observatory, Image of the Day for 15 February 2021, URL:

80) ”A Short Journey to the Center of the Earth,” NASA Earth Observatory, Image of the Day for 12 February 2021, URL:

81) ”Trading Surfboards for Snowboards,” NASA Earth Observatory, Image of the Day for 9 February 2021, URL:

82) ”Eloise Floods Mozambique,” NASA Earth Observatory, Image of the Day for 3 February 2021, URL:

83) ”Gold Mining in Russia ’s Central Aldan Ore District,” NASA Earth Observatory, Image of the Day for 30 January 2021, URL:

84) ”Brunt Breaking Up with Antarctica this Year?,” NASA Earth Observatory, Image of the Day for 20 January 2021, URL:

85) ”Curious Clouds in the Transantarctic Mountains,” NASA Earth Observatory, Image of the Day for 19 January 2021, URL:

86) ”Rising Seas in Charleston,” NASA Earth Observatory, Image of the Day for 11 January 2021, URL:

87) ”An Outburst from Popocatépetl,” NASA Earth Observatory, Image of the Day for 7 January 2021, URL:

88) Steve Cole, Kate Ramsayer, Jon Campbell, “Landsat 8 Satellite Begins Watch,” NASA Release 13-160, May 30, 2013, URL:

89) Kate Ramsayer, “NASA's Landsat Satellite Looks for a Cloud-Free View,” NASA, May 22, 2013, URL:

90) Jon Campbell, “Landsat Images Provide the Gold Standard for New Earth Applications,” USGS News rooms, May 9, 2013, URL:

91) Rebecca Moore, “A picture of Earth through time,” Google Official Blog, May 9, 2013, URL:

92) Steve Cole, “New Public Application of Landsat Images Released,” NASA, May 9, 2013, URL:

93) “LDCM Mission Updates,” NASA, URL:

94) “Landsat Thermal Sensor Lights Up from Volcano's Heat,” NASA, May 6, 2013, URL:

95) “Landsat Data Continuity Mission,” NASA, News and Features, URL:

96) “LDCM Status Update for May 2, 2013,” NASA, May 15, 2013, URL:

97) Matt Radcliff, Rob Simmon, Jesse Allen, Holli Riebeek, Paul Przyborski, “Come Fly With the Newest Landsat,” NASA Earth Observatory, URL:

98) Tom Holm, “Landsat: Building a Future on 40 Years of Success - April 14, 2013, 705 km Orbit,” 12th Annual JACIE (Joint Agency Commercial Imagery Evaluation) Workshop , St. Louis, MO, USA, April 16-18, 2013, URL:

99) “LDCM Underfly with Landsat 7,” USGS, March 29, 2013, URL:

100) Steve Cole, Jon Campbell, “First Images Released From Newest Earth Observation Satellite,” NASA, USGS, March 21, 2013, URL:

101) “A Closer Look at LDCM's First Scene,” NASA, March 21, 2013, URL:

102) “LDCM Status Update for Feb. 21,” NASA, Feb. 21, 2013, URL:

103) “NASA Completes Critical Design Review Of One Landsat Instrument,” Space Daily, May 28, 2010, URL:

104) Bryant Cramer, “USGS Perspectives on LDCM and Landsat,” Landsat Science Team Meeting,” Jan. 19-21, 2010, Mountain View, CA, USA, URL:

105) Tom Loveland, “Landsat and LDCM Status,” 2008 NASA Carbon Cycle & Ecosystems Joint Science Workshop, April 28-May 2, 2008, University of Maryland, Adelphi, MD, USA

106) James Storey, Michael Choate, Kenton Lee, “Geometric performance comparison between the OLI and ETM+,” Proceedings of the Pecora 17 Memorial Remote Sensing Symposium, Denver, Co, USA, Nov. 16-20, 2008

107) Jeanine Murphy-Morris, “Operational Land Imager ,” Landsat Science Team Meeting, Sioux Falls, SD, Jan. 8, 2008, URL:

