TY - JOUR

T1 - Influence of the resolution of forest cover maps in evaluating fragmentation and connectivity to assess habitat conservation status

AU - Hernando, A.

AU - Velázquez, J.

AU - Valbuena, R.

AU - Legrand, M.

AU - García-Abril, A.

PY - 2017/8

Y1 - 2017/8

N2 - The analysis of fragmentation and habitat connectivity is important in determining their conservation status and ensuring their long-term survival. However, the reliability of assessments on habitat conservation status may depend on the resolution of forest cover maps used as inputs. The aim of this paper is to quantify differences in the results of habitat fragmentation and connectivity analysis found when using three different forest cover maps of various resolutions, and discusses their effect in the assessment of habitat conservation status. The study was conducted in a Natura 2000 habitat (9120:Atlantic acidophilous beech forests) in Spain. To measure fragmentation, we carried out a morphological spatial pattern analysis (MSPA) which provided a very detailed spatial landscape description (core, islet, bridge, loop, branch and perforation elements). We compared the habitat total area (Hta) with the habitat area without edge width (Hwe), which correspond to the obtained previous cores. To measure connectivity, we used the probability of connectivity index (PC). We used three different forest cover maps with different spatial resolutions: (1) a 2 m map derived from remote sensing using very high resolution satellite imagery (GeoEye) processed with object-based image analysis (OBIA_layer); (2) a 10 m map derived from fieldwork and aerial photo-interpretation at 1:10,000 scale (Forest_layer); and (3) a 50 m map obtained by a similar method at 1:50,000 scale (Atlas_layer). Our results confirm results obtained by previous studies showing that the resolution of input forest cover maps substantially influences MSPA results. The habitat area proportion classified as core decreased as the resolution of input forest cover maps increased, whereas the amount of islets, bridges, loops, branches and perforations increased. The spatial resolution of forest cover maps influences the assessment of habitat conservation status. Habitat conservation status was assessed as being ‘unfavourable inadequate’ (the middle rank out of three) with the coarse Atlas_layer, and as ‘unfavourable bad’ (the bottom rank out) when using the higher resolution Forest_layer and OBIA_layer. This can be critical for European environmental funding. Results obtained also show that resolution of input forest cover maps influence the calculated values of PC index. The use of high-resolution forest cover maps is critical to study habitat connectivity, since otherwise the outcome presents no appreciable result. We conclude that using remote sensing techniques together with OBIA is the most appropriate and cost-effective method for analyzing forest fragmentation and connectivity for habitat conservation status assessments. The source and/or the method of generation of the habitat data layer used (also the spatial resolution) as well as the connectivity analysis method applied must at all times be reported in such analyses.

AB - The analysis of fragmentation and habitat connectivity is important in determining their conservation status and ensuring their long-term survival. However, the reliability of assessments on habitat conservation status may depend on the resolution of forest cover maps used as inputs. The aim of this paper is to quantify differences in the results of habitat fragmentation and connectivity analysis found when using three different forest cover maps of various resolutions, and discusses their effect in the assessment of habitat conservation status. The study was conducted in a Natura 2000 habitat (9120:Atlantic acidophilous beech forests) in Spain. To measure fragmentation, we carried out a morphological spatial pattern analysis (MSPA) which provided a very detailed spatial landscape description (core, islet, bridge, loop, branch and perforation elements). We compared the habitat total area (Hta) with the habitat area without edge width (Hwe), which correspond to the obtained previous cores. To measure connectivity, we used the probability of connectivity index (PC). We used three different forest cover maps with different spatial resolutions: (1) a 2 m map derived from remote sensing using very high resolution satellite imagery (GeoEye) processed with object-based image analysis (OBIA_layer); (2) a 10 m map derived from fieldwork and aerial photo-interpretation at 1:10,000 scale (Forest_layer); and (3) a 50 m map obtained by a similar method at 1:50,000 scale (Atlas_layer). Our results confirm results obtained by previous studies showing that the resolution of input forest cover maps substantially influences MSPA results. The habitat area proportion classified as core decreased as the resolution of input forest cover maps increased, whereas the amount of islets, bridges, loops, branches and perforations increased. The spatial resolution of forest cover maps influences the assessment of habitat conservation status. Habitat conservation status was assessed as being ‘unfavourable inadequate’ (the middle rank out of three) with the coarse Atlas_layer, and as ‘unfavourable bad’ (the bottom rank out) when using the higher resolution Forest_layer and OBIA_layer. This can be critical for European environmental funding. Results obtained also show that resolution of input forest cover maps influence the calculated values of PC index. The use of high-resolution forest cover maps is critical to study habitat connectivity, since otherwise the outcome presents no appreciable result. We conclude that using remote sensing techniques together with OBIA is the most appropriate and cost-effective method for analyzing forest fragmentation and connectivity for habitat conservation status assessments. The source and/or the method of generation of the habitat data layer used (also the spatial resolution) as well as the connectivity analysis method applied must at all times be reported in such analyses.

KW - Fragmentation

KW - Connectivity

KW - Habitat conservation status

KW - Spatial resolution

KW - Forest cover maps

KW - OBIA

KW - GIS

KW - Remote sensing

KW - Natura 2000

U2 - 10.1016/j.ecolind.2017.04.031

DO - 10.1016/j.ecolind.2017.04.031

M3 - Erthygl

VL - 79

SP - 295

EP - 302

JO - Ecological Indicators

JF - Ecological Indicators

SN - 1470-160X

ER -