Copyright copy 2013 by the author(s) Published here under license by the Resilience AllianceZimmerling J R A C Pomeroy M V dEntremont and C M Francis 2013 Canadian estimate of birdmortality due to collisions and direct habitat loss associated with wind turbine developments AvianConservation and Ecology 8(2) 10 httpdxdoiorg105751ACE-00609-080210

Research Paper part of a Special Feature on Quantifying Human-related Mortality of Birds in Canada

Canadian Estimate of Bird Mortality Due to Collisions and DirectHabitat Loss Associated with Wind Turbine Developments

Estimation de la mortaliteacute aviaire canadienne attribuable aux collisions etaux pertes directes drsquohabitat associeacutees agrave lrsquoeacuteolienJ Ryan Zimmerling 1 Andrea C Pomeroy Marc V dEntremont and Charles M Francis 1

ABSTRACT We estimated impacts on birds from the development and operation of wind turbines in Canada considering bothmortality due to collisions and loss of nesting habitat We estimated collision mortality using data from carcass searches for 43wind farms incorporating correction factors for scavenger removal searcher efficiency and carcasses that fell beyond the areasearched On average 82 plusmn 14 birds (95 CI) were killed per turbine per year at these sites although the numbers at individualwind farms varied from 0 - 269 birds per turbine per year Based on 2955 installed turbines (the number installed in Canada byDecember 2011) an estimated 23300 birds (95 CI 20000 - 28300) would be killed from collisions with turbines each yearWe estimated direct habitat loss based on data from 32 wind farms in Canada On average total habitat loss per turbine was123 ha which corresponds to an estimated total habitat loss due to wind farms nationwide of 3635 ha Based on publishedestimates of nest density this could represent habitat for ~5700 nests of all species Assuming nearby habitats are saturated and2 adults displaced per nest site effects of direct habitat loss are less than that of direct mortality Installed wind capacity isgrowing rapidly and is predicted to increase more than 10-fold over the next 10-15 years which could lead to direct mortalityof approximately 233000 birds year and displacement of 57000 pairs Despite concerns about the impacts of biased correctionfactors on the accuracy of mortality estimates these values are likely much lower than those from collisions with some otheranthropogenic sources such as windows vehicles or towers or habitat loss due to many other forms of development Speciescomposition data suggest that lt 02 of the population of any species is currently affected by mortality or displacement fromwind turbine development Therefore population level impacts are unlikely provided that highly sensitive or rare habitats aswell as concentration areas for species at risk are avoided

REacuteSUMEacute Nous avons eacutevalueacute les impacts de la construction et de lrsquoopeacuteration de parcs eacuteoliens sur les oiseaux au Canada enconsideacuterant la mortaliteacute attribuable tant aux collisions qursquoagrave la perte drsquohabitat de nidification Nous avons estimeacute la mortaliteacute parcollision agrave partir de donneacutees provenant de la recherche de carcasses dans 43 parcs eacuteoliens et avons inclus des facteurs decorrection tenant compte de lrsquoefficaciteacute des observateurs de la persistance des carcasses et du fait qursquoelles pouvaient se trouveragrave lrsquoexteacuterieur de lrsquoaire couverte par les recherches En moyenne 82 plusmn 14 oiseaux (IC agrave 95 ) ont eacuteteacute tueacutes par eacuteolienne par anneacuteedans ces parcs quoique ce nombre a varieacute de 0 agrave 269 oiseaux par eacuteolienne par anneacutee lorsque les parcs ont eacuteteacute pris individuellementFondeacute sur 2 955 eacuteoliennes construites (bilan en deacutecembre 2011 au Canada) nous estimons que 23 300 oiseaux (IC agrave 95 = 20000 agrave 28 300) seraient tueacutes agrave la suite de collisions avec les eacuteoliennes chaque anneacutee Nous avons ensuite estimeacute la perte directedrsquohabitat agrave partir de donneacutees de 32 parcs eacuteoliens au Canada En moyenne la perte drsquohabitat par turbine srsquoeacutelevait agrave 123 ha cequi eacutequivaut agrave 3 635 ha drsquohabitat perdu au profit drsquoeacuteoliennes dans lrsquoensemble du Canada Selon des estimations de densiteacute denids publieacutees une quantiteacute drsquohabitat de cet ordre pourrait contenir 5 700 nids toutes espegraveces confondues En preacutesumant que lesmilieux voisins sont satureacutes et que deux adultes sont deacuteplaceacutes par site de nidification les effets de la perte drsquohabitat directe sontmoins importants que ceux qui sont attribuables agrave la mortaliteacute directe Or la puissance eacuteolienne installeacutee augmente rapidementet on preacutevoit qursquoelle se multipliera par plus de dix drsquoici 10 agrave 15 ans ce qui pourrait se traduire par la mortaliteacute directe drsquoenviron233 000 oiseauxanneacutee et le deacuteplacement de 57 000 couples Malgreacute les preacuteoccupations quant aux impacts des biais des facteursde correction sur la preacutecision des estimations de la mortaliteacute ces valeurs sont vraisemblablement moins eacuteleveacutees que cellescalculeacutees pour les collisions avec drsquoautres sources anthropiques comme les fenecirctres les veacutehicules ou les tours ou encore cellesassocieacutees agrave la perte drsquohabitat attribuable agrave de nombreuses autres formes de deacuteveloppement Les donneacutees sur la compositionspeacutecifique indiquent que lt 02 des populations peu importe lrsquoespegravece est en reacutealiteacute toucheacutee par la mortaliteacute ou le deacuteplacement

1Canadian Wildlife Service

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

occasionneacute par le deacuteveloppement eacuteolien Ainsi les reacutepercussions sur le plan des populations sont peu probables pourvu que lesmilieux tregraves sensibles ou rares de mecircme que les aires de concentration drsquoespegraveces en peacuteril soient eacuteviteacutes

Key Words bird populations Canada collision mortality correction factors habitat loss wind turbine

INTRODUCTIONAlthough wind power is widely viewed as a clean alternativeto fossil fuel-based energy generation there has been someconcern regarding the impact of wind farms on birds (Kernand Kerlinger 2003 Langston and Pullan 2003 Kingsley andWhittam 2005 Drewitt and Langston 2006 Barclay et al2007 Smallwood 2013) Birds can be killed through collisionswith turbines and other ancillary structures such asmeteorological (met) towers or power lines and through nestmortality if vegetation clearing required for projectdevelopment occurs during the nesting season (Band et al2007 Smallwood and Thelander 2008) Constructionassociated with wind turbines can also lead to loss of nestinghabitat thus reducing the carrying capacity or productivity ofa site in the longer term

Most turbine collision studies have reported low levels ofoverall bird mortality (Drewitt and Langston 2006 2008)especially when compared to mortality from other man-madestructures such as communication towers (Kerlinger et al2011 Longcore et al 2012) Results from mortality studies atvarious sites in the United States and Europe generally suggestthat annual bird collisions range from 0 to over 30 collisionsper turbine although data collection protocols experimentaldesign and analysis methods varied substantially among windfarms making many studies difficult to compare (Kuvleskyet al 2007 Sterner et al 2007 Stewart et al 2007 Ferrer etal 2012) Erickson et al (2001) estimated 33000 birds killedper year based on 15000 turbines in the United States for anaverage of 21 birdsturbineyear Manville (2009) suggestedthat this may be a considerable underestimate and provideda number of 440000 birds per year with 23000 turbinesinstalled but did not provide a rationale for the revisednumber Smallwood (2013) estimated 111 birdsmegawattyear (or 222 birds per turbine for a typical modern 2 MWturbine) which would represent about 573000 birds killedeach year across an installed capacity in 2012 of 51630megawatts One report from Spain (Atienza et al 2011)suggested mortality rates could be as high as 300 - 1000 birdsturbineyear but little justification was provided for thisestimate

