RT Journal Article
T1 GIS-based detection and quantification of patch-boundary patterns for identifying landscape mosaics
A1 Hardt, Elisa
A1 López de Pablo, Carlos Tomás
A1 Martín de Agar, Pilar
A1 Dos Santos, R. F.
A1 Pereira-Silva, E. F.
AB The study of boundaries between patches allows us to understand the complexity of landscape interactions, especially those involved in the anthropic use of natural resources, which is a common source of environmental problems when harnessing landscape services. The study of the relationships between those two elements makes it possible to identify distinct homogeneous environmental areas in which the same ecological interactions occur. These areas are the mosaics that make up a landscape. This paper presents a GIS-based procedure to identify and quantify the boundaries of land use/cover patches and to record those data in matrices of patches by boundaries. These matrices, by means of a multivariate analysis, allow us to recognize landscape mosaics. This semi-automated procedure contributes to making the concept of landscape mosaics operative and enabling its application to landscape management. To exemplify its possibilities, we tested three alternatives for quantifying boundary measures: presence/absence, frequency and length. They each describe interactions with different details and provide different nuances in interpretations of landscape organization. In the study case, the frequency data provided a more easily understandable interpretation of the mosaic identification and characterization of landscape heterogeneity because these data are less conditioned by the spatial distribution, size or length of rare boundaries. Irrespective of the boundary measure used, a large central mosaic is always identified, highlighting the influence of landscape homogeneity and fragmentation on mosaic identification and the robustness of the tested procedure.
PB ALÖKI Applied Ecological Research and Forensic Institute
SN 1589 1623, ISSN (Online): 1785 0037
YR 2018
FD 2018
LK https://hdl.handle.net/20.500.14352/18523
UL https://hdl.handle.net/20.500.14352/18523
LA eng
NO Ministerio de Ciencia e Innovación (MICINN)
NO Universidad Complutense de Madrid-Banco de Santander
NO FAPESP-BR
NO CAPES-BR y Fundación Carolina-ES
DS Docta Complutense
RD 17 abr 2025