Spectral indices for the detection of salinity effects in melon plants

Abstract

Water scarcity and soil salinization affect large semiarid agricultural areas throughout the world. The maintenance of agricultural productivity implies better agricultural practices and a careful selection of resistant crops. A proper monitoring of the physiological status of plants can lead to better knowledge of plant nutritional requirements. Visible and near-infrared (VNIR) radiometry provides a non-destructive and quantitative method to monitor vegetation status by quantifying chemical properties using spectroscopic techniques. In this study, the capability of VNIR spectral measurements to detect salinity effects on melon (Cucumis melo L.) plants was tested. Melon plants were cultivated under multiple soil salinity conditions (electrical conductivity, (EC)1:5: 0.5, 1.0 and 2.5 dS m-1). Spectral data of leaves were transformed into vegetation indices indicative of the physiological status of the plants. The results showed differences for N (p < 0.05), K and Na content (p < 0.01) due to salinity suggesting different degrees of salt stress on the plants. Specific leaf area increased with salinity levels (p < 0.001). The capabilities of VNIR radiometry to assess the influence of soil salinity on melon physiology using a non-destructive method were demonstrated. A normalized difference vegetation index (NDVI750-705), and the ratio between water index (WI) and normalized difference vegetation index (WI/NDVI750-705) showed significant relationships (p < 0.01) with the salinity. Therefore, this method could be used for in-situ early detection of salinity stress effects.