Mapping the Duration and Severity of Drought Impacts on Grasslands in the Southern Great Plains through a Water-related Vegetation Index Derived from MODIS Imagery

Abstract:

Grasslands in the Southern Great Plains of the United States are sensitive to drought. Previous studies have utilized a variety of methods and indices to monitor drought and assess the impacts. These drought indices were based on climate data such as precipitation and surface air temperature. However, water-related vegetation indices such as land surface water index (LSWI) have not been widely used as an indicator of drought severity at large scales. In this study, we analyzed land surface temperature (LST) data (MYD11A2) and LSWI data (MOD09A1) from 2006 and 2007 (H10V05, covering parts of Oklahoma and Kansas) to assess drought. First, LST data from an entire year were used to define and map the temperature-defined plant growing season (start and end dates as well as duration of LST > 5 °C). Second, LSWI data from within the LST-based growing season were used to assess vegetation growing condition and delineate the LSWI-based growing season (between the first date and the last date when LSWI > 0). The number of days with LSWI < 0 within the LSWI-based growing season is around zero in a year without summer droughts (e.g. 2007), but increases substantially in a year with severe summer droughts (e.g. 2006), varying between 100 to 150 day. Four levels (severe drought, moderate drought, abnormally dry, and no drought) of drought indicators were used to define the drought severity of grasslands based on the length of the LSWI-based growing season and the total number of LSWI < 0. Preliminary results revealed that the frequency of LSWI < 0 within the LSWI-defined growing season corresponded well with the drought condition in 2006. The LSWI-based approach developed in this study, at a spatial resolution of 500-m and a temporal resolution of 8-day, may provide an additional drought indicator in the Southern Great Plains.