Population changes in elephants of North Bunyoro Uganda

The elephant population in the North Bunyoro region of Uganda had been affected during 1925-1965 through controlled shooting by license. A minimum of 14,416 elephants had been killed during this 40-year period according to official records. In addition, an unknown number probably succumbed to injury sustained during these operations and to ivory poaching. The population structure of these elephants was thus already distorted considerably by 1965 when the Ugandan National Parks began a further round of cropping to keep a check on the population size. During 1965-1967, another 2,000 elephants were culled in the Murchison Falls National Park. These provided the material for detailed scientific investigations by Richard Laws and his collaborators.

When a population in near equilibrium with its habitat is reduced through hunting, we can expect compensating factors. As a result of increased availability of food and other resources per capita, the remaining individuals in the population could show one or more of the following: faster growth, earlier puberty, increased fecundity, and lower natural mortality. These demographic traits of the culled sample of elephants from Murchison Falls examined by Richard Laws and coworkers, however, showed exactly the opposite tendencies. There was slower growth, an (presumed) increase in age of puberty (to 16.3-17.8 years) in females, decreased fecundity (intercalving interval of 5.6 and 9.1 years in the northern and southern populations, respectively), and increased mortality (by over 50%) of calves compared to the more productive African elephant populations. The sharp reduction in recruitment clearly showed up in the age structure of the culled sample as a deficiency in the younger age classes. From the age structure, it seemed as though the decline in fecundity had commenced about 20 years earlier, that is, around 1946.

In an attempt to explain what may have happened to the Murchison Falls elephant population, Laws and his team used a simple deterministic model based on the fundamental equation of population growth modified for their purpose, Nt = Noe~zt, where Nt is the abundance of age t, No is the number of newborns, e is the natural exponent, and z is the instantaneous mortality rate.

They took No, the number of newborns, to be an arbitrary 1,000 (later adjusted to 1,197), sexual maturity in females to be 12 years, an intercalving interval of 4 years, and an equal ratio of male to female calves at birth. Taking the calculated adult mortality rates, but with a little adjustment of calf mortality rates, they generated the age structure of a "steady-state" population, in which the births were balanced by deaths. They believed this represented the age structure of the 1946 elephant population, before the decline in recruitment and increase in calf mortality started (fig. 7.6).

The model was then run to describe the 1966 situation. Sexual maturity in females was changed to 18 years and the intercalving interval to 7 years. By 1966, the number of newborns had declined (from 1,197 newborns in 1946) to 457 individuals. Based on observed age structures from the culled sample, the model further indicated that calf (0-4 years old) mortality rates had increased anywhere between 50% and 100% of the pre-1946 rates. There was no indication that postweaning (after 4 years old) mortality rates had changed. The model was then taken forward to the year 1971, which predicted further decrease in the number of recruits (407 newborns) and the abundance of elephants up to 4 years of age.

Summing the population sizes from various age classes, beginning with an adjusted value of 1,197 newborns, the model indicated a total population size in 1966 of 21,195 individuals compared to the 9,400 elephants indicated by

Figure 7.6

Hypothetical population models for the North Bunyoro elephant population in Uganda at different time periods. (a) The solid lines indicate population structure estimated from sample cropping in 1965-1967, while the broken line makes allowance for numbers taken in control shooting operations since 1946. The scale on the right represents an initial assumed value of N0 = 1,000 (from Laws et al. 1975). (b) Pseudostable age distribution of the simulated population including that expected at stability under different elephant densities per square kilometer. (From Fowler and Smith 1973. Reproduced with permission granted by the Journal of Wildlife Management, and its publisher The Wildlife Society, U.S.A.)

Figure 7.6

Hypothetical population models for the North Bunyoro elephant population in Uganda at different time periods. (a) The solid lines indicate population structure estimated from sample cropping in 1965-1967, while the broken line makes allowance for numbers taken in control shooting operations since 1946. The scale on the right represents an initial assumed value of N0 = 1,000 (from Laws et al. 1975). (b) Pseudostable age distribution of the simulated population including that expected at stability under different elephant densities per square kilometer. (From Fowler and Smith 1973. Reproduced with permission granted by the Journal of Wildlife Management, and its publisher The Wildlife Society, U.S.A.)

actual census figures. The relative abundances of various age classes for 1966 were thus proportionately scaled down by a factor of 0.44. The model-derived population for 1946 was now calculated to be about 16,000 elephants. To this, the 6,000 elephants killed during 1946-1966 in controlled shooting and sport hunting had to be added, giving a total of 22,000 elephants for the 1946 North Bunyoro population. The overall population trends at North Bunyoro were 22,000 elephants in 1946, declining to 9,400 by 1966 and further projected to 7,900 by the year 1971.

A model such as this has limitations because of several assumptions made in respect to functional relationships and a stationary and stable aged population in 1946. Drawing inferences about demography from cross-sectional sampling of age structures is always a tricky business. Nevertheless, the model was useful in at least describing qualitatively the demographic trends in the North Bunyoro elephant population over a period of a quarter century. In this case, the standing age structure during the 1965-1967 cull clearly showed that recruitment had declined substantially, and it strengthened the conclusions drawn independently on reproductive rates from postmortem examinations of the elephants. There still remained the question of why the demography of this population had slowed rather than sped up after a population reduction.

Laws and his colleagues hypothesized that density-dependent compensatory mechanisms served to slow the demography of this population. In spite of a substantial decline in elephant numbers since 1946, there was also a decrease in habitat area. From 6,300 km in 1946, the elephant range contracted by almost half to 3,200 km2 by 1966. The elephant density thus decreased only marginally from a mean of 3.50 elephants/km2 in 1946 to 2.94 elephants/ km2 by 1966.

At the same time, there was an increase in the mean weight of an elephant over this time, from 1,894 kg in 1946 to 2,234 kg in 1966 (with reduced recruitment, there was a preponderance of older, heavier animals). A biomass calculation showed that the unit weights of elephants changed from 6,633 kg/km2 in 1946 to only 6,561 kg/km by 1966, or a 1% reduction. During this time, there was significant decline in habitat quality (conversion of woodland to grassland), strongly suggesting that density-dependent regulatory factors were responsible for the reduced birth rate and increased calf mortality rate.

0 0

Post a comment