Historical and future antarctic precipitation based on cmip5 models, reanalysis data and in situ measurements / Malcolm Tang Siong Yii

Malcolm Tang , Siong Yii (2020) Historical and future antarctic precipitation based on cmip5 models, reanalysis data and in situ measurements / Malcolm Tang Siong Yii. PhD thesis, Universiti Malaya.

	PDF (The Candidate's Agreement) Restricted to Repository staff only Download (203Kb)
Preview	PDF (Thesis PhD) Download (3572Kb) | Preview

Abstract

The study of precipitation in Antarctica is one of the most studied research topics today. The reliability and accuracy of climate models in simulating Antarctic precipitation, however, are still debatable. Measuring precipitation in Antarctica poses many distinctive challenges. This work attempts to establish a comprehensive study of precipitation in Antarctica. The first part of the study assesses a year-long measurements from five precipitation instruments with reanalysis datatsets and satellite data. The tipping bucket gauges (TBGs) were observed to be less sensitive compared to laser-based sensors (LBSs). Case studies of the daily precipitation and seasonal precipitation measurements showed VPF-730 to be the most reliable precipitation sensor among the instruments. The reanalyses had a positive correlation with wind speed, in particular the precipitation measurement from the Japanese 55-year Reanalysis (JRA-55). During strong wind events, the GPCP 1-Degree-Daily (1DD) was unable to fully capture the effect of wind, and hence the relatively low precipitation amount. The Laser Precipitation Monitor (LPM) and Campbell Scientific-700 (CS700H) had instrumental errors during the study. Installing multiple LBSs at different locations (in close proximity) can help identify inconsistency in the readings. For the second part of the study, we assess the precipitation and surface air temperature (SAT) of Antarctica (90 ⁠oS to 60 ⁠oS) using CMIP5 models and the European Centre for Medium-range Weather Forecasts “Interim” reanalysis (ERA-Interim); the National Centers for Environmental Prediction Climate Forecast System Reanalysis (CFSR); JRA-55; and the Modern Era Retrospective-analysis for Research and Applications (MERRA) datasets for 1979–2005. For precipitation, the time series show that the MERRA and JRA-55 have increased significantly from 1979 to 2005, while the ERA-Interim and CFSR have insignificant changes. The reanalyses also have low correlation with one another (generally less than +0.69). 37 CMIP5 models show increasing trend, 18 of which are significant. The resulting CMIP5 multimodel mean (MMM) also has a significant increasing trend of 0.29 ± 0.06 mm year⁠−1. For SAT, the reanalyses show insignificant changes and have high correlation with one another, while the CMIP5 MMM shows a significant increasing trend. The variability of precipitation and SAT of MMM could affect the significance of its trend. One of the many reasons for the large differences in precipitation is the resolution of the CMIP5 models. The final part of the study involves using CMIP5 models to predict precipitation and SAT trends in Antarctica under four different representative concentration pathways (RCP): RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5 for the year 2006-2100. The study shows that for precipitation, the Peninsula has the highest trend regardless of scenarios. For SAT, the interior region has the highest trend, followed by the Weddell Sea off the coast of Halley. For the projection of SAT, the consistency of the CMIP5 models in simulating accurate historical SAT climatology no longer appear in the RCP experiments. However, the consistency of the projection from different models increases as the Greenhouse gases (GHG) concentration increases. The correlation between SAT and precipitation increases as the GHG concentration increases.

Actions (For repository staff only : Login required)