Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218
Performing Department
WILDLIFE ECOLOGY
Non Technical Summary
Habitat fragmentation is a primary threat to biological diversity, and mammalian carnivores are particularly vulnerable to fragmentation effects. This project evaluates the degree of habitat fragmentation and landscape connectivity through radio-telemetry and simulation modeling of carnivore movement.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Goals / Objectives
Habitat fragmentation has been targeted as one of the most serious threats to biological diversity worldwide, and in areas with increasing urbanization, fragmentation is virtually inevitable. Intensive development in southern California over the past century has destroyed most of the native coastal sage scrub and chaparral habitats. This massive habitat loss, in conjunction with the high levels of local endemism of native species, has helped create a "hot-spot" of extinction in the region. In particular, mammalian carnivores are particularly vulnerable to extinction in fragmented habitat and might not be able to persist in fragmented landscapes, especially those that are exceedingly small or isolated. This research program will evaluate landscape-level connectivity in the highly fragmented landscape of coastal southern California. We will focus on large mammalian carnivores because top predators are ecologically pivotal organisms whose status is indicative of the
connectivity of ecosystems. The research project will consist of two components: 1.) Field surveys of carnivore movement and dispersal, including GPS radio-telemetry, track, and remotely-triggered camera surveys, and 2.) Development of spatially-explicit, individual-based computer simulation models of carnivore movement through the fragmented landscape. The simulation movement models will be developed and validated using our field surveys along with movement data from previously published studies of carnivores. Our movement models will allow us to identify, design, and protect wildlife movement corridors and landscape connectivity not just in southern California but other fragmenting landscapes.
Project Methods
Field Surveys: We will use a suite of field survey methodologies, including track surveys, remotely-triggered camera surveys, and GPS radio-telemetry, to assess the distribution, abundance, movement, and persistence of carnivores in the fragmented landscape of southern California. Track and camera surveys will follow methodology we have successfully developed and implemented in regional carnivore surveys throughout coastal southern California. In addition, GPS radio-collars will be placed on mountain lions and bobcats to document their movement patterns and responses to landscape elements; we plan to gather data from up to 50 GPS radio-collared animals, each radio-tracked for up to 2 years. Study sites will include urbanizing landscapes near Los Angeles. Simulation Modeling: We will use our field data of carnivore movement to inform spatially-explicit, individual-based computer simulation models of carnivore movement through the fragmented landscape of southern
California. We will use these models to assess the degree of landscape connectivity, to generate predictions of the effects of future landscape alterations on wildlife populations, and to help guide monitoring and management plans in the region. Our movement models will be composed of four parts: a spatial component, an individual component, an observation component, and a function minimization component. The spatial component will consist of input files that contain data on landscape features and functions that act as an interface between the spatial data and other components of the models. The southern California landscape will be visualized using GIS models of features that likely impact carnivore movement, including vegetative characteristics, land use types, road density, development pressures, and habitat suitability models for mountain lions and bobcats. In the individual component, animals have an explicit location in space at each time step in the simulation. Where the animal
is located from one time step to the next depends upon the "rules of movement" that individuals follow, and how they respond to variation in the landscape. Movement rules for our simulation models will be initially formulated using movement data from prior carnivore studies in southern California and other regions. Based on these field data, we will formulate alternative models for animal movement in response to landscape elements. These models will then be validated using our high-resolution movement data collected through GPS radio-telemetry of carnivores in the field (see above). The observation component will simulate the sampling of data from the movement model output to mimic field data collection. The function minimization component will be used to determine parameters for each model that best describe the movement data. We will then select the "best" movement model using a maximum likelihood approach.