Launch of the Minotaur rocket that carried COSMIC, 15 April 2006. (Courtesy Orbital Sciences Corporation.)
Editor's note: On the evening of 14 April, ten years of planning came to fruition with the successful launch of six microsatellites of COSMIC (Constellation Observing System for Meteorology, Ionosphere and Climate), called FORMOSAT-3 in Taiwan. In this installment of President's Corner, Rick Anthes—a principal investigator since the project's inception—provides a first-person look at the launch and explains how scientists can sign up to work with the data now beginning to stream in. We'll provide follow-up coverage in future issues of the Quarterly as COSMIC unfolds.
It was the proverbial dark and stormy night on 14 April at Vandenberg Air Force Base in California. A Minotaur rocket carrying six identical microsatellites for COSMIC waited quietly in the rain as the countdown proceeded toward a projected liftoff at 5:10 p.m. Pacific Daylight Time. I had never witnessed a launch in person before, but I knew that many problems could arise at the last minute, causing delays or cancellation, and was amazed as the countdown passed one hour, 30 minutes, then 10 and 5 minutes without incident. Then, without warning, only 90 seconds from launch, the Taiwanese and U.S. visitors in the control room were stunned by three sharp words from the mission commander: "Abort. Abort. Abort." A pressure reading from the rocket engine did not agree with a written specification. Given the threatening weather and the short launch window (about three hours), the launch was clearly at risk.
Fortunately, the problem was with the written specification, not the rocket. After the batteries on the payloads were recharged, the countdown resumed, this time without interruption. The satellites were successfully launched into low-Earth orbit (an altitude of 512 kilometers or 318 miles) at 6:40 p.m. The launch marked a major milestone in an exciting and complex journey toward use of the Global Positioning System (GPS) for atmospheric sounding for research and operational weather prediction. The journey, which began a decade ago, has involved many partners in Taiwan and the United States and scientists from more than a dozen countries around the world. It has been a rocky road in spots, but the bumps were forgotten when the Minotaur rocket shot into space and all six satellites were safely in orbit an hour later. A little over a week later, the first soundings were being obtained (see graphic).
A launch announcement and an introduction to COSMIC appear in the 25 April issue of EOS. The mission is a Taiwan-U.S. collaboration carried out by Taiwan's National Space Organization (NSPO) and UCAR. The primary science goal is to obtain near-real-time stratospheric and tropospheric soundings as well as electron density profiles in the ionosphere using the radio occultation technique. The data will be used to support operational global weather prediction, climate monitoring, space weather forecasting, and ionosphere and gravity research. (For additional details, see the special edition of Terrestrial, Atmosphere and Oceanic Sciences, March 2000.)
The payloads aboard COSMIC
There are three science payloads on each satellite: a GPS receiver, developed at NASA's Jet Propulsion Laboratory (JPL), that tracks GPS signals; a Tiny Ionospheric Photometer (TIP) payload that measures the night sky photon emission; and a Tri-Band Beacon (TBB) payload that transmits phase coherent signals with three frequencies and measures the total electron density along the path from the satellite to the ground receivers.
These two soundings, collected on 21 April, show the first profiles of electron density obtained by COSMIC.
The radio occultation receiver measures the propagation time of radio waves from a GPS satellite to a COSMIC satellite. As the radio waves slice through Earth's atmosphere, they are refracted (bent), resulting in the slowing of the signal and a shift of frequency. These changes are related to the vertical gradients of electron density in the ionosphere, temperature in the stratosphere, and temperature and humidity in the troposphere. The vertical profiles of bending angles, or refractivity, will be directly assimilated into numerical weather prediction models at operational centers around the world. Vertical profiles of temperature and moisture can also be deduced with the use of auxiliary information. If all satellites work as planned, about 2,500 vertical profiles of globally distributed atmospheric soundings will be obtained each day.
The history of radio occultation began with the leadership of scientists at JPL and Stanford University in the early 1960s, who used the technique to probe the atmospheres of other planets. In the late 1980s, scientists at JPL proposed a radio occultation mission to sound Earth's atmosphere, but their proposal was not funded. In the early 1990s, UCAR took advantage of a mission of opportunity with special firmware provided by JPL to launch a commercial GPS receiver into space on the MicroLab-1 spacecraft on 3 April 1995. The GPS-MET experiment was incredibly successful and proved the concept of obtaining high-quality atmospheric soundings of Earth. Yet there were three major issues with the GPS-MET soundings.
- About half the soundings failed to penetrate into the lower troposphere below an altitude of about 6 km (4.5 mi).
