Detecting soil salinity with MODIS time series VI data |
| |
| Institution: | 1. School of Resources and Environment, Xinjiang University, 14 Shengli Rd, Xinjiang 830046, China;2. Earth and Environmental Studies, Montclair State University, 1 Normal Ave., Montclair, NJ 07043, United States;1. Ecosystem Management, School of Environmental and Rural Science, University of New England, Armidale, NSW 2351, Australia;2. Department of Environment and Agricultural Natural Resources, King Faisal University, Hofuf 31982, Saudi Arabia;1. Chinese-Israeli International Center for Research and Training in Agriculture, China Agricultural University, Beijing 100083, PR China;2. Center for Agricultural Water Research, College of Water Resources & Civil Engineering, China Agricultural University, Beijing 100083, PR China;3. Agricultural and Biological Engineering, Purdue University, 225 South University Street, West Lafayette, IN 47907, USA;1. Desert Research Center, Cairo, Egypt;2. Department of Plant and Soil Science, Texas Tech University, Lubbock, TX, USA;3. Ramakrishna Mission Vivekananda University, Kolkata, India;4. Louisiana State University, Baton Rouge, LA, USA |
| |
| Abstract: | Mapping of salinization using the satellite derived vegetation indices (VIs) remains difficult at broad regional scales due to the low classification accuracy. Satellite derived VIs from the Moderate Resolution Imaging Spectroradiometer (MODIS) have more potential because the MODIS balances the requirements of spatial detail, spectral and temporal density and tends to reflect vegetation responses through time. However, the relationship between MODIS data and salinity may be underestimated in previous studies because the MODIS time series data were not investigated thoroughly, especially regarding vegetation phenology. This study assessed the applicability of MODIS time series VI data for monitoring soil salinization with a series of MODIS pixels selected in the Yellow River Delta, China. The hidden information in vegetation phenology was investigated by improving the quality of VIs time series data with the Savitzky–Golay filter, extracting the phenological markers and differentiating VIs time series data based on vegetation types. The results showed that the quality of the enhanced vegetation index (EVI) time series data were improved by the Savitzky–Golay filter, which could provide more accurate thresholds of phenological stages than the empirical definition. The seasonal integral of EVI (EVI-SI) extracted from the smoothed EVI time series profile was verified as the best indicator of the degree of soil salinity. Additionally, the correlation of EVI-SI and soil salinity was highly dependent on land cover heterogeneity, and the ranges of correlation coefficients were as high as 0.59–0.92. EVI-SI was linearly correlated with ECe in cropland with a high model fit (R2 = 0.85). The relationship of EVI-SI and ECe fit best with a binomial line and EVI-SI was able to explain 70% of the variance of ECe. Despite the poor fit of the linear regression model in mixed sites limited by spatial resolution (R2 = 0.32), MODIS time series VI data, as well as the extracted seasonal parameters, still show great potential to assess large-scale soil salinization. |
| |
| Keywords: | MODIS time series data Phenological metrics Seasonal integral vegetation index The Yellow River Delta Vegetation heterogeneity |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
