Assessing shallow groundwater level using RTI model and long-term MODIS data in Ejina Basin, Northwest China

The present study aims to exploit the thermal inertia model to assess the shallow groundwater level distribution, as an attempt to address agricultural water shortage and ecological degradation in arid and semi-arid regions. Apparent thermal inertia (ATI) model is one of the most commonly used and successfully implemented models, whereas in some cases, the scales of the results achieved on different dates are inconsistent, which poses a serious difficulty when monitoring soil moisture over time. To address this restriction, we introduce a soil moisture indicator, called RTI, which derives an approximation of soil thermal inertia, from flexible multi-temporal MODIS observations covering a 2-year period. Taking the Ejina Basin in arid area as an example, to more accurately assess the shallow groundwater level distribution, ATI and RTI models were adopted based on the long time series MODIS data to assess the multi-temporal soil moisture content in the study area. The data of groundwater level measured by observation well and thermal inertia were employed to generate scatter diagrams. Subsequently, a nonlinear equation was developed. Based on the thermal inertia, the temporal and spatial distribution map of the shallow groundwater level in the whole study area was generated with the inverse calculation of the nonlinear equation. It was found that the shallow groundwater level distribution in Ejina Basin tended to be shallower from south to north, as well as from summer to winter. As revealed from the results, both ATI and RTI models can effectively assess the shallow groundwater level in arid areas. The RTI model could better reflect the seasonal changes of soil moisture content and groundwater level.