The URL can be used to link to this page

Ord No 21-21-2411and ORDINANCE NO. 21-21-2411 An Ordinance amending Section 20-2.3 Definitions and Section 20-3.6 Supplemental Regulations of the City of South Miami Land Development Code to add provisions regarding the use of bird-safe design features. WHEREAS, birds provide valuable and important ecological services; and WHEREAS, the City of South Miami has been designated a Bird Sanctuary since 1927; WHEREAS, birding is a hobby enjoyed by 46 million Americans with an annual $107 billion total industry output in the United States; and WHEREAS, as many as one billion birds may be killed by collisions with windows every year in the United States; and WHEREAS, new buildings can be designed to reduce bird deaths from collisions; and WHEREAS, after a public hearing on July 13, 2021, the Planning Board's motion to approved the ordinance failed by a vote of two (2) ayes to five (5) nays; and WHEREAS, the City Commission desires to adopt regulations for the incorporation of bird-safe features in new construction and renovations. NOW, THEREFORE, BE IT ORDAINED BY THE MA YOR AND CITY COMMISSION OF THE CITY OF SOUTH MIAMI, FLORIDA: Section 1. The foregoing recitals are hereby ratified and incorporated by reference as if fully set forth herein and as the legislative intent of this Ordinance. Section 2. read as follows: * * * Land Development Code Section 20-2.3 Definitions is hereby amended to Bird Activity Zone: The zone that falls between 0 -100 feet above grade. * * * Bird-Safe Glazing Treatment: Bird-safe glazing treatment may include fritting, netting, permanent stencils, frosted glass, exterior screens, physical grids placed on the exterior of glazing or UV patterns visible to birds. To qualify as Bird-Safe Glazing Treatment vertical elements of window patterns should be at least Y4 inch wide at a minimum spacing of four inches or horizontal elements at least 1/8-inch-wide at a maximum spacing of two inches. Page lof3 Ord. No. 21-2 1 -24 11 * * * Section 3. fol lows : Land Development Code Section 20-3.6 I S hereb y amended to read a s * * * (X) Bird-Friendlv D es ign. ill Glazing . (i) In a ll zo ning districts other than RS and RT, one hundred percent of the window and door g lazing in the Bird Activity Zo ne must use Bird-S afe G lazing Treatment. Other gl aze d areas within th e Bird Activity Zo ne that have unbroken glazed segments twenty-four square feet and larger must al so be constructed us in g Bird-Safe G la z ing Treatment. Areas abo ve the Bird Acti vity Zone do not need to use Bird-Safe G lazing Treatment regardless of size. (2) Lightin g. Lighting sha ll be s hield e d. No up -li ght in g s hall be used . Searchlights are pro hibit ed. Section 4. Co rrections. Con forming language or technical s cri ve ner-ty pe co rrection s may be mad e by the C it y Attorney fo r an y co nformin g amendments to be in co rp o rated into the final ordinance for sig nature. Section 5. Codification. The provi s ions o f this ordinance are to beco me and be mad e part of the C ity of South Miami Code of Ordinances as amended ; th e sect ions of tlli s ordinance ma y be renumb e red o r re-Iette red to accomplish s uc h int enti on; and th e word "ordin a nc e" m ay be ch an ge d to "sect ion " o r o ther app ropri ate wo rd. Section 6. Severability. Ifan y sec tion , clause , sentence , or phrase of thi s o rdinanc e is for any reason held in va lid or unc ons titution a l by a co urt of com pe t ent jurisdiction, thi s holding wi ll not affect th e va lidity of th e remaining portions of thi s ordinance. Section 7. Ordinances in C onflict. A ll o rdinance s o r part s of o rdin ances and a ll sec tion s and pa rt s of ord in a nc es in direc t conn ict he rew ith are hereby repea le d. Section 8. Effective Date. This o rdinanc e is effect ive up o n e nactm e nt. PAS SED AN D ENACTE D T HIS ~ da y of Sep te mbe r, 202 1. Arr EST: A PPROV ED: Page 2 of 3 Ord. No . 21-21-24~1 1st Reading: 8117 /21 2 nd Reading: 917 12 1 READ AN D APl'R OVED AS TO FORM , LANGUAG EXECU~T;;1 'A-~~ COMM IS S ION VOTE: Mayor Phi l ip s : Com mi ssioner Corey : Commissioner Gil: 4-1 Nay Yea Yea Com mi ss io n er H a rri s: Yea Commissio n er Liebman: Yea Page 3 of 3 City Commission Agenda Item Report Meeting Date: September 7 , 2021 Submitted by : Jane Tompkins Submitting Department: Planning & Zoning Department Item Type : Ordinance Agenda Section : Subject: Agenda nem No:21 . An Ordinance amending Section 20-2 .3 Definitions and Section 20-3.6 Supplemental Regulations of the City of South Miami Land Development Code to add provis io ns regarding the use of bird -safe design features. 3/5 (Mayo r Philips ) Suggested Action: Approval Attachments : Cover M e mo re Bird Safe De sign 0907202l.docx PB-2l-009 Final PB Regular Meeting Minute s Excerpt -07-13 -202l.pdf Bird-F rie nd Iy -Bui I ding-Design . pdf Loss Scott Bird Building Collisions . pdf Vi racon email re availability.docx Viraco n email re cost .docx 1 THE CITY OF PLEASANT LIV I NG CITY OF SOUTH MIAMI OFFICE OF THE CITY MANAGER INTER-OFFICE MEMORANDUM To: Hon o r ab le Mayor and Members of the City Co mmissio n VIA: Shari Kamali, City Ma nager FROM: Jane K. Tompkins, AICP, Pl annin g Director DATE: August 17, 2021 UPDATED September 7,2021 SUBJECT: An Ordinance amending Section 20-2.3 Definitions and Section 20-3.6 Supplemental Regulations of the City of South Miami Land Development Code to add provisions regarding the use of b ird -safe design features. BACKGROUND: Collisions wit h gla ss kill millions of birds eac h year. Accor din g t o one report!, between 365 and 988 million bird s are lik ely killed in th e U.S. each year as a result of co lli sion s w ith buildings. The annua l bird mortality at residences is est im ated to be between 159 and 378 million birds; between sixty-two (62) and 644 milli on birds at buildings from four (4) to e le ve n (11) sto ri es ta ll ; an d betwee n 104,000 and 1.6 m illion birds at hi gh-r ise buildings more than twelve (12) sto ri es tall. Building co lli sion s are second only to feral and free -rangin g pet ca t s as the largest human- ca used mortality for birds in the U.S " Ther e are two r easons bird s hit w ind ows. First, they do n't recognize glass ; th ey don't learn the same v isua l clues (s uch as door frames and even dirt) as humans that help t h em d et ect glass. Consequently, r efl ect i ons of trees, shr u bs or the sky are perceived to be r ea l and birds try to reach them. Al so, beca use their eyes are placed more on the sides of their heads, th ey have little de pth pe r cept i on but ex t ens ive fields of view to the side and be hind . Their focus is thus not necessar il y o n w hat's in front of them 3 . Bird strikes can be prevented through the in corpora ti o n of certain d es ig n feature s. Essentially, t he goa l is to create v isua l obst ru ct io ns th at th e bird und ersta nd s can't be passed throu gh. In , Loss . Scott R .. Sara S. Loss. Pete r P. Marra and Tom Wi ll . 2014. Bird-build ing Co lli sio n s in the United States: Est im at es of Ann u al Mortality and Spec ies Vuln erab ility. Condor 2 American Bird Conserva n cy . (2015). Bird Friendly Building Design. 3 American Bird Cons ervancy. 2 Loe Amendment re bird-safe desisn September 7,2021 Page 2 of 2 new construction, this is generally done with glass that has ceramic dots, other types of "frits," and other materials silkscreened or otherwise applied. Research by the American Bird Conservancy shows that patterns covering as little as five percent (5%) of the total glass can deter most strikes under experimental conditions.4 Most birds will not attempt to fly through horizontal spaces less than two inches high or through vertical spaces four inches wide or less. Patterned glass is most effective in preventing strikes because they are always visible, even with strong reflections. Other techniques for preventing strikes include window screens, netting, grills, louvers and exterior shades. Awareness of this problem has been growing in the u.s. New York City and San Francisco have adopted bird-friendly design requirements and Congress has reviewed legislation, the Bird-Safe Buildings Act, that would require public buildings incorporate these design features. Locally, several buildings that are in the design-phase or are already under construction include bird safe glass. These include a residence hall at Florida International University and several buildings on the University of Miami and the University of Florida campuses. Mayor Philips requested the ordinance be prepared. ANALYSIS: The proposed ordinance adds a requirement that bird-safe design features be incorporated into new construction. It's based on a model ordinance prepared by the American Bird Conservancy (ABC) and an ordinance adopted by the City of San Francisco in 2011. The two documents take very different positions in their applications. The San Francisco ordinance only requires bird-safe glass in areas near open spaces of two or more acres and on certain features (such as skywalks and balconies) that have unbroken glazed segments twenty-four (24) square feet or larger. If the City followed this approach, the requirement would only apply to areas around some of our parkss, some balconies and perhaps the proposed pedestrian bridge. San Francisco also exempts residential structures that are less than forty-five (45) feet in height and that have less than fifty (50) percent glass on a fa~ade. In South Miami, single-family homes are limited to twenty-five feet (25') in height so only facades that are more than fifty percent (50%) glass would have to include bird-safe glass. In staff's observation, most homes do not exceed this threshold and would thus qualify for the exemption. The ABC ordinance takes a more stringent approach and requires one hundred percent (100%) bird-safe glazing from grade up to one hundred (100) feet in height on all new construction and major renovations. In South Miami, the maximum building height in all but the TODD (MU-6), (MU-M), and the Downtown SoMi (OS) zoning districts is one hundred (100) feet. Adopting the ABC approach would that require almost all construction and renovation include bird-safe design. 4 Bird Friendly Building Design" published by the American Bird Conservancy, 2015. 5 The following parks are greater than two acres: Dante Fascell, Fuchs, Marshall Williamson, Murray, Palmer, and South Miami. 3 LDC Amendment re blrd-safe design September 7,2021 Page 3 of 2 Only those portions of buildings above one hundred (100) feet in those three districts would be exempt. The proposed ordinance presents something of a compromise between the extreme positions presented above. In the residential districts, bird-safe glazing treatment would only be required on glazed segments that are twenty-four (24) square feet or larger. In a typical home, a sliding door would exceed this size, but a window would not. In all other zoning districts, one hundred percent (100%) of the glazing below a height of one hundred (100) feet must be treated. Also, all areas that have unbroken glazed segments twenty-four (24) square feet and larger would require bird-safe glazing treatment. Bird-safe glazing treatment is defined to include fritting, netting, permanent stencils, frosted glass, and exterior screens or grids. Certain dimensional standards apply to the patterned treatments. The cost of bird-safe glazing treatments varies depending on the type of treatment (e.g., exterior screens versus patterned glass). According to one manufacturer, the additional cost to apply a silkscreen pattern to the type of glass typically used in South Florida is about $7/sq. ft., or about twenty-eight (28) percent over the base cost of the glass. In addition to the glazing treatment requirements, the ordinance also requires that lighting be shielded and prohibits up-lighting and searchlights. Many bird species migrate by night. Thus, these restrictions are included because artificial lighting seems to disrupt birds' orientation mechanisms and cause them to deviate from their flight paths. As they fly near light sources, they may become disoriented and land in undesirable locations. PLANNING BOARD RECOMMENDATION: After a public hearing on July 13, 2021, the Planning Board voted against a motion to approve the item. Two members voted in favor of the ordinance, and five members voted against it. CITY COMMISSION FIRST READING: During the first reading of the ordinance on August 17, 2021, Commissioner Liebman requested that answers be provided to four questions. 1. Validate the number of birds killed by glass strikes. Response: The studies referenced in the footnotes are attached. 2. Validate the cost of the glass. Response: See attached email from Jeff Rigot with Viracon Architectural Sales re the cost. 4 LOC Amendment re bird-safe design September 7,2021 Page 4 of 2 3. Is the glass difficult to procure? Response: See second attached email from Jeff Rigot regarding availability. 4. Can it be hurricane safe glass? Response: See second email from Jeff Rigot confirming that the screen frit that makes a window bird-safe can be applied to hurricane resistant glass. Please also note that bird-safe glass is only one option for complying with the code. Based on the proposed definition, other techniques such as window screens are also permitted in lieu of the specially treated glass. STAFF RECOMMENDATION: Staff recommends approval of the proposed ordinance. Attachments: • Draft Ordinance • City Notice of Public Hearing • Legal Ad • Excerpt of Planning Board meeting minutes • loss, Scott R., Sara S. loss, Peter P. Marra and Tom Will, 2014. Bird-building Collisions in the United States: Estimates of Annual Mortality and Species Vulnerability. Condor • American Bird Conservancy. (2015). Bird Friendly Building Design. • Two emails from Jeff Rigot with Viracon Architectural Sales 5 CITY OF SOUTH MIAMI PLANN ING BOARD Regu l ar Meeting Minutes Exc erpt Tuesd ay, July 13, 2021 CITY COMMISS ION CHAMBERS 07 :00 P.M. The City of South Miami Code of Ordinances, Section 8A -5, requires that aI/lobbyists, as defined in thot section, must registe r with the City Clerk before engaging in any lobbying activities and in most cases pay an annual fee of $500.00 and an additional $100 for each additional issue. This applies to all persons who are re ta ined with or without compensation to influence any action, decision, recommendation of someone with the city, including the city manage', city attorney, department heads, city personnel, or members of the city commiss ion or members of any city board, concerning a matter thot could joreseeably be address by the city commission or a city board. The re are some exceptions and exemp tions. The following are not considered to be lobbyist: a representotive of a principal at 0 quasi-judicial hearing, experts who present scientific or technical information at public meetings, representatives of a neighborhood association without compensation and representatives of a not-for-profit community based organization for the purpose of requesting a grant who seek to influence without s pecial compensation. Individuals who wish to view or listen to the meeting in its entirety, audio and video versions of the meeting can be found on the city's website (www.southmiamifl .gov). I. Call to Order Action: Ms . Bonich ca ll ed th e meeting to order at 7:02 P.M . Mr. Pepe pro v ided the Board and pu bli c w it h the rules for the mee t in g. II. Roll Call Board Members Present Constituting a Quorum: Ms. Li sa Bo nich (Chairperso n), Mr. Jay Mi ll er (Vice- Chairper son), Mr. Orlando Borges, Mr. Lee Jacobs, Mr. Subrata Ba su, M s. Michelle Readon , and Ms . Guirla H. Dodard . Board Members Absent: None City Staff Present: Ms. Shari Kamali (City Manager), Ms. Jane Tompkins (Plannin g Directo r), and Mr. Marcus Li g htfoot (Senior Planner/Zoning Administrator). City Staff Absent: No ne. City Attorney: Mr. Thomas Pepe. 1 9 III. Administrative Matters 1. Introduction of New Planning Board Member Ms. Bonich introduced the Board to Ms. Michelle Readon and Ms. Guirla H. Dodard, the newly appointed Planning Board members. IV. Public Hearings: 1. PB-21-009 Applicant: The City of South Miami An Ordinance amending Section 20-2.3 Definitions and Section 20-3.6 Supplemental Regulations of the City of South Miami Land Development Code to add provisions regarding the use of bird- safe design features. Applicant: The City of South Miami, represented by Ms. Jane Tompkins, AICP (Planning Director) Ms. Bonich read PB-21-009 into the record. Ms. Tompkins presented the item to the Board. Mr. Borges inquired about who requested that the item be drafted, to which Ms. Tompkins responded that it was Mayor Philips. Ms. Bonich asked if there were enough bird deaths outside of San Francisco, New York City and the City that would trigger the creation of an ordinance that would place an additional cost on the installation of glass in the City for residents. Mr. Pepe stated that when he investigated installing the glass on his own residence, he looked at using the least reflective glass and haven't had any bird strikes. He then stated that because the more energy efficient glass is typically more reflective, they are prone to more bird strikes. Based on Mr. Pepe's statement, Mr. Miller responded that the ordinance may be lacking information since the Ordinance makes no mention of less reflective glass. He then stated that because none of the Board members have not experienced any bird strikes, this ordinance may be a waste of time. Mr. Borges asked what the increase in cost would be, to which Ms. Bonich responded that it would be an increase of 28% or more for windows and doors. Mr. Pepe added that the increase in cost is for the glass only, not the glass and frame. The Chairperson opened the floor to public comments on PB-21-009 . • None The Chairperson closed the floor to public comments on PB-21-009. 2 10 Mr. Jacobs informed the Board that he has had birds strike his windows in the past. Mr. Jacobs then stated that a lot of migrating birds come through South Florida. Because of that, it is a problem. With no further discussion, the Board made a motion on the item. Motion: Mr. Jacobs moved to approve PB-21-009 as presented. The motion was seconded by Mr. Mr. Basu. Vote: Yes 2. No 5 (Readon. Miller. Bonich. Dodard. Borges) Ms. Readon: No Mr. Miller: No Ms. Bonich: No Ms. Dodard: No Mr. Borges: No Mr. Basu: Yes Mr. Jacobs: Yes The motion to approve PB-21-009 failed to be approved by the Board. V. Public Comments/New Business The Chairperson opened the floor for public comments and any new business. Public Comments Section There were no Public Comments. New Business Section There was no business discussed. The Chairperson closed the floor for public comments and any new business. VI. Approval of the Minutes 1. Planning Board Regular Meeting Minutes of May 11, 2021: Motion: Mr. Miller moved to approve the meeting minutes as presented. The motion was seconded by Ms. Bonich. Vote: Yes 5. No 0 (None) Mr. Borges: Yes Ms. Bonich: Yes Mr. Miller: Yes Mr. Basu: Yes Mr. Jacobs: Yes Ms. Dodard: Abstain Ms. Readon: Abstain 3 11 The motion to approve the meeting minutes was unanimously approved by the Board. VII. Future Meeting Date: August 10, 2021 VIII. Adjournment The meeting was adjourned at 10:18 P.M. 4 12 ':0. V:US(J r's Bridge fo r I.a bo rat ory Scie nces, s hown here unde r const ru ction In Octoher 20 1S. The hulld lng i< sc h eduled to open InJanuary 20 16 . Co\'c r rende ring and photos co urt esy or Enn ea d Arc hit ecls , • , II • - Cover re nderi n g and p h oto t h is P "8e: The n ew Bridge fo r l .. 'l boratory Scie n ces bu ild in g nl Vassa r Co lleg e. d es ig ne d b ~' Ri ch"rd Olcon /Ennead Ar c hilects, red e fine s the id e ntity o ll h e 'ici l!n ces o n t h e Coll ege's hi s to ri c Ci l l1lpU'i and provi d es tech nolog ically adva n ced facill t ic!I for '!tudcnls. facu lty, and resea rch e rs . Fundame ntal to the b uild in g's d es ig n is its ~e aml ess int egratio n with the natura l landsca pe . scal e, and cam p us .. es t hetic of t h e Coll e ge , In t hi o;: n at ura l wood ed se ttin~, the n eed fo r strategi es to re duce h ird colli s io ns with th e b uilding wall apJl ilfc n t. In res po nse, t h e build in g Wil \ d es ign e d to comply wit h LEE D Pil o t C redil 55: Sird C.o lll \lo n J)N c rrl'nce, Ennead m anaging partne r G uy Max we ll is it n a li o n al1 y recog ni zed c ham p io n of b ird-fri end)}' des ign and has led Lnncad\ inlltlV allvc approac h 10 make the b uil ding'S g laz ing safe r fo r birds, e m pl oying pa tt ern ed g l<ls'>, SW ..'l'n s a n d \umhades, .U1 d Ornilux g lass , a speciillty gln ~s product thai uws a U\' coatin g vi sihh' to h ln l<; h ut 11 01 hu n lil n ~. By fr a m in g an d s h ()w('a ~i ll g vi ews of Ill e In lldsc;lp c , the hu i ldin g u'!ehra l l'\ <lnd COll llt'C t S \Iutl c ll t s w ith the surro und i ng c ll\'lro nmc nl , wh il e th e ove rall d e ve lopmen t o f the p recinc t rc.'purpuscs .In unde rulili ze d :;e cll1r of c all1 pll ~, 14 Exec utive Sum ma ry Introdu cti o n .. W h y Bird s Maller The Legal Landscape Gl a ss: The In visi bl e Thre al... .. .4 .... 6 .... 7 ................. ... 7 ... 7 Lighti n g: Exace rbat ing the Threat .. 8 Bi rds and t h e Huitt Environ m en t .. . ... 8 Impact o f Co JI L~ions o n Hird Pop ulati ons.. . ...... 9 Bird Co lli sions and Sustaina hl e Arc h itecture ............ 9 Defin in g W h at's Good For Bi rds Problem : G la ss . ................. 1 1 ... 12 Pr operties of G la ss Henect io n Tra ns pare n cy ................ . .......... 1} ............................. 13 ...... 13 Ula c k Hole or Pa ss ag e Effect ................. 13 Factors Affecti ng Ha tes of Hird Co lli sio n s ..... 14 for a Pa rtic ular Huil di ng Bu il ding Des ign .................... . ...... 14 Building Sile ... ....... 14 Orienta ti o n and Siti ng ...... . Time of Day .................. . Green Roofs and Wall s ............................. . .. 14 .. 16 . 16 So luli ons: G lass ... IS Ne tt in g, Scree ns, Grilles, Shutters, E.x terior Sha des ..... 19 Aw nings and Overhangs ...................................... 2 0 Angled Gl ass............ . ......................... 20 Patt e rn s o n Gla ss .... . UV Pa Uerned Glass ....... . Opaque and Translu ce nt Gl a ss Window Fil ms So luti o n s Appli ed to Int e rior G la ss Decals and Tape ............................ . Temporary Solutio n s ................. . ............... 2 0 ........................ 22 ......................... 22 .. 24 ....... 24 ... 24 .............. 20 ltemediation Case Study: Javits Ce nt e r ... 27 Ught : Pro b le m s and So lu ti ons So lution s .... Li g ht s Out Programs ................. . ...................... 2H ........... 3 0 . .. 31 Solu ti ons: Policy ............. . Leg is lation Pri orities for Policy Direc ti ves ... Sustainabi li ty Haling Prog r<1lll s . Mode l Ordinance .... Th e Sc ie n ce o f Bird Co lli sion s ................ . Magnitud e of Colli s ion Deat h s .. Patte rn s of Mortalily ........ . Species a t Risk ..... . ...... 32 ...... 33 . .... 34 ....... 34 ......... 35 . ..... 36 ........... 37 ............. 38 ... 38 C h aracte ri s tics o f Buil d in gs ............................. 39 Amount of G la ss .............. . T im e of Da y .. Local Landscape Avia n Vis io n a nd Colli sions ......... 39 ........ .40 .40 ...... .4 1 Av ia n Orie ntation and the Earth's Mag netic Fi e ld .. .42 Bird s and Li g ht Pol luti o n .............. . ......... .4 2 Li gh t Colo r a n d Avia n O ri el1l"al"i o n ..... . ..... .44 . ... .45 Hese arch : Dete rrin g Colli sio n s T h e 2 x 4 Hule ........................... . Eva luating Col li sion Pr oblems ........ . -A Toolkit fo r Building Ow n ers So luti ons .... Se a so nal Timi n g ... Wcat he r .... Diurnal T iming ................... . Loca ti on. Local Bird Po pulations .......... . i'os t·Miti gation Mo nitorin g . Refere n ccs ..... Acknowledglllenis .. Discl aim e r AHC's Hir d ·Fric n dly Building Standard ................. .47 . .... .48 .... .49 . ....... 5 0 ........ 50 .............. 50 .................... 5 0 ......... 5 1 .. 5 1 . ........ 52 .. 57 ....... 57 . ... 59 The Mea of gl as s a ll a fa~-ade is the stronge..~t predictor of threa t to birch . Then' are also other reason~ to limit glas s. Skidmore Owings Mt:rrll 's Bronx, New York , EmCIgenc), Call Ce nter is 3 handso me e....:ample of creative deSign with re s tri c ted glass. for a building inte n ded to be hoth secure and blast-re sis tan t. Photo by Chrl~ Sheppard . ABC For updates a nd n ew inform ati o n , see collisions.abcbirds.org 15 A bird. prohah!y a dove, hit Ihc window of an Indiana home hard enough 10 !ca\'e this ghost I)" image on th e glass. I'holo hy David F,lIlchl'r Executive Summary Co lli s io n with g lass claim s th e lives o f hundreds o f milli o n s of birds each year in til e Un it ed Sta tes. It is second o n ly to domestic cats a s a so urce of morta lity linked direct ly t o 11uman actio n . Bird s that h ave s li ccesshill y flow n t h ousa nd s o f mi les o n mi g ration can die in seco nds on a p ane o f glass; im p acts kill fledgling s b efo re th ey CCl n truly ll y. Because g la ss is dallge ro u s for s trong, h ea lt h y, breeding ad ul ts, as well a s sick o r yo ung birds, it call ha ve a pa rt ic ular ly ser io u s im pact o n populati o n s. Bird ki ll s occur at build in gs across t h e United States and a rou nd the world. We know m os t a bout morta li ty patter n s in c iti es, because that is w h e re most monitoring takes p lace, b u t v irtuall y <1I1y bui lding w ith g lass poses a threat whereve r it is. Th e dead b i rd s d ocumented by m o nito rin g p rog ram s o r prov ided t o museu m s co n s titu te merel y a fra c tion of the bird s actua ll y kill ed. The mag nitude of t hi s prob lem C(l n be discou rag in g , but there a rc a lready effec ti ve sol u t io n s and an increasi n g co mmerc ia l co mm it m e nt to d eve loping n ew so luti o n s, i f peo p le can be co n vinced t o adop t t h em. That mtHicial li g htin g at ni ght plays a Sig nifica nt part in m o rt a l ity fro m g la ss is wide ly accepted , but ofte n mis u nde rstood . Th e major it y of co lli s io ns with buildings take p lncc dur in g daylight. Th e re arc many well -docu me n ted inst a n ces of b ri g h t li g h ts at ni g ht d iso r ienting large numbers o f b ir d s-u sua ll y n ig ht- mi g rati n g passer ines b ut also seab ird s-some of wh ic h m ay c ir cle in th e li g h t, so m e time s until daw n . Nocturnal m o rta lity associated with ci rc ul ation events is ca u se d by col li sion wit h guy w ir es and oth e r st ructures . Such even t s we re d esc rib ed sta rt in g in the late 19th century at li g hth ouses, a n d la ter at the Washington Mo num e nt, Statu e of Libe rt y, and Empire Sta t e Building, w hi c h were the o nl y b ri g htly lit st ru ctures i n the ir a reas. T oday, s uc h eve nts occur mos t ly at offs h ore d rillin g p la t fo rm s a nd co mmuni catio n towers. These situ a ti ons h ave in com m on bright li ght surro und ed by d arkness, and th e i r freque n cy h a s d ec reased in cities as areas o f d arkness arou nd b ri g ht stru c tures h ave a lso b eco m e li t. However, there arc stro ng in d icati ons that birds a re s tili being di so riented by urban lig hts and th at li g hts arc lin ked to mor tality, eve n th o u g l, m o rta li ty patterns ha ve c h a n ged. Adva nces i n g la ss techno logy and production since the mid-twen tie t h ce ntury ha ve m ade it possi bl e t o co n s truct s ky sc ra pe rs wit h a ll -glass wa ll s, homes wit h huge picture w ind ows, and mil es of tran s paren t n o ise- bar ri ers on h ig h wa ys. Th ere ha s bee n a gene ral in crease in the amo unt of g la ss us ed in co n stru c ti o n -a n d t h e amou n t of g la ss on a bu il ding is t h e bes t" p red ic t or of : =-----;:;:;;~-;;=;:;;=:~ -- -:::;er==:! r... _ ... ..:... .. F --... :;.: ............. _. --... - Ne who u se Ill, design cd by I'olshek Pa rtnerShip Arc h itec ts, i s pari of Syra(use Ul1 ivc rs lt}"s S.1. Ncwhousl' Sc h ool of l'ubli ( Com munications. T his h uilding in corpo rales an umlllialin g, frilt ed g la ss {"(ild e wi th t h c word s o f Ihe fir st amcndmelU e tched in leHe rs s ix fe e t hi g h ill o n g Ih e basc. Photo by C hr is tin c Sheppa rd, AII C 16 the numbe r o f bird s it wi ll kil l. However, wh il e g lass is im po rtant for bringing light in to b uildin gs, a fa~ade with over 30-4091, g lass dramatica ll y increases e n eq;y use for heatin g a nd coo tin g. Bird-frie n d ly d esig n is becoming re cogni ze d as par t of sustainable desig n, requ ir e d inc reas i ng ly by leg islat ion across Nort h Am e r ica. New co n stru ct ion can in corpo rate from t h e begi nning bird-friend ly design strategies that ar e cost n e u tral. Th e re a rc many ways to red u ce m o rt ality from existi n g buildin gs, w ith more so lutions being developed a ll the time. Be ca u se the sc ie n ce is co n s tant ly evo lvi ng , a n d beca u se we wi ll a lways w ish for more informatio n than we h ave, the temptatio n is to postpone action in th e hope that a panacea is ju s t around t h e corner. But we ca n 't wa it to act. We have t ile tools a n d t ile st rateg ies to make a difference n ow. Arc h itects, designers, city The s teel mesh clH'e lo ping Zurich's COcoon in Switzerland, designed b~' Ca men zinti Evolution , I.td, prm'ides privacy, r educes heating and cooli ng CO\t~, and r r otect~ hinis, but s till permits ()cc up o\!1 '~ tn SCI.' nut. Photo h~' Anton Volgger pla nn e rs, and leg is lators are key t o so lv ing thi s problem. Th ey n o t o nl y h ave access t o til e latest building co n stru ctio n m a t er ials and concepts; they are a lso thoug h t leaders and tre n d se tters in the way we b uil d o ur co mmunities and p ri ori ti ze bu il ding design iss u es. Thi s p ubli cat io n aims to prov ide p lanne rs, arc hitec t s, and deSigners, bird advocates, and loca l, mun iCipa l, and fed- e ra l author iti es, as we ll as the ge n e ra l pub li c, with a dear un ders t a ndi ng of t h e n ature and magnitude o f th e threat glass p oses to birds. Si n ce th e first editi on, in 201 1, th e re ha s been increased awareness of collisions, ev idenced by n ew ord inan ces a nd g u idelin es for b ird-friend ly co n st ru c- ti on, new material s Lo re t rofi t existing build in gs, Clnd pro- moti on by th e g lass indu s try of bird-frie ndl y ma te ri a ls. Th is ed iti on includes an updated revi ew of the un derly- ing scie n ce, examples of so lution s t h at can be app li ed to both n ew co n stru ction and existin g bui ldings, and an explanation of what info rmati o n is still n ee ded. We h ope it w ill sp ur ind iv iduals, b u s inesses, comm uni t ies, SC i e nti st s, and gove rnments to address t h is issuc and make their bu il dings safer fo r b ird s. Const ru cting bird- fri e n d ly build ings a n d e limi nating the worst existing threats req u ir e only im agi n ative desig n , effcctive retro- fits , and recogniti o n that birds h ave in trin s ic and cultur- a l as we ll as eco nom ic and ecological va lu e to human ity . Ame ri can Bird Co n servancy's Co lli sio n s Pr og ra m wo rks at t h e n at io n a l levcl to redu ce bird mortality by coo rdinatin g w ith orga ni za tions and gove rnm ents, developing educational programs a nd too ls, eval u a t in g and deve loping so lu t ions, crea tin g ce n tra li zed reso urces, a n d ge n e ratin g aware n ess. The fa~ad e of Sauerbruch Hutt o n's Brandho rst Mu se um is a brillian t ex ampl e of m ixing glass and non-glass material s. Ph oto by To n~' Brady Why Birds Matter For m any people, birds and nature ha ve intrin sic worth. Birds have been im portant to hum ans throughout histo ry, ofte n sy mbo li zing c ul tu r;a l va lu es su c h a s p eace, f reedom , and fidelit y. I n addition to th e pleasure th ey ca n bring to peop le , we depend on them for c ritica l eco l ogica l functions. Birds consume vast quantities of insects a nd control rodent populations, reducing damage to crops and fo rests <:Ind h elping limit" the triln s mi ssio l1 of diseases s uc h as West Nile virus, dengue feveT, a nd malaria. Bird s playa vita l role in regenerating hab itats by pollinating plants and dispersing seeds. Birds arc also a direct economic resource. Accordi n g to the U.S. rish and Wildlife Serv ice , bird watch ing is o n e of th e fa stest grow in g le is ure activities in Nor th America, an over $40 billion industry accounting for many jobs. The Legal Landscape At the start of the 20th c entury, fo ll ow in g th e ext in c- tion of the Passenger Pigeon and the n ear ext in ction of other bird s pecies due to unregulated hunting, laws were passed to protect bird populations. Among them was the Migrato ry Bird T rea ty Act (M IlT A), w hi ch made it ill ega l to kill a migratory bird without a permit. The scope of this law, w hi ch is still in effect today, extends beyond hunting, s uc h that a n yo n e causi ng the death of a mi gra- tory b ird, even if unintentionally, ca n be prosecuted if that d ea th is d ee m ed to have been foreseeab le. At pres- e nt, th e scope of the MBTA is und e r c hall e n ge in federal court and it is im possible to say whether it wi ll ever be used to c ur b g la ss co lli sions. Howeve r, courts in Canada have ruled that building ow n e rs arc respo n sib le fo r m or- tality ca use d by g lass. Vio lations of th e MIlTA ca n res u lt in nn es of up to S500 per incident and up to six months in prison. The Bald (O p po~ite) The Whil e·tJlr o at ed Sparro w is the IlI OU fr cq uc rH "Ic tlm o f "olll si ons rCIKJrt l'd by milan lTl o nltorin.: program s. I'hoto hy Itolx-rt Ito)'se and Gol d c n Eagle Protecti o n Act (orig in a ll y the Ba ld E.'gle Pr o tecti on Act of 1940), the Enda n gered Species Act (1973), and the Wild Bird Conserva ti on Act (1992) provide further protect io n s fo r b ir ds that ma y apply t o building co lli s ions. Rece nt Icgis l"lt'i o n , p rima r il y at th e city a nd state level s, h as addressed the problem of mor- tality from buildin g co lli sio n s and li ght polluti o n . Start- ing with Toronto, Canada, in 2009 and San FranCiSCO, Ca lifo rnia , in 20 10 an increasing numbe r of s t~lte s and muni c ipalities have passed taws mandating bird -fr ie ndl y design, whi le oth e r author ities h ave passed vo lunta ry measures . Glass : The Invisible Threat G lass is in v is ibl e t o both birds and hum ans. Hum ans learn to see glass through a combination of experien ce and vi s ual c u es like mullions and even dirt, but bird s are un ab le 10 lise Ihese Sig nal s. Most birds ' first encou nte rs w it'h g lass are fatal w h en Ihey collide wi I'l l iL at fu ll flight s peed. As pects of bird vision contribute to the problem. Whereas human s have eyes in th e front o f th e ir h eads and good depth perception, most birds' eyes are pla ced at the sid es of th e ir heads. Birds thu s have littl e d ept h perception beyond the ra n ge of their bi ll s but extensiv e fields of view t o the sid e and behi nd . T h ey judge their night speed by th e passin g of objects to their s id es, so th e ir fo c ll s in flight is n o t n ecess(uil y ahead. Bes ides s im- p ly u s in g designs w ith Icss g lass, we can protect bi rd s by us in g sc rcens, shutters, and details th at partl y o bsc ure g la ss while sti ll providing a view, or by using two-di- m ens iona l patterns th a t birds perceive as actua l barriers. However, b ir ds ha ve poor co ntrast se n s itiv ity co mp.Hc d to humnns: s h apes at a d ista n ce m e rge into a b lur at closer range for b ir ds. Thi s m ea n s that m ost sig nal s that make g lass safe for birds wi ll probab ly be read il y v isible to peop le. Refl ec ti o ns o n h o me window s arc a sig nlfi canl source o f bird m o rtalit y. The partially o pened ve rli ca l blinds here Im lY hreak lip the lencc tl o n en o ug h 10 redu ce Ih e hazard 10 hlrds . Ph o lo by Chrhtlnc: Sheppard . An c . ,. , ........ -.-I -,,-,-' -. .. "..,... .• -I ' ~ . , ...... i\4 ~·S ..', III I' .: :', i :J'.:~.T i ~ ~ . 'J 'it ll·· 't· , I ,. ~ ~ '~~ . -;. ~ ~ ~ ~'I' .' 1 ~; ~.~·" .. ~l .":' , :. ~. trA' I . :-*, ' ,' .. ~ ~.:J . : !-:_,_";.:, .~-4. ...- ::. -:. " .... -: ;;.-.-: -',---.. -_ .. _. __ . ~.-.. ~ .. ---.--- IIirds ma y try to rea c h vegetatl o n seen Ihro ugh IWO or more gla ss wall s or windows; th e s ingle decal h ere Is n o l enoug h to so lve Ih e pro bl em . bul two o r threc cou ld d o the trick . Photo by Christine Sheppard. AII C U g ht ;1\ ni g ht can di sori e nt b ird s, amI the problem is n o t restrict ed ('0 ta ll b uil din gs . This sce n e o( t Ol S Vega s h y flight d e p ic ts ;I tlHC;1\ 10 ;'l n y bi rd m igr'll in g nca r b ~r at !light. l'h o to by IIr C-Il th:J Si gnatu lc, \Vikiptdia Lighting: Exacerbating the Threa t Most birds, with obv io u s excepti o n s, a rc a c tive by day, with eyes best adapt ed for day li g ht s ight. However, m a n y bird s pec ies migrate by night, allow ing them to lise da y li g ht h o urs fo r f ce d ing. We st ill d o n 't kn ow eve ry thin g ab o u t how ni ght-flying b irds navigate. We do know that birds probab ly h ave two special sc n ses t hat a ll ow th e m to d e termine loca ti on and d irect io n u sin g th e Earth 's m agn eti c fi e ld. O n e o f th ese, located in t h e eye, may a ll ow birds to "sec" m agnet ic lin es in th e p rese n ce of dim b lue li g h t. Sta r ma ps, la ndmarks, a n d o ther mec h an is m s a rc a lso invo lved. Artificia l ni ght li g htin g see m ing ly d is rupts o ri e ntati o n m ec han is m s evo lved t o work w it h dimmer, n a tura l li g ht s o ur ces and ca n ca u se b ird s to d ev ia t e fr om the ir night paths. Til e problem is co m po und e d fo r b irds n y i ng in mi st o r cl o ud, w hi c h can ca use them to fl y lowe r and closer to artifi cia l light sources, depri ving th e m of ce les tia l and m agn etic c u es. As b ird s fl y n ear li g ht sources, they m ay become di so ri e nte d imd eventu a ll y la nd in th e built e nvironm e n t. T h e maj o rity o f co lli s io n s w ith bu il dings actuall y take place by day . As b irds seek food t o fu e l th e ir next migra- to ry ni g h t, th e y fa ce a Illa ze o f struc tures, and man y, unabl e t o di s tin gu is h betwee n h abitat and re n ec tion s, h it g la ss. Th e amo un t o f li g h t e mitted b y a b uil ding is a stron g predictor of t h e numbe r of co ll is io n s it wi ll cause, m ore so than bu il d in g h e ig ht. Patte rns of li g ht i ntens ity il CroS S a n o c turn a l land scape may influ c llce t h e patte rn o f birds landing in that landscape at t h e e nd of migra- ti o n stages. Thu s, red u c ing li ght tres pa ss from all leve ls o f bu il din gs and their surroundi n gs is an im po rtant pa rt of a s trateh'Y to red u ce co lli s io n s with g lass. T h e re is s o m e rece nt ev id e n ce t h at electromagnetic radiatio n out- s ide the v isi ble spec trum m ay a lso disori e nt birds . Birds and t he Built En viro nment HUlIlall S first bega n u s in g g la ss in Egypt arou ll d 3500 Be E. G la ss b low in g, invented b y the Ro mans in t h e ea r ly first ce ntury e E, g reatly increased th e ways glass co u ld be used, includin g th e fir s t crude g lass w in dows. T il e 17th ce ntu ry saw th e developm en t o f th e n o at p rocess, e n - abli n g p ro du c tion of la rge s h eets of g lass. Tili s techn o l- oh'Y became more so phis ti ca ted, eve ntu a ll y mak in g g lass w indows a vai lab le on a la rge sca le by the 1960s. I n th e 1980s, development of n ew productio n and co n st ru c ti o n techn o logies c u lminated in toda y's g la ss s kyscr ape rs and increas in g u sc o f glass in a ll types of constru ction. Spraw lin g land-use pattern s and intens i fi ed urbani zati o n deg rade t h e quality and quant it y of bird h abitat a c ross 20 th e g lobe . C iti es a n d to w n s e n c roach on rive rb a n ks a n d s h o rel in es. Su bur b s, farm s, and rec rea ti o n a reas in c reas- in g ly in frin ge upo n we t lands <lnd wood land s. So me b ird s pecies sim p ly abando n d isturbe d h ab itat. Fo r re sid e n t s pecies t h a t can t o le ra te d istu rban ce, g lass is a c o n s tant t h reat, as th ese b ird s a rc se ld o m far fro m h um a n s tru c- tu res. Migrat i ng b ird s are oft e n fo rce d t o la n d in tr ees linin g o ur Si d ewa lks, c ity pa rks, wa te rf ro n t b us in ess d is- tr ic ts, a nd oth e r urba n g ree n patc h es th a t h ave rep la ced t h e ir t radit io nal stopover s ites . Th e am o unt o f g lass in a b u ildin g is t h e s tro n ges t p re d iC- to r of h ow d a n gero u s it is t o b ird s. However, eve n sm all are a s o f g la ss can be le tha l. W hil e bird k ill s a t hom es ar e es timated a t o n e to 10 b ird s pe r ho m e pe r year, t h e la rge n um ber of h o m es m ul tip li es t hat loss to m illi ons o f b ird s pe r year in th e United States, rep rese nt in g ovc r 46 (J.'h of t h e tota l p ro bl e m . O th e r fa c to rs can in crease o r dec re a se a b uild in g's im pa c t, in cl ud in g th e d e n sity and s pec ies co m pos iti o n of loc a l b ir d po pu latio n s; loca l geogra p h y; th e type, loca ti o n , a n d exte n t o f la n dsc(lp in g and n ca rby hab ita t ; preva ilin g w ind a n d we a t h e r; and patte rn s o f mi grat io n th ro ug h th c ar ea. All m u st be co n s id e red whe n p lann in g b ir d -fri c nd ly bu il din gs . Impact o f Collision s o n Bird Populations Abo u t 25% of s pec ies a rc n ow o n t h e U.S. 'vV atc h Li st of birds o f co n servati o n co n ce rn (a b cb ird s.o rgl bird s/wa t c h list ), a nd eve n man y co mmon s p ec ies a rc in d e clin e. Habita t d estructio n o r altera ti o n of bo t h b ree d in g a nd w inte r in g g ro un ds re m a in s the mos t ser io u s m a n-mad e probl e m , b ut co lli sio n s w it h b u il din gs arc seco nd o n ly to d o m es ti c cats as d ir ect fat a li t y t hre ats. Ne arl y o n e-t hi rd of t h e bird speci es fo und in t h e Un ited States-mo re than 258 s pec ies, from hum m in gb ird s to fal co n s-a rc d ocum c n te d a s vic ti ms o f co ll is io n s. Un li ke n a tu ral h aza rd s t h at pr ed o minan tl y ki ll weake r i ndi v id ua ls, co ll is io n s ki ll a ll c atego r ies o f b ir d s, in clu d in g so m e of t h e stro n ges t , h ealth ies t bird s that wo u ld o t h e rwi se su rvive to prod u ce o ff s pri n g . W it h o ut ac t io n , t h e cu mu lat ive e ff ec t o f t h ese d eaths w ill res ult in sig ni fi ca n t po pu la t io n d eclin es. Mos t o f th e m o rta li ty is avo idabl e. Thi s d ocum e nt is Oll e p iece o f a strat e!,'Y t o kee p b u il d in g co lli s io n s from in crea s in g and, ulti m a tel y, to reduce them. Bird Collisions a nd Sustainable Archit ecture In rece n t d ecad es, a d va n ces in g lass t ech no logy a n d p ro- du ct io n h a ve m ad e it p oss ib le to co n s t ruct ta ll b u ildings w ith a ll -g la ss wa ll s, and we h a v e see n a genera l in c re a se in the a m o un t of gl ass u se d ill a ll ty p es of co n s tru c ti o n . Thi s is manifes t in a n in c rease in p ic t ure w in d ows in priva t e h o m es, g lass balco ni es and ra ili n gs, b us sh e lte rs, and ga ze b os. New a p p li cati o n s fo r gla ss are be in g d ev el- o p ed a ll t h e t im e. Unfo rtu n ate ly, as th e a m o u n t o f g la ss in c rea ses, so d ocs t h c in c id e n ce of b ird coll isio n s. The Cape May campus of Atlantic Cape Cummunity College inher ited a building with large areas of gl ass that d ie! nOl have CO <lti n gs o r film 10 con l rollcrnperalure fi n d glarc-.1lld there were many collisions. The addition of Cn lljdcsppc has elimina ted the threa t to birds while reducing healing and cool in g cost~. I'hoto b~' Lisa Apcl .Gendron In recent dec ades, growing concern for th e environmen t h as s timulated the creat io n of IIg re en" stiHld a rd s and rat- in g sys tems for d evelopm e nt. The best kn ow n is th e U.S. G ree n Building Council (USGBC) Lead e rs hip in Ene rgy and En viron m e ntal Des ig n , or LEED. Whi le the USGIlC co n c urred that susta inab le bu ildings s h o u ld n ot kill hird s, it was initi a ll y d iffl c ulL lo creale recommendati o n s w ithin th e LEED cre dit sys tem. The so lut ion was ba sed o n (j t ec hnique Glil ed "tunn e l te s ting," a non -lethal m e th od us in g liv e birds that pe rmits a re lative t h reat sco re to be a SS ig n ed to patte rned g lass and other materi- als. (The sec tion o n Research in C hapter 6 reviews th e wor k und e rl y in g the assignment of threat sco res .) On Oc tober 14, 20 11 , USG BC added Pilot C redit 55: Bird Colli sio n Deterre nce to it s Pil ot C re dit Li brary. Th e c redit was draft ed b y American Bird Co n serva n cy (AG e), m e mbers of the llird-Safe G lass f o undatio n , and the USG BC Sus t a in able Sites Technical Advi sory Group (TA G). Building d evelopers that w ish to ea rn thi s c redit mu st quantify the threat leve l to birds posed by var ious material s and d es ign deta il s. Th ese threat fa ctors are use d to calc ulate i]l1 ind ex, o r we ighted average, repre se ntin g the buildin g's fa \,adei that index mus t be be low a standa rd val ue to emn th e cred it . The ind ex is intend ed to prov id e w ide latitud e in c rea ting d es ig n s that m eet the c riteria . Th e c redil a lso requires adopting in terior and ex te ri or li ghting plans and pos t-co n struction m o nito rin g . Pilot C red it 55 is o ne of the m ost widely u sed cred its in th e pil o t libr ary. A revi se d vers io n o f th e c redit, posted in the fall of 20 15, expa nds its availab ili t y to all LEED rating sys tems except "n eig hborh oods." AB C is a registered provider of th e American In s titu t e o f Architects (AlA) Co ntinuing Education System, o f- fe r ing classes on bird-friend ly d es ign and LEED Pil o t C re dit 55 in face-to-fac e and webinar format s. Contact C hri s tin e Sheppa rd , cs h eppa rd@abcbird s.o rg, for m ore information . Defin ing W hat's Go od for Birds It is in c rea s ing ly co mm o n t o sec t h e te rm "bird -fr ie ndly " used in a va ri ety of s ituati o n s t o demonstrate that a part ic ul ar produc t , build ing, legis lati o n , etc., is n o t harmful to bird s. All t oo o ften , however, th is te rm is unaccompanied by a clear d e finition anclla cks a sound sc ientifi c foundation to unde rpin it s use. Ull imately, d e fining "bird-friendl y" is a subj ective ta sk. Is bird- fr iendliness a co ntinuum, and if so, w h e re docs friend ly become u nfriendly? Is "bird-fTi c ndl y" th e sa m e as "bird- safe?" Il o w does th e definiti o n c h a n ge from u sc to u sc, si tuation t o s itua t ion? It is imposs ibl e to know exactly h o w man y birds a particular building wiII ki ll before it is built, and so, rea li stically, we cann o t declare a building to be bird-friendl y befo re it h as b een ca re full y monitored fo r seve ral years. There arc facto rs that ca n h e lp u s pred ict whet h e r (I build ing w ill be part icularly harmful to b irds or generall y ben ig n , and we ca n a cco rdin g ly defin e si mpl e IIb ir d -frie n dly bu il d ing standards" that, if fo ll owed, sig nifi ca ntly redu ce p o te ntial h aza rd to b ird s. Th at said, a 75% re ducti o n of m o rtality at a s tru c ture that kill s 400 bird s a yea r means that s tru c ture will still ki ll 100 birds a year. Becau se wi ndow kill s affect re productivel y active ad ult bird s, th e cumu lative effect o f savi n g so m e birds is amplified by th e ir reproduc tive o utput. Bec ause a 100% reduction in m o rta lity JIIa y be diffi c ult t o a c hi eve, ABC takes the position that it is be tte r t o take re aso n able ava il ab le actio n s immediately tha n to put off takin g action until a perfect so luti o n is poss ib le o r to take no ac tion at a ll. Harirl I'onla rln l Arc hit ects with Hohb ie/Youn g + Wri ght Ar ch itec ts used bolanlcallmagcry In 3M I:llnlnates 10 d c pici Ihe pla nl s th a t produ ce man)' of the com pounds u sed by s tud en ts al Ihe Un ive rsity of Waterloo School of I'hilfllla cy, Canada. I'holo b)' C hri stine Sh e ppard, ABC Pl :)11\1 'P11!dd"qS "lI !I~IlID ,{q (J IOIl,! -;V.0([1 ) '('1~ ilu!p "U.1J \I! 1101 .\\ ~I ! .1 P!\U ! ~H II!l d OJ SI)J!rt ffU!MCJp 'pJ I!;(I!l( ,'[{IIIOP I! s asod mflp l c "J'O 'U()I ::IU!q ~'I!M S!ll! II! SSC [~ "'11 . 111 A a , I , I H • o on U sse 1,:) Properties of Glass G lass, as a st ru ctura l m a te rial , ca n ran ge in appea ran ce fro m transparent to mi rro red to opaque. It s s urface (eln com pl et ely r efle c i li g ht or l et virtually I 00% of li g ht pa ss through. A parti c u lar piece of glass wi ll c h a n ge appear· <l n er d epe nd i n g on en v ironm ental facto rs, including pos iti on re lat ive to th e SlIn , the diffe re n ce between exte- rior and illterior li ght leve ls, wha t-Illay be reflec ted, and the ang le at which it is v iewed. Co mbin a ti o n s of these factors ca n ca u se g lass to look li ke a mirror or a dark passageway, o r be comp letely in v isib le. Humans do not actually "sec" clear gla ss, but a rc c u eci by co ntex t s uc h a s Th e glassowalled towers o f the Time Warner Cente r in New York City appear 10 birds as just .ulotllcr piece of the s ky . Pho to by C hristine Sheppard, ABC mulli o n s, d ir t, o r window fram es. Bird s, h oweve r, do n ot p e rceive right a n g les a nd ot her a rc hitec tura l sig nal s as in d ica to rs of o bs ta c les o r artificia l e nvironm e nts: they take w hat they sec lite rall y. W hil e loca l birds ma y be- come fa mili a r with individual p ieces of g la ss, they d o n ot eve r g ra sp th e co n cep t "glass." Refl ection Under th e ri g ht co nditi o n s, eve n tran spare nt glass o n build in gs ca n ro rm a mirror, rcnec Lin g sky, clouds, or n ea rb y habitat attractive to birds. When bird s tr y to fly to the r en ec t ed Il ab itat, th ey hit th e g l ass, Renected veg- etation is the most dangerous, but birds a lso attempt to f'l y past rel1ected buildings o r t'h ro u g h reflected passage- ways, w ith fa tal resu l ts. Transparency Birds stri ke tra n spare n t w indows as they a ttempt to ac- cess pote n t ial perc h es, p lants, fo od or water sources, o r oth er lures see n tllfough t h e g la ss, whet h er in s ide or outside. Large planted at ri a arc frequent problems, as are g lass ba lc o n y ra ilings and "skywa lk s" joining buildings. T h e in creas ing tre nd toward g lass u sed in landscapes, as wa ll s around roof ga rden s, as handrail s o r wa lkway di v id ers and eve n gazebos is d a n gero us because bird s pe rce ive a n unobstru c ted ro ute throu g h thelll to h abitat beyo n d, Black Hole or Pa ssage Effect Birds o rten fly th roug h srnall gaps, such as sp aces be- tween le a ves or branc hes, into nest caviti es, o r through othe r sma ll open ings tha t th ey e n co unter. In so m e li ght, th e space be hind g lass ca n appear b la ck, crea ting t h e appearance o f ju st su c h a cavi t y o r "passage" w ith unob- stru cted a ccess through w hi c h birds tr y to fl y. Tran sparent handrails are il dangerous trend for birds, especially when they front vegetation. I'h olo by Christine Sheppard, t\RC I.arge faci ng pnncs of glass can appcnr \0 be a dear pathwOlY. Ph olo bl' C hristine Sheppa rd , ABC The same glass can appea~ tran sparent or highly reflective , depending 0 weathe~ or time of day. rhoto' by Chnstine ShCPP.1ld, ARC Factors Affec ting Rates o f Bird Co lli si o n s f o r a Pa rti c ula r Building Every site and every bu ildi n g can be characterized as a u n ique comb i nation of r isk factors for collis i ons. Some of t h ese, particul a rl y aspect s of a bu ilding's design , arc very building-s pecific. Many prob le m d es ign features can be read il y improved, Of, in new constructio n , avo ided. Others of t hese-fo r exa mp le, a build ing's location re lative to m ig ration stopover sites, regi o na l eco logy, and geograp h y-arc dimcu lt if not impossible to m od ify. Building D es ign Peop le li ke g lass and i t h a s become a pop ul ar bu il ding mater ia l. All -g lass buildings have beco m e m o rc and m ore common a s gla ss has become a low-cost mate rial fo r co n struction. G lass causes virtually all b ird collisions w ith bu ildin gs. Studies ba sed on monito rin g data have sh ow n a direct re lationship between the am o u nt of g lass on a bui ld in g and th e nu m ber of co lli s io n s at that s itc- the m orc glass, the m o re bird deaths. Mi rro re d glass is ofte n u se d inte n t io nall y to make a buildin g "blend" in to a vegetated a rea by re n ec t in g its su rr ou ndin gs, making those bui ldin gs especia lly dead ly to b i rd s. However, cill -giass b uil d ings a re com- in g in c reasi n g ly into q uestio n. According to g roups like the Ame rican Society of Heating , Refrigerating and Ai r- Co ndit io ni ng Engi n ee rs (AS H RAE) and the In ternat io nal Code Council, wh e n t h ere is more than 30-40911 g lass on a fat;ade, h ea ting and coo lin g cos ts begin to in c rease. B uildin g Si ze G la ss s kys crapers, becau se o f the ir h eig h t and v is ib ility, ar c o ft e n the main focus of collision documentati o n, and th ey do a ccoun t for m o re co ll is ions per bu il d ing than s ma ll er structures. Ho wevc r, because t h ere a rc many more homes and low-risc bui ldin gs, t h e latte r a cco unt for m o re total m o rtality. A study pub li s hed by s cientists at the Smi t hsonia n in 2014 es t imated 508,000 annua l bird deaths fo r hi gh-rises, 339 mi ll ion for low-rises , and 253 mi lli on for h omes. More co lli sion s p robably occ li r a t g la ss on lower !loors, w h ere 1Il0 s t b i rd act iv ity takes p lace, but w h en mon ito rs have had access to setbacks and roofs, they have co n Sis tently found at least some carcasses, ind icating that g lass at any leve l can be a t h rea t. Ori e ntation a nd Siting Because migrat ing birds are frequent co lli sion Vict ims, it is ofte n assumed that more collisions will occur o n no rt h -and sout h -facing fa~ade s. Howeve r, most bui ld- ing co ll isio n s take p la ce during the da y, and b u ildin g orientation in re lat ion to co mpass direction has not b een implicated as a factor. Siting of buildings with re- spect to su rround ing habitat and la n d scap ing has more Bhds flying fr o m'l meado w 011 the left are challneled toward tlie gla ss door s of Ihi s Imildin g b~' a rocky outcro p to th e right of t he path . I'holo by Chri stine Sh eppar d, AII C 26 I'lantings on se tbacks and rooftops l:all aUra ct b i rds to g lass they mig ht otherwise avoid . Chris Sh e ppard, AB C imp li cations. Physical fea tures lik e walkways that pro- v ide an open night path thro llg ll vegetated landscape, or obstacles lik e outcrops of rock or b e rm s, can channe l birds toward or away fr o m glass and s hou ld be cons id- ered ear ly in th e design p hase . Movement patterns of birds w ithin s urroundin g ha bitat may cause unanticipat- ed colli sion s . Bird s ofte n fl y be twee n landscape feature s, for example, betwee n two stand s o f trees, and may be at ri sk fr o m stru cture s a long th e ir route. Gla ss that reflects s hrubs and tree s ca uses more co ll i- sion s than g lass that refl ec ts pave ment o r g ra ss. Studies that measured vegetation w ithin o n ly 15 to 50 feet of a fcll;:ade have led to the mi sco n ceptio n that plantings b eyo n d a ce rtain di stance don't ill f"lue n ce co lli Sions, but vegetatio n at mu ch g reate r di s tances ca n eas ily be v is ib le in re fl ec ti o n s. Vegetati o n around buildings will bri n g m o re bird s into th e v ic inity o f th e build ing; th e re fl ec- tion of that vegeta tion brings m o re bird s into the g las s. Tall e r trees and shrubs co rr e late w ith mo re co lli sions. It sh o uld be kept in mind that vegetation o n slopes near a building wi ll reflect in w ind o ws above ground le vel. Stud ies u s in g b ird feeders (K lem etal. 199 1 ) have s hown that fatal co lli sions res ult when b i rd s fl y toward g lass from more than a few fee t away. T im e of Day Co lli s io n s te nd t o ha ppe n m ost wh en birds a re most a c - tive . Man y studi es have documented that although colli- s ion s peak during the ea rly morning, th ey can h a ppe n at almost any tim e of day . Most m o nito rin g program s have focllsed on ea rl y mornin g before cleaning crews have swept s id ewa lk s be c a use of th e increase cllikc li h ood of finding bird s and b e cau se it is ea s ier to obtain vo luntee r sea rch ers in the pre-work h o urs. G r ee n Roofs an d Wa lls G reen roofs bring c le m ents attractive to birds to high er levels, b ut ofte n they ar c built in close proximity to g la ss. How ever, recent work s h ows that well -designed green roofs ca n be co me fun ctiona l ecosys tem s, p rov id ing food and even nest s it es fo r birds . Sitin g of g ree n roo fs, as we ll as g ree n wa ll s and rooftop garden s, shou ld therefore be carefull y conS id e re d , and g la ss adjacent t o th ese feature s s hou ld ha ve protection for birds. G ree n roo fs ,md walls ca n provide food and other re so urces to birds, b u t the}' call also ,Ittrac t birds to g la ss that t hey mi ght not o the rwi se enco unte r . Emiliu Amuasz 's i\C n OS building in hlkuoka, Japan, is an interesting example. I'hoto by J<cn ta Mabuc hi 2 8 '" N l/) c o It is poss ib le to d esig n b Uildin gs that can reas on a b ly b e ex pe c ted to kill few or n o birds. N umero ll s examples alr eady ex ist, n o t n ecessaril y d es ig ned w ith birds in mind but s impl y to be functi o na l and attra ctive. These buildings may ha ve m a n y windows, but t h e ir scree n s, latti cewo rk , lo uve rs, and o ther d ev ices o uts id e, o r patte rn s integ rate d into th e gla ss, \",'arn bird s b e fo re th ey co ll id e. Findin g g lass t rea tm e n ts th a t ca n e limin a te or g rea tl y reduce bird mortality, whi le minimall y o bsc urin g th e glass itse lf , h as bee n th e goal o f severa l rese arc h e rs, in c ludin g Martin Ross le r, Da ni e l Kle m , and C hri stin e Sh e ppa rd . Th e ir wor k, di scussed in m o re d e t a il in t h e Sc ie n ce c hapte r, h as foc use d prim a ril y o n the s p a cin g, le n g th , Wid th , opacity, co lo r, a nd o ri e ntati o n o f cl emen t s m a rk ed o n g la ss, anc! h as s h ow n that pa tte rn s cove rin g a s littl e as 5% o f th e to tal g la ss surface can d e t e r most stri kes u n de r ex p er i me nta l co nditi o n s. Th ey ha ve s h ow n that as a ge n e ral rul e, Ill os t so n g bird s will n ot atte m p t to f1y th ro ug h h or izo nta l s paces less than 2 in c h es hi g h o r through ve rtical spa ce s 4 in c h es wide o r less. We re fer to t hi s as th e 2 x 4 rul e, a nd it is cl e a rl y re la ted lo lh e s ize and s hape o f b ird s in Oi g lll . (See c hart a ll pa ge 47). Des ig nin g a n ew s tru c ture to be bird -f ri e ndl y d oes n o t re quire res tri c tin g th e im a ginati o n o r adding t o the cos t o f co n s tru cti o n. Arc hitec t s a round th e g lo be h ave crea t- e d fa scin a tin g and impor ta nt s tru c tures that in co rp o ra te litt le or n o d a n ge rou s g la ss . In so m e c a ses, in s pirati o n h as b ee n bo rn e o ut of fun c ti o nal n ee d s, su c h as sh a d - in g in h o t climates; in o th e rs, from a es th e tic s . Be in g b ird -fri e ndl y us u a ll y h as bee n incid e nta l. Now, h oweve r, bui ldin gs are be ing de sig n ed with birds in mind , and m a te ri a ls d es ig n ed fo r thi s purpose are mul t ipl y in g _ Un - til rece ntl y, re lTofittin g ex istill g bui ldin gs ha s bee n m o re (O ppmi tt') The ext{'rnalllla~~ ~crccn o n th e roSA llC'ginnal Field Offi ce in l lo us ton, Tl'xas, deslg ill'{l b y Page So utherland Page, helps control heat b ut also reduces the li keli hood or co lli sio ns_ Photo hy Timothy I lu rsley diffi c ult and cos t ly th a n it is lo da y. Ho weve r, n ew m a te- rial s arc appearing and cos t s c an be co ntroll ed b y t a rg e t- in g probl e m areas rather than e ntire b uildin gs . Bird -fr ie n d ly m a te r ial s and des ig n feat-ur es often ove rlap in fun cti o n w ith mate rial s to control heat and li g ht, sec urity m eas u res, a nd d ecorati ve d es ig n d e tail s . Bir d- frie nd ly b uildin g-des ign stra t eg ies a lso fa ll into t hree ge n e ral c atego ri es , a lth o u g h all th ree co ul d be co mbin ed in a s in g le stru c ture. These a re: 1 . Us in g minimal g lass (Bron x Call Cente r, U.S . Mi ss io n to th e Un ited Natio n s) 2. Pla ci n g g la ss be hind so m e type o f scree ning (d e Yo ung M u se um, Coope r U ni o n ) 3 . Us ing g la ss w ith inhe re nt prope rt ies t h a t re du ce colli sio n s (Bro o kl y n Bo tanic Ga rd e n Vis itors Ce nte r; Stude nt Ce nter a t Rye rso n Un-ive rs ity, Toro n to; an d Ca th e dral o f C hris t t h e Li g ht) Awnings and Overhangs (Not Recommended) O ve rhan gs have b ee n fr equ e ntl y reco mm e nded to redu ce co lli sio n s . Howev e r, th ere are man y s itu atio n s in w hi c h overhall gs d o not e limin ate refle c ti o n s alld on ly b loc k gla ss fro m th e v iew o f birds f1 y in g a bove. T h ey a rc thu s o f li m it e d e ff ec ti ve n ess a s a ge n e ra l stratq,ry . Ove rhan gs wo rk bes t w h e n g la ss is s had owed fro m all s id es . run c ti o n a l c lem e nts s u c h as bal co ni es and ba lu s tr a d es can bl oc k the v iew o f g la ss, p rotectin g b ird s w hil e p rov idin g an am e nity fo r res id e nts. Ang led Glass (Not Recommended) In a stud y (Kle m e t al., 2004) co mpa rin g bird co lli s io n s w ith ve rt ic al pa n es of g la ss to th ose tilted 20 o r 40 d e- g rees, th e an g le d g la ss res u lt ed in le ss mortality . Kl c m sp ec ula te d that thi s was becau se th e g la ss re ll ec te d th e The IJrooklyn Botanic Garden 's Visitors Cen ter, des igned by Wei ss/Ma n fredi, waS" int e nded to be bird-friendly frD llI lts inceptio n-a challenge, as it l1l ake~ ex te nsiv e use of glass. Photo (gl Albe r Vecerka, ESTO Gl ass walls and doors at tbe I\rooklyn Botanic Garden's Visitors Cen te r incl ude a c ustom frittin g pa tte rn l ha t Illeets bird -friendly c riteria . ~\'I onitoring for collis io n s after the building o p ened in d icates thatl hc design was successful. Ph oto by Christi n e Sh e p pa rd , AII C Over ha ng s bl oc k viewing of gla ss fro m some angles, hut do nol nccessarily elimi n atc reflections . I'hoto by Christine Shep p,ud, !\lIe Refle c ti on s in Ihls a ngled f .. ~·ad e ca n he seen clearly O\·cr a lon g distance, and birds co m ap proach the !-!Ia ss hom any angle . Pho to by Christinc She ppa rd , ARC 20 Bi rd-Fr iendly BUildin g DeSign g ro u nd , n o t vege t a ti o n. Us in g a n g led g la ss ha s b eco me a commo n reco mm e ndati o n a s a b ird-fri e nd ly feature. Howeve r, w hi le an g led g lass m ay be u se ful in spec ial cir c um stan ces, th e b irds in t h e stu dy we re fl y in g p a ra ll e l t o t il e grou nd from n earb y fccde rs, hi tt in g th c g lass at ac u t e a n g les, w it h less fo rce than a pe rpe nd ic ul ar s tr i ke. In most s itu a ti o n s, h oweve r, b ird s m ay a p p ro a c h g la ss from an y a ngle . N e tting, Scr ee n s, Grilles, Shutters , Ex t e rior Shade s T h e re are m a n y ways to co m b in e t h e ben e fit s o f g la ss with b ir d-fri c nd ly d es ig n by in cor poratin g e le m e nts t h at pre clud e co ll isio n s w hil e p rov idin g li g ht and v iews. So m e a rc h itec t s h ave d es ig n ed d eco ra ti ve fa~a des th a t w rap entire str uc tu res. Deco ra ti ve g rill es a re a lso part o f m a n y a rc hitec tura l t ra dit io n s . Ex t e ri o r, m o t o ri ze d so la r scree n s a nd s h ad es are e ffec ti ve a t co ntrol lin g h ea t a nd li ght, inc rea se sec u rity, an d ca n be ad jus t ed t o m a x im ize v iew or b ird a nd sun p ro tect io n at di ffere n t t i m es. Ne t- t in g, g rill es, and s h utte rs a re co mm o n c le m e n t s that can ma ke g lass safe fo r b ird s o n b uil d in gs o f any sca le. T h ey ca n be lI se d in re t ro fit or b e a n integ ra l pa rt o f an o rig i- n a l d es ig n a ll d call sig nif icantl y red u ce bi rd m o rta li ty. Befo re t h e c ur rent age of un open a b le w ind ows, sc reens protec t ed birds in add iti o n t o se rv in g th e ir p rim a ry pu r pose of kee pin g b u gs o ut. Scree n s arc st ill am o n g th e mos t cos t-effec ti ve met h ods for pro tec t in g b ird s, a nd , if insec ts a re not a n iss u e, n ea rl y in v isib le net ting ca n o ft e n b e in sta ll e d . Scre e n s and n e ttin g sh o uld be in stall ed a t so m e re m ove from t h e w ind ow so th a t t h e im p ac t of a strike does n o t ca rry birds into t h e g la ss. Sevc ra l co m pan ies se ll scr ee n s that ca n be attac h ed w ith s uc ti o n cups o r eye h oo ks fo r s ma ll areas o f g la ss . Oth e rs spec ia lize in muc h large r in st a ll ati o n s. (Find sOllT ces at co il isio n s.a be bi rd s.o rg). Patt e rn s on Gla ss Ce ra mic do ts, o th e r ty p es o f "frit s,1I a nd o th e r m ate ria ls can b e sc ree n e d , prin t e d, o r o t h e rwi se a p p lied to g lass s ur fa ces. Thi s is o ft e n d o n e t o re du ce t h e t ransmi ss io n o f li g ht a n d h ea t and ca n a lso provid e d es ign d e t a il. In so m e cases, frit patte rn s are h a rdl y v iSib le, b ut w h e n d es ig n ed a ccord in g to t h e 2 x 4 ru le (see p . 