Fig 1.
The locations of the 16 study cities (a) and, as an example, the habitat loss during urbanization in Paris from 1800 to 2000 (b).
Table 1.
List of landscape metrics used in the study, all of which, except normalized total core area and normalized nearest neighbor distance, were based on McGarigal et al. [32] and Wu et al. [33].
Fig 2.
Changes in built-up area and habitat in 16 study cities from 1800 to 2000 at two extents: (a) the urban region and (b) the central city area. The study cities are ordered by urban population in 2000.
Fig 3.
Different forms of the relationship between habitat loss and habitat fragmentation during urbanization, derived from historical landscape pattern analysis (See Figs A–J in S4 Appendix for details).
*Landscape metrics include: (1) area metrics, i.e., mean patch size (MPS), total core area (TCA), and normalized TCA (NTCA); (2) density metrics, i.e., patch density (PD) and edge density (ED); (3) shape metrics, i.e., landscape shape index (LSI) and perimeter-area fractal dimension (PAFD); and (4) connectivity metrics, i.e., mean Euclidean nearest neighbor distance (NND), normalized NND (NNND), and Cohesion (See Table 1 for details).
Table 2.
Different forms of the relationship between habitat loss and habitat fragmentation during urbanization in Paris, as an example, derived from historical landscape pattern analysis at the urban regional extent (Number of samples = 10) (See Figs A–J in S4 Appendix for details).
Fig 4.
Different forms of the relationship between habitat loss and habitat fragmentation during urbanization, derived from space-for-time analysis (See Figs K–AD in S4 Appendix for details).
*Landscape metrics include: (1) area metrics, i.e., mean patch size (MPS), total core area (TCA), and normalized TCA (NTCA); (2) density metrics, i.e., patch density (PD) and edge density (ED); (3) shape metrics, i.e., landscape shape index (LSI) and perimeter-area fractal dimension (PAFD); and (4) connectivity metrics, i.e., mean Euclidean nearest neighbor distance (NND), normalized NND (NNND), and Cohesion (See Table 1 for details).
Table 3.
Different forms of the relationship between habitat loss and habitat fragmentation during urbanization in Paris, as an example, derived from space-for-time analysis at the extent of 64 by 64 pixels (Number of samples = 818) (See Figs K–T in S4 Appendix for details).
Fig 5.
Different forms of the relationship between habitat loss and habitat fragmentation, representing different hypotheses reported in the literature (See S1 Appendix for the detailed sources).
*Landscape metrics include: (1) area metrics, i.e., mean patch size (MPS), total core area (TCA), and normalized TCA (NTCA); (2) density metrics, i.e., patch density (PD) and edge density (ED); (3) shape metrics, i.e., landscape shape index (LSI) and perimeter-area fractal dimension (PAFD); and (4) connectivity metrics, i.e., mean Euclidean nearest neighbor distance (NND), normalized NND (NNND), and Cohesion (See Table 1 for details).
Fig 6.
Spatial patterns of habitat and corresponding values of landscape metrics with decreasing percentages of habitat in Paris, as an example, derived from both historical landscape pattern analysis and space-for-time analysis.
*The habitats in the historical analysis were derived from data at the central city area extent in Paris in 1800, 1880, 1928, 1955, and 1987. The habitats in the space-for-time analysis were derived from data in Paris in 2000 with the extent of 64 by 64 pixels.