African Journal of Range & Forage Science

ISSN: 1022-0119 (Print) 1727-9380 (Online) Journal homepage: http://www.tandfonline.com/loi/tarf20

Linear structures in the Karoo, South Africa, and their impacts on biota

W Richard J Dean, Colleen L Seymour & Grant S Joseph

To cite this article: W Richard J Dean, Colleen L Seymour & Grant S Joseph (2018) Linear structures in the Karoo, South Africa, and their impacts on biota, African Journal of Range & Forage Science, 35:3-4, 223-232 To link to this article: https://doi.org/10.2989/10220119.2018.1514530

Published online: 22 Nov 2018.

Submit your article to this journal

View Crossmark data

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tarf20 African Journal of Range & Forage Science 2018, 35(3&4): 223–232 Copyright © NISC (Pty) Ltd Printed in South Africa — All rights reserved AFRICAN JOURNAL OF RANGE & FORAGE SCIENCE ISSN 1022-0119 EISSN 1727-9380 http://dx.doi.org/10.2989/10220119.2018.1514530

Review Paper

Linear structures in the Karoo, South Africa, and their impacts on biota§

W Richard J Dean1,2*, Colleen L Seymour1,3 and Grant S Joseph1,4

1 FitzPatrick Institute of African Ornithology, DST/NRF Centre of Excellence, Department of Biological Sciences, University of Cape Town, Cape Town, South Africa 2 South African Environmental Observation Network (SAEON): Arid Lands Node, Kimberley, South Africa 3 South African National Biodiversity Institute, Kirstenbosch Research Centre, Claremont, Cape Town, South Africa 4 SARChI-Chair on Biodiversity Value and Change, Department of Zoology, School of Mathematical and Natural Science, University of Venda, Thohoyandou, South Africa * Corresponding author, email: [email protected]

Linear structures include fences, roads, railways, canalised water ways and power lines, all man-made. Fencing as a way of managing livestock began in the late 1800s, and by the early twentieth century was almost fully implemented throughout the Karoo sensu lato. The advent of these fences, and now in many instances, ‘game proof’ (~2 m high) and electric fences have impacted native flora and fauna in various ways. Roads influence the quality and quantity of vegetation along corridors throughout the Karoo, with impacts on wildlife through increased mortality, but also provide foraging opportunities. Road-side structures, such as transmission poles, offer nest sites and perches in otherwise treeless landscapes, benefiting certain avifauna. Railway lines create similar corridors, carrying ungrazed vegetation along them. Their associated structures – culverts and (historically) steel frame bridges – provide nest sites for and some mammals. Transmission cables and pylons along power lines have mixed benefits to conservation, providing nest and perch sites, but can also cause bird mortalities. Unquestionably, these linear structures have had marked effects on the biota of the Karoo, through their effects on wildlife and livestock movement and mortality, plant demography and the spread of native and alien .

Keywords: fencing, high tension pylons, railways, roads, road verges

Preamble

Linear structures are not ‘natural’ in most ecosystems, and been reviewed for temperate regions (Morelli et al. 2014). are among the main threats to natural habitat worldwide Furthermore, mitigation measures and research gaps (Geneletti 2006). Linear structures include fences, roads, need to be identified. railways, power lines and pylons, and canalised rivers, This review has three aims. Firstly, we examine the and share common impacts, including direct habitat loss available literature to assess the positive and negative and habitat fragmentation (Andrews 1990; Trombulak impacts of linear structures on biodiversity in the Karoo; and Frissell 2000; Spellerberg 2002; Forman et al. 2003; secondly, we assess which taxa are most impacted, and Coffin 2007), hazards for many species of (Ledger which life history or behavioural traits might be associ- and Annegarn 1981; Bevanger 1994; Ledger and Hobbs ated with these impacts; and finally, we review possible 1999; Janss 2000; Lehman et al. 2007; Shaw et al. 2010; mitigation measures and research gaps for future work. We Jenkins et al. 2011; Shaw 2013; Collinson et al. 2015) and outline impacts associated with linear structures in general, corridors for invasive biota (Rahlao et al. 2010). Yet certain in the Karoo, and then review each of the main categories linear structures can also be beneficial to biodiversity, for of linear structures in turn. example, fences can minimise human–wildlife conflict The Karoo consists of two biomes, the Succulent Karoo (Hayward and Kerley 2009) and reduce the likelihood of and the Nama-Karoo, which together make up about 30% wildlife–vehicle collisions (Collinson and Patterson-Abrolat of South Africa and a small part of Namibia (Dean and 2016). Despite their ubiquity, which is ever-increasing Milton 1999; and see map in Walker et al. 2018). Both with infrastructure development, the effects of linear biomes are semi-arid, but differ in rainfall seasonality structures on biota are poorly known, particularly for arid and amount (Desmet and Cowling 1999), and in vegeta- and semi-arid regions of the world, although these have tion (Mucina and Rutherford 2006). Our review covers the

§ This article is from the ‘Karoo Special Issue: Trajectories of Change in the Anthropocene’.

African Journal of Range & Forage Science is co-published by NISC (Pty) Ltd and Informa UK Limited (trading as Taylor & Francis Group) 224 Dean, Seymour and Joseph