4 7), patte rn s o n gl ass ca n a lso prevent bird s t r ik es. Patterns o n th e outside s urface o f glas s d e ter co lli s io n s most effec ti ve ly becau se th ey a rc a lways v is ib le, eve n w ith st ro ng re- fl ectio n s . Th is ty p e o f d es ig n , u se ful prim ari ly fo r n ew co n struction, is c urre ntl y m o re co mm o n in Eu rope a nd ,\ cus to m hit patte rn was d es igned by Enne<l d Ar c h it ec ts for Vassa r Co llege's Hridg e for l.aborator}' Sciences hUildin g. Ele m e n ts of th e pattcrn occur on IwO se pa rate su rfa ces, in creaSi n g vis ibili ty 10 hirds in ni gh t , who will see a constantly c hangi ng pattern thaI ma y appear to m ove. Photo by Ch risti ne Sh ep pard , AI~C 32 Whil e so m e inte rnal fr ill e d gla ss patterns can be ovc rcomc b )' rc nec t io n s, Fran!.: Ge h ry 's I/\C headqua rters in Manhallan is so den se that tile g l a s~ appear s o paque. I'holo hy Chri stine Sheppard, AB C D.w.ll..u.o. Mikad o's paltern re flect s UV wave length s. The spiderweb dkc t is visib le o n l)' from vcry limitClJ viewing angles. Photo courlcs), of Arn o ld Glass Asia, but is be in g offe red by an in c rea sin g numbe r o f manufac tu re rs in t h e Uni te d States. New t ec hn o log ies a ll o w in g printin g o f cera mi c ink s o n th e o u ts id e surfa ce o f g lass m ay g rea tl y in crease op ti o n s fo r b ird-fr ie ndly des ign in th e U.S. More co mm o n ly, frit is app li ed t o a n in te rn a l s urface o f in s ul a ted g lass units. Thi s type of des ig n m ay n o t be v isib le if th e a m o unt o f li g ht re n ected b y th e Irit is i n suffi cie ll t to overco m e re l1 ec ti o n s on t h e ou tsid e s ur face of th e g lass or if frit is a p p li ed as d ots be low th e v isu a l thres h o ld of b ird s. So m e inte rn a l frit s m ay o n ly h el p brea k u p re fl ec ti o n s wh en viewed fro m so m e a n gles a nd in ce rt ain lig h t co nditi o n s. Howeve r, wit h th e ri g ht cOIlliJi n ation o f su rf ace re fl ec ti v ity a n d fr it a p p lica ti o n , a pattern o n a n inside s u rface ca n st ill be effec ti ve. T h e headqua rters of th e in te rn e t co m pa n y lAC in New Yo rk Ci ty, d es ig n ed by Fra n k Ge h ry, is composed e n tirely of fritted g lass, m ost of h ig h de n sity and a lways vis ible. No co lli sio n m orta lit ies h ave bee n reported a t t h is b ui ld in g afte r two yea rs of m o ni to ri n g by New Yo rk City Audubo n . r:Xr:OWLE's J aco b Ja vits Ce n te r, a lso in Ma nhattan, reduced co lli sio n s by as m uc h as 90% w ith a re n ova ti o n th a t eli m in ated some d angero us glass and rep laced oth er g la ss w ith a v isib le fri t patte rn . An oth e r ex am p le o f a visible intern a l fri t patte rn is see n in Skid mo re Ow ings Me rril's Cath edral o f C hri st th e Lig ht in Oa kla nd, Cali fo rni a . UV Patt erne d Gl ass So n gb ird s, gull s, parrot s, a nd oth er b irds ca n see in to t h e u ltrav iolet (UV) spec trum of li g h t, a ra n ge la rgel y in v is ib le to hum a n s (sec p age 4 1). Oth e r b ird t ypes, i ncl udi n g rap lors, kin g fi sh e rs, hu m m in g b ir d s, a n d p igeo n s, are less se n siti ve to UV . Ultra v io let re fl ec t ive a nd /or abso rb ing patte rn s "in v is ibl e t o h um a n s b ut v isib le t o b irds" ar e fr e qu e nt ly s uggested as th e o ptimal so lut io n for m a n y b ird co ll is io n p ro bl e m s, b ut few suc h p rod u ct s a re ava il abl e co mm e rc ia ll y as o f 2015 . Prog ress in deve lop m e nt o f bird-fri e n d ly UV g lass h as bee n slow, b u t w ith legis latio n in m u lti ple loca ti o n s m a nd a tin g b ir d-fri e nd ly d es ig n , g lass m a nu fac ture rs (lIl d d is t ri buto rs, as we ll as w in dow-film m an ufac tur e rs, a re taki n g a n active ro le in d eve lo pi ng n ew so lu t io n s fo r th is a p p li ca ti o n . Rese ar c h indi ca t es tha t UV pa tte rn s n eed stro n g contrast t o be effec ti ve, especia ll y in the ear ly m o rn in g an d la te a fte rnoon , w h e n UV in s un li g h t is at low levels. Howeve r, UV p atterns m ay b e ineffec ti ve fo r m a n y species th a t h ave bee n reported as v icti m s o f co lli sio n s w ith g lass, includin g h u mm i n g bird s, fl yca tchers, AllI er ica ll Woodcock, a n d woodpec kers. Opaque a n d T ra n s lu ce n t Gl ass Opaque, etc h ed, sta in ed, o r frosted g lass a n d g lass b lock arc exce ll e n t o pti o n s t o recl u ce or e li m inate co lli s io n s, a nd m a n y a t trac ti ve arc h itectura l app licatio n s ex ist . T h ey can be lIsed in ret ro flt s bu t a rc m o re com m on ly u se d in n ew co n stru c t io n . Froste d g la ss is c reated by ac id etchi n g o r sa n db las tin g t ransp arent glass. Frosted areas a re t ra n sl u ce n t, bu t variou s fin ish es a re a va il a bl e w it h diffe rin g levels of lig ht t rans m iss io n . An entire s u rface can be frosted , o r fros t ed pa t te rn s can be applied . Patte rn s sh o uld co nfo rm to th e 2 x 4 ru le d es crib ed o n page 47. For re t ro fit s, g lass a lso ca n be fros ted by san dbla st in g o n sit e. St a in ed g lass is t yp icall y seen in re lat ive ly s mall a re a s but ca n b e extrc m c ly a ttrac ti ve a nd is n ot co n d ucive to co lli s io n s. Glass b lock is ve rsa t ile , can be use d as a d es ign d e t a il o r pri mary co n stru ct io n m a te ri al , a n d is a lso un li ke ly to ca u se co lli s io n s. An o th e r promisi n g m a te r ial is p h otovolta ic g lass, w hi c h h as b ee n u se d in s t a in ed-g lass w ind ows a nd hi g h way n o ise ba rr iers. Th is so lution is especia ll y in teresting, beca u se 34 ~ , i.) • ~ .,~ -. ~l )' 24 e'rd-Frlencly BUIldIng Des Ign tran spa rent hi ghway noi se ba rri e rs ca n ca use co lli sio n s, a nd su ch ba rri e rs arc begi nning to be in sta ll ed in t h e United Stiltes. Window Films Mos t patterned wind ow film s were initia ll y int ended fo r li se in s id e stru c tures as d es ign e le m e nts or fo r privacy . l OW, o uts ide surfa ce appli ca ti o n s intended to reduce bird co lli sio n s <lrC co min g onto the mark e t, and so me h ave prove d hi g hl y e ffe c ti ve and popu la r. Th e o ld es l su ch pr o du ct c reates a n o paqu e w hite s urffl ce o n th e o ut- si d e o f g la ss th at still pe rmits v iew in g from the in si d e. Pattern s G ill be pr in t e d o n thi s m ater ial, a lt h o ug h im- ages o f trees and o th er h abitat are n o t rec o mm ended. A fi lm with a pattern o f n a rrow, h o ri zo n ta l stripes ha s c liminated co ll isio ns at the Philad e l p h ia Zoo Bear Co un - t r y exh ibit ror ove r Av e yea r s (see ph o t o o ppos ite) alld h as been similar ly success ful in oth e r in st a ll atio ns w h en app li ed to o uts id e surfa ces of g las s. In th ese cases, th e res ponse has bee n posi tive. Anot h e r o ption is t o apply vi n y l patt e rn s lik e w ind o w fi lm but wit h th e t ra n spare n t ba c kin g re m oved. Solutions Applied to Interi or Glass Li gh t co lored shades have been reco mmended as a way t o dete r co lli s io n s. Howeve r, w h e n vi s ib le, th ey do not e ffectiv e ly re du ce re n ecti o n s, a n d rcn ec ti o n s may m ake th e m comple te ly in v isib le. C losed bli n ds h ave th e sa m e prob le m s, but if v is ib le a nd partl y o p e n , th ey can pro- duce th e appea ran ce o f a 2 x 4 pattern . If an ex t e ri o r so- luti o n is not poss ib le and tap e o r sti cky notes a re a pp li ed to th e in s id e of Wi nd o ws, be sur e to c h ec k t h e w indows seve ra l tim es a da y t o e n sure t h at these material s ar e v is ible. Decals and Tape Decal s are probabl y th e m os t fam ili a r so luti o n to b ir d colli sio n s, but th e ir effec tive n ess is wi d e ly mi sund e rst ood. Bi rd s do not recog n ize d eca ls as A Zl'!1 Wind C Urllli n is all lncxpclls!w hut cx tr e mely cfrcctivc WOl)' 10 d e ler coll isi ons. l.e ll!;lh s of pa r.l c hul C (O rd or ~i milar rn alc ri "ls arc strull g vc rtlc:lll y. C\"C I), four in c ht's , in front of prob lcm g l:.ss. Cleating bot h a \'is u:II and a phys ical bar rier. Phot o lIy Gkllil Phillips 36 . . - ABC. with support fro the Ruslnow Fa i1y Foundation. . . .. ake home windows safer for • • . . .. . . - lets birds see glass w ile letting . . .. . .. -. and lasts up to four years. For more infor atlon. VISit .. . ..... . l'hot05 br Darimz ZdzlebkowskJ , ABC 26 B,rd-Fn enaly AUlldln g DeSign si lh o uettes of falc ons, spide rwebs, o r othe r natural obj ec ls, but s imply as obs tacles that they Illay tr y to ny around. DCGl ls can be vc ry effecti ve if app li ed fo ll owing th e 2 x 4 rul e o n th e outside o f g lass, but in genera l, th ey mu st be repla ced frequ e ntl y, at least annua ll y. Tape is genera ll y mo re cost effe ctive and quicker to apply, but most household tap es don't sta nd lip we ll to the e le m e nts. Tape intended to last fo r severa l years on th e o u tside of wi nd ows has beco m e co mmercia ll y ava il able and is effective w h e n applied following t h e 2 x 4 gu id e. Thl! Consili llnl Towers, a mirro r·glas s complex in Toro nto, o nce killed thou sands o f bird s eac h ye ar. Aft e r being tilken to (O utt , it s owne rs re tr ofitte d the lower GO feel o f gl a ~~ w ith II Fl'alhcr Fri <'ll dly dot patt e rn that h<ls greatl y reduced bi rd m o rt a lity . lll'ne c led in Ih i~ gla~~ i~ Mic ha c!1\Ie,urc, th e fo unde r of Tnro nt o '~ Fatal Ughl AWMc n ess Progra m. Photo s by C hri stine She ppard, AII C Temporary Solutions In some circums tan ces, espec ia ll y for homes and s mall build ings, quick, l ow~cost, temporary so luti o n s, su c h as mak in g patterns on glass wit h paint, stickers, o r eve n p os t~i ts , can be very effec tive in th e short term. Even a m odes t effo rt can reci u ce co ll is ions. Su c h measures can be applied w h e n n eeded and arc m ost effective f o ll ow~ ing th e 2 x 4 ru le. (For m ore infor mati o n, see AB C's fly e r "Yo u Can Save Birds from Fl y in g into W ind ows" and other sources at co lli s io n s.abcb ir ds.org). !\IIC lIirdTape was effe ctiv e :lt the Forest BC,1Ch Mig ratory Hesccvc in Wisconsin (left). ,Ind al so performed well in tunnel tes ts co ndu c ted in ,\mt ri a. Photo h}' Ch ristine Sheppard, AII C 38 REMEDIATION CASE STUDY : Javits Center In 2009. the New York Ci ty Audu b on Society id e ntifi ed t he Jacob K. J avits Convention Center as having one of the highest bird-co lli sion morta l ity rat es in New Yo rk C i ty. A m ajor renova tion and expansion. designed by the bird-friendly a r chitectural firm of FXFOWLE. was com - p lete d in 20 14. So m e es pe c ia ll y d ea dly g lass at stree t leve l was re pla ced w it h opaque panels. Large panes of clear fr itt ed glass wi th va rying surface characteristics were brought to t h e site a nd co mpa r ed to find th e right comb inati on for bird s an d peo ple . A 6.7S-acre green roof. w ith adjac e nt t ran sl ucent gla ss. c rowns the bui lding and is already providing re so urces f or birds. Bes t of all . collisions at t h e now mu c h larg e r sit e have be en reduced by 90%. i -. Bird s evo lved co mpl ex co mpl e mentary syste m s for o r i- e ntatio n and v is ion long before human s deve loped art i- Ii cial li g h L We still h ave muc h m ore to learn, b ut recent science has begun to c la rify h o w a rtifi cial li ght poses a threa t to birds, espec iall y n octurn a l mig rants. Alth o u g h most g lass co lli s io n s tak e p lace durin g day li g ht h ours, artificia l li g htin g a t ni g llt pla ys a role in th e num ber and d ist ributio n of co lli sio n s across th e built e n v ironment'. Un fort lln a te ly, th e d e ta il s of h ow birds respond to ni g ht li g htin g are less we ll unde rs tood than ha s bee n co m - mo nl y be li eve d . Ma n y co lli sio n v ic tim s, especia ll y so ngbird s, are ord in ar il y a c ti ve by d ay a n d ha ve eyes spec ialized for colo r vision a nd bright lig h t. But a l t h o u g ll t h ey mi g ra te at night, these birds helve poor night vis ion. In stead, t h cy h ave m ag n e ti C se n ses that a ll ow th e m to n aV ig ate lIsing the Earth's magnetiC field. O n e of these is located in t h e retin a a nd req uires d im blue natural lig ht to func t ion. Red wavele n gth s fo und i n most artificial li g ht h ave been sh ow n to disrupt that m ag n et iC se n se. Studi es in Ge rman y and Ru ss ia ha ve d ocum e nted b ird s !l y in g throu g h bea ms o f li g ht and d iver tin g fr om th eir co urse anyw h e re from a few degrces to a full c ir cle. Are a s wi l'h sign ifica ll t. li g ht p o lluti o n llI ay b e cO lll pl etely di so ri e ntin g to birds. Bird s a re <Ittracted to re lative brig h tness, a nd by day often o rie nt toward t h e sun. If a son gb ird ni es into a h o m c, da r ken in g th c roo m a nd openi n g a bri g ht w in - dow is th e best way to re lease it. Bi rds a rc thought t o be attrac ted to artificia l li ght a t ni g ht, but we don't kn ow w h at li g h t leve l at w h at distance is su ffi cie nt to cause attract io n o r o th e r be h avior<ll im p ac ts. Gau threa u x a n d Be lse r, di sc uss in g im pac ts of night li ghting o n b ird s, spec ul a t ed that in fa c t , birds affected by night li ghtin g ma y sim p ly be o n co ur se to pass ove r th e ti g hts, n ot n ecessarily attracted from a di s tan ce . Marqu e ni e a nd Van d e Lelar, stud y ing b ird s a n d li gh ts o n a d rillin g ri g in the o rth Sea, estima ted t hat when all th e li gh t s o n th e platfor m wer e lit, they affecte d b irds li p to 3 to 5 kil o m e- te rs a way, ca us ing man y t o c ircl e the platfo rm . T h e sc ie n ce is inconclusive: Li g hts ma y on ly impact bird s as they e nd a mi g ra to ry s tag e and co m e d ow n close to th e b uilt e n v i ro n me nt, or li ghts ma y dive rt bird s that wou ld ord inaril y pa ss b y. Bad weat h e r ca n c au se bird s t o fl y lower and close r to li g hts, wh il e a lso e liminating any vis ual cu es. The inte ractio n s that prod u ce co rr e lations between bui lding li gh t e mi ssio n s (m d co lli sions ma y take p lace at re la tive ly close range . Once bi rds come close t o a li g ht so urce, th e e lec t ro m agne ti c radi atio n active ly interferes with t h eir magnetiC or ie ntatio n mechan is m. I 100 Iston s k)'Jine at nig ht. Photo by Jeff Woodman . . .. -"", ti"", 11 0111"'-jJU I ' I .elf .. . , . .. " ,1111'-- 1/ j ',:11 I I 11 11111 illWIIlU~ ~.lill.·:::.::J. : . ~; . !,.., '!;- Overly li t ImildinRs wasle e lect ri ci ty ,m el increase g ree n· house gas cmissloJlS an d ,Iir pollution Icvel s. Thc), a lso pose a threat to birds. Ph o to by Matthew Haines ExamQles Unacceptable I Discouraged FIlcII.fw, M oroduee ~.,w;t UgNlfeSplln Acceptable FlanI .... 5hieIII1tIo1itjt~to~GIontrdll!tot~ aoIllO IadIitII:Ie beI:IfIf \/1IlOO at nighI Il Cp rirltl'cI co urlcsy o f Da rk SkySO(:kly.o r ~ So m e co mbinatio n of attrac ti o n and d i so rientati on may re sult in large r number s of birds in th e vicinity of b ri ghter build ings a nd th us, by day, in Illorc co lli s ions. Interesting ly, th e re see m to be no reports of li g hts attra c ting o r di so ri enting mi g rants as th ey take o ff o n a n ew m igrato ry sta ge. There ha s been a t e ndency to associa te colli s ion eve nts with very ta ll st ru c tures, th o u g h publi s hed reports c learl y d ocum e nt im p act from light at a ll leve ls. Ea rl y re ports o f thi s p h e nomen o n came from li g hth o ll ses. Co nte m po rary re p o rts of Ji ght*ils soci<Jt cd circling eve nts arc co mm o n at ocea ni c drilling rigs , and di so r i ented b irds h ave been repo rte d at ni g ht sk iin g sites. A s tud y in Toronto, u s in g th e number of lighted windows o n a se ri es of bui ldings as an in dex of emitted li g ht, fo und that the amo unt of li g ht emitted, n o t th e hei g ht o f th e bu ildin g, was th e best pred ic tor of bird mortality . Solutions Po o rly des ig n e d o r imprope rl y in s tall e d ou tdoor fi x tures a dd ove r Sl billi on to e lectr ica l costs in the United States eve ry year, acco rdin g to the In te rnational Dark Skies As so- c iat io n. Rece nt s tudi es es timate that over two-third s of the world's population ca n n o longe r see th e Milky Wa y, ju st one o f th e nighttim e wo nde rs t hat con n ect people w ith nature. G lare fro m p oo rly shielded o utdoor li ght fixtures decreases v is ibility and can c reate dan gerous co nditions, es p eciall y for o ld er people, a n e! recent studies sugges t that lo ng-t er m exposur e to night li ghtin g can in crease t h e ri sk of breast" ca n ce r, depress ion, diabetes, o bes ity, and sleep dis o rde rs. To gethe r, th e e co logi cal, fi n anC ia l, a nd c ultural impacts of excess ive buildin g li gh ting are co mpe llin g rea- so n s to reel u ce and re fine Ii g hl us ag e. Redu c in g exterior building and s ite li g htin g h as prove n e ffe c ti ve at reducing mortality of ni g ht migrants at 42 individua l bu il din gs, but ac hi ev i n g ove ra ll red u c tion i n co lli sions w ill req u ire appl ying th ose pri n ci ples o n a w ider sc a lc. At th c sa mc time, t h ese meas ures redu cc bu il din g e n ergy cos t s and decre a se air and li g ht pol- lution. EFf-ic ie n l d es ig n o f ligh t-i n g sys te m s p lu s ope ra - t io na l strateg ies to red u ce light tr espa ss o r "s p ill li ght" fro m buildings w hi le m axim iz in g use fu l li g ht a rc bot h important st rateg ies. In additi o n , an in creasi n g b o d y o f ev iden ce s h o ws that reel li ght and w hite li g h t (w h ic h co ntain s re el wave le n gth s) particularly co n fuse bird s, w h ile gr ee n and blue lig ht m ay h el ve far less im pact. Lig ht po llu ti o n is la rge ly a res ulL or in effi c ie nt ex te ri o r li gh tin g, and improving li g htin g des ign u suall y produ ces sav in gs g reat e r t h a n t h e cost or cha n ges. r o r cxa mp le , g lo be flxture s pcrmillilli e co ntro l o f li g ht , w h ic h s h in es in all dir ec t io n s, res ultin g in a loss o f as m uc h as 50% of e n e rg y, as we ll as poo r illumin a ti on. C ut-o ff sh ie ld s ca n red u ce li g h t in g lo ss a nel pefmit u se o f lower powe red b u lbs. Mos t IIv anity li g htin g" is un necessa ry. How ever, when it is used, down-li g htin g causes less tr espass than u p-li g htin g. W h ere li g h t is n eeded fo r safe t y and sec u rity, red u c in g th e a mo unt of li g ht tres pa ss o uts ide of the n eed ed areas can h e lp by e li min at in g s hadows. Spotlights a nd se arc hli g hts s h o ul d not be lI se d d uri ng bird migration. Communities that ha ve imp lemented p rog ram s t o redu ce light po llut io n h a ve n o t fo und an inc rease in c rim e. Us ing automatic contro ls, includin g timers, photo- se nso rs, and infrared and motion d etec t o rs, is fa r m o re effective than rel y ing on e mployees turnin g o ff lights. Th cse dev ice s ge n e ra ll y pay for th e m se lves in e n e rgy sa v ings in less than a ye ar. Workspace li g ht i n g s h o uld be in stall ed w h ere n ee d e d , rath er than in lar ge area s. In a re as wh ere ind oor li g h ts w ill be on at ni g ht, minimi ze per imete r li g htin g and /o r draw s hades afte r da rk. Sw it c hing to daytime cleaning of o ff·ice buildin gs is a simp le way t o reduce li g htin g w hil e a lso reducin g cos ts. Lights O u t Programs Des pite th e co mpl ex ity o f redu c in g b ird co lli sio n s w it h g lass, th ere is one simpl e way to d ecrea se morta li ty: turn li g hts off. Across th e Un it ed Sta tes and Ca n a da , "Li g hts O ut" program s at t h e munic ipa l a n d s tate level s e n co ura ge building owners a nd occ upa nts to turn o ut li g hts visibl e from o ut side d urin g sp rin g and fall mi g ra ti o n. T h e first o f th ese , Li g hts O ut C h icago, was s tarted in 1995, fo ll owed by To ro nto in 1997. Th e progra m s t h e m selves are dive rse. Some are d ir ec ted by e nvironme ntal groups, ot h e rs by government departments, and still ot h e rs b y partners hips of o rgani zat io n s, Part icipation in most, su c h as Ho uston 's, is vo luntary. Mi nn eso ta mandates turnin g off li g hts in state-ow ned and lea se d bu il dings . Ma n y jur isdi c t io n s have m o n itor ing compo n e nts. Mon i- to rin g program s can prov ide im po rtant info rmation in additi o n to quantifyin g co llision leve ls and doc um e nt- in g so lut io n s. Id eall y, li g hts-out prog rams would be in eff cc t ye ar-ro und and be applied wi d c ly, sav in g b ird s and e n e rgy cos ts and re du c ing e miss ions o f g reen h o u se gases. ABC stand s rcady to he lp d eve lop new prog ram s and to s u ppor t and expa n d ex is tin g prog ram s. Powe rful uca m s of li g ht , cve n ill a lamls c,.pc o f urban light po lluti o n , can e ntrap m igrating birds, see n h e re circlin g in IIIl' beams o f the 9/1 1 Me m o rial Tribu te in Ught in Ne w York Ci ty. Bec au se bird s mOl }' cir cl e fo r hours, m o nit o rs wal c h itll night. alld till.: light ue,ullS arc tem por :u ily turn ed off to release 10lrge ilccumulOl- tions of bird s. Phot o by Jason NapolitOl ll o Blro-FrLendlY BUilding Oes;gn 314 3 Legislation C h a n gi n g hum a n be ha v ior is ge n e rall y a s low process, eve n w h e n the c hange is un controvers ial. Legislation c an be a powerful tool fo r m o difying be havio r. Co n se rvati o n legislation h as c rea ted resclVCS, reduced pollutio n , a nd protected threaten ed species and ecosystems. Polici es that promote bird-fri en dly d es i g n and reduction o f lig ht pol - lu ti o n ha ve recent ly proliferated across th e Uni ted States and Ca nada, foll ow in g the ea rl y exa mples of Toronto and Sa n fran cisco. Th ey vary co n Siderabl y in sco pe and d e tail, ofte n ren edin g l o cal po liti cs. (A rea l-tim e d ata ba se of or- dinances and oth e r in s truments mandating o r promoting b ird -fri e n dly actioll, including links to so urce la n g uage, G lil b e found at co lli s io ll s ,abcb irds.o rg ). An early c hall e n ge in creating effecti ve legis la t io n was the lack o f o bj ective m easures that arc hitects co uld usc to accom p li s \, the ir task. f or exa mple, a co mm o n recom - m e ndati o n , to "in crease v isuClln o ise," because it was unquantifi ed and und e fin ed, m ad e it d iffi c ult for arc hi - tects and planne rs to k n ow w h e th e r a parti cu lar d esig n co mpli e d w ith re qU irem e nts. Mate rial t es ting (see p. 45) has mad e it poss ibl e to ass ign relative threa t factors to va ri o u s bu ildin g fa <;:ade mate rial s and to u sc th ose sco res to c rea t e quantitative g Uid e lin es and mandates. The illu s tration t o the r ig ht broad ly compares Sa n Fran- c isco's Bir d-s afe Buildi ng Standard w ith LEED Pil o t C re d - it 55, bo lh based o n th e u se o f mate rial s with quantifi e d three,lt leve ls. Sa n r:rancis co 's sta ndard appli es ge n era ll y t o n ew co n struc tion a nd is res tric t ed to fa .;ades within 300 feet of a t wo-acre park or pond. T h e LEED c re dit is inte nti o nall y very flex ible . It appli es to all building fa- cades and all ows fo r restr ict ed a m o unts of hi g h -threat g lass, or large r a m o unts o f bird -fri e ndl y g la ss . Beca u se b ird s are fo und through o ut the bui lt environment, ABC (Opposite] Unit ed Stat es Capit ol, WaS h ing to n. D.C . I'hoto by Sh)c k.xcllllf,: prefe rs the LEED model. (A BC's m ode l legis lati o n ca n b e found o n page 3 5.) Bird love rs across th e co untry arc propos ing bird-friend ly d es ign ordina n ces at both loca l a n d s t ate leve ls. A BC is ready to act ive ly s uppo rt s uch e ff o rt s. Bo th m a ndato ry and vo luntary instrum e n ts can be e ffec ti ve. Vo lunta ry g uid e lin es a rc easi e r t o m o dify if th ey prove to h ave un- intended co n sequ e n ces and ca n lead t o a mandate, but can al so b e ig n ored. Ge n era ll y A BC reco mm e nds manda- tory g Uid e lin es, begin nin g w ith a small subse t o f build- in gs and expa nding as com munity support increases a nd res ista n ce de c reases. In co rpo ra ting bird-friendl y des ign iss u es in to loca l su st a inability p oliCies is a n o th e r way t o drive c han ge .. An inte restin g exa mpl e o f this is t h e r:a ir fax Co unty, Virg inia, proffe r sys t em. New co n s tru c ti o n proj ect s arc req uire d to ad dress a se ri es of s u sta ina b ili ty issues, i.ncludin g po tential bird mortality, and e ith e r to d esc ribe -_ ZONE 1 The design of the G range Insurance Audubo n Center in Columbus, Ohiu, includes many p;lIlds of g la ss. rritted wit h th e ~ilh {}u e tt cs o f speci es of birds in fIlJ.lht. Phuto b}' Ch rist ine Sheppard , AB C LEED PILOT CREDIT 55 SAN FRANCISCO BIRD-SAFE BUILDING STANDARD courtesy of Deborah L,urc1 or Its new Visitors Center in Sempach. opened in May 2015. the Swiss • .. . . .. . ... mandala from bird silhouettes (below) ... . .. . .. . . ." . .. .. '11·.· .• · .•• --. discussion a ong viSitors. an achievement second only .. - -• • •• • h ow these w ill be add ressed by the pro ject o r ex plain w h y sti c h action i s n ot poss i ble. Pri oriti es for Policy Directives ABC ge n era ll y reco mm e nds aga in st attempti n g to map l o c ation s w h er e bird-fri en d l y desi g n is r eq uired be ca u se b ird s can be foun d in alm ost eve r y environment, eve n in see min gly inhosp itable o nes. Howeve r, th e re may be occa s ion s w h en it is n ecessary to compro mi se o n the scope of legi s lat io n . In such cases, it mus t b e re cogn ized that prox imi ty to und eveloped la nd , ag ri cu ltural areas, parks, and wa t e r o ften co rr es po n d to in c reased b ird pop ul a ti o n s a n d therefore in c re a se d ri sk of co llisions. In add iti o n , areas located in be tween landscape feat ur es des ir able to bird s may a lso pose higher risk s . For e xamp le , in Nc \·v York Ci t y so m c ev ide n ce su ggest s t h at birds a pproach Ce ntra l Park fr om due sout h d ur in g spr in g mig ra tion, creat in g a grea t er ri s k zone di rec tl y south of the park. Al s o, bu il ding feat ures s uc h as g reen roofs s h ou ld be co n sidered when deter min ing g rea t er risk zo n es for po li cy p urposes . Sustainability Rating Programs Anot h e r dr iver of b ird-fri end ly policies co n s is t s of sli sta inab ili ty rating p rogram s lik e th e G ree n Buildi n g Cou n c il's LEED p rog ram, G ree n G lobes, Living Building C h a ll e n ge, a n d o th e rs. Th ere is ge n e ral agreement that su sta inab le b uil dings sh ou ld n ot kill bird s. This te n et a p pears with differ ing levels of robustness in different sys te m s, w ith th e m ost spec ifi c be ing th e LEED program, wh ic h gra nts Pi lot C re dit 55 : Bird Co lli s io n Deterrence. Til e cred it is calc ulat ed using a weig hted average of th e re lative threat r<lUng of eac h mate ri a l on a build in g's fa~ade. Th e credit has attracte cl a lo t of atte n tion, wit h m any projects clp p ly in g for it". Th e new Vassar Brid ge fo r Lab ora tory Scie n ces on th e cover of thi s publicat ion was o n e of t h e fir s t to be d es ig n ed wit h th e cred it in mind (lnd to earn th e credi t . Because a number of gla ss-wa ll ed buildings ha ve been awa rd ed LEE!) ce rtifi ca ti o n a t the hi g h est level, at o n e poin t t h e re was concern that su stainab le des ign was not co m patible wit h bird-fri e n d ly design. Thi s was iro n ic, as in add iti on to prov idin g n atura l li ght, g las s o n su stainab le b uildings is intended to link peop le in side w ith the natural wo rld o uts id e. However, accord in g t o bot h AS HR AE a n d ICC, cost s for h eating a nd coolin g increase w hen t ota l g lass su rface exceeds 30-40% o f the t ota l bu ilding e n ve lope, depen ding o n cl im ate. Thi s is m o re than s uffi c ie n t for prov iding li g ht a n d views when g lass pla ce m ent is cons id e red thoughtfu ll y. Th is is a great p lace to sta rt tile d es ign of a bird-fri en dl y stru ct ure . The fa ,adc o f Ill c W D nTH Build in g in Switzerla m l is l11 osH}' glass, laminalcd willl a fabri c Illat is bla c k on lll e in ~idc bul aluminium·coaled outs id e. The in n er surfan' delivc rs good vi silJilHy, and Ih e fallri( pro vid es Sll;ldc and IIl1er· c SUng vi su<l l cffe c t ~ o uts ide. l'rclllllill:J.r}' s tud ies h}' til e Swi ss Ornithologica l JlIStilut e s ugg est Ih"\lhe Illilterials used in Ih is building rnay <lI sa dele r bird colli sio n s. Photo Il}' Hans Sc hmid 46 Model Ordinance for Bird-Friendly Construction [O RD INANCE Na m e[ Spo n sored b y: [li s t n ames [ W H EREAS, bird s prov id e vCl luabl e and impo rtant eco logica l se rv ices, W H EREAS, [Ioca ti o n[ has recorded [ [species of res ident a n d migrato ry b ird species, W HER EAS, b ird ing is a h obby e nj oyed by 64 million Ame ricans and gene rates m ore than $40 bi lli o n a ye m in eco n o mi c act iv it y in t h e U n it ed St a tes, WHEREAS, a s man y as one b illi o n bird s ma y be ki ll ed by coll isions wit h win d m ·vs eve ry year in t h e United Sta t es, W H EREAS , reducing li gh t p o lluti o n ha s been s h ow n to red u ce bird deat h s from co lli sio n s w ith w indows, W H EREAS, n ew b uild ings can be design ed t o reduce b ird deaths from co lli s ions w ith out add it iona l cost, W HERE AS, t h ere ex is t strateg ies to mitigate co lli s io n s o n ex is t in g bu il d in gs, W H EREAS, mo re than 30t}1, glass o n a fa~ade usuall y increases costs fo r h C':l ting and coo ling WHE REAS, bird -f ri e ndl y p ra c ti ces ofte n go h a n d-in -h and w ith e n e rgy e fti cien cy imp rove m e nts, An d W H EREAS [a n y add it ions s p eci"c to the particul ar IOGlt io nl NOW, TH EREFORE, BE IT O RD AIN ED , by lacting agencyllt it lc of leg islat io n and oth er necessary lan g u agel (a ) In th is sec ti o n t h e term "Le':ld e rs hip in Ene rgy and En v ironm e n t a l Des ig n (LEEDY' m e ans a g reen b uild ing ra t in g sys te m prom u lga t ed by the U nite d States G ree n Build ing Coun c il (USG Jl C) that prov ides s p ec if i c prill ci pl es a nd p ractices, some mandato ry b ut t h e ma jo rit y discretio n ary, t h at m ay b e app li e d d uri ng the desig n , co n s tr u c ti o n , and o peratio n phases, wh ic h e n able the build i ng to be awarded po ints fro m reaching prese nt sta n da rds of e ll v iro l1J ne ntal e fn c iency so th a t it m ay a c hi eve LEED ce r t ifi catio n f ro m t h e USGUC as a "gree n " bu il d in g. b) [actin g agencyi clocs hereby o rder [acti n g d e partm e ntl to take t h e steps n ecessa ry to assurc t h at a ll n ew ly co n s tr u cted b uil d in gs an d all b ui ldin gs sciled uled for cap ital improvement are d es ig n ed, bu ilt, <Ind o p e rated in acco rda n ce w ith th e s tandards and req uire m e nts of t h e LEED G ree n Bu ildin g Rating Sys t e m Pil ot C re dit 55: Bird Co lli s ion Dete rrence. (c) T h e USG BC rel eases revised ve rs io n s of th e LEED G ree n Bu il din g Ratin g Sys t e m o n a regu la r basis; a nd lacting depa rtm e nt! s h a ll refer to th e most c ur rent ve rs io n o f the LEED w llcn beg in ning a new bu il di n g constr u ct io n permit proj ect or re novat io n . (d) New co n s tru cti o n and major re n ovatio n proj ec ts s h all in co rpo rate bird-fri e ndl y b uild ing m a t e ri a ls a nd d es ig n features, in clud in g, b ut not li m ite d t o, those reco mm e nd ed b y th e Am e ri ca n Bird Co n se rva n cy pub lic<lt io n Bird-FrieJJd/y Blli/dillg Des ig ll . (e) la c t in g ciepa rtm entl s hall make exist in g b uildin gs b i rd -fri e ndl y w h ere p racticab le . T h e Stu di o G;l11g 's /\qua Tower in C h iG lgo was desig ned with birds In milld . St rategies included frlttcd )(Iass and balcony halmtradcs. 