Controlling the distribution of elephant

1 Chapter Seven 2 Controlling the distribution of elephant 3 Lead Author: C. C. (Rina) Grant 1 4 Authors: Roy Bengis2, Dave Balfour3 & Mike Peel4 5 Contributing Authors: Warwick Mostert5, Hanno Killian6, Rob Little7, Izak Smit1, 6 Marion Garai8 & Michelle Henley9 7 8 1 South African National Parks 9 2 National Department of Agriculture 10 3 Eastern Cape Parks 11 4 Agricultural Research Council 12 5 Venetia-Limpopo Nature Reserve 13 6 Welgevonden Private Nature Reserve 14 7 World Wild Fund for Nature, South Africa 15 8 Space For Elephant Foundation 16 9 Save the Elephants Transboundary Elephant Research Project 17 18 19 Introduction...............................................................................................................4 20 Purpose of fencing.....................................................................................................5 21 The containment of wildlife....................................................................................5 22 Disease control........................................................................................................6 23 Diseases that can be transmitted from wildlife to domestic stock......................6 24 Protection of livestock and crops............................................................................9 25 Consequences of fence breakages..........................................................................11 26 Loss of animals .....................................................................................................11 27 High costs involved in returning the animals or to destroy them .....................11 28 Loss of scarce or expensive species..................................................................12 29 Domestic stock entering wildlife areas.................................................................12 30 Economic impacts.............................................................................................12 SECOND DRAFT Assessment of South African Elephant Management 1 1 Ecological effects of domestic stock in wildlife areas......................................12 2 Other conservation uses of fences..........................................................................12 3 For protection of vegetation..................................................................................12 4 For understanding system function.......................................................................13 5 For the protection of individual trees....................................................................13 6 For the protection of infrastructure.......................................................................14 7 Efficacy of fences.....................................................................................................15 8 To contain elephant in a wildlife enterprise..........................................................15 9 In disease control ..................................................................................................18 10 Consequences of restriction of movement by fences............................................20 11 Fences and elephant welfare. .................................................................................24 12 Technical specifications for fences and their maintenance.................................25 13 Ensuring efficiency of fences..................................................................................28 14 Alternative methods to change elephant distribution .........................................29 15 Surface water manipulation as elephant management tool...................................29 16 Disturbance as a managing tool ............................................................................33 17 Changing behaviour as a management tool .........................................................34 18 Repellents..............................................................................................................36 19 Buffer Crops .........................................................................................................37 20 Moats and ditches .................................................................................................37 21 Stone walls............................................................................................................37 22 Effects of fence removal or the lack of fencing ....................................................37 23 Legal obligations for fencing..................................................................................40 24 Legal consequences of fencing ...............................................................................40 25 Legal obligation of owners of fenced properties. .................................................40 26 Conclusion ...............................................................................................................40 27 Research Gaps.........................................................................................................41 28 SECOND DRAFT Assessment of South African Elephant Management 2 1 2 Figures 3 Figure 7.1: A map indicating the foot and mouth disease control areas in South 4 Africa. Fence breakages in these areas have serious consequences for disease 5 control. ........................................................................................................................8 6 Figure 7.2:. Wire netting can be used to protect large trees from ring barking but 7 does not stop trees from being pushed over or broken. Photos from Mapungubwe 8 National Park ............................................................................................................14 9 Figure 7.3: Male elephant returning to KNP over border fence ..............................16 10 Figure 7.4: Reports of elephant breakages of fences between Kruger and the 11 Limpopo province from January 2001 to October 2004...........................................17 12 Figure 7.5: Animal migration routes prior to the erection of the foot and mouth 13 fence (Whyte, 1985) .................................................................................................22 14 Figure 7.6: Diagram of electric wires for elephant proof fence with an example of 15 such a fence in Mapungubwe National Park. ...........................................................26 16 Figure 7.7: Distribution and density patterns of elephant in Kruger. Note the 17 concentration along the drainage lines (Courtesy of Sandra MacFayden) ...............30 18 Figure 7.8: Changes in elephant numbers in Kruger with the change in management 19 policies: waterpoint closure and stopping of culling. ...............................................31 20 Figure 7.9: The was a slow increase of elephant in Sabie Sand Wildtuin with the 21 removal of the fence between Kruger and the private reserves. Elephant don’t move 22 in numbers to new areas............................................................................................38 23 Figure 7.10: Season movement of three elephant families between Sabie Sand 24 Wildtuin and Kruger. ................................................................................................39 25 26 Tables 27 Table 7. 1: Case study illustrating the ecological and economic effect of fencing 28 and water provision on the ecology of areas of small size. ......................................23 29 Table 7.2: Specifications, erection and maintenance costs for different types of 30 fences. All electrified strands must have a minimum voltage of 6000 V and must be 31 have sufficient energisers to supply power to maintain this voltage over a distance 32 of 8 km. .....................................................................................................................27 SECOND DRAFT Assessment of South African Elephant Management 3 1 2 3 Introduction 4 The general impression left on my mind was that, with civilization closing in 5 on all sides, ultimately something must be done to segregate the game areas 6 from those used for farming; otherwise sooner or later some excuse for 7 liquidation of the wild animals will be found … North of the Letaba River 8 the country West of the Park consists mainly of native locations and areas. 9 Here the Park itself might be fenced off. 10 11 Of course, a suitable fence over 200 miles long would be a most expensive 12 undertaking, and its upkeep considerable. It would have to traverse all kinds 13 of country, including stony hill ranges, and dense bush, but to my mind one 14 of the chief difficulties would lie in the wide sand rivers running from west to 15 east, and subject to annual heavy floods, which would carry away any kind 16 of fence, and on their subsidence leave the way open for animals to pass 17 freely up and down the river bed. 18 19 (National Parks Board of Trustees. 1946. Annual Report of Warden, Kruger 20 National Park –1945. J. Stevenson-Hamilton, KNP Warden. Dated 23rd 21 January 1946, pp 11-12) 22 23 Physical barriers, such as fences are passive control measures (Cumming & Jones, 24 2005) and are often seen as the most effective approach to containing elephant. 25 There are several reasons for the containment of wildlife and particularly elephant. 26 One of the most important reasons for using fences in a large conservation area such 27 as the Kruger National Park (KNP) is animal disease control (Freitag-Ronaldson & 28 Foxcroft, 2003) – to protect livestock from wildlife-associated diseases, and also to 29 protect wildlife from diseases of domestic species. Containment of wildlife is 30 another very important reason for fencing, to protect neighbouring communities 31 from damage by wildlife especially elephant and predators as well as infrastructural 32 damage caused by wild animals. Furthermore, by fencing a property ownership of 33 the species present is understood and animals are protected from illegal hunting. SECOND DRAFT Assessment of South African Elephant Management 4 1 Fences are however not the only option to create barriers and several other options 2 are discussed here. 3 4 5 Purpose of fencing 6 The containment of wildlife 7 Fenced landscapes allow people and elephants to share a landscape without the 8 problems associated with human elephant conflict (Hoare, 2001) (Chapter Four). 9 Extensive wildlife areas in South Africa are distributed amongst ranching areas with 10 domestic stock and crops which make it imperative to use fences to contain the wild 11 animals. When elephant are included in the wildlife area the fences have to be 12 upgraded to be able to contain the wildlife efficiently. The fencing also has to be 13 more efficient and sturdy when relatively small conservation areas are located 14 within agricultural areas These fences have to comply with legal requirements that 15 are stipulated for the different species that may be included in a wildlife area and 16 have to be maintained according to these specifications. 17 18 It is only in southern Africa, and South Africa in particular, where fencing plays a 19 large role in the wildlife and conservation industry (South African Savannas 20 Network, 2001). In most other parts of Africa the existing national parks and game 21 reserves have never been fenced, and have always maintained and supported 22 wildlife populations. In addition, many of these conservation areas also seasonally 23 support pastoral tribes making use of shared grazing. These communities had to 24 adapt to the activities of their wild neighbours, and many types of localised (village 25 level) physical barriers and deterrents (thorn bomas and ditches) as well as noise 26 and smell, have been used to protect crops and livestock. 