|
|
Gustafson, Eric, J., and Robert H. Gardner, 1996, Dispersal and Mortality in a Heterogeneous Landscape Matrix, International Journal of Geographical Information Sciences, Vol 10, pp. 254-268.
This article extends the traditional simplifying assumptions how metapopulations disperse in an environment where landcovers are disrupted. Metapopulations are a set of local populations within some larger area, where typically, migration from one local population to at least some habitat patch is possible. A patch is a continuous area of space with all necessary resources for the persistance of a local population and separated by unsuitable habitat from other patches. The main traditional assumption that this article extends is that the exchange rate among local populations is so low that migration has no real effect on local dynamics.
Studies have looked at movement of populations to patches but not the spatial arrangement of patches or the mortality rates of land cover between patches. In this study the authors modelled self-avoiding random walkers (i.e. individuals could not return to a previously visited patch until after 2 steps) and mortality rates to determine the probability of migration from one patch to another.
Four motality scenarios were modelled;
The 11 classes of land use/land cover data were derived from a Landsat TM image of Indiana. Three different landscapes were selected. One dominated by a surface mine, one dominated by a river and one dominated by agriculture. A landscape dispersal coefficient (LDC) was calculated based on simulated mortality rates and a maximum number of steps from land cover to land cover.
The results showed that constant mortality rates, even low ones, resulted in low LDC's. The habitat-dependant case showed highed LDC's because individuals could choose land covers with lower mortality risk. The time-dependant case was intermediate, as might be expected, because the initial mortality rate used was lower than average and the rate increased the longer the individual was migrating.
ANOVA showed that migration success was mainly due to differences in mortality functions (50%) and distance to nearest neighbor (25%).
