1. Theoretical and empirical research has shown that increased variability in demographic rates often results in a decline in the population growth rate. In order to reduce the adverse effects of increased variability, life-history theory predicts that demographic rates that contribute disproportionately to population growth should be buffered against environmental variation. To date, evidence of demographic buffering is still equivocal and limited to analyses on a reduced number of age-classes (e.g. juveniles and adults), and on single sex models.
2. Here we used Bayesian inference models for age-specific survival and fecundity on a long-term dataset of wild mountain gorillas. We used these estimates to parameterize two-sex, age-specific stochastic population projection models that accounted for the yearly covariation between demographic rates. We estimated the sensitivity of the long-run stochastic population growth rate to reductions in survival and fecundity on ages belonging to nine sex-age-classes for survival and three age-classes for female fecundity.
3. We found a statistically significant negative linear relationship between the sensitivities and variances of demographic rates, with strong demographic buffering on young adult female survival and low buffering on older female and silverback survival and female fecundity. We found moderate buffering on all immature stages and on prime-age females.
4. Previous research on long-lived slow species has found high buffering of prime-age female survival and low buffering on immature survival and fecundity. Our results suggest that the moderate buffering of the immature stages can be partially due to the mountain gorilla social system and the relative stability of their environment.
5. Our results provide clear support for the demographic buffering hypothesis and its predicted effects on species at the slow end of the slow-fast life history continuum, but with the surprising outcome of moderate social buffering on the survival of immature stages. We also demonstrate how increasing the number of sex-age-classes can greatly improve the detection of demographic buffering in wild populations.

The study was carried out in Volcanoes National Park in Rwanda, on the groups of habituated mountain gorillas monitored by the Dian Fossey Gorilla Fund’s Karisoke Research Center, often referred to as the Karisoke subpopulation. Since 1967, groups in this subpopulation have been monitored and protected on a near daily basis. Through the mid 2000s, the Karisoke groups generally numbered three but over the last decade, group fissions and new group formations resulted in an average of 10 groups in the region (see Caillaud et al, 2014). During daily observations, detailed demographic data were recorded, such as dates of birth and death, dates and types of individuals’ entry (immigrants) and departure (emigrants) from the study population, group composition, and maternal relatedness (for further details see Strier et al. 2010 and Granjon et al. (2020). In particular, groups were frequently monitored (daily between 2010-2016), and the arrival of a new individual to a monitored group was recorded as immigration. When individuals were lost to follow, depending on age, sex, health and group movement individuals could be classified as emigrated. However, when in doubt, the fate was recorded as unknown (Granjon et al. 2020).

The Survival data consist of the number of individuals per age and sex (F = female, M = male, U = unknown), and per detection class. These are: a) 'Alive' when individuals entering the age interval survived until the end of the interval; b) 'Died' when individuals entering the age interval died during the interval; c) 'Censored' individuals known to have survived but were then not seen again (e.g. out-migration); and d) 'Disapeared' individuals that were not detected but their fait was unknown (i.e. they could have out-migrated or die).

The reproduction data consist of the number of females of a given age present at the study population within the study period, as well as the total number of infants produced by females of that age. 

The data consist of three columns: Age, Females, Infants. The Age column starts at the age at sexual maturity for mountain gorillas until the last age a female was present in the study. Columns Females and Infants depict the total number of each category for each female age.