High Nature Value farmland identification from satellite imagery, a comparison of two methodological approaches

G.W. Hazeu, P. Milenov, G.B.M. Pedroli, V. Samoungi, M. van Eupen, V. Vassilev

Research output: Contribution to journalArticleAcademicpeer-review

13 Citations (Scopus)

Abstract

While the identification of High Nature Value (HNV) farmland is possible using the different types of spatial information categories available at European scale, most data used is still too coarse and therefore only provides an approximate estimate of the presence of HNV farmland. This paper describes two promising methods using remote sensing – one for HNV farmland identification and one for change detection within HNV farmland. The performance of the two methods is demonstrated by detailed results for two case studies – the Netherlands for the HNV farmland identification, and Bulgaria for change detection within HNV farmland. An estimation of the presence of HNV farmland or of HNV farmland change can well be based on high-resolution satellite imagery, but the classification method must be adapted to regional characteristics such as field size and type of landscape. The temporal variability and bio-climatological characteristics across Europe do not allow for a simple European classification of HNV farmland. Also comparison between years is complicated because of the large impact of seasonal variation in the land cover expression and the complexity of the HNV farmland definitions. Although HNV farmland detection methods are promising, remote sensing alone does not yet provide the appropriate tools for adequate monitoring.
Original languageEnglish
Pages (from-to)98-112
JournalInternational Journal of applied Earth Observation and Geoinformation
Volume30
DOIs
Publication statusPublished - 2014

Keywords

  • land

Fingerprint Dive into the research topics of 'High Nature Value farmland identification from satellite imagery, a comparison of two methodological approaches'. Together they form a unique fingerprint.

Cite this