Main Article Content
Crop identification is vital to make an inventory of the crops grown in a given area and their cultivation period. The Remote sensing (RS) techniques can provide information on the distribution of cultivated land, crop types, and areas for the agricultural sector's effective management.
In remote sensing, various vegetation indices (VI) can analyze and evaluate multiple phenomena and themes. The Normalized Difference Vegetation Index (NDVI) is an essential and highly significant remote measurement widely used in agriculture for phenological monitoring and crop health (Ray and Dadhwal, 2001).
In this work, we present a methodology for the contribution of NDVI from Landsat 7 (TM) and (ETM+) images to crop mapping in the Gharb region using a classification based on the pixel approach and estimating crop coefficient from NDVI.
The classification results concern six main types of crops planted in this region (beet, maize, sugar cane, market gardening, cereals, and rice). The classification map showed differences in agricultural practices adopted by farmers in crop spatial distribution. The classification results showed the ability of this methodology to discriminate between crops. The precision of classification for all classes is 90%, and the kappa indicator index is 0.88, which indicates a good classification.
Crop coefficients were deduced from the NDVI extracted from the images. Due to meteorological data collected from the meteorological station TCSC of SK Tlet, the estimation of the reference evapotranspiration was made and subsequently the potential evapotranspiration of each crop during the agricultural season 2019-2020.
The highest values for ETC were obtained when the crop was in its full development when water was mainly lost through transpiration after a slight decrease in the ratio values observed during the phase of the vegetative cycle (maturity).
The water requirements (daily, monthly and annual) for the crops were determined and their electrical energy consumption.