1'11 010 I,y T im Bloolllqui st 47 I)UI1lJrl-ili of spccll's of hirds arc kllll'tl by collisions . These birds were colll'ctl'd by munltors with FI.Ar In Toronto, Canada rhoto by Kl'nnl'th Ilcrdy Magnitude of Collision Deaths Th e number of bird s k ill ed by collis i o ns with g l ass every year is astronomica l. Q uantify ing mortality leve ls and i mpa c ts o n p o pula li o n s ha s b ee n d ifrl c ull , howeve r. Until rece ntl y , loc al m o rtalil y s tudi es---<l esp ite produc- i n g va luab l e information-aimed more at do cumenting m o rtality than quantifying it, and did not follow ri gor- ous protoco ls. Lo ss el al. (20'12 ) crea ted m e thodo logy and techniques of ana lys is t o d e termine th e mag nitude of anthropoge n ic m o rtality, u s in g ex istin g data se ts . The authors comprehensively a cquired pub li s h e d and unpub li s h e d data sels o n co ll is ions with bui ldin gs (Lo ss el (If., 2013). Dat a sets were lilter ed u s in g a va r ie t y o f cri- ter ia t o e n s ure that they co u ld be u sed in si n g le a n alyses . Lo ss el. al. (20 14b) ha ve a lso co mpre hens iv ely d escribed how to co lle ct m ea ningful data o n co ll isio n s. Th e a uth o rs calcul ated the m ed ian all nu al m orta li ty at h o m es at 253 mi ll ion , or 2 .1 birds p e r st-ruc t·ure. Urban re s id e nces wit h o ut feed e rs a cco unt for 33% of this m o r ta lity c umu la t ively , as t h ere arc more s u c h reSide n ces, even though reside n ces w ith feeders produce m o re co ll isi oll s indi Viduall y. Rural res iden ces wit h out feed e rs acco unt for 3 1'M, o f re side nliaim o r talily, fo llowed by urban res id e nces with feeders (19%) and rura l res ide n ces with fee d e rs (1 7IMI). Median m o rtal ity at low-ri sc bu il dings (4 to 11 s to r ies), c alcu lated from two data sets, was ave ra ged as 339 million , or 2 1.7 bird s pe r bui lding . Hi g h-rises, a lth o u g h co ll ec ti ve ly ca u s ing th e least m o rtality (508 ,000), indi v idually had t h e hi g hes t m edian rate of 24 .3 bird co lli s io n s pe r bu il d in g. Co mbin ing a ll building classes produces an es tim ate of 365 and 988 (m ed ian 5 99) milli on birds ki ll ed annua ll y in th e U ni ted States. Machtans, e t (/1. (2013) est im ated that about 25 million (ranging from 16 to 42 million ) birds are kill ed by co ll id in g wit h windows i n C anada annually, wit h 90% o f building -re lated morta liti es c au sed b y hou ses, s li g htl y less than 10% by low-r ise bu il dings, and approximate ly 1% by tall buildings. In bo th ca ses, the total m o rtality cause d by hOllses is a fun ction of their large numbe r compare d t o the two o th e r classes of bu il dings. Prev io u s ly, Dun n (1 993 ) s urveyed 5 ,500 p eo ple who fe d bird s at th e ir hom es and re cord ed window co lli s ion s . Sh e d er iv ed an esti mate of 0.65-7.7 bird death s per hom e per yea r for No rth America, Klem (199 0 ) estimated that each building in the United States kill s o n e to 10 birds p e r year . Us in g 1986 U.S. ce n s u s data , h e combined numbe rs of h o mes, sc h oo ls, and com mercial buildings for a m axi mum total of 97,563/626 build in gs, produc- ing a n es timate of 100 million to one billi o n b ird s kill ed annually. Kle m eL (/1. (2 009a) us e d data from New York C ity Audu- bon is m o nitorin g o r 73 Manhattan bui ld in g fa c;ades to es timate 0.5 coll ision deaths p e r acre per yea r in urban e nvironm e nts, for a tota l o f about 3 4 million migra- tory birds annuall y co lli ding with c it y build in gs in th e A sample o f co lli s ion v ictlms from BaltIm ore. Photo b}' Daniel J. Lcbbin, Ali C T hi s Ham Swall o w illu st ra tes th e ty pe o f ilCT o lJalk fly ing Ihal may keep swallows from be ing freq u ent colli sio n vi ctim s. If bird s do identify gl a ss ,1$ a ba rri er at close rang e, pe rhaps by sound or air lII o vc m cnts, most s pecies ma ), be unable t o re ac t fa s t e nough to avoid strikin g th e s urfa ce. Ph o to by Ke ith Ri n gland Sharp-shi n ned I la wl\. I'hol0 by Ted t\rdlcy 38 B/ro-f! enaly AU/id·r.g Des,gr Unite d Sta t es . Howeve r, t h e re co ul d be maj or d iffe re n ce s in c o ll isio n pa tte rn s in cit ie s across t h e U nited Sta te s, a nd th ese n Uln be rs s h o u ld b e co nfir m ed u sin g d a t a fro lll ad d i t io n a l lo ca ti o n s. In T ile A m eric(l 1I Bi rd CO II Se rl'a ll c), G /li de to Bird COllse rv a t ioll (Le bb in et a l ., 20 I 0 ) th e a u t h ors s tate rI ••• we h a ve re a c h e d (] p o in t in hi s to ry w h en th e impact s o f hu ma n act iv it ies a rc so profo u n d a n d far-r ea c h in g th at from 11 0W a ll , it w ill a lwa ys be im po ss ib le to u nta n g le t h e co m p le te ly n a t u ra l d ecl in es fro m t h ose that a rc partia lly or co mp le te ly an t h ro poge ni c. r:rom a co n se rva ti o n stand poi nt, it is la rge ly i rr e leva nt, an ywa y_ An y h u m a n -ca u se d s t ress that we G ill a ll ev ia te from a d ec li ll ill g speci es call po te ntia ll y iJ e n e fi t its po p ul a tio n , a nd we s h o ul d tak e a c ti o n to lesse n t h at stres s if we c an ." T h is is a b u ndant ly t ru e fo r b ird m o rta lity fro m g la ss beca use t h e re ar c a c ti o n s t h a t ma n y, if n o t Ill os t , indi v idual s can t a ke th e m selves, d irec t ly, to redu ce the to ll ta ke ll b y ex is ti n g g la ss. Patterns of Mortality It is d iffic u lt to ge t a co mp le t e and a cc ur ate p ic t ure of av ian m or ta li ty fro m co lli s io n s w it h gla ss . Co lli sio n d ea th s ca n occ u r a t a n y t im e of da y o r yea r. Mo n it o rin g p rog ra m s foc u s on c it ies, .m el eve n in t e n s ive m o nito ri n g prog ra m s cover o nl y a po rtion of a ci t y, u s u a ll y v is iting t h e gro un d le ve l o f a g ive n s it e a t m os t o n ce a da y a n d ofte n o n ly duri n g m ig rati o n se a so n s. Man y city b ui ld- in gs h a ve s t e p pe d ro of se t ba c ks tha t arc in a ccess ib le to m o ni t or in g tea m s. So m e stud ies ha ve focu se d o n repo rt s fro m h o m eow n ers o n ba c kyard b ird s (K le m , 1989i Dunn, 1993) o r o n m o rta lit y o f m ig ra n t s i n an ur ba n en v iro n - m e nt (Ge lb an d De lacre toz, 2009; Kle lll el ai., 2009a ; Ne wto n , 199 9). O t h e rs ha ve a n a lyze d co ll is io n v ic t im s prod u ce d by s i n g le , lar ge -m ag n itud e in cid e n t s (Sc al y, 1985) o r th a t h ave bec o me part o f mu se um co ll ec ti o n s (Sn yde r, 19 4 6; Bl e m et a l ., 1998; C o d o n e r, 1995). Th e re is ge nera l s u ppo rt for t h e fa c t tha t b ird s kill ed in co lli- s io n s a rc n o t d ist in g ui s h ed by a g c} sex, s ize, o r h eal t h (fo r ex a m p le: Ble m a n d W ill is, 1998 ; Co d o n e r, 1995; Fink an d Fren c h, 19 7 J ; Ha ge r et (II ., 2008; Kle m , 1989), b ut th e ma jo r ity o f wo rk h as fo c u sed o n d a t a tak e n d u r- in g mi gr a to ry pe rio ds} pr im a ril y ea st of th e Mis s iss ip pi Riv e r. Spec ies a t Risk Sn yd er (19 46), ex am inin g w in d o w co lli s ion fa ta lit ies a t t h e Roya l O n tar io Mu se um , n o t ed th a t t h e m ajor it y were llIi g ra n t s all d "tun ll e l Il ye rsll -sp ec ies t l1 at fr eq u e n tly tl y throu g h s m a ll s p aces in d e n se, u nd e rs t ory h ab itat. Co n- ve rse ly, res id e n t s p e c ies we ll adapted to a n d comm o n in urba n a rea s, s u c h as th e Ho u se Sp a rr ow a nd Eu ropean Star lin g, a re n o t p ro mi n e n t o n li sts o f fat a li t ies, possib ly bec a u se ind iv id u a ls s urv iv in g t h e ir lir st co lli sio n m a y te a c h o ffs pr in g t o avo id w in d ows . It is we ll kn ow n t h a t zoo bi rd s in ex h ib it s with glass wall s ca n a nd d o le a rn a b ou t s p ec ific p ieces o f g las s, b u t bird s do n o t le a rn a bo ut g la ss as a ge n e ra l co n ce pt. Dr. Da ni e l Kl e m mai n t a in s rUll ni n g t ota ls o f t h e n u m - be r o f sp ec ies re p o rt ed in co ll is io n eve n t s in co un t r ie s arou n d t h e wo rl d. (T h is info rma ti o n ca n be fo u n d at 11 ll p :l/tin yur l.co m !ob3 n c 4s). In 20 15 , th e site id e n tifi e s 8 6 8 spec ies g lo ba ll y, wi th 274 f ro m th e Un ite d States. T h e inte n sity o f m o nitori n g a nd re portin g prog ra m s var ies w id el y fro m co untry to c o u n t ry, h oweve r. I-la ge r e/ al. (2 0 08 ) co m pa re d th e num be r of s peci es a n d in div id ua l b ird s kill ed at bu il di n gs a t AlI gli st a n a Co ll e ge in Illi n o is w ith t h e d e n Sity and d ive rs ity of bi rd s p ec ies in t h e s ur rou n d in g area. Th e a uth ors co n cl ud ed t h a t t h e 50 total w ind ow clre ,l, the habitat immedia t e ly adjace nt to windows, and b e hav ioral differenc es amon g s p ec ie s were th e bes t predicto rs o f m o rtality patterns, rath e r than th e mere s ize (lnd co mpositi o n of the loca l bird pop ul ation. Kah le el (II . (20 15) rea c h e d si mila r conclusions in an ana lysis o f live years of data at t h e C alifo rnia Acade m y of SC ien ces, a lso rll1di n g l hat mi gra nts do no t make up the prepo nd e ra n ce of birds kill e d a nd t h at m a les Cl re ove rr ep rese n ted re la ti ve t o t h e ir ab und a n ce in h ab ita t s ad jace n t to th e m useu lll . Du n n (1993), a n a lyz in g w in- ter data from h o m es with b ird fecders, fou n d t h a t the frequency di s tribu ti o n of bird s at t h e feeders close ly para ll e led th e d istributi o n of s peci es kill ed by n earby windows. Dunn fo und few co ll is io n s o n windows of less than o n e square m e ter, and an increase in co lli sions w ith an incrc.'lse in w ind ow s ize. Specics suc h as the White-throa t ed Sparrow, Ove nbird, and Co mm o n Yellowthroat appear consistently o n top 10 lists from urban areas. It is poss ib le that th ese species respo nd m o re readily to li ght and thus are more li kel y to Com lU on Vd lowthro .. t. I'holu by Owen Deu lsch end migrato ry stages in the built e nvironm e nt, but thi s n eeds 10 be con linned_ Add itionally, Loss et al_ (20 -13) n o ted that Go lden-winged Warbler, Painted Bunting, Ca nada Warb le r, Wood Thru s h ), Kentu cky Warbl e r, and Wo rm-ea ting Warb ler-species idcnlifled as birds o f co n se rvat ion co n cern -we re a lso disproportionate ly represe n te d in b uil d in g kills_ I-lage r (2 009 ) n oted that w in d o w-stri ke m orta li ty was repo rted for 45'}1, of raptor s pecies fou n d freq u e n t ly in urba n a re as of t h e United 'states a n d was th e lead in g so ur ce of m o rta lit y fo r Sh a rp- s h in ned Haw ks, Cooper's Hawks , Me rlin s, a n d Pe regr in c Falcons. Beca use mos t d ata o n g las s co lli sions a rc fro m th e eastern ha lf of th e United States, these li s ts arc pre- sumabl y bia se d toward species occurring in that ran ge. Character istics of Buil di ngs Am o unt of G l ass From a stu d y o f Illult iple bu ildings in Manhattan, Klem et al. (2009.) concluded that bo th the proportion and absol ute amount of g lass on a build ing fa c;ade best p red ic t mo rtal ity rates, ca lcu lati n g t h at every in c rease of 10% in t h e expa n se of glass cor re la tes to a 19 91J in crease in bird m orta lity in spri n g a n d 321M. in fa ll. How we ll t h ese eq uatio n s p redi c t m o rta lity in oth er c it ies re lTl ili ns to be tested. Coll in s and I-lo rn (2008), s t u dy ing co ll isio n s at Mi lli kin Un ivers ity in Il li n o is, concluded that total g lass area and t h e prese nce /abse n ce o f larg e expa n ses of g lass predicted morta li ty leve l. I-lager et nl_ (2 008, 20 14 ) cam e to the s ame conclUSion, as did Dunn (199 3) and Kah le etal. (2015). Howeve r, th e "pa tc hiness" of g la ss across a fa c;a d e-how many pi eces, th e ir s ize , how they arc se pa rated, etc. (another wa y of sa yi ng IIv isual n o isc")-has n o t yet been exp lo red in deta il but cou ld be importa n t. The fi\r;ade o f the Kew York Times building, by FXFQWLE and !le nzo I'iano, Is composed of ce ramic rods, spa ced to let occul , .. nIS sec Oul w hil e mini m izi ng th e ex tcn! of exposed gl "s~l-\ood for Ctm l rollinl-\ heat and light. :1Ilt! safe for bi rds. I'hOlo II)' C hri stin e Sh e ppard . AIIC Sn nh c lta 's Studcnt Learnin g Ccntre at ](ye rson Univ c rsit y is oll e of lhe fir st constructed under To ronto's desig n] .. \\,. Ph o to h y I{ick l.i g thcl m Time of Day Mo s t mo ni to r ing programs focus on ea rl y m orn ing h o urs to doc um e n t morta lity during. migrat io n , often st a rti ng m o ni to rin g ro u tes at claw ll , before si d ewa lks a re cleared. T hi s can , h oweve r, lead to th e m is perccptio n tha t n ig h t-nying m igra nts ar e cras hin g i nto lig h te d bu il d in gs at ni g ht, o r o nl y in ear ly m orn in g, w llereas in fact m ost co llisio n s take p lace d u rin g the d ay. It sh o ul d be note d t h at "da\.-vn " is a t ime t h at va r ic s a lll ong species (Th omas et (/1. (2002), with some b ird spec ies ac ti ve befo re hu man s s t a rt to se c li g h t in t h e s ky . I-la ge r a n d C raig (2 0 14), in a study of re sident pop ul a ti o n co lli s io n s in n ort h we ste rn illin o is be t wee n Ju ne a n d ea rl y Augus t , fo u nd t h at 6 61MI of bi rd s died be twe e n s un r ise and 4 :00 p.m 'l wit h n o colli sio n s between 4:00 p.m. and su n se t . Dc\'lC n.:ta z a nd GcJb (2 006) fo u nd co lli sio ns fro m early m o rn in g u ll t il m id -afternoon, but with a peak during m o rn ing h o u rs. This findin g is co nfirm ed by m o ni to ri ng p rogra m s like th a t of Pe nn sy lva n ia Aud u bo n , where ro u te s we re follo we d th ree t imes in success ion ea rl y eac h d ay, with birds fo und at eac h pass (Ke ith Ru sse ll , pers. co m m .) a nd w h e re peo pl e li v in g o r work in g in build ings rep o r t window st rikes t h roug h aft e rnoo n h Oll rs (O lso n , pe rs. co mm ). Local La ndscape Gelb and De lacre t a z (200 6, 2009) e va luated d a t a from colli s io n m orta li ty a t Ma nha ttan b uil d in g fa<;,:a d es . Ti ley fo u nd t h at sites wh ere glass refl ec te d exte n s ive vegeta- t io n were assoc iated w it h more co lli s ions tha n gla ss re ncc tin g littl e or n o vegetatio n . Of t h e 10 b uildin gs re s ponsib le for t h e mos t co ll isio n s, fo ur were "low-rise." Kle m (2 009 ) m Cclsurcc\ var iab les i n t h e space imm ed i- ate ly associa t ed with b uil d in g fa<;:acl es in Man h a tta n as ris k fac to rs for co lli s io n s. Bot h in creased Ileigh t of tr ees a n d in c rease d h e ig h t o f vegeta t io n in crease d t h e ri sk of co ll is io n s in fa ll. Te n pe rcen t increa ses in tree h e ig ht and t h e h eig h t of vegetat io n co rr espo n ded to 30% and 13% in creases in co lli s io n s in fa ll. In spr in g, o nl y t ree h e ig h t h ad a s ign ifi ca nt illllu cnce, with a 10 % in cre ase corre- s pond in g to a 2291) in crease in co ll is ions. C o n fus ing lYI in c reasi n g "fac in g a rea ," defi n e d as th e d ista n ce to th e nearest str u cture, co rre sp o n ded st rongly w it h in c reased co ll ision s in spring <mel wit h red u ced co lli sio n s in fa ll. Pr es u mab ly, vegeta ti o n in c rea ses risk bot h by attrac t ing more b i rd s La an area a n d by be ing re fl ec led in g lass. Bayne et nl. (20 12) co n fir m ed tha i th e ris k of bird-window co lli s io n s va ries acco rdin g to locatio n (ur ban ve rs us rura l, h ome ve rs us apartment , with or without feed e rs, and age o f n e ighborhoo d ). Th ey u se d on line surveys a n d det e rm in ed t h at ru ra l res id e n ces Ilad m o re co ll is ions th an urban ones and re s id e n ces wit h feede rs had a lm ost tw ice a s man y co ll is ions as th ose w ithout feede rs. f or urban dwe lli ngs, incidence o f co lli sio n s increased w ith age of nei g hborh ood , associated w it ll presen ce of mature tr ees. r=r eq uency of colli sions varied se a so n a ll y: 24% in fall, 35 % su m mer, 25% sprin g, 16% w in ter. Mo rta li ty pa t te rn s were sim il ar: 26% fa ll , 3191) s u mm er, 26% sp ring, 17% w inter. r=or ty -e ig h t s pecies were reported. I-lage r et (/1. (2013) n o t ed t h at es t im ates of b ird -co lli s io n m o rtalit y ofte n pos tulate a re lat ively co n st ant ra n ge of co lli sion s at a ll b uil d in gs (for example, Klem, 1990). Howeve r, t h ey su gges t ed t hat eac h buildin g in a land - sca pe h a s its own m ortality "S ig n ature," ba se d not only o n c h arac t eris t ic s of t h e s t ru cture b ut also o n t h e dis· tr ibu ti o n of reso ur ces th ro u g h o u t th e loca l lan d scape, in cl ud in g la n d cove r, h abi t at type, water, an d pave me n t. T h ei r pro t oco l selected b uil d ings at rando m and h a s rece n t ly bee n expa nd ed t o m ul t ip le oth er s ites across No rt h Ame ri ca . 52 Avian Vision a nd Collision s Bird spec ies li ke falcon s arc famou s for their a c u te vis ion, but takin g a "bird's-eye view" is m uch more co mpli ca t ed than it so und s . To st ar t w ith , where hum a n co lor v is io n relic s o n three types of se n so rs , b irds have four , plu s an array of co lor til te rs that toge th e r a ll o w bird s t o d iscri mi- nate betwee n many morc co lo rs than people (Vare la et 01. 1993) (see ligure thi s page). Tllc re is a lso var iation in vision am o ng di ff e rent gro ups of b i rds. While so m e bird s sec on ly into the vio l et range of light, m any bird s, including most passcrin es (O d ee n and l-lii s tad , 2003, 2 0 13) see into the ultravi o le t spectrum (UV S s pecie s). U l t ra vio l et can be a co mponent of any co l or (C uthill et a/. 2000). \,y h c reas humans sec reci, ye llow, o r re d + ye llow, birds ma y sec red + yell ow, but a lso re d + ultra · Vio let, ye ll ow + u lt rav iolet, and red + ye ll ow + ult ravio- le t-co lo rs for whic h we ha ve n o n ames . Every o bject abs o rb s, re nee ts, and transmits ul t ravi o let li g ht a lo n g wit h the other wavelengths in th e visible spect rum. UV patterns o n g lass are ofte n c ited as des ir ab le so l utions to co lli s ions-v isib le to b irds but n ot to h uman s . Howeve r, aside from manufacturi ng comp lexities, man y bi rd taxa that collide frequent ly with glass, in cluding rap to rs, p igeo n s, woodpeckers, and 11um m in gb ird s, may n o t be ab le to perce ive UV patte rn s (H ~lstad and Odeell , 20 14 ). Additiona ll y, birds are o ften active in ea rl y m o rnin g, when UV light levels are low. Human s and o th er primat es have re latively Oat fac es, with eyes c lose together. Th e over lap o f vi s ual r,e ld s m eans t h at hum ans h ave good depth perception and a tend e n cy to focu s o n what is ahead. Most bird s have eyes at t h e sides of the ir h ead s, giv i ng t h e m exce ll ent per i phera l v is ion but poo r depth pe rception , ofte n lim ited t o the le ngth o f th e ir b eaks, pres umabl y to judge potential food items. T h ey may be mu ch les s in tent on w h at is in front of them (Ma r t in 201 1, 2012) but ab le to watch for potenti a l predators to tile s ide o r b e hind t llem. Ma n y spec ies' most a c ute v is ion is to the s ide. W ith o ut much 3D v is io n , birds li se a mecha n is m ca ll ed "v isua l n ow fie lds" to judge th e ir speed and rate of progres s in fl ig ht b y t he passage of e n v iro nm e ntal features to th e ir sides (Bh agclva tuia et (II. 201 1). Colli sio n s with glass may be pa rtl y a re sult of bird s ex pecti n g o pe n a ir ahead, co m- b in ed w it h re lativel y poor forward vision. Birds process images fa s t er than human s; where we see co ntinuo u s Ill o ti o n in a m o vi e, b ir ds wo uld see ni c ke ri n g ima ges (D'Eath, 1998; Grec n wood et (/1.2004; Evan s et at. 2 006). Th is s peed h e lp s man y bird s mancuver q u ick ly in Painled Hu nti ng. I'holo by Ted Ard lc), Comparison of Human and Avian Vision nm 350 400 450 500 550 600 Bilscd all artwork h~r Sheri Wil li amson 650 Contr;l st \el1 sitivi ty is a m eas ur e of t he lim it of visihility for lo\\'-contra~t pattcrn~. E<l c h person 's con trast sensitiv- ity can bl' measlIr<''t 1 h}' Ihe cxtClllto which he or she can see the bal~ that form an arch in thi s pholograph. ·111e exac t IOCilli o n of till' peak of Ihe curV(' vari es with o ne's di stance fr o rn the image; the area within the ar ch is larger whe n Dill' is cl o se r. For a given distann', th e area under t h e arc h is ~mall('f for hirds. Image courtesy of Izumi O:r..aWiI , lkrkelc:}' Ne l,To st"le llt"e Lil ooratory res p o n se t o un e xp ect ed o bs t a cl es as Lhey fl y thro u g h co mplex h ab itats. In o n e res p ec t h owevcr- spa ti a l co ntra s t se n s iti v it y-human vis io n o u tperf o rm s a v ian (G him an d Ho d os, 2006). Co ntrast se n siti v ity is "th e a bility o f th e o bse rve r to di sc rimina te be twee n adi ;:l(e nt stimuli o n th e ba sis o f th e ir differ e n ces in re lati ve lumin os it y (c ontrast ) rath e r than t h e ir a b solute luminan ces ." Bird s' la c k o f co ntra st se n siti v it y ma y be a n im pe d ime n t t o cre atin g s ig mll s to preve n t co lli s io n s t h a t arc effect ive for b ir ds bu t no t v isuall y in t rus ive t o hum ans. Avian Orientation and the Earth's Magnetic Field In t h e 1960s, it was di scovered t h a t m ig ra t in g bi rd s po s- sess th e abi lity t o orie n t t h emse lves us in g c u es from t h e SUIl , pol a r ized li g ht, s ta rs, th e Ea rth 's mag n e ti c fie ld , vis ual lan d m a rk s, an d poss ib ly even od ors to fin d th e ir way . Exac tl y h ow thi s wo rks -and it lik e ly vari es a m o ng spec ies -is still be in g in ves ti gated , (r o r a c om p re h e n s ive rev iew of th e m ec hani s ms invo lv ed in av ian o ri en tati o n, sec W ilt sc hk o and Wiltsc hk o, 200 9). Th e Ear th 's lII ag li etic fi e ld ca n p rov ide both dir ec l"i o n a l a nd p o s it io n a l inForma li on. It ap pe ar s that ni g ht-fl y in g mi gr an ts , and pe rha ps all bird species, h ave ma g n et ic fi e ld-d e tec tin g s tr uctu res in th e reti n a of t h e eye th a t depen d a ll lig ht for fun ct io n a nd prov ide co m pass o rie nta ti o n . Thi s ma g n e ti c se n se is wavcl e n gt h- depe nd e n t. Expe rim e n ts Ilave s h ow n th a t the co mpass is d is rupte d by lo n g wave le n g th li g ht but re q u ires low-inte n s ity sh o rt wa ve le n gt h li g ht (W ilt sc h ko et a/. 2007). Thi s resear c h h as tak e n p la ce o nl y in labo ra to ri es, a nd it is impo rtan t to d e termin e h ow it tr a n s la t es t o th e rea l wo rld. In a dditi o n , a n t hropogeni c e lec t ro ni c n o isc, fo und th ro u g h o ut urb a n e n v iro nm e nts, ha s rece ntl y bee n sh ow n t o di s rupt mag n eti C co mpa ss o ri e nta ti o n in Europea n Ro bin s at very low inte n s iti es (En ge ls et (/1. 20 14). Thi s filldin g may hav e se ri o u s impl ic ati o n s fo r strateg ies aim ed at re ducin g co lli s io n s b y redu c ing a rtificia l ni g ll t li g htin g a lon e and s h o uld be a pri o rity fo r a dd iti o nal wo rk. A second mag n e ti c m ec h a ni sm , p rov idin g bird s with pos iti o n a l infor m a ti o n , h as bee n postu la t ed , but its d e t a il s ha ve n ot bee n d e t e rmin e d . (Fo r a rev iew of m ag n e to receptio n a nd its use in avia n mig rati on, sec Mo uri tse n , 20 1 S.) Birds and Light Pollution Th e earli est re po rts o f mass av ia n mo rt a li ty ca u se d by li g h ts were fro m li g h t h o u ses, but thi s source o f m o r ta li ty esse nti a ll y di sap pear e d w h e n st eady-burnin g li g hts we re re pla ced by rotatin g beams Oon cs a nd Fran cis, 2003). Fl as hin g o r inte rru p t ed bea m s ap pa re ntl y illl oweei b ird s t o co ntinu c t o naVi ga te, w hi c h ha s al so bee n fo unei m orc recentl y at ce ll t owe rs with s tro be li g htin g (Ge hrin g or 01.20 09). Th e emphas is o n t a ll s tru c tures by Li g hts Out p ro g ram s ig n o res th e fac t that li g h t fro m mt.lI1 y so ur ces, from urban spraw l t o pa rkin g lo ts, ca n affec t bi rd be ha v io r and pot e n t ia ll y s tra nd bir ds in th e b uil t e n v iro nm e nt (G au t hrea u x and Be lse r, 2006). Eva n s- Og d e n (2002) sh owe d that li g ht e mi ssio n level s of 16 b Uildin gs, ran g in g in h e ight from 8 lo 72 fl oo rs a nd ind exe d by th e numbe r o f lighte d w indows o bse rve d a t n ig ht, correl ated directl y w ith bird m o rt a lity, a nd 54 t h a t t h e a m o unt of li g h t e mitted by a structure was a better p re d ic t o r of morta li t y leve l than buildin g height, a l th o ug h h e ig h t was a factor. Pa rki n s et al. (20 15) mad e similar fJlld in gs. Mas s colli s io n eve nts o f mi g rants assoc iat ed w ith li g ht and often \vith fog o r storm s h ave been fr e qu e ntl y reported (We ir , 1976; Avery ef al. 1977; Av ery ef al. 1978; Cra wford, 198 1a , 198 1 b; Ga uthre au x a nd Bel se r, 2 006; New ton, 2007). But tl,e se are n o lo nger the predo min ant so ur ces of morta li ty, poss ib ly becau se t il e ni g h t landscape h as c h anged radicall y si n ce early re po rts of m Cl SS co llision eve n ts at t a ll s tru c tures lik e the Wa s hin g to n Monullle nt a nd Statu e of Libe rt y . These and o th e r s t ru c tu res were once beaco n s in areas of relativ e da rk n ess, but arc n ow surro un ded by square miles of li g h t p o ll ut ion. W hil e collisions at structures li ke cell tow e rs co ntinu e to take p lace at ni g ht, th e ma jority of co lli sio n s with buildings n ow take place duri n g the day . (I-la ge r, 2 0 14; Kahl e etal., 20 15; O lson , p e rs. camlll.) Pattern s of li g ht inten s it y see m to pla ya ro le in th e di s tr ibu ti o n of co lli s ions in t h e b uil t env ironment, how- eve r. Bird s may land in patte rn s di c tate d b y t h e pattern of lig ht intens ity in an area, so th e b rig hte s t b uildings a rc th e Ill os t lik e ly to cause co lli s io n s early in t h e day. As bird s move through the landscape seekin g food, pat- te rn s re lated to distribution of veget ation appear. Studi es us in g radar t o map m ovem e nt of bir ds th ro u g h the built e n v ironm e nt a rc s tartin g t o appear, b ut we n eed in fo r- mation CIt the leve l of spec ies a nd indiv id ual s to trul y und e rs tand h ow li g ht is im pacti n g b irds. It is o ften sa id that birds arc attracted to li g h t s at ni g ht (Gauth rea u x a nd Be lser, 2006; Poot ef al. 2008). J-IO\\'cve r, we d o n o t ha ve direct ev id ence that b irds are, in fact, attract ed t o li g h t s; t h ey may si mpl y respolld t o li g hts th ey encou nter. G authrea u x a nd Bel se r q uo te Ve rh eije n as su ggest in g that "capture " m ig ht be a better wo rd for b irds' respo n se to night li g htin g. Wh il e "cap ture" d ocs se em appropriate to describe t h e ph e nom e n on of bird s c ircl in g d rilling p latform s, o r in th e li g hts o f th e 9/11 Me m o ri al 's Tr ibute in Light in Ma nh attan, tl d iso ri e ntati o n 'l is a term t h at cove rs morc o f t h e s p ect ru lll of be h aviors see n w h en birds inte ra c t w ith li g ht at ni g ht. Ga uthrea u x an d Be lse r (2 006), re porting u npu b li sh ed data l s tated that "exposu re to a li g ht fie ld ca u ses alte rati o n o f a s t raight Ri ght pat h (for exa mpl e Il overing, s low in g clown , s hiftin g direc t io n, or c ir cl ing)," and t hi s ha s bee n repo rted b y othe r aut h o rs . Lark in and Fra se (1988, in Ga uthreau x and Belser, 2006) u sed p o rt a bl e tra c kin g radar to reco rd fl igh t paths of birds near a broad cast towe r in Michigan. Bird s s h owed a ran ge o f res p onse, rrom Circ lin g to arcs to li near ni g ht. Ha u pt and Sc hill emeit (20 11 ) described th e path s of 2 13 birds fly in g thro ugh up-li g htin g fro m several d iffe re n t o utd oo r lig htin g sc h e m es. O nl y 7.SC}b s h owed n o c h ange in be h av io r, w hi le the rema inder dev iated fro m t h eir courses by vary in g d egr ees, from mi ni m a l co ur se d ev ia- t ion through Circ lin g . It is n o t known w h eth e r respo n se diffe re n ces a rc s pec ie s re lated. Il o ls h akov et al. (2010) d eve lo ped t h e Optica l-E le ctroni c Dev ice to stud y n oc turnal migra ti o n behavio rs of so n g b ird s. In sp ired by the more limited tec hniqu es of moo n watc hin g and watc ll in g b ird s cross ce il 0 11l c te r li ght bea m s, th e dev ice u ses sea rchlig h ts to illumin ate b ird s from th e grou n d, w hil e a reco rdi n g unit d ocum e nts th e bird s' m ovemen ts. W ith th is tec hn iqu e, they can stud y 1) ground-and a ir speed ; 2) com p e n sation for w ind drift o n the ba sis of di rec t m easurement.s of headings a nd track dire c tions o f in d iv id ua l bird s; 3) win g-beat pattern a nd it s variation dependin g o n Swainson's Thrus h . Photo by O wen Deutsch Th e gla ss \\'alls of th is atr ium , cou pl ed with nighttime illumination, creat e a n ex tre m e collisi o n hazard for h irds. I'h o to co urt esy o f New York C it y I\uduho n Cal\;lda W;ub1cr . Photo by Ted i\rdle~' w ind directi o n a nd ve loc it y. In so m e cases, spec ies c an be i de ntin ed. Bo lsha kov et 01. (20 13) ex amin ed th e e ff ec t s o f w ind co nditi o n s o n num be rs of bird s al oft ancllli g ht traj ec to ri es of b ird s c ross in g th e li gt'lt bea m f ro m the ap para tus . Th ey d e t e rmin ed th a t num bers o f b ird s d o diffe r with wind stre n g th, but that b ird s m ay be a ttrac t e d to the li g ht b ea m und e r calm conditi o n s . Th ey al so found that th e li g ht beam di s turbs s trai ght ni g ht traj ec to ries, es pe ci a ll y in calm w ind co nditi o n s . Regress io n m o d e ls su gges t th a t th e prob a bility o f curved lli g llt t raj ec to ri es is gr ea ter for s lllall b ir ds, es p ec ia ll y w h e n th ere is li ttl e o r n o m oo n . Bul y uk el al. (20 14) u se d th e sa m e d ev ice to co m pa re be- h av io rs of n ig h t -mi gratin g passe rin es u nde r n a tu ra lnoc - turna l illu mination (at t h e Cou ris h Spi t of t h e Ba lti c Sea) wit h bi rds pass in g t h ro u g h an urban li g h t e n v iro nm e nt (i n s ide th e ci t y li m its o f st. Pe te rs bu rg, Ru ss ia ). So n g birds we re d isti ngui sh ed as e ith e r smal l p asse rin es o r thrush es . Th e illumin ated ba c kgro u n d ca u se d a d ecrease in im age q ualit y. Th e s h a pe of ni g ht tra c ks was co mpared fa r th e two groups, a n d a la rge r propo rti o n o f small so ng bi rd s ch an ged ni g ht path w h ile cro ss ing t h e lig h l. This co u ld be exp lain ed by nigh t type or fli g h t speed. Th e propo r tion of so n gb ir ds c h a n g in g fl ig h t trajecto ry ill t h e lig hte d co ndi- ti o n was m uc h sm a ll e r th a n und e r t h e d a rk co n d it io n. T o un ders t a n d exactl y h ow li g h t affects b ir ds a n d w h a t ac ti o n s m ust b e t a ke n to red uce th ose effec t s, we n eed to kn ow mu c h m o re . fo o r ex a m pl e, a t w h a t ran ge (h o ri zon- t a l a nd ve rti cal ) and und e r what co nditi o n s d o b ird s fee l di sru p ti o n from li g ht, and o f w h a t inte n s ity and wa ve- le n g th co mpos iti o n? How d o th ese felc t o rs c h a n ge th e ir be h av io r? Does ni g ht li g h t in g ha ve an y effec t o n b ird s d epa rt in g a t th e beg innin g o f mi g ra to ry s t ages? Do we eve r a ctu a ll y see bird s c h a n g in g co urse t a m ove t oward a b ri g ht li g ht so ur ce? Light Color a'nd Avian Orientation Startin g in th e 1940s, ce il o m ete rs-powcrful b ea m s o f li g ht u se d t o m eas ure th e h e ig ht of c loud cove r-ca m e int o li se and we re associat e d w ith s ig nifi ca nt bird k ill s. Fi lt e rin g o ut lo n g (re d ) wa ve le ngth s an d u s in g th e blu el gr een ran ge g reatl y re du ce d mortality, alth o ug h we d o n 't kn ow w h e th e r th e inte n s iti es of tl1e se two co lo rs o f li g hts we re equal. La te r, re pla ce m e nt of lix e d-b ealll ce il o m e te rs w ith ro tatin g be ams esse ntiall y e limin ated th e impac t o n migratin g bircl s (Las key, 1960). A co m p l ex series o f la b o ra t o ry stu dies in t h e 1990s d e m onst ra t ed t h a t b ird s re quired li g ht in ord e r t o se n se th e Eart h 's m ag n e ti C fle Jd. Bird s co uld o ri en t correc tl y un der mono- c hro m a ti c bl ue or g ree n li g ht, bu t lo n ger wave length s (ye ll ow an d re d ) ca u sed d isor ientati o n (Rapp li et al " 2000; Wi l tsc hko et (I I., 1993,2003,2007). W iltsc hko et a l . (2007) s h owed th a t a bove inte n Sity thres h o ld s t h a t d ec rea se fro m green to UV, bird s sh owed di so ri e nta ti o n . Di so ri e nta tio n occ u rs at' lig l1t le ve ls th a t arc s ti ll re la- ti ve ly low, e qui va le nt t o less than half an h o ur b efore s unri se und e r cle ar s ky. Poot et 01 . (2008) d em o n st ra te d th at mi g ra t ing b irds ex- p osed to vari o u s co lo re d li g h ts in t h e fi e ld res p o nd ed th e sa m e way as th ey do in th e labo ra t o ry. Bird s respo n ded strong ly t o w h ite a nd re d li g hts a nd app ear ed di so ri e nt- e d b y t h e lll , especia ll y under overcast sk ies. Green li g ht provoked less res po n se a nd minim a l di so ri e ntati o n ; b lu e li g ht attrac te d few birds a nd did n o t di so ri e nt th ose th at it did a tt rac t . Bird s we re n o t attrac te d t o infrare d li g ht. Ev an s et (II. (2 0 07) also tes t ed different light co lors but did n o t see agg reg ati o n unde r red li g ht. Ho we ve r, th ey s u bse qu e ntl y d e te rmin ed that th e in t e n sity o f re d li g ht u se d was less than fo r oth e r wa ve le ngth s, a nd wh e n th ey re pe ateel t h e tr ial w it h hi g h e r in te n Sity red , th ey did sec ag gregat io n (Evans, pe rs. co mm . 