Population reconstruction is a technique that uses harvest-at-age data and backward addition of cohorts to estimate minimum population size over time. Management agencies use population reconstruction because it uses data that are commonly collected for managed species, particularly for bear and deer populations. However, this technique had not been rigorously evaluated for accuracy or precision. We used computer simulations to evaluate the impact of life history parameters, harvest rate, sampling error, and violated assumptions on Downing population reconstruction estimates. This technique was robust to collapsing age classes if harvest rates for the oldest two age classes in the reconstruction were similar. Harvest and natural mortality rates were the driving factors in the accuracy of population reconstruction estimates. The technique was most accurate when harvest rate was high and natural mortality was low. Use of population reconstruction to detect population trends was confounded by changes in harvest and natural mortality rates. Gradual changes in these rates over time could potentially mask underlying population trends. Large annual variations in harvest rate also could make trend detection difficult. Given these findings, population reconstruction appears to be a promising management tool for species with high, relatively constant harvest mortality, such as many hunted bear and deer populations.