Karoo sensu lato – the effects and impacts of linear been introduced to control black-backed jackals Canis structures are, in general, not confined to one or the other mesomelas (Heard and Stephenson 1987) and many farms of the two biomes. are now using single-wire electric fences to manage small groups of livestock (Macdonald 2005). Linear structures in the Karoo Fencing has recently become even more of a barrier in the landscape, given an increase in the number of Linear structures bring about direct habitat loss by removing ‘life-style’ farms, ‘game’ farms and private nature reserves long strips of habitat, and indirect habitat degradation, in the Karoo (Brandt and Spierenberg 2014). Between through resultant fragmentation, disruption of migration 1993 and 2000, the percentage of agricultural land in routes and drainage lines, or by partitioning ephemeral South Africa used for game farming increased from 8.5% wetlands. These structures (particularly railways, roads and to 12.5% (van der Merwe and Saayman 2003), and is power lines) introduce an additional mortality factor into likely to increase, based on observed trends. The fencing ecosystems where have no evolutionary history of required for these farms is higher and more complex than avoiding fast-moving vehicles or anticipating difficult-to-see that required for managing commercial livestock, and may structures such as electrical cables. The impacts of power be electrified and up to 2.4 m high (CapeNature 2015). line structures, including cables and pylons, have been When farms are fenced following policy and recommen- relatively well studied in the Karoo (Ledger and Annegarn dations, the owner may be exempted from legislation 1981; Jenkins et al. 2010, 2011; Shaw 2013; Shaw et al. applicable to the hunting, catching and selling of game, 2016, 2017), where the rising tide of investment in solar and and thus there are benefits to the land owner in terms of wind farms has needed a closer examination of the effects purchasing and selling game (Province of the Northern of power cables across flight paths of birds, the resulting Cape 2011; CapeNature 2015). The high capital invest- mortality of certain species, and the potential impact of ment in accommodation, other infrastructure and vehicles losses of individuals to populations. Although hydraulic (Anon 2003) associated with game farming necessitates fracturing (fracking) in the Karoo for the extraction of natural protecting the investment against predators and poachers, gas (Scholes et al. 2016) does not strictly involve linear which in turn means predator-proof fences, either electric structures, the associated infrastructures and increased fencing, mesh fencing or both. traffic will nevertheless fragment habitats and are as much a conservation cause for concern as linear structures. Impacts of fencing Most negative impacts of fencing increase in severity with Fencing the ‘impenetrability’ of the fencing, governed by fence An understanding of fencing in the Karoo would be height, fencing type and the use of electrification. The incomplete without knowledge of its origins and develop- fences associated with game farming tend to rank highly ment. Historically, livestock were managed by nomadic in all of these, involving multiple electrified strands at pastoralism or herding to a central place for shelter during different heights, which prevent or restrict the movements, the nights. Neither system confined livestock except in or kill, a number of ‘non-game’ species, including reptiles the night shelters where a small area may have been and mammals (Burger and Branch 1994; Beck 2010; closed off with stone walls. In the late 1800s, however, the Macray 2017). advantages of subdividing farms into smaller units, each The negative impacts of fencing that are particularly with a watering place, became apparent, and fencing- relevant to the Karoo include (inter alia) disruption of off paddocks began in the districts of Beaufort West, dispersal patterns (Vanak et al. 2010; Cassidy et al. 2013; Fraserburg, Carnarvon and Victoria West (Talbot 1961). Woodroffe et al. 2014), increased risk of local extinction due This was encouraged by the Fencing Act (Act 30 of 1883), to random demographic, genetic and environmental events but a conservative attitude and lack of funds delayed wide (Hayward and Kerley 2009; Kesch et al. 2014; Woodroffe implementation until just about the turn of the century. et al. 2014), collision with fences, increased mortality rates In addition to gaining control and ownership of the land (Boone and Hobbs 2004; Beck 2010; Macray 2017), and through fencing (van Sittert 2002), livestock managers reduction of carrying capacity through overgrazing and saw many other advantages: periodic resting of rangeland, decreases in habitat quality (Hoare 1992; Ben-Shahar 1993; control of ‘vermin’ (Nattrass et al. 2017), prevention of Boone and Hobbs 2004; Kesch et al. 2014; Woodroffe et al. the spread of diseases (Talbot 1961), and curbing of the 2014). The need for biota to move in response to climate periodic movements of springbok (Antidorcas marsupialis), change likely increases the severity of all of these threats. which were seen as a threat to grazing resources Fences can also be positive for biodiversity, because (Roche 2004). they can help reduce human–wildlife conflict, enabling the The 1922 amendment to the fencing act covered the existence of predators in protected areas within the Karoo. inclusion of vermin-proof fences and gave livestock Fences can exclude or control human access (Hayward managers another weapon in the war against predators and Kerley 2009), and allow for restoration or protection (Talbot 1961). The Karoo became a network of paddocks of certain habitats of concern (Boone and Hobbs 2004), enclosed by stock-proof and jackal-proof fences; indeed, prevent spread of diseases (Hayward and Kerley 2009) and the whole of South Africa ‘has a fence culture, with reduce wildlife road mortalities by preventing game from thousands of kilometres of game [and stock-proof] fencing reaching roads (although game can also become trapped dividing farms [and] national parks’ (Collinson et al. within the road reserve with fencing denying them a route 2015). More recently, electrified stock-proof fences have away from the road; Eloff and van Niekerk 2005). African Journal of Range & Forage Science 2018, 35(3&4): 223–232 225

Taxa most affected by fencing include raising the minimum height of electrified strands to The way that animal taxa interact with fences is driven 200–250 mm, although landowners are not likely to support primarily by how they move and how they respond to this solution as higher strands reduce the ability to control threats. Although most mammals and birds would likely ‘problem’ species such as jackal (Canis mesomelas) and jump away on receiving an electrical shock from electrical porcupine (Hystrix africaeaustralis) (Macray 2017); rock fencing, tortoises withdraw into their shells, and snakes may aprons (Beck 2010) and barriers in front of fences (Burger actually wrap themselves around the livewire (Beck 2010 and Branch 1994); periodically switching off the current and and references therein). controlling the voltage could also reduce mortalities (Macray Most information on electric-fence mortalities in the Karoo 2017). Barriers and aprons carry the disadvantage that they is on tortoises, with some data on mammal mortalities. may actually trap tortoises next to wires and also effectively Species susceptible to injury and/or death for the De Aar reduce the height of the electric wire (Macray 2017). and Middelburg areas of the Karoo are given in Table 1. No Furthermore, Macray (2017) found that for some species large ‘walking’ birds are listed, but there is the potential that of tortoise, although fences were linked to higher mortality, species such as the large and smaller korhaans whether or not the fence was electrified or not made no could be susceptible to injury or death as a result of difference to mortality rates. Other measures, such as fence encountering electric fences. It is unlikely that these species design, underway tunnels and virtual fencing to control would attempt to move directly through the fence unless the problem animals (see Macray 2017 for more details) are all spacing of the wire strands appeared to be sufficiently open. areas of research that may provide useful solutions for not Fences also introduce indirect increased predation only tortoises but also for other species. pressure. For example, tortoises may spend more time near fences, as they search for a way to get through, Roads but the fences themselves provide lookout perches for There is an extensive network of roads, both tarred predatory birds such as Pied Crows (Macray 2017). (sealed) and gravel, in South Africa, including the Karoo (Kannemeyer 2016). Roads, everywhere, provide an Possible mitigation measures and research gaps for fencing array of goods and services for animals and plants (Coffin The number of issues, both positive and negative, 2007, van der Ree et al. 2015), especially in low produc- associated with fencing means that a number of mitigation tivity semi-arid ecosystems such as the Karoo (Milton et al. strategies exist to address some threats, but we are unlikely 2015) and inland arid Australia (Lee et. 2015). Roads act to address all of them, and research into the possibilities as corridors through arid lands, providing resources not presented by methods such as virtual fencing, e.g. where an freely available in the surrounding rangelands. Runoff water affordable collar is used to control animal movement, and from road surfaces, spillages of grain, food discarded by use of scent marking of habitat to discourage undesirable travellers, decomposing road kills, dust and inputs of carbon species from an area, could be used to do away with much dioxide from motor vehicles all combine to make roads and of the physical fencing present in the landscape today. Until their verges more productive than the surrounding habitat such solutions are affordable and in widespread use, other (O’Farrell and Milton 2006). solutions are needed, however. Fencing can be made to be more visible to larger animals by attaching items to fences. Impacts of roads on biota Issues of isolation and genetic inbreeding can be addressed Roads affect animal and plant populations in many ways, with translocation of individuals periodically, to increase including habitat fragmentation, inhibiting movement, and genetic diversity. Recently, studies have suggested as sources of mortality and modifications to behaviour measures to reduce tortoise mortality associated with (reviewed by Andrews 1990; Trombulak and Frissell fencing (e.g. Beck 2010; Macray 2017). These measures 2000; Spellerberg 2002; Forman et al. 2003; Coffin 2007; chapters in van der Ree et al. 2015). For birds, there are both disadvantages and benefits from roads and associated Table 1: Species potentially at risk from electrocution from electric infrastructure (Morelli et al. 2014). fences at two Karoo sites, De Aar and Middelburg. Data from Beck (2010) Taxa affected by roads and road verges Most, if not all, taxa are affected by roads and road verges. Vernacular name Scientific name The roads themselves represent treacherous stretches Dwarf Tortoise Chersobius [Homopus] boulengeri where fast-moving vehicles collide with species poorly Greater Dwarf Tortoise Homopus femoralis equipped to avoid them. Some species even have adapta- Tent Tortoise Psammobates tentorius tions that make them particularly vulnerable to becoming Leopard Tortoise Stigmochelys pardalis road kill, such as the imitation of shaking leaves when in Southern Rock Monitor Varanus albigularis open spaces, in which chameleons engage, increasing the Eastern Cape Dwarf Chameleon Bradypodion [karooicum] ventrale time they spend in the path of oncoming traffic. Ectothermic Puff Adder Bitis arietans species, such as snakes and other reptiles, are attracted to Rinkhals Hemachatus heamachatus the warmth provided by tar roads, and many species benefit Cape Cobra Naja nivea from, but also become, road kill. Black-backed Jackal Canis mesomelas Structures along roads, such as ‘telephone’ poles in an Ratel/Honey Badger Mellivora capensis otherwise almost tree-less landscape, offer nest sites Aardvark Orycteropus afer to certain bird species, particularly crows (Malan 1994; 226 Dean, Seymour and Joseph