27 28 In contrast, the wildlife industry in southern Africa has evolved and expanded 29 during the last two decades (Smith & Wilson, 2002, South African Savannas 30 Network, 2001). Much of this expansion took place right in the middle of existing 31 agricultural areas, or close to community settlements. Furthermore, where most of 32 the remaining large wildlife used to be conserved in the larger national and SECOND DRAFT Assessment of South African Elephant Management 5 1 provincial parks, smaller private reserves and game farms are playing an 2 increasingly important role in the conservation of these species 3 4 It is now the responsibility of the landowner or manager of the particular 5 conservation area, whether state or privately owned, to ensure that the animals they 6 keep in the conservation areas do not interfere with the neighbouring communities’ 7 livelihoods, or do damage to their property or crops. The landowner has a legal 8 obligation to all adjacent owners for damage that escaped animals can cause, as well 9 as public liability in case of death or injuries or damage to property in the event of 10 the animals breaking through the perimeter fence (Chapter Eleven). 11 12 Disease control 13 Diseases that can be transmitted from wildlife to domestic stock 14 Certain indigenous animal diseases that are carried and maintained by wild animals, 15 can be very infectious to livestock thus form a threat to the livestock industry. In 16 southern Africa, the use of fencing and other disease control measures such as 17 proclamation of animal disease control zones, and permit requirements, to strictly 18 control the movement of wildlife and livestock, has played an important role to 19 enable the countries of the region to access beef and other livestock markets in 20 Europe and elsewhere in the developed world. Directly contagious diseases such as 21 rinderpest, foot and mouth (FMD) and malignant catarrhal fever as well as diseases 22 transmitted by flightless vectors such as African swine fever and Corridor 23 disease{Theileriosis} ) can be effectively managed by barrier fencing (Bengis et al. 24 2002). In contrast, barrier fences are ineffectual when dealing with diseases 25 transmitted by winged vectors, such as trypanosomiasis, African horse sickness, 26 bluetongue and Rift Valley fever. 27 28 Foot and mouth disease, rinderpest and African swine fever have the potential for 29 very rapid spread, and are listed by the Organisation International Epizooties (OIE = 30 World Organisation for Animal Health) as important animal health threats, because 31 these diseases, may have serious local, national and international animal health 32 implications. These diseases do not only cause local losses during outbreaks, but SECOND DRAFT Assessment of South African Elephant Management 6 1 due to their epidemic potential, they affect international trade in livestock and their 2 products, as well as other agricultural commodities, with serious socio-economic 3 consequences. 4 5 In southern Africa, buffalo are the most important disease risk to cattle because they 6 carry several diseases that affect livestock production, including foot and mouth 7 disease, Corridor disease, bovine tuberculosis and brucellosis. The Animal Diseases 8 Act (35 of 1984) highlights specific responsibilities of owners or managers of 9 properties with buffalo, including effective containment. 10 11 In South Africa, foot and mouth disease only occurs in the lowveld buffalo 12 population (Fig. 7.1) of Mpumalanga and Limpopo Provinces. This highly 13 contagious “trade sensitive” disease is therefore controlled by law (Standing 14 Regulations of the Animal Diseases Act 35 of 1984) and was one of the major 15 reasons for the erection of the animal disease control fence on the western and 16 southern boundaries of Kruger by the Department of Agriculture between 1961 – 17 1963. At that time, the fence was constructed to be effective for containing cloven- 18 hooved ungulates, including buffalo. Elephants, at that time were at relatively low 19 density (population estimate around 3000), especially in the southern and central 20 districts, with minimal pressure on the fences. Structurally, the fences that were 21 erected were simple 1,8 metre fences consisting of 10 strands of barbed wire with 22 no electrification (that technology did not yet exist), and were found to be adequate 23 to prevent the movement of most ungulates. After the erection of these fences, the 24 number of outbreaks of FMD in neighbouring cattle fell progressively, and not a 25 single outbreak was detected in livestock adjacent to Kruger during the period 1983 26 – 1999 SECOND DRAFT Assessment of South African Elephant Management 7 1 2 Figure 7.1: A map indicating the foot and mouth disease control areas in South 3 Africa. Fence breakages in these areas have serious consequences for disease 4 control. 5 6 Wildlife-maintained diseases which are a danger to livestock, or could have public 7 health implications increase the importance of containing wildlife in areas that are 8 surrounded by domestic livestock. Even though elephants play no direct role in the 9 transmission of these diseases per se, fence breakages caused by elephant create 10 conduits that allow disease-carrying species to escape, resulting in a serious threat to 11 the health of the domestic stock. 12 13 It is therefore important to realise that since the 1995 moratorium on lethal elephant 14 population management in Kruger, the total elephant population has almost 15 doubled, and pressure on the Kruger fences has increased significantly. As an 16 indirect result, during the short period from 2000 to 2006, five outbreaks of foot and 17 mouth disease have occurred in cattle adjacent to Kruger, four of which could be 18 directly linked to buffalo escaping through fence breaks. This in spite of the fact that 19 the fences had been upgraded to 2,4 meter, 20 strand fence that was electrified at 5 20 levels. 21 SECOND DRAFT Assessment of South African Elephant Management 8 1 Some other examples of indigenous animal diseases that are carried by wildlife and 2 become problematic when fences are broken include:African swine fever (ASF) - 3 Warthogs are the mammalian carriers of African swine fever virus which causes a 4 fatal viral infection in domestic pigs. Warthogs also serve as transport hosts for the 5 vector - a tick. In addition to foot and mouth disease, buffalo are also carriers of 6 theileriosis (Corridor disease) which is transmitted by brown ear ticks and is a fatal 7 disease in cattle. Buffalo may also be carriers of brucellosis, which is an animal and 8 human health threat. Blue wildebeest are carriers of alcelaphine malignant catarrhal 9 fever virus, which causes a fatal viral infection in cattle. 10 11 12 13 Protection of livestock and crops 14 With increasing densities of elephants and depletion of natural foods in conservation 15 areas (Smith & Kasiki, 2000), especially during dry seasons, the pressure for 16 elephants to break out and look for more nutritious food sources increases 17 (Naughton-Treves, 1998, O’Connell-Rodwell et al, 2000). Most of these fence 18 breakage culprits are single bulls that are brazen and strong enough to break the 19 fence. Often conflicts with expanding human habitation displace elephants which in SECOND DRAFT Assessment of South African Elephant Management 9 1 turn come to depend on crop-raiding to survive in resource poor habitats (Tchamba, 2 1995). Cultivated crops are the perfect attractant for elephants, they are often highly 3 nutritious (grains), and/or taste good (fruits and vegetables). The result is that 4 elephants become crop raiders (Masilwa, 2001). 5 6 Fences do decrease the incidence of crop raiding. In Negande, crop raiding incidents 7 dropped by 65% after the erection of an elephant proof fence but this dropped to 8 42% the following season. A small circular fence erected around irrigated crops was 9 also successful in avoiding crop loss. However, in spite of agreeing to the project, 10 villagers were reluctant to maintain the fence after the first success. The economic 11 benefit of the erection of elephant proof fences is questionable and the main benefit 12 may be that fewer animals are killed because of causing damage (Taylor, 1994). 13 In addition, elephants breaks in the protected area fences create conduits for large 14 carnivores to exit protected areas and predate on livestock. 15 16 Thus farming of crops and livestock in areas which contain free ranging elephant 17 and lion results in increased human-wildlife conflict. In arid environments, 18 communal agricultural activity is concentrated along riparian zones which are also 19 favoured by elephant and has thus resulted in a higher impact on the sensitive 20 riparian zones and an increase in elephant damage. 21 22 It must also be appreciated that elephant habitat expansion corridors will also 23 predictably increase the human contact interface, and in most situations, such 24 corridors will need to be fenced. 25 26 In smaller protected areas that have elephants (e.g. Addo), more substantial and 27 robust fences are needed because the rate of contact of elephant with the fence 28 increases as the length of the fence decreases. These type of fences do not have to 29 be electrified to be effective if the animals are trained to respect the fence 30 (Anderson, 1994). Simple electric fences with only three strands and a voltage of 31 5.5 kV has been successful in controlling damage causing animals in Mwea district 32 Kenya, but there was very active community involvement and a full time fence 33 attendant, paid by an funding agency (Omondi et al., 2004) SECOND DRAFT Assessment of South African Elephant Management 10 1 2 3 Consequences of fence breakages 4 Over and above the negative consequences that elephant break outs may have due of 5 crop raiding, or creating conduits for large carnivores or disease-carrying wildlife to 6 exit protected areas (as discussed above), there are several other consequences that 7 need to be mentioned 8 9 Loss of animals 10 Despite the obvious reasons for not wanting elephants to escape, for example 11 maintaining good neighbourly relationships, there are other reasons why the fence 12 breaking of elephants can create problems. 13 14 High costs involved in returning the animals or to destroy them 15 Animals that escape from conservation areas either have to be returned to the 16 conservation area, or need to be destroyed where they are. In the case of elephants, 17 the costs can be high (Lubow, 1996). The capture and transport of elephant needs 18 specialised equipment, helicopters and vets experienced in elephant capture to be a 19 success (Nelson et al., 2003). If the elephants are close enough to the conservation 20 area, they could possibly be chased back (Hoare, 2001), but again, a helicopter is 21 most often necessary for this to succeed, although elephant have been chased back 22 from horseback in Etosha. Another option, and one which is often used, is to 23 destroy the animal/s (SANParks, 2005). It is much easier in the case of a single 24 elephant, but if a whole herd has escaped, the cost of removing the carcasses could 25 also mount up. Also, if a herd escapes from an area like KNP where high numbers 26 of elephants are present, the destruction of one herd is of no conservation 27 significance. However, if the only resident herd, at great cost happens to break out, 28 the owners of the elephants will be less keen to have the herd destroyed. 29 SECOND DRAFT Assessment of South African Elephant Management 11 1 Loss of scarce or expensive species 2 Several reserves which have elephant also have expensive, rare or endangered species 3 like rhino, roan, sable, tsessebe etc. These species have usually been introduced at great 4 costs to the reserves, and although these species would seldom cross fences by 5 themselves, they could escape through fences that were damaged by elephants. 6 Proof of ownership of escaped animals can become a problem and can result in 7 conflict and increasing tension between different landowners. 