Minimize FormoSat-3

FormoSat-3 / COSMIC (Constellation Observing System for Meteorology, Ionosphere and Climate)

Overview    Spacecraft    Launch    Mission Status    Sensor Complement   Ground Segment   References

The ROCSat-3/COSMIC (Republic of China Satellite-3 / Constellation Observing System for Meteorology, Ionosphere and Climate) is an international collaborative project between NSPO (National Space Program Office) of Taiwan and UCAR (University Corporation for Atmospheric Research) of the United States of America. Initiated in December 1997, the project will launch a LEO constellation of six microsatellites to collect atmospheric remote sensing data for operational weather prediction, climate, ionospheric (space weather monitoring), and geodesy research. NSPO is the prime sponsor and owner of the satellites. UCAR, located at NCAR in Boulder, CO, is primarily sponsored by NSF (National Science Foundation). Other partners in the project include JPL, NRL, USAF, NOAA, CWB (Central Weather Bureau of Taiwan), industry from both countries, universities, and other research organizations from the US, Taiwan, and other countries. 1) 2)

The overall objective of ROCSat-3/COSMIC is to extend the low-cost research approach of refractive GPS radio occultation measurements (to derive important weather and climate research parameters, including atmospheric temperature, moisture, and pressure), that began with the GPS/MET instrument on Microlab-1 (launch April 3, 1995), to the next step by testing the ability of a constellation of six "ROCSat-3/COSMIC microsatellites with GPS/MET heritage" to provide the data needed to fully evaluate the impact of this promising new observational tool.

A goal is also to demonstrate the utility of atmospheric/ionospheric limb soundings in operational weather prediction, space weather monitoring and space geodesy. In addition to carrying an advanced version of the JPL-developed GPS receiver for occultation measurement, each satellite will carry two tiny, simple secondary instruments (tri-band-beacon and photometer) which synergistically enhance the accuracy and utility of the ionospheric observations. A global data collection network and operations center will process space and ground observations and deliver products to users in real-time for operational impact studies. 3) 4) 5) 6) 7) 8)

Note: A public naming competition took place in Taiwan in 2004 with regard to the ROCSat satellite program. At the end of this contest, the ROCSat program was given the new name of FormoSat in December 2004. Hence, the ROCSat-3 constellation became FormoSat-3. In USA, FormoSat-3 is known under the name of COSMIC (Constellation Observing System for Meteorology, Ionosphere and Climate).


Figure 1: Artist's illustration of FormoSat-3 / COSMIC spacecraft (image credit: NSPO)


Orbital Sciences Corporation (OSC) of Dulles, VA, USA is the prime contractor for the microsatellites, selected by NSPO. A joint team of the OSC and NSPO engineers designed and developed the satellites, with the early phase of the work performed at OSC and integration and test performed at NSPO. Several domestic industry companies of Taiwan were selected to participate in the project by providing satellite components. The following list gives an idea of the involvement: 9) 10) 11)

• Satellite computer (Acer Technologies Inc.)

• Mission interface unit (Acer Technologies Inc.)

• Solar sensor (Shihlin Electric & Engineering/eBright Corp.)

• Rechargeable storage battery (Shihlin Electric & Engineering/eBright Corp.)

• Current converter (Shihlin Electric & Engineering/eBright Corp.)

• Satellite antennas (Victory Industrial Corp.)

• Receiving coupler (Victory Industrial Corp.)

• Transmitting filter (Victory Industrial Corp.)

• Satellite heating elements (Yung Tien Industrial Co.)

The S/C structure is of Orbcomm heritage (MicroStar bus), a cylindrical shape of 1.03 m diameter and about 18 cm in height (width). On orbit, two solar panels deploy on each side of the satellite. All spacecraft are identical, with a mass of about 61 kg (including fuel). Each S/C features on-board propulsion to reach its final destination orbit. The ACS (Attitude Control Subsystem) provides attitude knowledge (±5º roll and yaw, ±2º pitch) with an Earth limb sensor and a magnetometer. The ACS is comprised of the ACE (Attitude Control Electronics) and a suite of sensors and actuators. The ACE, driving the core ACS algorithms and interfacing to sensors and actuators, is comprised of a 68302 processor, analog and serial I/O, and discrete signal outputs. The ACS uses two primary calculation routines, a fast routine that runs every 2 seconds and a slow routine that runs every 10 seconds. The fast routine performs control and estimation calculations while the slow routine runs the ARS (Attitude Reference System) filter gain calculations.


Figure 2: Functional block diagram of the ACS (image credit: NSPO)

Power (46 W) is provided by a solar array and 10 Ah batteries. The propulsion system consists of two tanks to store propellant (hydrazine) and 4 small monopropellant thrusters located in the four quadrants of the x-z plane. The design life is 2 years (5 years expendables).

RF communications: The communication subsystem consists of an S-band receiver, an L-band transmitter, and a set of S-band and L-band antennas.


Figure 3: View of the stacked FormoSat-3/COSMIC spacecraft (image credit: NSPO, UCAR) 12)

Spacecraft mass

416 kg for all six spacecraft


Single string

Solar array

207 W EOL, single axis articulated


3-axis, Zero Momentum Bias, nadir and target pointing


<1.7º control, <1.4º knowledge

Payload data storage

128 MB

Uplink, downlink

S-band: 32 kbit/s, S-band: 2 Mbit/s

Propulsion capability

Monopropellant blowdown system, 6.7 kg hydrazine

Table 1: Spacecraft specifications


Figure 4: Inside view of the FormoSat-3/COSMIC spacecraft (image credit: NRL, UCAR) 13)


Figure 5: Photo of six FormoSat-3/COSMIC satellites mated to the Minotaur launch vehicle (image credit: Orbital ATK)


Launch: The FormoSat-3/COSMIC constellation of 6 identical spacecraft was launched on April 15, 2006 (UTC) from VAFB, CA on Minotaur-1 vehicle. The low-cost nature of the mission required a special launch and deployment design of the constellation: 14) 15)

• All six spacecraft were launched in a single shot by a Minotaur vehicle of OSC into one orbital plane. Minotaur has the capability to lift the six satellites into an initial circular parking orbit of about 500 km altitude with an inclination of 72º.