Dean and Milton 2000; Hockey et al. 2005), and benefit roads benefits plants by reducing the impact of some insect species such as Greater Kestrel (Falco rupicoloides), that herbivores (van den Bosch and Telford 1964), leading to use abandoned crow nests (Hockey et al. 2005). Road higher productivity of plants in the verge, and to increases kills attract predators and scavengers that may become in the abundance of herbivorous invertebrates on plants in road kills themselves (Dean and Milton 2003; Dean et al. road verges (Lightfoot and Whitford 1991). 2005; Collinson et al. 2015; Lee et al. 2015; Shilling et al. Compared to rangelands, vegetation on road verges 2015), frequently in the Karoo (Dean and Milton 2003) has remained relatively pristine (Milton et al. 1994; and in other southern African ecosystems (Bullock et Siegfried 1999) and in the Karoo is usually separated al. 2011). Table 4 in Dean and Milton (2003) shows a from rangelands by fences; verges often have remnants number of small carnivorous mammals that had been of vegetation types not represented, or rare, on the other killed on roads in the Karoo. These include (no. of records) side of the fence (Esler and Milton 2006; O’Farrell and African Wild Cat (Felis lybica) (7), Black-footed Cat (Felis Milton 2006). Road verges may also be places where alien nigripes) (1), domestic cat (Felis sp.) (1), Bat-eared Fox invasive plants are common, however. Runoff water, plus (Otocyon megalotis) (158), Cape Fox (Vulpes chama) (9), disturbance, along edges of roads in arid areas such as Black-backed Jackal (Canis mesomelas) (1), Striped the Karoo lead to weed invasions (Milton and Dean 1998; Pole-cat (Ictonyx striatus) (58), Small-spotted Genet Rahlao et al. 2010), and the current practice of clearing (Genetta genetta) (6), Suricate (Suricata suricatta) (7), vegetation on verges (Petersen 2009) makes space Yellow Mongoose (Cynictis pencillata) (7), Small Grey available for weeds (Milton and Dean 2010). This can lead Mongoose (Galerella pulverulenta) (7) and Water Mongoose to more lush foraging sites that attract animals, and that in (Atilax paludinosus) (2) (Dean and Milton 2003). turn may lead to road kills. The implication is that the animals were foraging on the road and not simply crossing from one side to the other, Mitigation and future research to reduce impacts of roads although there is no certainty that this was always the case. Despite the problems presented to biodiversity by fencing, Despite the obvious problem of roadkills, surprisingly little is it also reduces wildlife–vehicle collisions (Eloff and van known about the effects of road kills on animal populations Niekerk 2005) near roads. Other measures could include in southern Africa (Collinson et al. 2015). Roads and road signage or traffic calming measures in areas where collision kills may represent an important source of food for crows rates with wildlife are known to be high, and building of in the Karoo (Siegfried 1963) and other arid ecosystems ‘under road’ passes in areas for migration. Future research in southern Africa (Winterbottom 1975; Macdonald and should focus on identifying collision hotspots; landuse, Macdonald 1983; Malan 1994; Joseph et al. 2017), habitat and road types associated with higher incidences leading to higher numbers of crows than would ordinarily of roadkill (e.g. Clevenger et al. 2001, 2003), the best be supported (Fincham et al. 2015). This in turn leads to design for under road passes, as well as the possibilities increased predation pressure on groups such as tortoises, of for virtual fencing. Given that roads also carry indirect which some species are of conservation concern (Fincham effects associated with increased predation pressure from and Lambrechts 2014). In other arid ecosystems, both in predatory birds that have increased their distributions in the Southern and Northern Hemispheres, roads through response to habitat (Dean et al. 2005), climate change arid and semi-arid areas similarly represent sources of food, (see, e.g., Cunningham et al. 2015) and driver behaviour particularly for crows (Austin 1971; Conner and Adkisson (e.g. Joseph et al. 2017), research into how to reduce their 1976; Williams and Colson 1989; Knight and Kawashima densities is also an important field of enquiry. 1993; Knight et al. 1995) but also for some Accipitridae A particular gap in our knowledge is the effects of dust (Meunier et al. 2000). Raptors (Acciptridae sensu stricto), from gravel roads in the Karoo and the impacts this may such as Pale Chanting Goshawks (Melierax canorus) have on plants and invertebrates, and, for that matter, (Malan 1992), Cape Crows (Corvus capensis), Pied the impact of dust on cyanobacterial crusts in adjacent Crows (Corvus albus), and White-necked Ravens (Corvus rangeland (Pointing and Belnap 2014). Similarly, the effects albicollis), regularly collect fur and hair from road-killed of carbon dioxide on roadside biota requires investigation. mammals for nest-lining material (Dean and Milton 2003), so road killed animals are not only a source of food. Railway lines Despite increased predation associated with increases ‘Railways alter landscapes in simple yet profound ways’ in crows and other predators, rodents, reptiles and (Dorsey et al. 2015). Despite this alteration to the landscape other animals may also benefit from roads through arid and the many kilometres of rail in southern Africa (Anon and semi-arid areas. Animals may be attracted to grain 2016) – South Africa had 20 500 km of railway (approxi- spills that infrequently provide small patches of food for mately the distance from the North to the South poles) in granivorous ants, rodents and birds. Verges provide 2014 (https://tradingeconomics.com/south-africa/rail-lines- shelter and food for the smallest of mammals, such as total-route-km-wb-data.html) – there are very few data on shrews and mice; this source of potential prey attracts the impacts or benefits to biota in the Karoo, or in arid or snakes and mongooses, some of which are killed on the semi-arid ecosystems anywhere else in the world (Dorsey road (Dean and Milton 2003). Other animals attracted to et al. 2015; Santos et al. 2017). roads, for which there are very little data for the Karoo, include granivorous and scavenging ants (Tshiguvho et Impacts of railways on biota in the Karoo al. 1999), predatory beetles and invertebrate detritivores There are hardly any data on interactions, benefits and (WRJD pers. obs). In south-western USA, dust from gravel impacts of railways and biota for the Karoo. Elsewhere, African Journal of Range & Forage Science 2018, 35(3&4): 223–232 227 most studies have focused on train–large mammal not been any concomitant studies that show an advantage collisions, particularly large ungulates, bears and wolves; to plants and animals, as shown by van den Bosch and such studies often lack quantitative data (Dorsey et al. Telford (1964) and Lightfoot and Whitford (1991) for 2015; Borda-da-Àgua et al. 2017). Furthermore, almost all emissions from road traffic. There is an urgent need for studies on the impacts of railways on the biota have been studies on the ecological footprint of railways in southern done in the Northern Hemisphere, including studies on birds Africa, particularly on the importance of rail reserves as and railway lines in wetland habitats (Godinho et al. 2017a, benchmark sites for rare plants in the Karoo, and the effects 2017b; Lucas et al. 2017; Múrias et al. 2017). Studies in of emissions throughout the rail network. drier habitats have been done in the Czech Republic (Havlin 1987), Chicago, USA (Heske 2015), and Spain (Malo et al. Taxa impacted by railways 2017), but no studies have been done in semi-arid and arid Amongst the groups most negatively affected by railways ecosystems (Borda-da-Àgua et al. 2017). Havlin (1987), in are small and less agile species, such as chelonians. a study in the Czech Republic, found that 96 bird species Although there are no data about the effects of railways on were associated with the tracks, and with some correlation chelonians in the Karoo, a study in India found that spotted between birds and the number of woody species along the pond turtles (Geoclamys hamiltonii) enter the tracks at tracks. The observations by Heske (2015) were on mortality places where the road is level with the tracks and are then of birds, mammals and reptiles in an urban nature reserve trapped, unable to climb over the rails (Dorsey et al. 2015). in Chicago and showed that mortality at the railway study This situation is likely to be present in the Karoo, where site, depending on taxa, was similar or significantly less farm entrance