Simulation Study Of The Effect Of Dispersal Habitat Area And Configuration On Landscape-genetic Relationships

Recently, the increasing habitat loss and fragmentation has become the key driver causing loss of biodiversity. Landscape genetics, an interdiscipline of landscape ecology and population genetics, is suited for testing the effect of structural landscape connectivity on functional landscape connectivity. These studies play an increasingly important role in the conservation of biodiversity and the management of reserve. The core of landscape genetics is to connect the spatail pattern of genetic variation with the highly random space-time process. Computer simulations can be used in this filed to integrate the randomness of individual with environment. It could explicitly control the pattern-process relationships, which ensures the rigorous distribution of the factors caused the genetic differentiation and the evaluation for these factors. It has provided effective tools in landscape genetics.Population connectivity is a key factor for the survival of local species. It is significant for the conservation of biodiversity to increase population connectivity. Dispersal habitat can effectively increase the migration of organisms from one habitat patch to another. Its area and configuration influence the genetic differentiation and diversity. In this paper, we used spatially explicit, individual-based simulation modeling to explore the relationships of landscape structure and genetic pattern with different areas and configurations of dispersal habitat. We quantified the effect of fragmentation on genetic diversity. We chose patch density (PD), correlation length (GYRATE_AM), the clumpy index of class aggregation(CLUMPY), patch cohesion (COHESION), and aggregation index (AI) to analyze landscape pattern with different fragmentation. We evaluated the effect of different dispersal habitat areas and configurations on the relationship of landscape-genetic by regression analysis. Finally, we built the most effective model for the prediction of landscape-genetic relationships.The results indicate that:(1) There are thresholds in fragmentation effects. Fragmentation will not have an remarkable effect on genetic diversity when dispersal habitat resistance R≦4 and dispersal habitat area P≧0.25. When P≦0.15 or R=8, the effect of landscape fragmentation on genetic diversity have a negative relationship with dispersal habitat area and a positive relationships with dispersal habitat resistance.Landscapes of moderate fragmentation have largest influence on genetic diversity.(2) When change the degree of fragmentation, the variation of landscape metrics are little under large dispersal habitat area; While the changes are larger under small dispersal habitat. When change the size of dispersal habitat area, the changes of landscape metrics are little in the slight fragmentized landscape; While the changes are much larger in severe fragmentized landscape.(3) Genetic differentiation is mainly influenced by dispersal habitat area and configuration when habitat areas are same. Dispersal habitat area is the strongest individual predictor of genetic differentiation. And correlation length of dispersal habitat is the second strongest individual predictor. The effect of habitat configuration on genetic differentiation is increase with the increase of dispersal habitat resistant. While the effect of dispersal habitat configuration on genetic differentiation is decrease with the increase of dispersal habitat resistance. Genetic differentiation is mainly influenced by dispersal habitat area and configuration when dispersal habitat area and correlation length are small. While dispersal habitat area and correlation length are large, it mainly influenced by dispersal habitat resistance.(4) The most effective model of predicting partial mantel r includes resistance, patch density, correlation length of dispersal habitat, and correlation length of habitat.