Passerines typically comprise 80 of all fatalities most ofwhich involve nocturnal migrants (Mabee et al 2006Kuvlesky et al 2007) but some of the greatest concern hasrelated to raptors The behavior of some diurnal migratingbirds such as raptors makes them vulnerable to collisionswith wind turbines particularly if they are hunting (Higginset al 2007 Garvin et al 2011 Martiacutenez-Abraiacuten et al 2012)Raptor mortality rates were particularly high at some early

large-scale wind energy facilities in California eg Altamont(Orloff and Flannery 1992) Although some reports suggestthat newer wind facilities are generally associated with lowerbird fatality rates (Erickson et al 2001 2005) others proposethat the more modern turbine designs with taller towers andlarger blade lengths with higher tip speeds pose highercollision risks (Morrison 2006) For nocturnal migrants thereis little evidence that particular species of birds are morevulnerable than others and mortality is thought to beproportional to the relative abundance of each species (Drewittand Langston 2006) Current levels of mortality are notthought to impact most individual bird populations (Kuvleskyet al 2007 Arnold and Zink 2011) Nevertheless thesecollisions contribute to the cumulative mortality of birds(Gauthreaux and Besler 2003) and potentially could influencethe long-term viability of some populations Particular concernhas been expressed that even low levels of mortality for specieswith low population densities and slow reproductive ratessuch as raptors could have population level impacts (Manville2009)

Nest mortality and the displacement of breeding pairs throughloss of habitat may also occur when habitat is removed for theconstruction of turbine pads access roads and power lines(Pearce-Higgins et al 2012) Such impacts are similar to thosefrom many other types of development but neverthelessrepresent a potential impact of a wind farm Newer facilitiesare constructed with larger more efficient turbines that resultin fewer turbine pads and roads per unit of energy In Canadamost wind farms have been built in agricultural areas or otherdisturbed areas but some have been built in areas that wereformerly relatively undeveloped eg contiguous forest

Wind energy development in Canada has grown dramaticallyin recent years and has been subject to considerable effort forenvironmental assessment relative to many other forms ofdevelopment but a comprehensive analysis of data fromCanadian sites has not yet been undertaken (Smallwood [2013]used data from only two Canadian sites) Installed capacity ofcommercial wind power in Canada increased by 750between 2005 and 2011 as of December 2011 there were over135 wind farms with 2955 wind turbines in the country andthis number is expected to increase tenfold over the next 10 -15 years (CANWEA 2011) The high effort required forenvironmental assessments has been due partly to uncertaintyabout the impacts of a relatively new and changing technologyas well as concerns about wildlife impacts observed at someof the early facilities such as those in California As a resultmany facilities have undergone lengthy baseline studies formultiple seasons and years to determine wildlife use of the site

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

(Stantec unpublished report Jacques Whitford Ltdunpublished report) The results of these studies have at somesites influenced the location of turbine locations within a windfacility (Golder Associates Ltd unpublished report) andoccasionally influenced the decision whether or not toproceed with development of a wind facility (eg NationalWind Watch 2011) Once constructed many sites have carriedout extensive postconstruction studies to monitor directmortality caused by wind turbines to birds and other wildlife

We analyze available data on bird mortality frompostconstruction monitoring reports in Canada to derivenational and provincial estimates of total bird mortality fromcollisions with wind turbines and compare them withpublished estimates from other countries and estimates of birdmortality from other forms of development We also estimatethe loss of breeding habitat and potential rates of nest mortalityas a result of construction activities Although weacknowledge that other wildlife particularly bats may alsobe impacted by wind power projects we limited our currentanalyses to impacts on birds

METHODS

Collision mortality dataWe obtained all available postconstruction monitoring reportsfor sites across Canada that had been submitted to NaturalResources Canada as part of the postconstruction monitoringrequirements of an environmental assessment We alsoobtained data from postconstruction monitoring reportspublished on wind developersrsquo or their consultantsrsquo websitesas well as some confidential data directly from developers ortheir consultants on the condition that they could be used forthis analysis but could not be publically released Thesereports provide the results of carcass searches conducted atthe base of turbines at operating wind farms however fewinclude mortality data from meteorological towers or aboveground power lines so those were excluded from our studyWe obtained data for wind farms in eight provinces but wereunable to obtain relevant data for wind farms in the YukonManitoba or Newfoundland and Labrador There are currentlyno wind farms operating in the Northwest Territories orNunavut

Most data in postconstruction monitoring reports werecollected by environmental consultants contracted by the windenergy developer Data collection methodologies used toestimate mortality were not standardized and therefore variedsomewhat between wind energy developments andconsultants especially prior to 2007 when federalenvironmental assessment guidelines for birds were published(Environment Canada 2007ab) The lack of standardizationin data collection protocols means that the raw carcass countsmay not be comparable among studies because the proportion

of carcasses retrieved may vary among study Neverthelessby taking into account appropriate correction factors specificto the methodology in each study it is possible to estimatemortality rates and make direct comparisons among studies

We estimated collision mortality based on data collected fromsearching for carcasses around a fixed distance usually 50 mfrom the base of turbines and corrected for incompletedetection Carcass searches are only expected to find some ofthe birds killed by turbines Some carcasses may be removedby scavengers others may land outside the search area whileothers may be overlooked by the searcher Each of these factorsmay vary among studies depending upon the terrain the areasearched and the individual searching in the field We applieda standardized approach to correcting the raw data usingEquation 1 which is similar to that used for postconstructionmonitoring studies in some provinces (eg OMNR 2011)

C = c (Se Sc Ps Pr Py) (1)

where C = corrected number of bird mortalities c = numberof carcasses found Se = proportion of carcasses expected tobe found by searchers (searcher efficiency) Sc = proportionof carcasses not removed by scavengers over the search period(scavenger removal) Ps = proportion of area searched withina 50 m radius of turbines (assuming a uniform distribution ofcarcasses within the 50 m radius) Pr = proportion of carcassesexpected to fall within the search radius and Py = proportionof carcasses expected during the times of year that surveystook place Our analyses assumed that all carcasses found werekilled as a result of collisions with turbines

All but three of the studies we analyzed estimated bothsearcher efficiency and scavenging rates specific to their studydesign and study area There was substantial variation in theestimated values for searcher efficiency (range 030 to 085)and scavenger removal rates (range 010 to 091) highlightingthe importance of using site-specific values wheneverpossible Differences in searcher efficiency are expectedbecause of variation in the habitat being searched whichvaried from gravel pads to agricultural fields to regeneratingvegetation as well as differences in observers and theirexperience Scavenger rates are also expected to vary amongsites depending both on the habitat which affects the easewith which scavengers can find carcasses as well as thescavenger community in any given area which potentiallyincludes birds mammals and invertebrates such as ants andburying beetles (Labrosse 2008)