roads often cross and are level with the rails than mortality on roads. Although roads and railway lines (‘level crossings’). Large tortoises, such as Stigmochelys are similar corridors running through the landscape, railways pardalis, would probably be able to get over the tracks, but offer a different set of hazards and benefits to the biota. Psammobates, Chersobius and Homopus species are too Some possible reasons for lower mortality rates on railways small to climb over the overhanging lip on rails. are that rail traffic is far less frequent than road traffic, the As already pointed out, birds interact with railway lines noisier approach and vibrations of the rails may give some primarily through the power lines, and not so much with the warning to animals, and the narrower track of railway lines tracks themselves. Collisions with railway power lines have decreases the risk when crossing the tracks. Some smaller been reported by Shaw (2013), but without any estimate of animals can become trapped between the lines, and thus frequency. As with roads, crows appear to have made use cannot take any avoiding action, however (Dorsey et al. of the railway lines for food (Siegfried 1963) and nesting 2015). Lower mortality rates also reduce the number of opportunities. Other bird species, and bats, also gain potential scavengers, reducing the probability that the nesting and roosting sites associated with rock cuttings and scavengers themselves would be killed. bridge infrastructure. As with roads, the vegetation in the Possible benefits to mammals and birds of railways arise rail reserve is often quite different to that of the surrounds, from ‘track kills’ providing food for scavengers, and grain and these areas are often beneficial to both native and spills, providing food for granivores, although currently invasive species (see section above on Impacts). Havlin’s agricultural products appear to be largely transported (1987) study in the Czech Republic, in which he found across the Karoo by road rather than trains (Havenga and gallinaceous birds to be most common on the railway, could Pienaar 2012; Anon 2014). be paralleled in the Karoo if there are grain spills, although Apart from collisions with animals, overhead power lines this is possibly unlikely, given the move from rail to road for electric trains can electrocute birds, but also provide transport over the last two decades. steel structures for nests and shelter. Cuttings through bed rock and through deep soils on humps may be used by Mitigation and research to reduce the impact of railway lines birds that nest in rock crevices or by birds that tunnel into Given that many of the challenges posed to biodiversity by sandbanks. Another possible benefit to the avifauna is the railways are similar to those presented by roads, many of availability of nest sites for birds and diurnal roost sites for the mitigating measures and fields for future research would bats in the steel lattice girder bridges over railway tracks in be the same as those presented for roads. In addition, South Africa (see examples in Walters 2014). The space mitigation measures would need to include systems to between the rail track and the adjacent fences (‘rail reserve’) enable animals trapped between the rails to escape, and is relatively undisturbed and provides food and shelter for these would also require some careful design and testing to invertebrates, micro- and small mammals and birds. The maximise their effectiveness. vegetation, sometimes distinctive in these rail verges, may also provide refuges for plant species threatened by Canals grazing, and some species assemblages in the rail verges The only canalised water-way in the Karoo is the ‘Orange- have now formed new alliances and communities (Cilliers Fish River Tunnel’ (OFRT) leading water 83 km from the and Bredenkamp 1998). However, rail reserves (and railway Gariep Dam in the Northern Cape province to the Fish stations) are also places where invasive alien vegetation River Irrigation Scheme in the Eastern Cape province can establish (Cilliers and Bredenkamp 1998). (Cambray et al. 1986). This can be considered as a linear Carbon compounds, and other environmental pollutants structure that has similar characteristics as roads and from diesel train exhaust fumes, as well as dust from railway lines. On a smaller scale, a number of Karoo towns freight and the rolling stock, may disadvantage plants and have ‘lei’ water systems, i.e. water furrow systems within animals in the rail reserve (Lucas et al. 2017). There have villages. This water is usually fed from a town reservoir or 228 Dean, Seymour and Joseph dam, and channelled to properties in the town through a APLIC 2012), with the dangers to birds influenced by both system of furrows, creating a fairly reliable source of water design and voltage of the power line structure (Jenkins et for biota in villages. al. 2010). Power lines can also be positive for some species of birds, offering elevated nest, perching and roosting sites, Impacts of canals particularly for raptors, in otherwise largely treeless areas Most of the OFRT is covered, and apart from the initial such as the Karoo (Kemp 1985; Anon 1993; Boshoff 1993). damage to the environment during the construction phase, impacts of the tunnel have been on fish and invertebrate Taxa affected by power lines species assemblages and distributions (see below). There Cables on power lines (and to some extent pylons) are does not appear to have been any research conducted on hazardous for large flying birds, particularly those that the effects of lei water on biota in Karoo villages, but it is are blind or have a blind spot in the direction of travel possible that the water provides resources for species such (Martin and Shaw 2010). Birds are electrocuted when they as birds, attracting them to villages, where they may or physically bridge the air gap between live and earthed parts may not be prone to increased predation by domestic cats. of a power line structure (van Rooyen 2000; Lehman et al. Standing water may also see increases in invertebrates 2007). Large species, such as eagles, vultures, storks and such as mosquitoes. herons, are capable of perching on power structures in such a way as to bridge the gap; smaller species, such as crows Taxa affected by canals and kestrels, that commonly use power lines as roosts, Once the OFRT was completed, subsequent impacts of perches or nest sites are seldom electrocuted, although the tunnel have been on fish and invertebrate species two of nine electrocutions recorded by Shaw et al. (2010) assemblages and distributions (Cambray and Jubb 1977; were Cape Crows. Large species of birds rather than small Petitjean and Davies 1986; Snaddon 1998). Nothing on the birds appear to be more at risk from colliding with power impact of the OFRT on other taxa has been published, but lines (Shaw et al. 2010) and mortalities may be species- there are anecdotal accounts of the mortality of mammals specific (Janss 2000). The largest number of collision trapped in the uncovered parts of the canal. mortalities in the Karoo were by three species of large birds (total number of carcasses recovered in brackets): Ludwig’s Mitigation and research into measures to minimise impacts ( ludwigii) (468), Kori Bustard ( kori) of canals (63) and Karoo Korhaan ( vigorsii) (42) (Shaw The large OFRT has had impacts on land fauna but the 2013). There is some uncertainty about whether mortalities main groups affected by interbasin transfer (IBT) are caused by power line collisions are causing a decline in aquatic. The nature of the aquatic environment makes the Ludwig’s Bustard population – the authors could not mitigation for the negative consequences of IBT difficult, conclude that there had been a change in population size and a recent review by Zhuang (2016) suggests that the between the counts done in the 1980s and the counts done ecological risks of IBTs are so great that it would be better in the 2010s, but nevertheless suggested that its Red List to avoid them altogether, rather focusing on water use status should remain at Endangered (Shaw et al. 2016). efficiency, rainwater harvesting and desalination of sea The extent of power line pylons (and the availability of water as solutions to water shortages. For lei water canals, pylons) in South Africa, including the Karoo, is shown by there appears to be no research to date on the effects Madden (2013). Nests of White-backed Vultures (Gyps of these on biota, but this could be an interesting area to africanus) and (Polemaetus bellicosus) have investigate. been found on pylons in the Karoo (Dean 1975; Boshoff 1993; Anderson