Atmospheric delay modeling for satellite laser altimetry

Abstract

NASA's Ice, Cloud, and Land Elevation Satellite (ICESat) is a laser altimetry mission with the primary purpose of measuring the mass balance of the ice sheets of Greenland and Antarctica. It will provide 5 years of topography measurements of the ice, as well as land and ocean topography. In order to accurate topography measurements the laser altimeter ranges must be corrected for certain biases. Atmospheric delay is one such bias. As the laser pulse travels through the atmosphere it will be refracted, introducing a delay into the travel time. This delay must be estimated to correct the ranges and the delay estimations need to be validated. Of particular concern are errors in the delay estimates that have the same characteristics as the expected mass balance variations. The main focus of this dissertation is to formulate algorithms for calculating the ICE-Sat atmospheric delay and estimate the expected delay values and errors. Our atmospheric delay algorithm uses numerical weather model data to estimate delay values. We have validated these algorithms using Automatic Weather Stations (AWS) in the polar regions and GPS data over the globe. The GPS data validation was also augmented by in-situ meteorology measurements at some the stations. The GPS validation process additionally allowed us to investigate the estimation of precipitable water vapor using GPS data. The validation studies have shown that our atmospheric delay algorithm errors are well within the ICESat error budget of 20 mm. The overall global delay errors are estimated to be approximately 5.4 mm and the polar delay errors are 12.2 mm. There are no discernible biases in the error and the seasonal variations in error magnitudes are well characterized.