We acknowledge the possibility that some estimates ofdetection probabilities and scavenger removal rates may bebiased in various ways In some studies multiple observerswith different levels of experience may have carried out thecarcass searchers but only a single value of searcher efficiencywas provided and it was not clear whether this was anappropriately weighted average across observers Some

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

reports indicated that carcasses used for searcher efficiencyand or scavenger trials were not always placed in the samehabitat types where carcass searches were conducted if theywere concentrated in habitats that were easier or harder tosearch this could have created a bias The type of carcassesused in the searcher efficiency and scavenger removal trialsalso varied among projects most studies used migratory birdsthat had previously been found around turbines but a few usedyoung chickens and quail of varying ages Labrosse (2008)demonstrated that detection probability associated withcarcasses increases with the contrast and color against thebackground If domesticated birds are used for searcherefficiency trials this could lead to biased estimates if they aremore conspicuous than typical wild birds Furthermoresearcher efficiency trials should be done ldquoblindrdquo so that thesearcher is unaware when they are being tested to avoidchanges in search patterns on testing days but if the searcherfinds a domestic bird heshe will immediately be aware of thetrial Scavenger removal trials may also have been biased ifthey were only carried out during part of the season if domesticbirds instead of wild birds were used if they were not placedin all of the habitats being searched or if some of the carcasseshad already been dead for a while freshly dead birds are likelyto be scavenged much faster than birds that have been deadfor several days and are partly dehydrated (Van Pelt and Piatt1995) For the purpose of this study we have accepted thevalues provided in the individual studies but recognize thatin some cases this may lead to bias an issue which we considerin the discussion

Most studies did not estimate the proportion of birds expectedto fall beyond the typical 50 m search radius (Pr) so we usedstandardized estimates based on a limited number of studiesthat monitored a larger search area The distribution ofcarcasses in relation to distance from the turbines is unlikelyto vary much among sites because it is affected mainly by theheight of the turbines (Hull and Muir 2010) which is verysimilar for most Canadian turbines and is not affected byhabitat searcher efficiency or scavengers Hull and Muir(2010) used a Monte-Carlo approach based on ballistics tomodel the proportion of carcasses that would be thrownvarious distances from a turbine assuming that birds acquirea forward momentum based on the speed of the blade and areequally likely to be hit anywhere along the length of the bladeFor turbines with an 80 m high nacelle and 45 m long bladessimilar to most Canadian turbines they suggested that 99 ofsmall birds (10 g) would land within 71 m of the turbine baseThis distance increases to nearly 90 m for midsized birds (680g) and 115 m for large birds (4500 g) Because most birdskilled at Canadian turbines were small to midsized weassumed that a radius of 85 m would include nearly allindividuals and used empirical data from postconstructionmonitoring reports that had searched areas up to 85 m aroundturbines in open habitats All of these studies used equal-width

transects to provide uniform coverage through the search areaIn three studies a 120 x 120 m square grid was searched aroundeach turbine while one used a 160 x 160 m square grid Allof these studies provided complete coverage out to 60 m butto estimate the number of carcasses that fell between 60 - 85m from the turbine we extrapolated the number of carcassesfound per square meter in the corners of the square search gridto a circle with a radius of 85 m Averaged across these fourstudies we estimated that only 482 of carcasses fell withina 50 m radius This estimate is very close to an estimate basedon our own unpublished study in spring 2013 that searchedthe entire 85 m radius around turbines in an agricultural areaand found 58 carcasses of which only 465 were within a 50m radius We applied a correction factor of 48 (a weightedaverage of these studies) to the estimated mortality for all windfarms that only searched up to a 50 m radius

Most sites were only surveyed for part of the year generallyin seasons when the highest risk of mortality was anticipatedusually the spring and autumn migration periods In one caseonly three months ie one season of postconstructionmonitoring data were collected Most of the reports providedthe total number of carcasses and estimated total mortality persite only for seasons that were surveyed To extrapolateestimates to an annual total (Py) we used data from four windfarms in Alberta and one in Ontario that were surveyed for upto two years throughout the annual cycle and for whichmortalities were reported for each month Using these datawe estimated the monthly distribution of mortality throughoutthe year We then estimated annual mortality for other sitesby dividing the corrected number of bird mortalities per turbineby the proportion of mortalities expected during the actualdates that surveys took place eg in Alberta on average 95of mortalities occurred between April and November inOntario 985 of mortalities occurred during this time periodFor sites in Ontario and east we used the Ontario estimatewhereas for sites in Manitoba and west we used the Albertaestimate of proportion of mortalities expected during the actualdates that the surveys took place We acknowledge that thereare limitations to extrapolating results from one season to otherseasons based on only five studies because seasonal patternsof mortality rates may vary among locations Howeverbecause most studies concentrated their efforts during seasonswith the highest expected mortality and because relativelyfew birds are found during the summer and winter monthsany error associated with extrapolation to these seasons isunlikely to have much impact on our estimates

We applied the appropriate correction factors (Equation 1) toall 43 wind farms using the data provided in reports on searcherefficiency scavenger rates and area searched Where multipleyears of data were collected at a site we used the average ofall years for the analysis In some studies there wereinsufficient data to apply correction factors and estimatemortality for each season ie spring and fall so we used

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

average factors At one wind farm in Alberta searcherefficiency was not reported and at two wind farms scavengerimpacts were not reported In these cases we applied theaverage values for wind farms in Alberta (Se = 065 and Sc =061) We did not include data from any reports in our analysiswhere both searcher efficiency and scavenger impact datawere absent or where surveys occurred for less than threemonths throughout the year

To determine if there was any significant variation in estimatedmortality among provinces we conducted a one-way Analysisof Variance (ANOVA) using estimated average mortality foreach of the 43 wind farms as the sampling unit and provinceas the predictor variable We estimated avian mortality foreach province based on estimated average mortality perturbine in that province for studies for which we had datamultiplied by the total number of turbines in the province Forprovinces without any collision data we used the estimatedaveraged mortality per turbine across Canada We estimatedtotal collision mortality for wind farms across Canada as thesum of the provincial estimates

We did not analyze whether specific turbine model types orcharacteristics posed a higher risk to birds relative to othertypes because there was relatively little variation in the totalheight of wind turbines (hub height plus rotor radius) in thesample for which we had data Of the wind farms analyzed inthis study 81 started their postconstruction monitoringstudies after 2007 and based on a review of turbinespecifications we found that the total height (tower andblades) of nearly all of the turbines erected since 2007 wasbetween 117 m and 136 m with a nameplate capacity of 15MW to 30 MW

Species-specific population impactsWe estimated the population relevance of mortality as thenumber of individuals of each species lost from collisions peryear with wind turbines relative to the total estimatedpopulation size of that species Of the 43 reports reviewed 37reported species composition of the fatalities Becauserelatively few individuals of any one species were detected ateach site we pooled species-composition data across all sitesto estimate the percentage of mortalities that were representedby each species The annual mortality estimate of each speciesat a national scale was calculated using Equation 2

Mortality Species X = Mortalityturbine Turbines carcasses Species X (2)