and Hohne 2007), and Pied Crows Power lines frequently nest on pylons in this region (Madden 2013). The interactions between power lines and animals have been well studied in southern Africa and in many other parts Mitigating measures and research gaps of the world, including the Karoo (Ledger and Annegarn Nest sites provided by power line pylons pose threats 1981; Bevanger 1994; Ledger and Hobbs 1999; Jenkins et to rare and endangered species in the Karoo. Research al. 2010, 2011; Shaw 2013; Shaw et al. 2016, 2017). There into measures to make the poles supporting power are approximately 350 000 km of power lines in South Africa lines unattractive as nesting and roosting sites would (Shaw 2013) and both Shaw and Rudman et al. (2017) likely be useful. provide maps showing the main transmission lines across A recent, comprehensive review of the literature highlights the Karoo. factors that influence collision risk for birds with power lines, covering an array of issues, from bird behaviour to habitat Effects of power lines on biota characteristics and climatic variables (Bernardino et al. Although the extent and coverage of power lines across 2018). Recommendations for reducing power-line collisions the Karoo are relatively minor compared with other parts of include route planning, power-line marking, changes to South Africa, there is nevertheless some cause for concern. electric wire configuration, and habitat management, but The costs and benefits of power line structures have been underground cabling is the only solution that would avoid summarised by D’Amico et al. (2018) and others including collisions altogether (Bernardino et al. 2018). Underground Shaw (2013) and Shaw et al. (2017). Power lines, pylons cabling is highly unlikely in South Africa, and in the Karoo in and associated structures, such as stay wires anchoring particular, given the vast distances that such cabling would pylons, are a major source of mortality in birds (Janss 2000; need to traverse, and the other pressing socio-economic African Journal of Range & Forage Science 2018, 35(3&4): 223–232 229 imperatives for the South African fiscus. For research Anon. 1993. Martials, pylons and the Karoo. Gabar 8: 43. gaps, again, Bernardino et al. (2018) have compiled a Anon. 2003. Game ranch profitability in southern Africa. Rivonia: comprehensive list of issues for research, many of which SA Financial Sector Forum. are germane to the Karoo. These include assessment of Anon. 2014. Freight shift from road to rail. Unpublished report. Pretoria: Department of Environmental Affairs. the effectiveness of different devices, including ‘glow-in- Anon. 2016. Transnet integrated report 2016. Pretoria: Department the-dark’ devices, optimal spacing of devices to alert of Transport. birds to the presence of power lines and acoustic scaring APLIC (Avian Power Line Interaction Committee). 2012. Reducing (Bernardino et al. 2018). Although optimal configuration and avian collisions with power lines: the state of the art in 2012. routing would also be useful to assess, again, in a country Washington, DC: Edison Electric Institute and APLIC. such as South Africa with limited resources, it is likely that Austin GT. 1971. Roadside distribution of the common raven in the there are few options for routing power lines in a way that is Mojave Desert. California Birds 2: 98. both economically and ecologically favourable. Beck A. 2010. Electric fence induced mortality in South Africa. MSc thesis, University of the Witwatersrand, South Africa. Conclusions Ben-Shahar R. 1993. Does fencing reduce the carrying capacity for populations of large herbivores? Journal of Tropical Ecology 9: 249–253. The increasing trend in establishing game farms (more Bernardino J, Bevanger K, Barrientos R, Dwyer JF, Marques AT, fences), more traffic and more roads in the Karoo, is Martins RC, Shaw JM, Silva JP, Moreira F. 2018. Bird collisions likely to have increasing negative effects on the biota with power lines: state of the art and priority areas for bird of the Karoo. Compounding this is the need to move in collisions with power lines. Biological Conservation 222: 1–13. response to climate change, and whilst fences may be Bevanger K. 1994. Bird interactions with utility structures: collision inhibiting this movement, other linear structures, such and electrocution, causes and mitigating measures. Ibis 136: as roads and power lines (Cunningham et al. 2015) are 412–425. facilitating it. Although there are small benefits to the biota, Boone RB, Hobbs NT. 2004. Lines around fragments: effects of the ‘goods and services’ caused by processes associated fencing on large herbivores. African Journal of Range and Forage Science 21: 147–158. with most linear structures is poor compensation for what Borda-de-Água L, Barrientos R, Beja P, Pereira HM. 2017. Railway has been lost in the past and likely to be lost in the future. ecology. Cham: Springer Nature. There are many gaps in our knowledge of the impacts of Boshoff AF. 1993. Density, breeding performance and stability of fences, roads, railways and power lines on populations Martial Eagles Polemaetus bellicosus breeding on electricity of animals, and attempts to close the gaps should be a pylons in the Nama Karoo, South Africa. In: Wilson RT (ed.), focus for research. Estimates of mortalities per kilometre Proceedings of the 8th Pan African Ornithological Congress: per year per different type of linear structure, for different birds and the African environment. Tervuren: Royal Museum for taxa, along with population modelling to enable assess- Central Africa. pp 95–104. ments of the size of the effects of mortalities, should be Brandt F, Spierenburg M. 2014. Game fences in the Karoo: made, as well as research into the relative effectiveness of reconfiguring spatial and social relations. Journal of Contemporary African Studies 32: 220–237. mitigating measures to reduce mortalities. Research on the Bullock KL, Malan G, Pretorius MD. 2011. Mammal and bird road indirect effects of these structures on ecological processes mortalities on the Upington to Twee Rivieren main road in the is sorely needed; linear structures are not all-bad – verges southern Kalahari, South Africa. African Zoology 46: 60–71. along roads and railway lines may support the only popula- Burger M, Branch WR. 1994. Tortoise mortality caused by electric tions of certain plant taxa (Esler and Milton 2006), providing fences in the Thomas Baines Nature Reserve. South African seed banks and input into animal populations, although the Journal of Wildlife Research 24: 32–37. other side of the fence may just be a sink. Much depends Cambray JA, Davies BR, Ashton PJ. 1986. The Orange-Vaal river on careful management of the biota associated with linear system. In: Davies BR, Walker KF (eds), The ecology of river structures – carefully managed power-line corridors in systems. Monographiae Biologicae 60. Dordrecht: Dr W Junk. Australia (Clarke et al. 2006) and carefully managed road pp 89–112. Cambray JA, Jubb RA. 1977. Dispersal of fishes via the Orange- verges in the Karoo (Esler and Milton 2006) have shown Fish Tunnel, South Africa. Journal of the Limnological Society of that it is possible to include biodiversity considerations southern Africa 3: 33–35. into practical measures. Keeping the biodiversity is the CapeNature. 2015. Fencing and enclosure of game and predators challenge for the future. in the Western Cape province. Resolution no. 18 of 2014/2015. Cassidy L, Fynn R, Sethebe B. 2013. Effects of restriction of wild Acknowledgements – We thank Lance van Sittert for references herbivore movement on woody and herbaceous vegetation in to fencing, and we thank two anonymous reviewers for their the Okavango Delta Botswana. African Journal of Ecology 51: suggestions for improving the presentation of results and general 513–527. comments on the text. Cilliers SS, Bredenkamp GJ. 1998. Vegetation analsis of railway reserves in the Potchefstroom municipal area, North West References province, South Africa. South African Journal of Botany 64: 271–280. Anderson MD, Hohne P. 2007. African White-backed Vultures Clarke DJ, Pearce KA, White JG. 2006. Powerline corridors: nesting on electricity pylons in the Kimberley area, Northern degraded ecosystems or wildlife havens? Wildlife Research 33: Cape and Free State provinces, South Africa. Vulture News 57: 615–626. 44–50. Clevenger AP, Chruszez B, Gunson KE. 2001. Highway mitigation Andrews A. 1990. Fragmentation of habitat by roads and utility fencing reduces wildlife-vehicle collisions. Wildlife Society corridors: a review. Australian Zoology 26: 130–141. Bulletin 29: 646–653. 230 Dean, Seymour and Joseph

