Mapping the evapotranspiration (ET) in Sussex County using satellite remote sensing data

Project Status
Project Contact(s)

DGS is collaborating with climate scientist Kevin Brinson (DEOS) and Tracy DeLiberty (Dept. of Geography) to develop and test methods to estimate and map annual and seasonal distribution of evapotranspiration (ET) for Sussex County, Delaware.   Remotely sensed data from Landsat 7 ETM+ and MODIS platforms will be used to estimate regional energy balance and water flux.  These estimates are calibrated by comparison to ET estimates determined by direct point measurements (Eddy Covariance and atmometer) and models driven by meteorological data such as temperature, relative humidity, wind speed, and soil moisture.  The results have the potential to improve accuracy and precision of ET models and will be valuable for efforts that use water budgets for resource management, agriculture, wetland assessment, and research.  
A final report was submitted for publication as a DGS Report of Investigations in January 2019.

Evapotranspiration (ET), the sum of evaporation (E) from the soil and transpiration (T) water loss from plants, is a main component of the water cycle. In Delaware, ET could account for approximately two-thirds of the annual average water budget (Johnston, 1976). Estimate of ET at the monthly - or seasonal - and regional scales are valuable for water resource management and watershed hydrologic studies. Direct measurement of ET is difficult. It requires not only specific devices such as a weighing lysimeter or eddy covariance instrument (ECI), but also skilled research personnel to get accurate data in a small area (Allen et al., 1998). To overcome these limitations, a variety of indirect methods have been developed in the past decades. Among them, remote sensing based methods are regarded as the best technology to efficiently provide regional estimates of ET. In March 2015, The Delaware Geological Survey (DGS), collaborating with Delaware Environmental Observing System (DEOS), installed an ECI at Warrington Farm in Sussex County.

An example of NDVI map that was generated from Landsat 7 data obtained on 8/14/2002

The new ECI can accurately measure ET from an upwind area that has a mix of crop and forest land typical of southern Delaware. Combining information derived from remote sensing in conjunction with DEOS weather station data and ECI measurements, we will develop and test a method to estimate the annual/seasonal ET for Sussex County. In this study, the Surface Energy Balance Algorithm for Land (SEBAL) model will be used to compute actual evapotranspiration (ET_a) from visible, near infrared, and thermal infrared bands of Landsat 7 ETM+ or MODIS satellite imagery (Landsat 7 ETM+ data are produced on a 30m pixel and MODIS data have a 500m pixel). A seasonal evapotranspiration map that covers an entire growing season then can be derived by extrapolating the instantaneous ET_a to the whole season in combining with climatic data collected from densely distributed weather stations in Sussex County, Delaware.

References: Allen, R. G., Pereira, L. S., Raes, D. and Smith, M., Crop Evapotranspiration: Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper No 56, FAO, Rome, 1998. Johnston, R. H., Relation of Ground Water to Surface Water in Four Small Basins of the Delaware Coastal Plain: Delaware Geological Survey Report of Investigation No. 24, 1976.