Effect of Urbanization on Extreme Climate Indices and Compound Events in Kerala

Abstract

Studies on historical patterns of climate variables and climate indices have attained significant importance because of the increasing frequency and severity of extreme events worldwide. While the recent events in the tropical state of Kerala (India) have drawn attention to the catastrophic impacts of extreme rainfall events leading to landslides and loss of human lives, a comprehensive and long-term spatiotemporal assessment of climate variables is still lacking. This study investigates the long-term trend analysis (119 years) of climate variables at 5% significance level over the state using gridded datasets of daily rainfall (0.25° x 0.25° spatial resolution), temperature (1° x 1° spatial resolution) at annual and seasonal scales (south-west monsoon, north-east monsoon, winter, and summer). Five trend analysis techniques, including the Mann-Kendall test (MK), three modified Mann-Kendall tests, and innovative trend analysis (ITA) test, were performed in the study. It is evident from the trend analysis results that more than 83% of grid points were showing negative trends in annual and south-west monsoon season rainfall series (at a mean rate of 39.70 mm and 28.30 mm per decade, respectively). All the trend analysis tests identified statistically significant increasing trends in mean and maximum temperature at annual and seasonal scales (0.10 to 0.20 °C/decade) for all grids. The K-means clustering algorithm delineated 59 grid points into five clusters for annual rainfall, illustrating a clear geographical pattern over the study area. There is a clear gradient in rainfall distribution and concentration inside the state at annual and seasonal scales. The majority of annual rainfall is concentrated in a few months of the year. That may lead the state vulnerable to water scarcity in non-rainy seasons. Land Use and Land Cover (LULC) analysis gives essential information on how the region has evolved over time. Due to adverse environmental effects, the significant and widespread changes in the LULC resulting from human activities have become a pressing issue for decision planners and the Government. Kerala, an ecologically diverse state in India characterized by complex topography, has experienced substantial LULC transformations due to rapid urbanization. These changes were assessed by analyzing Landsat images from 1990 to 2020, utilizing two distinct machine learning classification techniques, namely Random Forest (RF) and Classification And Regression Trees (CART), within the Google Earth Engine (GEE) platform. ii Normalized Difference Vegetation Index (NDVI), Normalized Differences Built-up Index (NDBI), Modified Normalized Difference Water Index (MNDWI), and bare soil index (BSI) are the indices used in addition to aid the accurate LULC classification. Results showed that the performance of RF is better than CART in all the years. Thus, RF algorithm outputs are used to infer the change in the LULC for three decades. The changes in the NDVI values indicate the loss of vegetation for the urban area expansion during the study period. The increasing value of NDBI and BSI in the state indicates growth in high-density built-up areas and barren plains. The slight reduction in the value of MNDWI indicates the shrinking water bodies in the state. The results of LULC showed the urban expansion (158.2%) and loss of agricultural area (15.52%) in the region during the study period. It was noted that the area of the barren class and the water class decreased steadily from 1990 to 2020. The study adopted a dynamic classification method using time-varying land cover data to classify the urban and rural grids. This approach provided a more comprehensive understanding of the urbanization process in Kerala over the past three decades. The study incorporated 24 extreme climate indices, including 12 temperature extremes and 12 precipitation extremes, tailored to the unique climatic features of Kerala state. Various statistical methods were applied to investigate changes in long-term trends and spatial variation of heat wave characteristics. Additionally, the research aimed to examine the spatio-temporal variation of heat wave-drought compound events in Kerala from 1951 to 2020. A comparative analysis was conducted to assess the intensity and duration of heat waves within compound events compared to individual heat wave events, providing insights into their distinct characteristics and patterns. By comparing the trends between urban and nearby rural grids, the study employed a method to estimate the impact of urbanization on climate extremes. This approach yielded valuable insights into the changes occurring in extreme weather events in Kerala and their connection to urbanization, enhancing our understanding of this relationship. Long-term trend analysis of extreme climate indices showed that more than 60% of the grids in the region had experienced a decrease in indices such as CWD, R10, R20, R25, RX5 day, PRCPTOT, and SDII, whereas 80% of grid points showed an increasing trend in indices like R95p, R99p, RX1 day, and R50. The rising incidence of CDD, coupled iii with a declining number of CWD occurrences in the state, signifies an extended period of drought conditions in Kerala. An increasing trend is observed in the extreme hot temperature indices. In contrast, all extreme cold indices demonstrate decreasing trends, indicating a concerning rate of decline in cold extremes alongside the rise in warm extremes. Results pointed out that urbanization has decreased light rain and increased extreme precipitation in urban areas. It is seen that urbanization has a statically significant positive effect on heat waves and compound events, with more substantial impacts observed for HWD, HWF, and HWL compared to HWA.