Clevenger AP, Chruszez B, Gunson KE. 2003. Spatial patterns and Railway ecology. Cham: Springer Nature. pp 103–116. factors influencing small vertebrate fauna road-kill aggregations. Godinho C, Catarino L, Marques JT, Mira A, Beja P. 2017b. Biological Conservation 109: 15–26. Assessing bird exclusion effects in a wetland crossed by a railway Coffin AW. 2007. From roadkill to road ecology: A review of the (Sado Estuary, Portugal). In: Borda-de-Água L, Barrientos R, Beja ecological effects of roads. Journal of Transport Geography 15: P, Pereira HM (eds), Railway ecology. Cham: Springer Nature. 396–406. pp 179–196. Collinson W, Patterson-Abrolat C. 2016. The road ahead: guidelines Havenga JH, Pienaar WJ. 2012. The creation and application of a to mitigation methods to address wildlife road conflict in South national freight flow model for South Africa. Journal of the South Africa. Johannesburg: Endangered Wildlife Trust. African Institution of Civil Engineering 54: 2–13. Collinson W, Parker D, Patterson-Abrolat C, Alexander G, Davies- Havlin J. 1987. On the importance of railway lines for the life of Mostert H. 2015. Setjhaba SA, South Afrika: a South African avifauna in agrocoenoses. Folia Zoologica 364: 345–358. perspective of an emerging transport infrastructure. In: van der Hayward MW, Kerley GIH. 2009. Fencing for conservation: Ree R, Smith DJ, Grilo C (eds), Handbook of road ecology. restriction of evolutionary potential or a riposte to threatening Oxford: John Wiley and Sons. pp 439–447. processes? Biological Conservation 142: 1–13. Conner RN, Adkisson CS. 1976. Concentration of foraging common Heard HW, Stephenson A. 1987. Electrification of a fence to control ravens along the Trans-Canada Highway. Canadian Field the movements of black-backed jackals. South African Journal of Naturalist 90: 496–497. Wildlife Research 17: 20–24. Cunningham SJ, Madden CF, Barnard P, Amar A. 2015. Electric Heske EJ. 2015. Blood on the tracks: track mortality and scavenging crows: powerlines, climate change and the emergence of a rate in urban nature preserves. Urban Naturalist 4: 1–13. native invader. Diversity and Distributions 22: 17–29. Hoare RE. 1992. Present and future use of fencing in the manage- D’Amico M, Catry I, Martins RC, Ascensão F, Barrientos R, Moreira ment of larger African mammals. Environmental Conservation F. 2018. Bird on the wire: landscape planning considering costs 19: 160–164. and benefits for bird populations coexisting with power lines. Hockey PAR, Dean WRJ, Ryan PG (eds). 2005. Roberts birds Ambio 47: 650–656. of southern Africa (7th edn). Cape Town: Trustees of the John Dean WRJ. 1975. Martial Eagles nesting on high tension pylons. Voelcker Bird Book Fund. Ostrich 46: 116. Janss GFE. 2000. Avian mortality from power lines: a morphologic Dean WRJ, Milton SJ (eds). 1999. The Karoo, ecological patterns approach of a species-specific mortality. Biological Conservation and processes. Cambridge: Cambridge University Press. 95: 353–359. Dean WRJ, Milton SJ. 2000. Directed dispersal of Opuntia species Jenkins AR, Shaw JM, Smallie JJ, Gibbons B, Visagie R, Ryan PG. in the Karoo, South Africa: are crows the responsible agents? 2011. Estimating the impacts of power line collisions on Ludwig’s Journal of Arid Environments 45: 305–314. Bustards Neotis ludwigii. Bird Conservation International 21: Dean WRJ, Milton SJ. 2003. The importance of roads and road 303–310. verges for raptors and crows in the Succulent and Nama-Karoo, Jenkins AR, Smallie JJ, Diamond M. 2010. South African perspec- South Africa. Ostrich 74: 181–186. tives on a global search for ways to prevent avian collisions with Dean WRJ, Milton SJ, Anderson MD. 2005. Use of road kills and overhead lines. Bird Conservation International 20: 263–278. roadside vegetation by Pied and Cape Crows in semi-arid South Joseph GS, Seymour CL, Foord SH. 2017. The effect of infrastruc- Africa. Ostrich 77: 102–104. ture on the invasion of a generalist predator: pied crows in Desmet P, Cowling R. 1999. The climate of the Karoo – a functional southern Africa as a case-study. Biological Conservation 205: approach. In: Dean WRJ, Milton SJ (eds), The Karoo: ecological 11–15. patterns and processes. Cambridge: Cambridge University Press. Kannemeyer L. 2016. South African road network condition, needs pp 3–16. and funding. The South African National Roads Agency Report. Dorsey B, Olsson M, Rew LJ. 2015. Ecological effects of railways Available at https://sarf.org.za/wp-content/uploads/2016/08/ on wildlife. In: van der Ree R, Smith DJ, Grilo C (eds), Handbook Louw-Kannemeyer.pdf. of road ecology. Oxford: John Wiley and Sons. pp 219–227. Kemp AC. 1985. Pylons, crows and raptors. In: Bunning LJ Eloff PJ, van Niekerk A. 2005. Game, fences and motor vehicle (ed.), Proceedings of the Birds and Man Symposium held in accidents: spatial patterns in the Eastern Cape. South African Johannesburg, 10–15 April 1983. Johannesburg: Witwatersrand Journal of Wildlife Research 35: 125–130. Bird Club. p 166. Esler KJ, Milton SJ. 2006. Towards best practice in management of Kesch KM, Bauer DT, Loveridge AJ. 2014. Undermining game road, power line and rail reserves. CIB Occasional Paper no. 1. fences: who is digging holes in Kalahari sands? African Journal Stellenbosch: Centre for Invasion Biology, Stellenbosch University. of Ecology 52: 144–150. Fincham J, Lambrechts N. 2014. How many tortoises do a pair Knight RL, Kawashima JY. 1993. Responses of raven and redtailed of pied crows Corvus alba need to kill to feed their chicks? hawk populations to linear right-of-ways. Journal of Wildlife Ornithological Observations 5: 135–138. Management 57: 266–271. Fincham JE, Visagie R, Underhill LG, Brooks M, Markus MB. 2015. Knight RL, Knight HAL, Camp RJ. 1995. Common ravens and The impacts of the Pied Crow Corvus albus on other species number and type of linear rights-of-way. Biological Conservation need to be determined. Ornithological Observations 6: 232–239. 74: 65–67. Forman RTT, Sperling D, Bissonette JA, Clevenger AP, Cutshall Ledger JA, Annegarn HJ. 1981. Electrocution hazards to the CD, Dale VH, Fahrig L, France R, Goldman CR, Heanue K, Cape Vulture (Gyps coprotheres) in South Africa. Biological Jones JA, Swanson FJ, Turrentine T, Winter TC. 2003. Road Conservation 20: 15–24. ecology: science and solutions. Covelo: Island Press. Ledger JA, Hobbs JCA. 1999. Raptor use and abuse of power lines Geneletti D. 2006. Some common shortcomings in the treatment in southern Africa. Journal of Raptor Research 33: 49–52. of linear structures on natural habitat. Environmental Impact Lee E, Croft DB, Achiron-Frumkin T. 2015. Roads in the arid lands: Assessment Review 26: 257–267. issues, challenges and potential solutions. In: van der Ree R, Godinho C, Marques JT, Salgueiro P, Catarino L, Osório de Castro Smith DJ, Grilo C (eds), Handbook of road ecology. Oxford: John C, Mira A, Beja P. 2017a. Bird collisions in a railway crossing Wiley and Sons. pp 382–390. a wetland of international importance (Sado Estuary, Portugal). Lehman RN, Kennedy PL, Savidge JA. 2007. The state of the In: Borda-de-Água L, Barrientos R, Beja P, Pereira HM (eds), art in raptor electrocution research: a global review. Biological African Journal of Range & Forage Science 2018, 35(3&4): 223–232 231

