Influence of population survey protocol on wood turtle (Glyptemys insculpta) detections
Open Access
- 6 November 2026
- journal article
- Published by CSIRO Publishing in Wildlife Research
- Vol. 53 (11)
Abstract
Context: The wood turtle (Glyptemys insculpta) is a species of conservation concern and currently being considered for listing under the US Endangered Species Act. Two regional survey protocols were developed to facilitate standardized monitoring across the wood turtle’s geographic distribution due to population declines throughout its range. It is currently unknown whether these survey designs are compatible for broad-scale analyses. Aims: We assessed two aspects of the protocols, constrained versus unconstrained surveys and aquatic-focused versus terrestrial-focused surveys, using four survey designs, i.e. aquatic constrained, aquatic unconstrained, terrestrial constrained, and terrestrial unconstrained, to determine whether they result in similar estimates for detection probability and expected count during surveys. Methods: Between 2021 and 2023, we completed comparisons at 21 sites in 9 states. We estimated adult detection probability and expected count for aquatic-focused and terrestrial-focused surveys by using N-mixture models and generalized linear mixed models. Key results: For aquatic surveys in medium-sized streams, estimated detection probability was 0.08 (95% CI: 0.05–0.12) and 0.05 (95% CI: 0.03–0.08), and expected count was 5.5 (95% CI: 3.6–8.4) and 3.4 (95% CI: 2.2–5.2) for constrained and unconstrained surveys respectively. For terrestrial surveys, estimated detection probability was 0.32 (95% CI: 0.23–0.43) and 0.24 (95% CI: 0.17–0.33), and expected count was 2.0 (95% CI: 1.2–3.5) and 1.6 (95% CI: 0.9–2.7) for constrained and unconstrained surveys respectively, at the mean air temperature (20.3°C). Conclusions: Our study results indicate that constrained and unconstrained surveys are both effective population survey techniques for wood turtles, but survey type should be accounted for in count or abundance models. In contrast, mean detection probability substantially differed between aquatic and terrestrial surveys. Thus, restricting all survey replicates within sampling periods at monitoring sites to only aquatic or terrestrial surveys is optimal for use in population models. Implications: Our results can be used by agencies to optimize within and across wood turtle monitoring programs, including personnel resources, to meet objectives including continuous monitoring and evaluation of wood turtle populations if the species is listed as threatened under the Endangered Species Act.Keywords
This publication has 29 references indexed in Scilit:
- Influences of Temperature and Time on Habitat Use Patterns of a Semi‐Aquatic TurtleEcology and Evolution, 2025
- Influences of aquatic and terrestrial habitat characteristics on abundance patterns of adult wood turtlesThe Journal of Wildlife Management, 2024
- Long-term monitoring in transition: Resolving spatial mismatch and integrating multistate occupancy dataEcological Indicators, 2022
- Large-scale collaboration reveals landscape-level effects of land-use on turtle demographyGlobal Ecology and Conservation, 2021
- Survey and analysis design for wood turtle population monitoringThe Journal of Wildlife Management, 2017
- Conservation needs for the endangered New Zealand sea lion, Phocarctos hookeriAquatic Conservation: Marine and Freshwater Ecosystems, 2017
- Intrinsic heterogeneity in detection probability and its effect on N‐mixture modelsMethods in Ecology and Evolution, 2016
- A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterionBehavioral Ecology and Sociobiology, 2010
- AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisonsBehavioral Ecology and Sociobiology, 2010
- Large scale wildlife monitoring studies: statistical methods for design and analysisEnvironmetrics, 2002