• This is being followed by a 13 month constellation deployment/distribution sequence. Each satellite will be separated from the launch vehicle individually. After in-orbit checkout, each satellite will be boosted by on-board thrusters to different altitudes ranging up to 800 km. The corresponding different rates of orbit nodal precession will then gradually drift the orbit planes apart, until a more-or-less even distribution of six orbit planes is achieved. The satellites will already start collecting atmospheric soundings during the orbit-adjustment and constellation distribution period. 16)

• The spacecraft deployment occurs in an inverse sequence (6th satellite first). The total deployment time is estimated to be 387 days. During the transition phase, there are 121 days for the 5th and 4th satellites in tandem flight, and 198 days for the 3rd and 2nd satellites in tandem flight. These temporary configurations are being used for the gravity study to determine the high order harmonics of the geopotential from the GPS data.


Figure 6: Spacecraft distribution in orbit after a series of initial orbit adjustments in 2006 (image credit: NSPO)

Orbit of constellation: Circular orbits, altitudes of 800 km, inclinations of 72º; there are 6 operational planes with 1 satellite per plane, spaced 24º apart.


Figure 7: Illustration of the FormoSat/COSMIC spacecraft constellation (image credit: Orbital ATK, Ref. 11)



Mission status:

• August 29, 2016: Due to the success of the FormoSat-3/COSMIC-1 constellation, U.S. agencies and Taiwan have decided to move forward with a follow-on RO mission, FormoSat-7/COSMIC-2, that will launch six satellites into low-inclination orbits in 2017, and another six satellites into high-inclination orbits in 2020. U.S. agencies, lead by NOAA (National Oceanic and Atmospheric Administration) are now actively partnering with Taiwan's NSPO (National Space Organization) to execute the FormoSat-7/COSMIC-2 program. 17)

• April March 9, 2016: FormoSat-3/COSMIC is the first large space collaboration program between Taiwan and the U.S. The program employed 6 microsatellites formed in global constellation that utilize the RO (Radio Occultation) technique to receive the L-band signals from the GPS satellites. The retrieved GPS RO signals then processed into the useful weather data to form the near-real time global weather observation network. The FormoSat-3 constellation will reach the 10th Anniversary since its launch on April 15, 2006. The accumulated retrieved RO soundings have exceeded 10 million profiles (6 million for the atmosphere and 4 million for the ionosphere) in the past decade. These RO data have been continuously provided to more than 3,000 registered users in 83 countries. The GPS-RO data has been demonstrated to be valuable to the climate, meteorology, and space weather communities. 18)

• Nov. 2015: The FormoSat-3/COSMIC constellation is operating. The contribution of FormoSat-3/COSMIC to weather prediction is considered to be "significant" by the Taiwan CWB and U.S. NOAA/NWS (National Weather Service) and represents an immense benefit to worldwide forecasting capability.

• Oct. 2014: Five of the six spacecraft (the FM-3 spacecraft is dead since Aug. 2010) of the FormoSat-3/COSMIC constellation are still operating, eight years after launch (mission life: 5 years) and continue to provide 1,500-2,000 GPS radio occultation soundings/day. Registered users: 2766 from 79 countries. The CDAAC (COSMIC Data Analysis and Archive Center), Boulder, CO, is providing NRT (Near Realtime) data to NWPs (Numerical Weather Prediction), AFWA (Air Force Weather Agency) of USAF. 19)


Figure 8: About 4.5 million FormoSat/COSMIC profiles were provided in realtime in the period Apr. 21, 2006 to Sept. 21, 2014 (image credit: UCAR COSMIC Program)

• In May 2013, FORMOSAT-3 / COSMIC is providing ~ 1600 daily RO (Radio Occultation) profiles in the neutral atmosphere, ~1300 daily electron density profiles even after near 7 years performance in orbit. Both CDAAC (COSMIC Data Analysis and Archival Center ) in the U.S. and TACC (Taiwan Analysis Center for COSMIC) in Taiwan have been processing the GPS RO data in near real time, and > 90% of RO profiles are delivered to operational weather centers within 3 hours of observation. The value of utilizing the FORMOSAT-3 / COSMIC GPS RO data have been demonstrated to be eminent among the weather and space weather communities. 20)

• The FormoSat-3/COSMIC constellation is operating "nominally" in 2012 as an experimental "Research Mission" for demonstrating the usefulness of GPS-RO (Global Positioning System - Radio Occultation) in operational numerical weather prediction, climate monitoring, and space weather forecasting. 21)

- The constellation has proven to increase the accuracy of the predictions of hurricane/typhoon/cyclone behavior, significantly improve long-range weather forecasts, and monitor climate change with unprecedented accuracy. The success of the FormoSat-3/COSMIC mission has initiated a new age for operational GPS-RO soundings.

- The FormoSat-3mission has retrieved 2.7 million soundings, serving more than 1,600 registered users from 57 countries. On average, FormoSat-3/COSMIC provides 1,500 ~ 2,000 GPS-RO soundings per day, uniformly distributed around the globe. Approximately 90% of the data are available within 3 hours of observations to support operational numerical weather prediction. Several worldwide weather operation centers have ingested FormoSat-3 data into their operational weather forecast models and forecasting abilities have been significantly improved. These include the NCEP (National Centers for Environmental Prediction) of NOAA, ECMWF, EUMETSAT, CWB (Central Weather Bureau) of Taiwan, UKMO, JMA, US AFWA, Canada Met, Météo-France, the Bureau of Meteorology of Australia, and others.

- The FormoSat-3/COSMIC constellation has successfully improved global weather analyses and predictions, the accuracy of weather prediction models, and the understanding of tropical, mid-latitude and polar weather systems and their interactions (Ref. 21).