8 9 Domestic stock entering wildlife areas 10 Economic impacts 11 Domestic stock entering wildlife areas, especially when aimed at tourism, can have 12 a negative effect on the product on offer. Studies done in the Zambesi valley named 13 wild animal species roaming free, indigenous plant species and lack of people as 14 important factors in the perception of the tourist of an area to be wild. Pollution, 15 litter, vehicles, noise and the presence of domestic animals are factors that 16 negatively influence tourists’ perceptions of wilderness (Wynn, 2003). 17 18 Ecological effects of domestic stock in wildlife areas 19 Apart from the disease threat posed by domestic stock, it has been shown that 20 domestic animals such as goats can cause a significant los of indigenous plants and 21 can have a more severe effect on the vegetation than even elephant in the Eastern 22 Cape (Stuart-Hill, 1992). The presence of cattle inside Masai Mara has constrained 23 black rhino distribution and carrying capacity (Walpole, 2001). 24 25 26 Other conservation uses of fences 27 For protection of vegetation 28 Fences have been successfully used to exclude sensitive areas or representative 29 vegetation. In Addo, exclosure fencing has been used effectively to protect endemic 30 plants from utilization by elephant. Due to the value of these exclosures, this SECOND DRAFT Assessment of South African Elephant Management 12 1 approach will also be considered for the new properties that have been included. 2 (Johnson, 1998, Johnson et al., 1999). Thus a core of five botanical reserves was 3 identified within the Park which would represent 91% of the Park’s special plant 4 species in less than 8% of it is area (Lombard et al., 2001). Exclosures can thus be 5 used to protect sensitive areas or plants that are sensitive to elephant impact. Mature 6 plants within such exclosures could also then act as valuable seed banks to populate 7 surrounding areas (Western & Muitumo, 2004). 8 9 For understanding system function 10 Exclosures have been very useful to study the effect of browsers and grazers on 11 selected areas in Kruger.. This information is essential for management decisions 12 such as avoiding mistakenly controlling elephant populations to address impact 13 concerns that they are not responsible for. Differences especially in tree numbers 14 and height help to understand the possible effect of elephant on the vegetation. 15 (Trollope et al., 1998) 16 17 Exclosures are also useful to develop an understanding of the time needed for 18 different plant types to recover after heavy use by elephant and other browsers 19 (African Elephant Specialist Group Meeting, 1993). In Addo, such exclosures have 20 contributed sustancially in our understanding of how the thicket vegetation responds 21 to elephant use (Kerley & Landman, 2006) 22 23 For the protection of individual trees 24 Individual large trees can be also be individually protected from elephant use. In 25 East Africa and in the Associated Private Nature Reserves (APNR) on the western 26 boundary of Kruger, 13mm mesh wire netting was wrapped around the trunk of 27 mature tree stems and thus have prevented such trees from being extensively bark 28 stripped by elephants (Gordon, 2003; Henley & Henley, 2007) (Fig. 7.2). Heavy 29 wire netting was more efficient in protecting trees against debarking and required 30 less maintenance but was also more visible than 13mm mesh wire at distances 31 further than 5m from the protected tree. Wire netting techniques did not protect 32 trees from being uprooted or broken. Results from these studies indicate that the SECOND DRAFT Assessment of South African Elephant Management 13 1 absolute use or avoidance of protected trees may not be as important as the degree 2 to which the wire-netting prevents extensive bark-stripping and consequently 3 increases the survival rate of trees that are susceptible to bark-stripping by 4 elephants. 5 6 Figure 7.2:. Wire netting can be used to protect large trees from ring barking but 7 does not stop trees from being pushed over or broken. Photos from Mapungubwe 8 National Park 9 10 In Mapungubwe national park, individual trees have also been individually 11 protected, but no surveys have been done to determine the efficacy of this process. 12 13 For the protection of infrastructure 14 Sturdy and well maintained fences are extensively used for the protection of 15 property. Such fences are specifically designed to protect infrastructure such as 16 water tanks, pipelines, windmills, dams, weirs, and buildings from elephant. In 17 addition, tourist facilities and landing fields need barrier protection. 18 SECOND DRAFT Assessment of South African Elephant Management 14 1 In the Mwea region of Kenya, where there were 48 elephants in 1995 the situation 2 deteriorated to such an extent that some had to be translocated to Tsavo East 3 National Park to minimize conflicts. This did not solve the problem, and an electric 4 fence was erected. Before fence construction, most human deaths caused by wildlife 5 were attributed to elephants with an average of three people killed yearly. 6 Immediately after the fence was completed, incidents of human elephant conflict 7 decreased substantially and no elephant-related death has been reported since. 8 (Omondi, 2004) 9 10 11 Efficacy of fences 12 To contain elephant in a wildlife enterprise 13 Experience has shown that elephants are capable of going through the most 14 sophisticated barriers, including highly electrified fences when they are determined 15 to do so (Fig. 7.3). This is a real challenge as the long-term existence of especially 16 smaller wildlife areas will probably be dependant on the efficacy of such barriers to 17 prevent animals escaping and avoiding conflict with neighbouring communities. 18 Well maintained fencing, especially electric fencing, appears to be the most 19 effective barrier to restrict movement for most of the larger wildlife species (Nelson 20 et al., 2003). Elephants in particular are difficult to restrict as a result of their large 21 size and ease with which they can break fences, which make them the most 22 important fence breaking species (SANParks, 2005). Their home ranges are large 23 and migration and movement patterns often extend not only beyond park or reserve 24 boundaries, but national boundaries as well (Craig, 1997) (Chapter Two). 25 SECOND DRAFT Assessment of South African Elephant Management 15 1 2 Figure 7.3: Male elephant returning to KNP over border fence 3 4 Elephant most often cross fences because of the availability of water and food in 5 adjacent areas (Buss, 1961). Studies on crop raiding by elephants at Kibale Forest 6 National Park, Uganda showed that crop raiding occurred throughout the year with 7 peaks in dry seasons when crop availability was high. Bananas and maize were the 8 main crops raided. Monthly crop raiding incidences were not influenced by forage 9 quality but by ripening of maize. Crop availability seems to be a more important 10 driver of elephant breakages in forest habitats, whereas in savanna habitats large 11 seasonal fluctuations in forage quality have a greater influence on temporal patterns 12 of crop raiding. (Chiyo et al., 2005). (Osborn, 2004) also found that the point at 13 which the quality of the available forage declines below the quality of crop species 14 corresponded to the movement of bull elephants out of a protected area and into 15 fields. Fence breakages in the fence adjacent to the Limpopo province is illustrated 16 in Figure 7.4 and seem to also coincide with periods where forage may be scare in 17 the park. 18 SECOND DRAFT Assessment of South African Elephant Management 16 1 2 Figure 7.4: Reports of elephant breakages of fences between Kruger and the 3 Limpopo province from January 2001 to October 2004. 4 5 There seems to be a spatial and temporal correlation between elephant densities and 6 the number of fence breaks. The elephant population of Kruger has almost doubled 7 in the past 10 years. Using the incomplete reports available, (Anthony, 2006) 8 recorded 386 incidents of damage causing animals in the area between the 9 Shingwedzi and Klein Letaba rivers between October 1998 and October 2004 (Fig. 10 7.4). Elephant caused 56 of these incidents and 8 of the elephant had to be 11 destroyed. The most common problem animals were buffalo (137), lion (72), 12 elephant (55), hippopotamus (33) and crocodile (18). It is important to note that 13 many of the problem buffalo, lion and even hippos probably exited through elephant 14 fence breaks. 15 16 17 SECOND DRAFT Assessment of South African Elephant Management 17 1 2 3 Standard electric fences work well to protect smaller areas for experimental 4 purposes or to protect infrastructure. The maintenance of the fence is essential. 5 Breakages rarely occur and then only to get to water. For example the water source 6 is enclosed in the case of N’washitshumbe camp in Kruger, where roan antelope are 7 being bred and are protected against predators, while waterpoints outside the 8 enclosure have been closed for ecological reasons (Grant, 2002). Other breakages 9 that occurred into this camp was due to failure of the electric fencing. 10 11 In disease control 12 Between 1983 and 1999, the elephant density in Kruger was relatively low (about 13 0,4 elephant /km2) and in that period no outbreaks of foot and mouth disease were 14 detected in livestock adjoining the park. In addition, it is important to remember that 15 elephant numbers were stabilised (at approximately 7000) by lethal population 16 management during the period between 1969 and 1994, and fence-breaking bulls 17 and problem peripheral herds were frequently targeted as part of problem animal 18 and border control management. Therefore during this period, elephant fence 19 breaking activities were sporadic and rapidly dealt with. 20 21 However with the increasing elephant density (0,46 – 0,62 elephant /km2 between 22 2000 and 2006, five major foot and mouth outbreaks were occurred in these 23 livestock populations. Four of these outbreaks ( Bushbuck Ridge -2001, Masisi - 24 2003, Mopani 2004 and Thulamela 2006) could be linked directly to buffalo exiting 25 Kruger through fence breaks. 26 SECOND DRAFT Assessment of South African Elephant Management 18 1 The Bushbuck Ridge outbreak cost the tax payer R20 million, the Masisi outbreak 2 cost R4 million, and the Mopani outbreak cost a massive R90 million to control. 3 Mass vaccination in and around the outbreak as well as road blocks and the erection 4 of additional cordons and barrier fences were necessary to avoid the further spread 5 of the disease. Further costs of such outbreaks include indirect costs to farmers due 6 to movement restrictions on agricultural products. Additional financial losses would 7 have been incurred if the outbreak was not contained within the declared foot and 8 mouth control area, as a result of trade barriers and millions of rands lost in export 9 earnings. 10 11 There has been a striking spatial and temporal correlation between the number of 12 elephant fence breaks and the number of vagrant buffalo incidents (State 13 Veterinarian – Skukuza, quarterly reports 2005 – 2007). There is also a striking 14 temporal correlation between number of fence breaks and elephant densities. ). In 15 the winter of 2005, up to 35 elephant fence breaks were recorded per day in the 12 16 km section of fence stretching from Sawutini to Naladzi. (State Vet – Second 17 quarterly report 2005). These elephants were breaking out to drink and bath in one 18 of the few remaining pools in the klein Letaba river. 19 20 Corridor disease (theileriosis), with close to 100% mortality of infected cattle, was 21 also sporadically reported in areas where buffalo that have crossed fences broken by 22 elephant and had dropped infected ticks. A less frequent occurrence, is the outbreak 23 of wildebeest associated malignant catarrhal fever, an inherently fatal herpes virus 24 infection in cattle which occurs where cattle and wildebeest are able to mix after 25 wildebeest moved out through broken fences. 