Owl

for Nocturnal Owl Monitoring in North AmericaM A R C H 2 0 0 1

D. Lisa Takats • Charles M. Francis • Geoffrey L. Holroyd • James R. DuncanKurt M. Mazur • Richard J. Cannings • Wayne Harris • Denver Holt

D. Lisa Takats • Charles M. Francis • Geoffrey L. Holroyd • James R. DuncanKurt M. Mazur • Richard J. Cannings • Wayne Harris • Denver Holt

PLEASE CITE AS:D. Lisa Takats, D. L., C. M. Francis, G. L. Holroyd, J. R. Duncan, K. M. Mazur, R. J. Cannings, W. Harris,D. Holt. 2001. Guidelines for Nocturnal Owl Monitoring in North America. Beaverhill Bird Observatoryand Bird Studies Canada, Edmonton, Alberta. 32 pp.

PUBLISHED BY:Beaverhill Bird Observatory and Bird Studies Canada, March 2001

ISBN 0-9688736-0-X

For more information or to receive a printed copy, contact:

Lisa Takats7th floor, O.S. Longman Building6909-116 StreetEdmonton, ABT6H 4P2

Phone: 780.427.1249E-mail: [email protected]

Information on the North American Raptor Monitoring Strategy is available at:http://www.im.nbs.gov/raptor/raptor.html

Copies of this document can be downloaded from: http://www.bsc-eoc.org

Environment�Canada�

EnvironnementCanada

i

T A B L E O F C O N T E N T S

GREAT GRAY OWL • GORDON COURT

Executive Summary ............................................................................................................... iii

Acknowledgements............................................................................................................... iv

List of September 1999 Workshop Participants ................................................................. v

Introduction ............................................................................................................................ 1

Review of Existing Owl Surveys in Canada ......................................................................... 3Manitoba .......................................................................................................................... 3

Ontario .............................................................................................................................. 3Alberta .............................................................................................................................. 4

Saskatchewan................................................................................................................... 4

British Columbia .............................................................................................................. 4Nova Scotia ....................................................................................................................... 5

Montana ........................................................................................................................... 5

Guidelines for Survey Protocols ........................................................................................... 7

Survey Objectives............................................................................................................. 7

Survey Methods ............................................................................................................... 7Route Selection ................................................................................................................ 7

Route Design .................................................................................................................... 8Georeferencing ................................................................................................................ 8

Number and Timing of Surveys ..................................................................................... 8

Silent Listening ................................................................................................................ 9Playback (optional) ......................................................................................................... 9

Time of Night ................................................................................................................. 10Environmental Conditions............................................................................................ 11

Owl Positions.................................................................................................................. 11

Surveyor Training ................................................................................................................. 12

Data Computerization ......................................................................................................... 13

Data Analysis......................................................................................................................... 15

Further Research Needs ....................................................................................................... 16

Literature Cited .................................................................................................................... 17

ii

GUIDELINES FOR NOCTURNAL OWL MONITORING IN NORTH AMERICA

APPENDICES

Appendix I: Recommended Equipment ................................................................. 19Appendix II: Scientific Names and Codes of North American Owls ................... 21

Appendix III: Beaufort Scale Translations to Wind Speeds ................................... 22

Appendix IV: Noise Level Descriptions ..................................................................... 22Appendix V: Random Route Selection .................................................................... 23

Appendix VI: Sample Data Sheet .............................................................................. 24

REMOVABLE DATA SHEET ................................................................................................... 25

iii

Information on distribution, abundance, and population trends of all North American

bird species is required for developing sound conservation strategies, identifying

species in need of particular conservation action, and evaluating the effectiveness of

current management programs.

these parameters over time. The key elementsare as follows:

• routes should be selected using appropriaterandomizations (if possible) to ensure thatthey are representative of the area beingsurveyed, within the constraints of aroadside survey.

• routes should consist of at least 10 stations,spaced at least 1.6 km apart, that can besurveyed in a single night.

• routes should be surveyed once per year atthe time when the majority of species inthe region are most active vocally.

• the starting position, and preferably allstations along a route should begeoreferenced to allow linking of owlrecords to locations for habitat analysis.

• the protocol at each station should startwith a 2-minute silent listening period.

• optionally, playback may be used at astation if particular species of owls arebeing targeted that may respond well toplayback.

• the field data form should be designed sothat the intervals in which each owl isdetected (i.e. before or after playback ofvarious species) are recorded.

• record the approximate direction anddistance to the first location where eachowl was detected.

E X E C U T I V E S U M M A R Y

Most species of nocturnal owls are poorlymonitored by existing multi-species surveys,such as the Breeding Bird Survey, MigrationMonitoring, and Christmas Bird Counts.Several regions of Canada and the UnitedStates have established volunteer-basednocturnal roadside surveys for breeding owls.These appear to be an effective means ofmonitoring many species of owls, but there isconsiderable variation in the methods used.

In February 1997, participants in a workshopon nocturnal owl monitoring discussed theproblems of current owl surveys (Holroyd andTakats 1997). In September 1999,representatives from the main volunteersurveys in Canada met in Winnipeg, Manitobato develop a set of standards for owlmonitoring that would allow data to beintegrated across surveys, while recognizinggeographic variation in target species andsurvey objectives. The outcome of thatmeeting was agreement on a set of standardcomponents that should be incorporated intoroadside surveys for breeding owls. Thesemeetings, with subsequent discussions, haveled to development of guidelines for surveyprotocols that we hope will be adopted by allorganizations running nocturnal roadsidesurveys for owls.

These procedures are designed for broadscale monitoring of relative abundance,distribution, habitat use, and changes in

LONG-EARED OWL • GORDON COURT

iv

This document is a product of a National Nocturnal Owl Monitoring workshop held in

Winnipeg on September 27-28, 1999.

Memorial Fund; Saskatchewan NaturalHistory Society; University of ManitobaAlumni Fund; Manitoba Conservation DataCentre; Boreal Wilderness Guides Ltd.;Partners in Flight Manitoba; Susann Myers(Parks Canada). We would particularly like to

thank all of the volunteers who havehelped by participating in the

existing surveys.

The national meeting washosted by Manitoba

Conservation (WildlifeBranch). Comments on thedocument from Jeff Adamyk(Alberta ConservationAssociation), Jim and BarbBeck (Beck Consulting) PeterBlancher (Canadian WildlifeService), Roger Brown,Gordon Court (Alberta

Environment), Sue Cotterill(Alberta Environment), Ray

Cromie, Tony Diamond, LoneyDickson (Canadian Wildlife

Service), Erica Dunn (CanadianWildlife Service), Bruce Falls, Hardy

Pletz, Charles Priestley, Steve Sheffield(US Fish and Wildlife Service), and Dave

Stepnisky were greatly appreciated. Thankyou to Elson Olorenshaw (Treasurer,Beaverhill Bird Observatory), who took careof all the funding received for this projectthrough Beaverhill Bird Observatory. Finally,a big thank you to all who donatedphotographs for the document.

A C N O W L E D G E M E N T S

Although not all participants are included asauthors, they all provided valuable input, andthe complete list of names and addresses ofattendees is given on the following page. Wegreatly valued the interactions and inputfrom everybody present at the workshop,including Bob Nero’s unreleasable GreatGray Owl, Lady Grayl, who patiently satthrough the workshop, reminding usconstantly why we were there.

An earlier draft of this documentwas presented at the annualRaptor Research Conference inLa Paz, Mexico in November1999, where we receivedvaluable feedback and inputand added Denver Holt (OwlResearch Institute), whoconducts extensive owl surveysin Montana.

