Field-derived spectra of salinized soils and vegetation as indicators of irrigation-induced soil salinization

Salinization is a major cause of soil degradation ill the Murray-Darling Basin of Australia. The objective of this research is to evaluate the utility of field-derived spectra of saline soils and related vegetation for characterizing and mapping the spatial distribution of irrigation-induced soil salinization. A FieldSpee FR hand-held spectrometer was used to measure the spectra of a range of salinized soils and associated vegetation. Strategies for mapping field-derived spectra using hyperspectral (HyMap) imagery were assessed, and a continuum-removed Spectral-Feature-Fitting (SFF) approach adopted. Field-derived spectra of the vegetation comprising of samphire, sea blite, and native grass species are also useful indicators of salinization: however, their absence is not necessarily ail indicator of healthy soils. Distribution maps created using the SFF method and a restricted wavelength range of field-derived spectra provide ail accurate record of the distribution of both vegetation and soil indicators of salinization at the time of image acquisition. Salinized soil and vegetation indicator class maps show a similar spatial distribution to soil salinization as mapped by ground-based geophysical surveys. (C) 2002 Elsevier Science Inc. All rights reserved.