2 0 1 1). 56 Scie nti s ts work in g in the Gulf o f Mexico (Ru sse ll , 2005 ), th e No rth At lanti c (W iese et (/1. 200 I), and t h e Nor th Sea (Poot et 01.2008) report that bri g h t lights of ocea ni c drill ing rigs in duce circl in g be ha vior and mortality in birds at ni g ht. Worki n g o n a rig in the Nor th Sea, Ma rq ucnic et(/I. (20 13), estimate d that b ir ds were i'lfrcc Lcd u p LO nve kil ome le rs away. Replaci n g about half th e lights w ith n ew bu lbs e mittin g minimal red light red uc ed c ir cling be havi o r by about 50%. The authors spec ul ate that com pletel y re-Iampi n g th e platform wou ld redu ce b ird aggrega ti o n by 90%. Ge hring et al. (2009) d e m o n s t rated that m o rta li ty at co mmunication t owe rs was grea tl y reduced if s trobe li g h t in g was u se d as opposed to steady-bu rnin g w llite, or especia ll y red li g hts. At t h e 9 /1 1 Memorial Tr i bute in Li g ht in Man h attan, w h e n birds (]ggregate and ci rc le in the b ea m s, m o nil o rs (urn lh e Ii g hl s o ul brieOy, re leasi n g th e b ir d s (Elbin, 20 15, pe rs. com m .). Reg ul a r, s h ort inte rva ls of da rkn ess , o r rep lacement of s t eady-burn in g warn ing ~U ~.1I1 F.lbin t e~t ~ oJ bird in Ihe tunne l althe Carneg ie ~"'us e um 's I'owci e rrnlll l1<1mlln g StatI o n In m lllhwcs lcrn "cnnsy lv,1nia . I'h o lo by C hristine Sh e ppard, AlI C li g hts w it h in ter mi tten t li g hts, arc exce ll ent opti ons for protecting bird s, .1Ild manipulatin g lig ht co lor also has promise, alth o ug h add ili o nal field tria ls for colored lights arc need ed. Research : Deterring Collisions Sys temati c efforts to ide nti fy Sig n als that ca n be used to make glass v is ib le to birds began with th e wo rk of Dr. Dan ie l Kle m in 1989 . Testing glass panes in the neld and u s in g a dic h o to moll s c h o ice prolocol in an aV iary, Kle m (1990) de m onstrated t h at po pular d evices like IId iving falc o n" silh ouettes were effective o nl y if tlley we re applied d e nse ly, s paced two to fo u r inch es apart. Ow l decoys, blinking h o lida y li g hts, and pictures of ve rtebra te eyes we re among it ems found to be in effe c ti ve. G rid and stripe patte rn s made from wh ite mate ri al , o ne in ch wiel e, were tes ted at different spacin g intervals. O nl y t hree were effective: a 3 x 4-in c h grid; ve rtica l strip es s pa ced fo ur in c h es apart; and h orizont al G lass panes arc b eing tes ted at the I'owderrnill Tunnel, a s seen from th e outside. Ph oto by C hrI st in e Sh e p pard, A BC T h e tunnel : <In a pp;H<ltu~ for 5'1ft'l y te sting c ffe c tiv e n e~s o f mate rial s a nd deSig ns for d e terr in g bird colll ~J o m . I'holo by C hri sti n e Shcppard, AII C A bird's.e}'e \'iew o f gla ss In the tunne l. Ph o to b y C hri stln e She ppard, AII C The tunnel at I'owcl c rlllill . shOl"ing tht" rralll e w o rk whNC Ihe ba Ck ground will h e ll1 o lllut:'d . Photo by Ch ri stine Sheppard. Aile st ri pes s pa ced about an in ch apart across the entire surface. (A summa ry of Klcm 's res ult s can be fo und at coli isionS.a bcb i rd s.o rg). Bu il ding 011 Kle ill 's lilldillgs, Hoss le r deve loped a tes ting program in Austria startin g in 2004 and co ntinuing to the present (Hossler and Zuna-Kr at ky, 2004; Ros s le r, 200S; Ro ss le r, et (11.,2007; Ro ss ler and Laube, 2008; Ros s ler, 20 10; Rossler, 2012; Ro ssler, 20 13 ). The banding center at th e Ho h e nau Rin gclsdorf Bio log ical Station outside Viell na , Austria, o ffered a large sampling of birds fo r eac h tes t, in so m e in stances permitting compariso n s of a particu lM pattern under differing intens iti es of li g h t in g. Thi s progra m ha s foc used p rimarily o n geo m e tri C patte rn s, eva luatin g the impa c t of spac in g, or ie ntat io n , a nd dim e n s io n s. Bird s arc p la ced in a "tun ne l," w h ere lh ey ca n v iew tw o pie ces of glass: one unmodifi e d (t h e co nlro l) and l h e o lh er v,'ith the pattern to be tes t ed. Birds Il y down th e tunn el and are scored accordi n g to whether the y try to exit through the control o r th e pattern e d glass. A mi st n et ke eps th e bird fro m hitting the g lass, and it is th en released. The pro ject fo cu ses n ot only o n findin g pilttern s effec tive for deterring collis io n s, but al so o n effect ive patterns that cove r a minima l part of the g la ss s urface . To date, so m e patterns th a t cove r o n ly 5% of the g lass h ave bee n fo und to be hi g h ly e ff ec t ive. (1\ s ummary of Ro ss ler 's res ults ca n be fou nd at co lli s ions. abcbirds.org). Buildin g on Ross le r's wo rk , ABC co ll aborated w ith th e Wild life Co n serva ti o n SOC iety, New York C ity Audubon, a nd th e Carneg ie Museu m to const ruct a tunne l at Powd e rmill Na ture Rese rve 's banding stati o n, primarily to te st co mmerCially ava ilable material s. Re sults fro m the fir st seaso n sh owed that making an ent ire surface UV- re ll ec l"i ve was not an e ffective way to deter birds. With UV material s, contrast see ms to be important. G la ss fritted in patterns co nform in g to the 2 x 4 ru le, howeve r, sco re d we ll a s dete rr e nts. (A summary of results from Po wcl e rmill can be found at co lli sio n s.abcbirds.o rg). Most clear g la ss Illad e in the United States tran s mits about 96 % .In d re fl ects about 4% of light falling perpendicular to the o utside surfa ce. Th e amou nt of li g ht re ll ected in creases at s llarper a n g les: clear glass reflec ts abo u t-50% o f in c id e nt light at a n g les over 70 degrees. Li g h t o n the insid e o f the glass is also partl y re fle c ted and par ll y tr ansmitted. Th e re lative inte nsiti es of li ght tr a n s lllitted from th e in si d e and re fl ec ted from th e o utsid e surfa ces o f g la ss combined with the viewing angle d e termine whether th e g la ss appears transparent or mirrors the s urr o unding e n v ironm ent. Patterns o n the in s ide s urfa ces of glass and objects inside th e g la ss may n ot always be v isibl e . These c han geab le optical properties s up port th e argum e nt that patterns applied to the o uter s urfa ce of g lass are more effec tive than pattern s applied to the inn e r s urfa ce. Efforts Ilave b ee n made t o model freesta ndin g g lass, g la ss in s tall e d o n a building, and re fl ec ti o n s o n g lass in some tTials . (Th e t es tin g protocol fo r fre es tanding g lass, d eve loped at Hoh e n au, and th e t esting protoco ls use d at Powde rmill ca n be fo und at co lli sion s .abcbirds.o rg). 58 H o ri zo n ta l lines with a m ax imum spacing o f 2 in ches 2'~ Hed-br easted N ulh.,\c h . [,holo b}' Itoy 1 [ancliff The 2 x 4 Rule Research on son gbird s. the most numerous victims of co lli - sio ns. has shown that horizontal lines m ust be two o r fewer inch es apart to deter th e majo rity o f birds. Vertical spaces must be four or fewer inches apart. This difference p resum- ably has to do with the shape of a flyi ng bird. (Narrowe r spacing is required to d eter collisions by hummingbirds.) Schiff ner et al. (2014) showed that budgies have a very pre- cise understanding of their own physical dimensions. Trained to fly in a tu nnel. t he b ird s w e re then challenged to pass through eve r narrowing gaps . They were able to assess the Vertical lines with a maximum spaci ng of 4 in c h es .. 4" width o f the gaps re lative to their body size and adjust their flight behavior accordingly. It seems likely that this is a gene ra l avian trait. usef ul for navigating co m plex envi ro nm e nts at fli ght spee d. Bhagavatula e t al. (2011) used the same tunne l setup to investigate how opti ca l flow cues gU id e fl lg llt. It appears that birds balance the speeds of images perce ived by both eyes. in this case. images to the birds' sides . This reinforces the suggestion of Ma rti n (2011) that humans expe rience the world as something ahead o f th em. while fo r birds In fli gh t. what IS ahead of them is not necessari ly their pri mary focus . I Lf) E Q) ..0 o '---- D- +--' CD =s ro > w o o f- « Ofte n , on ly part of a b uilding is respons ibl e fo r ca u sin g m os t of th e co ll isio n s. Evaluation and d ocum e ntation ca n h e lp in th e deve lop m ent of a p rog ram o f re m ed ia- ti o n target ing that ar ea . Remediati on G ill be a lm ost a s e ffe c ti ve as modify ing th e e ntire buildin g, a s we ll a s less ex pe n s iv e. Documentation of patte rn s of m o rt a li ty and e nvironm e ntal features th a t ma y be co ntri bu tin g t o col - li s io n s is esse ntial. Ope rati o n s p e rso nn e l ar c o ft e n good sou rc es of info rm ation for co mm erci al bu il d in gs, as they m ay come across bird carcasses w hil e perfor ming regu lar m a inte nan ce activi ti es. People who work n ea r windows are oft e n aware of bird s hitting th e m. Regular mo ni to rin g n o t o nl y produces data o n th e m agnitude and patterns of m o rtality, but al so prov id es a ba se lin e fo r d e m onstra ting improve m e nt. The best m o nitor in g progra m s fea ture con siste nt e ff o r t, careful doc um e ntation of coll is io n locati o n s, and acc urate id e ntifi cati o n of v ic tim s. Eff ective m o nito rin g Sil ou ld doc um e nt at least 18 m o nth s of co lli s io n s befo re mitigation is a t temp t ed, unl ess colli sio n rates are especia ll y hi g h. (Reso ur ces for mon it ori n g, from simple to sop hi s ti cated , can be fo und at co lli sions.abcbi rd s.o rg). So l utions Ma n y factors co m e into p lay in se le ctin g h ow to make g la ss sa fe fo r b ird s . Th e tabl e belo w co mpares co mm o n solution s a cco rd in g to th e ir effectiv e n ess, app earance, re lati ve cos t, case of a pplicati o n, lon gevity, a nd require d m a intena n ce. Effe ctive patte rn s o n the exte ri o r s urface o f g lass w ill co m bat reflectio n , tran spare n cy, and pa ssage e ff ec t. Within th e 2 x 4 g uide lin es, ho weve r, co n sid e rabl e variati o n is po ss ible when d ev is in g bird -fri e ndl y patte rn s. We re com m e nd th at lin es b e at lea st lA-inch wide, but it is n ot n ecessa ry that th ey be o nl y ve rtica l or h o ri zo ntal. Co ntrast b e tw ee n pattern and backgro und is im portant, however, and deSig n ers s h o uld be aware that t h e backg round-b uil d in g inte ri o r, sky, vegetat ion- ma y c h a nge in a p peara n ce t hro u g hout the da y. COMPARISON OF RETROFIT OPTIONS Material Effectiveness Cost Application Appearance longevity Upkeep Seasonal. ****. S • • na na temporary solutions Netting ***** SS .. .,. **** ... Window fitm ***** SSS **** ***** ._-**** Screens ***** SS ... ***. ***** **** Shutters ***** SSS ._. **** ***** **** Gri ll es ***** SSS **** ****-11 ***** **** Rep lace glass ***** SSSSS ***** ****. .**** *.** 5 stars/dollars h ighly expensive easy attractiVe long-lasting minimal effective T h is sec ur ity g rill~ cr~ates a patt e rn tbat will deter birds fr o m fl yi ng 10 refl ec tion s. Ph o to by Chris tine Sh e ppa rd , AB C Fog increa se s the d ange r o f light bOlh hy ca us in g bi rds \0 ny lo wer and b y refra c t ll1 g li g ht so It is visih le o ve r a lar ge r ar ea . Ph o to by C h rist in e Sh eppard . ABC 50 81fd-he.,aly Au:la''-'9 Des'9~ T ile fo ll owing qu es ti o n s can guide the eva luation and d ocum e ntation process by h e lpin g to identify feat·tlr es li ke ly t o cau se co ll isio n s and oth er impo rtant factors. Seasona l T i ming Do co l li sio n s hilpp e n mos tl y during mi gr ati o n o r n edg- in g pe ri o d s, in winte r, o r ye ar roundl l f co lli s io n s hap- pe n o n ly d ur in g a s h o rt tim e p e r io d , it ma y b e p oss ibl e to appl y in ex pe n s ive , t e mp o ra ry so lution s during that tim e and re m ove th e m fo r the rest o f th e year. Som e b ird s w ill attack th e ir ow n re nec ti o n s, es peciall y in sprin g. Thi s is n o t a tru e co ll isio n . Territorial ma les, es p ec ia ll y Am e ri ca n Ro bins and No rth e rn Cardinal s, perce ive t h eir re fl ec t-io n as a rival mal e. They are un- li ke ly to inju re th e m se lves, and te mp o raril y b loc kin g reflec ti o n s in th e offe ndin g wind ow (a nd th ose n earby) fr o m th e o utsid e s h ould reso lve the proble m . Tap i n g up pa p e r and s m earin g a soa p pa st e ca n bot h be effect ive. Weather Do co lli sio n s co in c id e w ith parti c ul ar wea th e r co n cli- ti o n s, s u c h as fog gy or overcast da ys? Suc h colli sion s ma y be li g ht-re lated , in w hich ca se a n e mail notifica- ti o n syste m, ask in g bui ldi n g perso nn el to turn o ff li g hts w h e n bad weather is fo reca s t, is advisab le . D iu rna l T imi n g Do co lli sio n s happe n at a p a rt icu lar tim e o f da y? Th e appearan ce o f g la ss ca n c h a n ge sign ifi cantly with diffe r- e nt li g ht leve ls, direc t o r ind irec t illum ina ti o n , and sun angl es. It m ay be poss ibl e to simpl y li se shad es or shu t- te rs durin g cr iti cal tim es. La WN -fl oor windo ws ar c Ih o ught 10 b e more dangerou s to bird s because Ih ey a rc m ore likel y \0 rencct vege ta tio n. Ph o to by C hri st ine Sh ep pard, AIl C Locatio n Are there particular w indo \/',"s, gro ups of windows, or bu ild in g fa.;ades t h at account fo r most co llisions? If so , it m ay b e cost e ffec ti ve to modify o n ly t h ose sect ion s o f g lass. Is glass lo cated where birds fly be tween roos ting or n es tin g and feed in g sites'? Are th e re area s w h ere plants ca n be see n through g lass-for example, an atrium , cou rty ard, or g la ss building co nn ect o rs? Arc th e re a rc hitec tural o r lan dsca pin g features that te nd to dir ect birds toward g la ss? Suc h fea tures mi g ht includ e a wa ll o r roc k o utc ropping or a path way b ord e re d by dense vege tation. So luti ons include u sing a sc ree n or t re lli s t o diver t fli g ht paths. Are th ere fruit tr ees, be rr y bu s h es, or o ther plants n ca r windows that ar c lik e ly to attrac t birds cl oser to g lass? Th ese w indows s hould be a hi g h pr ior it y for re m e diation. Th e gla ss itself ca n be mod i fi e d , but it may al so b e poss ibl e to u se li ve o r inan im a te land scap in g el e ments to block th e v iew betwee n food so ur ces a n d w indmvs. 6 2 Local Bird Popula tio n s What ty p es of birds a rc lIsuall y fo und in an area? Loca l bird gro ups or vo lunteers may be able to hel p c haracterize loca l and tran s itory bird popu lations, as we ll a s Ih e m os llike ly ro utes ro r bi rd s making short fli g hls arou nd th e arca . The Am e ri ca n Birding Association , Bird 1,Vatcl,ers D isest, Audubo n c hapters, an d Birding.com a rc good p la ces to s ta rL fillding s ll c h resource s. U n iversiti es, co ll eges, and mu se um s may al so be h e lpfu l. --- Post-Mitigation Monitoring M o nitoring efforts sh ou ld co ntinue f o r at l east 18 months after m itigatio n efforts arc m ade, and fo r at lea s t tw o p eak co lli sion sea so n s (o ft e n the (a ll in urban area s, but s pring and Slimmer ITI cl Y a lso be p ea k se a so n s in m o rc rllfal locati o n s). Colli si o n rates vary al o n g w ith l oca l b ird popUlat io n s, so a ye ar of hi g h population and hig h co llisi o n s may b e followed by a year of l ow populati o n s and l ow co lli si o n s, reg ard l ess of th e effective n ess o f (In y mitigation. Use of g la ss with .. hl~h l y effec ti ve hurlz"!]Ii\J frit !!iU1!.au. together wit h SUll s h (lc lcs, e arn ed th is reuofitlcd huildill~ oll lhc SUN Y Brockp o rt Ci llllJlU S th e l.EE\) "co lli si on d e te rren cc" c rcllil. Ph nlo h y 1',lU I Tanke l T his Oven bi rd survive d a collisio n a nd was recovered alive during II Lights Out monitoring effort In lIalllmore , Marylill1d . Phuto by Daniel J. I.cbbln, AIlC References , Avery, M.L., P.F. Springer andJ.F. Cassel, 1977. Weather influences on nocturnal bird mortality at a North Dakota tower. Wilson Bulletin 89(2):291-299. Avery, M.L., P.F. Springer and N. S. Daily, 1978. Avian mortality at man-made structures, an annotated bibliography. Fish and Wildlife Service, U.S. Dept. of the Interior: Washington, D.C. 108 pp. Bayne, Erin M., Corey A. Scobie and Michael Rawson, 2012. Factors influencing the annual risk of bird-window collisions at residential structures in Alberta, Canada. Wildlife Research http://dx.doLorg/l0.1071/WR11179 Bhagavatula, Partha S., Charles Claudia nos, Michael R. Ibbotson and Mandyam V. Srinivasan, 2011. Optic Flow Cues Guide Flight in Birds. Current Biology 21: 1794-1 799. BIem, C.R. and B.A.Willis.1998. Seasonal variation of human-caused mortality of birds in the Richmond area. Raven 69(1):3-8. Bolshakov, C(lsimir V., Michael V. Vorotkov, Al- exandra Y. Sinelschikova, Victor N. Bulyuk and Martin Griffiths, 2010. Application of the Optical- Electronic Device for the study of specific aspects of nocturnal passerine migration. Avian Ecol. Behav. 18: 23-51. Bolshakov, Casimir V., Victor N. Bulyuk, Alexandra Y. Sinelschikova and Michael V. Vorotkov, 2013. Influence of the vertical light beam on numbers and flight trajectories of night-migrating songbirds. Avian Ecol. Behav. 24: 35-49. Bulyuk, Victor N., Casimir V. Bolshakov, Alexandra Y. Sinelschikova and Michael V. Vorotkov, 2014. Does the reaction of nocturnally migrating song- birds to the local light source depend on backlight- ing of the sky? Avian Ecol. Behav. 25:21-26. Codoner, N.A. 1995. Mortality of Connecticut birds on roads and at buildings. Connecticut Warbler 15(3):89-98. Collins and Horn, 2008. Bird-window collisions and factors influencing their frequency at Millikin University in Decatur, Illinois. Transactions of the Illinois State Academy of Science 101(supplement):50. Crawford, R.L, 1981a. Bird kills at a lighted man- made structure: often on nights close to a full moon. American Birds (35):913-914. . Crawford, R.L, 1981b. Weather, migration and autumn bird kills at a North Florida TV tower. Wison Bulletin, 93(2):189-195. Cuthill, I.C., J,C. Partridge, A.T.D. Bennett, C.D. Church, N.S. Hart and S. Hunt, 2000. Ultraviolet vision in birds. Advances in the Study of Behavior 29:159-215. Davila, A.F., G. Fleissner, M. Winklhofer and N. Petersen, 2003. A new model for a magnetoreceptor in homing pigeons based on interacting clusters of superparamagnetic magnetite. Physics and Chemistry of the Earth, Parts AlBIC 28: 647-652. Dunn, E.H. 1993. Bird mortality from striking residential windows in winter. Journal of Field Ornithology 64(3):302-309. D'Eath, R.B., 1998. Can video images imitate real stimuli in animal behaviour experiments? Biological Review 73(3):267-292. Elbin, Susan, 2015. Pers. comm. Evans, W.R., Y. Akashi, N.S. Altman, A.M. Manville II, 2007. Response of night-migrating songbirds in cloud to colored and flashing light. North American Birds 60, 476-488. Evans, W.R., 2011 Pers. comm. Evans, J.E., I.C. and A.T. Cuthill, D. Bennett, 2006. The effect of flicker from fluorescent lights on mate choice in captive birds. Animal Behaviour 72:393-400. Evans-Ogden, 2002. Effect of Light Reduction on Collision of Migratory Birds. Special Report for the Fatal Light Awareness Program (FLAP). Fink, L.c. and T.W. French. 1971. Birds in downtown Atlanta-Fall, 1970. Oriole 36(2):13-20. Fleissner, G., E.Holtkamp-Rotzler, M. Hanzlik, M. Winklhofer, G. Fleissner, N. Petersen and W. Wiltschko, 2003. Ultrastructural analysis of a putative magnetoreceptor in the beak of homing pigeons. The Journal of Comparative Neurology 458(4):350-360. Fleissner, G., B. Stahl, P. Thalau, G. Falkenberg and G. Fleissner, 2007. A novel concept of Fe- mineral-based magnetoreception: histological and physicochemical data from the upper beak of homing pigeons. Naturwissenschatten 94(8): 631-642. Gauthreaux, S.A. and C.G. Belser, 2006. Effects of Artificial Night Light on Migrating Birds in Rich, C. and T. Longcore, eds, 2006. Ecological Consequences of Artificial Night Lighting. Island Press. Washington, DC. 259 pp. Gauthreaux, Sidney A. Jr. and Carroll G. Belser, 2006. Effects of Artificial Night Lighting on Migrat- ing Birds in Ecological Consequences of Artificial Night Lighting, Catherine Rich and Travis Longcore eds. Island Press, Washington, D.C. 458 pages. Gehring, J., P. Kerlinger, and A.M. Manville. 2009. Communication towers, lights, and birds: successful methods of reducing the frequency of avian collisions. Ecological Applications 19:505-514. Gelb, Y. and N. Delacretaz. 2006. Avian window strike mortality at an urban office building. Kingbird 56(3):190-198. Ghim, Mimi M., and William j. Hodos, 2006. Spa- tial contrast sensitivity of birds.J Comp Pilysiol A 192:523-534 Gochfeld, M.,1973. Confused nocturnal behav- ior of a flock of migrating yellow wagtails. Condor 75(2):252-253. Bird-Friendly Building Design 5~ 65 Greenwood , V., E.L. Smith, A.f{. Golds mith , I.C Cuthill, 1...1-1. Cr isp , M.B.W. Swan and A.T.D. Bennett, 2004. Does the flicker frequency o f liuorescel1 t lighting affect the welfare of captive European starlings? Applied A llilllal Behaviullr Sci ellce 86: 145-159. I-la ger, S.B., H. Trudell, K.J. McKay, S.M. C randall , L. Mayer. 2008. Bird den sity and mortality at windows. lIVi/sulI /ollru al ofOru ilh%S), 120(3):550-564 . Ha ger, Stephen B., 2009. HUlllan ·re lated th reats to urban raplors. f. R(/plor Re s. 43 (3):2 10-226. Ha ge r S.B., Cosenti n o 13 . .1 ., McKay K.j., Mo n son c., Zuurdceg W., and B. Blevins, 20 13. Window Area and Development Drive Spa ti al Variat ion in Bird·Window Colli sion s in an Ur ban Land scape. Thc Wo rld Tradc Ccnter or New Orleans, de signed by Ed ward Durre ll St o ne, m C$ a simple hird-friend]}' stratcHY; ;lirnost all windows havc exterior shutt ers. Ph u to 11)' C hri stine Sl1 e pP;lId , AII C !'LoS ONE 8(1): e533 7 1. do i: 10.137 1 /jou rna l. pone .005337 I Download at: htt"J)://peo pl e.hws.eclu/cose nt"ino/ publ ica tions_fi les /PL o 51X,200 N 1:.1111 2020 13 (Vt1 20 Hager.pdf I-lager S.Il., Cra ig M.E. (2014). U;rd -w indow coll isio n s in the s umlll er breeding se<lso n. Peer j 2: e460 h ttps :! Idx.doLorg/1 O. 77 1 7/peerj.460 Ha upt , 1-1. and U. SchilJcmcit, 20 1 1. Skybetlmcr und Gc iJ ii udca nstrahlun ge n bringen ZugvogcJ vom Kurs ab: N eue Untersuchungen lind cine rec htli che Bewertung dieser Li c htanlagen. Nu L 43 (6),20 1 I , 165-1 70. (Search/spo\" Li ghts and Building Lighting Divert Mig ratory Birds orr Course: New investigations <md a le gal evaluatio n of t h ese li g hting system s) Herber t, A.D., 1970. Spatial Disorientation in Rird s . Wi/SOli Bllllelill 82(4):400-419. Jones, J. and C M. Fran c is, 2003. Th e effe c ts of light c haracteristics on avian mortalit y allightholl ses. jOllmal o( A viml BioloS), 34: 328-333. Kahle , Logan Q., Maure en E. Flann e ry and John P. Dumbachcr, 20 15. Bird·window c ollisions a t a west coa s t urban parklan d: a n a lyses of bird biology a nd window a ttribut es from Golden G al c Park, San Franc isco. In press, PLOS Olle. Kerlingcr, P., 2009. I-Imv Bird s Migrate, second ed ition , revisions by Ingrid Johnson. Stackpole Books, Mechanicsville, PA. 230 PI'. Klclll , D., Jr., 1990. Collis ion s b e twee n bird s and windows: Mortality and prevention. /ollma/ of Field Omit/IU/os), 61 (1):120-128 . Klem , D., J r., 1989. Bird·window collisions. I'Vi/so ll /JII//etill 101 (4 ):606-6 20. Klein, D.,Jr., 1991. G lass and bird kills: An overview and sugges teel pl a nnin g and design methods of preventing ,I fatal h azard. Pp. 99·104 in L W. Adams ane! D. L.. Leedy (Eds.), Witd life C o n servatio n in MetropOlitan Environments. Nat l. Ins1. Urban Wild \. Sy mp . Scr . 2, Colu mbia , MD. Klem, Daniel Jr., and Pete r G. Saengcr, 20 13. Evaluating th e Effectiveness of Select Visua l Sig nal S to Prevent fiird·window Co llisi o n s. TIle W i lsoll 10/lmo/ of Om ilil%g)' 125(2):406-4 I. Download at: http://www.mu hl enberg.eelli / 111 a in/acaelcll1 ics/biology /f aeu I t Y /k I c m/ aco/ Bird-wi nelow.h tm l Klem, D. Jr., D.C. Keck , K.L. Marty, A.j. Miller Ba ll , E.E. Nicill, C T . Platt. 2004. Effects of window <Ingling. feeder placement, and scavcngers on avian mortality at plate glass. \,Vil.'WII BIII/erill 11 6(1):69-73. Klelll, D. Jr., c.J. Farmer, N. De lacretaz, Y. Gelb and P.G. Sacnge r, 2009a . Architectural and Landscape Ri s k Factors Associated with Bird-Glass Collision s in an Urban Environment. \rVi/sOII/ullmal of Omilil%S), 12 1(1 ):126-13 4. Klell1, D. Jr ., 2009. Prcventing l3ird·Window C ollision s . \'Vi/soll/ullmal ufOmillw/ogy 12 1 (2):314-3 2 1. Laskey, A., 1960. Bird migration casua ltie s and weather conditions, Autumns 19 58, 1959,1960. T/l e Misra/lt 3 1 (4): 6 1-65. I.e bbin, Daniel J., Michael J . Parr and George 1-1. FenWick, 2010. Tile Alllericm l Bird CUIISerV(lII c), G llide 10 Bird COlIs er va tiuII. University of C hicago Press, C hicago. 447 pages. \.ongcore, Travis, Cathe rin e Ri ch, Pierre Mineau, Beau MacDonald, Daniel G . Bert , Lauren M. Sul li van, Erin Mutrie, Sidney A. Gauthreaux Jr, Michael L. Avery, Robert L. Crawfo rd , Alb e rt M. Manville I I, Emilie R. Travis and David Drake, 2012. An estimate of avian mortality at co mmuni ca ti on towers in the United 5t"ates and Ca nada. Longcore, Travi s, Catherine Rich , Pierre Mineau, Beau MacDonald, Daniel G. Bert, Liluren M, Sul livan, Erin Mutrie, Si dney A. Ga uthreau x Jr., Michael L. Ave ry, Robert L. Craw fo rd, Albe rt M. Manville 11 , Emilie R. Travis, and David Drake, 20 13. Avian mortality at comm uni Gltion towers 66 in the United States and Canada: which species, how many, and where? Biological Conservation 158:410-419. Loss, Scott R., Tom Will and Peter P. Marra, 2012. Direct human-caused mortality of birds: improving quantification of magnitude and assessment of population impact. Frontiers in Ecology and the Environment, September, Vol. 10, No.7: 357-364 Loss, Scott R., Tom Will, Sara S. Loss and Peter P. Marra, 2014. Bird-building collisions in the United States: Estimates of annual mortality and species vulnerability. Condor 116:8-23. Loss, S.R., Loss, S.S., Will, T., Marra, P.P. 2014. Best practices for data collection in studies of bird-window collisions. Available at http://abcbirds.org/?p=10399 Machtans, Craig S., Christopher H.R. Wedeles and Erin M. Bayne, 2013. A First Estimate for Canada of the Number of Birds Killed by Colliding with Building Windows. Avian Conservation and Ecology 8(2): 6. http://dx.doi.org/l0.5751/ ACE-00568-080206 Muheim, R., j.B. Phillips and S. Akesson, 2006. Polarized Light Cues Underlie Compass Calibration in Migratory Songbirds. Science 313 no. 5788 pp. 837-839. Muheim, R., 2011. Behavioural and physiological mechanisms of polarized light sensitivity in birds. Phil Trans R Soc B 12 March 2011: 763-77. Marquenie, j., and F.j.T. van de Laar, 2004. Protecting migrating birds from offshore production. Shell E&P Newsletter: january issue. Marquenie, j.M., M. Donners, H. Poot and Steckel, 2013. Adapting the Spectral Composition of Artificial Lighting to Safeguard the Environment. Industry Applications Magazine, IEEE 19(2):56-62. Mouritsen, H., 2015. Magnetoreception in Birds and Its Use for Long-Distance Migration. Pp. 113-133 in Sturkie's Avian Physiology, sixth edition, Colin G. Scanes ed. Academic Press, Waltham, MA, 1028 pp. Newton, I., 2007. Weather-related mass-mortality events in migrants. Ibis 149:453-467. Newton, I., I. Wyllie, and L. Dale, 1999. Trends in the numbers and mortality patterns of Sparrowhawks (Accipiter nisus) and Kestrels (Falco tinnunculus) in Britain, as revealed by carcass analyses. Journal of Zoology 248: 139-147. adeen and HAstad, 2003. Complex Distribution of Avian Color Vision Systems Revealed by Sequencing the SWS1 Opsin from Total DNA. Mol. Bioi. Evol. 20(6):855-861. 2003. adeen and Hastad, BMC Evolutionary Biology 2013, 13:36. The Phylogenetic Distribution of Ultraviolet Vision in Birds www.biomedcentral. com/1471-2148/13/36 HAstad and adeen (2014), A vision physiological estimation of ultraviolet window marking visibility to birds. Peerj 2:e621i DOl 10.7717/peerj.621 O'Connell, T. j. 2001. Avian window strike mortality at a suburban office park. Raven 72(2):141-149. Parkins, Kaitlyn L, Susan B. Elbin and ElIe Barnes, 2015. Light, Glass, and Bird-building Collisions in an Urban Park. Northeastern Naturalist 22(1): 84-94. http://dx.doLorg/10.1656/045.022.0113 Poot, H., B.j. Ens, H. de Vries, M.A.H. Donners, M.R. Wernand, andj. M. Marquenie, 2008. Green light for nocturnally migrating birds. Ecology and Society 13(2): 47. Rapplill R., R. Wiltschko, P. Weindler, P. Berthold, and W. Wiltschko, 2000. Orientation behavior of Garden Warblers (Sylvia borin) under monochromatic light of various wavelengths. The Auk 117(1):256-260. Rich, C. and T. Longcore, eds, 2006. Ecological Consequences of Artificial Night Ligizting. Island Press. Washington, DC. Richardson, W.J., 1978. Timing and amount of bird migration in relation to weather: a review. Oikos 30:224-272. Rossler and Zuna-Kratky, 2004 Vermeidung von Vogelanprall an Glasflachen. Experimentelle Versuche zur Wirksamkeit verschiedener Glas- Markierungen bei Wildvogeln. Bilogische Station Hohenau-Ringelsdorf [available for download from www.windowcollisions.info ]. Rossler, M. and T. Zuna-Kratky. 2004. Avoidance of bird impacts on glass: Experimental investigation, with wild birds, of the effectiveness of different patterns applied to glass. Hohenau-Ringelsdorf Biological Station, unpublished report. (English translation: available from ABC.) Rossler, 2005. Vermeidung von Vogel an prall an Glasflachen. Weitere Experimente mit 9 Markierungstypen im unbeleuchteten Versuchstunnel. Wiener Umweltanwaltschaft. Bilogische Station Hohenau-Ringelsdorf [available for download from www.windowcollisions.info ]. Rossler, M. 2005. Avoidance of bird impact at glass areas: Further experiments with nine marking types in the unlighted tunnel. Hohenau-Ringelsdorf Biological Station, unpublished report. (English translation available from ABC.) Rossler, M., W. Laube, and P. Weihs. 