We used the Partners in Flight (PIF) landbird populationdatabase (httprmboorgpif_dblaped) Status of Birds inCanada (httpwwwecgccasoc-sbcindex-engaspxsL=eampsY=2010) and the USFWS Waterfowl Population Status2012 report (httpwwwfwsgovmigratorybirdsNewReportsPublications PopulationStatusWaterfowl

StatusReport2012_finalpdf) to obtain estimates of the totalpopulation size of each of the 10 most abundant species incarcass searches as well as any species listed as Endangeredor Threatened under the Species At Risk Act (Government ofCanada 2002) Population impact for each species was thencalculated as the total estimated mortality for that speciesdivided by its estimated population size

Loss of nesting habitatWe used information from 32 environmental assessments andpostconstruction reports for wind energy developments inCanada to estimate total habitat loss from power lines roadssubstations laydown areas and turbine pads at each site Thereis some evidence that habitat loss increases in anapproximately linear fashion with wind farm size eg numberof turbines (Stantec unpublished report) As such we summedhabitat loss from each component and presented the value ashabitat loss (ha) per turbine Average habitat loss per turbineper wind farm was calculated and extrapolated to predicthabitat loss per turbine at the remaining 103 wind farms

We used ecological land classification habitat mapping orremote-sensing imagery provided in the environmentalassessments to classify habitats for the 32 wind farms asagriculture grasslands deciduous forest coniferous forest ormixed forest Habitat data provided in many environmentalassessments were insufficient to provide finer-scaleclassifications The remaining 103 wind farms for whichenvironmental assessments could not be obtained were locatedbased on their geographical coordinates and assigned to oneof these broad habitat classifications based on SPOT satelliteimagery for the location Nest densities in these habitats wereestimated based on previous Canadian studies of incidentaltake due to forestry operations (Hobson et al 2013) mowingand other mechanical operations in agricultural landscapes(Tews et al 2013) We used the MODIS land coverclassification layer to calculate the total area of each broadhabitat type and province and the Hobson et al (2013) andTews et al (2013) nest densities to estimate total number ofnests The percent of nesting habitat lost for a given habitattype in a province was calculated by dividing the predictedhabitat lost from wind energy developments by the total areafor that habitat type across the province

RESULTS

Collision mortality estimatesEstimates of collision mortality among the 43 wind farmsvaried between 0 and 269 birds per turbine per year Onaverage estimated mortality (plusmn 95 CI) was 82 plusmn 14 birdsper turbine per year There was no significant variation inestimated mortality per turbine among provinces in Canada(F735 = 152 p = 0191) Based on 2955 installed turbines (thenumber installed by December 2011) the estimated annual

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Table 1 Estimated bird mortality per turbine from collisions at wind farms with available carcass search data and estimatedtotal mortality per province based on the number of installed turbines

ProvinceTerritory No ofWind Farms

No of Turbines No of Wind FarmsAnalyzed

Estimated MortalityTurbine

Predicted EstimatedMortality

Yukon (YK) 2 2 0 82dagger 16British Columbia (BC) 3 83 2 84 697Alberta (AB) 26 588 7 45 2646Saskatchewan (SK) 5 132 3 101 1333Manitoba (MB) 3 123 0 82dagger 1009Ontario (ON) 38 965 19 108 10422Quebec (QC) 15 672 2 52 3494New Brunswick (NB) 4 113 2 24 271Nova Scotia (NS) 26 161 6 112 1803Prince Edward Island (PE) 9 90 2 152 1368Newfoundland (NF) 4 26 0 82dagger 213Total 135 2955 43 23273daggerWhere no data were available for a particular province the weighted national estimate was used

mortality across Canada was 23300 birds (95 CI 20000 -28300) Nearly half of all collisions occurred in the provinceof Ontario where the highest numbers of turbines are installed(Table 1)

Species-specific population impactsOverall the 37 reports that recorded species compositionduring postconstruction mortality surveys identified 1297 birdcarcasses of 140 species The most frequently recoveredspecies were Horned Lark (Eremophila alpestris) Golden-crowned Kinglet (Regulus satrapa) Red-eyed Vireo (Vireoolivaceus) European Starling (Sturnus vulgaris) and TreeSwallow (Tachycineta bicolor Table 2) For the mostcommonly recovered carcasses and also for species at riskcollision mortality was estimated to have an annual impact ofless than 08 of any population at a national level (Table 2)

Loss of nesting habitat and nest mortality estimatesOn average total habitat loss per turbine at 32 wind farms inCanada was 123 plusmn 072 ha Based on this average thepredicted total habitat loss for wind farms nationwide was3635 ha (Table 3) Using the nest density estimates providedin Hobson et al (2013) and Tews et al (2013) the total numberof potentially affected nests in each habitat for all representedprovinces was 5715 We had few data on the timing ofconstruction activities to estimate the number of nests thatmight have been disturbed or destroyed during constructionThe amount of nesting habitat disturbed or destroyed wasestimated to vary from 0002 of coniferous forest in theYukon Territory to almost 8 of mixed forest in PrinceEdward Island (Table 3)

DISCUSSION

Collision mortality estimatesOur estimates for average annual mortality per turbine at windfarms in Canada of 82 birds is higher than most estimates

derived from reports from individual studies in Canada Thisis due mainly to incorporation of two additional correctionfactors the proportion of birds likely to fall outside a 50 msearch radius and the proportion of birds killed at other timesof year Nevertheless our estimates are lower than some recentestimates for bird mortality in the United States Manville(2009) suggested that annual bird mortality from wind powerprojects in the United States was 440000 birds which equalsabout 19 birds per turbine based on an estimated 23000turbines at the time However he did not provide any detailson how this estimate was derived Smallwood (2013)undertook a detailed assessment of correction factors basedon data from 60 different reports and estimated an averagemortality of 11 birds per MW per year implying 22 birds perturbine for a 2 MW turbine Extrapolated to an installedcapacity of 51630 MW in the United States this implies573000 bird fatalities per year Most of the difference inSmallwoodrsquos estimates compared to ours appears to be dueto differences in the correction factors rather than a differencein the number of carcasses found in the data he analyzed Forexample he used larger corrections for birds falling outside a50 m radius by assuming a logistic distribution of carcasseshe concluded that carcasses could fall up to 156 m away froman 80 m turbine though this is farther than Hull and Muir(2010) suggested is likely Furthermore his analysis assumesthat mortality is proportional to the rated capacity of theturbines but particularly for newer turbines this seemsunlikely for example there is only a 19 increase in the bladeswept area between a 15 and 30 MW turbine (httpsitege-energycomprod_servproductswind_turbinesen15mwspecshtm httpwwwvestascomenwind-power-plantsprocurementturbine-overviewv90-30-mwaspxvestas-univers) A reportfrom Spain estimated that annual mortality was between 300 -1000 birds per turbine (Atienza et al 2011) but examinationof the underlying data suggests similar number of carcasseswere found in Spain as in Canada Their very high mortality

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Table 2 Reported mortality and estimated annual collision mortality and percent of Canadian population impacted for the 10species most frequently reported as casualties as well as two species at risk

Species No of Carcasses Proportion of Total Total PredictedMortalitydagger

Canadian PopulationEstimateDagger

of Population

Horned Lark(Eremophila alpestris)

135 010 2327 30000000 0008

Golden-crowned Kinglet(Regulus satrapa)

