The pressing issue of global climate change is being rigorously examined, with urban heat islands (UHIs) identified as a contributing factor. A UHI is a city or town area exhibiting a temperature variance from its surrounding environment. Researchers employ various relative UHI parameters to model UHI data and predict temperature fluctuations. For the study area of Srinagar City, Jammu & Kashmir, India, land surface temperature (LST) data and its correlated parameters were sourced from satellite imagery. The LST data for the region were analyzed to understand the development and investigate the UHI effect and its variations. Utilizing an 8-day revisit period during the busiest season month, the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite provided LST data from 2001 to 2021. To address the large dataset, which includes 16 samples of LST data per square kilometer each year measured in Kelvin (K), different machine learning (ML) techniques were employed to establish associations for UHI modeling. These clusters were then compared with established scientific categories in the study area. The long short-term memory neural network (LSTM) model, using time-series data, predicted changes in urbanized areas, vegetation cover, wetlands, and other factors. For land use land cover (LULC) prediction, the neural network (NN) model outperformed all others, with Regression (R) = 0.897, Validation = 0.912, and Training = 0.931, and mean squared error (MSE) = 2.012, Validation = 0.191, and Training = 1.124. This paper strives to identify and analyze the relationship between LULC change and LST variations in the context of urbanization. Initially, it examines the correlations between LST and variables such as vegetation, man-made structures, and agriculture, employing built-up indices within each LULC category and vegetation cover. Subsequently, it assesses the impacts of LULC change and urbanization on UHI using hot spot and urban landscape analyses. Finally, it proposes a model employing non-parametric regression to predict future urban climate trends, considering anticipated changes in land cover and land use.