Censusing large mammals in Kibale National Park: evaluation of the intensity of sampling required to determine change

Abstract Monitoring programmes are essential for management of large mammal populations because they can detect population change. It is vital that we have the means to evaluate the effectiveness of protected areas. Kibale National Park is a stronghold for large mammal conservation in Uganda. Past wildlife surveys in Kibale focused on specific taxa or areas, but our large mammal survey covered the entire protected area and we evaluated the intensity of sampling required to determine population change. Using line transect sampling, we found that the distribution of large mammals was nonrandom and related to habitat‐type. However, confidence intervals of population estimates revealed that much more intensive sampling was required to detect changes in population density at a time scale reasonable for management. For many species, populations would have to decline by 40–60% for this method to detect population change. Post‐stratification decreased confidence intervals of density estimates slightly, increasing our ability to detect change. However, confidence intervals of estimates were still too large to detect a meaningful population change on a time scale that would allow management to take action. Most incidences of illegal activity were about 5 km from the park boundary; however, animal densities were not lower in this area.