108) Edward J. Knight, “OLI Overview and Status,” Landsat Science Team Meeting, July 15, 2008, Reston, VA, URL:

109) Bill Ochs, “Status of the Landsat Data Continuity Mission,” Landsat Science Team Meeting, July 15, 2008, Reston, VA, URL:

110) Edward J. Knight, Brent Canova, Eric Donley, Geir Kvaran, Kenton Lee, “The Operational Land Imager:Overview and Performance,” 10th Annual JACIE ( Joint Agency Commercial Imagery Evaluation) Workshop, March 29-31, 2011, Boulder CO, USA, URL:

111) Esad Micijevic, Ron Morfitt, “Operational Calibration and Validation of Landsat Data Continuity Mission (LDCM) Sensors using the Image Assembly System (IAS),” Proceedings of IGARSS (IEEE International Geoscience and Remote Sensing Symposium) 2010, Honolulu, HI, USA, July 25-30, 2010

112) Brian L. Markham, Philip W. Dabney, Edward J. Knight, Geir Kvaran, Julia A. Barsi, Jeanine E. Murphy-Morris, Jeffrey A. Pedelty, “The Landsat Data Continuity Mission Operational Land Imager (OLI) Radiometric Calibration,” Proceedings of IGARSS (IEEE International Geoscience and Remote Sensing Symposium) 2010, Honolulu, HI, USA, July 25-30, 2010, URL:

113) Brian Markham, “LDCM On-Orbit Cal/Val Considerations,” Proceedings of the Landsat Science Team Meeting, Mesa, AZ, USA, March 1-3, 2011, URL:

114) “Ball Aerospace Completes CDR For Landsat's Operational Land Imager,” Nov. 26, 2008, Spacemart, URL:

115) “NASA Completes Critical Design Review of Landsat Data Continuity Mission,” Science Daily, June 1, 2010, URL:

116) Dennis Reuter, Cathy Richardson, James Irons, Rick Allen, Martha Anderson, Jason Budinoff, Gordon Casto, Craig Coltharp, Paul Finneran, Betsy Forsbacka, Taylor Hale, Tom Jennings, Murzy Jhabvala, Allen Lunsford, Greg Magnuson, Rick Mills, Tony Morse, Veronica Otero, Scott Rohrbach, Ramsey Smith, Terry Sullivan, Zelalem Tesfaye, Kurtis Thome, Glenn Unger, Paul Whitehouse, “The Thermal Infrared Sensor on the Landsat Data Continuity Mission,” Proceedings of IGARSS (International Geoscience and Remote Sensing Symposium), Honolulu, Hawaii, USA, July 25-30, 2010, URL:

117) Ramsey L. Smith, Kurtis Thome, Cathleen Richardson, James Irons, Dennis Reuter, “Terrestrial Applications of the Thermal Infrared Sensor, TIRS,” URL:

118) M. Jhabvala, D. Reuter, K. Choi, C. Jhabvala, M. Sundaram, “QWIP-Based Thermal Infrared Sensor for the Landsat Data Continuity Mission,” Proceedings of the QSIP (Quantum Structure Infrared Photodetector) 2009 International Conference, January 18-23, 2009, Yosemite, CA

119) M. Jhabvala, D. Reuter, K. Choi, M. Sundaram, C. Jhabvala, A. La, A. Waczynski, J. Bundas, “The QWIP focal plane assembly for NASA's Landsat Data Continuity Mission,” Proceedings of the SPIE, 'Infrared Technology and Applications XXXVI,' edited by Bjørn F. Andresen, Gabor F. Fulop, Paul R. Norton, Volume 7660, April 5-9, 2010, Orlando, FLA, USA, pp. 76603J-76603J-13, doi:10.1117/12.862277

