Supplemental structured surveys and pre-existing detection models improve fine-scale density and population estimation with opportunistic community science data

Abstract

Density and population estimates aid in conservation and stakeholder communication. While free and broadly available community science data can effectively inform species distribution models, they often lack the information necessary to estimate imperfect detection and area sampled, thus limiting their use in fine-scale density modeling. We used structured distance-sampling surveys to model detection probability and calculate survey-specific detection offsets in community science models. We estimated density and population for 16 songbird species under three frameworks: (1) a fixed framework that assumes perfect detection within a specified survey radius, (2) an independent framework that calculates offsets from an independent source, and (3) a calibration framework that calculates offsets from supplemental surveys. Within the calibration framework, we examined the effects of calibration dataset size and data pooling. Estimates of density and population size were consistently biased low in the fixed framework. The independent and calibration frameworks produced reliable estimates for some species, but biased estimates for others, indicating discrepancies in detection probability between structured and community science surveys. The calibration framework produced reliable population estimates with as few as 10 calibration surveys with positive detections. Data pooling dramatically decreased bias. This study provides conservationists and managers with a cost-effective method of estimating density and population.

Figure 1

Stringently filtered survey locations in the community science (green) and structured (orange) datasets for American Robin. After stringent filtering and geographic sampling, 1060 community science surveys remained for this species. The structured survey dataset was sampled without replacement to match survey number.

Figure 2

Workflow for analyses including (a) the frameworks and datasets, (b) the zero-inflated density modeling method, and (c) the calculation of detection probability offsets used within density models. The fixed framework incorporates no offsets and assumes a constant area surveyed of 200 m and perfect detection.

Figure 3

Results from the zero-inflation portion of density models for each framework, including AUC (A, B, and C) and estimated area of suitable habitat (D, E, and F), compared against a best-practices reference (benchmark). To allow for summarization across species, for each species, the results of the ten iterations within a framework were adjusted to the percentage of the median species-specific reference value. Results are divided into species that are common (A and D; 8 species), uncommon (B and E; 4 species), and rare (C and F; 4 species) within our study area as rarer species had insufficient data for the use of larger calibration datasets (Table 1).

Figure 4

Results from the density portion of density models for each framework, including mean density (A, B, and C) and estimated population (D, E, and F), compared against a best-practices reference (benchmark). To allow for summarization across species, for each species, the results of the ten iterations within a framework were adjusted to the percentage of the median species-specific reference value. Results are divided into species that are common (A and D; 8 species), uncommon (B and E; 4 species), and rare (C and F; 4 species) within our study area as rarer species had insufficient data for the use of larger calibration datasets (Table 1).