26 27 Outbreaks of African swine fever are generally related to warthogs exiting the Park 28 and making contact with domestic pigs. Warthogs do not necessary go through 29 elephant breaks and frequently creep under fences that are slightly elevated due to 30 wind or water erosion. Once there has been direct contact between domestic and 31 wild pigs and the disease has been transmitted to the domestic pigs, a horizontal pig 32 to pig contagious cycle developes with large scale deaths. To give an idea of the 33 potential scale of African swine fever outbreaks, one that was well documented in 34 southern Mozambique in 1997 resulted in the deaths of an estimated 180,000 pigs. SECOND DRAFT Assessment of South African Elephant Management 19 1 2 However, it must be emphasised that the more sporadic outbreaks of wildlife 3 diseases in livestock are totally under-reported, because local communities 4 frequently consume the carcasses, and no diagnosis can be made. 5 6 7 Consequences of restriction of movement by fences 8 In the African context restriction of elephant movement is generally a result of 9 human encroachment or habitat change (Hoare & Du Toit, 1999). In South Africa, 10 movement is mostly restricted by fencing which has been erected with the express 11 intent of restricting the animals to a certain area. Contrary to the situation in open 12 landscapes where animals are not restricted and can select from all available 13 resources and habitats, fences restrict direct access to other resources. Some of these 14 may be key resources such as water e.g. elephant in Tembe Elephant Park which no 15 longer have access to the Pongola River only four kilometres away. Apart from the 16 fact that these restrictions may have a significant effect on the elephant population 17 dynamics (Illius and O’Connor, 2000) the ecology of the animals may be affected 18 (van Aarde & Jackson, 2007). The relative importance of the different resources 19 change with climatic and seasonal changes and the long term effect of fencing in 20 this regards is not well understood and requires further targeted research 21 (OwenSmith et al., 2006). 22 23 In recent literature the increase in density and what is commonly called 24 “overabundance” of elephant has been attributed to fences restricting elephants to 25 confined areas (Gillson & Lindsay, 2003; van Aarde & Jackson, 2007). By 26 preventing the spatial flux in density they argue, the natural regulators of elephant 27 populations are reduced if not lost and this results in excessive impact and 28 homogenisation of the local biodiversity, particularly the vegetation (Owen-Smith et 29 al., 2006). Mechanisms of this are not well understood, but may be linked to 30 elephants being very adaptable in their ability to eat poor quality food (Owen-Smith, 31 1988). Thus even when confronted with a forage limited by choice in quality and 32 quantity, they can continue to be productive. However there seems to be general 33 agreement that fencing elephant into smaller areas will have a greater negative SECOND DRAFT Assessment of South African Elephant Management 20 1 effect on the natural system variability, possibly because larger areas have a wider 2 range of different habitats than smaller areas (Owen-Smith et al., 2006). Another 3 argument is that the range of habitats which elephants normally have access to, 4 include areas that serve as sink (sensu Dias 1996) and that by preventing animals to 5 move into these areas, one is immediately exposing the contained area to the effects 6 of the enhanced population growth (van Aarde & Jackson, 2007) that can lead to a 7 loss in habitat variability. 8 9 Furthermore, a massive build-up of elephant numbers in smaller fenced off areas is 10 often followed by a decline in woodland density due to a combination of tree 11 destruction by elephants and the effects of fire (Laws, Parker & Johnstone, 1975; 12 Caughley, 1976; Barnes, 1980 & Lewis, 1982; Owen Smith, 1983). In a number of 13 Parks this has led to the disappearance of large areas of Acacia and Commiphora 14 woodland and local extinction of tree species like Baobab (Adansoni digitota), 15 which is highly favoured by elephant. The elephant were not capable of responding 16 to these self-inflicted radical changes in the food supply because of the fences 17 (Jachmann & Bell, 1984). 18 19 Prior to the erection of a veterinary fence western boundary in the 1960s there was 20 an east-west seasonal migration of herbivores from the western part of what is now 21 the Kruger National Park towards the Drakensberg mountains (Fig. 7.5). With the 22 initial erection of the fence, many animals were killed such as giraffe, wildebeest 23 zebra, and kudu (Whyte & Joubert, 1988; Albertson, 1998). In Botswana the disease 24 control veterinary fences also prevented vital wildlife movements, fragmented 25 populations, separated young animals from herds and caused death of animals that 26 get stuck in the fence (Albertson, 1998). Fences do not only affect the migration 27 routes of animals between resource areas, but also affect the natural drivers such as 28 fire. Before the erection of the fence, fire was probably important in maintaining the 29 wooded grassland savanna, with a continuous grass cover interspersed with mature 30 trees and shrubs (Frost et al., 1986; Trollope, 1992). Increased grazing pressure due 31 to the confinement of animals, led to reduction in the frequency of hot fires and this 32 often precipitates bush thickening (Peel, 2005). Wildlife-based tourist operations in 33 the region are adversely affected by such bush encroachment because the dense 34 woody layer reduces game visibility. SECOND DRAFT Assessment of South African Elephant Management 21 1 2 Figure 7.5: Animal migration routes prior to the erection of the foot and mouth 3 fence (Whyte, 1985) 4 5 The erection of the veterinary fence between the Kruger and private land to the west 6 necessitated the provision of water in previously seasonally waterless areas. Water 7 shortages in such a confined area with inadequate surface-water may increase fence 8 breakages, conflict with humans (especially around water sources) and risk of 9 disease spread. Artificially provided water sources will counter this effect, but alter 10 the spatial and temporal foraging and trampling patterns of both elephants and other 11 water-dependent animals (Chamaille-Jammes et al., 2007a; Smit et al., 2007a). This 12 may ultimately influence the vegetation (e.g. Thrash, 2000; Brits et al., 2002), soil 13 (e.g. Thrash, 1997) and nutrient patterns (e.g. Tolsma et al., 1987; Nsinamwa et al., 14 2005) on multiple scales (multiple piosphere effect). Additional permanent water 15 sources has also been blamed for influencing predator/prey relationships 16 (Harrington et al., 1999; McLoughlin and Owen-Smith, 2003; Mills and Funston, 17 2003), creating unnaturally high herbivore numbers with consequent population 18 crashes during droughts (Walker et al., 1987), compromising system resilience 19 (Grant et al., 2002) and degrading the quality of the herbaceous layer (Parker and SECOND DRAFT Assessment of South African Elephant Management 22 1 Witkowski, 1999). The effects of fencing and water provision can also be seen in 2 the change in the status of impala which did not occur west of 31030’E in the 1800’s 3 (Kirby, 1896) and were in fact not found west of the Orpen Gate until the 1920’s 4 (Porter ,1970) and are now the most prolific herbivore in the Lowveld. Both 5 elephant and impala are strong competitors, have a great impact on areas they 6 inhabit and are ultimately able to change the habitat to suit their requirements, 7 switching easily from their preferred grazing to browsing. Weak competitors such 8 as roan, sable and tsessebe cannot compete with species such as elephant and impala 9 and are only successful within intensive breeding camps such as found at Selati 10 Game Reserve. (Table 7.1). 11 12 Table 7. 1: Case study illustrating the ecological and economic effect of fencing 13 and water provision on the ecology of areas of small size. 14 15 16 One further consequence of fencing is that depending on the timing of the fence 17 erection, it may split a population of elephant (e.g. in the case of the Tembe 18 Elephant population split between South Africa (Tembe Elephant Park) and 19 Mocambique (Maputo Special Elephant Reserve). This may lead to genetic 20 fragmentation, although for a species with such a long a generation time it is 21 unlikely to have an effect in a short period. 22 23 Lastly fences separate local communities from resources such as water and 24 medicinal plants and this can have important implications for both the plants and the 25 humans (Chapter Four). 26 27 SECOND DRAFT Assessment of South African Elephant Management 23 1 Fences and elephant welfare. 2 3 Fences have become necessary in wildlife management, in particular, in the 4 comparatively densely settled South Africa (Damm, 2002). In the case of elephants, 5 this largely means to prevent the animals from being able to use the landscape to the 6 full potential of its food and water resources and thus importantly to curb the ability 7 to disperse across the landscape. This behavioural impact can be on either daily 8 patterns of resource use, seasonal range expansions or contractions, or the ability for 9 recruitment dispersal. Current methods for conserving both wild and captive 10 elephant populations fail to preserve elephant social systems. Elephant aggression is 11 not an isolated event and in heavily impacted areas, intraspecific mortality among 12 male elephants may account for up to 90% of all male deaths, compared with 6% in 13 relatively unstressed communities. Elsewhere, including Asia, there are reports of 14 poor mothering skills, infant rejection, increased ‘problem animals’, and elevated 15 stress-hormone levels (Bradshaw et al., 2005). 16 17 From a welfare perspective, an elephant population can be kept in and could be 18 satisfied in an enclosed area as small as 150 km2 as indicated by the fact that they 19 did not immediately increase their ranges when boundary fences were removed 20 (Druce et al., 2007). The problem is however, fundamentally based on limited space 21 and human population numbers are as much cause for concern as that of the 22 elephant population. 23 24 Elephant have a complex social environment that stretches over four different tiers 25 from mother- daughter relationships to very loose incidental group relationships 26 (Wittemeyr et al., 2005). The extended herd structure is important for their well- 27 being, and without it elephants are being deprived of a great amount of sociality and 28 socialization opportunities as well as behavioural repertoires. To achieve such a 29 social environment, small groups should be eliminated and larger populations and 30 corridors promoted. In principle, the notion of dividing the land surface into distinct 31 parcels has become so familiar to us that the understanding of the physical elements 32 often overpowers the realization of the abstract elements. The abstract elements are 33 that we design the way to behave regarding the land parcels, but unfortunately this 34 behaviour is not bought into or indeed understood by the natural biota. SECOND DRAFT Assessment of South African Elephant Management 24 1 2 As one follows actual case studies of the behaviour of bulls and matriarch cows in 3 the natural biota, it becomes evident that their aggressive behaviour towards man 4 diminishes the less aggressive man is towards the elephant. For this reason it is 5 evident that tourism, a passive utilisation of the habitats supporting elephant, cannot 6 under any circumstances, be shared with the aggressive utilisation of the same 7 habitat through hunting of elephants. Ethical visitor viewing and the approach 8 towards the elephant comfort zones is also an essential factor to reduce human / 9 elephant conflict. 