Precipitable Water Vapor (PWV) Estimation from GNSS Modelled Tropospheric Delay: A case of SURVNET and IGS Stations in Uganda.

Abstract

Context and background

 The Lake Victoria basin in Uganda is prone to frequent weather extremes, including floods and droughts, which have devastating impacts on agriculture, water resources, and human settlements. Accurate estimation of Precipitable Water Vapor (PWV) is crucial for predicting these weather events. However, traditional PWV estimation methods using radiosondes and weather satellites have limitations in terms of spatial coverage, temporal resolution, and cost. In recent years, Global Navigation Satellite System (GNSS) technology has emerged as a promising alternative for PWV estimation, leveraging the tropospheric delay of GNSS signals. This study explores the potential of GNSS-modelled tropospheric delay for PWV estimation in the Lake Victoria basin, utilizing data from SURVNET and IGS stations in Uganda.

Goal and Objectives:

The goal of this study was to investigate the feasibility of estimating Precipitable Water Vapor (PWV) from GNSS-modelled tropospheric delay by assessing the accuracy of GNSS-derived PWV estimates and evaluating the performance of SURVNET and IGS stations in Uganda,

Methodology:

This study utilized a combination of GNSS data from SURVNET and IGS stations in Uganda, located within the Lake Victoria basin. The GNSS data was processed using the GAMIT/GLOBK software to estimate the tropospheric delay, which was then converted to PWV using empirical formulas. The PWV estimates were validated against MODIS PWV data from NASA. Additionally, spatial and temporal analysis of the PWV estimates was performed to assess the variability of atmospheric water vapour over the study area.

Results:

The results showed that the GNSS-derived PWV estimates from SURVNET and IGS stations in Uganda correlated well with MODIS-derived PWV data, with a standard deviation of 2.23 mm and a correlation coefficient (R) reaching 0.94. The spatial analysis revealed minimal variability in PWV across the Lake Victoria basin, with higher values observed during the wet season than in the dry. Overall, the results demonstrated the potential of GNSS-modelled tropospheric delay for accurate PWV estimation in the Lake Victoria basin.