Assessing the feedback relationship between vegetation and soil moisture over the Loess Plateau, China

The study of the feedback relationship between vegetation and soil water is of great significance for improving the understanding of hydrological process, soil and water conservation and water resources utilization. However, the mutual feedback between vegetation and soil moisture had remained unresolved. To this end, the feedback relationships between NDVI and soil moisture (SM) over the Loess Plateau (LP) were studied by using the Granger causality method. Three layers of soil moisture data were used, with depths of 0-10 cm (SM1), 10-40 cm (SM2) and 40-100 cm (SM3), respectively. The spatial distribution characteristics of causality between NDVI and SM1/ SM2/SM3 were further analyzed. Results indicated that (1) NDVI and SM1 were positively correlated in the whole region of the LP, while NDVI and SM2/SM3 were negatively correlated in the southeast of the region; (2) compared with SM2 and SM3, SM1 and NDVI had a more significant mutual feedback relationship; (3) after the program of Grain for Green on the LP, the vegetation increased, the feedback relationship between NDVI and SM1 was further enhanced, and the mutual feedback relationship was the most obvious in the key areas of revegetation. In general, this study sheds new insights into the feedback between vegetation and soil moisture, which can help guide the adjustment of local agricultural structure and ecological restoration.

Automated summary

This study investigated the feedback relationship between NDVI and soil moisture over the Loess Plateau using the Granger causality method. The results showed that NDVI was positively correlated with soil moisture at the 0-10cm depth, while it was negatively correlated with soil moisture at the 10-40cm and 40-100cm depths. After the Grain for Green program, the feedback relationship between vegetation and soil moisture at the 0-10cm depth was further enhanced.