• The FormoSat-3 / COSMIC (simply referred to as FS-3/C) constellation is operating "nominally" in 2011 (Ref. NO TAG#. The satellite system is providing global data in near real-time to over 1,000 users worldwide, including NOAA. 22) 23)

- On April 15, 2011, FS-3/C celebrated its fifth anniversary on orbit. This is a major milestone. COSMIC has accomplished all the science objectives it set out to do, and has convincingly demonstrated the power of GPS radio occultation for research and operations. 24)

- In 2011, FS-3/C is providing up to 2,000 daily RO profiles in the neutral atmosphere, electron density profiles and total electron content arcs in the ionosphere, GPS scintillation observations, and TIP radiance products. The data have already demonstrated their value for operational weather forecasting, hurricane forecasting, and investigations of the atmospheric boundary layer. The data have been used extensively to test ionospheric models and their use in operational space weather models is under development. COSMIC GPS-RO data also have the potential to be of great benefit to climate studies due to their demonstrated high precision and global and diurnal sampling coverage. 25)

- The FS-3/C satellites have performed successfully for over 4 years now. It is not a perfect constellation for an operational system, but it has achieved more than satisfactory results for an experimental system operating in a semi-operational manner. The FS-3/C satellites are degrading as anticipated; however, NSPO assesses these satellite will continue to operate into the next few years. All six satellites have experienced some anomalies in the electric power subsystem and/or payload instrument performance causing onboard electronic power shortages and payload duty-cycle reduction. The SOCC and the operation team used operational methods to reduce the impacts of the anomalies and increase the payload data output. - The success of the FS-3/C mission has initiated a new era for near real-time operational GNSS-RO soundings. NSPO is committed to continuing the FS-3/C satellite constellation operation to collect RO data to minimize the data gap duration. 26)

NSPO and NOAA will proceed with the follow-on FormoSat-7 / COSMIC-2 (or simply FS-7/C-2) joint mission implementation - a 12 spacecraft mission which is in the definition phase in the 2010-2011 timeframe with a first planned launch in 2014.

• Performances overview of the mission in September 2010 (Ref. 46) :

- The FormoSat-3 / COSMIC satellites have performed successfully for over 4 years now. It's not a perfect constellation for the operational system, but it has achieved more than satisfactory expectation for an experimental system in an semi-operational usage.

- The FormoSat-3 / COSMIC satellites are degrading as anticipated; however, NSPO assesses these satellites will continue to operate into the next few years.

- While NSPO and NOAA are proceeding to the FormoSat-7 / COSMIC -2 joint mission implementation, NSPO is committed to continue the FormoSat-3 / COSMIC satellite constellation operation to collect the RO (Radio Occultation) data to minimize the data gap duration. 27)

• In February 2010, observations from MetOp/GRAS and GRACE-A were added to the FS-3/C operational observing system. 28)


Figure 9: FormoSat-3/COSMIC spacecraft operation status in September 2010 (image credit: NSPO, NOAA, UCAR, Ref. 46)

• The FormoSat-3 constellation is operating nominally in 2009.

• In January 2009, the status of the 6 satellite constellation remains largely unchanged. Almost three years into the project operations are going well with over 3.2 million quality checked neutral atmospheric (1.5 M) and ionospheric (1.7 M) profiles delivered. Depending on the state of the constellation, between 1400 - 2300 good soundings/day are being obtained. 29)

Two spacecraft (FM2 and FM3) continue to have power problems but show no sign of further degradation. The issue of the signal-to-noise ratio (SNR) drops from some of our antennas still remains unsolved. A detailed study trying to correlate these SNR drops with a multitude of spacecraft parameters such as solar angle, receiver temperatures, power conditions etc. did not lead to any conclusive explanation so far.

• Five out of the six satellites in the constellation had reached their final mission orbits of 800 km as of November 2007. 30)

• The constellation is on the way to its final orbit configuration. All satellites remain healthy except spacecraft flight model No 2 (FM2) with a problem of power shortage and FM3 currently staying at an orbit of 711 km due to a mechanism issue to be solved. The current sounding profiles retrieved from the GPS occultation measurements are over an average of 1800 daily. The sounding profiles have been used to study atmospheric and ionospheric structures and total electron content, and assimilated into numerical atmospheric and space weather predictions models to improve the accuracy of prediction. All satellites are in good health and providing initial data. 31)

As of May 31, 2008, the FormoSat-3 constellation has collected over 1,020,000 atmospheric sounding data sets and 1,270,000 ionospheric profiles. The most noticeable needs for weather model changes are being observed in the Polar Regions. The scientists have identified the errors of the Antarctica regional model and first collected the temperature structure over the Antarctica. The global and frequent nature of the FormoSat-3 constellation data collection offers a remarkable contribution to weather forecast models. - Several weather centers such as NCEP (National Centers for Environmental Prediction) of the United States, ECMWF (European Centre for Medium-Range Weather Forecasts), UKMO (UK Meteorology Office), Météo France, and JMA (Japan Meteorological Agency) have injected the assimilation data into their weather forecast models to improve the accuracy of weather prediction. Numerous case studies have shown the data make positive impacts on the path prediction of typhoons and hurricanes. - The average 135-minute data latency achieved make FORMOSAT-3/COSMIC a quasi-operational constellation. 32)

• As of Aug. 2007, the satellite constellation is approaching final deployment with only one more spacecraft, FM1, remaining its initial 500 km orbit. Presently the system is producing 1500-1700 good neutral atmospheric soundings per day with an average latency of about 2 hours. 33) 34)

• Maneuvers continue to move the satellites into their final orbits. As of Jan. 2007, FM2 and FM5 are at 800 km altitude while FM6 is at 716 km. FM1, FM3 and FM4 are still at 518 km.

• The satellites are averaging about 1,200 soundings a day, in a nearly uniform global distribution, providing independent data over vast stretches of ocean and ice where there are no weather balloons. As the satellites approach their final positions, they will increase their output to about 2,500 soundings a day. 35)

• The first set of occultation data from FormoSat-3 was obtained on April 21, 2006.

• Cosmic data became available to the public on July 28, 2006. JPL and its partners have begun processing Cosmic data into temperature and water vapor profiles of the atmosphere and measurements of the electron content of the ionosphere.


