Assessing the Accuracy of Large-Area Land Cover Maps: Experiences from the Multi-Resolution Land-Cover Characteristics (MRLC) Project

摘要

Large-area land-cover maps create unique challenges for thematic map accuracy assessment. We describe desirable characteristics for the design of large-area assessments, and propose two-stage cluster sampling as a general framework possessing the flexibility needed to achieve these characteristics. The sampling theory supporting use of the two-stage design is briefly described. We identify a need for further research to evaluate specific design options within this general sampling framework, and we describe some practical and philosophical problems that must still be resolved to enhance the effectiveness of accuracy assessments.