Spectroscopic Characterization of Plant Cover in El-Fayoum Governorate, Egypt

Document Type : Original Article

Authors

National Authority for Remote Sensing and Space Sciences (NARSS), 23 Joseph Tito Street, El-Nozha El-Gedida, P.O. Box: 1564 Alf Maskan, Cairo, Egypt.

Abstract

EL- Fayoum region is the unique Egyptian western desert depression provided by surface Nile water. It’s one of the global oldest agricultural provinces. Hyperspectral remote sensing plays an important role in assessing the biophysical characters of plant. The current investigation is one of the early attempts to study the hyperspectral characteristics of cultivated crops and wild plants in El-Fayoum Governorate. The work assists in understanding the status of the vegetation structure responding to water scarcity in the study area. Spectral reflectance of cultivated & wild plants was recorded ASD field spec spectroradiometer device and integrated with lab analyses using statistical analyses of SPSS and JMP Software. Optimal waveband and optimal wavelength were calculated by ANOVA and Tukey’s analysis for discriminating of plants in the study area. Remotely sensed hyperspectral data were processed to produce spectral indices of plants to estimate the plant's vitality. Results indicated that Near Infrared (NIR) and Short Wave Infrared II (SWIR II) spectral regions were optimal to discriminate investigated taxa. In the Blue spectral zone no significant reflection was noticed. Spectral characteristics for the Mangifera indica (infected plant) indicated water stress. Spectral Reflectance analysis for Ablmoschus esculentus indicated plant suffered chlorophyll decrease. Plant Senescence Reflectance Index (PSRI) and Moisture Stress Index (MSI) for Mangifera indica (infected plant) and Abelmoschus esculentus (infected plant) were high and indicated that two plants suffer canopy stress. Also, results showed that Citrus sinensis has the highest value of NDVI (0.82) and CRII (7.99) between other plants. It could be concluded that the study of spectral signature is rather valuable in characterizing vegetation cover. Also, growth conditions and the environment can be predicted via spectral characterization curves.

Keywords

Main Subjects