Assessment of MODIS-Derived NDVI and EVI for Different Vegetation Types in Arid Region: A Study in Sirjan Plain Catchment of Kerman province, Iran

Abstract

Trend of vegetation in typology scale has remained uncertain, especially in arid
ecosystems. Therefore, we aimed to assess two vegetation indices including the Normalized
Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI). Trends and their rates
of indices in vegetation types were analyzed by the Mann–Kendall test and Theil–Sen trend
estimator. To this end, we determined the typology maps of Sirjan plain catchment in Kerman
province, Iran based on field operations; then, Moderate Resolution Imaging Spectroradiometer
(MODIS) monthly data with the spatial resolution at 250 × 250 m from 2000 to 2019 were
organized into the NDVI and EVI time series to investigate the trend. The results showed that some
vegetation types which had no distinct trend were random and time-independent; however, the
Alhagi camelorum type demonstrated the EVI and NDVI annual increase by 0.0009 and 0.0011
year
-1
, respectively. The other types that changed dimensionally and were associated stochastically
had different slopes of increase according to Theil–Sen estimate. The most increased rate of the
NDVI was attributed to Artemisia sieberi- Amygdalus lycioides- Ebenus stellata type (0.0015 year
-
1
) whereas for the EVI Artemisia aucheri-Astragalus parrowianus, Zygophyllum eurypterum-
Artemisia sieberi- Astragalus arbusculinus, Dendrostellera lessertii-Noaea minuta, Artemisia
sieberi- Zygophyllum eurypterum and Cornulaca monacantha-Salsola orientalis types were
increased by 0.0008 year
-1
. Of note, sudden NDVI and EVI changes (out of the confidence level of
Theil–Sen estimate) were observed for the majority of types especially in 2008 and 2017. More
investigations on environmental data and human activity in the mentioned years affecting as
turbulator of majority of types are recommended.