Application of Remote Sensing Techniques for conserving Scarce Water Resources: A Case Study from Pakistan

Pakistan, which was once a water surplus, is now a water deficit country according to Malin Falkenmark criteria. The conventional wisdom of managing canal water supplies, which usually results in over- or under-irrigation, is not sufficient to meet the challenge of water demand in future. This paper introduces the use of modern tools like Remote Sensing (RS), Geographic Information Systems (GIS) and CROPWAT to improve the management of the existing irrigation systems. This study was conducted for the Pehure High Level Canal (PHLC) and the Upper Swat Canal (USC) system in the North Western Frontier Province (NWFP) of Pakistan. Crop identification at distributary level was made from multi-temporal Remote Sensing satellite images, using various image processing techniques, such as supervised, unsupervised classification and spectral mixture analysis. Cropped areas were calculated for each individual crop from these classified images, and then crop water requirement at distributary level was estimated using CROPWAT. Assuming all other parameters of the CROPWAT model optimistic, the calculated crop area was of major concern. The supervised classification with support of unsupervised classification and ground truth information has proven to be the best option and cost-effective technique for calculating the actual cropped area. The results of this study can be used while devising guidelines for water managers to release the canal supplies based on crop water requirement. This practice will help in avoiding wastage of canal water at farm level, which can be optimally used for increasing irrigated areas and crop productivity in the area.