To maximize crop production, reference evapotranspiration (ET₀) measurement is crucial for managing water resources and planning crop water needs. The FAO-PM56 method is recommended globally for estimating ET₀ and evaluating alternative methods due to its extensive theoretical foundation. Numerous meteorological parameters, needed for ET₀ estimation, are difficult to obtain in developing countries. Therefore, alternative ways to estimate ET₀ using fewer climatic data are of critical importance. To estimate ET₀ with alternative methods, difference climatic parameters of temperatures, relative humidity (maximum and minimum), sunshine hours, and wind speed for a period of 20 years from 1996 to 2015 were used in the study. The data were recorded by 11 meteorological observatories situated in various climatic regions of Pakistan. The significance of the climatic parameters used was evaluated using sensitivity analysis. The machine learning techniques of single decision tree (SDT), tree boost (TB) and decision tree forest (DTF) were used to perform sensitivity analysis. The outcomes indicated that DTF-based models estimated ET₀ with higher accuracy and fewer climatic variables as compared to other ML techniques used in the study. The DTF technique, with Model 15 as input, outperformed other techniques for the most part of the performance metrics (i.e., NSE = 0.93, R² = 0.96 and RMSE = 0.48 mm/month). The results indicated that the DTF with fewer climatic variables of mean relative humidity, wind speed and minimum temperature could estimate ET₀ accurately and outperformed other ML techniques. Additionally, a non-linear ensemble (NLE) of ML techniques was further used to estimate ET₀ using the best input combination (i.e., Model 15). It was seen that the applied non-linear ensemble (NLE) approach enhanced modelling accuracy as compared to a stand-alone application of ML techniques (R² Multan = 0.97, R² Skardu = 0.99, R² ISB = 0.98, R² Bahawalpur = 0.98 etc.). The study results affirmed the use of an ensemble model for ET₀ estimation and suggest applying it in other parts of the world to validate model performance.

Land use and land cover (LULC) changes are significantly impacting the natural environment. Human activities and population growth are negatively impacting the natural environment. This negative impact directly relates to climate change, sustainable agriculture, inflation, and food security at local and global levels. Remote sensing and GIS tools can provide valuable information about change detection. This study examines the correlation between population growth rate and LULC dynamics in three districts of South Punjab, Pakistan—Multan, Bahawalpur, and Dera Ghazi Khan—over a 30-year period from 2003 to 2033. Landsat 7, Landsat 8, and Sentinel-2 satellite imagery within the Google Earth Engine (GEE) cloud platform was utilized to create 2003, 2013, and 2023 LULC maps via supervised classification with a random forest (RF) classifier, which is a subset of artificial intelligence (AI). This study achieved over 90% overall accuracy and a kappa value of 0.9 for the classified LULC maps. LULC was classified into built-up, vegetation, water, and barren classes in Multan and Bahawalpur, with an additional “rock” class included for Dera Ghazi Khan due to its unique topography. LULC maps (2003, 2013, and 2023) were prepared and validated using Google Earth Engine. Future predictions for 2033 were generated using the MOLUSCE model in QGIS. The results for Multan indicated substantial urban expansion as built-up areas increased from 8.36% in 2003 to 25.56% in 2033, with vegetation and barren areas displaying decreasing trends from 82.96% to 70% and 7.95% to 3.5%, respectively. Moreover, areas containing water fluctuated and ultimately changed from 0.73% in 2003 to 0.9% in 2033. In Bahawalpur, built-up areas grew from 1.33% in 2003 to 5.80% in 2033, while barren areas decreased from 79.13% to 74.31%. Dera Ghazi Khan expressed significant increases in built-up and vegetation areas from 2003 to 2033 as 2.29% to 12.21% and 22.53% to 44.72%, respectively, alongside reductions in barren and rock areas from 32.82% to 10.83% and 41.23% to 31.2%, respectively. Population projections using a compound growth model for each district emphasize the demographic impact on LULC changes. These results and findings focus on the need for policies to manage unplanned urban sprawl and focus on environmentally sustainable practices. This study provides critical awareness to policy makers and urban planners aiming to balance urban growth with environmental sustainability.

In arid regions, irrigated agriculture is mainly dependent on groundwater. In Pakistan, 73% of agricultural land is directly or indirectly irrigated through groundwater. In Punjab (Pakistan), 1.2 million private tube wells are operating, mainly extracting 90% of the country’s groundwater. Most of these wells are poorly designed due to improper site investigations and poor estimations of the aquifer’s hydraulic parameters. As a result, most wells become dry, causing considerable financial losses to farmers. Hence, optimizing the well-designed parameters through proper soil investigations is essential. This research aims to develop a statistical model for estimating the hydraulic conductivity of soil through on-site investigation: five sites were selected in Multan (Pakistan), and seven samples were collected at each location from 3, 6, 9,12,15,18, and 21 m depth. For hydraulic conductivity, soil texture, and porosity, soil laboratory tests were carried out. Finally, a statistical model was developed using hydrological parameters such as average grain size distribution (D₅₀), uniformity coefficient (U), and porosity (n). Statistically computed hydraulic conductivity was verified with experimentally measured and empirically derived hydraulic conductivity. Statistically measured hydraulic conductivity showed closer agreement with experimentally measured hydraulic conductivity than the empirically measured hydraulic conductivity: root mean square error (RMSE), correlation coefficient (Cc), and mean absolute error (MAE) are, respectively, equal to 0.013, 0.93, and 0.011.