- Many of the soundings that did penetrate into the lower troposphere, especially in the moist tropics, showed a negative bias in the retrieved refractivity.
- Only setting occultations (those in which the low-Earth orbiting satellite carrying the GPS receiver was setting behind Earth relative to the transmitting GPS satellite) could be retrieved.
How to access the data
Two data centers will receive and process the raw data from the three payloads:
The data products will be ready for distribution within two hours from the end of day of collection. After a checkout period of about two months, all raw data and data products will be freely and openly available from CDAAC and TACC to the international science and operational weather prediction communities. The products include vertical profiles of bending angles, refractivity, electron density, temperature, pressure, and water vapor in the atmosphere, as well as radiances observed by the TIP. Users of the data products are required to register at the TACC Web site by submitting a Data Use Agreement electronically to NSPO for approval. An online version of the agreement is available on the NSPO TACC and the UCAR CDAAC Web sites for users to access electronically.
The difficulties associated with GPS-MET were largely solved with the development by JPL and UCAR of an improved signal tracking technique known as open loop tracking. This technique uses the predicted Doppler shift, based on an atmospheric model, which eliminates many of the problems associated with the previous tracking method. With open loop tracking, we expect that more than 50% of the GPS RO soundings will penetrate below 1 km (0.6 mi) above Earth's surface (in the absence of mountains) in the tropics, and an even higher percentage globally. Moreover, rising as well as setting occultations will be obtained, doubling the number of soundings obtained per day, and most of the negative refractivity bias will be eliminated. Tests with the Argentine SAC-C satellite mission verified that open loop tracking works as expected.
The GPS radio occultation technique can also obtain vertical profiles of electron density in the ionosphere. The three-dimensional electron density distribution between 90 and 800 km can be inferred with enough detail to study the ionospheric structure, scintillation, and irregularities in electron density (see graphic). Thus, COSMIC will give an unprecedented look at the ionospheric electron density structure, and provide a rich data set that can be assimilated into space weather models.
The six COSMIC satellites were separated from the launch vehicle one after another into the same initial orbit of 512 km (318 mi) and 72° inclination. During the year following the launch they will gradually be boosted into their final orbits of 800 km (about 500 mi). Through differential precession, they will, along the way, gradually separate into different orbital planes spaced 24° apart. Thus, during the first several months after launch the satellites will be close together, providing a dense swath or band of soundings across the globe as the constellation orbits Earth every 100 minutes. In the days immediately following the launch the satellites are very close together, affording the rare opportunity for comparison of retrievals from independent instruments and platforms at essentially the same time and place (see sidebar).
It will be many more months before the full success and impact of COSMIC can be determined, but the launch and the initial data look very promising. Reaching this milestone has been enormously gratifying for me and the COSMIC/FORMOSAT team, and I express my deeply felt thanks to all of the team members. I would especially like to thank Jay Fein, NSF's COSMIC program manager, who has supported GPS radio occultation research for many years, beginning with the GPS-MET program.
The sponsors of COSMIC/FORMOSAT-3 are Taiwan's National Science Council, NSF, NASA, NOAA, the U.S. Air Force, and the Office of Naval Research. The UCAR data analysis and archiving team of Doug Hunt, Bill Kuo, Chris Rocken, Sergey Sokolovskiy, Bill Schreiner, and Stig Syndergaard kindly provided the initial COSMIC soundings.
The first soundings
These vertical profiles of "dry" temperature (black and red lines) were made by two independent receivers on separate COSMIC satellites (FM-1 and FM-4) at 0007 UTC on 23 April 2006. The satellites were about five seconds apart, which corresponds to a distance separation of the soundings of about 1.5 km (0.9 mi). The latitude and longitude of the soundings are 20.4°S and 95.4°W. The extremely close agreement of the two COSMIC soundings demonstrates the precision of the GPS radio occultation technique.
The blue profile labeled "AVN" is the temperature obtained from the National Centers for Environmental Prediction (NCEP) analysis, interpolated to the time and place of the COSMIC soundings. The difference between the NCEP profile and the two COSMIC profiles near the tropopause at 17 km is due to the much higher vertical resolution of the COSMIC soundings. The differences below about 6 km are due to the neglect of water vapor in computing the "dry" temperature from the refractivity; thus they are a measure of the water vapor present in the atmosphere.
"Dry" temperature is calculated from the observed refractivity with the assumption that water vapor is zero. Dry temperatures would not be used in applications such as weather models, but they are useful as a simple means of comparing soundings of refractivity, which will be used in such applications.