Trend of Future Rainfall Estimates Using PRECIS Scenario A2 & B2 in Northeastern After Being Bias Adjusted with Spatial Monthly Average by Kriging Method

Abstract

Since there are variations in the future rainfall estimates using the regional climate model PRECIS Scenario A2 & B2 comparing to measured rainfall from rain gauge stations at a certain time and area due to seasonal influence and different geography in each area, hence, the objective of this study is to analyze and find spatial monthly average of bias adjustment with 1 km x 1 km grid cell in each month and to apply the analyzed monthly bias adjustments of 265 rain gauge stations located at the same position in the article “Analysis of Bias Adjusted Rainfall Estimates Using PRECIS Scenario A2 & B2 in Northeastern with the Data Collected from Rain Gauge Stations” to analyze and find spatial monthly average of bias adjustment with 1 km x 1 km grid cell in each month using Simple Kriging, Ordinary Kriging, Universal Kriging, Co- Simple Kriging, Co-Ordinary Kriging and Co-Universal Kriging respectively. The Cross-validation result is used to select the appropriate bias adjustment and it shows that the rainfall estimates using PRECIS Scenario A2 is suitable with Simple Kriging method for bias adjustment in January and February, Ordinary Kriging method is suitable for bias adjustment in April, Universal Kriging method is suitable for bias adjustment in July, Co-Simple Kriging method is suitable for bias adjustment in August, November and December, Co- Ordinary Kriging method is suitable for bias adjustment in May, June, September and October, Co-Universal Kriging method is suitable for bias adjustment in March whereas the rainfall estimates using PRECIS Scenario B2 is suitable with Simple Kriging method for bias adjustment in November and December, Ordinary Kriging method is suitable for bias adjustment in July and September, Co-Simple Kriging is suitable for bias adjustment in February, August and October, Co- Ordinary Kriging is suitable for bias adjustment in January, April and June and Co-Universal Kriging is suitable for bias adjustment in March and May respectively. The result of spatial monthly average of bias adjustment with 1 km x 1 km grid cell in each month has been used for bias adjustment of future rainfall amount during the year 2013-2069 from the regional climate model PRECIS Scenario A2 & B2 in 20 Provinces: Maha Sarakham, Nakhon Phanom, Roi Et, Sakon Nakhon, Si Sa Ket, Nakhon Ratchasima, Nong Khai, Bueng Kan, Nong Bua Lam Phu, Amnat Charoen, Udon Thani, Ubon Ratchathani, Buri Ram, Chaiyaphum, Mukdahan, Surin, Yasothon, Kalasin, Khon Kaen and Loei in order to analyze the trend of future rainfall amount. The study finds that the future rainfall estimates using PRECIS during the year 2013-2069 tend to have an increase in annual rainfall average, rainy season rainfall average and dry season rainfall average by 7%, 8% and 5% respectively for A2 scenario whereas B2 scenario has been increased by 10%, 12% and decreased by 8% respectively comparing to the base year 1970-2012.