We would like to thank all of theindividuals and agencies who havebeen supporting owl monitoringinitiatives and this workshop inCanada, including: Steve Brechtel andGordon Court (Alberta Environment-Fisheries and Wildlife ManagementDivision); Alberta Sport, Recreation, Parks,and Wildlife Foundation; Beaverhill BirdObservatory; Canada Trust Friends of theEnvironment Fund; Canadian Wildlife Service(Environment Canada); Merlin Shoesmith,Brian Gillespie, and Carol Scott (ManitobaConservation); Bird Studies Canada; OntarioMinistry of Natural Resources; James L. Baillie


LADY GRAYL• GEOFF HOLROYD

v

albertaLisa Takats, Beaverhill Bird Observatory7th Floor O.S. Longman Building6909-116 Street, Edmonton, AB T6H 4P2PHONE: (780) 427-1249 / FAX: (780) 422-9685E-MAIL: [email protected]

Geoff Holroyd, Environment CanadaCanadian Wildlife Service, Room 2004999-98 Avenue, Edmonton, AB T6B 2X3PHONE: (780) 951-8689 / FAX: (780) 495-2615E-MAIL: [email protected]

manitobaJim Duncan, Manitoba ConservationBox 24, 200 Saulteaux CrescentWinnipeg, MB R3J 3W3PHONE: (204) 945-7465 / FAX: (204) 945-3077E-MAIL: [email protected]

Kurt Mazur, Partners in Flight ManitobaBox 24, 200 Saulteaux CrescentWinnipeg, MB R3J 3W3PHONE: (204) 945-6816 / FAX: (204) 945-3077E-MAIL: [email protected]

Bob Nero, Manitoba ConservationBox 24, 200 Saulteaux CrescentWinnipeg, MB R3J 3W3PHONE: (204) 253-7115 (h) / FAX: (204) 945-3077

Heather Hinam, Dept. of ZoologyUniversity of ManitobaWinnipeg, MB R3T 2N2PHONE: (204) 253-7115 (h)

ontarioCharles M. Francis, Bird Studies CanadaP.O. Box 160Port Rowan, ON N0E 1M0PHONE: (519) 586-3531 / FAX: (519) 586-3532E-MAIL: [email protected]

Doug Gilmore, Ontario Ministry of NaturalResourcesBox 5003Red Lake, ON P0V 2M0PHONE: (807) 727-1336 / FAX: (807) 727-2861E-MAIL: [email protected]

Neil Dawson, Ontario Ministry of NaturalResourcesRR#1, 25th Side RoadThunder Bay, ON P7C 4T9PHONE: (807) 939-3120E-MAIL: [email protected]

saskatchewanWayne Harris, Saskatchewan Environment &Resource Management350 Cheadle St. WSwift Current, SK S9H 4G3PHONE: (306) 778-8218 / FAX: (306) 778 8212E-MAIL: [email protected]

british columbiaDick Cannings, Bird Studies Canada1330 Debeck Road, S.11, C.96, RR#1Naramata, BC V0H 1N0PHONE: (250) 496-4049E-MAIL: [email protected]

montanaU.S. Representative from November 1999 RaptorResearch Conference

Denver Holt, Owl Research Institute, Inc.P.O. Box 39Charlo, MT 59824PHONE: (406) 644-3412E-MAIL: [email protected]

S E P T E M B E R 1 9 9 9 W O R K S H O P P A R T I C I P A N T S

L I S T O F

BARRED OWL • GORDON COURT

vi


1

Information on distribution, abundance, and population trends of all North American

bird species is required for developing sound conservation strategies, so that species in

need of particular conservation action can be identified, and the effectiveness of

current management programs can be evaluated.

abundance. Migration Monitoring may havethe potential to monitor populations of someof the more common migratory species, suchas Northern Saw-whet Owls (Dunn 1999) (seeAppendix 1 [page 19] for scientific names),but the precision and reliability of trendestimates has not yet been demonstrated.Furthermore, it does not provide informationon breeding distributions of owls, and is notuseful for most species, as they do notmigrate.

Broadcast surveys are one of the most widelyused techniques to locate and survey owls(Bondrup-Nielsen 1978, Johnson et al. 1981,Smith 1987, Mosher et al. 1990). Owls vocalizeto communicate with their mates anddelineate territory (Johnsgard 1988).Imitating or broadcasting tape recordings ofowl vocalizations can invoke vocal responsesfrom many species of owls (Mosher and Fuller1996). This survey technique has been used todocument the range and status of several owlspecies in North America (Duncan and Duncan1997), and can also be used to determinehabitat associations (Laidig and Dobkin 1995,Lehmkuhl and Raphael 1993, Mazur et al.1997, Proudfoot et al. 1997, Duncan andKearns 1997, Takats 1998a). Unfortunately,despite their value for detecting owls,broadcast surveys have some limitations as anational monitoring tool. Playback calls willnecessarily vary among regions dependingupon the target species. Furthermore,

In the past few decades there has beenincreasing concern over the status of bothdiurnal and nocturnal raptors (Newton 1979).Birds of prey are high on the food chain,making them vulnerable to manyenvironmental factors, such as toxins (e.g.,DDT, Carbofuran) as well as habitat loss (Fyfeet al. 1976, Gutierrez et al. 1984, Noble et al.1993, Wellicome 1997, Whelan 1996). As such,they may be valuable indicators ofenvironmental health (Oliphant 1994) andmany species of raptors have been chosen asindicator species in various regions (Allen1987, Bosakowski 1994, Gutierrez and Carey1984, James et al. 1995, Johnson 1987,Duncan and Kearns 1997).

Relatively little is known about theabundance and population trends of mostspecies of nocturnal owls in North America.Most owls are not adequately monitored bythe existing multi-species continent-widesurveys in North America (Downes et al.1999). The Breeding Bird Survey takes placeoutside of the breeding season for most owls,and at a time of day (early to mid-morning)when most owls are relatively silent.Christmas Bird Counts are also conducted at atime of year when most owls are relativelyquiet, and the effort expended searching forowls is not standardized. Moreover, manyowls may shift their home ranges in winter, soChristmas counts do not necessarily provideinformation on breeding distribution and

GREAT HORNED OWL • GORDON COURT

I N T R O D U C T I O N

2


variation in the type of call (call note,duetting, song), quality of the recording,sequence of calls, species included in playback,effect of timing on response, and nature ofthe broadcast equipment may all affectnumbers of owls detected.

In Canada, volunteer owl surveys have beenestablished in Manitoba (Duncan and Duncan1991), Ontario (Francis and Whittam 2000,Lepage et al. 1999), Alberta (Takats 1998b),Saskatchewan (Harris, pers. comm.), and PrinceEdward Island (Susann Myers, pers. comm.).Although there are some similarities in theprotocols that have been adopted, there arealso a number of differences. If these surveys,as well as any new surveys that may bedeveloped, could adopt a standard protocol,then the data from different regions can beintegrated to provide national or continentaltrend and relative abundance information.

A Nocturnal Owl Monitoring Workshop washeld in February 1997 to discuss thedevelopment of a strategy to determine thestatus and assess trends of nocturnal owls inCanada and the USA (Holroyd and Takats1997). In September 1999, representativesfrom the major volunteer owl surveys inCanada met to develop guidelines forstandardizing owl monitoring surveys thatcould be used throughout Canada. Eventually,we hope that these guidelines will beintegrated into the North American RaptorMonitoring Strategy (http://www.im.nbs.gov/raptor/raptor.html). The goal of this document

is to present these guidelines for owlmonitoring protocols for North America. Wehope they will be adopted by all of theexisting surveys in Canada and elsewhere inNorth America, and also by any new surveysthat might be established.

These guidelines require a number ofcommon elements, but provide someflexibility to support regional needs. They areintended primarily for large-scale, extensivesurveys, particularly those involving volunteersurveyors. Volunteers are particularly suitablefor owl surveys, because most areas haverelatively few species of owls and volunteerscan be trained relatively easily todifferentiate common species. Inclusion ofvolunteers also helps increase publicawareness of owls and related conservationissues.

These procedures could also be used byorganizations such as consulting companies oragencies with objectives such as assessingrelatively large regions for owl populations.They are less suitable for small-scalemonitoring, which would generally requirealternative, more intensive procedures.

In this document, we first review the mainexisting owl surveys in Canada and Montana.We then define the objectives of surveysbased on the proposed standard protocol,each aspect of the protocol, and someconsiderations for data computerization andanalysis.

LISA

TA

KA

TS

3

ManitobaIn 1991, Jim and Patsy Duncan in cooperationwith Manitoba Conservation initiated the firstextensive, systematic volunteer owl survey inCanada. Initially consisting of 21 routes insoutheastern Manitoba and adjacentMinnesota, the survey grew to 77 routes by1998 and covered a larger area (Duncan andDuncan 1998). This survey’s goals were: 1) toestimate relative abundance and distributionof each owl species; 2) to estimate species-specific habitat associations; 3) to estimateyear to year population fluctuations; and 4) toprovide an opportunity for volunteers tocontribute to the understanding of owlecology. This owl survey has been run everyyear from 1991 to 1999. Although all specieswere recorded, the only playback calls usedwere of Boreal and Great Gray Owls. Surveyswere conducted between 30 min. after sunsetand 30 min. before sunrise, using a standardprotocol (Table 1).

Ten species of owls have been detected onthese surveys: Barn Owl, Barred Owl, BorealOwl, Eastern Screech-owl, Great Gray Owl,Great Horned Owl, Long-eared Owl, NorthernHawk Owl, Northern Saw-whet Owl, andShort-eared Owl. By 1999, over 400 volunteershad participated in this survey (Duncan, pers.comm.).