10 11 Work done by Space For Elephant Foundation indicates that escapes from 12 conservation areas coincide with the rainy seasons when cloud formations are 13 consistently low, allowing easier communication and when there is an abundance of 14 vegetation and Marula berries. Elephant attempts to escape seem to be due to 15 confrontational stress and they seek to return to the area from whence they came. 16 Elephant face excessive trauma when the group is captured, with translocation and 17 release into a foreign area (Space For Elephants Foundation See website 18 www.space4elephants.org). 19 20 21 Technical specifications for fences and their maintenance 22 Given the present state of technology, well-applied electric fences can act as a 23 powerful deterrent to elephant entry and trespass (Hoare, 1992). However, an often 24 neglected aspect of electric fences is their cost-effectiveness, and to this end, 25 economic cost benefit analyses are essential prerequisites. 26 A typical electrified game fence is illustrated in Figure 7.6. SECOND DRAFT Assessment of South African Elephant Management 25 1 2 Figure 7.6: Diagram of electric wires for elephant proof fence with an example of 3 such a fence in Mapungubwe National Park. 4 5 6 The different types of fences and their efficacy is summarised in Table 7.2. SECOND DRAFT Assessment of South African Elephant Management 26 1 Table 7.2: Specifications, erection and maintenance costs for different types of 2 fences. All electrified strands must have a minimum voltage of 6000 V and must be 3 have sufficient energisers to supply power to maintain this voltage over a distance 4 of 8 km. 5 6 SECOND DRAFT Assessment of South African Elephant Management 27 1 Ensuring efficiency of fences 2 Long-term success using fences to contain elephant is dependent on meticulous 3 routine maintenance and the use of solid, durable material that is well anchored.. 4 Electric fencing technology is simple and definitely deters elephants, but has to be 5 kept up continuously to be efficient (Hoare, 2003). Added to that the damage that 6 elephants do and the direct costs involved in fixing and/or replacing fences that 7 were destroyed by elephant, and the labour costs makes fencing very expensive as a 8 management tool (WWF 1998, Hoare 1995). 9 10 The following points need to be adhered to, to ensure that fences keep elephant 11 in/out: 12 13 1. Fences should be patrolled every day on a rotational system to effect fence repairs. 14 2. Have sufficient trained staff and transport available. 15 3. Define responsibilities for maintenance and costs associated clearly and 16 budget appropriately. 17 4. Have agreements with neighbouring communities about the importance of 18 the fences for them to avoid parts of the fence being removed for their 19 private use. 20 21 22 23 5. Avoid human interference by using specific structures such as cable instead of wire. 6. Insure that there is a reliable electric supply thus suffcient electric power points to deliver the required current. 24 7. Remove vegetation around fences to avoid shorts in the electric current be 25 physically clearing the area or controlled by judicious use of herbicides 26 annually. 27 8. Check fences after fires as they may damage fences, after flash floods which 28 may cause shorts in electrical fences crossing drainage lines, and after heavy 29 lightning. 30 31 9. Ensure that gates at river crossings providing water to neighbours are securely closed. SECOND DRAFT Assessment of South African Elephant Management 28 1 10. Consider re-opening strategic boreholes during the dry season to reduce 2 fence breaks in areas where elephant movements are associated with 3 accessibility to water. 4 5 6 Alternative methods to change elephant distribution 7 Surface water manipulation as elephant management tool 8 Elephant distribution is often associated with the distribution of surface-water and 9 rivers (Stokke & du Toit, 2002; Redfern et al., 2003; Chamaillé-Jammes et al., 10 2007; Smit et al., 2007 a&b). It has been shown that the addition of surface water to 11 areas with limited natural water availability can increase the density of elephants 12 (Cumming, 1981) and expand their spatial distribution (Chamaillé-Jammes et al., 13 2007). Surface water manipulation (e.g. closing down and/or moving artificial 14 sources) has therefore been proposed as a “non-intrusive and natural” management 15 tool with which elephant density and distribution patterns can be restricted (Owen- 16 Smith, 1996; Gillson & Lindsay, 2003; Chaimaillé-Jammes et al., 2007; Chaimaillé- 17 Jammes et al., in press). However, considering the mobility of elephants (e.g. 18 Viljoen & Bothma, 1990; Verlinden & Gavor, 1998), it is arguable how effective 19 surface water manipulation will be as an elephant management tool in areas like the 20 Kruger National Park where water is usually widely available (South African 21 National Parks, 2005; Redfern et al., 2005; Owen-Smit et al., 2006; Smit et al., in 22 press). This argument is supported by the fact that the elephant density did not 23 respond to the closure of a considerable number of artificial waterholes in the park 24 (Fig. 7.7). However, the closure of many artificial waterholes might have decreased 25 the ultimate density at which water or forage resources within walking distance 26 from water would start to regulate the elephant population effectively through 27 density-dependent mechanisms. It is anticipated that surface-water manipulation 28 may only have a significant influence on Kruger’s elephant density if the population 29 is allowed to grow further, if more waterholes are closed and/or if a very serious 30 drought ensues. It is unsure, however, at what elephant density, and under what 31 drought conditions, water would become a limiting resource for Kruger’s elephants 32 (Smit et al., in press). Smaller water-dependent species that cannot travel as far as SECOND DRAFT Assessment of South African Elephant Management 29 1 elephants and that have more specific habitat and forage requirements, may be more 2 affected by changes in artificial water provision in their habitats than elephants 3 which are highly mobile and adaptable (Redfern et al., 2005; Smit et al., in press). 4 Furthermore, by digging for water in superficially dry riverbeds (Dudley et al., 5 2001) or by monopolising the last remaining water sources (Joubert, 2006), 6 elephants sometimes have access to water that other species cannot utilise. 7 8 9 10 Figure 7.7: Distribution and density patterns of elephant in Kruger. Note the concentration along the drainage lines (Courtesy of Sandra MacFayden) 11 12 Artificial provision of water may also influence the area of elephant impact 13 (Chamaillé-Jammes et al., 2007). For example, in the Addo Elephant National Park 14 the impact on the endemic subtropical thicket has been very extensive around the 15 artificial waterholes while areas far from the waterpoints have been substantially 16 less used (Knight et al., 2002). Other studies have also indicated that the use of 17 elephants of vegetation is higher in closer vicinity to water (e.g. Ben-Shahar, 1983, 18 Nelleman et al., 2002). Therefore, if water is artificially provided, it should 19 preferably be restricted to natural supply areas, minimising spatial alterations to SECOND DRAFT Assessment of South African Elephant Management 30 1 grazing patterns (Pienaar et al., 1997). Water provision should enhance that natural 2 variability of the landscape, and should allow for the development of the full 3 elephant density range (from high density impact areas to low density “refugia”). A 4 uniform distribution of water by the addition of artificial water sources will suppress 5 the natural variability brought about by natural water availability. This is not 6 desirable for biodiversity conservation (Owen-Smith, 1996; Knight et al., 2002). 7 However, depending on the availability of natural water sources, artificial 8 waterholes may not influence large-scale elephant distribution patterns as much as 9 the local-scale activity patterns (i.e. piosphere effect). For example, it has been 10 found that the landscape-scale dry season distribution of elephants in Kruger is 11 more closely linked to the river system than to the artificial waterhole network (Smit 12 et al., 2007a&b) (Fig. 7.8) Considering this, together with the mobility of elephants 13 to move between the (usually wide-spread) ephemeral and permanent water sources, 14 it is arguable how effective the density and distribution patterns of the elephants 15 could be manipulated under normal conditions by means of water provision in 16 Kruger (Redfern et al., 2005; Smit et al., in press). 17 18 19 Figure 7.8: Changes in elephant numbers in Kruger with the change in management 20 policies: waterpoint closure and stopping of culling. 21 SECOND DRAFT Assessment of South African Elephant Management 31 1 It must also be taken into consideration that the closure of water points close to the 2 perimeter fence of a protected area may increase the frequency of elephants 3 breaking the fence to access water outside. 4 5 Therefore, although water provision can influence elephant density and distribution 6 in fenced areas, the effectiveness of surface-water manipulation as a management 7 tool will depend, inter alia, on (1) natural surface-water availability, (2) forage 8 quality, (3 local densities, (4) position of a population on its growth trajectory, (5) 9 size of the confined area, and ultimately (6) the elephant management objectives 10 (e.g. are objectives defined by density-dependence or impact?) (Smit et al., in 11 press). Surface water manipulation will be most effective as a management tool in 12 large systems with very limited natural water distribution. In such systems both the 13 density and the distribution patterns may potentially be influenced by water 14 provision. However, the effect of water limitation on elephant populations may be 15 limited as extreme climatic events, such as those associated with the El Niño - 16 Southern-Oscillation (ENSO), seem to have a very pronounced effect on elephant 17 reproduction. The ecological variability of these climatic events leads to the 18 characteristic fluctuations in elephant populations, rather than the mortality-driven 19 fluctuations which are commonly seen in many ungulate populations. (Wittemeyr et 20 al., 2007). In small enclosed areas with adequate natural water, artificial water 21 provision can also be expected to have a relatively small and localised effect. Even 22 though artificial water provision may be essential for survival in a small system with 23 no reliable water source, any water provided will effectively be within walking 24 distance for elephants. 25 26 Finally, although water provision may be considered a potential management tool in 27 certain systems, managers should guard against employing surface-water 28 manipulation to achieve elephant-specific management objectives, when water 29 manipulation may in fact have more pronounced effects on other components of 30 biodiversity (i.e. a single species- versus systems- approach). 31 SECOND DRAFT Assessment of South African Elephant Management 32 1 Disturbance as a managing tool 2 Disturbance methods may be used to deter elephant, but elephants soon become 3 habituated (Bell, 1984; De Boer & Baquete, 1998; Hoare, 1995; O'Connell-Rodwell 4 et al., 2000; Osborn & Rasmussen, 1995; Thouless, 1994), especially if the same 5 animals are regularly involved (Hoare, 1999a). These methods require trained 6 personnel and they can be dangerous because of proximity to the elephants. 