Evaluating multispectral remote sensing and spectral unmixing analysis for crop residue mapping Anna Pacheco ⁎, Heather McNairn Agriculture and Agri-Food Canada, 960, Carling Avenue, Ottawa, Ontario K1A 0C6 abstract article info Article history: Received 11 December 2009 Received in revised form 21 April 2010 Accepted 25 April 2010 Keywords: Tillage Crop residue Non-photosynthetic vegetation SPOT Landsat TM Spectral unmixing analysis Endmembers Tillage practices can affect the long term sustainability of agricultural soils as well as a variety of soil processes that impact the environment. Crop residue retention is considered a soil conservation practice given that it reduces soil losses from water and wind erosion and promotes sequestration of carbon in the soil. Spectral unmixing estimates the fractional abundances of surface targets at a sub-pixel level and this technique could be helpful in mapping and monitoring residue cover. This study evaluated the accuracy with which spectral unmixing estimated percent crop residue cover using multispectral Landsat and SPOT data. Spectral unmixing produced crop residue estimates with root mean square errors of 17.29% and 20.74%, where errors varied based on residue type. The model performed best when estimating corn and small grain residue. Errors were higher on soybean fields, due to the lower spectral contrast between soil and soybean residue. Endmember extraction is a critical step to successful unmixing. Small gains in accuracy were achieved when using the purest crop residue- and soil-specific endmembers as inputs to the spectral unmixing model. To assist with operational implementation of crop residue monitoring, a simple endmember extraction technique is described. © 2010 Published by Elsevier Inc. 1. Introduction Tillage practices affect the long term sustainability of agricultural soils as well as a variety of soil processes that impact the environment. Tillage serves a number of purposes including preparing the soil for seeding, mitigating soil compaction, controlling weeds and incorpo- rating fertilizers into the soil. However, tilling also disintegrates soil aggregates and reduces crop residue cover on the soil. Retention of surface crop residues is an important conservation practice as these residues protect the soil from wind and water erosion. Residues incorporated into the soil by tilling decompose more quickly. Consequently all other factors being equal, soils under no-till or conservation tillage management have higher levels of organic matter and sequester more carbon. The environmental benefits along with the savings in labour and fuel costs associated with reducing or eliminating tillage have resulted in increasing implementation of these practices. Yet growing interest in the use of crop biomass and crop residues for biofuel production may counter the benefits gained in the adoption of conservation practices (Lal & Pimentel, 2007). Consequently monitoring changes in residue management in re- sponse to policy and market influences is important. Information on tillage activities and residue cover assists in implementing policies and programs to promote beneficial manage- ment practices (BMPs), and in monitoring the success of these initiatives. Crop residue estimates are also a critical parameter in estimating soil carbon and in modeling and monitoring improvements in carbon sequestration that follow from adjustments in land manage- ment approaches. The National Agri-environmental Health Analysis and Reporting Program (NAHARP) of Agriculture and Agri-Food Canada, Canada's agri-environmental indicator initiative, require tillage infor- mation as input to 15 of the existing 29 indicators. Since 1896, much of this information has been gathered through census surveys implemen- ted every five years. These data are collected at the farm scale but survey results are reported in aggregate. Spatial allocation of the census data to the landscape, interpretation of census survey questions, and infrequent surveying (once every 5 years) can make it difficult to capture the spatial and temporal variability in tillage management practices (Lobb et al., 2007). Other field methods such as roadside visual surveys or line- point transects (Morrison et al., 1993) are often unable to characterize the variability of crop residue cover across an agricultural field. These methods are also tedious, time consuming and prone to human judgment errors. Rapid, accurate and objective methods to measure percent crop residue cover are thus required to meet the needs of policy, programs, land management decision-makers and carbon modelers. With access to an increasing numbers of satellites, Earth observation can play an important role in providing residue and tillage information at spatial and temporal resolutions that support monitoring and modeling at regional and watershed scales. Several remote sensing methods have been developed to quantify percent crop residue cover, including a number of approaches that rely on Remote Sensing of Environment 114 (2010) 2219–2228 ⁎ Corresponding author. E-mail address: anna.pacheco@agr.gc.ca (A. Pacheco). 0034-4257/$ – see front matter © 2010 Published by Elsevier Inc. doi:10.1016/j.rse.2010.04.024 Contents lists available at ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse