Using Landsat Archive for Retrospective Estimates of Conservation Tillage Practices.

The Environmental Quality Incentives Program (EQIP) provides financial assistance to encourage producers to adopt conservation practices. Upon contract completion, it is unclear how many producers continue those practices without payment, or to what extent these producers adopt similar practices on fields that were never enrolled in a contract. Likewise, it is uncertain whether neighboring producers adopt the practices in response to a nearby producer under EQIP contract in a peer effect. One practice of interest is the use of conservation tillage as observed by increased crop residue cover left on the soil surface. Completion dates for spring planting vary, possibly by as much as three months, due to a wide range of reasons- different crops require different planting dates (e.g. corn precedes soybean by weeks), and readiness of soils (frozen or too wet). To best capture the landscape mosaic seen in crop residue cover in the landscape, a compilation of images is required. The Landsat platforms can provide this temporal and spatial coverage, with 16-day global coverage beginning in the mid-1980s. Here we used the Normalized Difference Tillage Index (NDTI) which has proven robust and accurate. It utilizes multispectral satellites available in higher temporal and spatial coverage. We completed 10 years of residue maps for the 150,000 square kilometer study area in SD, ND, and MN and compared the results to the ARMS data provided by USDA ERS. The overall accuracy was between 70-80% with additional improvement when spurious points were removed. Now that the Landsat Archive exists at no cost, we can do retrospective studies to help answer how conservation practices persist after contract completion or spread through peer effects.