92 007 1629 23000000 0007

Red-eyed Vireo(Vireo olivaceus)

80 006 1396 96000000 0001

European Starling(Sturnus vulgaris)

66 005 1164 30000000 0004

Tree Swallow(Tachycineta bicolor)

53 004 931 12000000 0008

Red-tailed Hawk(Buteo jamaicensis)

40 003 698 582000 0120

Ring-billed Gull(Larus delawarensis)

27 002 465 1000000 0047

Mourning Dove(Zenaida macroura)

26 002 465 5300000 0009

Mallard(Anas platyrhynchos)

20 002 465 7200000 0006

Purple Martin(Progne subis)

20 002 465 523000 0089

Canada Warbler(Cardellina canadensis THR)sect

4 0003 70 1350000 0005

Chimney Swift(Chaetura pelagica THR)sect

4 0003 70 145000 0048

dagger Extrapolation based on 2955 turbines across Canada and assuming that species composition of sampled sites is representative of all sites ignoringvariation in habitat and species distributionsDagger Based on estimates from various sources (see Methods)sect Threatened on Schedule 1 of Species At Risk Act

estimates were based on assumptions that searcher efficiencywas extremely low scavenger rates were very high and largenumbers of carcasses fell outside the search areas Howeverno evidence was presented to support those assumptions andit is quite possible that mortality rates were not actually anyhigher than those in Canada

We estimated total mortality across all sites in Canada at about23300 birds per year based on 2955 turbines Installed windcapacity is growing rapidly in Canada and is predicted toincrease more than tenfold over the next 10 - 15 years whichcould lead to direct mortality of approximately 233000 birdsper year Based on Smallwoodrsquos (2013) analysis currentmortality in the United States is estimated at 573000 birds peryear which with a projected sixfold increase over the sametime period could lead to direct mortality of over 3 millionbirds per year Even at these levels estimated mortalityassociated with wind turbines would still be lower than thosefrom some other anthropogenic sources Erickson et al (2005)estimated 500 million birds killed annually due to collisionswith residential buildings 130 million for collision with powerlines 80 million for collisions with vehicles and 100 milliondue to domestic and feral cats To some degree thesedifferences in impacts are due to the much larger numbers of

other structures in the landscape For example in the UnitedStates there were an estimated 100 million residentialbuildings and average mortality is estimated to be five birdsper building (see Klem 1990) However mortality perstructure is also higher for many other structures than windturbines For example in Canada the average annual mortalityrate from communication towers is estimated to be 28 birdsper tower (Longcore et al 2012) compared to 82 birds perwind turbine Several studies have suggested that manymigratory birds exhibit avoidance behavior when approachingmodern wind turbines (eg Erickson et al 2002 Zdawczyk2012) which may partly explain relatively low mortalitycompared to other structures

We found substantial variation among sites in the estimatedmortality per turbine ranging from 0 to 27 birds per year butlittle variation among provinces although for severalprovinces we only had data from a few sites Some naturalvariation in mortality estimates is expected because of site-specific characteristics that may concentrate migratory birdsin some areas and not in others For example landscapefeatures such as promontories and large bodies of water aremore likely to concentrate migratory birds along the shoreline(eg Diehl et al 2003) as are largely forested landscapes

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Table 3 Estimated habitat loss number of nests lost or breeding pairs displaced and percent of nesting habitat lost for eachhabitat by province

Province Habitat No ofwindfarms

No ofTurbines

Predictedhabitat

loss (ha)

Nestsha Estimated No ofnests lostpairs

displaced

Area of habitat (ha) of nestinghabitat lost

Yukon Coniferous Forest 2 2 25 4 10 123200 0002British Columbia Mixed Forest 3 83 1021 5 510 132000 0077Alberta Agriculture 21 532 6544 02 131 88700 0737Alberta Grassland 5 56 689 08 55 75300 0092Saskatchewan Agriculture 5 132 1624 02 32 122800 0132Manitoba Agriculture 3 123 1513 02 30 30600 0494Ontario Agriculture 32 875 10763 02 215 32300 333Ontario Deciduous Forest 2 86 1058 78 825 110000 0096Ontario Urban 4 4 49 02 1 2700 0183Quebec Agriculture 10 506 6224 02 124 20800 299Quebec Mixed Forest 5 166 2042 62 1266 247900 0082New Brunswick Coniferous Forest 1 33 406 72 292 4600 0891New Brunswick Mixed Forest 3 80 984 6 590 25400 0386Nova Scotia Mixed Forest 25 127 1562 62 969 16500 0945Nova Scotia Deciduous Forest 1 34 418 57 238 25400 0165Prince Edward Island Agriculture 4 62 763 02 15 2300 336Prince Edward Island Mixed Forest 5 28 344 62 214 430 796Newfoundland Mixed Forest 4 26 320 62 198 59000 0054

TOTAL 135 2955 3635 5715

relative to predominately agricultural landscapes (Buchanan2008) Based on data from the postconstruction monitoringreports we were unable to identify factors other thancorrection factors that may explain variation in mortalityamong sites

The accuracy of collision mortality estimates depends both onthe quality of the carcass search data and the accuracy ofcorrection factors used to account for incomplete carcassdetections If search effort is only sufficient to detect a few (orno) carcasses estimates will be unreliable regardless ofcorrection factors Potential biases in correction factors couldlead to either over or under estimates of mortality Severalfactors could lead to overestimates of searcher efficiencyincluding use of inappropriate carcasses that may be moreconspicuous or larger than species that would be expected tobe found during carcass searches (Labrosse 2008)concentrating carcasses in more exposed habitats within thesearch area and failing to ensure that searcher efficiency trialsoccur without the knowledge of the observer Scavenger ratesmay be biased if carcasses used in the trials are not fresh(Smallwood 2013) are not representative of the species beingdetected or if too many carcasses are used at one time fortrials ie scavenger swamping (Smallwood et al 2010) Allof these could result in underestimates of mortality On theother hand many studies estimate scavenger removal over thetotal search interval eg three days This may lead to anoverestimate because on average one would expect carcassesto be exposed to potential scavengers for only half the searchperiod eg for a typical three-day search interval birds wouldbe equally likely to be killed one two or three days beforethe search leading to an average exposure of 15 days Our

correction factors may also be biased low if some birds fallbeyond 85 m (Jain et al 2007 2009 Smallwood 2013)although ballistic modeling suggests very few birds areexpected beyond that distance (Hull and Muir 2010) On theother hand we assumed that all birds found during carcasssearches died as a result of colliding with the turbines If someof these birds died from other sources of mortality unrelatedto wind turbines (eg see Nicholson et al 2005) this wouldlead to an overestimate of the impacts of turbines

The net effect of these various potential biases both positiveand negative is difficult to predict because some may canceleach other but overall we believe that our estimates areprobably reasonable Furthermore we note that despite somepotential biases the wind industry has some of the mostreliable data for estimating incidental mortality to birds of anyindustrial sector in Canada Even if we have underestimatedsome of the correction factors and the actual mortality isdouble what has been reported it is still low compared to otherindustrial sectors