Figure 10: The deployment timeline of the FormoSat-3/COSMIC constellation (image credit: NSPO)



Sensor complement: (IGOR, TIP, CERTO/TBB)

The FormoSat-3/COSMIC constellation produces about 3000 soundings (minimum requirement of 2500/day) of bending angle and refractivity globally in all weather each day for at least one year after the spacecraft are placed in their final orbits. From these soundings, estimates of electron density in the ionosphere and temperature, water vapor and pressure in the stratosphere and troposphere will be derived. Desirable characteristics of these data include such items as: high accuracy, high vertical resolution, all weather (clouds and aerosols do not affect measurements), no calibration of instrument required, no instrument drift, require no first guess, modest cost.


Science requirement


Vertical GPS occultation resolution

0.3-1.5 km

0.3 km near surface, 1.5 km at 45 km

Horizontal GPS occultation resolution

300-600 km

Smaller for fronts with large slope

Bending angle profile

1.5 x 10-6 rad

Limited mainly by residual ionospheric error and data noise

Refractivity profile


As good as 0.2% between 10-30 km

Temperature profile

1º C

In 0-40 km height range, assuming dry air

Water vapor profile


Between 0-5 km, assuming model temp. error of 1-2º, errors between 5-10 km are larger

Geopotential height vs pressure

10-20 m

Between 0-30 km (worse at solar maximum)

Electron density profile


Profile quality near the F2 peak, can be worse elsewhere; limited by horizontal gradients

GPS TEC (Total Electron Content)

0.001 TECU (Total Electron Content Unit)

Relative TEC (absolute about 3-5 TECU)


0.003 TECU

Relative TEC (absolute about 1-3 TECU)

TIP peak for F2 density, NmF2


Performs better at high electron densities
Works only on the night side of the globe

Magnetometer measurement

10 nT

Precision at 1 Hz rate, 500 NT accuracy

Scintillation GPS


100 Hz SNR data used to determine S4

Scintillation TBB


50-1000 Hz SNR data

LEO position knowledge

about 10 cm

3-D rms position error (not critical)

LEO velocity knowledge

<0.1 mm/s

Relative velocity error for paired satellites




Table 2: Science requirements of the FormoSat-3/COSMIC constellation

Number of occultation measurements

About 2500 soundings/day on a global scale

L1/L2 phase measurements

About 2 mm (10 s interval for precise orbit determination)

GPS phase sampling rate

0.1-50 Hz

GPS vertical range neutral atmosphere

Surface to 60 km (50 Hz sampling)

GPS vertical range ionosphere

90-800 km (10 Hz sampling)

TBB phase measurement

<32 mm at 150 MHz (ground receiver)

TBB sampling rate

>50 Hz (ground receiver)

TIP measurement

<10% (uncertainty in photon count)

TIP footprint

125 km x 25 km (at 400 km height of F2 layer)

TIP resolution

0.1-10 s averaging


10 nT precision, 500 nT accuracy

Table 3: Observational requirements of FormoSat-3/COSMIC


IGOR (Integrated GPS Occultation Receiver):

IGOR is based on the BlackJack GPS occultation receiver design of JPL and flown on such missions as CHAMP, SAC-C, and GRACE. IGOR, built by Broad Reach Engineering of Tempe, AZ, is the primary science instrument of the FormoSat-3/COSMIC constellation. The IGOR receivers, on FormoSat-3/COSMIC will be able to track all GPS satellites in view simultaneously, including two or more occulting satellites. It will operate fully autonomously, scheduling when to track which satellites and at what sampling rate based on its own known position and those of the GPS satellites. The instrument reports high-rate (50 Hz) dual frequency carrier phase measurements on the occulting links with sub-millimeter precision for accurate, high resolution profiling. Lower rate (0.1 Hz) phase measurements of all satellites in view are being collected for precise orbit determination (POD) at the 5-10 cm level.

IGOR tracks both GPS carrier frequencies (L1, L2) to separate the frequency-dependent (dispersive) ionospheric delay from the non-dispersive refractive delay of the neutral atmosphere. A patented "semi-codeless" technique is used to obtain precise measurements of the L2 signal, both carrier phase and pseudorange, with anti-spoofing turned on. In addition to these measurements, the GPS instrument can record GPS signal amplitudes for on-orbit ionospheric scintillation monitoring and correction of signal diffraction effects in post-processing. The instrument mass is 4.6 kg; size of about 20 cm x 24 cm x 10.5 cm; power of 16 W nominal, 23 W peak, antenna inputs: 4. 36)

Note: IGOR is also referred to as GOX (GPS Occultation Experiment).


Figure 11: Illustration of the IGOR instrument (image credit: Broad Reach Engineering)

The GPS radio occultation technique is based on the following principles: As a signal travels through the atmosphere it is retarded and bent. This results in a phase and Doppler shift, which can be measured very accurately by the GPS receiver aboard the LEO FormoSat-3/COSMIC satellites. Since the transmitter and receiver positions and velocities are accurately know from precise orbit determination, the signal bending angle alpha as a function of impact parameter, can be computed from the Doppler shift observed at LEO. From the basic bending angle versus impact parameter data, vertical profiles of refractivity as a function of tangent point radius can be derived. Further analysis converts refractivity to electron density in the ionosphere.


Figure 12: Occultation scheme of GPS signals with LEO satellites (image credit: Broad Reach Engineering)

In-orbit reprogramming of IGOR: In October 2005, a newly launched GPS satellite (SVN 53, PRN 17) began broadcasting the L2C signal along with the traditional L1 and L2 signals. The term L2C refers to a bi-phase modulated carrier at the L2 frequency. Occultation measurements from orbit require the high-accuracy provided by dual-frequency signals. The IGOR GPS science instrument on FormoSat-3/COSMIC uses a code-enhanced technique to track these encrypted signals without being "keyed" but with a loss of precision relative to a keyed receiver. These techniques have all involved compromises. In particular, SNR is less using codeless processing. Unlike the P(Y) modulation present on all other GPS satellites, L2C is broadcast without encryption and thus available to civil users. Tracking L2C should benefit scientific measurements of the ionosphere and atmosphere from space due in part to the recovery of the L2 carrier phase with full Signal to Noise Ratio (SNR). 37)

The IGOR GPS receiver has software radio features that allow reprogramming of some signal processing functions (design includes an FPGA). Engineers at JPL working with scientists at JPL and UCAR remotely modified the IGOR payload on one of the COSMIC weather satellites to track the L2C signals.

