2021 9th International Conference on Agro-Geoinformatics (Agro-Geoinformatics) | 2021

Comparison of TVDI and soil moisture response based on various vegetation indices

 
 
 
 
 

Abstract


Soil moisture is an important factor that affects crop growth, development and yield. Relative soil moisture (RSM) can express the effect of water deficit on crop growth wonderfully. Temperature vegetation drought index (TVDI) is one of the most widely used methods in agricultural drought remote sensing monitoring at present. In order to improve the accuracy of soil moisture inversion by TVDI, the performance of TVDI calculated by different vegetation indices used in drought monitoring was compared in Northeast China on the 8day scale of the whole growth period (May to September). Enhanced vegetation index (EVI), ratio vegetation index (RVI), modified soil-adjusted vegetation index (MSAVI) and leaf area index (LAI) were selected to construct TVDI. The relationship between these indices and situ relative soil moisture (RSM) were analyzed every 8 days in 2009. The results showed that the correlation between TVDIs and RSM at 20cm depth was stronger than 10cm depth averagely. All the TVDIs obtained by different vegetation indices had a certain negative correlation with soil moisture, indicating that the higher the TVDI, the lower the soil moisture and the drier. Taking the soil moisture at a depth of 20cm as an example, four indices showed relatively consistent trend: the absolute correlation coefficients $\\vert r\\vert$ in May were less than 0.3, and from June to September $\\vert r\\vert$ were around 0.5 overall. Especially in the middle of the growing season (the second half of June and Early September), $\\vert r\\vert$ were greater than 0.5. On the whole, TVDI performs best in the middle of the growing season, followed by the late growing season, and the early growing season is unsatisfactory. Among them, vegetation indices performed best when taking LAI, followed by RVI, and TVDI constructed by EVI and MSAVI were roughly the same. TVDI when LAI was calculated was higher correlated with soil moisture than RVI, EVI, MSAVI, but generally, all indices performed well after June.

Volume None
Pages 1-5
DOI 10.1109/Agro-Geoinformatics50104.2021.9530348
Language English
Journal 2021 9th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)

Full Text