Figure captions

Figure 1. For spatial count (SC) and spatial partial identity models (SPIM), yellow cells show expectations of model performance across different values of individual-level aggregation and cohesion (i.e., non-independence). Red crosses and green check marks indicate patterns identified in simuations. Values of aggregation and cohesion considered in simulations are shown in parentheses.
Figure 2. Expected probabilities that two draws from a population with 4 partial identity covariates (coat color, sex, GPS collar, antler points, in various population proportions) would be distinct from each other, given an increasing number of those covariates used for SPIM estimation. Closed circles show all combinations, while open squares highlight those evaluated in simulations. Colored lines indicate the mean probability of identity for a given partial identity covariate for each specified number of total covariates used.
Figure 3. Example patterns of detections for N = 140 individuals (colored circles) for select combinations of aggregation and cohesion with detection probabilities of p0 = 0.05 (top) and p0 = 0.20 (bottom), illustrating how non-independence affects the distribution of individuals and their detections across the statespace (rectangle). Lines connect individuals to sampling locations (crosses) where detected. The sampling array has 3-unit spacing with a 9-unit buffer to the statespace edge.
Figure . Numbers of unique individuals detected (top), total detections (middle), and traps with detections (bottom) from simulating detection data for populations of N = 140 under 9 different non-independence scenarios with detection probability p0 = 0.05, generated by varying aggregation (group size) and cohesion.
Figure . Calculated overdispersion factor (\(\hat{c}\)) of simulated detection data with detection probability p0 = 0.05 when individuals exhibit non-independence through varying levels of aggregation and cohesion.
Figure . Point estimates of abundance, N (top row), and spatial scale of individual movement, σ (bottom row), with SCR, SC, and SPIM across a range of aggregation and cohesion values and combinations of partial identity covariates, with detection probability p0 = 0.05. Dotted lines indicate truth: N = 140, and σ = 3.
Figure 7. Coverage of estimates of abundance, N (top row), and spatial scale of individual movement, σ (bottom row), for SCR, SC, and SPIM, across a range of aggregation and cohesion values and combinations of partial identity covariates, with detection probability p0 = 0.05. Dashed line indicates 95% coverage.