On FormoSat-3/COSMIC there are four microstrip style antennas which receive L1 and L2 frequencies. Two of the antennas are high-gain arrays directed fore and aft towards the Earth's limb. These are intended for atmospheric occultations. The other two antennas provide a wide field of view for precise orbit determination and ionospheric science.


Figure 13: IGOR receiver modification (image credit: JPL, UCAR)

The IGOR is programmed to simultaneously track up to 16 GPS satellites while processing amplitude range and phase from L1CA, L1P and L2P signals. Each L1P and L2P tracking loop is primarily guided by the L1CA tracking loop allowing narrower loop bandwidths on those lower SNR signals. The FPGA and processor sections were modified (Figure 13) to process up to two L2C signals simultaneously. These can be enabled only for PRNs 12, 17 & 31 and only during setting occultation observations. The L2C tracking loop was configured to be fully independent from the L1CA loop with the same loop bandwidth (unlike L1P and L2P).

This receiver modification can be commanded ON/OFF from the ground and during the evaluation periods with COSMIC FM2 data from PRNs 12, 17 & 31 was not transferred to weather centers by UCAR.


Figure 14: 1 second SNR for L1CA, L2P and L2C during atmospheric occultation from COSMIC FM1 on May 24, 2008 (image credit: JPL)

The initial data of Figure 14 show that L2C is eminently useable as a signal for spaceborne science receivers in general and atmospheric occultations specifically. In particular, the higher SNR available from using L2C phase, relative to the encrypted P2(Y2), give much more robust and precise observables. In addition L2C occultation data provided bending and refractivity profiles consistent with the large scale structure of L1CA profiles.


TIP (Tiny Ionosphere Photometer):

TIP is a nadir-viewing instrument, designed and built by NRL (Naval Research Laboratory), Washington, DC and Praxis Inc. TIP and TBB provide measurements of electron density, an important parameter of the upper atmosphere. The readings of TIP and TBB complement the primary IGOR instrument so that 3-D fields of electron density gradients between 90 and 750 km can be inferred.

TIP is a compact, narrow-band, ultraviolet photometer operating at the 135.6 nm wavelength (UV radiation). This emission is produced by recombination of O+ ions and electrons, which is the natural decay process for the ionosphere. At night, the strength of the emission is proportional to the product of the square of the peak electron density; during the daytime the emission is dominated by photoelectron impact excitation of atomic oxygen and is not useful for ionospheric sensing. 38) 39) 40) 41)


Figure 15: TIP electronics module (top) and sensor module (image credit: NRL)

In particular TIP provides horizontal gradients in ionospheric electron density at the peak of the F2 layer, along the satellite orbit track. TIP measures the naturally occurring nighttime emission of neutral oxygen at 135.6 nm. This emission (airglow) is produced by the recombination of O + ions and electrons and is proportional to the square of the electron density in the ionospheric F region. Since horizontal gradients of electron density are a limiting error source for occultation inversions in the ionosphere, the combined analysis of TIP and GPS data promises improved retrievals of nighttime ionospheric profiles.

TIP is nadir-pointing with a 3.8º circular FOV providing a 30 km horizontal resolution from an orbital altitude of 800 km. The TIP sensor module consists of:

• Photomultiplier tube observing UV light

• Strontium fluoride filter passes 131-160 nm emissions

• Very high sensitivity ~150 counts/s/Rayleigh


Figure 16: Illustration of the TIP filter wheel (image credit: NRL)


CERTO/TBB (Coherent Electromagnetic Radio Tomography/Triband Beacon Transmitter):

CERTO/TBB was designed and built at NRL. CERTO/TBB transmits phase data measurements at 150, 400 and 1067 MHz (VHF, UHF, L-band) which can be received at ground stations worldwide. These data are converted to line-of-sight TEC (Total Electron Content) observations that can be processed with 2-dimensional ionospheric tomography techniques. CERTO/TBB data can also be combined with the other ionospheric observations in tomographic and physical data assimilation models to compute global four-dimensional electron density fields. 42) 43) 44) 45)

The FormoSat-3/COSMIC instrument suite permits three-dimensional tomography of the ionosphere with unprecedented resolution and accuracy. FormoSat-3/COSMIC data will be highly complementary to other satellite sounding systems, including radiometric sounders on the POES and GOES series satellites of NOAA. The independence and the high-vertical resolution of the radio occultation soundings complement the high horizontal resolution of the radiometric soundings and together the two systems can likely be combined to yield composite soundings of temperature and water vapor with unprecedented accuracy, horizontal and vertical resolution, and global coverage.


Figure 17: Photo of the CERTO instrument (image credit: NRL)


Figure 18: CERTO/TBB accommodation on FormoSat-3/COSMIC (image credit: NRL)


Figure 19: Joint CERTO/TBB, GPS-GOX, TIP operations on FormoSat-3/COSMIC (image credit: NRL)



Ground segment:

When FS-3/C was launched, ground station support was contracted with the USN (Universal Space Network) through their stations at Poker Flats, Alaska and Kiruna, Sweden. USN performed very well for 2 years, but in an effort to reduce operational costs NOAA made a decision to employ indigenous resources. NOAA assets were established for FS-3/C at Fairbanks Command and Data Acquisition Station (FCDAS) as well as Wallops Command and Data Acquisition Station (WCDAS), and services were 10 contracted with Kongsberg Satellite Services (KSAT) at their Tromsø Satellite Station through NOAA agreements with the Norwegian Space Center (Ref. 26).