Conservation 136: 159–174. 15: 921–938. Lightfoot DG, Whitford WG. 1991. Productivity of creosotebush Peterson MS. 2009. Letter from South African National Roads foliage and associated canopy arthropods along a desert Agency Limited in response to a query about mowing of road roadside. American Midland Naturalist 125: 310–322. verges. Ref #401890v1, SANRAL Southern Region, Port Lucas PS, Gomes de Carvalho R, Grilo C. 2017. Railway Elizabeth. Disturbances on wildlife: types, effects, and mitigation measures. Petitjean MOG, Davies BR. 1988. Ecological impacts of inter-basin In: Borda-de-Água L, Barrientos R, Beja P, Pereira HM (eds), water transfers: some case studies, research requirements and Railway ecology. Cham: Springer Nature. pp 81–99. assessment procedures in southern Africa. South African Journal Macdonald I. 2005. Use of electrical fencing in agriculture. [s.l.]: of Science 84: 819–828. KwaZulu-Natal Department of Agriculture and Environmental Pointing SB, Belnap J. 2014. Disturbance to desert soil ecosystems Affairs. contributes to dust-mediated impacts at regional scales. Macdonald IAW, Macdonald SA. 1985. The demise of the Biodiversity and Conservation 23: 1659–1667. solitary scavengers in southern Africa – the early rising crow Province of the Northern Cape. 2011. Northern Cape Nature hypothesis. In: Bunning LJ (ed.), Proceedings of the Birds and Conservation Act No. 9 of 2009 (2011). Province of the Northern Man Symposium held in Johannesburg, 10–15 April 1983. Cape, Provincial Gazette Extraordinary no. 1566, 19 December Johannesburg: Witwatersrand Bird Club. pp 321–335. 2011, vol. 18. Proclamation No. 10, 2011, General Notice of Macray MB. 2017. Tortoise mortalities along fences in the south- 2011 and Annexure 1. eastern Karoo, South Africa. MSc thesis, University of Cape Rahlao SJ, Milton SJ, Esler SJ, Barnard P. 2010. The distribution Town, South Africa. of invasive Pennisetum setaceum along roadsides in western Madden CF. 2013. The impacts of corvids on biodiversity. MSc South Africa: the role of corridor interchanges. Weed Research thesis, University of Cape Town, South Africa. 50: 537–543. Malan G. 1992. Nest-lining used by pale chanting goshawks in the Roche CJ. 2004. “Ornaments of the desert”: springbok treks in the Little Karoo, South Africa. Gabar 7: 56–61. Cape Colony, 1774-1908. MA thesis, University of Cape Town, Malan G. 1994. Distribution of nests of the pied crow Corvus albus South Africa. on telephone poles in the Upper Karoo. Ostrich 65: 44–45 Rudman J, Gauché P, Esler KJ. 2017. Direct environmental Malo JE, García de la Morena EL, Hervás I, Cristina Mata C, impacts of solar power in two arid biomes: an initial investigation. Herranz J. 2017. Cross-scale changes in bird behavior around a South African Journal of Science 113: Art. #2017-0113. high speed railway: from landscape occupation to infrastructure Santos SM, Carvalho F, Mira A. 2017. Current knowledge on use and collision risk. In: Borda-de-Água L, Barrientos R, Beja wildlife mortality in railways. In: Borda-de-Água L, Barrientos P, Pereira HM (eds), Railway ecology. Cham: Springer Nature. R, Beja P, Pereira HM (eds), Railway ecology. Cham: Springer pp 117–134. Nature. pp 11–22. Martin GR, Shaw JM. 2010. Bird collisions with power lines: failing Scholes B, Lochner PA, Schreiner G, Snyman-Van der Walt L, to see the way ahead? Biological Conservation 143: 2695–2702. de Jager M (eds). 2016. Shale gas development in the Central Meunier FD, Verheyden C, Jouventin P. 2000. Use of roadsides by Karoo: a scientific assessment of the opportunities and risks. diurnal raptors in agricultural landscapes. Biological Conservation Report no. CSIR/IU/021MH/EXP/2016/003/A. Stellenbosch: CSIR. 92: 291–298. Shaw JM. 2013. Power line collisions in the Karoo: conserving Milton SJ, Dean WRJ. 1998. Alien plant assemblages near roads Ludwig’s Bustard. PhD thesis, University of Cape Town, South in arid and semi-arid South Africa. Diversity and Distributions 4: Africa. 175–188. Shaw JM, Jenkins AR, Ryan PG, Smallie JJ. 2010. A preliminary Milton SJ, Dean WRJ. 2010. Plant invasions in arid areas: special survey of avian mortality on power lines in the Overberg, South problems and solutions: a South African perspective. Biological Africa. Ostrich 81: 109–113. Invasions 12: 3935–3948. Shaw JM, Jenkins AR, Allan DG, Ryan PG. 2016. Population size Milton SJ, Dean WRJ, du Plessis MA, Siegfried WR. 1994. A and trends of Ludwig’s Bustard Neotis ludwigii and other large conceptual model of arid rangeland degradation: the escalating terrestrial birds in the Karoo, South Africa. Bird Conservation cost of declining productivity. BioScience 44: 70–76. International 26: 69–86. Milton SJ, Dean WRJ, Sielecki LE, van der Ree R. 2015. The Shaw JM, Reid TA, Schutgens M, Jenkins AR, Ryan PG. 2017. function and management of roadside vegetation. In: van der High power line collision mortality of threatened bustards at a Ree R, Smith DJ, Grilo C (eds), Handbook of road ecology. regional scale in the Karoo, South Africa. Ibis 160: 431–446. Oxford: John Wiley and Sons. pp 374–381. Shilling F, Perkins SE, Collinson W. 2015. Wildlife/roadkill observa- Morelli F, Beim M, Jerzak L, Jones D, Tryjanowski P. 2014. Can tion and reporting systems. In: van der Ree R, Smith DJ, Grilo C roads, railways and related structures have positive effects on (eds), Handbook of road ecology. Oxford: John Wiley and Sons. birds? – A review. Transportation Research Part D 30: 21–31. pp 492–501. Mucina L, Rutherford MC (eds). 2006. Vegetation of South Africa, Siegfried WR. 1963. A preliminary evaluation of the economic Lesotho and Swaziland. Strelitzia 19. Pretoria: South African status of Corvidae and their control on sheep farms in the Great National Biodiversity Institute. Karoo. Cape Department of Nature Conservation Investigational Múrias T, Gonçalves D, Lopes RJ. 2017. Evaluating the impacts Report 4. Cape Town: Department of Nature Conservation. of a new railway on shorebirds: a case study in central Portugal Siegfried WR. 1999. Human impacts. In: Dean WRJ, Milton SJ (Aveiro Lagoon). In: Borda-de-Água L, Barrientos R, Beja P, (eds), The Karoo: ecological patterns and processes. Cambridge: Pereira HM (eds), Railway ecology. Cham: Springer Nature. Cambridge University Press. pp 239–241. pp 197–214. Snaddon CD. 1998. Some of the ecological effects of a small Nattrass N, Conradie B, Drouilly M, O’Riain MJ. 2017. A brief history inter-basin water transfer on the receiving reaches of the upper of predators, sheep farmers and government in the Western Berg River, Western Cape. MSc thesis, University of Cape Town, Cape, South Africa. CSSR Working Paper no. 398. Cape Town: South Africa. Centre for Social Science Research, University of Cape Town. Spellerberg IF. 2002. Ecological effects of roads. Enfield, NH: O’Farrell P, Milton SJ. 2006. Road verge and rangeland plant Science Publishers. communities in the southern Karoo: exploring what influences Talbot WJ. 1961. Land utilization in the arid regions of southern diversity, dominance and cover. Biodiversity and Conservation Africa. Part I: South Africa. In: Stamp LD (ed.), A history of land 232 Dean, Seymour and Joseph

