NDVI saturation adjustment: A new approach for improving cropland performance estimates in the Greater Platte River Basin, USA

• Published in: Ecological indicators Vol. 30; pp. 1 - 6
• Main Authors: Gu, Yingxin, Wylie, Bruce K., Howard, Daniel M., Phuyal, Khem P., Ji, Lei
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01-07-2013; Elsevier
• Subjects: Animal and plant ecology; Animal, plant and microbial ecology; Applied ecology; Biological and medical sciences; Conservation, protection and management of environment and wildlife; Cropland performance; Ecosystem productivity; eMODIS; Fundamental and applied biological sciences. Psychology; General aspects; Greater Platte River Basin; NDVI saturation adjustment; Parks, reserves, wildlife conservation. Endangered species: population survey and restocking; RVI; Synecology; Vegetation index; United States; North America; America; Greater Platte River Basin; NDVI saturation adjustment; Vegetation index; RVI; Cropland performance; Ecosystem productivity; eMODIS; Productivity; Cultivated field; Ecological indicator; River basins; Ecosystem; Performance; Environmental monitoring; Environmental protection
• ISSN: 1470-160X; 1872-7034
• DOI: 10.1016/j.ecolind.2013.01.041

► We adjusted near-saturated NDVI based on the relationship between NDVI and RVI. ► We mapped cropland performance (CP) in the GPRB using saturation adjusted NDVI. ► The new saturation adjusted CP map improves the quality of the original CP map.
In this study, we developed a new approach that adjusted normalized difference vegetation index (NDVI) pixel values that were near saturation to better characterize the cropland performance (CP) in the Greater Platte River Basin (GPRB), USA. The relationship between NDVI and the ratio vegetation index (RVI) at high NDVI values was investigated, and an empirical equation for estimating saturation-adjusted NDVI (NDVIsat_adjust) based on RVI was developed. A 10-year (2000–2009) NDVIsat_adjust data set was developed using 250-m 7-day composite historical eMODIS (expedited Moderate Resolution Imaging Spectroradiometer) NDVI data. The growing season averaged NDVI (GSN), which is a proxy for ecosystem performance, was estimated and long-term NDVI non-saturation- and saturation-adjusted cropland performance (CPnon_sat_adjust, CPsat_adjust) maps were produced over the GPRB. The final CP maps were validated using National Agricultural Statistics Service (NASS) crop yield data. The relationship between CPsat_adjust and the NASS average corn yield data (r=0.78, 113 samples) is stronger than the relationship between CPnon_sat_adjust and the NASS average corn yield data (r=0.67, 113 samples), indicating that the new CPsat_adjust map reduces the NDVI saturation effects and is in good agreement with the corn yield ground observations. Results demonstrate that the NDVI saturation adjustment approach improves the quality of the original GSN map and better depicts the actual vegetation conditions of the GPRB cropland systems.