Source: Applied GeoSolutions, LLC submitted to
COVER CROP INFORMATION SYSTEM: USING REMOTE SENSING AND MODELING TO MAP DETAILED INFORMATION ABOUT COVER CROPS ACROSS WIDE REGIONS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1019388
Grant No.
2019-33610-30492
Cumulative Award Amt.
$100,000.00
Proposal No.
2019-00535
Multistate No.
(N/A)
Project Start Date
Sep 1, 2019
Project End Date
May 31, 2020
Grant Year
2019
Program Code
[8.4]- Air, Water and Soils
Project Director
Hagen, S.
Recipient Organization
Applied GeoSolutions, LLC
87 Packers Falls Road
Durham,NH 03824
Performing Department
(N/A)
Non Technical Summary
Agricultural row crops occupy over 240 million acres of land in the United States. Decisions regarding the implementation of management practices in these agricultural areas have a significant effect on other environmental outcomes, including soil erosion, water quality, and carbon sequestration. Fractional cover and biomass of the winter cover crop are the primary determinants of how effectively a cover crop protects the soil and immobilizes the nutrients, and while information regarding the presence or absence of cover crops is rare, information regarding the quality of cover crops is essentially non-existent. We propose to use remote sensing and modeling to extract detailed information on cover crops across large regions, and believe that this system will represent a cost-efficient solution.
Animal Health Component
50%
Research Effort Categories
Basic
10%
Applied
50%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1120199208070%
9030199303030%
Goals / Objectives
We will evaluate the feasibility of using fused multi-sensor observations for operational cover crop monitoring that will systematically provide detailed information about the dynamics of cover crop systems. Our initial focus will be on three agricultural test areas: the Maumee River Watershed within the western basin of Lake Erie, the Vermilion Headwaters and Mackinaw watersheds in Illinois, and the Choptank River Watershed in Chesapeake Bay region, with the ultimate goal of providing a continental scale system. We will assess the feasibility of mapping cover crop attributes (location, fractional cover, green-up date, kill date, and biomass) with high frequency time series information from a combination of modeling and multiple satellite sensors including Landsat 8, Landsat 7, Sentinel 2, and MODIS, and gain additional insights about the cover crops by integrating satellite remote sensing observations with a soil biogeochemical model.
Project Methods
We have constructed a project team that includes remote sensing experts, cover crop experts, and web programming experts. Through this project team, we have cover crop information, both ground and remote sensing-based, from diverse agricultural regions in the United States. Our technical objective will be met through the completion of three parallel tasks.Task 1. Demonstrate the technical feasibility of mapping characteristics of cover crops using multi-sensor remote sensing data, using our Geospatial Image Processing System (GIPS; open source software ) to acquire multi-temporal Landsat, Sentinel-2A and 2B, and MODIS data for the 2015-2017 locations. These data will be used to create a 5-day 30-m gap-filled NDVI time series product which will be a foundational data set for cover crop class, fractional cover, timing information, and biomass. The generation of cover crop class, fractional cover, and timing will be done with a decision tree algorithm. Biomass will be estimate via NDVI time series integration with the DNDC biogeochemical model. All estimates produced will be validated against ground observations.Task 2. Develop a protocol for systematic processing and validation of cover crop map products by building upon our open source GIPS system.Task 3. Understand user needs and design a prototype web-based data visualization and delivery system by working closely with an advisory team that will provide feedback on data requirements, including acceptable product accuracy and latency, as well as preferred data formats.

Progress 09/01/19 to 05/31/20

Outputs
Target Audience:Under this project, the team successfully reached two target audiences. (1) Researchers at USDA-ARS in Beltsville, MD. Due to this project, in large part, Dagan established a CRADA with Dr. Steven Mirsky's lab at ARS in Beltsville, MD. Under this agreement, Dagan and Mirsky's lab will share research results as we work to build a fully operational cover crop mapping system with comprehensive information. (2) Stakeholders seeking comprehensive information about cover crop performace over wide areas and back through time. Notably, we had substantive discussions about requirements with The Nature Conservancy, the Ecosystem Service Market Consortium, and OpenTEAM. p { margin-bottom: 0.1in; line-height: 115%; background: transparent } Changes/Problems:Our demonstration regions shifted slightly due to data availability. This change had no significant effect on the project outcomes. What opportunities for training and professional development has the project provided?While no graduate or post-doc research was funded through our Phase I project, we will need a full-time post-doc to extend the system demonstrated through our Phase I research. How have the results been disseminated to communities of interest?Our team conducted webinars with two stakeholders: Steven Mirsky's lab (ARS-Beltsville) - our discussions have resulted in a new CRADA; The Nature Conservancy - interest is growing in using an improved cover crop mapping system to set baselines and evaluate conservation program performance; Ecosystem Service Market Consortium - Improved maps of cover crop performance can help with project verification; What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We accomplished the goals outlined in our proposal. Specifically, we generated a time series of fused remote sensing observations using MODIS, Sentinel 2, and Landsat satellite observations in two study locations - Beltsville, MD and Tippecanoe County, IN to map crop calendar and estimate biomass through the winter cover crop growing season in a flexible framework that allows for wide-area deployment These remote sensing derived estimates were then used as input to the DNDC soil biogeochemical model to improve estimates from carbon and nitrogen cycling.

Publications