Year Established: 2017 Start Date: 2017-03-01 End Date: 2018-02-28
Total Federal Funds: $5,000 Total Non-Federal Funds: $10,000
Principal Investigators: Tyson Ochsner, Chris Zou
Abstract: Soil moisture is an essential variable which affects climatic, hydrological, agricultural, and ecological systems. Due to the impact of soil moisture on important earth processes, in-situ soil monitoring networks are becoming more prevalent. However, the majority of soil moisture monitoring networks consider only one land cover type, usually grasslands, which limits the use of these data for areas with mixed land cover types. The Oklahoma Mesonet has monitored soil moisture at over 100 grassland sites for nearly two decades, but large areas of forest (12 million acres, or 28% of the state’s land area), cropland (~8 million acres, or 18%), and other land cover types have gone largely unmonitored (McPherson et al., 2007; National Agricultural Statistics Service; Oklahoma Forestry Service). While the current long-term soil moisture record is useful for a number of applications in many research areas, a major limitation of the current data is that it has been collected exclusively in grassland ecosystems and does not reflect soil moisture conditions under other land covers. However, remote sensing by satellites has led to the availability of high-resolution vegetation indices (VI) data, and we hypothesize that these data, along with in-situ meteorological data from the Mesonet, may be incorporated into a simple water balance model to effectively estimate root-zone soil moisture at sites throughout Oklahoma. These estimates may then be used to train a computational model to estimate soil moisture across the entire state, regardless of land cover. The proposed generalized soil moisture estimation method would provide new, much needed information relevant to a number of disciplines, including hydrology, water resource planning, climatology, and agriculture. The long-term goal for this project is to increase scientific understanding of the variability of soil moisture under the many cover types found throughout Oklahoma and to create a new, general method of large-scale soil moisture estimation and mapping. We will reach this goal by 1) utilizing vegetation indices (e.g., NDVI or ERI) data collected by the MODIS satellite and Mesonet meteorological data to develop an efficient computational model capable of estimating soil moisture under various land cover types found in Oklahoma, and 2) validating estimated soil moisture values using in-situ soil moisture monitoring in multiple vegetation types throughout Oklahoma.