NEST-BUILDING BY LOWLAND GORILLAS IN THE LOPE-RESERVE, GABON - ENVIRONMENTAL-INFLUENCES AND IMPLICATIONS FOR CENSUSING

We analyzed data from 373 fresh nest-sites (containing 2435 nests) of lowland gorillas (Gorilla g. gorilla) during a 4-year period in the Lope Reserve, Gabon, to determine whether the observed variability in nest building was due to environmental influences. We recognized and defined seven types of nest in terms of the degree of construction and the raw materials used. Overall, nests built on the ground from herbaceous planes are the most common type (40%), followed by tree nests (35%). Frequencies of the different nest-types vary significantly between eight habitat-types. In habitat-types with high densities of understory herbs, ground nests predominated, but when herbs were rare, the majority of nests were in trees. A general preference for sleeping in herbaceous ground nests is indicated since trees are abundant in all habitat-types, except savanna. The frequency of nesting in trees shows a significant positive correlation with rainfall, but effects of climate are confounded by seasonal variation in use of different habitat-types. When elephants were attracted to the same localized food sources as gorillas, many tree nests were built even when herbs were available. We conclude that different nest-types reflect a variety of solutions to maximize comfort, depending on available raw materials and the probability of rainfall or disturbance by elephants or both factors. Nests are a powerful tool for population censuses and demographic studies of great apes, but problems exist in interpreting data on lowland gorilla nests. Results from this analysis show that only a third of nest-sites accurately reflects group size (of weaned individuals) and that 26% of an gorilla nest-sites could be mistaken for those of chimpanzees, as all nests, or all those visible from a transect, were in trees. Gorilla nests at Lope were nonrandomly distributed with respect to habitat-types, and nest construction varied seasonally, thereby introducing sources of bias to transect nest counts. We discuss these problems and ones related to assessing the decay rate of nest-sites and make recommendations relevant to census work.