Application of statistical weather generators

• Main Author: Wang, Qin Yu
• Other Authors: Qin Xiaosheng
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Civil engineering
• Online Access: https://hdl.handle.net/10356/154394
• Institution: Nanyang Technological University

WeaGETS is a MATLAB-based versatile random day-to-day weather generator. It can produce everyday rainfall amount, the maximum and minimum temperature for one station for any length of time. Because WeaGETS generates data over a long time, it is ideal for assessing agricultural and hydrological risk. It also allows for weather simulation in unknown regions. Furthermore, it can be used as a low-cost method to investigate the impact of climate change on a specific place. In this report, we use the first-order Markov model to generate the frequency of rainfall. Gamma distribution to produce everyday rainfall amount. Precipitation generating parameters have been smoothed using second-order Fourier Harmonics. Tmax and Tmin are generated under a conditional scheme. WeaGETS is being used to simulate twenty-three years of data from the Year 1894 to the Year 2006. We show all the details of data analysis for the first Year 1984 with the help of Excel and graph. For the other twenty-two years, data can be found in the appendix. We use MATLAB to run the WeaGETS. The coding we used had already been developed. Our primary target is to find the accuracy of the simulation data generated by WeaGETS then find the application of the WeaGETS. After comparing both data analyze based on yearly and monthly, we find WeaGETS underestimates the daily rainfall amount, frequency of rainfall, and minimum temperature. However, it overestimates the maximum temperature, so we hope the WeaGETS can be improved in the future. Moreover, we hope WeaGETS can develop to simulate not only for a single station. Finally, we also hope WeaGETS can add more climate parameters for the simulation.