Impacts of wind energy developments on wildlife: a southern hemisphere perspective

Wind energy is seen as a sustainable alternative to electricity generation methods that produce greenhouse gases, or are perceived by the public as being unsafe. Globally, installed wind capacity has risen year-on-year from 6.1 GW in 1996 to 238.3 GW in 2011 (GWEC 2011). In 2011, 42 times more generation capacity was added compared with 1996. Europe leads the world in wind energy generation with over 96 GW installed capacity. The greatest users are the Danes with 26% of annual electricity demand met by wind power. However, growth is strongest in Asia with China installing 26% of the world's new wind energy generation in 2011. Despite this rapid growth, wind energy only represents 2.5% of current global generation capacity. New Zealand has also embraced wind energy, producing 623 MW at the end of 2011, or about 4% of the country's annual generation (NZWEA 2012). This capacity comes from 16 operating wind farms located from Waikato to Southland, and east to the Chatham Islands. A further 26 facilities are either consented or have sites under investigation. The industry plans to increase generation to 3.5 GW by 2030, representing 20% of total generation capacity (NZWEA 2012).

Despite its potential for reducing greenhouse gas emissions and dependence on non-renewable energy sources, wind energy has been associated with negative effects on bats and birds, generally through collisions with turbines or other structures such as electricity transmission lines. Collisions between birds and turbines are perhaps the principle concern about the environmental impacts of wind farms. Birds are common, mobile, often migratory, and may fly at heights that put them at risk of collision with turbine blades. Some early wind farms were built at sites with high numbers of slow-flying raptors, which suffered high collision rates (most famously Altamont Pass in the USA; Smallwood & Thelander 2008), but more judicious placement of wind farms and use of larger but slower-turning turbines (Smallwood & Karas 2009) seem to have reduced mortality rates at newer farms. Nevertheless, birds continue to collide with turbines globally, and debate remains about the ability of birds to detect and avoid turbines under different weather conditions, and how bird movements through a wind farm translate into collision rates (e.g. Everaert & Stienen 2007; de Lucas et al. 2008; Martin & Shaw 2010).

Effects of wind turbines on bats first came to the notice of researchers in the mid-1990s when a study monitoring bird fatalities at a facility in Minnesota reported 13 bat fatalities in the first two years of operation (Osborn et al. 1996). Since then a number of other studies have investigated bat mortalities. The number of bats killed at wind farms varies widely. In the USA, 0.8–41.1 bats MW^-1 yr^-1 are killed (Johnson et al. 2003 and Fiedler et al. 2007, cited in Kunz et al. 2007a). Bats most at direct risk are migratory, tree roosting species (Kunz et al. 2007b) and account for 75% of fatalities at North American wind farms. Non-migratory species most at risk are fast-flying, open-air foragers that are generally less manoeuvrable (Arnett 2008; Natural England 2009). Injury or death results either through direct collisions with turning blades, or through barotrauma (tissue damage caused by a rapid change in air pressure; Baerwald et al. 2008). It is still not known why bats collide with turbine towers or blades. Current theories include bats being attracted to the turbines (due to the turbines, or their locations on ridge-tops, attracting insects or because of the ultrasound they produce), their placement along the migratory pathways of bats or because bats may perceive towers as potential roost trees (Kunz et al. 2007a). When combined with other negative anthropogenic effects on bats, such as habitat modification or clearance and disease, losses caused by wind turbines could negatively impact long-term population trends (Blehert et al. 2009; Borkin et al. 2011).

It is clear that wind energy development does have environmental costs, and this fact has had a negative effect on scientists' and the public's perception of the industry. A valid question, however, is whether the effects of wind energy generation are greater than those of other industries or activities. It has been estimated that in the USA, between 100,000 and 400,000 birds are killed by wind turbines each year. This compares well with buildings (100,000 to 1 million), agricultural pesticides (67 to 90 million) and domestic cats (365 million to 1 billion) (Subramanian 2012). In New Zealand, more blackbirds (Turdus merula) are likely killed on urban roads each year than at wind farms. Data extracted from the Civil Aviation Authority of New Zealand (2012) and Ministry of Transport (2010) show that, in 2010, 180 birds were killed by aircraft ‘on-aerodrome’ in the Auckland region. Smaller airports with fewer aircraft movements, such as Dunedin and Napier, recorded 49 and 115 likely deaths, respectively, in the same year. For comparison, Bull et al. (this issue) estimated collision rates of 5.8 and 4.6 birds per turbine at Project West Wind, Wellington, with overall annual mortality estimates of 363 and 289 birds over two years of monitoring. Despite the seemingly lesser effects of wind farms compared with other activities, researchers must also be aware of exactly which species are being harmed (Subramanian 2012). Often wind farms affect species that are slow reproducers (e.g. raptors), are already endangered (e.g. Californian condor [Gymnogyps californianus] and whooping cranes [Grus Americana]), or face other threats such as disease (e.g. white nose syndrome in bats). In New Zealand, the west coast of the North Island provides excellent wind conditions for electricity generation, but the major shorebird flyway in the country also runs along this coast (Fuller et al. 2009), putting vulnerable species such as wrybill (Anarhynchus frontalis) at potential risk.

