Study of precipitable water vapor from GPS

• Main Author: Tan, Jian Hong
• Other Authors: Lee Yee Hui
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/77527
• Institution: Nanyang Technological University

In this research project, extensive research and analysis were done to find a correlation between various parameter and cloud formation. The GPS dataset used is obtained from Nanyang Technological University Singapore (NTUS) Global Navigation Satellite System (GNSS) and is processed using GNSS-Inferred Positioning System (GISPY-OASIS) software. Global Mapping Function (GMF) was chosen to process the data after substantial considerations as it is the simplest to implement and the results derived strongly resembles the results obtained from numerical weather model (NWM) Mapping Functions (MF) such as Vienna Mapping Function (VMF1). The focus of this research project will be on the attempt to correlate post-fit residual and rainfall events as there is little to no research on the impact of cloud formation on post-fit residual. Observation of the heatmap and scatterplot of the elevation against post-fit residual values shows a significant amount of errors, mainly due to multi-path effects. Multi-path Stacking map (MPS) algorithm was used to eliminate or minimize the effects of multi-path on the post-fit residual values. The corrected post-fit residual show a good correlation with rainfall as the range of variation of the residual value increases significantly during rainfall events. This led to the use of Standard Deviation (SD) of the corrected post-fit residual values, along with rainfall data from the weather station and weather radar, to plot graphs in time series to observe the trends during rainfall events. It is observed that the SD of the corrected post-fit residual increases significantly during periods with rainfall happening and is comparatively low during non-rainy days. The results of both observation, using weather radar, weather station and GPS data show the potential that post-fit residuals can be integrated into existing algorithms to improve nowcasting’s rainfall prediction.