Maximum entropy modeling of giant kangaroo rat (Dipodomys ingens) populations

Graduation Date

2015

Document Type

Thesis

Program

Other

Program

Thesis (M.S.)--Humboldt State University, Environmental Systems: Mathematical Modeling, 2015

Committee Chair Name

Abeer Hasan

Committee Chair Affiliation

HSU Faculty or Staff

Keywords

Statistical model, Humboldt State University -- Theses -- Mathematical Modeling, Maximum entropy estimation, Macroecology, Carrizo Plain, Population estimation, Quadrat counts, Intraspecific spatial abundance distribution, Coleman random placement model, Giant kangaroo rat, Spatial point pattern, ALC, Range-area relationship, Maximum likelihood estimation

Abstract

Maximum entropy estimation is a non-parametric method of determining probability distributions from limited information. In the field of ecology, maximum entropy estimation has seen use in the last 10 to 15 years as a method for developing macroecoogical models. Some macroecoogical metrics, such as the intraspecific spatial abundance distribution and the range-area relationship, and methods, such as setting out a grid and counting individuals, are much more easily applied to plants than animals. However, some animal species do lend themselves to this technique, making it possible to determine these metrics. The giant kangaroo rat (Dipodomys ingens) is one such species, with large burrows that are easily identifiable from aerial photography. This provides an opportunity to test out the applicability of the maximum entropy model for the spatial abundance distribution and the range-area relationship, and compare it to the earlier Coleman random placement model. Scale is found to have a significant effect in the applicability of the maximum entropy model, while the Coleman random placement model fits well across a greater range of scales. In addition, it is shown that the Coleman random placement model can be used to estimate the total number of burrows in an area from limited data, with varying accuracy depending on scale.

https://scholarworks.calstate.edu/concern/theses/t435gg44g

Share

 
COinS