ESTIMATION OF FOLIAR NITROGEN OF SAL AND BANJ OAK FORESTS IN WESTERN HIMALAYA USING HYPERSPECTRAL REMOTE SENSING

摘要

Nitrogen (N) is an important element for plant growth and development and excellent indicator of forest health status. Hyperspectral remote sensing, a non-destructive technique helps in understanding the forest health by estimating N from the forest ecosystem. In the present study, N has been estimated in pure patches of Oak (Quercus spp.) and Sal (Shorea robusta) species in the parts of Uttarakhand, India. Vegetation indices using different spectral band combinations were used to identify the appropriate wavelength representing N absorption features. The spectral bands at wavelengths 660 nm, 1517.82nm and 1689.30nm were found to be most suitable for these species. Among all the nitrogen indices, log normalized (1/R) nitrogen index performed better and exhibited strong positive correlations with foliar N content. For Banj oak, N was estimated to be 0.29 to 2.79 ± 0.03 t ha at R~2 = 0.82 using linear regression equation. Similarly, for Sal species, R~2 achieved maximum at 0.92 and N was estimated to be 0.020 to 0.095 ±0.01 t'ha. The study indicates the utility of hyperspectral data to assess N concentration of Sal and Banj forest for quick monitoring of the forest health. The technique can also be used in indicating the disturbance and degradation in the forest ecosystem.