120) M. Jhabvala, K. K. Choi, C. Monroy, A. La, “Development of a 1 k × 1 k, 8–12 µm QWIP array,” Infrared Physics & Technology, Volume 50, Issues 2-3, April 2007, pp. 234-239

121) K. Thome, D. Reuter, A. Lunsford, M. Montanaro, R. Smith, Z. Tesfaye, B. Wenny, “Calibration overview for the Thermal Infrared Sensor (TIRS) on the LandsatData Continuity Mission,” 10th Annual JACIE ( Joint Agency Commercial Imagery Evaluation) Workshop, March 29-31, 2011, Boulder CO, USA, URL:

122) K. Thome, D. Reuter, A. Lunsford, M. Montanaro, R. Smith, Z. Tesfaye, B. Wenny, “Calibration of ther Thermal Infrared Sensor on the Landsat Data Continuity Mission,” Proceedings of IGARSS (International Geoscience and Remote Sensing Symposium), Vancouver, Canada, July 24-29, 2011

123) Jason Budinoff, Konrad Bergandy, Joseph Schepis, Adam Matuszeski, Richard Barclay, “Development of the Scene Select Mechanism for the Thermal Infrared Sensor Instrument,” Proceedings of the 14th European Space Mechanisms & Tribology Symposium – ESMATS 2011, Constance, Germany, Sept. 28–30 2011 (ESA SP-698)

124) Mary Pagnutti, Robert E. Ryan, Kara Holekamp, “Landsat Data Continuity Mission and Sentinel-2 Multi-Spectral Instrument Image Product Simulations for Sensor Comparisons and Data Fusion Research,” Proceedings of the 11th Annual JACIE (Joint Agency Commercial Imagery Evaluation ) Workshop, Fairfax, VA, USA, April 17-19, 2012, URL:

125) James Nelson, Robert Patschke, Howard Garon, Alan Ames, Claire Mott, Grant Mah, Jason Williams, James Joseph, “Landsat Data Continuity Mission (LDCM) Space to Ground Mission Data Architecture,” Proceedings of the 2012 IEEE Aerospace Conference, Big Sky, Montana, USA, March 3-10, 2012

126) Landsat Data Continuity Mission (LDCM), Ground System (GS) Integration and Test Plan,” USGS, LDCM-I&T-001, Version 1.1, September 2009, URL:
LDCM-I&T-001_GS%20Integration% 20&%20Test%20Plan_v1.1.pdf

127) Jonathan Gal-Edd, “LDCM Ground System - Network Lessons Learned,” SOSTC GSFC May 24-25, 2010, URL:

128) Dave Hair , Doug Daniels, “Landsat Data Continuity Mission (LDCM) USGS Project Status Report,” Proceedings of the Landsat Science Team Meeting, Mesa, AZ, USA, March 1-3, 2011, , URL:

129) Susan M. Good, Ann M. Nicholson, Mark A. Woodard, “Landsat Data Continuity Mission (LDCM) Flight Dynamics System (FDS),” Proceedings of SpaceOps 2012, The 12th International Conference on Space Operations, Stockholm, Sweden, June 11-15, 2012

130) G. R. Mah, H. Garon, C. Mott, M. O'Brien, “Ground System Architectures Workshop 2014, Landsat 8 Test as You Fly, Fly as You Test,” Proceedings of GSAW 2014 (Ground System Architectures Workshop), Los Angeles, CA, USA, Feb. 24-27, 2014, URL:

131) Del Jenstrom, “Status of the Landsat Data Continuity Mission,” Proceedings of the Landsat Science Team Meeting, Mesa, AZ, USA, March 1-3, 2011

132) Matthew Radcliff, Mike Carlowicz,”Landsat: Continuing the Legacy,” NASA Earth Observatory, 1April 2021, URL:

The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: ”Observation of the Earth and Its Environment: Survey of Missions and Sensors” (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates (

Spacecraft    Launch    Mission Status    Sensor Complement    Ground Segment    References    Back to top