Carlson Auditorium, 76-1125
December 4, 2019 at 3:30am

Presented by Dr. Alisa Coffin
Research Ecologist with the USDA-ARS Southeast Watershed Research Laboratory in Tifton, Georgia
Adjunct Assistant Professor, University of Georgia

Abstract

The Long-Term Agroecosystem Research (LTAR) network was established in 2012 to facilitate long term research into the sustainable intensification of agriculture in the United States of America (US). Research sites established at eighteen locations across the conterminous US are engaging in a common experimental strategy to evaluate existing and potential agricultural production systems across domains of production, environmental management and rural prosperity. Experimental work at LTAR locations occurs at plot, field and enterprise scales, providing highly detailed information in both space and time about the responses of production systems to experimental manipulations.

To accomplish this work, remote sensing is used to extrapolate local measurements to broader landscape and regional scales, and to record observations across broader areas where direct measurements cannot be obtained. Currently, sites within the LTAR network are using remote sensing for a variety of applications related to crop and forage production. In addition to freely available mid- and coarse- resolution satellite imagery, LTAR remote sensing tools include sensors affixed to towers, unmanned aerial systems, and field-portable spectroradiometers. Remote sensing products have been or are being used in LTAR to: model production in working lands across broad areas; estimate regional water use and stress, snowpack, soil moisture and productivity; map fine-scale vegetation patterns in grazing lands; assess pollinator habitat near crop lands; and capture continuous phenology of crop and forage species.

Considerable challenges being addressed by the LTAR network are related to: data coordination and management of big data across the network; identifying and testing workflows that provide statistically valid and meaningful results about agricultural sustainability and food security; and deploying artificial intelligence that enables the balanced integration of small, slow, clear data streams with big, fast, noisy data rivers to derive useful information that can guide US agricultural policy. Clearly, remote sensing is an indispensable tool for agricultural research, but the role of remote sensing in resolving these data integration issues for LTAR is still taking shape. Addressing these challenges will require considering how remote sensing contributes to LTAR data products, ensuring that image processing methods, including calibration and quality control, are incorporated into in LTAR research workflows, and considering how remote sensing products provide a spatio-temporal framework for integrating diverse data sources.