Using simulation models to evaluate ape nest survey techniques

• Published in: PloS one 2010-05, Vol.5 (5), p.e10754-e10754
• Main Authors: Boyko, Ryan H, Marshall, Andrew J
• Format: Article
• Language: English
• Subjects: Accuracy; Animal behavior; Animal populations; Animals; Approximation; Computational Biology; Computer Science/Applications; Computer Simulation; Conservation; Data Collection; Decay; Decay rate; Dissertations & theses; Ecology; Ecology/Population Ecology; Estimates; Hominidae - physiology; Mathematical analysis; Mathematical models; Matrix methods; Methods; Models, Biological; Monkeys & apes; Nest building; Nesting Behavior - physiology; Nests; Orangutans; Polls & surveys; Pongo pygmaeus abelii; Population decline; Population density; Population Dynamics; Population number; Reliability; Sampling; Simulation; Size estimation; Surveying; Surveys; Time Factors
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0010754

Conservationists frequently use nest count surveys to estimate great ape population densities, yet the accuracy and precision of the resulting estimates are difficult to assess. We used mathematical simulations to model nest building behavior in an orangutan population to compare the quality of the population size estimates produced by two of the commonly used nest count methods, the 'marked recount method' and the 'matrix method.' We found that when observers missed even small proportions of nests in the first survey, the marked recount method produced large overestimates of the population size. Regardless of observer reliability, the matrix method produced substantial overestimates of the population size when surveying effort was low. With high observer reliability, both methods required surveying approximately 0.26% of the study area (0.26 km(2) out of 100 km(2) in this simulation) to achieve an accurate estimate of population size; at or above this sampling effort both methods produced estimates within 33% of the true population size 50% of the time. Both methods showed diminishing returns at survey efforts above 0.26% of the study area. The use of published nest decay estimates derived from other sites resulted in widely varying population size estimates that spanned nearly an entire order of magnitude. The marked recount method proved much better at detecting population declines, detecting 5% declines nearly 80% of the time even in the first year of decline. These results highlight the fact that neither nest surveying method produces highly reliable population size estimates with any reasonable surveying effort, though either method could be used to obtain a gross population size estimate in an area. Conservation managers should determine if the quality of these estimates are worth the money and effort required to produce them, and should generally limit surveying effort to 0.26% of the study area, unless specific management goals require more intensive sampling. Using site- and time- specific nest decay rates (or the marked recount method) are essential for accurate population size estimation. Marked recount survey methods with sufficient sampling effort hold promise for detecting population declines.