2007. Investigations of the effectiveness of patterns on glass, on avoidance of bird strikes, under natural light conditions in Flight Tunnel II. Hohenau- Ringelsdorf Biological Station, unpublished report. English translation available for download from www.windowcollisions.info Rossler, M. and W. Laube. 2008. Vermeidung von VogeJanprall an Glasflachen. Farben, Glasdekorfolie, getOntes Plexiglas: 12 weitere Experimente im Flugtunnel II. Bilogische Station Hohenau-Ringelsdorf (available for download at www.windowcollisions.info ). Rossler M. and W. Laube. 2008. Avoidance of bird impacts on glass. Colors, decorative window- film, and nOise-damping plexiglass: Twelve further experiments in flight tunnel II. Hohenau- Ringelsdorf Biological Station, unpublished report. (English translation available from ABC.) Bird-Friendly Building Design S!67 Rossler, M., 20 10. Vcrmcidung von Vogelanprall an Glasmic hen: Schwarze Punh:le, Schwflrz-o ra nge Markierungc n, EckelI' 4nird®, Evonik Soundstop®, XT BirdGuard. (available for d ownload from www . wi n dowcoll isions . in fo ). Ru sse ll , K., 2009. Pers comm. Rus se ll , Keith, 2015. Co nversa tion on August 13. Ru ssell, R.W. 2005. Interac t ion s behveen migrat- ing birds and offshore oil and gas platforms in the northern Gu lf of Mexico: Final Report. U.S. Dept. of the Interi or, Mineral s Management Se rvice, G ul f of Mexico oes Reg ion, New O rl ea n s, LA. oes Stud y MM5 2005-009. 348 pp. www.data.boom.gov/I.I/ I'D FI m ages/ 1'.51'15/2/29 55. pd f Sch iffner, Ing o, Hong, 0 Vo, Panna S. Bhagavatula and Mandyam V Srinivasan, 20 14. r..'fin ding the gap: in-night body awareness ill birds. FroJlliers ill Zoo/OS), 20 14, 11:64 http://www.fron tiersinzoology. com/content/l I /l/64 Scaly, S.G.,1985. Analysis of a sample ofTennes- sec \I\'arblcrs window-killed during s pring migra- tion in Man itoba. North Americ(l1/ Bird BO/uter 10 (4):12 1-124. Snyder, 1..1.., '1946. "Tunneilliers" aJld ,·vi ndow fatalitie s. CO lldor 48(6):278. Th omas, Robert J., Tamas Szekely, Innes C. C utl l ill, David G. C. Harper, Stu art E. Newson, Tim D. Fray- ling and Paul D. Wall is, 2002. Eye s ize in birds and the timing of song at dawn. Prof. It Soc. LOlld. B (2 002) 269, 83 1-8 37. 001 I 0.W98/rspb.200 1.1 941 Van De Laar , F.j.T., 2007. Green I.i ght to Birds, In vestigation into the Effect of Bird-rriendly Li ghting. Ne derlandse Aardolic Maatschappi j, Th e Net h erlands . 24pp. Vare la, F.j., A.G. Palacios and T.I-I. Golds mith, 199:t Co lor v is ion of birds. In Visioll , 13m in , (Illd Behavior i ll Birds , 1-1 . P. Zeig ler and 1-1 . Bisc h of cds., chapter 5. Weir, R.D., 1976. Annotated b ibliog raphy of bird kills at man-made obstacles: a review of th e st ate of the art and so lutions. Department of Fi sheries and the Environment, Ca nadian Wildlife Se rvice, Ontario Re gion, 1976. Wiese, Franc is K'J W.A. Montcvecchi, G.K. Davoren, F. H ue Hman, A.W. Diamond a n d j . Lin ke, 200 1. Seabi rds at Risk aroun d Off-shore Oil Platforms in the No rth -west Atlantic. I\I/arilie POI/UtiOIl B/Illetill 42(J 2): 1285-1290. Wiltschko, W., R. Wiltschko and U. Munro, 2000. Li g ht-d e penden t magn ctoreception in birds: the effect of intensity of 565-nl11 green li g ht. Na l llnvi ssel lsc!m {tell 87 :366-369. Wiltschko, W.,U. Monro, 1-1. Ford and R. W iltsc hko, 2003. Magn e tic orientation in birds: non-compa ss responses und e r monochromatic light of increased intensity. PI'O c. R. Soc. LOII(I. 13:270 ,2 133-2 140. Wi ltschko, W.,U. Monro, 1-1. Ford and R. W il tschko, 2006. Bird na v igation: wh at type of informat-ion d ocs the magnetite-based receptor provide? Proc. 11. Soc. B 22 Nove mber 2006 vol. 273 no. 1603 28 15-2820. Wil tschko, W. and R. Willschko, 2007.Magneto- reception in birds: two receptors for two differen t ta sks. J. OmitilOlugy 148, Sup plem ent 1:61-76 . W il tscllko, R., K. Stapput, 1-1 . Bischof and W.Wiltschko, 2007. Light-depenclent magnetoreception in bird s: inc rcils ing intenSity of monochromatic light changes the nature of the response. Frolltiers ill ZooioSY 2007 4:5. doi: 10.1186/1 742-9994-4-5 W il tschko, R., U. Monro, 1-1 . Ford, K. Stapput and W. Wilt scilko, 2008. Light-dependent magnetoreception: orie n tation behaviour of migratory birds under dim rcd lig ht. Tile /oll(//al or Experilll elltal Biology 2 11 ,33 44-3350. Wiltschko, It and W. Will'schko, 2009. Avian Navigation. Ti,e A Il k 126(4):7 17 -743. For the I .. lllgley i\cadcmr in I ~erks hir e, U.K., Foster + l'arlncH used louvers to control li gh t and ventilation, also ll1aking the building safe fo r birds. Ph o to hy C h ris Phippen orIs 68 Authors and Editors vV ritten by Dr. C hris tine Sh eppa rd, Bird C ollisions Ca m - pai g n Manager and Gle nn Phillips, Bird C o lli s io n s and D cvclopmcnL Orficcr Additio nal c ontributio n s by: An n e Law, and M ike Par r. Edited by: George Fe n wick , C lare N iel se n , M ike Pa rr, Darin Sc h roecl e r, Mary Jane Kaplan , Barbara Ka n ccl baulll Des igned b y: Ge m ma Rad ko Recommended Citation C hristin e Sh e ppard and G le nn Ph illi p s . Bird-Friel/dl)' Buildil1g D es ig ll, 2 nd Ed. (T h e Pl a in s, VA : Am e ri c an Bird Co nserva n cy, 2015) Acknowledgments Ameri can Bird Conservancy (ABC) would lik e to thank the fo llowing fo r th ei r help in bringin g this d oc ument to fruition: Susan Elb in and th e staff of New York C ity Audubon , th e 'W ildlife Co nse rva ti o n SOC iety, Bi rd-sa fe G lass Foundation , Inte rnati o nal Dark Skie s As- sO ciatioll , Fata l Li g ht Awa ren ess Progra m , Jo anna Eckl es, Debo rah Laurel , G u y Maxwe ll , Dan Pi se lli , Br u ce Fowle, M ar c i a Fo w l e, and St efan Knust. VVe are especia lly grateful to th e Leon Levy Fo undatio n fo r its o n going suppo rt fo r J\BC's Collision s Progra m. Th i s document i s based o n gui de lines p ublish ed by th e New Yo rk C it y Audu bon SOC iety , Inc., M ay 2007, wit h su ppor t fr o m the U.S. Fi sh & W ildlife Se r v ice throu gh til e N eotrop ica l Migratory Bird Con se rvati o n Ac t. Disclaimer Tflis fJllblic(ltiUII is presenlcli ill suur / ra illl (lIIll is il/lelltlcd fur SC I/('rol XlIidtil lCe (111 )'. Tile materia l was dra ll'lI (ru m mil//}' so urces; eve ,)' effuf/ lVa s lIU1de to cit/: tllos e so urces, a llrl a llY omiss i ons are il/advert ellt. Tile cu/lle llL s uf tI,iJ fJllufica ticJI/ (1ft ' Iwl il/tcl/d ed (I S pro(css i cJl/nl (/e/ll ice. II HC, /f Ie lttllllOrs, (/1/(/ Nc lI' \fo rk Cit y lI uc/llu ull makc I/O f('Pft'SC l l l tlliulI or lI'armll/)" eit ll er e.ypres.~I)' or implied, (IJ 10 til e CtJIllf,let elle.'i .'i or (l cc /l raq o( tile cO lltellts. Users or OW,'1e s uide/il/es mllst m ake illdefJ CIUlellt dl'l"rmi//rlti{)l/s ti S to 1I1e suita bility or app/ictlbility ufllli' il/for/1/atio/l (o r tlw;r 011'/1 si/I/atio// or PllrlJOscs; ril e in for /1l {/tio /l is 11 01 illlcl/(ic(llo bc {/ slI iJs titute lor specific, t cc/mim/, or pru(ess iUl w/ advice or services. III I/O ('ve nt will (1I f' p!lblis/lets ur (Il/thors be r espu llsi/J /e ur l iabl e (ur c/(I/1I(l.\rt'S of (III>, /llIlII n' ur k ind w /mlSoel'er re.Hllling fru m (/1(' di .~tril)llti(}11 uf, lise of, IJr r eli(l n ce Oil tile C()lIt e tl t.~ o( Illis p"blicatioll . © Ame ri can Bird C o n se rvancy , 20 15 T h e lns tltllt Arabe du j\,·lo ncl e in Paris, Fr ance, provides light to the building interior without using glass. Ph o to by Joseph Ra dko, Jr. I American Bird Conservancy is the Western Hemisphere's bird conservation specialist-the only organization with a single and steadfast commitment to achieving conservation results for native birds and their habi tats throughout the Ameri cas. With a focus on efficiency and working in pa r tnership, we take on the toughest problems f aci n g birds tOday, innovating and building on sound science to halt extinctions, protect habi tats, eliminate threats, and build capacity for bird conservation . I ~-r- (/AMERICAN BIRD CONSERVANCY American Bi rd Conservancy 4249 Loudoun Avenue, P.O . Box 249 The Plains, VA 20198 540-253-5780 abcbirds.or g r r " + ,_ 0 !jjj fm W! HI! u ~ '" 0 fHI co '" Briefly. a bird-friendly bUilding IS one wnere ' At least 90% of the material in the exposed fa<;:ade from ground level to 40 feet (the prima!), bird collisio zone) has a threat score of 30 or less . derived fro controlled experiments At least 60% of material in the exposed fa<;:ade above the collision zone meets the above standard . All glass surrounding atria or courtyards meets the above standard . ;There are no "see through" passageways or corners . Outside lighting is appropriately shielded and directed to minimize attraction to night migrating or nocturnal birds . I terior lighting is turned off at night if not in Lise and designed to minimize light escaping throug windows during night operation. Landsca ing IS designed without features known to increase collisions . Actual bird ortality IS monitored and compensated or (for example . in I he Bu rj Qatar, d c~lgll cd by Jean NOtlvel , was /lamed lIest Tall Hulldlng Worldwide i n 201 2 . The fa ~rl c , cre ated with mulH·lay er ed \crcc ns, cx prcssc\ 10c.11 cultur e w hile p ro\t idlng pr o tecti o n fr o m hi gh U!I1lPl.!taIUt(!S a nd ,and. Ph o to h}' \1ar(' n c ~blJTdcs Printing costs for this publication have been kindly covered by an anonymous donor \ THE CONDOR '" Ornithological Applications RESEARCH ARTICLE www.coope r.org Volume 116, 2014, pp. 8 -23 001: 10.1650/CONDOR-13 -090.1 Bird-building collisions in the United States: Estimates of annual mortality and species vulnerability Scott R. LOS5",a* Tom WiII,2 Sara S. Loss,' and Peter P. Marra' 1 Migratory Bird Center, Sm ith sonian Conse rvatio n Biology In stitute, Nationa l Zoo logical Par k, Was hington, DC. USA 2 U.S. Fish and Wildlife Serv ice, Division o f Migratory Bi rds, Midwest Reg ional Offic e, Bloomington, Minnesota, USA a Cu rr ent address: Department of Natural Resource Ecology & Management, Oklahoma State University, Stillwater, Ok lahoma , USA Ii Co rr es pond ing author: scott.l oss@okstate.edu Received October 9,2013; Accepted October 17, 2013; Publi shed January 2, 20 14 ABSTRACT Building co lli sions, an d partic ula rly co lli sions w ith w ind ows, are a major anthropogenic threat to b ird s, w ith rough est ima tes o f bet wee n 100 mi ll ion and 1 bi ll ion birds ki ll ed annu ally in the United States. However, no current U.s. est ima tes are based on systema tic ana lysis of mul tiple data so urces. We reviewed the published li teratu re and ac quired unpublished data sets to sy stematica ll y quantify bird -b uil d ing co lli sion mortality and sp ecies-spec ifi c vulnerab ili ty. Based on 23 studies, we est im ate that between 365 and 988 million birds (me dian = 599 mi lli on) are kill ed annua ll y b y bui lding collis i ons in the U.S., with roughly 56% of mortality at l ow -ri ses, 44 % at residences, and <1% at high-rises. Based on >92,000 fatality records, and after controll in g for popu lation abundance and range over lap w ith st u dy sites, we identified several species that are disproportionately v uln erab le to co ll isions at all bui lding types. In add ition , severa l species li ste d as nationa l Birds of Conservation Co ncern due to their declining p opulations were identified to be highly vuln erable to build ing coll isions, including Go ld en -winged Warb l er (Verm;vora chrysoptera ), Pa inted Bun ting (Posserino ciris ), Canada Warbler (Cardellina canadensis), Wood Thrush (Hylocichla mustelino), Kentucky Warb ler (Geothlyp;s formosa), and Worm-eating Wa rbl er (Hefmith eros vermivorum). Th e identificati on of these five migratory species with geog rap hi c ranges li mited to eastern an d ce n tra l North America re fl ects seasonal and regional biases in the currently available bu ilding-collisio n data. Most samp ling has occurred during migration and in the eastern U.s. Furt her researc h across seasons and in underrepresented reg ion s i s needed to red uc e t hi s bias. Nonethe less, we prov id e quantitative ev id ence to support t he con clu sion that building coll isions are seco nd on ly to feral and free-ranging pet cats, which are estimated to kil l ro ug hl y four t im es as many birds each year, as the largest source of direct human-caused morta lity for U.s. bird s. Keyw ords: anthropoge nic morta li ty, Bi rds of Co nservat ion Conce rn , individual res iden ce, low-ri se, high-rise, sys tematic revi ew, w indow co lli sio n Colisiones entre aves y edificios en los Estados Unidos: Estimaciones de mortalidad anual y vulnerabilidad de especies RESUMEN Co li sones co n ed ifici os, en particul ar contra ventanas, presentan una amenaza an tropog en ic a importante para las aves, y se estima qu e ca u sa n la muerte de entre 100 m ill an a mil mill ones de aves anu almente. Si n emba rg o, no existen estima cio nes para los Estados Unidos que esten basadas en un ana lisis sistemat ico de datos provenientes de mult ipl es fue ntes. Rev isamos datos pub li ca d os y ta m bien adq ui rimos bases de da tos in ed itos para cua ntificar de un a man era siste mati ca la mortalid ad causada por co lis ones entre aves y ed ificios, y la vulnerabilidad d e d iferentes es peci es. Basado en 23 est udi os, est im am os que entre 365 y 988 millones de aves (promed io = 599 m ill ones) mueren anualmente como consecuencia de co lis iones con edificios en los Estados Uni dos, co n aprox imadamente 56 % de la mortalid ad en edificios de baja altura, 44 % en residencias, y <1% en edifi cios de muchos piso s. Basa d o en >92,000 f ata lidade s registradas, y lu ego do controlar por abund ancia pob laciona l y so lapa miento d e rango ca n area d e es tudio, id entifi camos varias especies que son desproporciona lmente vuln erables a co li siones con t od os los tipos d e edi ficio. Ad emas, varias especies lis ta d as naciona lme nte como Aves de lnteres para la Conservacian debido a sus poblaciones en decliv e fu eron id entificad as como alta men te v uln erabl es a co li siones, incluyendo Vermivora chrysop t era, Passerina ci ris, Cardellina canadensis, Hy/oc ichlo mustelina, Geothlypis formosa, y Hefmitlleros vermivorum. La iden t ifi cac ian de es t as cinco especies migratorias con rangos geograticos res tringido s a Norteamerica orie n ta l y central refle j a sesgos estacionales y regiona l es en la d ispon ibil idad de datos actua les dispon ib les de colisiones con edificios. La mayoria del muestreo ha ocur rid o durante l a epoca de migrac ian yen el este de los Estados Unidos. Hace n falta in vestigaciones adi cion ales a trave s d e estaciones y en regiones poco repre sentada s par reducir este sesgo . Sin emba rgo , prese ntam os © 201 4 Coop er Ornithologi cal Soci ety . ISSN 000 4-8 038, electr on ic ISS N 1938 -5 129 Direct all request s to rep rodu ce journal content to th e Cen tral Orn ith ology Pub licati on Office at aou cospub s@ gmail.co m 73 S. R. Loss, T. Will, S. S. Loss, and P. P. Marra u.s. bird-building co lli sions 9 evidencia cuant itativa que apoya la conclusion que, como causa de mortalidad ligada derectamente a los humanos en los Estados Unidos, las colisiones con edifjcios son superadDs so la mente por los gatos mascotas libres, los cuales matan aproximadamente cuatro veces la cantidad de aves anualmente. PaJabras clave: mortalidad antropogenica, Aves de Imeres para la (onservaci6n, residencia particular, ed ificia de baja altura, edificio de muchos pisos, revision sistematica, colision can ventana INTRODUCTION Co lli sions between birds and man-made structures, including commun ication towe rs, wind turbines, power lines, and buildings, coll ect ivel y result in a tremendous amount of bird morta li ty. Bu ildings are a glo b all y ubiquitous obstacle to avian flight, and collisions with buildin gs , espec ially t heir gla ss windows (F igure I), a re thought to be a major anthropogen ic threat to North American birds (K lem 1990a, 2009, Machtans et al. 2013). Estimates of annua l mortality from building collisions range from 100 million to 1 billion birds in the Uni ted States (Klem 1990a, Dunn 1993) and from 16 to '~2 m illi on bird s in Canada (Mach tans et 31. 2013). Th is magnitude of mortality would place buildings behind only free-rang in g domestic cats among sources of direct hum an-caused mortali ty of birds (B lancher 2013, Loss et al. 2013). Research on bird -building co llisi ons typically occurs at ind ividua l sites with little sy nth es is of data across stud ies. Conclusions about corre la tes of mortality and the total magnitude of mortali ty caused by colli s ion s are therefore spatiall y limited. \Xfithin stud ies, morta lity rates have been found to increase wit h the percentage and s urface area of buildings covered by glass (Col lin s and Horn 2008, Hager et aL 2008, 2013, Klem et aL 2009, Borden et aL 20 10), the presence and height of vegetation (Klem et al. 2009, Borden et a l. 2010), and th e amount of light emitted from FIGURE 1. A Swainson's Thrush killed by colliding with the window of a low-rise office building on the Cleveland State University campus in downtown Cleveland, Ohio . Photo cred it: Scott Los s windows (Evans Ogden 2002, Z ink and Eck les 2010). In the most extens ive building-co lli sio n study to date, per- building mortali ty rates at individual residences were higher in rural than urban areas and at residences with bird feeders than those with out feeders (Bayne et a l. 2012). However, compared with larger buildings in urban areas (e.g., skyscrapers a nd low-rise buildings on office and university campuses), detached residences appear to cause lower overa ll morta lity rates and relative ly high amounts of mortality during non-migratory p eriod s (Klem 1989, Dunn [993, O'Connell 2001, Klem et aL 2009, Borden et aL 2010, Machtans et a l. 20[3). Despite the apparently large magnitude of bird-build ing collision morta li ty and the associated conse rvation threat posed to bird populations , there currently ex ist no U.S. estimates of buildin g-co lli sion mortality that are based on systematic analys is of multiple data sources . The most widely cited estim ate (100 million to 1 billion fata li ties per year) was first presented as a rough figure along with qualificatio ns (K lem 1990a) but is now often cited as fact (Best 2008). Assessment of spec ies-specific vu ln erability to collisions is also critica l for settin g conservation priorities and understanding population impacts; how eve r, exist ing estimates of spec ies vuln erability are limited in spatial scope. In the most systematic U.S. assessment of building collisions to date, species vulnerability was calcu lated using data from only three sites in eastern North Ame ri ca, but vulnera bili ty va l ues from this limited sample were used to conclude that building colli s ions have no impact on bird populations co ntinent-w ide (Arno ld and Zi nk 2011, but see Schaub el aL 20 11, Klem et aL 20 12). We reviewed the publish ed li terature on bird -buildin g co lli sions and also accessed num erous unpublished data- sets from North A meri can bu il ding-collision monitoring programs. \Y/e extracted >92,000 fatality records -by far the largest build i ng co lli s ion dataset co ll ected to date-and (1) systematically quantified total bird collis ion mortality alon g with uncertainty estimates by comb inin g probability dist ributions of mortality rates with est imates of numbers of U.S. buil dings and carcass-detectio n and scavenger- removal rates; (2) generated estimates of mortality for d iffe rent classes of buildings (i ncludin g residences 1-3 stories tall, low-rise non -re s id ential buildin gs and residen- tial buildings 4-11 stor ies tall, and high-rise buildings 2:12 stories ta ll ); (3) cond ucted sensitivity analyses to identify which mode l parameters cont ribu ted t h e greatest uncer- tainty to our estilnates; and (4) quantif1ed s pec ies-specific The Co ndor: Ornithological Appli cations 116:8 -23, © 2014 Cooper Ornithological Society 74 10 U.S. bird -build ing co ll isio ns vu ln erabili ty to co lli s ions across all bu il dings a nd fo r eac h building type. METHODS Literature Search We searc hed Google Sc ho lar and t he Web of Sc ie nce database (us in g t h e Web of Knowl e dge sea rch e ngin e) to lo ca te pee r-r ev iewed publi cations ab out bird -building c o lli s io ns. We used the search term s "bird wi nd ow co lli s ion" a n d "bird building colli s ion" a nd both terms with "bird " re p laced b y "av ian." '1-/e che cked refe r e nce li sts a nd an a nnotated bibliograph y (Seewage n a nd Sheppard 20 12 ) to id entify additional stud ies. Data from co ll is io n -monito ring program s we re lo cated using a Google search with t h e term "window coll isi o n monitoring program" a nd by co ntacti ng progra m coordin ators li s ted o n project websites. \Ve cross-ch ec ke d the datasets we found with a com preh en s iv e li st of "Li gh ts Ouf' prog rams prov ided b y C. S heppard. Add it ion al unpublis h ed datasets were locat ed b ased o n our knowled ge of o ngoin g stud ies prese n ted at profess iona l co n fere n ces or in pu bli s h ed abstracts. Fi n ally , we lea rned of u np ub li sh e d datase ts when co ntac t in g fir st aut h o r s of pub lis h e d stud ies; t h ese additional datase ts we re e ither more extens ive vc rsio n s of auth o rs' publish ed datasets, complete ly n ew datasets, o r in one case, a dataset from an indepe nd e n t ci ti zen scie nti st. Inclusion Criteria and Definition of Fatality Diffe re nt studi es employed d iffe ren t samplin g des ign s a nd data co ll ection protocols. To reduce th is variability, to ensure a base lin e for t h e ri go r of studi es we u sed, and to min imi ze bias in o ur anal yses, we im pl e m ented inclu s io n crit eria to filter data at both t he study and record levels. Inclu s io n c rit eria were d ifferent for t he a na lyses o f tota l mortality and spec ies vulnerabili ty . As a first s tep, we only includ ed s tudi es for in -de p t h r eview if t h e)' we r e conducted in the U.S. or Ca n ada and provid ed o ri gi n al d ata on bird -bu il d in g co lli s io ns. \X/e im p le m ented study- leve l inclu sio n c ri te ria for t h e estimate of tota l m o r ta li ty as fo ll ows . \'{/e ex cluded studies th at were based o n sa mpling at a si ngle structure; th ese studies often fo c u s o nl y o n un ique building types with non-re prese n tative morta li ty ra t es (e.g., museums, convention cen ters, 01' exceptionally ta ll h ig h -ri ses). We included datas ets th a t were ba sed o n systematic carcass surveys or syste m atic surve ys of h ome- ow ne rs, but we excluded t h ose that were ba sed o n sam plin g in response to predicted bui ld ing kill s , incidental o bse rvati o n s, opportuni sticall y sa mpl ed collections, or undoc ume nt ed m etho ds. Because est im ating pe r -buildin g mortality rates was a major co mp o n ent of t he mortality es tim ate, we also excluded stud ies if t h ey d id n ot record numbe rs of buildings monitored or provide s t reet S. R. l oss, T. Will, S. S. Loss, and P. P. Ma rra addresses of buildings th at wo uld have allow ed us to estimate numbers o f buildings. Because th e s pecies vuln era bi li ty analys is wa s ba sed o n cou n t proportions rather t h a n on per-building morta li ty rates, we imple m e n ted a diffe re nt set of inclus ion c ri te ri a th an that used for t h e total mortality es tim ate. T hi s resulted in the use of so mc studi es t hat wer e excl uded from t he to t a l mortality estim a te. S tudi es were o nl y incl uded in the s p ecies ana lysis if th ey id e n t ifi ed carcasses to s pecies. \Ve excluded s tudi es documenting fewer t han 1.00 co lli s io n records b ecau se proportio ns based on s mall sam p les are m o re lik e ly to be abnorm a ll y h igh o r low. As wi th th e to tal mortali ty estimate, we ex clud ed data that were based on incidental or opportunistic samplin g or undocume n ted m e thods . However, we d id include stud ies even if data were based on sam plin g of a s in g le st ru cture o r if we cou ld no t determine the number of bu il din gs sam p led. Thus, we assume that s pecies compos ition with in a site is independent of t h e number of buildings sa mpled. T he s tudy -leve l inclus io n c ri te ri a res ulted in 23 a nd 26 datasets used fo r t h e tota l m o rtality and s p ecies vuln e ra - b ili ty estimates, res pective ly (Ta b le 1), Seve n s tudies we re ex cluded from a ll a n a lyses (Tabl e S I in Supplementa l Mater ial Appendix A). Many datase ts in clude so me co lli s io n reco rds t h at were co ll ected during sta nd a rd ized s urveys and oth ers fo und incidenta ll y. In additio n , definitions of fatalities differ a mong stud ies. \Xle th e re fo re a pp li ed inclu s io n c ri te ri a to filter individual records a nd set our own definition of what co nstitutes a fatali ty. T he record -leve l inclusi o n c r iteria were the same for a ll of our an a lyses. \Y,./e excl uded records clearly d e noted as in cidental finds (i.e. not coll ec ted durin g s urveys), records w ith a dispos ition of "ali ve" o r "sur- vived ," and records of released birds. 'vie a lso excluded records of blood and/o r feat her s pots o n windows w it h no c arcass found. From the remaining records, we defined fata li ties to include a ny record wit h a d is po s ition inclu di ng "dead," "coll ected," or any dispos ition indicatin g seve r e injUl), (e.g., "disabl ed," "squ as h ed ," "fracture," o r "in - jured"). All oth er r eco rd s we re cons id e re d to have un kn own di s position (e.g., "stunn ed," "ex h austed ," "weak," "dis -o ri e nted," o r a n y d is po s iti on indi cating a bird was se n t to re h a bili tatio n ) and we r e excluded from all a n a lyses. The record-leve l c rit e ri a resulted in 92.869 records that we used to ge nera te t o tal morta lity a nd s pecies vu lnerability estim ates. It was not poss ibl e to confirm wheth e r fata li t ies we re caused b y co lli sions with windows or with other non -reflective po rti o n s of build- ings; therefore, for t h e pu rposes of this study, we treated a ll records as bu il di n g-co lli sio n fatalit ies. No n eth ele ss , t h e majority of bird morta li ty at bu il dings lik e ly occu rs due to co ll is ion wit h windows 0 1' oth e r reflective s urfa ces (K lem 2009). The Condor: Orn ithologica l Appl ic ati ons 116:8 -23, ©2014 Coope r Ornithological Society 75 TABLE 1. Sampling coverage, number of buildings sampled, and mortality rates documented in studies meeting inclusion criteria for estimation of total annual U.s. Y' mortality from bird-building collisions and/or calculation of species-specific collision vulnerability. :c r- Used for Used for Mortality per building 0 VI Year-round mortality vulnerability Buildings .Y' Building class Location sampling? estimate? analysis? sampled Average Range Study :-I Residences Alberta Yes Yes Noc 1,747 0.7 0-43 Bayne et al. 2012 ~ (1-3 stories) U.s. & Canada No Yes Yes 1,165 0.85 0-21 Dunn 1993 Y' Duluth, MN No Yes Yes 42 2.3 f ? Bracey 2011 Y' r- Illinois Yes Yes Noc 242 1.5 ? Weiss & Horn 2008 0 VI Carbondale, IL Yes No" Yes h 1 33.0 NA Klem 1979 .Y' QI Purchase, NY Yes Noa Yes h 1 26.0 NA Klem 1979 ::J a. Low-rises Richmond, VA Yes Yes Yes 4 29.0 21-38 O'Connell 2001 ~ Cleveland, OH Yes Yes Yes 18 15.1 Borden et al. 2010 ~ Elsah,IL Yes Yes Yes 4 24.0 Hager et al. 2008 ~ Decatur,IL Yes Yes Yes 11 7.5 f ? Collins & Horn 2008 ~ .... Washington, DC No Yes Yes 21-38e 4.0 1-30 Lights Out DC 2010-2012 @ :r tD Rock Island, IL No Yes NOd 20 2.6 0.3-52.1 Hager et al. 2013 n 0 Decatur,IL No Yes NOd 11 4.8 Horn personal communication ::J a. Murray, KY No Yes NOd 13 1.6 0-7 Somerlot 2003 g Stillwater, OK Yes No" Yes 32.0 NA O'Connell personal communication 0 Rock Island, IL Yes Noil Yes 54.8 NA Hager et al. 2008 3 ;::;' Chicago,IL No Noil Yes 1,028.09 NA McCormick Place 1978-2012 :r Rochester, MN No NOb Yes ? ? ? Project Birdsafe Minnesota 2010-2011 0 5' San Francisco, CA Yes Noil Yes 1 47.29 NA California Academy of Sciences 2008-2012 to n' High-rises Indianapolis, IN No Yes Yes 48 3.3 1-14 Lights Out Indy 2009-2010 ~ » Atlanta, GA No Yes Yes 53 8.4 0-40 Sexton 2006 1:J Calgary, AB No Yes Yes 15-36 5.5 9 1-89 Collister et al. 1996, 1997, Booth & Collister 1998 '2. n' Baltimore, MD No Yes Yes 16-48e 7.19 1-81 Lights Out Baltimore 2008-2012 ~ Twin Cities, MN No Yes Yes 118 3.09 Project Birdsafe Minnesota 2007-2012 0' New York, NY No Yes Yes 17-31e 5.5 9 1-52 Project Safe Flight New York 2009-2011 ::J VI Philadelphia, PA No Yes Yes 10 13.29 ? Pennsylvania Audubon 2008-2011 ..... ..... Columbus, OH No Yes NOd 20e 1.4 0-5 Lights Out Columbus 2012 (J\ co Portland, OR No Yes NOd 21-44 1.09 ? Bird Safe Portland 2009-2011 I N Toronto, ON No Yes Yes 74-194e 17.49 1-535 Fatal Light Awareness Program 2000-2010 ~ @ Winston-Salem, NC No Yes Yes 16 3.69 0-10 Lights Out Winston-Salem 2011-2012 N Toronto, ON No Noa Yes 1 157.0 NA Ranford & Mason 1969 g Chicago,IL No NOb Yes ? ? ? Chicago Bird Collision Monitors 2002-2012 c: ~ n Milwaukee, WI No NOb Yes ? ? ? Wisconsin Night Guardians 2007-2011 ~ 0 Toronto, ON No NOb Yes ? Fatal Light Awareness Prog. 2007, 2011 C'" 0 a '0 New York, NY No NOb Yes Klem 2009 ~ I C'" 0 ., Study excluded from total mortality estimate because sampling conducted at a single building. S, 3 a: ;::;' b Study excluded from total mortality estimate because number of buildings sampled not recorded and no information provided to calculate this number. 3' :r 0 C Study excluded from species estimates because species data not provided. to 5' d Study excluded from species estimates because sample size < 1 00. n to ~ n' e Numberofbuildings is an estimate based on the average of potential minimum and maximum (see text); range indicates year-to-year variation in number of buildings sampled. iii' ~ f Mortality rate is corrected for scavenger removal and searcher detection rates. 0' Vl ::J 0 9 Mortality rate is an average per-building rate across all years of the study/monitoring program. VI n (D' h Study used for species risk assessment for building class but not assessment across all building classes (sample size < 1 00). ~ ..... 0) 12 U.s. bird -building collisions Data Extraction \Ve cla ss ified studies into three building classes thoug ht to cause different morta li ty rates (Mach tans et a l. 20 13) and for which data on the number of U.S. buildin gs is ava il ab le. These classes include residen ces 1-3 stories ta ll (detached houses and mu l ti~un i t residence s; hereafter, "residences"), low-rise non-resident ial buildings and res idential buildings 4-11 s tories ta ll (h ereafter, "Jow-r ises"), and high ~rise bu il dings 2:: 1 2 stor ies ta ll (hereafter, "h igh -rises"). For unpub li s h ed data from downtown areas of major cities, we assumed t h at a ll data came from hi gh-rises because it was not poss ib le to determine building height wit hout vis iting each site . For all other data so urces, we were able to con firm the building type from which data were co ll ected. Pub li shed studies that met our inclus ion criteria e ither reported an annual mortality rate per building (averaged across bu il dings) or presented both the number of dead birds found and the number of buildings sam pled, thus allowin g us to calcu late th is rate. For publis h ed stud ies, we extracted a single annual mortality rate for each study un less t h e study incl uded data from mo re t han one no n- adjacent site, in which case we extracted a separate rate for eac h site (e.g., Klem 1979). For unpublished datasets that included the number of buildings sampled, we always extracted a sing le mortality rate . Th is va lue was ge n erated by fir st calculating a s ingle-year per-building mortality rate (averaged acro ss buil dings) for eac h year of the study and then averag in g these rates across yea rs. In some cases, we determined t hat two or mo re source s presented dup li cate data whe n we observed that the data were collected at the same st ud y sites and during the same range of dates. In th ese instances, we extracted th e data from t he so urce t hat provid ed more detailed met hods or more ex tens ive fatality data, and we ex cluded th e dup li cated data when ext racting from t he oth er source. Data from collision~monitoring programs often inclu de the street address or in tersect ion where a ca r cass was found but not the number of buildings samp led . Sing le buildings can have more than one address, and a sing le address can inclu de more t han one building. In add ition, so me monitoring programs have no sys tematic protocol for recording addresses, resulting in multiple simil ar e n tries for an address (e.g., 1 Main , 1 Main St., and 1 Ma in-Smith Tower). To account for t hese issues, we entered addresses into Google Maps and used sate lli te view to determine if addresses referr ed to one or more buildings. If it was still unclear from mapping wheth e r an address referred to one or more bUild in gs, we ass umed it referred to one. Li kewise if we cou ld not confirm that two or more s imilar addresses referred to one building, we assumed they were separate bu il d in gs. If addresses with different card in al directions were possible (e.g., 1 Main E and 1 Main W), we assumed t hey referred to separate buildin gs , b ut if t hey were not possible (i.e. on ly I Ma in S. R. loss, T. Will, S. S. Loss, and P. P. Marra ex ists). we assumed data entry error and com bined addresses. Recognizing that these methods could not account for all duplicate addresses and data e ntry errors, we estim ated a min i mum a nd maximum num ber of buildings sampled in eac h yea r. \Y/e estimated a maximum number based on the number of unique addresses remaining after fo llo wing th e ab ove ste ps and t h e assumption t hat in te rsections referred to a number of buildings equal to the number of ca rcasses found up to four (i.e. four or more carcasses may result from colli s ion with four se parate bu il dings, one at eac h in tersectio n corner). 