Since April 2008, NOAA stations have been providing both uplink and downlink services and Tromsø has been providing downlink services only. Ground station support availability for FS-3/C was required to perform at 90% or better. Over the course of FS-3/C operations, ground stations services have performed at 95% or better with only minor interruptions due to occasional equipment issues (hung servers or processors, for example).

The MOC (Mission Operations Center) is located at NSPO. The MOC is embedded into NSPO's MMC (Multi-Mission Center). The MOC performs all S/C operations.

All science and some telemetry data is being sent to CDAAC (COSMIC Data Analysis and Archive Center) in Boulder, CO, and to TACC (Taiwan Analysis Center for COSMIC), a mirror site of CDAAC in Taiwan, located at CWB (Central Weather Bureau) in Taipei. The centers also receive data from a global network of ground GPS and TBB (Tri-Band Beacon Transmitter) receiving sites (the so-called fiducial network). The centers analyze the received data and distribute it to the principal investigators and to the science community for operational evaluation and research.

Downlink Parameters

Uplink Parameters

Receiver frequency

1690-1700 MHz (L-band)

Transmitter frequency

2025-2120 MHz


Left Hand Circular


Left Hand Circular

Data rates

Up to 2 Mbit/s

Data rate

Up to 32 kbit/s


CCSDS compatible


CCSDS compatible

Table 4: FormoSat-3/COSMIC communication characteristics

During the initial two years of the mission, the science data will be made available through the FormoSat-3/COSMIC project websites without charge.

POD (Precise Orbit Determination) of the constellation is required for data analysis.

FS-3/C mission data is distributed from data servers at the ground stations across the open internet via SFTP (Secure File Transfer Protocol) to the SOCC and CDAAC. Figure 4 shows the flow of data between the RTS, SOCC and CDAAC. Timeliness can vary but SFTP has been found to be a very reliable and inexpensive means for distributing the data globally. A typical post contact scenario consists of transferring real-time and non-real-time spacecraft data to SOCC, followed by the transfer of mission files to CDAAC and then to SOCC. Statistics show that mission data arrives at CDAAC for processing 15 min after spacecraft loss of signal (LOS) 97% of the time (Ref. 26).


Figure 20: System architecture of the FormoSat-3(COSMIC mission (image credit: NSPO, NOAA, UCAR) 46)


Figure 21: FormoSat-3(COSMIC ground station network and performance periods (image credit: NSPO, NOAA, UCAR)


Figure 22: Data flow in the FormoSat-3/COSMIC ground system (image credit: NSPO, NOAA, UCAR)


Science data processing: The CDAAC (COSMIC Data Analysis and Archival Center) at UCAR (2010/11) processes COSMIC data in near real time for operational weather centers and the research community. The CDAAC also reprocesses RO data in a more accurate post-processed mode (within 6 weeks of observation) for COSMIC and other missions such as: GPS/MET, CHAMP, SAC-C, GRACE, TerraSAR-X, (and METOP/GRAS in the near future).

The data processing at the CDAAC includes: GPS site coordinate and ZTD (Zenith Tropospheric Delay) estimation for a global ground-based reference network, high-rate (30 s) GPS satellite clock estimation, LEO precision orbit determination, computation of L1 and L2 atmospheric excess phases, retrieval of neutral atmospheric bending angles and refractivity for each LEO occultation event, estimation of absolute TEC (Total Electron Content), and retrieval of electron density profiles. The CDAAC also provides COSMIC TIP calibrated radiance products. All COSMIC products are made available freely 15 to the community at

Since launch of the FS-3/C constellation in 2006, COSMIC has provided a large amount of valuable payload science data to the operational and research communities. As of 1 September 2010, COSMIC and CDAAC have produced over 2.5 million high quality neutral atmospheric and ionospheric sounding profiles, over 2.6 million absolute TEC data arcs, S4 scintillation observations, over 16 000 hours of quality controlled TIP radiances, and a significant (but not centrally archived) amount of ground-based CERTO/TBB observations. On average, COSMIC currently produces around 1000 GPS-RO soundings per day. Approximately ninety percent of these are processed and delivered (via GTS) to operational centers within 3 h; the remaining ten percent have higher latency due to the satellites' inability to downlink every orbit (~100 minutes). The COSMIC RTSs (Real-Time Stations) are down-linking and forwarding the payload data to the CDAAC in less than 15 minutes on average. The CDAAC processes a single dump of payload data into profiles and forwards them to the GTS via NOAA in less than 10 min.


Figure 23: COSMIC operational processing (image credit: UCAR, NOAA) 47)

In 2011, the average latency of COSMIC data is approximately 75 minutes for single orbit dumps. The reliability of the RTS stations and the CDAAC near real-time processing system have been measured at greater than 95% and 99.5%, respectively (Ref. 26).


Figure 24: Overview of existing radio occultation missions and outlook (image credit: NSPO, NOAA, UCAR, Ref. 46)


1) L.C. Lee, C. Rocken, "Applications of Constellation Observing System for Meteorology, Ionosphere & Climate", R. Kursinski (Ed.), Springer, 2000, ISBN 962-430-135-2

2) C. Rocken, Y. H. Kuo, W. S. Schreiner, D. Hunt, S. Sokolovskiy, C. McCormick, "COSMIC System Description," Special issue of TAO (Terrestrial, Atmospheric and Oceanic Science), Vol. 11, No. 1, March 2000, pp.21-52

3) G. A. Hajj, L. C. Lee, X. Pi, L. J. Romans, et al., COSMIC GPS Ionospheric Sensing and Space Weather," Special issue of TAO (Terrestrial, Atmospheric and Oceanic Science), Vol. 11, No. 1, March 2000, pp.235-272

4) Y. K. Kuo, L. C. Lee, "A Constellation of Microsatellites Promises to Help in a Range of Geoscience Research," EOS Transcriptions, AGU, Vol. 80, No. 40, Oct. 5, 1999, pp. 467-471


