This paper provides an overview of the methodology of and describes preliminary results from an experiment called GPS/MET (Global Positioning System/Meteorology), in which temperature soundings are obtained from a low Earthorbiting satellite using the radio occultation technique. Launched into a circular orbit of about 750-km altitude and 70° inclination on 3 April 1995, a small research satellite, MicroLab 1, carried a laptop-sized radio receiver. Each time this receiver rises and sets relative to the 24 operational GPS satellites, the GPS radio waves transect successive layers of the atmosphere and are bent (refracted) by the atmosphere before they reach the receiver, causing a delay in the dualfrequency carrier phase observations sensed by the receiver. During this occultation, GPS limb sounding measurements are obtained from which vertical profiles of atmospheric refractivity can be computed. The refractivity is a function of pressure, temperature, and water vapor and thus provides information on these variables that has the potential to be useful in weather prediction and weather and climate research. Because of the dependence of refractivity on both temperature and water vapor, it is generally impossible to compute both variables from a refractivity sounding. However, if either temperature or water vapor is known from independent measurements or from model predictions, the other variable may be calculated. In portions of the atmosphere where moisture effects are negligible (typically above 5-7 km), temperature may be estimated directly from refractivity. This paper compares a representative sample of 11 temperature profiles derived from GPS/MET soundings (assuming a dry atmosphere) with nearby radiosonde and high-resolution balloon soundings and the operational gridded analysis of the National Centers for Environmental Prediction (formerly the National Meteorological Center). One GPS/MET profile was obtained at a location where a temperature profile from the Halogen Occultation Experiment was available for comparison. These comparisons show that accurate vertical temperature profiles may be obtained using the GPS limb sounding technique from approximately 40 km to about 5-7 km in altitude where moisture effects are negligible. Temperatures in this region usually agree within 2°C with the independent sources of data. The GPS/MET temperature profiles show vertical resolution of about 1 km and resolve the location and minimum temperature of the tropopause very well. Theoretical temperature accuracy is better than 0.5°C at the tropopause, degrading to about 1°C at 40-km altitude. Above 40 km and below 5 km, these preliminary temperature retrievals show difficulties. In the upper atmosphere, the errors result from initial temperature and pressure assumptions in this region and initial ionospheric refraction assumptions. In the lower troposphere, the errors appear to be associated with multipath effects caused by large gradients in refractivity primarily due to water vapor distribution.

Abstract.The

el; of Silurian soils as a result of pedoturbat i o~~, effectively mcreasing the average depth of soil CO, proiluction.Our results (Table 1) ~m p l y that atmospheric CO-, decllneil 1~y a factor of 10 from the Late S i h r i a n to the Early Permian, closely follow~ng (Fig. 4 ) a decline precllctecl hi; theoretical carbon lnais balance models (1). T h e largest decrease, hetween the Late Sil~lrian a11il Late Devonian. coincides with a of rapid evolution and diversificatlon of the terrestrial ecosystem (18).Estimates of atmospheric C02 levels from separated, time-equivalent ~-7aleosols are consistent, suggertlng that a coherent record of changing atlnospheric chem~stry is yreserl-eii In the ancient soil recorJ. REFERENCES AND NOTES1 R A Berner Science 261, 68 (1 993) At?? J SCI 294 56 (1 994) 2 T J Crowley and G R North, Paleoc!~~rato!ogy (Oxford UI?I\/ Press, Oxford, 1991) 3 R A Berner and R Ralswell, Geochim. Cosmochlm. Acta 47. 855 11983) L R. I

A reduced dynamic filtering strategy that exploits the unique geometric strength of the Global Positioning System(GPS) to minimize the effects of force model errors has yielded orbit solutions for TOPEX/POSEIDON which appear accurate to better than 3 cm (1 σ) in the radial component. Reduction of force model error also reduces the geographic correlation of the orbit error. With a traditional dynamic approach, GPS yields radial orbit accuracies of 4–5 cm, comparable to the accuracy delivered by satellite laser ranging and the Doppler orbitography and radio positioning integrated by satellite (DORIS) tracking system. A portion of the dynamic orbit error is in the Joint Gravity Model‐2 (JGM‐2); GPS data from TOPEX/POSEIDON can readily reveal that error and have been used to improve the gravity model.

No abstract