R E V I E W O F E X I S T I N G O W L S U R V E Y S

I N C A N A D A A N D M O N T A N A , U S A

R E V I E W O F E X I S T I N G O W L S U R V E Y S

I N C A N A D A A N D M O N T A N A , U S A

OntarioThe Ontario Nocturnal Owl Pilot Study wasstarted in 1995 in response to the Ministry ofNatural Resources need for information onowl populations in central and northernOntario forested regions to assess the impactof current forest management regimes(Francis and Whittam 2000, Lepage et al.1999). Bird Studies Canada (formerly LongPoint Bird Observatory) coordinates thissurvey, which presently has over 100volunteers This is a roadside survey, withdifferent protocols in northern and centralOntario (Table 1).

In northern Ontario, playback of Boreal Owl(to stimulate responses from Boreal andNorthern Saw-whet Owls) and Great GrayOwl has been used with a protocol similar tothe Manitoba surveys, except that 20 stopsare spaced 1.6 km. In central Ontario,playback of Boreal and Barred Owls is used,with only 10 stops, but a much longerplayback and listening period, becauseBarred Owls tend to respond slowly toplayback. Nine species of owls have beendetected: Barred Owl, Boreal Owl, EasternScreech-owl, Great Gray Owl, Great HornedOwl, Long-eared Owl, Northern Hawk Owl,Northern Saw-whet Owl, and Short-earedOwl.

Manitoba, Ontario, Alberta, British Columbia, Saskatchewan and Montana all have had

some relatively standardized owl monitoring programs running for various numbers of

years. Note that a number of these surveys were modified in 2000, to fit within the

guidelines presented in this document. Here we describe their protocols prior to those

changes.

NORTHERN SAW-WHET OWL • GORDON COURT

4


AlbertaIn 1988, Jim and Barb Beck organized the firstvolunteer owl surveys in the Alberta. BetweenFebruary 20 and March 22, 78 participants in31 parties heard and observed 528 owls of 7species in the Edmonton region. The surveywas relatively unstructured (effort was notstandardized) and gave primarilydistributional data (Beck and Beck 1988). TheAlberta Owl Prowl ran for one year, andcollected owl distributional informationacross the province (Beck and Beck, pers.comm.).

Standardized owl monitoring was initiated aspart of a Barred Owl study conducted in theFoothills Model Forest (FMF) 1995-1996(Takats 1998a), and as part of a forestfragmentation study on the Alberta PacificForest Management Area near Calling Lake(Court, pers. comm.). The objectives of theFMF study were to determine the distributionand habitat use of the Barred Owl in theregion. Standardized surveys were used tocollect distributional information, not only onBarred Owls, but on other species as well(Table 1). Surveys have continued along the10 routes that were set up in the first year ofthis project. Seven species have been recordedto date: Barred Owl, Boreal Owl, Great GrayOwl, Great Horned Owl, Northern Hawk Owl,Northern Pygmy Owl, and Northern Saw-whetOwl.

Primarily as a result of the work in the FMF, avolunteer nocturnal owl monitoring programwas initiated in 1998 (Table 1). This project isbeing coordinated through Beaverhill BirdObservatory with support from AlbertaEnvironment (Takats 1998b). The species ofparticular concern in the province are BarredOwl, Boreal Owl, Burrowing Owl, Great GrayOwl, Northern Pygmy Owl, and Short-earedOwl. Surveys are conducted anytime between30 min. after sunset and 30 min. beforesunrise. Boreal, Great Gray, and Barred Owltaped calls are played in the north, thefoothills, and the mountains, and NorthernSaw-whet, Long-eared, and Great Horned Owlare played in the southern prairie/aspen

regions. Ten species of owls have beenidentified including: Barred Owl, Boreal Owl,Great Gray Owl, Great Horned Owl, Long-eared Owl, Northern Hawk Owl, NorthernPygmy Owl, Northern Saw-whet Owl, Short-eared Owl, and Snowy Owl.

NORTHERN SAW-WHET OWL • GORDON COURT

SaskatchewanIn Saskatchewan, non-standardized surveyswere conducted along 31 different roadroutes from 1974 through 1991 (Harris, pers.comm.). Most routes were repeated atirregular intervals. These routes had variabledistance between stops, were variable inlength, and used a variety of human imitationowl calls instead of standardized tapeplaybacks.

From 1992 –1994 a volunteer survey programsimilar to that of Manitoba was carried out(Table 1). Although emphasis was on borealforest owls, routes were also completed in theaspen parkland, grassland (riparian corridors)and Cypress Hills regions. These surveys wereinitiated 30 min. after sunset and alsorecorded amphibians. Since that time, nocoordinated effort has been made, althoughsome volunteers continue to conduct theirroutes.

British ColumbiaIn the Okanagan area of British Columbia,non-standardized surveys for Western

5


Screech-owls, FlammulatedOwls, Northern Saw-whetOwls and Northern PygmyOwls have been conducted(Cannings, pers. comm.).These surveys were conductedin late February and March forthe Screech and Saw-whetOwls, and May-June for theFlammulated and Pygmy Owls.One minute of silent listeningwas followed byapproximately one minute ofvocal imitations, followed byone minute of silent listening;all repeated three times for atotal of approximately fiveminutes spent at each stop.

Nova ScotiaA volunteer owl survey wasinitiated on Cape BretonIsland, Nova Scotia in spring2000. The objective was todevelop an annual survey tocollect information onpopulation densities andfluctuations, distribution, and habitatpreferences of nocturnal owls (Myers, pers.comm.). Surveys were conducted from mid-March through mid-May. Surveys wereinitiated _ hour after sunset and started with2 min. of silent listening followed by tapedcalls of Boreal and Barred Owls, followed by afinal 2 min. of silent listening (Table 1).Routes consisted of 10 stops spaced at 1.6 kmintervals. Four species of owls were detectedincluding: Barred, Great Horned, NorthernSaw-whet, and Long-eared Owls.

MontanaThe Owl Research Institute in Montana (ORI)has been conducting standardized nocturnalowl surveys for over 10 years (Holt, pers.comm.). Species of regional concern includeBoreal, Flammulated and Great Gray Owls.


The survey does 3 min. silent listening atstations set at 1/4 mile (400 m) intervals alongroads in the western half of the state. Thesesurveys are conducted along roads and bysnow machines in higher elevations. The workis conducted from mid-May to mid-June forFlammulated Owls and from mid-February tomid-April for all other forest species.

Owls that have been recorded on these silentlistening surveys include: Eastern and WesternScreech Owl, Flammulated Owl, Barred Owl,Boreal Owl, Great Gray Owl, Great HornedOwl, Long-eared Owl, Northern Pygmy Owl,and Northern Saw-whet Owl. The first nestrecords and status reports in Montana forBarn, Boreal and Flammulated Owls are aproduct of the surveys. Additionally, relativehabitat associations for western Montana owlshave been identified.

6


PROTOCOL INFORMATION FOR LONG-TERM SURVEYS INMANITOBA, ONTARIO, ALBERTA, SASKATCHEWAN,BRITISH COLUMBIA, NOVA SCOTIA, AND MONTANA

Manitoba 20 or more 0.8 km3:40 min

1 visit- late March/early April

1:00 listening0:20 BOOW / 1:00 listening0:20 GGOW / 1:00 listening

RegionStations per

routeStation spacing/time at station

Surveys per year/timing

Broadcast protocol (minutes)& playback species*

Ontario -north

20 stations 1.6 km3:40 min(4:40 min in 1999)

2 visits (3 in 1999)- early-mid March- early-mid April- late April/early May

2:00 listening (2:00 in 1999)0:20 BOOW / 1:00 listening0:20 GGOW / 1:00 listening

Ontario -central

10 stations 2.0 km13:00 min(14:00 min in 1999)

2 visits (3 in 1999)- early-mid March- early-mid April- late April/early May

1:00 listening (2:00 in 1999)0:20 BOOW / 1:00 listening0:20 BARR / 2:00 listening (x2)0:20 BARR / 1:40 listening (x3)

AlbertaFoothills

10 stations 1.6 km15:00 min

3 visits (4 in 1995)- March 20 to May 5

2:00 listening0:20 BARR / 1:00 listening (x6)5:00 silent listening

Alberta 10 stations 1.6 km8:00 min(9:00 in 1999)

2 visits- March 20 to April 10- April 11 to May 2

2:00 listening0:20 BOOW/NSWO / 1:00 listening0:20 GGOW/LEOW / 1:00 listening0:20 BARR/GHOW / 3:00 listening

Saskatchewan non-standard 1.0 km or more5:00 min

April 1 to 10April 30 to May 8

1:00 listeningBOOW/GGOW/BARR imitations orother species in different habitats

BritishColumbia

non-standard variable All species: LateFebruary to MarchFlammulated/PygmyOwl: May to June

1:00 listening1:00 Vocal imitation1:00 listening(repeated over 5 minutes)

Nova Scotia 10 stations 1.6 km8:00 min

1 visit- March 17 to May 7

2:00 listening0:20 BOOW / 1:00 listening0:20 BARR / 2:00 listening0:20 BARR / 2:00 listening

Montana Standardroutes,differentlengths

0.25 mile (400 m)3:00 min

All owl species:February 15 to April 15Flammulated Owl:May 15 to June 15

3:00 listening

*OWLS: BOOW - Boreal / BARR - Barred / GGOW - Great Gray / NSWO - Northern Saw-whetGHOW - Great Horned / LEOW - Long-eared

table 1


7

Survey ObjectivesThese guidelines were developed to achievethe following objectives:

1/ Obtaining information on distribution ofowls.

