Monitoring Vegetation Phenological Cycles in Two Different Semi-Arid Environmental Settings Using a Ground-Based NDVI System: A Potential Approach to Improve Satellite Data Interpretation

In semi-arid environmental settings with sparse canopy covers, obtaining remotely sensed information on soil and vegetative growth characteristics at finer spatial and temporal scales than most satellite platforms is crucial for validating and interpreting satellite data sets. In this study, we used a ground-based NDVI system to provide continuous time series analysis of individual shrub species and soil surface characteristics in two different semi-arid environmental settings located in the Great Basin (NV, USA). The NDVI system was a dual channel SKR-1800 radiometer that simultaneously measured incident solar radiation and upward reflectance in two broadband red and near-infrared channels comparable to Landsat-5 TM band 3 and band 4, respectively. The two study sites identified as Spring Valley 1 site (SV1) and Snake Valley 1 site (SNK1) were chosen for having different species composition, soil texture and percent canopy cover. NDVI time-series of greasewood (Sarcobatus vermiculatus) from the SV1 site allowed for clear distinction between the main phenological stages of the entire growing season during the period from January to November, 2007. NDVI time series values were significantly different between sagebrush (Artemisia tridentata) and rabbitbrush (Chrysothamnus viscidiflorus) at SV1 as well as between the two bare soil types at the two sites. Greasewood NDVI from the SNK1 site produced significant correlations with chlorophyll index (r = 0.97), leaf area index (r = 0.98) and leaf xylem water potential (r = 0.93). Whereas OPEN ACCESS Remote Sens. 2010, 2 991 greasewood NDVI from the SV1 site produced lower correlations (r = 0.89, r = 0.73), or non significant correlations (r = 0.32) with the same parameters, respectively. Total percent cover was estimated at 17.5% for SV1 and at 63% for SNK1. Results from this study indicated the potential capabilities of using this ground-based NDVI system to extract spatial and temporal details of soil and vegetation optical properties not possible with satellite derived NDVI.