6) Information provided by Paul Chen of NSPO

7) E. B. Pavlis, C. Chao, C. Hwang, C. Liu, C. Shum, C. Tseng, M. Yang, "Geodetic applications of the ROCSat-3/COSMIC mission, Towards an Integrated Global Geodetic Observing System (IGGOS)," International Association of Geodesy Symposia, Vol. 120, editors: R. Rummel, H. Drewes, W. Bosch, H. Hornik, pp. 214-217, Springer-Verlag Berlin, Germany, October, 1998

8) Y. H. Kuo, C. Rocken, R. A. Anthes, "Constellation Observing System for Meteorology, Ionosphere and Climate (COMIC)," URL:

9) "COSMIC Fact Sheet," URL:

10) Chen-Tsung Lin, Ren-Young Liu, "On orbit performance and constellation development lessons learnt of FormoSat-3/COSMIC satellite attitude control subsystem," 7th International ESA Conference on Guidance, Navigation & Control Systems (GNC 2008), June 2-5, 2008, Tralee, County Kerry, Ireland

11) "FormoSat-3/COSMIC," Orbital ATK Fact Sheet, URL:

12) C. McCormick, C. Lenz, D. Smith, T. Yunck, "Community Initiative for Continuing Earth Radio Occultation CICERO," Proceedings of the 21st Annual AIAA/USU Conference on Small Satellites, Logan, UT, USA, Aug. 13-16, 2007, SSC07-I-3

13) P. A. Bernhardt, C. L. Siefring. A. Yau,"Space Based Systems for Ionospheric Density and Scintillation Mapping in Conjunction with Incoherent Scatter Radars," AMISR Science Planning Meeting, Asilomar, CA, Oct. 12, 2006

14) A.-M. Wu, C. J. Shieh, V. Chu, "ROCSat-3 Constellation Design and Data Simulation," Proceedings of 53rd IAC and World Space Congress, 2002, Oct. 10-19, 2002, Houston, TX, IAF-02-A.7.06

15) A.-M. Wu, A. Husiau, C.-T. Lin, "FormoSat-3 Constellation Deployment," 58th IAC (International Astronautical Congress), International Space Expo, Hyderabad, India, Sept. 24-28, 2007, IAC-07-B4.5.10

16) Note: Earth oblateness is the reason for the orbital plane drifts. The nodal precession, a well-known gravity phenomenon, is for instance being used by all spacecraft in sun-synchronous orbit to compensate for the Earth's revolution around the sun (about 0.9856º per day).

17)_____________________ "FORMOSAT-7/COSMIC-2 (COSMIC-2) Science Mission," UCAR, Aug. 29, 2016, URL:

18) "10th Launch Anniversary of FORMOSAT-3/COSMIC and 3rd ICGPSRO 2016," NSPO, March 9, 2016, URL:

19) Bill Schreiner and COSMIC Team, "COSMIC Data Analysis and Archive Center (CDAAC) - Current Status and Future Plans,"Eighth FormoSat-3/COSMIC Data Users' Workshop," Boulder, CO, USA, Sept. 30-Oct. 2, 2014, URL:

20) 2nd International Conference on GPS Radio Occultation (ICGPSRO), Aspire Resort, TaoYuan County, Taiwan, May 14-16, 2013

21) Nick L. Yen, Chen-Joe Fong, Kendra L. B. Cook, Peter Wilczynski, "Future Low Earth Observation Satellite - Radio Occultation Mission," Proceedings of the 2011 EUMETSAT Meteorological Satellite Conference, 5-9 September 2011, Oslo, Norway, URL:

22) Kendra L. B. Cook, Peter Wilczynski, Chen-Joe Fong, Nick L. Yen, G. S. Chang, "The Constellation Observing System for Meteorology Ionosphere and Climate Follow-On Mission," 2011 IEEE Aerospace Conference, Big Sky, MT, USA, March 5-12, 2011

23) C.-J. Fong, D. Whiteley, E. Yang, K. Cook, V. Chu, B. Schreiner, D. Ector, P. Wilczynski, T.-Y. Liu, N. Yen, "Space and ground segment performance and lessons learned of the FORMOSAT-3/COSMIC mission: four years in orbit," Atmospheric Measurement Techniques, 4, pp. 1115–1132, June 2011, URL:

24) "Six satellites, five years - COSMIC hits a milestone, looks ahead," UCAR, March 21, 2011, URL:

25) UCAR COSMIC Program Office, URL:

26) C.-J. Fong, D. Whiteley, E. Yang, K. Cook, V. Chu, B. Schreiner, D. Ector, P. Wilczynski, T.-Y. Liu, N. Yen, "Space and ground segment performance of the FORMOSAT-3/COSMIC mission: four years in orbit," Atmospheric Measurement Techniques Discussions, 2011, doi:10.5194/amtd-4-599-2011, URL:

27) Nick Yen, "FormoSat-7 /COSMIC-2 Joint Plan and Current Progress," OPAC-2010 (Occultations for Probing Atmosphere and Climate-2010), Joint OPAC-4, GRAS-SAF & IROW-1 Climate Workshop, Graz, Austria, September 6-11, 2010 URL:

28) Lidia Cucurull, Dave Ector, Estel Cardellach, "An overview of the COSMIC follow-on mission (COSMIC-II) and its potential for GNSS-R," GNSS-R Workshop, Barcelona, Spain, October 21-22, 2010, URL:

29) Chen-Joe Fong, Nick L. Yen, Chung-Huei Chu, Chun-Chieh Hsiao, Shan-Kuo Yang, Yao-Chang Lin, Shao-Shing Chen, Yuei-An Liou, Sien Chi, "In Quest of Global Radio Occultation Mission for Meteorology beyond 2011," Proceedings of the 2009 IEEE Aerospace Conference, Big Sky, MT, USA, March 7-14, 2009

30) Joe Fong Chen, Tzong Shiau Wen, Tsung Lin Chen, Chuan Kuo Tien, Huei Chu Chung, Kuo Yang Shan, N.L. Yen, Shing Chen Shao, Hwa Kuo Ying, An Liou Yuei, Chi Sien, "Constellation Deployment for the FORMOSAT-3/COSMIC Mission," IEEE Transaction on Geoscience and Remote Sensing, Volume: 46, Issue: 11, Part 1, Nov, 2008, pp. 3367-3379

