Occupancy versus colonization-extinction models for projecting population trends at different spatial scales.

Area of Science:

• Ecology
• Population Dynamics
• Conservation Biology

Background:

• Accurate prediction of population dynamics is crucial for ecological research and conservation efforts.
• Future population trends are sensitive to modeling assumptions, particularly concerning landscape and climate changes.
• Deadwood-dependent fungi serve as a model system for studying sessile species due to their varied dynamics and substrate requirements.

Purpose of the Study:

• To evaluate the impact of different modeling assumptions on projected population trends of sessile species.
• To compare the predictive performance of occupancy models versus colonization-extinction models.
• To identify optimal modeling strategies for predicting long-term population dynamics under environmental change.

Main Methods:

• Employed hierarchical Bayesian models fitted to extensive, repeated survey data of deadwood and fungi across 174 patches.
• Compared occupancy models with colonization-extinction models using varying spatial scales and deadwood data resolutions.
• Investigated the influence of key explanatory variables on species occurrence and colonization-extinction dynamics.

Main Results:

• Occupancy models projected higher occurrence probabilities and more positive trends than colonization-extinction models, especially for species with lower occupancy and colonization rates.
• The magnitude of projected occupancy increase was significantly influenced by the spatial modeling scale and resource resolution.
• Differences in projections between model types were more pronounced for species with lower occupancy, colonization rates, and colonization:extinction ratios.

Conclusions:

• Colonization-extinction models are recommended over occupancy models for predicting population trends of sessile species.
• Modeling should be conducted at the finest utilizable resource-unit resolution and the same spatial scale for fitting and projection.
• Incorporating key drivers of metapopulation dynamics, such as resource availability, habitat quality, and spatial connectivity, is essential for robust predictions.