2/ Estimating relative abundance of owlswithin regions and across North America.

3/ Estimating trends in populations ofnocturnal owls at scales ranging fromregional (ie. ecoregion, province, state) tocontinental.

4/ Determining habitat associations of owls.

Survey MethodsThe basic survey method being proposed is tolisten for calling owls along a predeterminedroute consisting of a minimum number ofevenly spaced stations (Bibby et al. 1992). Inmost cases, the routes will be along secondaryroads, with relatively little traffic, althoughoff-road routes could be developed in someareas. This basic sampling method is used bythe Breeding Bird Survey, and lends itself tolarge-scale surveys where the intention is toobtain data that can be analysed at a regionalor larger scale. It is less suitable for intensivesampling of small areas. As well, it has thedrawback that results may only beextrapolated to habitats along roads, wherepopulation trends may or may not be the sameas those away from roads.

Route SelectionRoutes need to be selected so that they arerepresentative of the region being surveyed,in order to make valid statistical inferencesabout owl populations in the region.

• The only way to ensure that routes arerepresentative is to select routes randomlyfrom within the survey area using some sortof stratified sampling scheme.

• Although some routes could be selectedaway from roads, for access by snowmobileor horse or even possibly on foot (thoughfew routes could safely be done on foot atnight), most routes will necessarily be alongroads. Suitable roads must be accessible inlate winter/early spring, should not haveexcessive traffic or heavy logging trucks (forsafety reasons and so that owls can beheard) and should go through potentiallysuitable habitat.

• Each route should be separated by at least 5km from any other route, to minimize therisk that the same owls will be heard onmore than one route (Anderson et al. 1979).

• The objective of random route selection is toensure that all suitable roads are equallylikely to be selected. One possible approachto selecting random routes is outlined inAppendix V (on page 23).

• Unfortunately, there are a number ofdifficulties in selecting random routes. Forexample, information on which roads aresuitable, especially with respect to winteraccessibility and habitat, may not beavailable centrally. Also, volunteers may notalways be willing to survey selected randomroutes. Furthermore, even if roads areselected randomly, habitats or owlpopulations near roads may differ fromthose away from roads.

• In many cases if may not be possible toselect routes in a fully random fashion.Provided that routes are selected withoutprior knowledge of the distribution of owls,we believe that data from such surveys arestill valuable, especially in the absence of

G U I D E L I N E S F O R S U R V E Y P R O T O C O L S

LONG-EARED OWL • LISA TAKATS

8


any alternative information. Nevertheless,the greater the element of randomization,the greater the statistical credibility of thesurvey.

• Existing programs, with non-random routes,should continue to run existing routes,because there is considerable value tomaintaining continuity, but should try toadopt a suitable randomization procedurefor selecting new routes. In analyses,random and existing routes should betreated separately, and if average densitiesor trends prove to differ on the two types ofroutes, it may be appropriate to phase outthe non-random routes and replace themwith random routes (e.g., by attrition,through replacing non-random routes withrandom routes when volunteers drop outand new ones join).

• If any off-road routes are developed, theyshould be clearly identified as such, as theymay require separate analysis, due at least inpart to differences in selection procedures.

• In reporting on the results of the survey, it isimportant to clarify the area that has beensampled, and the procedures used to selectroutes, as this needs to be taken account inthe analysis (e.g. for developing weightingfactors for routes) as well as in theinterpretation of results.

• Because routes without owls do notcontribute to trend analysis (and are unlikelyto interest volunteers) and routes withoutowls for two years in a row, could bediscontinued, but efforts should be made torun them again every five years or so, in caseowls have returned to the route (thisprocedure has been used by the MourningDove call survey in the United States).

• Selected routes should usually be ground-checked during the day, prior to starting thesurvey, to ensure that they are, in fact, safeand usable, and go through suitable habitat.

Route DesignEach route should have 10 stations, distributedalong the route at equal intervals of 1.6 km.

• If the listening/playback protocol is short[see below], and the length of suitable roadis adequate, then it is recommended thatanother route be run (continued from thefirst route, or in another area).

• The spacing of 1.6 km is intended to reducethe chances of detecting the same owl atmultiple stations, while not requiringsurveyors to spend too much time drivingbetween stations. Depending upon thetopography, some of the louder owls, suchas Barred Owl, can be heard at distances of2 km or more (Takats 1998a, Mazur pers.comm., Duncan pers. comm.), but other owlscannot be heard as far or as clearly. Inpractice, we have found that most smallowls are not heard at neighbouring stationsalong the route, if stations are spaced at 1.6km.

GeoreferencingThe location of the starting point of the route,and of each station along the route, should berecorded as precisely as possible, either using aGPS (Global Positioning System) or throughreading the coordinates from an accurate anddetailed map.

• Accurate locations help ensure that thesame route and stations can be relocated inthe future

• In conjunction with GIS habitat maps,accurate locations allow analysis of broadscale habitat associations of owl locations,through plotting precise owl locations onthe maps.

• Knowledge of station locations is required,in combination with accurate habitat mapsto enable post-hoc stratification, to ensurethat routes are appropriately weightedrelative to the amount of each habitat inthe region.

Number and Timing of SurveysEach route should be surveyed once per yearat the time of year when vocal activity of themajority of species is greatest. The survey

9


window should be relatively broad (e.g., 4weeks) to maximize the number of surveysthat can be conducted, and to include anyannual variation in phenology.

• A single survey per year would encouragemore surveyors to participate by reducingthe amount of time spent surveying. Highlymotivated volunteers could be encouragedto survey multiple routes per year thusallowing for a higher number of routes tobe surveyed.

• Surveying a route two (or more) times peryear would provide information on annualvariation in the peak time of owl calling,and would more accurately monitor owlspecies with peak calling at different timesof the year. However, for this generalsurvey, we do not believe these advantagesjustify the 2-fold (or more) increase in thesurvey effort required. For a more intensivesurvey or limited areas, more than onerepeat survey may be preferred.

• The optimal timing for surveys is likely tovary among regions. In Canada this mayrange from mid-February through Maydepending upon the location. Also, there issome variation in peak calling amongspecies (for example, in Ontario andAlberta, peak calling of Great Horned Owlsis earlier than for Barred Owls). In mostareas the calling period for each species isbroad enough that there are time periodswhen all species are potentially calling. Ifpossible, survey timing should be selectedto minimize the number of migrating owlsrecorded. The survey window should beclearly defined by the survey coordinators.

• Each route should be surveyed close to thesame date every subsequent year.

Silent ListeningAll protocols should start with a two-minutesilent listening period at each survey stop.

• This will allow data to be compared acrossthe continent, regardless of what playbackprotocols (if any) may be adopted. Two

minutes appears to be adequate for mostspontaneously calling owls to be detected,at least during the period of peak callingactivity. In Alberta, relatively fewadditional owls were detected during athird minute of listening (Takats, pers.comm.). In Ontario, more than 70% of 5species of owls that were detected over a 5minute period (included playback) weredetected in the first two minutes (Francispers. comm.).

• A relatively short silent listening periodallows for the possibility of incorporatingplayback, if desired, or for increasing thenumbers of stations to be surveyed, both ofwhich are likely to be more efficient than aprotracted silent listening period.

Playback (optional)It is well known that broadcasting recordingsof owl vocalizations can increase calling ratesor invoke approach from many species (Fullerand Mosher 1981, McGarigal and Fraser 1985,Duncan and Duncan 1991, Lepage and Francis1998, Mazur pers. comm., Takats pers.comm.), although this has not been the casein all studies.

• Regionally specific playback protocols, oradditional silent listening periods could beadded, provided that owls heard duringthese periods are recorded separately fromthose heard during the first two minutes,and the playback protocol is standardizedat each station.

