Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Research Project #427013

Research Project: Fusing MODIS, Landsat, and Geostationary Data for Daily Monitoring of Crop Condition and Water Use at Field Scales

Location: Hydrology and Remote Sensing Laboratory

Project Number: 8042-13610-029-85-I
Project Type: Interagency Reimbursable Agreement

Start Date: Sep 15, 2014
End Date: Aug 31, 2018

Objective:
The overall objective is to create high temporal and spatial remote sensing data products from Moderate Resolution Imaging Spectroradiometer (MODIS), Landsat, and Landsat-like sensors, along with geostationary datasets, for time-continuous monitoring of crop condition and water use at scales resolving individual farm fields (30 m resolution). Existing MODIS land products (surface reflectance, surface temperature and leaf area index) will be fused with Landsat data to produce MODIS-consistent data products at Landsat spatial resolution. The fused MODIS-Landsat data products from this project will be used to accomplish the following tasks: (1) Map crop phenologic metrics at field scale by fusing MODIS, Landsat, and Landsat-like surface reflectance products [e.g., from Sentinel 2 or Terra Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER)] to optimize temporal sampling of annual VI curve. (2) Map daily water use and stress at field scale by fusing evapotranspiration (ET) data streams generated using MODIS, Landsat, and geostationary thermal and reflectance information. (3) Investigate impacts of drought timing/severity on yield, by combining the phenologic and water stress datasets retrieved at field scale to investigate impact of timing of stress relative to phenological stage on crop condition and yield, and to assess spatial variability in resilience to drought over agricultural landscapes.

Approach:
Several existing tools and methodologies that have been developed and prototyped by the team will be integrated and refined, each of which combines information from multiple satellite platforms and observing wavelengths. These tools/methods include a) Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) data fusion, b) MODIS normalization of Landsat-like sensor reflectances, c) DMS thermal image sharpener, d) MODIS-constrained Landsat LAI products, e) ALEXI/DisALEXI multi-scale ET retrieval algorithms, and f) the Atmospheric-Land Exchange Inverse (ALEXI)-based Evaporative Stress Index (ESI). The end goal is an integrated phenology and ET mapping system that can be readily implemented over specified regions of interest. Validation and accuracy assessment methodologies have been designed for each stage of the analysis. This is a three year project. For year 1, the focus will be on data fusion software and model integration. For year 2, the focus will be on system testing for USDA and AmeriFlux sites. For year 3, the approach for regional application in the U.S. Cornbelt will be extended.