Project Funding: 2011 - 2014

NRA: 2009 NASA: The Science of Terra and Aqua   

Funded by NASA

Abstract:
Routine simulations of land-surface fluxes are often provided at too coarse spatial resolution to be of real value for applications that require both high spatial resolution and frequent coverage such as field-scale drought monitoring and water resource management, yield forecasting and crop-growth monitoring. To obtain high resolution in both time and space we propose to use a multi-scale and multi-sensor data fusion approach that capitalizes on the spatial detail of Landsat (30 m) and high temporal frequency of MODIS (MODerate resolution Imaging Spectroradiometer) (up to two times daily) observations. Spatial downscaling of land surface models (LSM) to perform routine carbon and water cycle predictions at very fine resolution requires a robust and realistic representation of soil-vegetation-atmosphere dynamics at that scale. LSMs that include a prognostic modeling of the soil water balance face serious challenges related to providing a spatially adequate soil surface characterization. Thermal infrared (TIR) data provide valuable information about the sub-surface moisture status, and land surface temperature can be an effective substitute for in-situ surface moisture observations and a valuable metric for constraining land surface fluxes at sub-field scales. Routine thermal-based flux mapping at high spatial resolution is however significantly hindered by the low temporal resolution (16-day revisit time) of thermal band sensors such as Landsat. As a consequence, we will use and refine the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) to obtain time-continuous datasets of land surface fluxes at Landsat spatial resolution. We propose to use a multi-scale thermal-based land surface modeling system to effectuate regional to local downscaling of carbon, water and heat fluxes by using a combination of TIR and shortwave reflective imagery from GOES (Geostationary Operational Environmental Satellite), MODIS and Landsat. Key biophysical vegetation properties will be retrieved at 30 m resolution using a blended surface reflectance dataset as input to the REGularized canopy reFLECtance (REGFLEC) tool. REGFLEC facilitates retrievals of leaf chlorophyll (Cab), a biophysical parameter that is being increasingly recognized as a key for quantifying variability in photosynthetic efficiency. Cab will be used to delineate spatio-temporal variations in nominal light-use-efficiency, which serves as a fundamental modulator of carbon and water fluxes in the adopted LSM. The thermal-based modeling system will be applied to targeted regions within the continental U.S. and flux simulations will be compared with flux tower observations and independent model output. We expect that the fusion of Landsat and MODIS data streams will facilitate high fidelity carbon, water and heat flux mapping at Landsat spatial resolution and at a temporal frequency not otherwise possible.

2013 NASA Terrestrial Ecology Science Team Meeting Poster(s)

• Mapping land-surface fluxes of carbon, water and energy from field to regional scales   --   (Mitchell Schull, Martha Anderson, Bill Kustas, Carmelo Cammalleri, Feng Gao, Rasmus Houborg)   [abstract]   [poster]

More details may be found in the following project profile(s):

• NASA Terrestrial Ecology (TE) Project Profile