'vie estimated a m ini mum number by comb in in g similar addresses t hat may ha ve been from one bu il ding, even if we could not confirm th is with mapp ing, and assuming that all intersections referred to one building. Vie used th e average of t he m inimum and maximum number to estimate per -bu il ding mortali ty rates. Quantification of Annual Mortality from Building Collisions The studies we used cover vary in g portion s of the yea r, but most focus a ll or most of sampli n g effort on migration periods. Using raw per ~building mortality rates wou ld therefore result in a national estimate that is only rel eva n t to sprin g and fa ll m igration periods. \Xfe so ught to account for partial-year samp lin g and to generate estimates that reflected the ent ire yeal~ because seve ral stud ie s have indicated that building co lli s ion mortality can be substan~ tial during summe r and winter (Dunn 1993, Klem 2009, Bayne et al. 2012, Hager et al. 2013). G iven enough year- rou nd studies, partial -yea r m ortality rates can b e stan- dardized to year-round estimates us in g year~ro und stud ies as a baseline (Longcore et a l. 20 12, Lo ss et a l. 2013). However, there we re few year -round studies that met inclusion cr it eria (Tab le I), so we co uld not adjust in d ividual studie s to year -round estimates. In stead, we accounted for this li m itation in our estimation model (deta ils be low) by only us in g a year~ro und study for residences, repeatin g estim atio n using a su bset of stud ies that sampl ed yea r-round fo r low -r ises, or incorporating a co rrection factor to account for mortality during periods other t han migration for high -ri ses , a buil ding ty pe for which little data exists for Slimmer and winter (see definition of and ratio n ale for t his cor rection factor in Suppl e me ntal M a te rial Appendi x B). Despite the limitation of appl ying a post hoc cor rection factor to the high-r ise estim ate. we argue that this approach is preferable to as suming that no mortality occurs during the summer and winter. We estimated mortality in eac h building class by multiplying data ~derived probab ili ty distributions of per~ bui lding mortal ity rat es by distributions of numbers of bui ldi ngs . For residences , we fo ll owed Machtans et a l. The Condor: Ornithological Applications 116:8-23, © 201 4 Cooper Ornithological Society 77 S. R. loss, T. Will , S. S. loss, and P. P. Marra (201.3), which based mortality rates on the only year ~round building collision survey to date that sampled across a large number of residences, a study of 1,458 A lberta res idents in s ingle and multi-unit res idences (Bayn e e t al. 20 12). This study documented hi gher mortality rates at rural reside nces compare d with urban residences and at residences with b ird feeders compared with those without feeders. The s tudy also d ocumented increas ing mortality with increasing age of urban residences. We incorporated these el e ments into our residence sub ~modcl: Mortali ty rl,lr.ll with feeder(M RF ) = Nre slden ce X R X F X Krl,lr a) with fee der X Dresi dcncc Mortal ity rural no ('('dl'r(M RNF ) = Nfl'sid l'ncl' X R X (1 -F ) X KruTa ) no fcede r X Dr l'sidcnc(' MortalitYurb an with fecdc r (M UF ) = N rcsidcncc{age ) X (1 -R) X F X Kurb :m with fee dcr {agc ) X Dr csi dcncc Mortality urban no feC'dc r (M U:-1F) = N rcsidc ncc(ag c) X (l -R) X (l -F ) X Kurban no rl't'der (agc ) X D rcsi dencc Mortality rcs idcn ces(M It) = M RF + MR NF + J\1 UF + M UNF (I ) (2) (3) (4) (5) wh e re N is the number of residences in th e U.S., R is the percentage of residences in rural areas, F is the p e rcentage of residences with bird feeders, K is the annua l per ~ building mortality rate, and D is a correction factor to account for two biases that le ad to underestimation of mortality (Hager et al. 201 3 ): removal o f carcasses by scavengers prior to fatality surveys and imperfect d e tec tion of the carcass es remain ing at the time of surveys. For Equations (3 ) and (4), we calculated mortality by bui lding age classes (0 -8, 9-18, and 19 -28 ye ars, and a ll ages 2:29 years ), and summed estimate s across age classes. Th es e age classes c orrespond close ly to those in Machtan s e t a l. (2013 ), but we s hifted classe s sl ightl y (e .g ., 9-18 ye ar s instead of 10-20 years) to match housing age data from the U.S. Census Bureau. For low-rises, we generated two s eparate e stimates of co lli sion mortality, one us ing mortality ra tes ba s ed on a ll e ight studies meeting our inclusion criteria and one ba sed only on four year-round studies. \X/e used t h e following sub-model for both est imates: u.s. bird-build i ng collis ions 13 For high-rises , there are no datasets based on year-round systematic s ampling. '\</e incorporated a correction factor (Y) into t h e morta li ty e s timation sub ~mode l to account for additional fatalities occurring outside of migration periods: MortalitYhigh _risc(M H ) = N hlgh_ri sc X f(high -r isc X Y X Dh igh _ri sc (7) 'v./e estimate d to t a l annual build in g co lli sion mortality b y summing estimates for individua l bui lding classes; we conducted estimation twice, once using each of the low - rise e s timates : Morta litY tot:l) = /vIn + MI. + MH (8) All of t h e above parameters were treated as probability di s tribut ions . Fro m the probability distribution of each parameter (se e Tab le 2 fo r specific distributions, Supple- mental Material Appendix B for rationale for all d is tribu- tions, and Table S2 in Supplemental Material Appe ndix C fo r numbers of bu il dings), we random ly drew one value and used the above formulas . We use d ··runiC" and ··rnbino m·· commands (for uniform and n e gative b inom ial distributions, respective ly ) in Program R and c onducted 10,000 iterations to generate a range of estimate uncer~ tainty. Sen sit iv ity Anal y sis \X /e u sed multip le linear regression anal yses a ssuming a norma l erro r distribution (fun c t ion ·'Im·· in Program R) to inves t igate t h e pe rcentage of uncerta inty in mortality e s timate ranges explained by each model parameter (Blan c her 2013 , Lo ss et aJ. 2013). \XIe treated t he 10,000 morta li ty-estimate replicates as the values of the depen~ dent var iab le and random ly drawn value s of each parameter a s values of predi c tor variab les. W/e used partia l R2 values to interpret the pe rcentage of variance in the es timate range explain e d by each parameter. \X/e repeated t h is re gres s ion ana lys is four tim es : o n ce for the tota l mortality e s t imate (including all paramete rs ) and once for each o f the three building class estimates (with each re gress ion mode l only including the parameters relevant to that building cla ss). Quan t ificat ion of Speci es Vulnera bility In addition to estimating total annllal mortality, we calculate d vulnerab il ity for species and taxonomic groups. \XIe fo ll owed Arnold and Z ink (201l), wh o identified "super-collide!"·' a nd "super-avoider·· species using colli - sion re cords from t h ree unpublishe d datasets. \'\Ie gre atly expanded upon th e earlier study by using 26 datasets from across N orth Am e rica (Table 1). All analyses desc ribed be low were conducted acros s all datasets to estimate Morta li tYlo w_r,s('(M d = N low-ri se X Klow-rise X D lo\.,'-ri5c (6) overall building collision vulnerability, as well as separately The Co ndor: Ornitholog ica l Applications 116:8-23, © 2014 Cooper Ornithological Society 78 14 U.S. bi rd -building colli si ons S. R. Loss, T. Will. S. S. Loss , and P. P. Marra TABLE 2. Probability distribut ions used to estima t e total a nnu al U.s. mo rtal ity from bird -building co ll isions. We defined unifo rm distributions for most parameters because not enough data exist to ascribe higher probability to particula r val u es in the d efi ned range. We defined n egative bin om ial distributions for the low-rise and high -rise morta lity ra te di s tribu tions because they a llo wed the majority of probability density to match the confidence intervals indicated by the data w h ile al so allowing for a small probability of higher collisio n morta lity rates, refi ecting the exceptionally high mortality rates t h at have been documented at so me low-rises and high-rises (see mortality rate s in Table 1). Dist ributi on Parameter type Reside nces (1-3 stori es) Numbe r of residences Uniform Pe rce ntage in urban areas Uniform Pe rc e n ta g e with bird feeders Uni form Mortality rat e Rural wi th feeders (all ages) Unifor m Rural without feeders (all ages) Unifo rm Ur ban with feeders Age 0-8 Unifo rm Age 9-18 Uniform Age 19 -28 Uniform Age 29 + Uniform Rural withou t feede rs Age 0-8 Uniform Age 9-18 Unif orm Age 19 -28 Unifor m Age 29 T Un iform Scavenging/detectability correction Un ifo rm Low-rises Number of low-rises Unifo rm Morta lity rat e (a ll studies) Neg. bin. Mortality rate (year-round studies) Neg. bin. Scavenging/detectability co rre ctio n Unifor m Hi g h-rises Number of high-rises Un iform Mortality rat e Neg. bin . Partial -year sa mpling correction Uniform Scavenging /detectability co rr ection Un iform Distribution parameters Varies by age (Supplemental Mate rial Appendix C) Min = 72.6 %; Max = 88.8 % Min = 15 %; Max = 25% Min = 2.17; Min = 4.03 Min = 0.98; Max = 1.82 Min = 0.28; Max = 0.52 Min = 0.42; Max = 0.78 Min = 0.56; Max = 1.04 Min = 0.63; Max = 1.17 Min = 0.11; Max = 0.20 Min = 0.18; Max = 0.33 Min = 0.25; Max = 0.46 Min = 0.28; Max = 0.52 Min = 2; Ma x = 4 Min = 14 .0 million; Max = 16.2 million n = 4.6; p = 0.35 n =5.1;p =0.26 Min = 1.28; Max = 2.56 Min = 19.854; Max = 21.944 n = 4.0; p = 0.37 Min = 1.05; Ma x = 1.20 Min = 1.37 ; Max = 5.19 Source U.s. Census Bureau 2011 U.s. Census Bureau 2012 Dunn 1993 Bayne et al. 2012, Machtans e t al. 2013 Bayne et al. 2012, Machtans et al. 2013 Bayne et al. 2012 , Machtans et al. 2013 Bayne et al. 20 12, Machtans et al. 2013 Bayne et al. 2012, Machtans e t al. 2013 Bayne et al. 2012, Mach tans et al. 2013 Bayne et al. 2012, Machtans et al. 20 13 Bayne et a l. 2012, Machtans et al. 2013 Bayn e et al. 2012 , Machtans et al. 2013 Bayn e et al. 2012, Machtans et a1. 2013 Dunn 1993 Mu ltiple sou rces (see Supplementa l Mate rial Appendix C) 95% of di stri bution prob. density = 4_18 :1 95 % of distribution prob. d ens ity = 5-28b Hage r et at 20 12. 2013 Sky Scraper Source Media 2013 70% of di st ribu tion prob. density = 4-11 b Additional 5-20% mortality outside of migration Ward et al. 2006, Hager 2012. 2013 iI Range represents 95 % confi d ence in terval of mortality rate s calculated across all eight studies of low-rises meeting inclusion criteria. b Rang e re presents 95% confidence in terva l of mortality rates ca lculat ed from four yea r-round s tudi es of low-rises meeti ng inclusion criter ia. C Range represent s 95 % confidence int e rval of mortality rates calculated from 11 studies o f tall buildings meeting inclusion criteri a. for each buildin g cl ass to est im ate class-specifi c vulnera - bility. As described previously, we o nly include d datasets with more tha n 100 records fo r t h e overall vu lnera bil ity analysis. However, b ecau se t h e r e were o nl y two datase ts for res id ences t h at had more than 100 records, we also included two sma lle r datasets to calculate co lli s io n v ulne rabi li ty fo r t hi s building class. Numb ers of fata li t ies can va,y a mong s pecies due to population abundance and the degree of ran ge ove rl a p w ith study locat ions (A rnold and Zink 2011). To account for p o pu lation a bundance, we extracted national popula- tion s ize estimates from t h e Partners in Fli g ht Po p u lat ion Est imates Database (Ri ch et a1. 2004), w h ich includes North A merican popu lation estimates gen erated using U.S. Breeding Bird Survey data (Sa uer et a l. 2012). We u sed North American abunda n ce rath e r than regional abundance becau se it is difficu lt to link stu dy s it es w h ere morta li ty occurs to the affected regional subsets of bird po pulations, especia ll y for s pecies t h at a r e ki ll ed primaril y d urin g mi grat io n (Lo ss et aL 2012). To a ccoun t fo r range overlap with s t udy s ites, we cou nted the nu m b er of s ites overlapping w it h eac h s pec ie s' breed in g, wintering, a nd/or m ig ration range (S ibley 2000). We followed A rnol d and Z in k's (2 011 ) approach for ca lcul atin g species vu ln e rabil- ity. To g ive each s ite eq u al weight ing, we first sta ndard- ized each dataset to 36,000, t h e largest s ingle-site tota l Th e Condor; Ornithologica l Applications 116:8 -23, © 2014 Cooper Ornithological Socie ty 79 S. R. Loss, T. Will, S. S. Loss, and P. P. Marra Residences 1 -3 stories 600 A 500 M 400 300 200 C1> 100 -IU 0 E :;: 100 200 300 400 500 rJ) C1> Mortality estimate (millio ns ) ..... 0 >. () c High-rises C1> ::J C" C C1> 600 .. U- 500 400 300 200 100 0 0 500 1000 2000 Mortality estimate (thousands) 600 500 400 300 200 100 0 500 400 300 200 100 o u.s. bird-building collisions 15 Low-rises B 0 200 600 1000 Mortality est imate (millions) Tota l wi n dow -collision morta lity 200 600 1000 Mortality estimate (millions) FIGURE 2. Frequency histograms for est imates of annua l u.s . bird mortality caused by collisions with (A) residences 1-3 stories tall , (8 ) low-rises (residences 4-11 stories tall and all non-residential buildings :;11 sto ries tall ), (C) high-rises (a ll buildings ~ 12 sto ries tall), and (0 ) all buildings. Estimates for low-rise s and for all buildings are based on the average of two estimates: one calc ulated with all eight low-rise studies meeting inclusion criteria and one calculated with a subset of four low-rise studies that conducted year- round sampling. number of fata li t ies, and then summed stan da rdized counts across studies for each species. \Y/e regressed logJO (X+l ) species cou nts (X + 1 transfo rm ation to account for zero counts for some species at some s it es) on 10glO popu lat io n s ize and IOglO ra nge overlap. Vu lnerability was estimated by fixing coefficients fo r popu lat ion size and range overlap to 1.0 (t hi s assumes that, for examp le. a LO -fold in crease in abundan ce is associated with a lO -fold incr ease in co lli s ion mortality, all else being equa l; A rn old and Zi nk 20 11). calc u lating re sidual s, and raising 10 to the power of the absolute value of residuals. This approach of fixing model coefficients was taken beca use there wa s an unknown leve l of e rror in both th e dependent and ind ependent variables and. therefore. standard regress ion mod e ls cou ld no t pr oduce unbiase d s lope est imates (\X1arto n et al. 2006. Arnold and Z i nk 2011 ). Calcu la ted vu ln erabi li ty va lues indicate the factor by which a species has a greater chance (posit iv e residua ls) or s mall e r c han ce (negative residuals) of experie ncing building co lli sion mortality compared with a species with average vuln e rability. \X'e estimated vu lnerability for taxonomic groups by averag- in g re s idua ls across species occu rring in at least two st udie s. RESULTS Estimates of Bird-Building Collision Mortality The 95% confidence interval of annua l bird mortality at reside nces was estimated to be between 159 a nd 378 million (med ian = 253 milli on) (Figure 2A and Table 3) after correcting for scave nge r re moval and imp erfect detectio n. This equ ates to a median annua l mortality rate of 2.1 bird s per building (95 % CI = 1.3-3.1). Reflecti ng the large number of re sidences in urban a reas and residences without bird feeders, we estimate that urban residences without feeders cumu lative ly account for 33 % of mortality at re sidences. followed by rural residences without feeders (3 1 %). urban re s id e nces wit h feeders (19%). and rura l residences wit h feed ers (17%). Th e Condor: Ornithol ogical Applications 116:8-23. ©2014 Coop er Ornithologica l Society 80 16 U.S. b ird-building collisions s. R. Loss, T. Will, S. S. Loss, and P. P. Marra TABLE 3. Es timates of annual bird mo rta lity ca u sed by building collisions at U.s buildi ngs. Fo r low-ri ses (a nd t herefore, fo r t he t otal m ort ality est i ma te), we generated two separate estimates of coll ision mortality, one usi ng mo rt alit y rates based on all eig h t low-rise studies meeti n g our inclus i on criteria and one based on a subset of four low-rise stu d ies that sa m p l ed mortality yea r-round. Poin t estimate 95% (I Building class Mean no. of bu i ld i ngs in U.S. Total Per build in g To t al Per build i ng Reside nces (1-3 stories) Low-rises 122.9 million 1 S.l million 253.2 mi llion 245.5 million3 409.4 mi ll ion b 2.1 16.3 3 27 .1 b 24.3 3.7 01 15 9.1-378.1 million 62.2 -664.4 millioni! 114 .7-1,028.6 mililo nb 1.3 -3.1 4.1 -44 .001 7.6 -68.1 b 5.0 -76.6 2.0 -6.801 2.5-9.4 b 20.900 508.000 Hi gh-ri ses Total . 138.0 mi l lion 507.6 millioni! 667.1 mlil ion b 4.8 b 104,000-1.6 mi llion 280.6-933.6 mi l lion 3 349.9-1,296 m illio n b 3 Estima t e based on low-rise estimate u sing al l eight studies meeting i nclusion criteria. b Estimate b ased on low-rise estimate u si n g subset of fou r yeaHound studies meet ing inclusion crite ria . T he 95% co n fidence interva l of annua l low-rise mortal- ity base d on all st u d ies meetin g in clus ion criteria was estimated to be between 62 and 664 mi ll ion b irds (median = 246 m il lion). The 95% confidence interva l based o n t he fou r year -r ou n d low-ri se stud ies was estim ate d to be betwee n U 5 millio n and 1.0 bi lli on bi rds (median = 409 m illi o n ). T h e ave rage of t he two medi a n fig ures is 339 milli o n (9 5% C I = 136-715 mill ion) (F igure 2B ). equating to a med ian a nnual rate of 2 1.7 birds per build in g (95% C I = 5.9 -55). T he 95 % confidence interval of high-rise mortality was estimated to be betwee n 104,000 and 1.6 m ill io n birds (med ian = 508.000) (Tab le 3 and Figure 2C) after correct in g fo r scavenge r removal, im pe r fect carcass de tect ion, a n d mo rta li ty d ur ing per iods other t h an m igrat io n . Despite caus ing the lowest tota l morta li ty, hi gh-ri ses had t he hi g hest med ian a n n ual morta li ty ra te: 24.3 b irds per bu il d ing (95% CI = 5-76). Combini n g est im at es from a ll b u il d in g cl asses (usi ng th e average of th e two low-ri se estimates) res ul ts in an estimate of 599 m illi on b irds kill ed an nua ll y across all U.S. buildi n gs (95 % c.1. = 365-988 mill io n) (F igure 2D). Factors Explaining Estimate Uncertainty Due to t he la rge n u m be r of low-rises and unce r ta inty about low-r ise mo rta li ty rates, se ns iti vity a n alyses i nd icat- ed t h at t he low-rise mo rta li ty rate ex pl ai ned a large amount of uncerta in ty for th e estimates of both low -ri se mortali ty (85%) and tota l mortality (75%). Other pa r am- eters ex pl a ining substa nt ial uncertainty for t he total est i mate incl uded t he correctio n factors for scavenge r remova l and car cass detec ti on at low -rises (10%) and res idences (9 %). For res idences, 70% of uncertai n ty was ex pl ain ed b y t h e co rr ectio n facto r fo r scave ng in g a n d detectio n and 15 % was exp la in ed by t he proportion of reside nces in u rban ar eas. For the h ig h -rise es ti mate, th e greatest u ncertai n ty was exp lained by the morta lity rate (67 %), followed by the correction facto r for scavengin g and detectio n (25%). Species Vulnerability to Building Collisions Of 92,869 reco rds used for an alys is, t h e spec ies most commonly repo rted as buil d in g kill s (collectively rep re- se n t in g 35% of all r eco rd s) we re W hi te-t h roa ted Sparrow (Zol1ot ric!Jia a/hicol/is). Da r k-eyed Junco (JUlICO hyema/is). Ovenbird (Seiunts aurocapilla), and Song Spa rr ow (/\1elo- spiza melodia). Howeve r. as expected. t here was a hi ghl y sign ificant correlat ion b etween fatality counts and popu - lat io n size (r = 0 .53. P < O.OO L df =2 13) a nd betwee n counts and range overlap wit h study sites (r = 0.25, P < 0.00 1. df = 223). Afte r accou n t in g fo r t hese factors. est im ated vu ln erabi li ty ac ross a ll b uild in gs was h ig h ly variable, ranging from 1,066 t imes more like ly to co ll ide t han ave rage to 273 t im es less lik ely to colli de t ha n ave rag e (hi gh vu l nerab ili ty species in Table 4; all values in Tables S3 -S6 in Suppl emental Material Appendix D). Severa l species ex h ibi t d isprop ort io nately h igh vu lner- ab ility to coll isions regard less of bu il di ng type, incl udi n g Ruby-t hroated Hummin g b ird (Arc hilochus co l ubris), Brown C reeper (Certhia americana), Ovenb ir d. Ye ll ow- be lli ed Sapsucker (Sphyrap;cus var;us). Gray Ca t b ird (Dumelella carolinensis), and Black-an d -w hi te \"Va rble r (Mniotilta varia). Seven spec ies that a re d isp ro po r t ion ately vu ln erab le to buil d ing colli sio n s a re natio n al Birds of Conservat io n Concern and 10 are listed reg iona ll y (Table 4; U.S. Fi s h an d W il d li fe Service 2008). Spec ies in t he former group include Golden-winged \Var bl er (\IenJlivorn chrysoplern) and Ca nada War b ler (Ca.rdellina canadel1s is) at low-rises. hi gh -ri ses. a nd overa ll , Pa inted Bu nting (J>asserilla ciris ) at low-rises and ove rall , Kentucky War bl er (Geothlypis formosa) at low-rises a n d hi gh -ri ses. Wo rm - eating \Varbler (Helmitheros vermivorum) at hi g h-rises. a nd \Xfood T h rus h (Hylocichla musleLina) at res ide nces. For species with vu ln era bili ty indices ca lc ulated fro m a The Condor: Ornithological Applications 116:8-23, © 2014 Cooper Ornithological Society 81 TABLE 4. Estim ates of species vuln erability to building co lli sions. Ri sk val ues indicate the factor by wh ich species are at a greate r risk of co lli sion com pared with a species with ave rag e risk. Species in boldfa ce ita lic s are Birds of Conserva ti on Co ncern at the nationa l leve l and spec ies in boldface are Bi rd s of Co nse rvation Concern in at lea st one U.S. region (U.S. Fi sh and Wildlife Service 2008). Scie ntifi c names are in Supplem ental Material Appendix D. Al l buildings Resi dences (1 -3 stories ) l ow-rises Hi gh-rises --< "" Species Risk Species Risk Species Ri sk Species Ri sk " () 0 ~ C- O " 0 2. 5- Anna's Hummingbi rd>! 1,066,4 Purple Finch 25 7,2 Go/den-winged Warbler 14 1.7 Townsend's So lit aire 167,4 Black-throated Blue Warbler 45.5 Ruby -th roated Hummingbird 174.7 Pointed Bunting 129.3 Black-throated Blue Warbler 78.5 Ruby-throa ted Hummingbird 37.0 Ovenbird 11 2, I Ruby-th roa ted Hummingbird 103,7 Connecticut Warbler 52,0 Townsend 's Solitaire 36.3 Brown Creeper 81.1 Black-th roated Blue Warbler 86.4 Brown Cre eper 44.3 Golden -winged Warbler 35.3 House Finch 80,1 Swamp Sparrow 50.6 Ovenbird 43.7 Painted Bunting 32 ,1 Bl ack-and-white Warbler 68,7 Canada Warbler 46.7 Ruby-throated Hummingbird 43.4 0 " '" ii' ~ Brown Creeper 26,2 Cedar Waxwing 50.5 louisiana Waterthrush 46.4 Worm -e ating Wa rbler 26.5 Connecticut Warbler 22,9 Fi e ld Sparrow 48.3 Brown Creeper 44.8 Canada Warbler 25.8 Ovenbird 21.8 Wood Thru s h 4 1.0 Yellow-b e llied Sa p sucker 38.3 Gra y Catbi rd 23,9 ,. " "2. Canada Warbler 17 ,9 Swainson 's Thrush 34 .7 Connecticut Warbler 35.7 Yellow-bellied Sapsucker 23.7 Swamp Sparrow 16.7 Northern Ca rdinal 27.5 Ovenbird 30.4 Golde n -winged Warbler 23.1 ii' ~ 5' Yellow-be llied Sap sucker 16,2 Blue Jay 26.5 Sharp -shinned Ha wk 27,8 Ame ri can Woodcock 22,1 louisi a na Waterthrush 14.3 W hite-breasted Nuthatch 25 .0 Ro se -b reas ted Gros b eak 24,1 Co mmon Yellowthroat 20.4 ~ ~ Gray Catbird 12,8 Yellow-bellied Sapsucke r 22,6 Gray Ca tbird 23,2 Sca rl et Tanag er 18.5 Pin e Grosbeak,} 12.4 Northe rn Waterthrush 22.5 Black -and-whi te Warbler 22,7 Bla ck-and-whit e Warble r 18.3 '" c;, , " ," American Woodcock 11.7 Nashville Warbler 22.2 Amer ican Woodcock 21.1 Swamp Sparrow 18,1 Pygmy Nuthatch '" 11.4 Gr ay Catbird 20,7 Kentucky Warbler 20,2 Ro se -breaste d Grosbeak 16,2 Bla ck-and-white Warbler 11.1 Northern Flicke r 20,2 Mourning Warb ler 19.3 Kentucky Warble r 14 .0 @ Pi ed-bill e d Grebe,} 11.0 D owny Woodpecker 18.7 Com mon Ye ll owthroat 18.4 Northern Goshawk 13,6 " ~ Common Ye ll owthroat 10.9 Bl ack-capped Chickadee 14,9 Cape May Warbler 16.7 Eastern Whip-poor-will 13.4 '" () 0 D Species is ranked fo r a ll bui ldings but not ind ivi dual classes because it occ urs in ~2 total studies, but <2 st udies with in bu ildi ng class. 0 " !'! 0 2. 5- 0 " '" ii' ~ V> 0 c, " .:< co N 18 U.S. bird-building collisions relatively small sample of studies (e.g., those noted with a superscript in Table 4), vulnerability indices may be biased. For example, the exceptionally high vulnerability value for Anna's Hummingbird (Caiypte anna) likely results from this species occurring in only two studies and experiencing exceptionally high mortality in one of these studies. Vulnerability estimates for taxonomic groups are in Table 5. Several high-risk bird groups are represented in our dataset by only one or two species (e.g., grebes, shorebirds, kingfishers, and gulls and terns); average risk values for these groups may not represent the entire taxonomic family. Other taxa, particularly the hummingbirds and swifts and the warblers, appear especially vulnerable to building collisions, with more than one species ranking in the overall high-vulnerability list. In particular, warblers experience disproportionately high collision risk, with 10 species ranking among the 25 most vulnerable species overall and 12 and 14 species ranking among the 25 most vulnerable species for low-rises and high-rises, respectively. Taxonomic groups with particularly low collision risk include ducks and geese, swallows, herons, upland game birds, and blackbirds, meadowlarks, and orioles. DISCUSSION Comparison of Mortality Estimate to Previous Estimates Our estimate of 365-988 million birds killed annually by building collisions is within the often-cited range of 100 million to 1 billion (Klem 1990a). Other estimates are either outdated (3.5 million, Banks 1979) or are simply a mid-point of the above range (550 million, Erickson et al. 2005). Our larger estimate of low-rise mortality based only on year-round studies suggests that total annual building collision mortality could exceed one billion birds, as suggested by Klem (2009). Using the year-round low-rise estimate results in an annual mortality estimate of up to 1.3 billion birds. Regardless of which figure is interpreted, our results support the conclusion that building collision mortality is one of the top sources of direct anthropogenic mortality of birds in the U.S. Among other national estimates that are data-driven and systematically derived, only predation by free-ranging domestic cats is estimated to cause a greater amount of mortality (Loss et al. 2013). A similar ranking has been made for anthropogenic threats in Canada (Blancher et al. 2013, Machtans et al. 2013). Major sources of direct anthropogenic bird mortality currently lacking systematically derived estimates include collisions with automobiles and other vehicles, collisions and electrocution at power lines, and poisoning caused by agricultural chemicals, lead, and other toxins. Additional systematic quantification of mortality is needed to allow rigorous comparisons among all mortality sources. S. R. loss, T. Will, S. S. loss, and P. P. Marra TABLE s. Average vulnerability of bird groups to building collisions across all building types. Risk values indicate the factor by which a species has a greater chance (for positive residuals) or a smaller chance (for negative residuals) of mortality compared with a species with average risk. Group Residual Risk Hummingbirds and swifts Grebes Shorebirds Kingfishers a Waxwings Warblers Gulls and ternsa Nuthatches, tits, and creeper Cuckoos Mimic thrushes Diurnal raptors Cardinaline finches Kinglets Thrushes Cardueline finches Night jars Woodpeckers Owls Doves and pigeons Sparrows House Sparrowa Wrens Coots and rails Flycatchers Vireos Starling a Corvids Blackbirds, meadowlarks, and orioles Upland game birds Herons Swallows Ducks and geese Gnatcatchersa a Values based on data from a single species. 