Downscaling of precipitation and temperature for climate change study in Singapore with LARS-WG

• Main Author: Lai, Hong Chern
• Other Authors: Qin Xiaosheng
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Civil engineering
• Online Access: http://hdl.handle.net/10356/71445
• Institution: Nanyang Technological University

Singapore has been experiencing the effects of climate change in the past few years in the form of increased dry spell length and increased number of warm days. Although the country has remained competent to cope with heightened water and energy demand, and spontaneous flash flood, strategic plans for building resilience and improving civil infrastructure should never be compromised as studies have shown evidence on continuous deterioration of Singapore’s climate. As General Circulation Models (GCMs) are known for their disadvantage of producing coarse grid information, statistical downscaling tools have been developed to provide finer and serviceable sub-grid information for hydrological and climate change study. In this study, outputs from 10 GCMs has been selected to generate 30 years of future weather series under three emission scenarios (A1B, A2 and B1) in three time periods (2011 to 2030, 2046 to 2065, and 2080 to 2099) by using Long Ashton Research Station – Weather Generator (LARS-WG), a stochastic climate downscaling tool that has been used intensively in climate change studies. LARS-WG performed well in downscaling precipitation and temperature of Singapore by using historical data (1979-2014) despite of falling short in modeling standard deviation of precipitation. The projected weather series showed that the most prominent change in precipitation happens to be during the Northeast Monsoon season of the year. Drier condition for the first quarter is expected. The temperature is expected to rise by the range from 0.2˚C to 3.14˚C by the end of the century.