Utilizing historical maps in identification of long-term land use and land cover changes

Ambio. 2023 Nov;52(11):1777-1792. doi: 10.1007/s13280-023-01838-z. Epub 2023 Feb 25.

Abstract

Knowledge in the magnitude and historical trends in land use and land cover (LULC) is needed to understand the changing status of the key elements of the landscape and to better target management efforts. However, this information is not easily available before the start of satellite campaign missions. Scanned historical maps are a valuable but underused source of LULC information. As a case study, we used U-Net to automatically extract fields, mires, roads, watercourses, and water bodies from scanned historical maps, dated 1965, 1984 and 1985 for our 900 km[Formula: see text] study area in Southern Finland. We then used these data, along with the topographic databases from 2005 and 2022, to quantify the LULC changes for the past 57 years. For example, the total area of fields decreased by around 27 km[Formula: see text], and the total length of watercourses increased by around 2250 km in our study area.

Keywords: Historical maps; Land use and land cover change; Semantic segmentation; U-net.

MeSH terms

  • Conservation of Natural Resources*
  • Databases, Factual
  • Environmental Monitoring*
  • Finland