Quantifying the Sensitivity of Land Use Land Cover Metrics Through Simulation Techniques

As human activities and climate change continue to reshape our landscape, understanding how land use changes over time is becoming increasingly important. Accurate ways to track and analyze these changes are essential for governments, businesses, and communities to make informed decisions. Monitoring agricultural land is particularly critical, as shifts in land use can impact food production and environmental pollutants. One of the primary tools used in the United States to monitor agricultural land is the Cropland Data Layer (CDL), an annual map created by the United States Department of Agriculture (USDA) from satellite images. While the CDL is highly accurate, the data product is still prone to misclassify certain land use types. These challenges can affect the reliability of any metrics calculated using the data. To address this, we have developed a new software package in the R programming language called cdlsim, which simulates land use changes to explore the sensitivity of land use metrics to variability in the input data. We demonstrate the utility of our software through two real-world applications: one in a low-agricultural-density region in Utah and another in a mixed agriculture and grassland landscape in South Dakota. This research not only enhances the credibility of CDL-based studies but also helps improve our understanding of land use trends, supporting better decisions for land management and environmental protection.