Remote sensing-based cropping pattern identification and its impact on groundwater use in canal command areas of an irrigated agriculture region in Pakistan

Efficient water and land management is crucial for sustainable agriculture, particularly in regions facing growing water scarcity and urbanization pressures. This study aims to evaluate the cropping patterns and irrigation water use in eight Canal Command Areas (CCAs) of Bari Doab, an agriculture-dominated region in Punjab, Pakistan. Based on Sentinel-2 satellite imagery and field survey results, cropping patterns across all CCAs during the Rabi and Kharif seasons from 2018 to 2023 were identified using the Random Forest (RF) algorithm, which achieved high crop classification accuracy with the overall accuracy reaching 89.9 % for the Rabi and 90.1 % for the Kharif season crops. Producer and user accuracies ranging from 90.3 %–90.8 % and 88.6 % underscore the reliability of the classification approach. Crop water requirements are estimated according to crop classification results in each CCA using the Penman-Monteith method, revealing higher crop evapotranspiration (ET c ) for Kharif crops than Rabi crops, which is driven by seasonal climatic differences. Spatial analysis indicated consistent cropland decline in CCA3 and CCA7 due to urbanization, with statistical cropland areas decreasing by 127 km² and 96 km² between 2018 and 2023, respectively. Groundwater abstraction increased steadily across all CCAs, with the highest increasing rates observed in southern regions cultivating water-intensive crops. Groundwater storage anomalies (GWSA) were estimated from observed groundwater level data and the terrestrial water storage (TWS) component of Gravity Recovery and Climate Experiment (GRACE) and its follow-on mission (GRACE-FO) by subtracting non-groundwater components acquired from the Global Land Data Assimilation System GLDAS v2.2. GWSA anomalies revealed long-term depletion trends in all CCAs, with the steepest declines in CCA7 and CCA8 despite moderate abstraction for crops. This indicates that population-driven groundwater stress plays a significant role in these areas. The findings provide valuable insights for policymakers and stakeholders to balance agricultural demands with water resource sustainability in arid and semi-arid regions. • Random forest was used in GEE for precise multi season crop classification. • Crop water requirement was estimated using the FAO Penman-Monteith method. • Groundwater abstraction was assessed through a comprehensive analysis and validated by using GRACE-derived GWSA in each CCA. • Urban expansion reduced cropland in CCA3 and CCA7. • Groundwater reliance increased in areas cultivating water intensive crops.