9

BROADCASTING • GORDON COURT

10


• Playback protocols, however, cannot bestandardized across the continent, becauseof variation in target species and thedifferences and changes in recordingquality, broadcast species, or broadcastequipment which could affect responserates and hence lead to long-term bias intrend estimates.

• Carrying and working with playback unitson a cold winter night can be a significanthassle. Playback can also potentially bedisruptive to owls (may increase risk ofpredation, disrupt foraging and courtship,and/or draw females off nests). In addition,playing calls can pull owls off theirterritories giving inaccurate information ontheir habitat use (Holroyd and Takats 1997).

• The benefits of broadcasts varyconsiderably among species, and need tobe balanced against the problems. Forexample, in Ontario, a 12-minute period ofalternating broadcasts and silent listeningincreased 3- to 6-fold the number of BarredOwls detected relative to the initial 2-minute silent listening period (Francis,unpublished). But for Northern Saw-whetand Boreal Owls, the relative increase incalling rates was much lower (because mostof them were calling spontaneously); forGreat Gray Owls there was no noticeableeffect of playback on calling rates.

• We recommend against the use of imitatedcalls (voice or whistling), as they cannot bestandardized, either across observers orover time.

• Playback recordings, if used, should be asclear and loud as possible withoutdistortion. Digital technology isrecommended (CD-ROM, solid state, ordigital tape) as the sound quality can bebetter controlled and is less likely todeteriorate over time. If cassette tapes areused, they should be replaced periodicallyto avoid deterioration of the tape. Theaudio equipment should be of sufficientquality that it will not distort the sound atloud volumes. We suggest the volume besuch that the recording can be heard at400m, but not at 800m (to minimize bias at

the next survey station due to owls hearingthe recording from the previous station). Ifpossible, the volume should be measured ata standard distance (e.g., 1m from thespeakers) using a decibel meter.

• If playback is used, a recording should beused that includes all of the playbacksequences and the silent listening periods.A soft ‘beep’ or other sound can be used toindicate the start of the first silent listeningperiod, and another beep to indicate theend of the final listening period. This willensure that the time is fully standardized ateach station, and reduce the need forparticipants to keep checking theirwatches. If a cassette tape is used, the tapelength should match the recording length,and the same recording put on both sides,so the tape can be flipped instead ofrewound.

Time of NightSurveys should be conducted between a halfhour after sunset and midnight. An attemptshould be made to conduct the survey at thesame time of night each year.

• Owl call rates can change significantlyduring the night (Palmer 1987, Takats andHolroyd 1997). Call rates of at least somespecies tend to be lowest in the middle ofthe night (midnight to 04:00) and resumeagain early in the morning (Takats 1998a).However, few volunteers are prepared to

GORDON COURT

11


complete a survey before dawn. As such, werecommend surveying routes in the evening.

Environmental ConditionsEnvironmental conditions such as windvelocity, precipitation, and temperature canaffect owl calling propensity and the ability ofsurveyors to detect owls (Fuller and Mosher1987, Takats 1998a). Song carries fartherduring certain meteorological conditions. Forexample, audibility can be high when a lowaltitude inversion is present, since the soundwaves are reflected downwards, but audibilitycan be low in unstable or windy conditions(Elkins 1983). Extremely cold or stormyweather also poses a safety risk for surveyors.

• Surveys should only be conducted underfavourable conditions: wind speeds <20 kmper hour (Beaufort 3 or less; Appendix III[page 22]) and no precipitation (includingrain and/or snow). Temperatures should beclose to the average for the season andefforts should be made to avoid extremelycold temperatures for reasons of volunteersafety and because of evidence that owlsmay be less vocal in very cold weather(Takats 1998a).

• If conditions deteriorate over the course ofan evening, surveyors must use theirjudgement whether the route should becompleted, or run again on anotherevening. Generally, light snow or drizzlestarting in the middle of a survey would notprevent completion of the survey.

• Some researchers have found moon phase tohave a significant correlation with owl callrates (Takats 1998a), however other studieshave found no clear relationship. Therefore,to avoid limiting the window of availablesurvey dates (and forcing changes in thedates from year to year) it is best not torestrict surveys to certain phases of themoon. Moon phase may be useful as acovariate, but does not need to be recordedas it is easily calculated from the date in acalendar or almanac.

• Cloud cover has not been found to have alarge effect on call rate (Takats 1998a), but

as it could, for example, interact with moonphase, we suggest that surveyors record it(as percentage of sky that is covered bycloud).

• Data on weather conditions should berecorded at the beginning and end of eachsurvey, and preferably at every station, sothat weather variables can be used ascovariates to reduce variance in countindices, or so that data from selectedstations can be excluded from certain typesof analyses if conditions exceeded thethresholds.

Owl PositionsSurveyors should be asked to estimate theapproximate direction and distance to the firstposition where they detect each owl. Whenpossible they should also plot estimatedlocations on maps provided (useful forchecking the estimates). Actual distances canbe estimated (possibly indicating anuncertainty ±50 m, ±100 m) or distances can begrouped into categories (i.e. 0-100 m, 100-300m, 300+ m). Distance/direction informationcan be helpful for several purposes.

• The location may help to determine whetherthe same owls are being detected atdifferent stations along the route.

• More precisehabitatmodeling canbe conducted,provided thatthe stationsthemselves aregeoreferenced.

• Distanceinformation can beused to adjust forsome of thevariation indetection rates,especially observervariation, usingdistance samplingmethods.

NORTHERN HAWK OWL • GORDON COURT

12

Volunteers may vary in ability to hear theowls; however it is important that they areable to identify what they are hearing(Sauer et al. 1994)

Surveyors must be able to identify, byvocalizations, all species of owls thatregularly occur in their area. Although manypeople may be familiar with the mostcommon vocalizations of each species, owlsmay give variants on their calls, and somespecies such as Long-eared Owls have a widevariety of vocalizations. As a result,participants shouldhave atraining tapeor compactdisk that

S U R V E Y O R T R A I N I N G

includes the vocalizations of all of theowl species likely to be encountered,preferably with some narrative oraccompanying text that highlights themajor distinguishing features. Similarsounds that might be encountered atthe same time of year and that could beconfused with an owl (for example, thewinnowing of snipe could be confusedwith the call of a Boreal Owl) could beincluded.

Surveyors must be able to determinewhere they are on a map, so that theycan report accurately the location oftheir route. Staking out the stops duringthe daytime prior to conducting thesurvey is recommended, and is useful toensure that the route goes throughsuitable habitat and is safe. Thevolunteer must also have a goodenough sense of direction to determinethe direction of any calling owls, eitherfrom a map, from the stars, or from acompass.


Owl surveys lend themselves to being run by volunteers, because relatively little

experience is required. The volunteers should be able to meet two basic requirements:

ability to identify a small suite of owl species, and ability to keep track of where they

are.

BOREAL OWL • GORDON COURT

13

We recommend that all data should becomputerized in a relational data baseformat. In the future, it may be possible todevelop a common database program thatcan be used for all surveys, but in themeantime, if a database is developed withthe following structure, it should be relativelyeasy to share data and convert them into acommon format for analysis.

The main tables would be as follows:

1/ ROUTE table (information on each routethat does not change with time)

• route identifier (number)

• route name

• nearest town and other locationinformation

• coordinates of first station (preferable inLat/Long, although if UTM is used then itis important to indicate datum and gridzone); if coordinates are available for allstations on the route, these should beput in a separate table

2/ SURVEY table (information specific toeach time the route is surveyed)

• route number (link to previous table)

• surveyor ID number (this should link to aSurveyor table that includes name andaddress information for each surveyor orassistant)

• surveyor’s assistant ID number

• date of survey

• start and finish time of survey

• weather information at start and end ofsurvey, if not recorded at individualstations (temperature, cloud cover,precipitation)

• broadcast equipment information (ifused) – make and model of equipment

3/ STATIONS table (conditions at eachstation that was surveyed)

• route number & survey date (link toprevious table)

• odometer reading (not needed ifcoordinates available for each station)

• start time at station

• wind conditions (Beaufort scale) atstation (see Appendix III on page 22)

• codes for background noise (vehicles,running water, machinery, frogs)

• comments (optionally computerized toinclude other species recorded, habitatnotes, etc.)

4/ OWLS table (one record for eachindividual owl detected)

• route number, survey date, station (linkto previous table)

• owl species – four letter code (seeAppendix II on page 21)

• owl number (if more than one of aspecies at that station)

• intervals when the owl was detected (ifthere are multiple intervals because of

D A T A C O M P U T E R I Z A T I O N


Data from the survey are only useful if they are efficiently stored and subsequently

analyzed. To analyze trends at a national or continental level, all data must be

computerized in a compatible format, so that they can be combined for analysis.

