Spatial Rainfall Variability and an Increasing Threat of Drought, According to Climate Change in Uttaradit Province, Thailand

This study presents the work commenced in northern Thailand on spatial and temporal variability of rainfall. Thirty years (1988-2017) rainfall data of eight meteorological stations were used for assessing temporal variability and trend analysis. The results showed decreasing trend in rainfall from its first half of the observed study period (1988-2002) to last half of the time period (2003-2017) in total average annual as well as monsoonal average rainfall by 14.92% and 15.50% respectively. It was predicted from linear regression results that by 2030 the average annual and monsoonal rainfall will drop by 35% and 34.10% respectively. All stations showed negative trend except Fakara met-station in annual rainfall. In the seasonal trend analysis, the results showed decreasing trend almost in all met-stations. Mann-Kendall trend test was applied to assess the trend. All met-stations show significant negative trend. To assess drought in the study area, Standardized Precipitation Index (SPI) was applied to 12-month temporal time period. The results predicted meteorological drought in the near future. The spatial distribution of rainfall presented changing phenomena in average annual, monsoonal, winter, and summer seasons in both analyzed periods.

Spatio-Statistical Analysis of Flood Susceptibility Assessment Using Bivariate Model in the Floodplain of River Swat, District Charsadda, Pakistan

Flood is one of the most predominant disasters around the globe and frequently occurring phenomena in the northern part of Pakistan. In this study, the effects of various divisions of flood inventory and combinations of conditioning factors were assessed for the preparation of final susceptibility map. The flood inventory map was prepared for Charsadda by visual interpretation of Landsat-7 image alongside the field survey and a total of 161 flood locations were mapped. The flood inventory was subsequently divided into training and validation datasets, 129 (80%) and 112 (70%) locations for training the model and 32 (20%) and 49 (30%) for validation of the model. In this study, nine conditioning factors were used (Elevation, Slope, Aspect, Curvature, Plan curvature, Profile curvature, Proximity to river, roads, and Land use/land cover) for the development of flood susceptibility map. All the conditioning factors were correlated with flood inventory map using the information value method. The final susceptibility maps were validated using prediction rate and success rate curve. The results from validation showed that the areas under curve in the prediction rate curve for the models are: Model A (99.47%), Model B (95.04%), and Model C (94.06%), respectively. The Area under curve (AUC) in the success rate curve obtained for the three models are: Model A (95.03%), Model B (86.91%), and Model C (89.67%), respectively. Eventually, the susceptibility maps were classified into five susceptibility zones. The success rate and prediction rate curve indicated that model A has more accuracy in comparison to model B and model C;though, the results obtained from prediction and success rate curve indicated that all the models are reliable and has no significant difference between the susceptibility maps. Consequently, results obtained from this study are useful for researchers, disaster managers, and decision-makers to manage the flood-prone areas in the study area to mitigate the flood damages.