Development of a New Soil Moisture Index Using SMOS Satellite Soil
Moisture Products: Case study in Southwestern Mongolia
Abstract
A new soil moisture index for monitoring drought occurrence and
intensity is presented. The index is based on the integration of
different remote sensing products and in situ observations. Due to a
shortage of precipitation, droughts reduce vegetation productivity, and
thus, aggravate the impact of moisture stress on pastureland. The
spatial distribution of soil moisture index with high-resolution images
in Mongolia is still being one of the essential goals in the remote
sensing and rangeland community. Specifically, we examined a new
composite Gobi soil moisture index (GI) based on the combination of
Ocean Salinity (SMOS) Soil Moisture, several products from the MODIS
satellite, and in situ Soil Moisture (SM) observations. A multiple
linear regression method was used for the estimation of GI soil moisture
index. The former includes the surface soil moisture from the Soil
Moisture and Ocean Salinity (SMOS) mission, the Moderate Resolution
Imaging Spectroradiometer (MODIS) derived land surface temperature
(LST), normalized difference vegetation index (NDVI), potential
evapotranspiration (PET). The latter includes a standardized
precipitation index (SPI) from in-situ data. The validation of the
approach is based on the relationship between SPI and in-situ soil
moisture (SM) observations, and their comparison to remote sensing (RS)
– derived indices. The results show that the correlation was
statistically significant between GI and in-situ SM observations from
the meteorological stations at 10 – 15 cm depths (p <
0.0001). The correlation between GI and SPI, as represented by the
correlation coefficient (r) was 0.64. The GI empirical equations that
utilize at least three key atmospheric variables are (a) NDVI, (b) land
surface temperature, and (c) potential evapotranspiration. The
established new GI soil moisture index was retrieved at the 1 km spatial
resolution for Southwest Mongolia from 2000 to 2018, and their two
summer months (July, August) were used for monitoring drought and
vegetation response to the varying soil/climatic conditions. Now, based
on the assessment of drought severity, the new soil moisture index
allowed us to assess a large-scale spatial coherence of droughts across
the Southwestern part of Mongolia.