Linear regression models to predict the tropospheric parameters at the Global Positioning systems’ sites over the East African region

Abstract

Currently, the East African tropical region has limited information about Precipitable Water Vapour (PWV) data and yet the region has a high potential for its utilization. This is on the grounds that the East African tropical region is profoundly prone to climate change and fluctuation. Existing studies need data on the detailing and performance evaluation of precipitable water vapour models within East Africa. This has been so as a result of the scattered Global Positioning System (GPS) networks and other alternative water vapour measuring equipments, enormous information gaps and the absence of surface meteorological data. The accessibility and precision of surface meteorological estimations is crucial in deriving accurate GPS PWV data. In this study, the daily average, PWV, pressure, temperature and weighted mean temperature () models have been developed utilizing one year (2013) GPS PWV and European Centre for Medium-Range Weather Forecasts (ECMWF) 5th Re- Analysis PWV (ERA5 PWV), total column water vapour (TCWV), surface pressure and 2 meter (2m) temperature data. The purpose of the developed models is to predict PWV over regions with data gaps where the computation of GPS Zenith Tropospheric Delays (ZTD) is impossible and in cases of station outages. In addition, the models will provide meteorological parameter where meteorological sensors are missing. The GPS PWV accuracy obtained with the developed models shows an average RMSE of 1.54 mm and MnB of 0.32 mm in comparison to the measured GPS PWV data. The ERA5 PWV accuracy obtained with the developed models shows an average RMSE of 0.33 mm and MnB of 0.01 mm in comparison to the measured ERA5 PWV data. Based on the RMSE, it was observed that the site-specific models developed can be utilized to provide estimates of nearly a similar degree of precision compared to the measured values at the thirteen stations.

Investigating the Impacts of Tropospheric Parameters on Received Signal Strength of the Mobile Communication System

“…The RSS is analyzed and correlated with each independent variable. A general formula for multiple linear regression (MLR) models includes multiple independent variables that are expressed in terms of dependent variables as follows (Orogun, 2021;Ssenyunzi et al, 2021):…”