31) Chen-Joe Fong, Nick Yen, Vicky Chu, Eddy Yang, Cheng-Yung Huang, Shao-Shing Chen, Yuei-An Liou, Sien Chi, "Constellation Challenges and Contributions of Taiwan Weather Monitoring Satellites," Proceedings of the 2008 IEEE Aerospace Conference, Big Sky, MT, USA, March 1-8, 2008

32) Chung-Huei Vicky Chu, Nick Yen, Chun-Chieh Hsiao, Chen-Joe Fong, Shan-Kuo Eddy Yang, Tie-Yu Liu, Mark Lin, Jiun-Jih Miau, "Earth Observations with Orbiting Thermometers - Prospective FORMOSAT-3/COSMIC Follow-On Mission," Proceedings of the 22nd Annual AIAA/USU Conference on Small Satellites, Logan, UT, USA, Aug. 11-14, 2008, SSC08-II-2

33) C.-H. Vicky Chu, S.-K. Yang, C.-J. Fong, N. Yen, T.-Y. Liu, W.-J. Chen, D. Hawes, Y.-A. Liou, B. Kuo, "The Most Accurate and Stable Space-Borne Thermometers - FORMOSAT-3/COSMIC Constellation," Proceedings of the 21st Annual AIAA/USU Conference on Small Satellites, Logan, UT, USA, Aug. 13-16, 2007, SSC07-VII-1

34) Latest Status update - Aug. 23, 2007, URL:

35) An-Ming Wu, Lance Wu, "Integrated Mission Planning for FormoSat-2 Imaging Satellite and FormoSat-3 Meteorological Constellation," Proceedings of the 57th IAC/IAF/IAA (International Astronautical Congress), Valencia, Spain, Oct. 2-6, 2006, IAC-06-B1.6.09


37) T. K. Meehan, Chi O. Ao, Byron Iijima, David Robinson, Doug Hunt, Christian Rocken,Bill Schreiner, Sergey Sokolovskiy, "A Demonstration of L2C Tracking from Space for Atmospheric Occultation," ION GNSS 21st. International Technical Meeting of the Satellite Division, 16-19, September 2008, Savannah, GA, USA

38) C. Coker, K. F. Dymond, S. A. Budzien, D. Chua, "First Observations of the Ionosphere Using the Tiny Ionospheric Photometer," Taipei, Taiwan, FormoSat-3/COSMIC Workshop 2006 - Early Results and IOP Campaigns, Nov. 28-Dec. 1, 2006

39) P. C. Kalmanson, S. A. Budzien, C. Coker, K. F. Dymond, "The tiny ionospheric photometer instrument design and operation," Proceedings of SPIE, `Instruments, Science, and Methods for Geospace and Planetary Remote Sensing,' Carl A. Nardell, Paul G. Lucey, Jeng-Hwa Yee, James B. Garvin, editors, Vol. 5660, Bellingham, WA, Dec. 2004, pp. 259-270

40) C. Coker, K. F. Dymond, S. A. Budzien, "Using the Tiny Ionospheric Photometer (TIP) on the COSMIC Satellites to Characterize the Ionosphere," American Geophysical Union (AGU) Fall Meeting San Francisco, CA, Dec. 6-10, 2002

41) K. F. Dymond, J. B. Nee, R. J. Thomas, 2000: "The Tiny Ionospheric Photometer: An Instrument for Measuring Ionospheric Gradients for the COSMIC Constellation," Terrestrial, Atmospheric and Oceanic Sciences, Vol. 11, 2000, pp. 273-290.

42) P. A. Bernhardt, C. E. Coker, "New TEC Data Sources from Radio Beacon Monitors of the Ionosphere," LWS Geostorm CDAW and Conference Florida Tech, Melbourne, FL, March 8, 2007

43) P. A. Bernhardt, C. A. Selcher, S. Basu, G. Bust, S. C. Reising, 2000: "Atmospheric studies with the Tri-Band Beacon instrument on the COSMIC constellation," Terrestrial, Atmospheric and Oceanic Sciences, Vol. 11, No 1, March 2000, pp. 291-312

44) P. A. Bernhardt, C. L. Siefring, "The CERTO and CITRIS Instruments for Radio Scintillation and Electron Density Tomography from the C/NOFS, COSMIC, NPSAT1 and STPSAT1 Satellites," The 2004 Joint Assembly (of CGU, AGU, SEG and EEGS), Montreal, Canada, May 17.21, 2004

45) P. A. Bernhardt, C. L. Siefring, T. W. Garner, T. L. Gaussiran, J. Secan, F. Smith, K. Groves, "First Results for the TBB/CERTO Beacon Experiment on FormoSat- 3/COSMIC," AGU (American Geophysical Union) Fall Meeting, 2006, San, Francisco, CA, USA, Dec. 11-15, 2006

46) Chen-Joe Fong, Doug Whiteley, Eddy Yang, Kendra Cook, Vicky Chu, Bill Schreiner, Dave Ector, Pete Wilczynski, Tie Yue Liu, Nick L. Yen, "Space & Ground Segment Performance of the FormoSat-3 / COSMIC Mission: Four Years in Orbit," OPAC-2010 (Occultations for Probing Atmosphere and Climate-2010), Joint OPAC-4, GRAS-SAF & IROW-1 Climate Workshop, Graz, Austria, September 6-11, 2010 URL:

47) Bill Schreiner, C. Rocken, X. Yue, B. Kuo, D. Mamula, D. Ector, "GNSS Radio Occultation Constellations for Meteorology, Ionosphere and Climate: Status of the COSMIC and Planned COSMIC-2 Missions," 2011 Space Weather Workshop, Boulder, CO, USA, April 26-29, 2011

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 (

Overview    Spacecraft    Launch    Mission Status    Sensor Complement   Ground Segment   References   Back to Top