At this stage it is difficult to make robust comparisons between wind farms and other sources of direct human impacts on wildlife in New Zealand, in part because there are so few data collected in standardised ways. The wind energy sector is, however, subject to regulatory processes that demand formal ecological assessments, and these have the potential to greatly advance our knowledge of the potential and actual impacts of wind farms in New Zealand.

There are three main parts of the regulatory system where scientists are (or should be) involved: pre-consent impact assessment; post-construction carcass monitoring; and (if present) monitoring of the effectiveness of any mitigation. Initial impact assessments may, depending on the proposed site and size of farm, involve detailed modelling of collision risk and require substantial investments of field data and radar work. Accurate carcass monitoring requires survey methods that are experimentally validated and appropriate in duration and scale. Mitigation for predicted kills may occur through local habitat enhancement or, in the case of migrants, through predator control at distant sites (e.g. braided rivers for endemic migrant shorebirds). In all of these, it is up to biologists and biostatisticians to communicate to consultants, lawyers and hearing panels, that robust methods are needed to evaluate the true risks and costs of wind farm developments. This job should become easier as more information about the New Zealand situation comes to light.

Application of best practice processes to the New Zealand situation, and indeed the situation in most countries worldwide, is hampered by a lack of basic understanding of the biology of the species at risk. This is exacerbated by uncertainty about how directly transferable overseas findings are to New Zealand bird species. Information on movement patterns, flight behaviour, population size, genetic structure, reproductive rates, etc, are fundamental to understanding, and thus mitigating for, potential effects of a wind facility. To achieve this aim industry must employ, directly or indirectly, researchers capable of gathering basic biological data that are both scientifically robust and applicable to their particular needs. In some situations, it may be that partnerships with universities provide a workable structure to channel expertise towards the wind energy industry (and there are international examples of doctoral-level projects associated with wind energy evaluations, e.g. Desholm 2006). Otherwise, universities can act as a feeder of well-trained graduates to the consulting world. Robust research now will reduce the cost of future developments by providing quantitative evidence of best practice accepted by the regulatory system, and by educating the next generation of biologists who are skilled in objective risk assessment.

The initial aim of this special issue was to provide a vehicle for the publication of studies on the effects of wind facilities in New Zealand. However, the quality, quantity and geographic distribution of studies submitted have given us the opportunity to widen the scope of the issue to encompass the southern hemisphere. This is opportune as few data exist from New Zealand, and we hope that the special issue will better inform decisions made for the hemisphere's unique species and conditions. In addition, publication of these studies will encourage others working on the effects of wind energy facilities to publish their work and thus contribute to the global information base on which regulatory decisions must be made.

Papers in this special issue cover a range of empirical and methodological issues to do with the assessment of wind farm impacts on bats and birds. Two are principally methodological. Seaton and Barea outline a framework for the assessment of risk to New Zealand falcon (Falco novaeseelandiae) which, if followed, would result in a cumulative build-up of knowledge about the risks to this high priority species nationally. Bernardino et al. discuss some of the issues and assumptions involved in the estimation of risk to bats and birds and the consequent limitations of predictions. The remaining four papers present quantitative data on collisions of birds and/or bats in New Zealand, Australia and South Africa. Bull et al. present the first data on bird strikes at a New Zealand wind farm. While they are only two years into a three-year monitoring programme, open dissemination of their data so far is a welcome addition to this volume. Hull and coworkers from Tasmania have contributed two papers. One asks whether there are specific taxonomic groups and/or foraging styles of Australian bird that are more prone to collision; the other looks at patterns of collision of bats, and the characteristics of the species involved. Finally, Doty and Martin document bat and bird mortality at a single pilot wind turbine in South Africa. It is inevitable that decisions about wind farms in New Zealand will be made largely on the basis of information derived overseas, but the opportunity is there for New Zealand researchers and consultants to publish their findings in the primary rather than the grey literature or in hearing submissions. It is therefore pleasing to be able to publish these papers in readily-accessible and searchable form. We hope that the papers in this special issue will serve as a starter for future submissions that will aid the decision-making about, and monitoring of, wind farms in New Zealand.