7 However, they are generally cheap to apply, they have been shown to have at least 8 some effect, they are non-fatal for the elephants and the involvement of the 9 authorities provides some public relations value (Nelson et al., 2007) 10 11 Villagers in Sumatra use powerful flashlights to deter elephants in combination with 12 noise and fire (Nyhus et al., 2000). Thunder-flashes and flares have been used in 13 Zimbabwe with initial success (Hoare, 2001a) 14 15 Firing weapons over the heads of crop-raiding elephants to chase them from fields 16 has been used in Zimbabwe (Hoare, 2001a), and Niassa Reserve in Mozambique 17 (Macadona pers.comm.). In Niassa, it is used successfully in combination with 18 electric fences. 19 20 O’Connell-Rodwell et al. (2000) experimented with trip alarms in villages (n=10) 21 made up of individual farms (n=44) in East Caprivi, Namibia. They found shorter 22 wires around individual farms to be effective in the short-term, but there was no 23 impact on the overall number of conflict incidents reported in a year as elephants 24 initially moved into neighbouring farms before becoming habituated. Each alarm 25 cost US$78, less than the average elephant crop-damage claim, while from 1993- 26 1995 an estimated US$1 800 was saved. 27 28 Massive disturbance (e.g. people, vehicles and/or helicopters) to drive elephants 29 away from a conflict area has been tried with some immediate, although short-term, 30 success in Zimbabwe (Hoare, 2001a). 31 32 SECOND DRAFT Assessment of South African Elephant Management 33 1 Changing behaviour as a management tool 2 While the needs for nutrients and a place to live are inherent, which foods animals 3 eat and which places they live are learned behaviours involving ongoing interactions 4 among genes, social, and biophysical environments over generations (Provenza et 5 al., 2003, Provenza and Villalba, 2006; Davis and Stamps, 2004).. Behaviour-based 6 management encourages humans to actively participate in management planning to 7 enhance ecosystem health, animal well being, and enterprise sustainability 8 (Provenza, 2003). This view of animal behaviour, and approach to managing 9 ecosystems, diverges from contemporary notions in ecology (Provenza, 2007). 10 11 Once understood, behavioural principles and processes can be transformed into 12 practices that provide an array of solutions to challenges people face in attempting 13 to manage landscapes for the well being of the many species of plants and animals 14 that depend upon them. The issue isn't if creatures are adapting to ongoing changes 15 in social and biophysical environments, they do so every day of their lives. The only 16 question is whether or not people want to participate in the process. 17 18 All animals behave based on the consequences: positive consequences increase and 19 negative consequences decrease the likelihood of behaviours reoccurring. 20 Consequences involve two general behavioural systems in animals -- skin-defense 21 systems evolved under the threat of predation and gut-defence systems evolved 22 under the threat of toxins in foods (Garcia et al., 1985). These two systems form the 23 basis for changing food and habitat selection behaviours in animals. Changing 24 food/habitat selection behaviours requires making the food/habitat an animal is 25 currently using less desirable (stick) relative to other foods/habitats (carrots). 26 27 Strategic Hunting: As an example of this approach: Elk are hunted in locations 28 where they are not wanted, such as the former feeding areas, and they are not hunted 29 in areas where they can stay. Hunting can have significant and lasting impacts on 30 the movement and distribution of game animals (Conner, 2002; Viera et al., 2003). 31 For instance, prior to 1986, both bull and cow elk at migrated to lower elevations on 32 the eastern portion of a ranch in Utah USA. In mid October in 1986, 100 hunters 33 were allowed access to the ranch to hunt cow elk; they harvested 86 cows in one SECOND DRAFT Assessment of South African Elephant Management 34 1 morning. For the past 20 years since that date cow elk have not migrated to lower 2 elevations until snow pushes them down later in November or December. Bull elk, 3 which have not been hunted in the lower elevations of the ranch, have continued to 4 migrate to lower elevations mid October. One of the most striking examples of this 5 involves a population of moose in central Norway that migrates from low-lying 6 summer areas to high-elevation winter areas, contrary to the general pattern of 7 migration (Andersen, 1991). Archaeological evidence shows their migratory 8 behaviour follows a traditional pattern unchanged since 5000 B.P. despite 9 deterioration in the quality of their winter range. Incongruously, there are no 10 physical barriers preventing the moose using better habitat. Rather, the barriers are 11 cultural, and they began 5000 B.P. when humans hunted (pit trapped) the moose. 12 Humans no longer pit trap the moose and the behaviours are held in place by 13 “culture”. 14 15 In making such major changes in management, a minimum of 3 years typically are 16 required to change the behaviours of long-lived social animals. The first year is the 17 most difficult, as none of the adults have any experience with the new system. The 18 second year is better because all those involved have a year of experience with the 19 new system and the animals that were unable to adjust to the new system have been 20 weaned. By the third year, all of the adults have two years of experience with the 21 new system and young animals born into the new system are becoming members of 22 the herd. In behavior-based management, people become agents of change over time 23 in animal cultures. Social organization leads to culture, the knowledge and habits 24 acquired by ancestors and passed from one generation to the next about how to 25 survive in an environment (De Waal, 2001). A culture develops when learned 26 practices contribute to the group’s success in solving problems. Cultures evolve as 27 individuals in groups discover new ways of behaving as with finding new foods or 28 habitats and better ways to use foods and habitats (Skinner, 1981). 29 30 Similarly, extended families with matriarchal leadership may provide a means for 31 changing elephant behaviour. Efforts could be focused on individual families, and 32 given the importance of the matriarch in behaviour of the family specific efforts 33 might be directed at the matriarch of each family. It may be best to test how to train 34 elephants using a variety of techniques with a small number of families. Long-term SECOND DRAFT Assessment of South African Elephant Management 35 1 mother-daughter associations should lead to the learning behaviour being 2 transferred thus limiting the time needed to train the animals to avoid certain areas 3 (Douglas-Hamilton, 1973, Moss & Poole, 1983). 4 5 Repellents 6 The use of chilli extracts has shown particular promise not only because Capsicum 7 based products are non-toxic and environmentally friendly, but specifically because 8 elephant’s advanced olfactory and memory capabilities make them suitable for 9 adverse conditioning (Osborn & Rasmussen, 1995, Osborn, 1997). Numerous 10 evaluations with chilli extracts have been completed – particularly in Zimbabwe 11 where the objective was to protect crops belonging to rural populations that adjoin 12 nature reserves or where elephants have caused extensive damage to crops (Osborn 13 & Parker, 2002, Osborn & Parker, 2003). These evaluations have been mainly 14 directed at a practical and cost effective means of applying Capsicum oleoresin in 15 different forms like sprays and treated ropes which are strung around crops. 16 Research has shown the effectiveness of chilli extracts as a spray, when 17 administered upwind of elephants and compared to traditional methods of trying to 18 deter elephants during crop raiding. When traditional measures are utilised, there is 19 normally an aggressive reaction from elephants, whereas in the case of aerial 20 spraying of Capsicum oleoresin, the response by the elephants was more rapid and 21 resulted in prompt withdrawal from the crops without aggression (Osborn, 2002). 22 Capsicum oleoresin has thus far functioned as a viable short term elephant repellent 23 in some areas but does not provide a long term solution to the problem of increasing 24 elephant numbers (Cumming & Jones, 2005). 25 26 Other ways to protect crops or particular specimens of vulnerable trees include the 27 placement of bee hives in strategic trees as elephants are sensitive to the sound and 28 sting of bees (Karidozo & Osborn 2005, Vollrath & Douglas-Hamilton, 2005a & 29 2005b). Using bees as a selective repellent offer the added benefit that as a 30 deterrent, bees could pay for themselves through the sale of honey (Vollrath & 31 Douglas-Hamilton, 2005a). 32 SECOND DRAFT Assessment of South African Elephant Management 36 1 Buffer Crops 2 Unpalatable crops such as tea, spiny plants such as sisal, timber plantations, and 3 Opuntia barriers have all been tried but none have deterred elephant (Hoare, 2003). 4 The cactus species Opuntia dillenii was used as a barrier in some parts of Laikipia 5 and Narok, Kenya. Its potential to spread as a weed, however, is a major limitation. 6 Another species, Mauritius thorn (Caesalpinia decapetala), has also been tried in 7 Transmara, albeit with little success (Omondi et al., 2004) 8 9 Moats and ditches 10 Ditches and moats have been tried in the past in Laikipia, Mt Kenya and Aberdares. 11 However, due to lack of proper maintenance, they have not been successful in 12 containing the elephants in protected areas. This method may be ideal only for 13 small-scale sites of 3 or 4 km and is not recommended for high rainfall areas as they 14 may cause considerable soil erosion (Omondi et al., 2004) 15 16 Stone walls 17 This method can only be considered where stones are available on site and the size 18 of the area to be fenced is not extensive. Stone wall are not effective for containing 19 elephants, as elephant soon learn to remove the rocks (Omondi et al., 2004). 20 21 22 Effects of fence removal or the lack of fencing 23 Elephant can disperse at rates of 7 – 10 km per year after the removal of a fence. 24 Hence the 20 000 km2 of Kruger was colonised within 50 years due to migration 25 from Mozambique and the establishment of breeding herds in the Kruger National 26 Park (Porter, 1970) after starting off with very few elephant in the early 1900s 27 (Kirby, 1896). Subsequently their numbers increased). Ten years after elephant were 28 recorded in the Serengeti National Park for the first time after an absence of 40 29 years, the population increased to 2000 individuals (van Aarde & Jackson, 2007). 30 SECOND DRAFT Assessment of South African Elephant Management 37 1 Elephant move slowly into new, unexplored areas as can be seen by the slow 2 increase in the elephant population in the private reserves next to Kruger after 3 removal of part of the western boundary fence in 1993. (Peel & Grant, Chapter 4 Eight (Fig.7.9 ). 5 6 Figure 7.9: The was a slow increase of elephant in Sabie Sand Wildtuin with the 7 removal of the fence between Kruger and the private reserves. Elephant don’t move 8 in numbers to new areas. 9 10 The most recent addition to the Associated Private Nature reserves is the Balule 11 Nature Reserve which was a low elephant density area. Numbers in this area have 12 increased from zero in the 1990’s to almost 500 in 2006 (Peel, 2006). Even though 13 it may still be too early to note the re-establishment of migration paths after the 14 removal of the fence between Kruger and Sabie Sand Wildtuin it does appear that 15 there is some seasonal movement in and out of areas such as the Sabi Sand (15 16 years). Satellite collared animals are followed over time and movement between 17 Kruger and Sabi Sand is already apparent in certain groups in both summer and 18 winter (Fig. 7.10). SECOND DRAFT Assessment of South African Elephant Management 38 1 Figure 7.10: Season movement of three elephant families between Sabie Sand 2 Wildtuin and Kruger. 