The importance of farmland for the conservation of the brown hyaena Parahyaena brunnea

1. What they did

Many protected areas are too small for wide-ranging carnivores, so populations living on farmland may matter a great deal for some species. The brown hyaena lives only in southern Africa, is listed as Near Threatened, and Botswana is thought to hold about half of the roughly 8,000 left. Most of what is known comes from protected areas, and a South African study had found fewer brown hyaenas on farmland than in protected areas.

The authors set out to count brown hyaenas on commercial farmland in the Ghanzi farm block of western Botswana, compare the numbers with protected areas, and see whether cattle farms or game farms suited them better. They used two methods:

• Spoor surveys. Two surveys of single-track farm roads, driven slowly from sunrise until 11:00, each until about 1,000 km had been covered. A skilled Bushman tracker sat on the front of the vehicle and recorded all fresh (24 hours old or less) tracks of medium- and large-sized predators. Survey 1 (six transects, wet season 2008 to 2009) covered a mix of cattle, small-stock and game farms; survey 2 (five transects, dry season 2009) was on one cattle farm.
• Camera-trap surveys. Three 62-day surveys in 2009 with 26 cameras on farm tracks, no bait or lure: one on the same cattle farm, one mainly on two game farms, and one on two other cattle farms. Most stations had a single camera, with up to eight extra cameras moved around to make paired stations, so that both flanks of each hyaena could be photographed and individuals recognised from their leg stripes, ear notches and other markings.

Spoor counts were turned into density with a published calibration for sandy soils. Camera data were analysed with a newer statistical method, spatially explicit capture–recapture, which takes account of where the cameras were.

2. What they found

Spoor surveys

• Brown hyaena spoor was recorded 81 times in survey 1 and 99 times in survey 2, giving 8.8 and 10.1 tracks per 100 km.
• The estimates became stable after about 20 to 30 spoor samples (about 170 km and 120 km of driving).
• Estimated density: 2.67 and 3.08 hyaenas per 100 km² (mean 2.88).

Camera surveys

• About 3,200 camera-trap days at 56 stations; brown hyaenas were photographed at 46 of them.
• Individually identified hyaenas used in the analysis: 9, 7 and 9 in the three surveys. One hyaena with a wire snare around their neck vanished from the photographs after three weeks and was left out.
• Estimated density: 2.81, 1.8 and 2.28 hyaenas per 100 km² (mean 2.3).

Comparisons

• These densities match or exceed those reported for the southern Kalahari (1.8 per 100 km²) and are comparable to Pilanesberg National Park in South Africa (2.8 per 100 km²).
• Both spoor estimates were higher than the camera estimates from the same area.
• Density was lower on game farms than on cattle farms: in the camera surveys by 21 to 36%.
• A 2007 spoor survey on the same farms, with the same tracker, gave 2.36 per 100 km², about 12% lower than in 2009, which suggests the population may have grown.
• Scaled up to the whole farm block, the estimates give about 345 (camera) to 430 (spoor) brown hyaenas, 4 to 5% of the estimated world population.

3. What they concluded

• Brown hyaenas appear to be thriving on Ghanzi’s commercial farmland, at densities similar to protected areas. Non-protected areas may hold a significant share of the world population and be critical for the species’ conservation.
• Livestock carcasses probably give a more reliable supply of food to scavenge, and fewer competing predators let hyaenas feed undisturbed. This supports the idea that livestock areas can benefit brown hyaenas, in contrast to the South African finding.
• Possible reasons for the difference from South Africa: farmers in Ghanzi seem less hostile to brown hyaenas than to leopards, lions, cheetahs and jackals; livestock range freely with little supervision at low stocking densities; very few people live in the district; and there is little large-scale cropping.
• Spoor surveys are cheap but time-consuming and can overestimate density through double counting. Camera surveys cost more to set up, but spatially explicit capture–recapture gives more realistic estimates; simpler capture–recapture methods gave higher or lower figures depending on the assumed area.
• Cautions: brown hyaenas live in clans, so their home range centres may not be independent, as the model assumes; nomadic hyaenas passing through may inflate estimates; the mixed single and paired camera design may bias identification; and the area surveyed was small compared with the whole farm block.

4. How tracking was or was not used

Tracking was central to the spoor surveys. A tracker riding on the front of the vehicle spotted and identified fresh tracks of medium- and large-sized predators on the road and judged whether they were 24 hours old or less. At each set of brown hyaena tracks, the spoor was examined and followed to see how far the hyaena had walked along the road and which way they went next. To avoid counting the same animal twice, fresh hyaena tracks within 1 km of an earlier set were treated as the same individual. Roads were not surveyed within 24 hours after rain or when another vehicle had driven on them in the previous 12 hours.

5. Tracker or observer certification

Reported but not validated. The tracker is described as “a skilled Bushman tracker”, supported by the statement that Bushmen are renowned for the accuracy and reliability of their tracking skills, but no certification or test of his skill is reported. The acknowledgements name him as Kaoga Xashe and thank him for his “peerless Bushman skills and knowledge”. He also did the 2007 spoor survey on the same farms. The spoor density estimates, and the comparison with the camera results, depend on his identification and ageing of tracks.

6. Other technologies, skills or innovations

• Cuddeback Digital Expert cameras on farm tracks, 25 to 40 cm above the ground, 2 to 2.5 km apart, checked weekly; photos managed in Camera Base.
• A mixed design of single and paired camera stations, with the extra cameras moved around, to cover more ground and still identify individuals with few cameras.
• Bayesian spatially explicit capture–recapture in SPACECAP, and ArcGIS buffers to estimate the area surveyed on farmland.
• Bootstrap analysis to judge how many spoor samples were enough.

7. Implications for conservation (from the paper)

• Farmland outside protected areas may be critical for brown hyaena conservation, much like Namibian farmland for cheetahs.
• Brown hyaenas are persecuted by trapping and poisoning, but much of this is aimed at other predators.
• Continued education on the benefits of brown hyaenas as “cleaners of the bush”, which helps prevent disease, together with good livestock and land management, could let commercial farmland make an important contribution to their conservation.
• More surveys outside protected areas may show higher densities than thought and raise the global population estimate.