Species-specific population impactsThe short-term population effects of wind power in Canadaon most species appear to be relatively negligible Generallythe birds recorded in carcass searches were abundant specieswith large populations For example Horned Lark Golden-crowned Kinglet Red-eyed Vireo European Starling andTree Swallow were the most commonly impacted species Forthe five most common bird species the effect of collisionswith wind turbines is unlikely to affect their conservationbecause the estimated mortality represents less than 001 oftheir Canadian populations

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Nevertheless mortality effects could be potentially importantfor individual species over the long-term eg 10 - 20 yearsespecially if wind farms are poorly sited For example at awind farm on Smola Island Norway prior to constructionapproximately 13 White-tailed Eagle (Haliaeetus albicilla)pairs occupied territories within 500 m of the proposed sitewhereas in 2009 only five pairs occupied territories (Nygardet al 2010) Between 2005 and 2009 36 White-tailed Eagleswere killed on the Smola Island suggesting that these collisionswere directly impacting the local population of eagles Someconcerns have been expressed that wind turbines in NorthAmerica could have negative impacts on long-lived specieswith low reproductive rates such as Golden Eagles (Aquilachrysaetos see Hunt et al 1998) though population levelquantitative data to support this concern have not beenpublished Mortality for species at risk is a particular concernbecause any incremental mortality could potentially slowrecovery However even for Chimney Swifts (Chaeturapelagica) a species that is threatened nationally wind turbinerelated mortality only affects 003 of the populationNevertheless as the number of turbines increases and giventhat many Canadian birds also migrate through the UnitedStates where they are exposed to many more turbinespopulation effects may eventually become an issue for somespecies if they are particularly vulnerable to turbines

Loss of nesting habitat and nest mortality estimatesThe other potential source of wind farm-related bird mortalityis the destruction of nests during construction Nest mortalitymight occur if vegetation containing nests is cleared ordestroyed during the bird breeding season If all constructionwas conducted during the breeding season based on nestdensity estimates approximately 5700 nests would have beendestroyed However limited available evidence frompostconstruction monitoring reports suggests that mostconstruction activities occur outside of the breeding birdseason For example in Ontario we estimated that only 20of projects conduct clearing activities during the breedingseason In the prairies up to 50 of projects may carry outsome vegetation clearing during the breeding season with thepotential to induce nest mortality The number of nestsdisturbed or destroyed from construction activities can beminimized if construction activities are conducted outside ofthe breeding season

In addition to collision and nest mortality birds may also beimpacted by the loss of nesting habitat as a result ofconstruction activities that remove vegetation for the turbinepads and infrastructure ie electrical lines substation accessroads According to our estimates approximately 123 ha ofvegetation is removed per turbine resulting in a loss ofapproximately 3600 ha an area equivalent to 500 km of atypical four-lane highway including shoulders and ditchesAssuming nearby habitats are saturated and two adultsdisplaced per nest site effects of direct habitat loss on reducing

bird populations through lost productivity the effects ofwhich are equivalent to nest mortality are lower than that ofdirect mortality At the provincial level effects of direct habitatloss from wind turbines may be more pronounced in lesscommon habitat types eg mixed forest in Prince EdwardIsland However our overall estimates of nesting habitat lossare still much smaller than habitat loss due to many other formsof development such as forestry agriculture and mining (seeCalvert et al 2013)

In addition to the direct loss of habitat birds may avoidforaging nesting and roosting habitats near wind farms duringconstruction activities and operation thus effectivelydecreasing habitat quality beyond the immediate footprint ofthe turbine (Band et al 2007 Higgins et al 2007) Theimportance of this indirect effect has rarely been measuredbut varies among species depending on their life historybehavior and habitat requirements (Desholm 2009) Forexample some species of birds breeding near wind farmshabituate to the presence and operation of the turbines overtime eg Pink-footed Goose (Anser brachyrhynchus Madsenand Boertmann 2008) while others tend to avoid these areasbecause turbines obstruct flight paths and feeding areas(Pearce-Higgins et al 2009) Some grassland breeding birdsfor example avoid nesting within 100 - 200 m of turbinesalthough at the Ponnequin Wind Energy Facility in Coloradograssland songbirds eg Horned Larks and WesternMeadowlarks (Sturnella neglecta) forage directly beneathturbines (cited in Kerlinger and Dowdell 2003) At theSimpson Ridge Wind Farm in Wyoming female survival nestsuccess and brood survival were not statistically different inareas with and without turbines although the authors cautionthat long-term data from multiple locations are needed tovalidate their results (Johnson et al 2012) Insufficient datawere available in postconstruction monitoring reports used inthis study to assess the effect of avoidance of foraging nestingor roosting habitats on birds in Canada

Conservation implicationsThe accuracy of collision mortality estimates depends stronglyon the reliability of correction factors used to account forincomplete carcass detections Environment Canadarsquos windenergy guidelines (Environment Canada 2007ab) providesguidance on the assessment of potential impacts of windenergy projects on migratory birds in Canada and the type ofcorrection factors that need to be considered duringpostconstruction monitoring studies However specific datacollection protocols were not included in these guidelines toaccommodate the diversity of landscape and habitat types thatexist across Canada Given concerns about the uncertainty andbiases associated with correction factors national standardsshould be established to ensure that correction factors arerobust and defensible and that the estimated impacts of windenergy developments on migratory birds are accurate Furtherdirected research on the expected carcass distribution in

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

relation to distance from turbines would reduce uncertaintyassociated with this correction factor

Overall based on the assumptions and limitation outlined inthis study the combined effects of collisions nest mortalityand lost habitat on birds associated with Canadian wind farmsappear to be relatively small compared to other sources ofmortality Although total mortality is anticipated to increasesubstantially as the number of turbines increases even atenfold increase would represent mortality orders ofmagnitude smaller than from many other sources of collisionmortality in Canada (Calvert et al 2013) Habitat loss is alsorelatively small compared to many other forms ofdevelopment including road development Population levelimpacts are unlikely on most species of birds provided thathighly sensitive or rare habitats as well as concentration areasfor species at risk are avoided

Although at a national level mortality associated with windenergy developments in Canada is unlikely to affect most birdpopulations based on the approach used in our study this maynot be true for other wildlife such as bats For example atsome wind farms in Canada the estimated mortality rate is gt45 bats per turbine annually It is uncertain whether this levelof mortality could have population level impacts because noreliable estimates are currently available of population sizesfor most species Those bat species experiencing significantpopulation declines because of White-nose Syndrome(Geomyces destructans) may be especially vulnerable

In some situations where a species population may bethreatened by wind turbine developments or where the rate ofmortality may be above provincial thresholds mitigation maybe required For example in the province of Ontariomitigation may be required when mortality estimates exceed14 birdsturbineyear or raptor mortality exceeds 02 birdsturbineyear (OMNR 2011) Mitigation measures to reducebird mortality may include the feathering of wind turbineblades when the risk to birds is particularly high eg at nightduring peak migration In extreme circumstances operationalmitigation techniques may include the periodic shutdown ofselect turbines during the highest risk periods At some windturbine developments in the United States modified marineradars have been installed to detect approaching bird activityassess mortality risk conditions and when necessaryautomatically activate the shutdown of all turbines Howeverthere are no published reports on the effectiveness of thisemerging technology to mitigate bird mortality at wind turbinedevelopments and the overall relatively low levels of avianmortality caused by wind turbines suggests this should notnormally be necessary

