Improved Change Detection with Trajectory-Based Approach: Application to Quantify Cropland Expansion in South Dakota

• Main Authors: Lan H. Nguyen, Deepak R. Joshi, Geoffrey M. Henebry
• Format: Article
• Language: English
• Published: MDPI AG 2019-04-01
• Series: Land
• Subjects: land cover/land use; agriculture; Cropland Data Layer; land surface phenology; South Dakota; trajectory-based change detection; multi-date
• Online Access: https://www.mdpi.com/2073-445X/8/4/57

The growing demand for biofuel production increased agricultural activities in South Dakota, leading to the conversion of grassland to cropland. Although a few land change studies have been conducted in this area, they lacked spatial details and were based on the traditional bi-temporal change detection that may return incorrect rates of conversion. This study aimed to provide a more complete view of land conversion in South Dakota using a trajectory-based analysis that considers the entire satellite-based land cover/land use time series to improve change detection. We estimated cropland expansion of 5447 km² (equivalent to 14% of the existing cropland area) between 2007 and 2015, which matches much more closely the reports from the National Agriculture Statistics Service—NASS (5921 km²)—and the National Resources Inventory—NRI (5034 km²)—than an estimation from the bi-temporal approach (8018 km²). Cropland gains were mostly concentrated in 10 counties in northern and central South Dakota. Urbanizing Lincoln County, part of the Sioux Falls metropolitan area, is the only county with a net loss in cropland area over the study period. An evaluation of land suitability for crops using the Soil Survey Geographic Database (SSURGO) indicated a scarcity in high-quality arable land available for cropland expansion.