use in arid regions. Arid Zone Research 17. Paris: UNESCO. van Sittert L. 2002. Holding the line: the rural enclosure movement pp 299–338. in the Cape Colony, c. 1865–1910. Journal of African History 43: Trombulak SC, Frissell CA. 2000. Review of ecological effects of 95–118. roads on terrestrial and aquatic communities. Conservation Vanak AT, Thaker M, Slotow R. 2010. Do fences create an Biology 14: 18–30. edge-effect on the movement patterns of a highly mobile Tshiguvho TE, Dean WRJ, Robertson HG. 1999. Conservation mega-herbivore? Biological Conservation 143: 2631–2637. value of road verges in the semi-arid Karoo, South Africa: ants Walker C, Milton SJ, O’Connor T, Maguire JM, Dean WRJ. 2018. (Hymenoptera: Formicidae) as bio-indicators. Biodiversity and Drivers and trajectories of social and ecological change in the Conservation 8: 1683–1695. Karoo, South Africa. African Journal of Range and Forage van den Bosch R, Telford AD. 1964. Environmental modification Science 35: 157–177. and biological control. In: de Bach P, Schlinger EI (eds), Walters D. 2014. Bridging the Eastern Cape: the life and work of Biological control of insect pests and weeds. London: Chapman Joseph Newey. East London: Coral Tree Press. and Hall. pp 459–488. Williams RD, Colson EW. 1989. Raptor associations with linear van der Merwe P, Saayman M. 2003. Determining the economic rights-of-way. In: Pendleton BG (ed.), Proceedings of the value of game farm tourism. Koedoe 46: 103–112. Western Raptor Management Symposium and Workshop. van der Ree R, Smith DJ, Grilo C (eds). 2015. Handbook of road Washington, DC: National Wildlife Federation. pp 173–192. ecology. Oxford: John Wiley and Sons. Winterbottom JM. 1975. Notes on the South African species of van Rooyen CS. 2000. Raptor mortality on powerlines in South Corvus. Ostrich 46: 236–250. Africa. In: Chancellor RD, Meyburg B-U (eds), Raptors at risk: Woodroffe R, Hedges S, Durant SM. 2014. To fence or not to proceedings of the V World Conference on Birds of Prey and fence. Science 344: 46–48. Owls, Midrand, Johannesburg, South Africa, 4–11 August 1998. Zhuang W. 2016. Eco-environmental impact of inter-basin water Berlin: World Working Group on Birds of Prey and Owls; Surrey, transfer projects: a review. Environmental Science Pollution BC: Hancock House Publishers. pp 739–749. Research 23: 12867–12879.

Received 26 July 2018, revised 8 August 2018, accepted 14 August 2018 Associate Editor: Tim O’Connor