摘要
The optimization of water resources management in arid region needs reliable information and knowledge on water resources and water requirements. Remote sensing and Geographic Information System techniques are used to estimate the Irrigation Water Requirements (IWR) in the Regueb-watershed for the summer season, when the peak of the water consumption is reached. The landsat 8 images were used to identify irrigated area and to estimate the IWR in dry season. Five methods were applied for the identification of irrigated area based on supervised classification and spectral indices. Two different approaches were applied to estimate IWR: K c-NDVI method based on the relationship between the Normalized Difference Vegetation Index (NDVI) and the crop coefficient (K c) and FAO approach based on the empirical equation of Penman–Monteith and the single K c given by FAO-56. The maximum likelihood classifier (MLC) performed the highest overall accuracy (85 %) and a kappa coefficient of 0.82. The IWR resulting values range from 10 to 14.5 Mm3 in summer season. To evaluate the methodology, five test areas were selected based on the diversity of crops in the whole study area. The validation of results indicates that the IWR values calculated using FAO method were in good agreement with the IWR values derived from remote sensing approach.