3 4 During August 2004, the boundary fences between Phinda Private Game Reserve 5 and two neighbouring reserves were removed. Initially family groups only moved 6 into the new area at night and spent minimal time there, while older bulls spent 7 longer periods of time, regardless of time of day. One year after the fence removal, 8 most of the elephants had only expanded their home ranges slightly into the new 9 area (Druce, Pretorius, & Slotow, 2007). Similarly, elephants that were introduced 10 into Marakele National Park in 1996 took few years to move to the adjacent 11 Marakele Pty limited after the fence was removed in 2001 (Bezuidenhout, 2004). 12 13 SECOND DRAFT Assessment of South African Elephant Management 39 1 Legal obligations for fencing 2 In any area where wildlife may be carriers of foot and mouth disease, the animals 3 diseases act (Act 35 of 1984) requires that the animals are separated from domestic 4 stock. 5 6 Any damage causing animal that can be clearly identified by marking, collars, 7 branding, microchip etc must be monitored and cases of damage need to be 8 investigated thoroughly using these identification techniques as proof. 9 10 The quality of fences for wildlife are legally stipulated and are defined for each type 11 of animal to be contained. 12 Please see Chapter Eleven for details. 13 14 15 Legal consequences of fencing 16 See Chapter Eleven 17 18 19 Legal obligation of owners of fenced properties. 20 See Chapter Eleven 21 22 23 Conclusion 24 Fences are probably the most efficient barriers to restrict elephant movement. 25 Electric fences can work very well if they are maintained at all times. These fences 26 have to be sturdy and durable as elephant will tend to re-cross a fence again once 27 they have been previously successful. Fences are more efficient when the animals 28 are trained in a boma to respect them. 29 30 Other barriers can be of some use, and may be cheaper then fencing, but 31 maintenance is also essential. SECOND DRAFT Assessment of South African Elephant Management 40 1 2 “Teaching” animals to avoid certain areas is worth investigating in our opinion as 3 barriers can not always be used to restrict elephant movement. Disturbance in the 4 form of noise or even local culling/ hunting could be a tool to teach the animals to 5 avoid certain areas. If this could be done successfully it may be possible to protect 6 sensitive areas at least to a certain extent without fencing or other barriers. 7 8 9 10 Research Gaps. Examine the use of learnt behaviour to control elephant movement. 11 12 Understanding the factors that determine distribution and density dependence of 13 elephant in enclosed areas. 14 15 Examine effective approaches to monitoring fences to establish the efficacy of the 16 fences and to identify the individuals that break the fences. 17 18 19 References 20 African Elephant Specialist Group Meeting. 1993. "Working Group Discussion 21 Three Elephant - Habitat Working Group. Pachyderm 17, 9-16. 22 Albertson, A. 1998. Northen Botswana Veterinary Fences: Critical Ecological 23 24 25 Impacts Unpublished manuscript. Andersen, R. 1991. Habitat deterioration and the migratory behaviour of moose (Alces alces L.) in Norway. J. Applied . Ecology. 28, 102-108. 26 Anderson, E.C. 1995. Morbillivirus infections in wildlife (in relation to their population 27 biology and disease control in domestic animals). Veterinary Microbiology. 44, 28 319-32 29 30 Anderson, J. L. 1994. "The Introduction of Elephant into Medium-Sized Conservation Areas." Pachyderm 18, 33-38. SECOND DRAFT Assessment of South African Elephant Management 41 1 Anthony, B. P. 2006. A View from the other side of the fence: Tsonga communities 2 and the Kruger National Park, South Africa, PhD Environmental sciences 3 and Policy, Central European University. 4 Bengis, R.G., R.A, Kock,. & J. Fischer, 2002. Infectious animal diseases : the 5 wildlife / livestock interface. Review. sci. tech. Off. Int.Epiz. 21(1), pp 53 – 6 65 7 8 9 Ben-Shahar, R. (1983) Patterns of elephant damage to vegetation in northern Botswana. Biological Conservation, 65, 249-256. Bezuidenhout, H. 2004. Report on the impact of elephants on the vegetation of the 10 Zwarthoek section, Marakele National Park. Arid Ecosystems Reserach 11 Unit. Conservation Services. 12 Bradshaw, G. A., A. N. Schore, J. L Brown, J. H. Poole, & T. Moss, 2005. Elephant 13 breakdown: Social trauma: early disruption of attachment can affect the 14 physiology, behaviour and culture of animals and humans over generations. 15 Nature 433, 807. 16 Chamaillé-Jammes, S., M. Valeix, & H. Fritz, 2007. Managing heterogeneity in 17 elephant distribution: interactions between elephant population density and 18 surface-water availability. Journal of Applied Ecology, 44, 625-633. 19 Chamaillé-Jammes, S., M. Valeix, & H. Fritz, (in press) Elephant management: why 20 can’t we throw the babies out with the artificial bathwater? Diversity and 21 Distributions. 22 Chiyo, P. I., E.OP. Cochrane, L. Naughton, &G.I. Basuta. 2005. Temporal Patterns 23 of Crop Raiding by Elephants: a Response to Changes in Forage Quality or 24 Crop availability? African Journal of Ecology 43, 48-55. 25 Conner, M.M. 2002. Movements of mule deer and elk in response to human 26 disturbance: 27 Research. Fort Collins, USA. 28 a literature review. Colorado Division of Wildlife, Mammals Cowling, R. & G.I.H. Kerley. 2002. Impacts of elephants on the flora and 29 vegetation of Subtropical Thicket in the Eastern Cape. In: Kerley, G.I.H., 30 Wilson, S. & Massey, A. (eds.) Proceedings of a workshop on Elephant 31 Conservation and Management in the Eastern Cape. Terrestrial Ecology 32 Research Unit Report 35, University of Port Elizabeth. SECOND DRAFT Assessment of South African Elephant Management 42 1 2 3 Craig, C. 1997. The ELESMAP Project Report, Namibia. Namibia Nature Foundation. Cumming, D. H. & B. Jones, B. 2005. Elephants in southern Africa: management 4 issues and options, Harare: WWF, WWF-SAPRO Occasional Paper Number 5 11. Harare: WWF. 6 Cumming, D.H.. 1981. The management of elephants and other large mammals in 7 Zimbabwe. In: P. Jewell, S. Holt & D. Hart (eds.). Problems in management 8 of locally abundant wild mammals. p 91-118. Academic press, New York. 9 10 11 12 13 Damm, G.R. 2002. The Conservation Game. Saving Africa’s Biodiversity. Safari Club International. Africa Chapter.Rivonia, South Africa. Davis, J.M. & J.A. Stamps. 2004. The effect of natal experience on habitat preferences. TREE 19, 411-416. De Vos, V., R.G. Bengis, N.P. Kriek, A. Michel, D.F. Keet, J.P. Raath & H.F. 14 Huchzermeyer. 2001. The epidemiology of tuberculosis in free-ranging 15 African buffalo (Syncerus caffer) in the Kruger National Park, South Africa. 16 Onderstepoort Journal of Veterinary Research. 68, 19-30. 17 18 19 20 21 22 23 De Waal, F. 2001. The Ape and the Sushi Master: Cultural Reflections of a Primatologist. Basic Books, New York, NY. Dias, P.C., 1996. Sources and sinks in population biology. Trends in Ecology and Evolution 11, 326–330. Douglas-Hamilton, I. 1973. On the ecology and behaviour of the Lake Manyara elephants. East African Wildlife Journal. 11:401-403. Druce, H. C., K. Pretorius, & R. Slotow. 2007. The response of an elephant 24 population to conservation area expansion: Phinda Private Game Reserve, 25 South Africa. Biological Conservation (in press). 26 Dublin, H.T., O.M. McShane, & J. Newby. 1997. Conserving Africa’s Elephants. 27 Current Issues & Priorities for Action. WWF International, Gland, 28 Switzerland. 29 Dudley, J.P., G.C. Craig, D. St.C. Gibson, G.Haynes, G. & J. Klimowicz, 2001. 30 Drought mortality of bush elephants in Hwange National Park, Zimbabwe. 31 African Journal of Ecology, 29, 187-194. SECOND DRAFT Assessment of South African Elephant Management 43 1 Freitag-Ronaldson, S. & L.C. Foxcroft, 2003. Anthropogenic Influences at the 2 Ecosystem Level Pp. 391-421 in The Kruger Experience: Ecology and 3 Management of Savanna Heterogeneity edited by J.T.Du Toit, K.H.Rogers, 4 and H.C.Biggs. Washington: Island Press. 5 Frost, P.G.H., J-C. Menaut, B.H. Walker, E. Medina, O.T. Solbrig, & M.J. Swift 6 (eds.). 1986. Responses of savannas to stress and disturbance. Biology 7 International Special Issue – 10. 8 Garcia, J., P.A. Lasiter, F. Bermudez-Rattoni & D.A. Deems. 1985. A general 9 theory of aversion learning. Pages 8-21 in Braveman N.S. and P. Bronstein 10 (eds.), Experimental Assessments and Clinical Applications of Conditioned 11 Food Aversions. New York Acad. Sci., New York, NY. 12 Gillson, L. & K. Lindsay, 2003. Ivory and ecology – changing perspectives on 13 elephant management and the international trade in ivory. Environmental 14 Science & Policy, 6, 411-419. 15 Gordon, C.H. 2003. The impact of elephants on the riverine woody vegetation of 16 Samburu National Reserve, Kenya. Unpublished report for Save the 17 Elephants. 18 Grant, C. C., T. Davidson, P.J. Funston, & D.J. Pienaar, 2002. Challenges Faced in 19 the Conservation of Rare Antelope: a Case Study on the Northern Basalt 20 Plains of the Kruger National Park. Koedoe 45, 45-66. 21 22 23 Hanks, J. 1979. A Struggle for Survival. The Elephant Problem. C. Struik Publishers, Cape Town, South Africa. Harrington, R., N. Owen-Smith, P.C. Viljoen, D.R. Mason, & P.J. Funston, 1999. 24 Establishing the Causes of the Roan Antelope Decline in the Kruger 25 National Park, South Africa. Biological Conservation 90, 69-78. 26 Henley, M.D. & S.R. Henley. 2007. Population dynamics and elephant movements 27 within the Associated Private Nature Reserves (APNR) adjoining the 28 Kruger National Park. Unpublished May Progress Report to the Associated 29 Private Nature Reserves. 30 31 32 Hoare R. E. 1995. Options for the control of elephants in conflict with people. Pachyderm 19: 54-63. Hoare, R, Hoare, R. 2003. Fencing an other barriers against problem elephants. SECOND DRAFT Assessment of South African Elephant Management 44 1 2 Hoare, R. E.; du Toit, J. T. 1999. Coexistence between people and elephants in African savannas. Conservation Biology, 13, 633-639. 3 Hoare, R.E. 2001. A decision support system for managing human-elephant conflict 4 situations in Africa. IUCN African Elephant Specialist Group Report. 5 6 7 8 9 Illius, A. & T.G. O'Connor, 2000. Resource heterogeneity and ungulate population dynamics. Oikos, 89, 283-294. Jachmann, H. & R.H.V. Bell, 1984. Why do Elephants Destroy Woodland? Pachyderm 3, 9-10. Johnson, C. F. 1998. Vulnerability, Irreplaceability and Reserve Selection for the 10 Elephant Impacted Flora of the Addo Elephant National Park, Eastern 11 Cape, South Africa. M.Sc. Thesis, Rhodes University, Grahamstown 12 Unpublished manuscript. 13 Johnson, C. F., R.M. Cowling, P.B. & Phillipson, 1999. The flora of the Addo 14 Elephant National Park, South Africa: are threatened species vulnerable to 15 elephant damage? Biodiversity and Conservation 8, 1456-1457. 16 Joubert, D. 2006 Hunting behaviour of lions (Panthera leo) on elephants 17 (Loxodonta africana) in the Chobe National Park, Botswana. African 18 Journal of Ecology 44, 279-281. 19 20 21 22 23 Kangwana, K. 1995. Human-elephant conflict: the challenge ahead. Pachyderm 19, 9-14. Karidozo, M & F.V. Osborn. 2005. Can bees deter elephants from raiding crops? An experiment in the communal lands of Zimbabwe. Pachyderm 39: 26-32. Kerley, G. I. H. & M. Landman. 2006. The impact of elephant on biodiversity in the 24 Eastern Cape Subtropical thickets. South African Journal of Science 102, 1- 25 8. 26 Kimball, B.A., F.D. Provenza & E.A. Burritt. 2002. Importance of alternative foods 27 on the persistence of flavor aversions: Implications for applied flavor 28 avoidance learning. Applied Animal Behaviour Science 76, 249-258. 29 Knight, M., G. Castley, L. Moolman, & J. Adendorff. 2002. Elephant management 30 in Addo Elephant National Park. In: G. Kerley, S. Wilson & A. Massey 31 (eds.). Elephant conservation and management in the Eastern Cape – SECOND DRAFT Assessment of South African Elephant Management 45 1 workshop proceedings. p 32-40. Terrestrial Ecology Research Unit 2 University of Port Elisabeth, Port Elisabeth. 