Responses to this article can be read online at httpwwwace-ecoorgissuesresponsesphp609

Acknowledgments

We would like to thank Natural Resources Canada forproviding many of the postconstruction monitoring reports tothe authors We are grateful to those wind developers that alsoprovided data for this manuscript Tom Levy provided helpfulcomments on early drafts of the analyses We would also liketo thank Mike Anissimoff who provided assistance in compilingthese data and Craig Machtans for providing leadership andfunding from Environment Canada for this project Finallywe would like to thank two anonymous reviewers for providingconstructive and thoughtful comments on this manuscript

LITERATURE CITEDArnold T W and R M Zink 2011 Collision mortality hasno discernible effect on population trends of North Americanbirds PLos One 6(9)e24708 httpdxdoiorg101371journalpone0024708

Atienza J C I M Fierro O Infante J Valls and JDominguez 2011 Directrices para la evaluacion del impactode los parques eolicos en aves y murcielagos (version 30)SEOBirdlife Madrid Spain

Band W M Madders and D P Whitfield 2007 Developingfield and analytical methods to assess avian collision risk atwind farms Pages 259-275 in M de Lucas G F E Janss andM Ferrer editors Birds and wind farms risk assessment andmitigation Quercus Madrid Spain

Barclay R M R E F Baerwald and J C Gruver 2007Variation in bat and bird fatalities at wind energy facilitiesassessing the effects of rotor size and tower height CanadianJournal of Zoology 85381-387 httpdxdoiorg101139Z07-011

Buchanan A 2008 Movements habitat use and stopoverduration of migratory songbirds in the western Lake Erie basinof northern Ohio Thesis Ohio State University ColumbusOhio USA

Calvert A C A Bishop R D Elliot E A Krebs T MKydd C S Machtans and G J Robertson 2013 A synthesisof human-related avian mortality in Canada AvianConservation and Ecology 8(2) 11 httpdxdoiorg105751ACE-00581-080211

Canadian Wind Energy Association (CANWEA) 2011Powering Canadarsquos future CANWEA Ottawa OntarioCanada Data accessed December 2011

Desholm M 2009 Avian sensitivity to mortality prioritisingmigratory bird species for assessment at proposed wind farmsJournal of Environmental Management 902672-2679 httpdxdoiorg101016jjenvman200902005

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Diehl R H R P Larkin and J E Black 2003 Radarobservations of bird migration over the Great Lakes Auk 120278-290

Drewitt A L and R H W Langston 2006 Assessing theimpacts of wind farms on birds Ibis 14829-42 httpdxdoiorg101111j1474-919X200600516x

Drewitt A L and R H W Langston 2008 Collision effectsof wind-power generators and other obstacles on birds Annalsof the New York Academy of Sciences 1134233-266 httpdxdoiorg101196annals1439015

Environment Canada 2007a Recommended protocols formonitoring impacts of wind turbines on birds EnvironmentCanada Gatineau Queacutebec Canada

Environment Canada 2007b Wind energy and birds aguidance document for environmental assessment EnvironmentCanada Gatineau Queacutebec Canada

Erickson W P G D Johnson M D Strickland and D PYoung 2005 A summary and comparison of bird mortalityfrom anthropogenic causes with an emphasis on collisionsPages 1029-1042 in C J Ralph and T D Rich editors Birdconservation implementation and integration in the AmericasProceedings of the Third International Partners in FlightConference General Technical Report PSW-GTR-191 USForest Service Pacific Southwest Research Station BerkeleyCalifornia USA

Erickson W P G D Johnson M D Strickland D P YoungK J Sernka and R E Good 2001 Avian collisions with windturbines a summary of existing studies and comparisons toother sources of avian collision mortality in the United States National Wind Coordinating Committee Washington DCUSA

Erickson W P G D Johnson D P Young D StricklandR E Good M Bourassa K Bay and K J Sernka 2002Synthesis and comparison of baseline avian and bat useraptor nesting and mortality information from proposed andexisting wind developments Bonneville Power AdministrationPortland Oregon USA httpdxdoiorg102172928474

Ferrer M M de Lucas G F E Janss E Casado A RMuntildeoz M J Bechard and C P Calabuig 2012 Weakrelationship between risk assessment studies and recordedmortality in wind farms Journal of Applied Ecology 4938-46httpdxdoiorg101111j1365-2664201102054x

Garvin J C C S Jennelle D Drake and S M Grodsky2011 Response of raptors to a windfarm Journal of AppliedEcology 48199-209 httpdxdoiorg101111j1365-2664201001912x

Gauthreaux S A and C G Besler 2003 Radar ornithologyand biological conservation Auk 120266-277

Government of Canada 2002 Species at Risk Act (SC 2002c 29) Government of Canada Ottawa Ontario Canada

Higgins K F R Osborn and D E Naugle 2007 Effects ofwind turbines on birds and bats in southwestern MinnesotaUSA Pages 153-175 in M de Lucas G F E Janss and MFerrer editors Birds and wind farms risk assessment andmitigation Quercus Madrid Spain

Hobson K A A G Wilson S L Van Wilgenburg and EM Bayne 2013 An estimate of nest loss in Canada due toindustrial forestry operations Avian Conservation andEcology 8(2) 5 httpdxdoiorg105751ACE-00583-080205

Hull C L and S Muir 2010 Search areas for monitoringbird and bat carcasses at wind farms using a Monte-Carlomode Australasian Journal of Environmental Management 1777-87 httpdxdoiorg1010801448656320109725253

Hunt W G R E Jackman T L Hunt D E Driscoll andL Culp 1998 A population study of Golden Eagles in theAltamont Pass Wind Resource Area population trend analysis1997 Report to National Renewable Energy laboratorySubcontract XAT-6-16459-01 Predatory Bird ResearchGroup University of California Santa Cruz California USA

Jain A P Kerlinger R Curry and L Slobodnik 2007Annual report for the Maple Ridge Wind Power Projectpostconstruction bird and bat fatality study - 2006 Curry andKerlinger Cape May Point New Jersey USA [online] URLhttpdocswind-watchorgmaple_ridge_report_2006pdf

Jain A P Kerlinger R Curry L Slobodnik and M Lehman2009 Annual report for the Maple Ridge Wind Power Projectpost-construction bird and bat fatality study - 2008 Curry andKerlinger Cape May Point New Jersey USA [online] URLhttpwwwbatsandwindorgpdfJain_2009bpdf

Johnson G D C W LeBeau R Nielsen T Rintz J Eddyand M Holloran 2012 Greater Sage-Grouse habitat use andpopulation demographics at the Simpson Ridge WindResource Area Carbon County Wyoming US Departmentof Energy Golden Colorado USA [online] URL httpwwwostigovbridgepurlcoverjsppurl=10377391037739pdf

Kerlinger P and J Dowdell 2003 Breeding bird survey forSearburgReadsboo expansion wind project BenningtonCounty Vermont Curry and Kerlinger Cape May Point NewJersey USA [online] URL httppsbvermontgovsitespsbfilesdocket7250DeerfieldPetition+SupportDocsKerlingerDFLD-PK-3_Breeding_Bird_Reportpdf

Kerlinger P J Gehring and J R Curry 2011 Understandingbird collisions at communication towers and wind turbinesBirding 4344-51

