Available at: https://digitalcommons.calpoly.edu/theses/3122
Date of Award
6-2025
Degree Name
MS in Biological Sciences
Department/Program
Biological Sciences
College
College of Science and Mathematics
Advisor
William T. Bean
Advisor Department
Biological Sciences
Advisor College
College of Science and Mathematics
Abstract
From Conflict to Clarity: Resolving Data Differences in Habitat Assessments to Inform Conservation Priorities for the Giant Kangaroo Rat
Caroline MacKenzie
Conservation strategies frequently rely on habitat suitability models, habitat assessments and species occurrence data to identify areas for protection. However, these methods often produce conflicting inference on species presence and habitat quality, complicating management decisions. This study builds on prior work which used three distinct methods to guide conservation planning for the endangered Giant Kangaroo Rat (Dipodomys ingens): habitat suitability modeling based on abiotic predictors, aerial imagery to assess habitat quality through visible burrow mounds and live-trapping data to confirm species presence. While each method captures a different aspect of habitat use, disagreements stemmed from temporal mismatches, inconsistent spatial resolution and outdated environmental inputs. Here, I use updated versions of each dataset to reassess areas of disagreement and investigate the root causes of conflict among methods. By correcting these issues, this reassessment provides clearer evidence of site persistence and suitability. I found that much of the previous misalignment between methods was driven by outdated inputs, differences in spatial scale and temporal mismatches in data availability. These findings highlight the importance of using multi-method comparisons not just to detect presence, but to understand persistence, site quality and underlying causes of misalignment.
Habitat suitability models updated with higher-resolution environmental layers from the POLARIS dataset (~30 m), as opposed to coarser-scale data from SSURGO, yielded more stable indicators of long-term habitat value than single-year trapping events. Aerial imagery was particularly effective in resolving mismatches when used with repeated surveys, confirming its value as a tool for detecting long-term site use. In cases where methods disagreed, suitability modeling was more often consistent with outcomes observed during reassessment. These findings highlight the importance of multi-method integration and underscore the need for continued investment in fine-scale—site-level, species-specific—data collection, particularly in areas of misalignment between methods, where resolving discrepancies often requires more detailed but costly survey efforts than coarse-scale regional datasets can provide. Conservation efforts grounded in current, locally validated datasets are more likely to reflect true ecological conditions and better support species recovery in rapidly changing landscapes.