14


additional silent listening period orbroadcasts)

• estimated distance to owl (metres oryards)

• direction to owl (in degrees, if necessaryconverted from N, NW, W, SW, S, SE, E,NE)

• comments (optional, i.e. indication ofsame owl from previous station, etc.)

Obviously, the field data sheets must bedesigned to ensure that the appropriate dataare collected. A sample data sheet is provided

in Appendix VI (page 24), but variations onthat theme may be required depending uponthe protocol adopted (e.g., with or withoutplayback). Even if coordinates are determinedby GPS, a detailed map of the stops should beprepared, so that the coordinates can bechecked, and so that this will be available if anew volunteer surveys the route in the future.Preparation of this map should be part of theroute selection procedure.


15

Data from the surveys described here aresimilar to those of the Breeding Bird Survey,and similar analysis methods may beappropriate, though some modifications maybe required in the future. A wide variety ofmethods have been developed for analysis ofBBS data (James et al. 1996, Link and Sauer1994, 1998), but there is still somedisagreement as to which methods are best(James et al. 1996, Link and Sauer 1994a, Linkand Sauer 1994b, Thomas 1996). There aretwo main methods currently being used bythe coordinators of the BBS. One involvesroute regression using estimating equations(Link and Sauer 1994), which assumes thattrends may differ among routes, andcalculates a weighted mean of the trendswithin routes. The selection of weightingfactors is strongly dependent upon thesampling scheme used to select routes. Analternate approach involves a generalizedlinear model assuming over-dispersed Poissonresiduals and a log-link function (Link andSauer 1998). This approach assumes thattrends are similar within a broader region,and allows more robust modelling of non-linear population changes (e.g., year to year

D A T A A N A L Y S I S

fluctuations). A simplified version of thislatter approach has been used for analysis ofpopulation trends in Ontario (Lepage andFrancis 1998, Francis and Whittam 2000), butit is not yet known whether this is the mostappropriate analysis method.

Finally, we recommend that relevant data bemade publicly available, preferably over theInternet. This will encourage volunteers byallowing them to see their resultsimmediately, and will encourage furtherresearch into analysis methods, thus ensuringthat maximum use is made of the data forconservation purposes. The data are collectedlargely by volunteers, and therefore shouldbe viewed as publicly available data.However, care should be taken to protectsensitive information, such as precise nestinglocations of rare species.

BOREAL OWL • GORDON COURT

To monitor owl populations, data must be analysed regularly, and the results made

available to the public, to managers, and especially to participants to encourage them

to participate in the survey. Regular correspondence at least twice a year is desirable to

maintain their interest.

16

Nevertheless, our knowledge of owls and owlbehaviour, as well as monitoring techniques,is far from complete, and it is quite likely thatas further research is done it will be possibleto develop new and better methods formonitoring owls in the future.

Some areas where further research would beuseful are as follows:

• Route selection: developing improvedmethods to select routes and/or analysedata to reduce bias due to non-random selection (includingroadside biases), whileremaining practical forsurveyors.

F U R T H E R R E S E A R C H N E E D S

• Observer effects: finding ways to reduce orcorrect for variation among observers,possibly through estimating theproportions of birds being detected.

• Playback: evaluating whether additionalplayback or silent listening periods mightimprove counts, especially for species orgeographic areas that have not yet beenwell studied.

• Survey methods: research optimal methodsfor documenting specific owl species or

owl species distribution, especiallyin geographic areas not yet

studied.

16

GREAT HORNED OWL • GORDON COURT

The methods recommended in this report are based upon the best information

available to us during compilation of this report. We believe they provide a sound

basis for developing owl monitoring protocols that will provide comparable data

across North America.

LONG-EARED OWL• GORDON COURT

17

Allen, A.W. 1987. Habitat suitability index models:barred owl. USDI, Fish and Wildlife Service BiologicalReport 82. 17 pp.

Anderson, D.R., J.L. Laake, B.R. Crain, and K.P.Burnham. 1979. Guidelines for line transect samplingof biological populations. Journal of WildlifeManagement 43(1): 70-78.

Beck, J. and B. Beck. 1988. The first ever greaterEdmonton owl prowl. Edmonton Naturalist 16(3): 6-20.

Bibby, C.J, N.D. Burgess, and D.A. Hill. 1992. Bird censustechniques. Academic Press Inc., San Diego, CA. 257pp.

Bondrup-Nielsen, S. 1978. Vocalizations, nesting andhabitat preferences of the boreal owl (Aegoliusfunereus) in North America. MSc. Thesis, Universityof Toronto, Ontario. 158 pp.

Bosakowski, T. 1994. Landsat reveals negative effect offorest fragmentation on barred owl distribution.Records of New Jersey Birds 20: 66-70.

Buckland, S. T., D. R. Anderson, K. P. Burnham, and J. P.Laake. 1993. Distance sampling: estimatingabundance of biological populations. Chapman andHall, London, U.K.

Downes, C., E.H. Dunn, and C.M. Francis. 1999.Canadian landbird monitoring strategy: Monitoringneeds and priorities into the new millennium.Partners in Flight – Canada, Ottawa. 34 pp.

Duncan, P.A. and J.R. Duncan. 1991. Nocturnal audio/playback owl survey of southeastern Manitoba andadjacent Minnesota, 1991 progress report. ManitobaDepartment of Natural Resources, Winnipeg,Manitoba. 8 pp.

Duncan, J.R. and P.A. Duncan. 1997. Increase indistribution records of owl species in Manitobabased on a volunteer nocturnal surveys using borealowl (Aegolius funereus) and great gray owl (Strixnebulosa) playback. In Biology and ConservationOwls of the Northern Hemisphere by J.R. Duncan,D.H.Johnson, and T.H. Nicholls (eds.). USDA ForestService General Technical Report NC-190. Pp. 519-524.

Duncan, J.R. and P.A. Duncan. 1998. Manitoba’snocturnal owl survey: 1998 progress report.Manitoba Conservation, Winnipeg, Manitoba. 8 pp.

Duncan, J.R. and A.E. Kearns. 1997. Habitat associationwith barred owl (Strix varia) locations insoutheastern Manitoba: a review of a habitat model.

In Biology and Conservation Owls of the NorthernHemisphere by J.R. Duncan, D.H.Johnson, and T.H.Nicholls (eds.). USDA Forest Service GeneralTechnical Report NC-190. Pp. 138-147.

Dunn, E.H. 1999. Recommendations for banding ofnocturnally migrating owls for the purpose ofmonitoring populations change. Draft for CanadianWildlife Service, Hull, Quebec.

Elkins, N. 1983. Weather and bird behaviour. T and ADPoyser, Carlton, UK. 239 pp.

Francis, C.M., and B. Whittam. 2000. Ontarionocturnal owl survey 1999 pilot study final report.unpublished report by Bird Studies Canada for theWildlife Assessment Program, Ontario Ministry ofNatural Resources. 34 pp.

Fuller, M.R. and J.A. Mosher. 1981. Methods ofdetecting and counting raptors: a review. In:Estimating Numbers of Terrestrial Birds by C.J. Ralphand J.M. Scott (eds.). Studies in Avian Biology 6. Pp.235-246.

Fuller, M.R. and J.A. Mosher. 1987. Raptor surveytechniques. In: Raptor Research ManagementTechniques Manual by B.A. Giron Pendleton, B.A.Millsap, K.W. Cline, and D.M. Bird (eds.). NationalWildlife Federation, Washington, D.C. Pp. 37-65.

Fyfe, R.W., S. Temple, and T.J. Cade. 1976. The 1975North American peregrine falcon survey. CanadianField-Naturalist 90: 228-273.

Gutierrez, R.J., D.M. Solis, and C. Sisco. 1984. Habitatecology of the spotted owl in northwesternCalifornia: implications of management.Proceedings of the Society of American Foresters1983. Pp. 368-373.

Gutierrez, R.J. and A.B. Carey. 1984. Ecology andmanagement of the spotted owl in the PacificNorthwest. USDA, Forest Service, General TechnicalReport PNW-185.

Holroyd, G.L. and L. Takats. 1997. Report on theNocturnal Raptor Monitoring Workshop. In: Biologyand Conservations of Owls of the NorthernHemisphere by J.R. Duncan, D.H.Johnson, and T.H.Nicholls (eds.). USDA Forest Service GeneralTechnical Report NC-190. Pp. 609-611.

James, P.C., K.M. Mazur, and S.D. Frith. 1995. Thebarred owl as an indicator of old forest and its rolein long-term forestry planning. Annual Report,Saskatchewan Environment and ResourceManagement. 27 pp.

