期刊名称：Journal of Geography, Environment and Earth Science International
印刷版ISSN：2454-7352
出版年度：2017
卷号：12
期号：4
页码：1-12
DOI：10.9734/JGEESI/2017/37283
语种：English
出版社：Sciencedomain International
摘要：Agriculture monitoring and managements is a key factor in the food production and food security. Mainly, crop identification and area quantification are most important factors in yield estimation and predictions. In arid lands, water is a limiting factor for the agriculture expansion and development. Conventional methods for both crop discrimination and crop water requirements are very expensive and unbearable economically. Remote sensing has been employed several decades ago in the different agricultural activities. Crop discrimination, water requirements and even weed and pest control could be achieved via remote sensing and geographical information systems. This paper utilizes remote sensing data in combination with ground meteorological data to calculate the Actual Crop Evapotranspiration (ETa) under modern irrigation systems conditions. Moreover, it also tries to discriminate between different crops and calculate area per crop type. Four Landsat8-OLI images were used to calculate the Land Surface Temperature (LST) during the different growth stages of the 2014 winter corps season. The dates of these satellite images were chosen to fall in the different growth stage of the crops in the study area. Ground meteorological data were used to estimate Reference Evapotranspiration (ETo) using the FAO Penman-Monteith (FPM) equation. Land surface temperature and Air Temperature (Tair) were used to observe the water distribution conditions of the study area by the means of mapping the Water Deficit Index (WDI). The WDI and Potential Crop Evapotranspiration (ETc) were used to calculate ETa. The supervised maximum likelihood classification method was employed for crop mapping using spectral signatures collected from different ground training sites through different field visits during the growing stages of the growing season. The use of multi-temporal Normalized Difference Vegetation Index (NDVI) resulted in a classification accuracy of 93% with a kappa coefficient of 0.90. The crop water requirement was affected by the decreasing surface and air temperature. Crop type and different growth stages were detected through applying Multi-temporal images.
