Abstract

Is it true that rural areas, like urban areas, experience temperature changes over time? The aim of this mission was to find the season-wise trend associated with land surface thermal alteration based on satellite imagery from the last 30 years using the least square regression method, as well as to determine the stimulus of LST for the years 2027 and 2037 using Artificial Neural Network and Support Vector Machine techniques. The average temperature in the winter, summer, and Monsoon months has risen by 0.11 °C/year, 0.19 °C/year, and 0.07 °C/year, respectively, according to the analysis. Additionally, the simulated models reveal which extreme finish temperature group (>37.13 °C) may include more areas than the current one. For example, in 2017, a total area of 28.95 km² was above the 37.13 °C temperature class, but this could increase to 37.91 km² in 2027 and 42.67 km² in 2037. Fragmentation analysis of the extreme temperature patches shows that the location of the high-temperature core steadily increases over time. The simulated water body, vegetation, built-up land, and bare land cover area show a decreasing trend in the first two parameters and an increasing trend in the last two, all of which influence temperature increase incident.

摘要

农村地区真的像城市地区一样会随着时间的推移经历温度变化吗？该任务的目的是使用最小二乘回归方法，根据过去 30 年的卫星图像，找出与地表热变化相关的季节趋势，并确定 2027 年和 2037 年 LST 的刺激因素使用人工神经网络和支持向量机技术。分析显示，冬季、夏季和季风月份的平均气温分别上升了0.11°C/年、0.19°C/年和0.07°C/年。此外，模拟模型揭示了哪个极端完成温度组 (>37.13 °C) 可能包括比当前更多的区域。例如，2017年，总面积28.95平方公里²高于 37.13 °C 温度等级，但到2027 年可能增加到 37.91 km ^{2和 2037 年 42.67 km 2}。极端温度斑块的破碎分析表明，高温核心的位置随着时间的推移稳步增加。模拟的水体、植被、建设用地和裸地覆盖面积前两个参数呈下降趋势，后两个参数呈上升趋势，均影响气温升高事件。