Kern J and P Kerlinger 2003 A study of bird and batcollision fatalities at the Mountaineer Wind Energy Center

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Tucker County West Virginia annual report for 2003 WestVirginia Highlands Conservancy Hillsboro West VirginiaUSA [online] URL httpwwwwvhighlandsorgBirdsMountaineerFinalAvianRpt-203-15-04PKJKpdf

Kingsley A and B Whittam 2005 Wind turbines and birdsa background review for environmental assessmentEnvironment Canada Gatineau Queacutebec Canada

Klem D Jr 1990 Collisions between birds and windowsmortality and prevention Journal of Field Ornithology 61(1)120-128

Kuvlesky W P Jr L A Brennan M L Morrison K KBoydston B M Ballard and F C Bryant 2007 Wind energydevelopment and wildlife conservation challenges andopportunities Journal of Wildlife Management 712487-2498httpdxdoiorg1021932007-248

Labrosse A A 2008 Determining factors affecting carcassremoval and searching efficiency during the post-constructionmonitoring of wind farms Thesis University of NorthernBritish Columbia Prince George British Columbia Canada

Langston R H W and J D Pullan 2003 Windfarms andbirds an analysis of the effects of windfarms on birds andguidance on environmental assessment criteria and siteselection issues Report written by BirdLife on behalf of theBern Convention Convention on the conservation ofEuropean wildlife and natural habitats Standing committee22nd meeting BirdLife International Cambridge UK[online] URL httpwwwbirdlifeorgeupdfsBirdLife_Bern_windfarmspdf

Longcore T C Rich P Mineau B MacDonald D G BertL M Sullivan E Mutrie S A Gauthreaux Jr M L AveryR L Crawford A M Manville II E R Travis and D Drake2012 An estimate of avian mortality at communication towersin the United States and Canada PLos One 7(4)e34025[online] URL httpwwwplosoneorgarticleinfodoi101371journalpone0034025 httpdxdoiorg101371journalpone0034025

Mabee T J B A Cooper J H Plissner and D P Young2006 Nocturnal bird migration over an Appalachian Ridge ata proposed wind power project Wildlife Society Bulletin 34682-690 httpdxdoiorg1021930091-7648(2006)34[682NBMOAA]20CO2

Madsen J and D Boertmann 2008 Animal behavioraladaptation to changing landscapes spring-staging geesehabituate to wind farms Landscape Ecology 231007-1011

Manville A M II 2009 Towers turbines power lines andbuildings ndash steps being taken by the US Fish and WildlifeService to avoid or minimize take of migratory birds at thesestructures Pages 262-272 in T D Rich C Arizmendi DDemarest and C Thompson editors Tundra to tropics

connecting habitats and people Proceedings 4th InternationalPartners in Flight Conference Partners in Flight

Martiacutenez-Abraiacuten A G Tavecchia H M Regan J JimeacutenezM Surroca and D Oro 2012 Effects of wind farms and foodscarcity on a large scavenging bird species following anepidemic of bovine spongiform encephalopathy Journal ofApplied Ecology 49109ndash117 httpdxdoiorg101111j1365-2664201102080x

Morrison M L 2006 Bird movements and behaviors in theGulf Coast region relation to potential wind energydevelopments Subcontract report NRELSR-500-39572National Renewable Energy Laboratory Golden ColoradoUSA httpdxdoiorg102172884690

National Wind Watch 2011 Xcel Energy pulls out of windfarm plan over rare bird concerns National Wind WatchRowe Massachusetts USA [online] URL httpswwwwind-watchorgnews20110407xcel-energy-pulls-out-of-wind-farm-plan-over-rare-bird-concerns

Nicholson C P R D Tankersley Jr J K Fieldler and NS Nicholas 2005 Assessment and prediction of bird and batmortality at wind energy facilities in the southeastern UnitedStates Final Report Tennessee Valley Authority KnoxvilleTennessee USA [online] URL httpwwwtvacomenvironmentbmw_reportbird_bat_mortalitypdf

Nygard T K Bevanger E Lie Dahl Oslash Flagstad AFollestad P Lund Hoel R May and O Reitan 2010 A studyof White-tailed Eagle Haliaeetus albicilla movements andmortality at a wind farm in Norway BOU Proceedings ndashClimate Change and Birds British Ornithologistsrsquo Union[online] URL httpwwwresearchgatenetpublication224-960531_A_study_of_White-tailed_Eagle_Haliaeetus_albicil-la_movements_and_mortality_at_a_wind_farm_in_Norwayfile9fcfd4fb367dfd4fbcpdf

Ontario Ministry of Natural Resources (OMNR) 2011 Birdsand bird habitats guidelines for wind power projects OMNRPeterborough Ontario Canada [online] URL httpwwwmnrgovoncastdprodconsumegroupslrmnrrenewabledocumentsdocumentstdprod_071273pdf

Orloff S and A Flannery 1992 Wind turbine effects onavian activity habitat use and mortality in Altamont Pass andSolano County wind resource areas 1989-1991 BioSystemsAnalysis Tiburon California USA

Pearce-Higgins J W L Stephen A Douse and R H WLangston 2012 Greater impacts of wind farms on birdpopulations during construction than subsequent operationresults of a multi-site and multi-species analysis Journal ofApplied Ecology 49386-394 httpdxdoiorg101111j1365-2664201202110x

Avian Conservation and Ecology 8(2) 10httpwwwace-ecoorgvol8iss2art10

Pearce-Higgins J W L Stephen R H W Langston I PBainbridge and R Bullman 2009 The distribution ofbreeding birds around upland wind farms Journal of AppliedEcology 461323-1331

Smallwood K S 2013 Comparing bird and bat fatality-rateestimates among North American wind-energy projectsWildlife Society Bulletin 3719-33 httpdxdoiorg101002wsb260

Smallwood K S D A Bell S A Snyder and J E Didonato2010 Novel scavenger removal trials increase wind turbine-caused avian fatality estimates Journal of WildlifeManagement 741089-1097 httpdxdoiorg1021932009-266

Smallwood K S and C Thelander 2008 Bird mortality inthe Altamont Pass Wind Resource Area California Journalof Wildlife Management 72215-223 httpdxdoiorg1021932007-032

Sterner D S Orloff and L Spiegel 2007 Wind turbinecollision research in the United States Pages 81-100 in M deLucas G F E Janss and M Ferrer editors Birds and windfarms risk assessment and mitigation Quercus MadridSpain

Stewart G B A S Pullin and C F Coles 2007 Poorevidence-base for assessment of windfarm impacts on birdsEnvironmental Conservation 341-11 httpdxdoiorg101017S0376892907003554

Tews J D G Bert and P Mineau 2013 Estimated mortalityof selected migratory bird species from mowing and othermechanical operations in Canadian agriculture AvianConservation and Ecology 8(2) 8 httpdxdoiorg105751ACE-00559-080208

Van Pelt T I and J F Piatt 1995 Deposition and persistenceof beachcrest of seabird carcasses Marine Pollution Bulletin 30794-802 httpdxdoiorg1010160025-326X(95)00072-U

Zdawczyk M E 2012 Assessing the potential avoidance ofwind turbines by migratory birds over Bowling Green OhioThesis Bowling Green State University Bowling GreenOhio USA