The changes in vegetation in the African semiarid regions, induced by elephant activity or other causes, have implications for how elephants have adapted to their changing landscape. Two regions where ecosystem change has been well documented are Tsavo and Amboseli in Kenya (chapter 6). The Tsavo landscape changed from predominantly woodland to grassland during the 1960s and 1970s. Larry Tieszen and coworkers reasoned that, as woodland changed to grassland, it was possible that elephants changed their diets from browse to grass. This would be reflected in the carbon isotope ratios of bone collagen. It was more likely that the dietary shift would be reflected in the bone collagen of younger elephants, rather than in older elephants whose isotopic values would reflect an earlier period when browse was still available to them in intact
-30.0 4 "» "t i i i i i i i i i i i t i i i i i r i i r i i i i i i i i i i i i 0 50 100 150 200 250 300 350 Annual rainfall (cm)
The relationship between rainfall and mean carbon isotope ratio, expressed as S13C (per mil), in bone collagen across several African (closed circle) and Asian (open circle and triangle) elephant populations. An approximate scale for proportion of C3 plants in diet is shown on the right Y axis. The data for African populations are based on van der Merwe et al. (1988, 1990), Tieszen et al. (1989), Vogel et al. (1990), and Cerling et al. (1999). The data for Asian elephants pertain to a single population in southern India (open triangle includes all ages and open circle includes only adults older than 25 years) and are based on Sukumar and Ramesh (1992). The rainfall data are based on various sources.
woodlands. When they analyzed a sample of 56 elephants representing all age groups from Tsavo, they surprisingly found no difference in the C/C ratios across this age range. Elephants of all ages obtained an average of 75% of their carbon from C3 plants. It was obvious that, in spite of dramatic change in the habitat from woodland to grassland, the elephants seemed to be conservative in their dietary preferences, perhaps changing their movement patterns to continue to browse to the same extent as previously.
A more recent study in Amboseli found that some elephants did change their diet in response to vegetation change. In the 1950s, the Amboseli basin contained a vegetation mosaic, including dense and open woodlands (with C3
browse), as well as swamps and grasslands (of C4 grasses and sedges). By the early 1970s, there was a 90% decline in woodlands in the basin. The slopes of the Kilimanjaro mountain range to the south, however, are still covered with woodlands. Cynthia Moss, one of the authors of this study, has been studying the elephants here since 1972 and had collected the lower mandibles of several identified elephants that died. Elephants that died in the early 1970s, when woodland was still present, showed an average S13C value of -18 per mil, as opposed to a value of -13 per mil for those that died in the late 1980s. Clearly, the proportion of C3 plant contribution to collagen synthesis dropped from 75% to 40% over this period. Among those that died during the late 1980s, younger animals, whose bone growth was more recent, showed higher proportions of C4 plants in their diets compared to older individuals whose bone growth would have largely occurred prior to the changes in vegetation.
More interesting insights came from microanalysis of elephant teeth. While bones hold a relatively longer-term record of an animal's diet because of slower turnover, the tooth dentin grows by adding layers and thus retains a subannual record of body chemistry. P. L. Koch and coworkers took micro-samples of sequential growth laminations from the molar roots of four elephants at Amboseli. Each microsample of dentin represented about 3 months of growth. Three of these elephants showed a regular cyclic S C variation, most likely the result of seasonal shifts in the proportions of browse and grass.
In spite of an overall shift in diets from browse to grass, two elephants at Amboseli defied this trend and continued to feed mainly on browse into the 1980s. Another female elephant, which died in 1987 at an estimated age of 49 years, showed a shift to browse just prior to death. These animals obviously found a way to obtain browse from areas outside the park boundary. More important, even those elephants that lived their entire lives in the Amboseli lake basin, transformed into grassland by the 1980s, still obtained between 35% and 50% of their protein from C3 plants. The Amboseli elephants thus engaged in relatively local movements to obtain browse, possibly adapting to the changing habitat through feeding at night in the woodlands at the southern edge of the park. A diminishing supply of browse was preferentially exploited by these elephants.
Was this article helpful?