L I T E R A T U R E C I T E D

NORTHERN SAW-WHET OWL • LISA TAKATS

18


James, F. C., C. E. McCulloch, and D. A. Wiedenfeld.1996. New approaches to the analysis of populationtrends in land birds. Ecology 77:13-27.

Johnson, B.T., B.T. Brown, L.T. Haight, and J.M.Simpson. 1981. Playback recordings as a specialavian censusing technique. In: Estimating Numberof Terrestrial Birds by C.J. Ralph and J.M. Scott, eds.Studies in Avian Biology 6: 68-75.

Johnson, D.H. 1987. Barred owls and nest boxes:results of a five-year study in Minnesota. In: Biologyand Conservation of Northern Forest Owls by R.W.Nero, R.J. Clark, R.J. Knapton, and R. H. Hamre(eds.). USDA Forest Service Technical Report RM-142. Pp. 129-134.

Johnsgard, P.A. 1988. North American owls, biologyand natural history. Smithsonian Institution,Washington, D.C. 295 pp.

Laidig, K.J. and D.S. Dobkin. 1995. Spatial overlap andhabitat associations of barred owls and greathorned owls in southern New Jersey. Journal RaptorResearch 29: 151-157.

Lehmkuhl, J.F. and M.G. Raphael. 1993. Habitatpattern around northern spotted owl locations onthe Olympic Peninsula, Washington. Journal ofWildlife Management 57: 302-315.

Lepage, D., C.M. Francis, and V. Deschamps. 1999.Ontario nocturnal owl survey 1998 pilot study finalreport. Report by Bird Studies Canada for OntarioMinistry of Natural Resources Wildlife AssessmentProgram. WAP-99-01. 21 pp.

Link, W.A., and J.R. Sauer. 1994a. Estimatingequations estimates of trends. Bird Populations2:23-32.

Link, W.A., and J.R. Sauer. 1994b. «New approaches tothe analysis of population trends in land birds»: acomment on statistical methods. Ecology 78:2632-2634.

Link, W.A., and J.R. Sauer. 1998. Estimatingpopulation change from count data: application tothe North American Breeding Bird Survey.Ecological Applications 8:258–268.

Mazur, K.M. 1997. Spatial habitat selection by barredowls (Strix varia) in the boreal forest ofSaskatchewan, Canada. MSc. Thesis, University ofRegina, Regina, Saskatchewan. 85 pp.

McGarigal, K. and J.D. Fraser. 1985. Barred owlresponses to recorded vocalizations. Condor 87:552-553.

Mosher, J.A. and M.R. Fuller. 1996. Surveyingwoodland hawks with broadcasts of great hornedowl vocalizations. Wildlife Society Bulletin 24: 531-536.

Mosher, J.A., M.R. Fuller, and M. Kopeny. 1990.Surveying woodland raptors by broadcast ofconspecific vocalzations. Journal of FieldOrnithology 61: 453-461.

Newton, I. 1979. Population ecology of raptors. ButeoBooks, Vermillion, South Dakota. Pp. 13-17.

Nichols, J. D., J. E. Hines, J. R. Sauer, F. W. Fallon, J. E.Fallon, and P. J. Heglund. 2000. A double-observerapproach for estimating detection probability andabundance from point counts. Auk 117: 393-408.

Noble, D.G., J.E. Elliott, and J.L. Shutt. 1993.Environmental contaminants in Canadian raptors.Technical Report Series No. 91. Canadian WildlifeService, Environment Canada, Ottawa, ON.

Oliphant, L. 1994. A report on results of national andregional ornithological surveys in Canada. Bird Trends4:1.

Palmer, D.A. 1987. Annual, seasonal, and nightlyvariation in calling activity of boreal and northernsaw-whet owls. In Biology and Conservation ofNorthern Forest Owls by R. W. Nero, R.J. Clark, R.J.Knapton, and R.H. Hamre (eds.). USDA Forest ServiceTechnical Report RM-142. Pp. 162-168.

Pendleton, B.A. Giron, B.A. Millsap, K.W. Cline, andD.M. Bird (eds.). 1987. Raptor managementtechniques manual. National Wildlife Federation,Washington, DC. 420 pp.

Proudfoot, G.A., J.L. Mays, S.L. Beasom, and R. Bingham.1997. Effectiveness of broadcast surveys indetermining habitat use of ferruginous pygmy-owls(Glaucidium brasilianum) in southern Texas. In:Biology and Conservation of Owls of the NorthernHemisphere by J.R. Duncan, D.H.Johnson, and T.H.Nicholls (eds.). USDA Forest Service General TechnicalReport NC-190. P. 338.

Sauer, J. R., B. G. Peterjohn, and W. A. Link. 1994.Observer differences in the North American BreedingBird Survey. Auk 111:50-62.

Smith D.G. 1987. Owl census techniques. In: RaptorResearch Management Techniques Manual by B.A.Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M.Bird (eds.). National Wildlife Federation, Washington,D.C.

Takats, D.L. 1998a. Barred owl habitat use anddistribution in the Foothills Model Forest. MSc. Thesis,Department of Renewable Resources, University ofAlberta, Edmonton, Alberta. 139 pp.

Takats, D.L. 1998b. Volunteer nocturnal owl surveys inAlberta, annual report. Beaverhill Bird Observatory,Edmonton, Alberta. 9 pp.

Takats, D.L. and G.L. Holroyd. 1997. Owl broadcastsurveys in the Foothills Model Forest. In: Biology andConservations of Owls of the Northern Hemisphere byJ.R. Duncan, D.H.Johnson, and T.H. Nicholls (eds.).USDA Forest Service General Technical Report NC-190.Pp. 421-431.

Thomas, L. 1996. Monitoring long-term populationchange: why are there so many analysis methods?Ecology 77: 49-58.

Wellicome, T. 1997. Status of the burrowing owl(Speotyto cunucularia hypugaea) in Alberta. AlbertaEnvironmental Protection, Wildlife ManagementDivision, Wildlife Status Report No. 11. 21 pp.

Whelan, P. 1996. Swainson’s hawk still in trouble. RaptorWatch 10(2): 1. Hawk Watch International.

19

APPENDIX I : RECOMMENDED EQUIPMENT

The following is a list of equipment that volunteer surveyors are likely to need to

conduct the nocturnal owl survey. Individual survey protocols should include a similar

list.

Equipment provided by the coordinating group:

Instruction Booklet This should detail why the survey is being conducted and describe thedetails of the protocol.

Training Tape/CD The tape should include the calls of all species of owls the surveyor is likelyto encounter, and even others that may not be expected. As well, it shouldinclude calls of other species of animals the surveyor is likely to hear (frogsand toads, snipe, woodcock, ruffed grouse). This can help the surveyordifferentiate between similar sounding species, but can also be used tocollect information on other species of interest.

Playback Tape/CD (if the protocol involves broadcasts of owl calls). Recording should bedigital, if possible, and should include the silent listening periods as well.If a tape is used, the same recording should be on both sides of the tape,so it does not need to be rewound.

Data Forms Data sheets should be simple and easy to use, but also readilycomputerized (see sample in Appendix).

Route Map Volunteers should be provided with a suitable scale map showing theroute location, on which they should mark the exact location of eachstation. They can also mark where they estimate each owl they hear on acopy of the map. The map should be included with the data forms whensubmitted.

Tax Relief Form When a non-profit group with charitable status is running the surveys, itmay be possible to provide volunteer surveyors with a tax receipt for theirout-of-pocket expenses while running the route. This could be usefulincentive to encourage more to participants. In Canada, in-kind donationscannot be receipted, so it is necessary for the volunteer to submit anexpense claim with a record of their food, mileage, and other expenses tothe organization. The organization then reimburses the expenses, whichthe volunteer then donates back to the organization to receive a taxreceipt (in practice, the organization will usually want to receive thereturn donation (post-dated if necessary) before issuing thereimbursement cheque).

Volunteer Form Many volunteer programs can be run under the auspices of anorganization that can cover individuals for General Liability, AccidentalDeath and Dismemberment. There are significant risks associated withtravelling, and standing at the side of, remote roads at night in late winter(e.g. winter storms, vehicle breakdowns, etc.). Because there is always thepossibility of an injury some coverage for volunteers is encouraged.

(optional)

20


APPENDIX I : RECOMMENDED EQUIPMENT. . .CONTINUED

Equipment provided by the surveyor:

Warm clothes Volunteers need to be reminded to bring extra warm clothes, even on awarm night, especially in case of changes in weather, storms, or vehiclebreakdown. Also, standing outside, without moving, listening for owlscan get quite cold. Several layers of warm clothes are advisable, includinga weather-proof jacket, wool or fleece clothes, mittens/gloves, a hat, awarm coat and boots.