3 Liversidge, R. 1978. It was all exaggeration. African Wildlife 32: 26–27. 4 Lombard, A. T., C.F. Johnson, R.M. Cowling, & R.L Pressey, 2001. Protecting 5 Plants From Elephants: Botanical Reserve Scenarios Within the Addo 6 Elephant National Park, South Africa. Biological Conservation 102, 191- 7 203. 8 Lubow, B.C. 1996. Optimal Translocation Strategies for Enhancing Stochastic 9 Metapopulation Viability. Ecological Applications 6, 1268-1280 10 management strategies. People and Wildlife Initiative. Wildlife Conservation 11 12 Research Unit, Oxford University. Moss, C.J. & J.H. Poole. 1983. Relationships and social structure of African 13 elephants. Pages 315-325 in R.A. Hinde (ed.) Primate Social Relationships: 14 An Integrated Approach. Sinauer. Sutherland, MA. 15 Naughton-Treves, L. 1998. Predicting Patterns of Crop Damage by Wildlife around 16 Kibale National Park, Uganda. Conservation. Biology. 12, 156-158. 17 Nelleman, C., R.M. Stein, & L.P. Rutina, 2002. Links between terrain 18 characteristics and forage patterns of elephants (Loxodonta africana) in 19 northern Botswana. Journal of Tropical Studies, 18, 835-844. 20 Nelson, A., P.Bidwell, & C. Sillero-Zubiri, 2007. A review of human-elephant 21 conflict management strategies Wildlife Conservation Research Unit, 22 Oxford University, People and Wildlife Initiative. Wildlife Conservation 23 Research Unit, Oxford University. 24 O'Connell-Rodwell C.E., T. Rodwell, M. Rice & L.A Hart. 2000. Living with the 25 modern conservation paradigm: can agricultural communities co-exist with 26 elephants? A five-year case study in East Caprivi, Namibia. Biological 27 Conservation 93, 381-391 28 29 Omondi, P., E. Bitok, & J. Kagiri, 2004. Managing human–elephant conflicts: the Kenyan experience. Pachyderm 36, 80-86. 30 Osborn, F. V. & G.E. Parker. 2002. Community-based methods to reduce crop loss 31 to elephants: experiments in the communal lands of Zimbabwe. Pachyderm 32 33: 32-38. SECOND DRAFT Assessment of South African Elephant Management 46 1 2 Osborn, F. V. & G.E. Parker. 2003. Linking two elephant refuges with a corridor in the communal lands of Zimbabwe. African Journal of Ecology 41: 68-74. 3 Osborn, F. V. & L.E.L.Rasmussen. 1995. Evidence for the effectiveness of an oleo- 4 resin capsicum aerosol as a repellent against wild elephants in Zimbabwe. 5 Pachyderm 20: 55-64. 6 Osborn, F. V. 2004. Seasonal variation of feeding patterns and food selection by 7 crop-raiding elephants in Zimbabwe. African Journal of Ecology 42, 322- 8 327. 9 10 Osborn, F. V. 1997. The ecology and deterrence of crop raiding elephants: final technical report. Unpublished report to USFWS. 23pp. 11 Osborn, F. V. 2004. Seasonal Variation of Feeding Patterns and Food Selection by 12 Crop-Raiding Elephants in Zimbabwe. African Journal of Ecology 423, 22- 13 27. 14 Osborn, F. V. Capsicum oleoresin as an elephant repellent: field trials in the 15 communal lands of Zimbabwe. 2002. Journal of Wildlife Management. 66, 16 674-677. 17 Owen-Smith N., G.I.H. Kerley B. Page R. Slotow. & R.J. van Aarde .2006 A 18 scientific perspective on the management of elephants in the Kruger 19 National Park and elsewhere. South African Journal of Science 102: 389 – 20 394. 21 22 23 24 25 Owen-Smith, N. 1996. Ecological guidelines for waterpoints in extensive protected areas. South African Journal of Wildlife Research, 26, 107-112. Owen-Smith, R. N. 1988. Megaherbivores: the influence of very large body size on ecology. Cambridge Studies in Ecology, 369. Owen-Smith, R.N. 1983 Dispersal and the dynamics of large herbivores in enclosed 26 areas: Implications for management. In: Management of large mammals in 27 African conservation areas (Ed. R.N. Owen-Smith). Haum, Pretoria. 28 Peel M.J.S. 2005. Towards a predictive understanding of savanna vegetation 29 dynamics in the eastern Lowveld of South Africa: with implications for 30 effective management. PhD thesis - University of kwaZulu-Natal. SECOND DRAFT Assessment of South African Elephant Management 47 1 2 Peel Mike 2006. Ecological Monitoring: Association of Private Nature Reserves. (Unpublished landowner report). 3 Pienaar, D.J., H.C. Biggs, A. Deacon, W. Gertenbach, S. Joubert, F. Nel, F., L.van 4 Rooyen, & F. Venter, F. 1997. A revised water-distribution policy for 5 biodiversity maintenance in the Kruger National Park. Internal report. South 6 African National Parks, Skukuza. 7 8 9 10 11 12 13 Porter, R.N. 1970. An ecological reconnaissance of the TPNR Private Nature Reserve. Unpublished report. Timbavati Private Nature Reserve, Hoedspruit. Provenza, F.D. 2003. Foraging Behavior: Managing to Survive in a World of Change. Utah State University., Logan. Provenza, F.D. 2007. What does it mean to be locally adapted and who cares anyway. Journal of Animal Science accepted. Provenza, F.D. & J.J. Villalba. 2006. Foraging in Domestic Vertebrates: Linking the 14 Internal and External Milieu. Pages 210-240 in V.L. Bels (ed.) Feeding in 15 Domestic Vertebrates: From Structure to Function. CABI Publ., 16 Oxfordshire, UK. 17 Provenza, F.D., J.J. Villalba, L.E. Dziba, S.B. Atwood & R.E. Banner. 2003. 18 Linking herbivore experience, varied diets, and plant biochemical diversity. 19 Small Ruminant Research 49, 257-274. 20 Redfern, J.V., C.C. Grant, A. Gaylard, & W.M. Getz, 2005. Surface water 21 availability and the management of herbivore distributions in an African 22 savanna ecosystem. Journal of Arid Environments, 63, 406-424. 23 Redfern, J.V., R. Grant, H. Biggs, W.M. & Getz, 2003. Surface-water constraints on 24 herbivore foraging in the Kruger National Park, South Africa. Ecology 25 (Durh.), 84, 2092-2107. 26 Skinner, B.F. 1981. Selection by consequences. Science 213,501-504. 27 Smit, I.P.J., C.C. Grant, & B.J.Devereux, 2007(a). Do artificial waterholes influence 28 the way herbivores use the landscape? Herbivore distribution patterns 29 around rivers and artificial surface water sources in a large African savanna 30 park. Biological Conservation, 136, 85-99. SECOND DRAFT Assessment of South African Elephant Management 48 1 Smit, I.P.J., C.C. Grant, & I.J. Whyte. 2007(b). Landscape-scale sexual segregation 2 in the dry season distribution and resource utilisation of elephants in Kruger 3 National Park, South Africa. Diversity and Distributions, 13, 225-236. 4 Smit, I.P.J., C.C. Grant, & I.J. Whyte.. (in press) Elephants and water provision: 5 6 what are the management links? Diversity and Distributions Smith R.J. & S.M. Kasiki. 2000. A spatial analysis of human-elephant conflict in 7 the Tsavo-ecosystem, Kenya. A report to the African Elephant Specialist 8 Group, Human-Elephant Conflict Task Force, of IUCN. Gland, Switzerland. 9 South African National Parks (2005) Report on the Elephant Management Strategy: 10 Report to the Minister: Environmental Affairs and Tourism on Developing 11 Elephant Management Plans for National Parks with Recommendations on 12 the 13 http://www.sanparks.org/events/elephants/strategy_19-09-2005.pdf 14 process to be followed. Accessible at: South African Savannas Network. The Status of Southern Africa’s Savannas. 15 Report to United Nations Environment Programme, University of London, 16 London. 17 Stokke, S. & J.T. du Toit. 2000. Sex and size related differences in the dry season 18 feeding patterns of elephants in Chobe National Park, Botswana. Ecography, 19 23, 70-80. 20 Stuart-Hill G.C. 1992 Effects of elephants and goats on the Kaffrarian Succulent 21 Thicket of the eastern Cape, South Africa. Journal of Applied Ecology 29, 22 699-710. 23 24 25 26 27 28 29 Taylor, R. D. 1994. Elephant Management in Nyaminyami District, Zimbabwe: Turning a Liability into an Asset. Pachyderm 18, 17-29. Tchamba, M.N. 1995. The problem elephants of Kaele: A challenge for elephant conservation in northern Cameroon. Pachyderm 19, 26-31. Thrash, I. 1997. Infiltration rate of soil around drinking troughs in the Kruger National Park, South Africa. Journal of arid environments 35,617-625. Tolsma, D. J., W.H.O. Ernst, & R.A. Verwey. 1987. Nutrients in Soil and 30 Vegetation Around Two Artificial Waterpoints in Eastern Botswana. 31 Journal of Applied Ecology 24, 991-1000. SECOND DRAFT Assessment of South African Elephant Management 49 1 Trollope, W. S. W., L.A. Trollope, H.C. Biggs, D.J. Pienaar, &A.L.F. Potgieter, 2 1998. Long Term Changes in the Woody Vegetation of the Kruger National 3 Park, With Special Reference to the Effects of Elephants and Fire. Koedoe 4 41, 103-112. 5 Trollope, W.S.W. 1992. Control of bush encroachment with fire in savanna areas of 6 South Africa. Special publication of the Grassland Society of Southern 7 Africa, Scottsville. 8 van Aarde, R. J.; T.P.Jackson, 2007. Megaparks for metapopulations: Addressing 9 the causes of locally high elephant numbers in southern Africa. Biological 10 11 12 13 Conservation, 134, 289-297. Verlinden, A. & I.K.N. Gavor. 1998. Satellite tracking of elephants in northern Botswana. African Journal of Ecology, 36, 105-116. Vieira, M.E., M.M. Conner, G.C. White, & D.J. Freddy. 2003. Effects of archery 14 hunter numbers and opening dates on elk movement. Journal of Wildlife 15 Management. 67, 717-728. 16 Viljoen, P.J. & J.D.P. Bothma, 1990. Daily movements of desert-dwelling elephants 17 in the northern Namib Desert. South African Journal of Wildlife Research, 18 20, 69-72. 19 20 Vollrath, F. & I. Douglas-Hamilton. 2005a. African bees to control African elephants. Naturwissenschaften 92 , 508-511. 21 Vollrath, F. & I. Douglas-Hamilton. 2005b. Elephants buzz off! Swara 25, 20-21. 22 Walker, B. H., Emslie, R. H., Owen-Smith, N., and Scholes, R. J. 1987. "To Cull or 23 Not to Cull: Lessons Form a Southern African Drought." Journal of Applied 24 Ecology 24381-401. 25 Walpole, M. 2001. Factors affecting the recovery of the Masai Mara black rhino 26 population. In: Walpole, M.J., G.G. Karanja, N.W. Sitati, N.W. & N. 27 Leader-Williams (eds). Wildlife and People: Conflict and Conservation in 28 Masai Mara, Kenya. 29 Wasilwa, N.S. 2001. Human elephant conflict in Transmara district, Kenya. In: 30 Walpole, M.J., G.G. Karanja, N.W. Sitati, N.W. & N. Leader-Williams 31 (eds). Wildlife and People: Conflict and Conservation in Masai Mara, 32 Kenya. SECOND DRAFT Assessment of South African Elephant Management 50 1 2 3 4 5 Western, D. & D. Muitumo. 2004. Woodland loss and restoration in a savanna park: a 20-year experiment. African Journal of Ecology 42, 111-121. Western, D., 1989. The ecological role of elephants in Africa. Pachyderm 12, 42– 45. Whyte, I. J. & S. C. J. Joubert. 1988. Blue wildebeest population trends in the 6 Kruger National Park and the effects of fencing. South African Journal of 7 Wildlife Research 18, 78-87. 8 Wittemeyr, G., I. Douglas-Hamilton, & W. M. Getz. 2005. The socio-ecology of 9 elephants: analysis of the processes creating multitiered social structures. 10 11 Animal Behaviour Manuscript number A9821, 1-15. Wittemeyr, G., H.B.Rasmussen, & I. Douglas-Hamilton. 2007. Breeding phenology 12 in relation to NDVI variability in free-ranging African elephant. Ecography 13 30, 42-50. 14 WWF 1998. Wildlife Electric Fencing Projects in Communal Areas of Zimbabwe - 15 Current Efficacy and Future Role. Price Waterhouse Coopers, Report for 16 WWF Southern Africa Programme Office, Harare, Zimbabwe. 17 WWF. 1997. Conserving Africa's elephants: Current issues and priorities for 18 action. In Dublin, H.T., T.O. McShane and J. Newby. World Wide Fund for 19 Nature International Report, Gland, Switzerland. 20 Wynn, S. 2003 Zambezi River: Wilderness and Tourism Research Into Visitor 21 Perceptions About Wilderness and Its Value. USDA Forest Service 22 Proceedings RMRS-P-27 200 23 24 Young, E. 1970. Water as faktor in die ekologie van wild in die nasionale Krugerwildtuin. PhD thesis, University of Pretoria, Pretoria. SECOND DRAFT Assessment of South African Elephant Management 51