|
Tapirs:
Status Survey and Conservation Action Plan
Published 1997
Status and Action Plan of the Lowland Tapir
(Tapirus terrestris)
Continued from Previous Page
Susceptibility of lowland tapir to overhunting in the western Amazonian basin
Several methods have been used to evaluate whether tapirs are overhunted, causing or threatening population declines. These include population growth models, population production models, density comparisons, age structure models, and comparisons of hunting effort.
Population growth models and a case in the Peruvian Amazon
Robinson and Redford (1991, 1994) developed a population growth model that evaluates whether hunting of lowland tapir is possibly sustainable if the population is at maximum game production. The model generates the maximum potential production (Pmax) of tapir using the equation
where K is the density at carrying capacity and fmax is the maximum finite rate of increase, which is the exponential of the intrinsic rate of increase (rmax). rmax of tapir was calculated using Cole's (1954) equation
where (a) is the age of the first reproduction, (w) is the age of last reproduction, and b is the annual birth rate. The age of first reproduction for lowland tapir was determined from the literature to be 3.7 years, the age of last reproduction 23.5 years, and the annual birth rate 0.38. This yielded an rmax value of 0.20 and a Pmax value of 0.16, using a carrying capacity of 1.61 ind./km2 that was calculated from the average density of lowland tapir (Robinson and Redford 1991).
Since little is known about the response of tropical fauna to hunting Robinson and Redford (1991) assumed that maximum production of lowland tapir would be at 60% of the carrying capacity due to density dependent effects. Thus, in the maximum production equation carrying capacity (K) is multiplied by 0.6. Robinson and Redford (1991) also assumed that the average life span would be a good index of the proportion of production that could go into harvests, with the remainder being left for natural mortality. In long-lived animals, such as tapir, natural mortality is usually low, and harvests must therefore take a small proportion of production if the population is to remain stable over time, i.e. sustainable. Robinson and Redford (1991) placed the lowland tapir into the long lived category which only permits 20% of production to be harvested in order to maintain a stable population. Therefore, Robinson and Redford (1991) estimated that 0.03 tapirs could be harvested per km2 at a sustainable level under maximum game production (0.2 x Pmax).
Robinson and Redford (1994) acknowledge that this model is most accurate when it evaluates whether a harvest is sustainable. The model, however, cannot tell whether a given harvest is sustainable, because it does not consider depleted game densities, less than maximum birth rates, or higher than minimum mortality rates, among other factors.
Alvard (1993) applied Robinson and Redford's population growth model by calculating the annual potential harvest in kg/km2 that could be harvested without causing a depletion in tapir populations. This value was estimated at 4.47kg/km2 (Robinson and Redford 1991). Alvard (1993) then compared this potential harvest to observed harvests of tapir in two Indian communities in Madre de Dios, Peru. Both of these communities, the Piro of Diamante and the Machiguenga of Yomiwato were harvesting at levels above those predicted to be sustainable, with the Piro taking an annual harvest of 14.1kg/km2 Of tapir and the Machiguenga taking an annual harvest of 10.6kg/km2.
Population production models in Peru and Bolivia
Production models are also used to evaluate whether lowland tapir populations have been overhunted. Bodmer (1994) developed a production model that compares productivity to harvest. Productivity is calculated by multiplying reproductive output (young produced/ind/yr) by the density of individuals (ind/km2). This yields a production value in terms of young produced/km2/yr that can be directly compared to hunting pressure (estimated at individuals hunted/km2/yr) to yield the percentage of production that is removed by hunting. Therefore, areas where hunters are taking more than 20% of tapir production are likely to be overharvested. This is because lowland tapir are long lived animals that can only have 20% of their production harvested in order to maintain a stable population (Robinson and Redford 1991).
Townsend (1995) used this production model to evaluate hunting by the Sirono Indians of Bolivia. She calculated tapir reproduction from examining female tapir harvested by hunters. 50% of female lowland tapir were reproductively active. This value was multiplied by 0.5 gestations per year, 1.0 young per litter, and a 0.5 sex ratio to yield a productivity of 0.13 young/ind/yr. Townsend (1995) estimated lowland tapir densities from the literature and used a minimum and maximum density estimate. This yielded an Upper and lower estimate of production. Townsend's lower estimate indicated that 45% of tapir production was being harvested by the Sirono and an upper estimate indicated that 179% of production was being harvested. Both of these harvests are well above the 20% limit of the production model, indicating that the Sirono were overharvesting lowland tapir.
Bodmer et al. (1993) also used this production model to evaluate hunting of lowland tapir in the northeastern Peruvian Amazon. By examining female tapir hunted by rural Ribereño people, Bodmer et al. (1993) estimated that 50% of female tapir were actively reproducing. This value was multiplied by the average number of gestations per year, estimated from the literature it 0.5, and the average litter size of tapir of 1.0 young per gestation. Assuming a 1:1 sex ratio, the average number of young/ind/yr was calculated at 0.12 young/ind./yr (Bodmer 1993). Surveys estimated an overall density of lowland tapir for the study area at 0.4 ind./km2 yielding a production of 0.05. Hunting studies in the same area revealed a tapir harvest of 0.08 ind/km2/yr. Thus, hunters were taking 160% of lowland tapir production, which is well over the 20% limit. Indeed, the harvest is taking more animals than the population is producing, which in turn should result in a rapid population decline from overhunting.
Density comparisons and age structure models in Peru
Bodmer et al. (1993) corroborate the decline of lowland tapir in northeastern Peru using density comparisons and age structure models. Lowland tapir densities were compared between slightly hunted and persistently hunted sites within the same habitats. In this study the density of lowland tapir decreased substantially from 0.4 tapir/km2 at the persistently hunted site versus 0.6 tapir/km2 at the slightly hunted site(Bodmer 1993).These differences appear to reflect a decrease in the populations of lowland tapir in the persistently hunted site. In contrast, rainforest peccaries, brocket deer, and large caviomorph rodents showed only slight differences in densities between the two sites.
Secondly, age structure curves of tapir may respond to hunting pressures and provide an index of survivorship that can be used to evaluate the condition of a population. The age structure of lowland tapir was compared between persistently and slightly hunted sites. In the persistently hunted site, populations of lowland tapir had lower proportions of older individuals than the slightly hunted site which may indicate overhunting in the persistently hunted site (Bodmer 1995). In contrast, the age structure curves of rainforest peccaries and brocket deer from these same sites depict populations that show little difference in the proportions of individuals in the older age class. This suggests that the peccary and deer populations in this region are not affected as greatly by hunting as the lowland tapir populations.