Playback Unit (if playback is part of protocol). Any tape or CD player should meetprotocol specifications in terms of broadcast volume, but should not betoo heavy to handle easily. It can run on batteries or on a cord that plugsinto the car cigarette lighter (cord should be long enough to reachoutside the vehicle). Volunteers using batteries should bring extras. Atowel can be placed underneath the unit to avoid scratching the vehicle.

Flashlight A safety item, in case of car trouble and also useful in case any thing isdropped outside the vehicle. A strong flashlight can be used to observean owl that has flown in, in response to playback, though we do notrecommend repeatedly scanning for owls, as this could scare them.

Thermometer A small thermometer to record the temperature during the survey.

Compass To determine directions to calling owls, especially if the stars areobscured by clouds, or the road is curving, or it is otherwise difficult todetermine orientation.

Watch/Clock A watch or clock with a second hand to time the listening period, if abroadcast recording is not used. An illuminated digital stopwatch wouldbe ideal, as it must also be easy to see in the dark.

Pen/Pencil Pencils are more reliable than pens in very cold temperatures, but mustbe dark enough to ensure legible writing on the form. If using a pen,ensure it is waterproof, in case the data sheets get wet (snow/water).

Reliable Vehicle/ Volunteers should be reminded that the best safety precaution is areliable vehicle that has been regularly maintained. A spare tire, withtools for changing tires, jumper cables, and road flares should always becarried. Standard safety items for the survey (or any other time duringwinter travel) include a candle, an empty can and matches (for heat andlight), a flashlight, a blanket or warm sleeping bag, a first aid kit, athermos of hot drinks, and some snacks. Volunteers should beencouraged to survey with somebody else, and to report their travelplans to somebody else.

Safety Supplies

21

365.0 Barn Owl Tyto alba BNOW

366.0 Long-eared Owl Asio otus LEOW

367.0 Short-eared Owl Asio flammeus SEOW

368.0 Barred Owl Strix varia BARR*

368.6 Spotted x Barred Owl Hybrid Strix occidentalis x Strix varia SBOH

369.0 Spotted Owl Strix occidentalis SPOW

370.0 Great Gray Owl Strix nebulosa GGOW

371.0 Boreal Owl Aegolius funereus BOOW

372.0 Northern Saw-whet Owl Aegolius acadicus NSWO

373.0 Eastern Screech-owl Otus asio EASO

373.1 Whiskered Screech-owl Otus trichopsis WHSO

373.2 Western Screech-owl Otus kennicottii WESO

374.0 Flammulated Owl Otus flammeolus FLOW

375.0 Great Horned Owl Bubo virginianus GHOW

376.0 Snowy Owl Nyctea scandiaca SNOW

377.0 Northern Hawk Owl Surnia ulula NHOW

378.0 Burrowing Owl Athene cunicularia BUOW

379.0 Northern Pygmy-owl Glaucidium gnoma NOPO

380.0 Ferruginous Pygmy-owl Glaucidium brasilianum FEPO

381.0 Elf Owl Micrathene whitneyi ELOW

Species # Common Name Scientific Name Code

* These are derived from the standard bird-banding codes, except for Barred Owl which should have thecode BDOW (BAOW conflicts with Barn Owl). When hand-written BDOW could easily be confused withBOOW, so we recommend use of BARR instead.

APPENDIX II: SCIENTIFIC NAMES AND CODES OF NORTH AMERICAN OWLS

22

0 < 2 (< 1) Smoke rises vertically

1 2 to 5 (1 to 3) Wind direction shown by smoke drift

2 6 to 12 (4 to 7) Wind felt on face, leaves rustle

3 13 to 19 (8 to 12) Leaves, small twigs in constant motion

4 20 to 29 (13 to 18) Raises dust/loose paper, small branches move

5 30 to 38 (19 to 24) Small trees in leaf sway

Beaufort # Wind Speed in km/hr (mph) Indicators of Wind Speed

APPENDIX I I I : BEAUFORT SCALE TRANSLATIONS TO WIND SPEEDS

1 Quiet

2 Some noise, but not distracting (dogs or coyotes barking/howling)

3 Significant noise that may have reduced owl detectability (ie. creek)

4 Constant noise (ie. heavy traffic, compressor station, roaring creek)

Noise Level Description

APPENDIX IV: NOISE LEVEL DESCRIPTIONS

23

APPENDIX V: RANDOM ROUTE SELECTION

A variety of methods could be used for randomizing route selection, to ensure that routes are asrepresentative as possible within the constraints of a roadside survey. One approach would be touse the same protocol as the Breeding Bird Survey (USGS). However, in regions where large sectionsof the landscape may not be suitable for owls (e.g., agricultural fields), it may be more efficient tomodify this approach to consider only areas with suitable habitat. Here we outline one approachthat may be able to achieve this, though we caution that it has not yet been widely used, and somemodifications may be necessary for implementation in a particular region.

• The survey area should be divided intorelatively large units, such as 1-degreeblocks, or 100 x 100km UTM blocks. Withineach block, a starting point should beselected randomly, by picking two randomnumbers (from a random number table, orusing a computer/calculator) to representcoordinates within that block (UTM orlatitude-longitude). Using a map showingpotentially suitable roads and habitatswithin the area, move from the randomstarting point to the nearest point on asuitable road, within suitable habitat, anduse that as the starting point of the route.Suitable roads must be accessible in latewinter/early spring and should not haveexcessive traffic or heavy logging trucks (forsafety reasons and so that owls can beheard). Select another random number toindicate the direction of travel on the road.If the road is not long enough in theselected direction, or reverts to unsuitablehabitat, take turns onto other roads ifnecessary, or move the starting pointbackwards along the road until a route canbe accommodated. Ideally, the whole routeshould remain within the sampling block,although allowing a limited portion of theroute to extend into the next block (e.g.<25%) could be allowed. If the selectedroad is not suitable (e.g. too short evenafter working both ways) then the nextnearest road to the sampling point should

be selected. If there are no suitable roadswithin a reasonable distance of therandomly selected point (e.g. 10 km) then anew random point should be selected.

• If insufficient information is available tosurvey organizers on the distribution ofsuitable habitat or roads, then volunteerscould assist with the route selection (e.g.,survey organizers could provide therandom starting point, and volunteerscould select the route using the same set ofrules). Survey organizers must work closelywith the volunteers to ensure that theyunderstand and follow the rules properly.

• Efforts should be made to ensure thatroutes are selected from as many blocks aspossible, to cover as wide an area aspossible with the survey.

• The definition of suitable habitat should besufficiently broad to cover the range ofhabitats used by any of the target speciesin the region, and not restricted to the besthabitat. This type of habitat-basedsampling scheme must be accompanied bya habitat-monitoring program to ensurethat changes in the extent of habitat (e.g.,loss of habitat, or regrowth of new habitat)are detected.

24

OP

TI

ON

AL

APPENDIX VI: SAMPLE DATA SHEET

Species Owl NumberDistance/Direction

DuringFirst Minute

DuringSecond Minute

AfterBroadcast Noise Level Comments

STATION: 1 ODOMETER: km/mile START TIME: WIND: 0 1 2 3 >3

Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4

Route Number: Route Name:

Surveyor: Assistant(s):

Date: Do you wish to participate again next year?

Temperature:

Precipitation:

START END

[ ] YES [ ] NO

[ ] ˚C [ ] ˚F Cloud Cover (%): START END% %

[ ] NONE [ ] LIGHT [ ] MEDIUM

Snow Cover: [ ] NONE [ ] PATCHY [ ] CONTINUOUS Max. Depth: [ ] CM [ ] INMin. Depth:

[ ] SNOW [ ] RAIN/

/

/

D A T A S H E E T

D AY M O N T H Y E A R/ /

Continued for 10 Stations per form.


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4

Route Number: Route Name:

Surveyor: Assistant(s):

Date: Do you wish to participate again next year?

Temperature:

Precipitation:

START END

[ ] YES [ ] NO

[ ] ˚C [ ] ˚F Cloud Cover (%): START END% %

[ ] NONE [ ] LIGHT [ ] MEDIUM

Snow Cover: [ ] NONE [ ] PATCHY [ ] CONTINUOUS Max. Depth: [ ] CM [ ] INMin. Depth:

[ ] SNOW [ ] RAIN/

/

/

D A T A S H E E T

D AY M O N T H Y E A R/ /


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4


DuringFirst Minute

DuringSecond Minute



Traffic Count

1 23 4

FRONT/BACK COVER PHOTOS:GREAT HORNED OWLS • GORDON COURT

GUIDELINES FOR NOCTURNAL OWL MONITORINGIN NORTH AMERICA

M A R C H 2 0 0 1

Documents

Owl