1.52 1.04 0.68 0.56 0.55 0.54 0.52 0.50 0.46 0.41 0.40 0.36 0.36 0.25 0.23 0.16 0.15 0.10 0.08 0.08 -0.15 -0.20 -0.24 -0.41 -0.55 -0.56 -0.61 -0.64 -0.77 -1.05 -1.07 -1.25 -1.68 33.2 11.0 4.7 3.6 3.6 3.4 3.3 3.1 2.9 2.6 2.5 2.3 2.3 1.8 1.7 1.4 1.4 1.3 1.2 1.2 1.4 1.6 1.7 2.6 3.6 3.6 4.1 4.4 5.9 11.3 11.6 17.9 48.1 A general pattern across and within building classes is that a large proportion of all mortality occurs at structures that kill small numbers of birds on a per-building basis but collectively constitute a high percentage of all buildings (e.g., residences compared to low-rises and high-rises; urban compared to rural residences; residences without feeders compared to those with feeders). This finding suggests that achieving a large overall reduction in mortality will require mitigation measures to be applied across a large number of structures (e.g., urban residenc- es). Our conclusion about the relative importance of residences for causing U.S. mortality is similar to that made for Canada by Machtans et al. (2013). This similarity arises because residences are estimated to comprise a similar proportion of all buildings in both countries (87.5% in the U.S and 95.3% in Canada). Even assuming the low- end mortality estimate for residences (159 million), total The Condor: Ornithological Applications 116:8-23, © 2014 Cooper Ornithological Society 83 S. R. Loss, T. Will, S. S. Loss, and P. P. Marra mortality at high-rises would have to be 100 times greater than our high-end estimate for that building class (1.6 million) for the two building classes to cause equivalent mortality. On a per-building basis, if each residence killed one bird per year, each high-rise would have to kill >5,800 birds per year to cause equivalent mortality. No evidence exists that high-rises kill this large number of birds. The species composition of window collision mortality also differs by building class. While the high risk group for individual residences includes several non-migratory resident species-including Downy Woodpecker (Picoides pubescens), Black-capped Chickadee (Poecile atricapillus), and Northern Cardinal (Cardinalis cardinalis)-nearly all high-risk species for low-rise and high-rise buildings are migratory. Compared with resident species, migratory species traverse longer distances, use a greater diversity of habitat types, and encounter more building types and total buildings during the annual cycle. Additionally, migratory species are attracted to large lighted buildings during their nocturnal migration; this attraction causes a large amount of mortality at low-rises and high-rises as birds either immediately collide with lighted buildings or become entrapped before later dying of collision or exhaustion (Evans Ogden 1996). The greater representa- tion of resident species in the high-risk group for residences may be due to the propensity for many of these species to congregate at bird feeders, a behavior that may place them at a greater risk of colliding with windows (Dunn 1993, Klem et al. 2004, Bayne et al. 2012). Despite the critical importance of reducing mortality at residences, mitigation measures targeted at a relatively small number of buildings with high per-building mortal- ity rates (e.g., some high-rises and low-rises) will likely result in large per-building reductions in mortality and therefore may represent a cost-efficient starting point for reducing mortality. The mortality proportions that we attribute to different residence types are similar to those estimated by Machtans et al. (2013). This result arises from both the previous study and ours basing analysis on Bayne et al. (2012), a Canadian study that provides a reasonable approximation of u.s. mortality rates as evidenced by rates documented in U.S. studies (Dunn 1993, Weiss and Horn 2008, Bracey 2011). Species Vulnerability to Building Collisions Our vulnerability analysis indicates that several species experience a disproportionately high risk of building collision mortality. Of particular concern within the list of high-risk species (Table 4) are those identified as national Birds of Conservation Concern (species likely to become candidates for listing under the u.s. Endangered Species Act without further action based on population trends, threats to populations, distribution, abundance, and relative density; u.s. Fish and Wildlife Service 2008). U.S. bird-building collisions 19 For species that are vulnerable to collisions at more than one building class or overall, including Golden-winged Warbler, Painted Bunting, Kentucky Warbler, and Canada Warbler, building collision mortality appears substantial and may contribute to or exacerbate population declines. For species identified as highly vulnerable to collision for one building class but not across building types (Wood Thrush at residences, Worm-eating Warbler at high-rises), building collisions may still represent a threat. However, risk rankings for these species are more likely to be inflated by high mortality rates at a few sites, and further research is required to clarify the degree to which populations of these species are threatened by collision mortality. Inferences about population impacts of a mortality source should ideally be based on incorporating mortality estimates into demographic models (Loss et al. 2012) or comparing estimates to population abundance (Longcore et al. 2013). Data limitations preclude intensive population modeling of building collision impacts. Sampling bias toward densely populated areas east of the Mississippi River, and therefore toward certain bird species, prevented us from estimating species-specific an~ual mortality. We initially attempted to apply average species proportions to the overall mortality estimate following Longcore et a1. (2013), but this method returned unrealistically high estimates for species that comprised a high percentage of counts in many studies (e.g., 140% of the total population of Ovenbirds estimated to be killed each year by building collisions). Our vulnerability estimates controlled for abundance and range overlap with study sites and therefore provide a less biased approximation of species- specific collision risk. Our vulnerability analysis expanded upon the analysis of Arnold and Zink (2011), which was based on three sites in the northeastern U.S. and adjacent Canada. Nonetheless, we documented some of the same vulnerable species, including Brown Creeper, Black-throated Blue Warbler (Setophaga caerulescens), and Swamp Sparrow (Melospiza georgiana), and similar high-and low-risk taxonomic groups (e.g., warblers and swallows, respectively), As in the previous study, the vast majority of highly vulnerable species were long-distance migrants. Unlike the previous study, we did not assess whether population trends were correlated with building collision vulnerability. This approach has received criticism (Schaub et al. 2011, Klem et al. 2012) and shifts focus away from identifying which individual species of conservation concern face a high risk of colliding with buildings. Research Needs and Protocol Improvements Sensitivity analyses indicated that more research of mortality rates at low-rises will contribute greatly to improving mortality estimates. Future research should sample a variety of low-rise types, including residential, The Condor: Ornithological Applications 116:8-23, © 2014 Cooper Ornithological Society 84 20 U.S. bird -bui lding co lli sions co mme rci al, a nd indust rial buildin gs. Researc h at low· ri ses ha s o cc urred mostly at bu il dings t hat are known to cause la rge numbers of fata li tie s (e.g., office o r univers ity campus bui ldings with man y windows a nd /or near favorable bird h abitat). Ra nd om se lection of b u il dings for moni to rin g (fo r all building classes) all ows for less· bi ased co nclu sio ns about local mortality r ates a nd more re li ab le exte nsion of res ults wit hin study a re as and across regions. Mo r ta li ty data specific to different low · ris e building types wi ll a ll ow improvement upon the current a pproa c h of assum ing that all low· ri se buil d in gs have s imilar morta li ty rates. Because we ba sed our low -ri se est imate on t h e numbe r of U.S. "esta bli s hments," and be cause t he re lation s hip betwee n n umbe rs of estab li s hm ents a nd numbers of build ings is unknown, we suggest t hat improved data be collected an d made ava il ab le for t he number of U.S. low-rise bUild ings. Non-re s identia l lo w-rises are no t curren tl y in cluded in assessments by t he U.S. Census Bur ea u. Sens it ivity an alyses a lso indica te that mo rta li ty estimates will benefit fro m qu an t ifi catio n of sea rch e r effi cienc y and scave nge r r emoval rat es. Rec ent resea rch has res ul ted in major advancements in und e r s tandin g these b iases. including studies that es timate ca rcass detection and /o r scave nge r rem ova l rates (Co llin s and Horn 2008, Hage r et al. 201 2, 2013) or apply meth ods to s imul taneous ly account fo r both bia ses (B ra cey 20 11, Etterso n 20 13). In th e future, st udies s h ould account fo r t hese bi ases whe n poss ibl e and in vest igate how t hese ra tes are affected b y s ize a n d spec ies of ca rcasses, a bu n dance and communi ty composition of sc av e nge rs, and ch aracte ri st ic s of vegeta- tion and habi tat near buildin gs. A la rge portion of t he unpubli shed data we used were co ll ected b y vo lu n teer-led co ll is ion· moni torin g programs in major c it ies. These c iti ze n -sc ie n ce pro grams have contributed greatly to t he understanding of bird -buildin g colli sions; h oweve r, sta nd ardization of data co ll ection a nd recording procedures is nece ssary to m ake t h ese data m o re co mparable across program s a nd ac ross years wit hin programs. As a first s tep , a ll monitoring prog rams s ho uld r ecord sa mplin g effo r t, including (1) a reco rd of a ll sur veys conducted , eve n th ose wit h zero fata li t ies found ; (2) t he number of perso n-h ours of sam p lin g in eve ry survey; (3 ) t he n u m ber o f buildin gs and bu il din g facades samp led; (4) street addresses of build ings (wit h atte ntion to avoiding mu lt ipl e add resses r efer rin g to one building and cl arifying when one address includes > 1 b u il ding); and (5) separate records of fata li ties found during s urveys o n officia l r o ut es and those found in c id e ntall y outs id e of survey periods and/or off of routes. Th is in formation wi ll a ll ow increased co mp ara bili ty of data among re g io ns, improved u nd e r- s tandin g of seasonal an d regio nal mortality patterns, and reduced bi as in est im ates of pe r · bu il d ing mo rta li ty rates and ove ra ll mortality. Co mbining effo r t-corrected mortal- ity data w ith in format io n ab o ut bu il din gs (e.g., height in S. R. Loss, T. Will, S. S. Loss, and P. P. Marra sto ri es and meters; orientation and area of build ing facades; glass a rea, type, exten t , and reflect ivity; vegetation presence, type, dens ity. an d h e igh t; and amount o f li g h t emitted), wi ll a ll ow id e ntifi cat ion of mo r ta li ty rat e co rre late s. pred iction of mo r tality rates from building c haracte ri stics, and impl e m entatio n of tec hn iques to r educe mortali ty. Mo ni torin g program s cou ld a lso ex pand to inco rp orate sam pling at multipl e building t ypes, in clud in g in d ividu a l residences and add iti o na l types of low-rises and high -r ises. A nat io n al reporting system and database fo r b ird morta li ty data would faci li tate standard · iza tion of data collection fo r building co ll isio n s a nd oth er morta li ty so urces (Lo ss et a l. 2012 ). Un til this t)'pe of co mpre hensive system is d eve loped and lau nc hed, window co lli s ion mon itor in g prog rams ca n u se s imp le user·defi n ed data e n t ry portals th at will increase standardizat ion of data record ing, formatting, and co mp il ation (see ex ample at h ttp s: II d ocs . go 0 g l e .c o m I s p r ea d s h eet/vi ew for 111 ? u s p= dr ive_we b &for mkey=d D A ldD V T SVUzS I NfXONxW m ZxTEctbHc6MQ #g id =O ), and t herefore benefit resea r ch t ha t sy n th es izes mu ltip le datasets. Model Limitations Because data co ll ec ti on methods va ri ed greatly among st udi es, we cou ld no t account for a ll differe nces amo n g th e datasets we syn th es ized . How th is li mitation influ e nced ou r esti mates is u ncl ear. No n ethe less , our inclu s io n criter ia removed stud ies that lacked a systematic com pon ent to samp lin g, and we acc ounted for parti al-yea r sa mplin g by either estimating mortality us ing only yea r ~rou nd studi es or ap pl ying correction factors to mo r ta li ty es timates. \Xfe also accou n ted for samp le size d ifferences w he n est im ating s pec ies vul ne ra b ili ty. However, th e data we ana lyzed ove r re presented th e easte rn U.S. and unde rreprese nted t he Great Plains , Inte ri or \X/est, and West Co ast. Because of t hi s data limi tation, t h e mortali ty rate di stribution s that we app li ed to all U.S. buildings we re p rimarily based on data from the easte rn U.S. This co uld ha ve bi ased our estimates if mor ta lity rat es in th e \"Vest diffe r co n sistently from th ose documented in the East ; however, t he la ck of western data prevents conclu sions about s uc h reg io na l va ri a ti o n. In add it io n . our species vu ln erabili ty estimates do not cove r s peci es with a large proport ion of t he ir ra n ge in th e West. Fu rth e r researc h of bird -buil d ing co ll isions in areas west of the Mississ ippi River is needed to docume nt wheth er pe r-buil d in g morta li ty rates differ cons iste ntl y from tho se in we ll -stud ied reg ions of the east and to assess building co ll ision vu lnera bili t ies for western bird spe ci es. Our morta lity estimates a re limi ted by t he ass um ption t hat all non-res ide n ti al esta bli shments li sted by the U.S. Ce nsu s Bureau are :::; 11 sto ri es ta ll and t hat all bu il d in gs sa mpl ed by mo ni to ri ng prog ram s in major downtow n areas a re > 12 stories ta ll. T hese assum pt io ns we re u navoidab le because U.s . low-r ise bui ldin g data a re n ot ava ilabl e a nd The Cond o r: Ornithological Applications 116:8 -23, ©20 14 Cooper Ornithological Society 85 S. R. Loss, T. Will, S. S. Loss, and P. P. Marra building height information was not recorded in most studies. . Our mortality estimates may be conservative because data from buildings that cause exceptionally high annual rates of collision were removed from our analysis before extending average rates to the scale of the entire U.S. Hundreds to greater than one thousand birds per year have been found at intensively monitored buildings in or near areas with a high concentration of birds during migration (e.g., Taylor and Kershner 1986, M. Mesure and D. Willard personal communication). Other factors that may have contributed to underestimation include crippling bias (e.g., an uncertain percentage of birds flyaway from sampling areas before dying) and sub-lethal effects that may influence social interactions and migration behavior even if not causing eventual death (Klem 1990b). Further research to quantify crippling bias and sub-lethal effects is crucial for continued improvement in the accuracy of mortality and species vulnerability estimates. Finally, we were unable to quantify seasonal patterns of mortality due to a limited sample of studies that surveyed throughout the year. Additionally, several studies employed varying sampling effort across seasons and did not record effort data that could be used to account for this variation. Among records meeting our inclusion criteria, 60.0% were found during fall migration (August-November) and 37.0% were found during spring migration (March-May). These figures are likely inflated relative to non-migratory periods because most studies sampled only during spring and fall. Despite varying sampling effort among seasons, mortality during fall migration appears to be consistently greater than during spring migration; this pattern was seen in most of the datasets and could be related to larger populations of birds in the fall due to presence of young-of-the-year birds. Notably, several studies have indicated substantial building collision mortality during periods outside of migration, including in winter at individual residences (Dunn 1993, Klem 2009) and in summer at low-rise buildings (Bayne et al. 2012, Hager et al. 2013). Our methods accounted for partial-year sampling by either using only year-round studies (for residences and low-rises) or applying a correction factor that assumed additional mortality during summer and winter (for high-rises, a building type for which little data exists for non-migration periods). Species vulnerability estimates were also likely to be influenced by seasonal sampling biases, with in-transit migratory species likely overrepresented compared with summer and winter residents. Additional year-round studies are needed at all building types to clarify how mortality rates and species composition of fatalities vary by season. Conclusions As human populations and numbers of buildings increase in the U.S. ~nd globally. actions to reduce bird mortality U.S. bird-building collisions 21 from building collisions will be necessary at all types of buildings. For residences, mitigation techniques could include reducing vegetation near windows, angling win- dows to reduce reflection, and installing netting, closely spaced decals, or UV light-reflecting glass (Klem et al. 2004, Klem 2006, 2009). For low-rises and high-rises, mortality can be reduced by minimizing light emission at night (Evans Ogden 1996, 2002) and incorporating bird friendly design elements into new and existing buildings (e.g., Brown and Caputo 2007, Sheppard 2011). A long- term approach to reducing mortality is the continued adaptation of Green Building certification standards to include bird collision risks (Klem 2009). We provide quantitative evidence of the large amount of bird mortality caused by building collisions in the U.S. Our estimates represent roughly 2-9% of all North American birds based on a rough estimate of 10-20 billion total birds in North America (U.S. Fish and Wildlife Service 2002). However, because our results illustrate that not all species are equally vulnerable to building collisions, and because considerable uncertainty remains regarding species-spe- cific mortality and population abundance, the actual impacts of collisions on population abundance are uncertain. Despite this uncertainty, our analysis indicates that building collisions are among the top anthropogenic threats to birds and. furthermore, that the several bird species that are disproportionately vulnerable to building collisions may be experiencing significant population impacts from this anthropogenic threat. ACKNOWLEDGMENTS We thank the following people and organizations for providing access to unpublished datasets from building collision monitoring programs: K. Brand (Lights Out Win- ston-Salem, Forsyth County Audubon Society & Audubon North Carolina), A. Conover (Lights Out Columbus, Ohio Bird Conservation Initiative & Grange Insurance Audubon Center), M. Coolidge (Bird Safe Portland, Audubon Society of Portland). S. Diehl and C. Sharlow-Schaefer (Wisconsin Night Guardians, Wisconsin Humane Society), J. Eckles, K. Nichols, and R. Zink (Project Bird Safe Minnesota, Audubon Minnesota & University of Minnesota), S. Elbin and A. Palmer (Project Safe Flight, New York City Audubon), M. Flannery (California Academy of Sciences), D. Gorney (Lights Out Indy. Amos W. Butler Audubon Society). A. Lewis and L. Fuisz (Lights Out DC, City Wildlife). M. Mesure (Toronto Fatal Light Awareness Program), W. Olson (Lights Out Baltimore. Baltimore Bird Club). A. Prince (Chicago Bird Collision Monitors. Chicago Audubon Society). K. Russell (Audubon Pennsylvania), and D. Willard (The Field Museum). A. Bracey, J. Ducey. M. Etterson. S. Hager, A. Harrington. D. Horn, G. Niemi. and T. O'Connell provided access to unpublished or otherwise unavailable data. R. Schneider and J. Rutter provided assistance with data collection and management; E. Bayne. C. Machtans, and C. Wedeles The Condor: Ornithological Applications 116:8-23, © 2014 Cooper Ornithological Society 86 22 U.s. bird-building collisions provid ed access to unpublished manusc r ipts; and M. Lynes and C. Sheppard assisted in locating datas ets. \Xfe give special thanks to D. Klem for providing access to nearly all of his window collision data, investing significant effort along with P. Saenge r t o digi t ize histori ca l r eco rds, an d for pioneering the study of bird -window collisio n s. S.R.L. was supported by a postdoctoral fellowship funded by the U.S. Fish and W'i l dlife Service through the Smithsonian Insti tuti on's Po stdoctoral Fellows h ip program. T h e fi ndin gs and opinions exp ressed in this paper are those of the autho rs and do not necessa r ily renect the opi n ions of the U.S. Fish and \Xfildlife Service or the Sm ithsonian I nstitution. LITERATURE CITED Arnold, T. W., and R. M . Zink (201 1l. Collision mortality has no discernible effect on population trends of North American birds. PLoS One 6:e24708. Banks, R. C. (1979). Human related morta l ity of birds in the United States. U.s. Department of the Interior, Fish and Wildlife Service. Special Scientific Report -Wil dlife No. 215. Washington, DC, USA. Bayne, E. M., C. A. Scobie, and M. Rawson-Clark (20 12 ). Factors in fl ue ncing the annua l risk of bird-window collisions at residential structures in Alberta, Canada. Wildlife Research 39: 583-592. Bes t , J. (2008). Birds-Dead and deadly: Why numeracy needs to address social construction . Numeracy 1 (1 ):art icl e 6. Blanche r, P. J. (20 13 ). Esti mated number of birds killed by house cats (Felis carus) in Canada. Avian Co n servation and Ecology 8(2):3. Booth, G., and D. Co ll ister (1998). Calgary Bird Banding Society 1997 Annual Technical Report. Calgary Bird Banding Society, Calgary, AB, Canada. Borden , W. c., O. M. Lockhart, A. W. Jones, and M. S. Lyons (2010). Seasonal, taxonomic, and loca l habitat compo nents of bird -window collisions on an urban uni ve rsity campus in Cleve land, OH. The Ohio Journal of Science 110:44-52. Bracey, A. M . (2011). Window related av ian mortality at a migrat ion corridor. M.s. thesis, University of Minnesota- Duluth, USA. Broughman,S. P., N. L. Swaim, and C. A. Hryczan iuk (2011). Characteristics of Private Schools in the United States. Results from the 2009 -10 Private Schoo l Universe Survey (NCES 2011-339). National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Educati on. Was hington, DC, US A. Brown, H., and H. Caputo (2007). Bird-Safe Building Guidelines. New York City Audubon Society, New York, NY, USA . Codone r, N. A. (1995). Mortality of Connecticut bi rd s on roads and at buildings. The Connectic ut Warbler 15:89 -98. Co ll ins, K. A., and D. J. Horn (2008). Bi rd-window collisions and facto r s influencing their frequency at Millikin University in Decatur. Illinois. Abstract. Tr ansactions of the 1IIinois State Academy of Sc i ence 101 (S upplement). Collister, D., G. Booth, and B. Couro nn e (1997). Calgary Bird Banding Society 1996 Annual Technical Report. Calgary Bird Banding Society, Calga ry. AB, Canada. S. R. Loss, T. Wi ll , S. S. Los s, and P. P. Marra Collister, D., G. Booth, G. Meye r, and B. Cou ronne (1996). Calgary Bird Banding Society 1995 Annual Technical Report. Calgary Bird Band i ng Society, Calgary, AB, Canada. Dunn, E. H. (1993). Bird mortality f rom st riking residential windows in w inter. Journa l of Field Ornithology 64:302 -309 . Erickson, W. P., G. D. Johnson, and D. P. Young, Jr. (2005). A summary and comparison of bird mortality from anthropo- genic causes w ith an emphasis on collis i ons. In Bird Conservation Implementation and Integ ration in the Amer- icas: Pr oceed ing s of the Third International Partners in Flight Conference (c. J. Ralph and T. D. Rich , Editors). USD A Forest Service Genera l Tech ni cal Report PSW-GTR-191, pp. 1029 - 1042. Enerson, M. A. (20 13). Hidden Markov models for est imating animal mortal i ty from anthropogenic causes. Eco l ogical Applications. doi:1 0.1890/12 -1166.1 Evans Ogden, L. J. (1996). Co ll ision Course: The Hazards of Lighted Structures and Windows to Migrating Birds. World Wildlife Fund Canada and the Fatal Light Awareness Program, Toronto, ON, Canada. Evans Ogden, L. J. (2002). Summary Report on the Bird Friendly Bu ilding Program: Effect of Light Reduction on Collision of Migratory Birds. Fatal Lig ht Awa reness Program, Toronto. ON, Canada. Fink, L. C. (1971 ). Bi rd s in downtown Atlan ta-Spring 1970. The Oriole 36:2-6. Gelb, Y., and N. Delacretaz (2006). Avian window strike mortal ity at an urban office build ing. Th e Kingbird 56 :1 90-198. Hager, S. B., B. J. Cosentino, and K. J. McKay (20 12 ). Scaveng in g affects persistence of avian carcasses result ing from window co llisions in an urban landscape. Journal of Field Ornithology B3:203 -211. Hager, S. B., B. J. Cosentino, K. J. McKay, C. Monson, W. Zuurdeeg, and B. Blevins (2013). Window area and develop- ment drive spatia l variation in bird -window collisions in an urban landscape. PLoS One 8:e53371. Hager, S. S., H. Trudell, K. McKay, S. M. Cranda ll . and L. Maye r (2008). Bird density and mortality at windows. The Wi lson Journal of Ornithology 120:550-564. Johnson, R. E., and G. E. Hudson (1976). Bird mortality at a g lassed -in walkway in Washington State. Western Birds 7:99- 107. Klem, D., Jr. (1979 ). Biology of collisions between birds and windows. Ph.D . dissertation, Southern Illinois Unive rsity. Carbondale, IL, USA. Klem. D .• Jr. (1989). Bird -window collisions. The Wilson Bullet in 101:606-620. Klem, D., Jr. (1990a). Collisions between birds and win dows: Mortality and prevention. Journal of Fie ld Ornithology 61: 120-128. Klem, D .• Jr. (1990b). Bird injuries, cause of death, and recuperation from collisions wi th windows. Journal of Field Ornithology 61:11 5-119 . Klem, D., Jr. (2006). Glass: A deadly conservation issue for b i rds. Bird Observer 34:73-81. Klem, D., Jr. (2009). Avian mortali t y at w indows: The second largest human source of bird mortality on earth. In Tundra to Tropics: Proceedings of the Fou rth Internat ional Partners in Flight Confere nce (T. D. Rich, C. Arizmendi, D. Demarest, and C. Thompson , Editors). Partners in Flight, pp. 244-251. The Condor: Ornithological Applications 116:8-23, © 2014 Cooper Ornithological Society 87 S. R. Loss, T. Will, S. S. Loss, and P. P. Ma rra Klem, D., Jr., K. L. DeG root, E. A. Krebs, K. T. Fort, S. B. Elbin, and A. Prin ce (2012). A second critique of Arnold and link 20 11 {P LoS One 6:e24708]. PL oS One Comment. http://www. p los o ne.o rg l a n no ta t io nI l i s t Th rea d .action ? root=4 9 19 7 Klem , D., Jr., C. J. Farm er, N. Del acretaz, Y. Gelb, and P. G. Sae nger (20 09). Architectural and landscape ri sk fac tors associated with b i rd -glass co lli sions in an urban environ- ment. The Wilson Jo urnal of Orn it hology 12 1:126-134. Klem , D., Jr., D. C. Keck , K. L. Marty, A. J. Miller Ba ll , E. E. Ni ciu, and C. T. Platt (2004). Effects of window angling, feeder placem ent, and scavengers on avian mortality at plate glass. The Wilson Bulletin 116:6 9-73. Lo ngcore, T., C. Rich, P. Mineau, B. MacDonald, D. G. Ben , L. M . Sullivan, E. Mutrie , S. A. Gauth reaux, Jr., M. L. Avery, R. L. Crawfo rd, A. M . Manville II , et al. (20 12 ). An estimate of morta lity at communication towers in the United States and Canada . PL oS ON E 7:e34025. Longcore, T., C. Rich, P. Mineau, B. MacDonald, D. G. Bert, L. M. Sullivan, E. Mutrie, S. A. Gauth reaux, Jr., M. L. Avery, R. L. Crawfo rd, A. M. Manville II , et al. (2013). Avian mortal ity at comm unication towers in North America: Which species, how many, an d where? Biological Conservation 158:410-419. Loss, S. R., T. Will, and P. P. Marra (20 12). Direct human-caused mortality of b i rds: Improving quantifi cat ion of magnitude and assessment of population impacts. Frontiers in Eco logy and the Environment 10:357-364 . Loss, S. R., T. Will, and P. P. Marra (2013). The impact of f ree- ran ging domestic cats on wild life of the Un ited Sta t es. Nature Communications 4:1396. M achta n s, C. S., C. H. R. Wedeles, and E. M. Bay n e (2013). A first estimate for Canada of the number of birds ki ll ed by co ll iding w ith buildings. Avian Conse rvatio n and Ecology 8(2):6. O'Connell, T J. (20 01 ). Av ian window strike mortality at a subu rban office park. The Raven 72:141 -1 49. Ran ford, R. B., an d J. E. Mason (1969). Nocturna l migrant mortalities at the Toronto-Do minion Ce nter. The Ontario Field Biolo gi st 23:26 -29. Rich, T. D., C. J. Bear dmore, H. Berlanga, P. J. Blancher, M. S. W. Br adst reet, G. S. Butch er, D. W. Demare st, E. H. Dunn, W. C. Hunter, E. E.l riigo-E li as, J. A. Kennedy, et al. (2004). Partners in Flight North Ame ri can Landbird Conservation Plan. Cornell Lab of Ornithology, Ithaca, NY, USA. Sauer, J. R., J. E. Hines , J. E. Fallon, K. L. Pardi eck, D. J. Ziolkowski, Jr., and W. A. Link (2012). The North Am er ican Breeding Bir d Su rvey, Results and Analysis 1966-2011. Version 07.03 .2013. US GS Patuxent Wildlife Re sea rch Center, Laur el, MD, USA. Schaub, M., M. Kery, P. Korner, and F. Ko rner-Nievergelt (20 11). Cri tique of Arnold and Zink 2011 [PLoS One 6:e24708[. PLoS ONE Co mm e nt. http://www.p l os on e .o rg /a n notat i on/ li st Th read.acti on ?ro ot=9659 u.s . bird -building co lli sions 23 Seewagen, C. L., and C. Sheppard (201 2). Bird Collisions w it h Windows: An Annotated Bibliography. American Bird Con - servancy, The Plains, VA, USA. Sexton, L. (2006). Bird str i kes at co mm erc ial bu ildings in Atlanta . MS t hesis . Univers ity of Glamorgan, Wales, UK. Sheppa rd, C. (201 1). Bir d-Frie n dly Bui ldi ng Design. American Bird Conse rvancy, The Pl ai ns, VA, USA. Sibl ey, D. A. (2000 ). Nation al Audubon Society The Sibley Guide to Birds. Alfred A. Knopf, New York, NY, USA. Sky Scraper Sou rce Media (20 13). Sky Scrape r Page. http)lwww . skysc rape rpage.co m /C ities Smallwood, K. 5., D. A. Bell, S. A. Sny der, and J. E. DiDonato (20 10). Novel scavenger removal trials increase estimates of wind turbine-ca used avian fatality rates . Th e Journa l of Wildlife Manage m ent 74 :1 089-1097. Somerlot, K. E. (2003). Survey of songbird mortality due to window collisions on the Murray State Uni ve rsity campus. Journal of Service Learning in Conservation Biology 1 :1-9. Taylor, W. K., and M. A. Ker shner (19 86). Migrant birds killed at the vehicle assembly build i ng (VA B), John F. Kennedy Space Cen ter. Jo urnal of Fi eld Ornithology 57:142 -154 . U.S. Census Bureau (20 08 ). 2007 Economic Census. http)1 fa ct f i n d e r2. ce n sus .g 0 v I f a ce sI t a b l ese rv i ce s/j s f I p a 9 es l p r oduc t v i ew.x h t m I ?pid =EC N _200 7 _ U S _OOA 1 & prodTyp""table U.s. Census Bur eau (20 11 ). American Housing Survey for the United States: 2009. Cu rr ent Housing Reports, Se ries H1501 09, U.S. Government Printing Office, Washi ngton, DC, USA . U.S. Census Bureau (201 2). 2010 Census Urban Fac ts. http:// www.cen su s.go v/geo /refe rence /ua /uafacts.htm I U.S. Fish and Wildlife Se rv ice (2002). Migratory Bird Mortality Fact Sheet. U.s. Department of the In terior, Fish and Wildlife Service, Arlington, VA, USA. U.s. Fish and Wil dl ife Service (200 8). Birds of Conservation Concern 2008. United States Dep artment of Interior, Fish and Wi ldl ife Service, Division of M igra tory Bird Management, Arlington, VA, USA. Ward, M. R., D. E. Stallknecht, J. Willis, M. J. Conroy, and W. R. Davidson (2006). Wild bird mortality and West Ni le vir us surve illance: Biases associated w i th detection, repo rting, and carcass persist ence . Journal o f Wildlife Diseases 42:92 -106. Warton, D. I., 1. J. Wright, D. S. Falster, and M. Westoby (2006). Bivariate line -fitting methods fo r all ometry . Biol ogical Re- views 81:259 -291. Weiss, R., and D. J. Horn (2008). Proj ect prevent coll i sion: A study of bird-wi ndow collisions in re si dential neighborh oods in Illinois. Abstract. Transactions of the Ill inois State Academy of Science 101 (S uppl ement). Zink, R. M., and J. Eckles (20 10). Twin Cities bird-building co lli sions : A sta tu s update o n "Project Birdsafe.'· The Loon 82:34-37. The Condo r: Ornithological Applications 116:8 -23, © 2014 Cooper Orn ithological Society 88 Good afternoon Jane, T hanks fo r t h e fo ll ow-u p .... g lad to hea r t he Bi,-d Safe Design is movin g forward. Hurri ca n e resistant glass is ava il ab le from mu lti ple Architectu r al com me r c ia l glass fab r icators w it h a bird safe desig n opt io n . Addi ng a b ird safe ceram ic silkscreen frit to existin g approved hu r rica ne g lass does not violate Code complia nce for hurricane res istance. The price p r e mi um for bird safe g lass var ies wide ly based on the type of b ird safe g lazing d esign, the scope of work fo r t h e project and t h e specifi c glass fabricato r 's p r od uction cost. Hope th is h e l ps .. ..Jet me know if yo u have any que s tions. Jeff Ri got, u:m AI'Il[).C, CCI'Il Viracon Architectural Sales 800 Park Drive I Owat onna MN 55060 P,239.872.2S25 jdgot@yjraco n corn I y jrilco o co m 89 Good afternoon Jane, The bird safe glass solutions th a t Viracon cu rrently offers are s ilk screen a ppli e d cera mi c fr it patterns o r di g ita lly a ppli e d cera mi c ink p atte rns . Th e co st add for silksc r ee n ap pli ed frit is $7.0 0 / s qft for s tandard sc r ee n patterns in monotone paint co lor. Di gitally ap plied ceramic ink is used w here Ar c hi tects d es ire multipl e co lo r s to the patte rn. The add co st for digital ink in a ge om etri c s h a pe pattern is $30.00 / sqft. Th e FlU Pa rkvi ew II Res idence Hall curr e ntly in co ns tru ction used Viracon s ilkscreen ap plied bird safe frit. Th e Univ e r s ity o f Florida Ma la c ho ws ky Data Science Center breaking g round Su mm e r 2021 w ill a lso use s il kscr-een app li ed bird safe frit. Vir acon doe s not c urr e ntl y offe r first s urface etc hed bird sa fe g lass or fir s t s u rface ultra violet refl ecting coatings .... 1 am aware th at both of these bird safe technologi es a r e s ignificantly more expensive than silksc r ee n app li ed frit patte rn s . Contact Ma rni and I if yo u h ave any qu estio ns . Th a nk you Jeff Ri go t. UWAI' 8[hC,CCI'1I Vi r-aeon Ar chitectural Sales 800 Park Drive I Owatonna MN 55060 P: 239.072.2525 jrjgot@yj ra coocom I vjracon com 90