Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Individual study: The effectiveness of traditional livestock husbandry as a conservation tool by reducing livestock-predator conflicts therefore enhancing coexistence with local herders, northern Kenya

Published source details

Woodroffe R., Frank L.G., Lindsey P.A., ole Ranah S.M. K. & Romañach S. (2007) Livestock husbandry as a tool for carnivore conservation in Africa's community rangelands: a case–control study. Biodiversity and Conservation, 16, 1245-1260

Summary

Conflict between people and wild vertebrate predators is a major issue in African savannas, where species such as African wild dog Lycaon pictus, cheetah Acinonyx jubatus, lion Panthera leo, leopard Panthera pardus and spotted hyaena Crocuta crocuta may kill livestock and are therefore themselves killed by local herders. This has led to the extirpation of these species from many areas. To investigate the possibilities for coexistence of people, livestock and large predators in rangelands, the effectiveness of traditional livestock husbandry in reducing predation by wild carnivores was assessed.

Study area: The study was carried out between January 2001 and June 2005 in Laikipia District (37º2'E, 0º6'N), and parts of adjacent Samburu, Isiolo and Baringo Districts, northern Kenya. The habitat is mainly semi-arid bush and savanna used for. Wildlife is abundant in some areas with leopards, lions, cheetahs and spotted hyaenas persisting, and wild dogs recolonising Laikipia in 2000, despite extensive livestock use (subsistence pastoralism and commercial ranching) and widespread human presence.

Data collection: Most data were gathered by 18 field staff employed in partnership with local non-governmental conservation organisations. They investigated reported livestock attacks to confirm the genuineness of each; no compensation was available for livestock losses, so farmers had no
financial incentive to misrepresent losses. The following were recorded:

i) location and time of an attack; ii) whether livestock were killed while grazing or when in a boma (enclosure), and iii) number and age of livestock killed or injured.

Data on livestock husbandry on the day of the attack were then collected for the herd or boma which had been attacked and from one to three nearby 'control' herds or bomas that had not been attacked. The following were recorded:

For grazing herds: the number of livestock and species; the number of herders and domestic dogs accompanying the herd; terrain (i.e. flat, rolling hills, single escarpment, hilly, mountainous);
density of the vegetation in the grazing area (i.e. open grassland, light bush, dense bush, very dense bush).

For bomas: construction i.e. thornbrush (constructed of whole trees or large branches cut and laid on their sides), solid (constructed of poles or stone), wire, or open (where livestock are bedded down with no barrier between them and the surroundings); the number of 'predator deterrents' at the boma on the attack night i.e. number of people, number of houses, number of domestic dogs, number of guns, number of fires burning outside bomas, and number of scarecrows (cloth hung in nearby trees or on boma walls); number of livestock present (cattle, sheep and goats, camels, donkeys); terrain and surrounding vegetation density.

The timing and nature of attacks varied among predators: hyaenas mainly attacked herds in bomas at night, cheetahs and wild dogs only herds grazing by day, and leopards and lions attacked both grazing herds and those in bomas.

Of 502 grazing herds (cases and controls combined) for which data were available, 97% were accompanied by herders; 24% were accompanied by one or more domestic dogs (average1.3, range 1–3). Daytime attacks were perceived to have been initiated by the predator approaching (rather than the herd stumbling upon a resting predator) in 95% of 19 cheetah attacks, 85% of 40 wild dog attacks, 91% of 11 hyaena attacks, 65% of 17 lion attacks and 65% of 49 leopard attacks. Leopard attacks were more likely for sheep and goat herds unaccompanied by domestic dogs (odds ratio 0.091), and those that grazed in more dense habitat (open and light bush vs. dense and very dense bush; odds ratio 7.33). There were no significant predictors of lion or cheetah attacks, although for cheetah there was a non-significant trend suggesting that attacks might be more common in sheep and goat herds unaccompanied by domestic dogs. In contrast, there was a non-significant trend suggesting that such herds accompanied by domestic dogs might be more likely to experience attacks by wild dogs

Of 491 bomas examined (cases and controls), 446 (91%) were constructed of thornbrush, 27 (5%) of poles or stone, 12 (2%) of wire, and six (1%) were open. The average thornbrush boma had 2.4 gates (range 1–12). Cattle were present at 62% of 477 bomas with on average, 54.1 (range 1–603) adult cattle and 9.5 (range 0–83) calves. Sheep or goats were present at 86% of 477 bomas with on average 144.1 (range 4–1300) individuals present. The average boma was occupied by 11.3 people. Domestic dogs (average of 2.0/boma) were present at 71% of 484 bomas. Fires were left burning outside 22% of 481 bomas, and scarecrows (average of 2.4/boma) were present at 44% of 483 bomas. Only 6% of 483 bomas had a gun. Contary to expectations, presence of scarecrows increased the likelihood of attack (odds ration 1.45), as did number of gates (1.40 odds ration); presecence of domestic dogs greatly reduced the risk of attack (dogs present vs. absent 0.412).

The authors suggest that these findings could lead to improvements in livestock husbandry which would reduce predation rates and contribute to the conservation and recovery of large carnivores in these community rangelands, although other measures e.g. prey conservation are also likely to be necessary.


Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: http://www.springerlink.com/content/ng4474577544307w/fulltext.pdf