dc.contributor | Universidade Estadual Paulista (Unesp) | |
dc.date.accessioned | 2014-05-20T15:23:57Z | |
dc.date.accessioned | 2022-10-05T16:25:07Z | |
dc.date.available | 2014-05-20T15:23:57Z | |
dc.date.available | 2022-10-05T16:25:07Z | |
dc.date.created | 2014-05-20T15:23:57Z | |
dc.date.issued | 1997-10-22 | |
dc.identifier | Forest Ecology and Management. Amsterdam: Elsevier B.V., v. 98, n. 1, p. 35-47, 1997. | |
dc.identifier | 0378-1127 | |
dc.identifier | http://hdl.handle.net/11449/34625 | |
dc.identifier | 10.1016/S0378-1127(97)00072-8 | |
dc.identifier | WOS:A1997YE09400003 | |
dc.identifier.uri | http://repositorioslatinoamericanos.uchile.cl/handle/2250/3906721 | |
dc.description.abstract | The purpose of this work was to study fragmentation of forest formations (mesophytic forest, riparian woodland and savannah vegetation (cerrado)) in a 15,774-ha study area located in the Municipal District of Botucatu in Southeastern Brazil (São Paulo State). A land use and land cover map was made from a color composition of a Landsat-5 thematic mapper (TM) image. The edge effect caused by habitat fragmentation was assessed by overlaying, on a geographic information system (GIS), the land use and land cover data with the spectral ratio. The degree of habitat fragmentation was analyzed by deriving: 1. mean patch area and perimeter; 2. patch number and density; 3. perimeter-area ratio, fractal dimension (D), and shape diversity index (SI); and 4. distance between patches and dispersion index (R). In addition, the following relationships were modeled: 1. distribution of natural vegetation patch sizes; 2. perimeter-area relationship and the number and area of natural vegetation patches; 3. edge effect caused by habitat fragmentation, the values of R indicated that savannah patches (R = 0.86) were aggregated while patches of natural vegetation as a whole (R = 1.02) were randomly dispersed in the landscape. There was a high frequency of small patches in the landscape whereas large patches were rare. In the perimeter-area relationship, there was no sign of scale distinction in the patch shapes, In the patch number-landscape area relationship, D, though apparently scale-dependent, tends to be constant as area increases. This phenomenon was correlated with the tendency to reach a constant density as the working scale was increased, on the edge effect analysis, the edge-center distance was properly estimated by a model in which the edge-center distance was considered a function of the to;al patch area and the SI. (C) 1997 Elsevier B.V. B.V. | |
dc.language | eng | |
dc.publisher | Elsevier B.V. | |
dc.relation | Forest Ecology and Management | |
dc.relation | 3.169 | |
dc.relation | 1,625 | |
dc.rights | Acesso restrito | |
dc.source | Web of Science | |
dc.subject | landscape ecology | |
dc.subject | fragmentation of natural forest formations | |
dc.subject | edge effect | |
dc.subject | fractals | |
dc.subject | fractal dimension | |
dc.subject | fractal models | |
dc.subject | remote sensing | |
dc.subject | geographic information system | |
dc.subject | vegetation indices | |
dc.title | A study of habitat fragmentation in Southeastern Brazil using remote sensing and geographic information systems (GIS) | |
dc.type | Artigo | |