![Fall migration routes of Platform Transmitting Terminal (PTT)-tagged MCP sandhill cranes of the 4 breeding affiliations: (A) West-central Canada–Alaska [WC–A]; (B) East-central Canada–Minnesota [EC–M]; (C) Western Alaska–Siberia [WA–S]; (D) Northern Canada–Nunavut [NC–N] based on their PTT-locations during fall 1998–2003. Closed circles with white rings represent breeding locations for cranes of each breeding affiliation.](https://www.researchgate.net/profile/David-Brandt-6/publication/229464327/figure/fig4/AS:267848941568018@1440871572477/Fall-migration-routes-of-Platform-Transmitting-Terminal-PTT-tagged-MCP-sandhill-cranes_Q320.jpg)
Content uploaded by David A. Brandt
Author content
All content in this area was uploaded by David A. Brandt on May 11, 2018
Content may be subject to copyright.
Geographic Distribution of the Mid-Continent
Population of Sandhill Cranes and Related
Management Applications
GARY L. KRAPU,
1
U.S. Geological Survey, Northern Prairie Wildlife Research Center, 8711 37th Street SE, Jamestown, ND 58401, USA
DAVID A. BRANDT, U.S. Geological Survey, Northern Prairie Wildlife Research Center, 8711 37th Street SE, Jamestown, ND 58401, USA
KENNETH L. JONES,
2
Department of Biological Sciences (M/C 066), University of Illinois at Chicago, Chicago, IL 60607, USA
DOUGLAS H. JOHNSON, U.S. Geological Survey, Northern Prairie Wildlife Research Center, 8711 37th Street SE, Jamestown, ND 58401, USA
ABSTRACT The Mid-continent Population (MCP) of sandhill cranes (Grus canadensis) is widely hunted in North America and is separated into
the Gulf Coast Subpopulation and Western Subpopulation for management purposes. Effective harvest management of the MCP requires detailed
knowledge of breeding distribution of subspecies and subpopulations, chronology of their use of fall staging areas and wintering grounds, and exposure to
and harvest from hunting. To address these information needs, we tagged 153 sandhill cranes with Platform Transmitting Terminals (PTTs) during 22
February–12 April 1998–2003 in the Central and North Platte River valleys of south-central Nebraska. We monitored PTT-tagged sandhill cranes,
hereafter tagged cranes, from their arrival to departure from breeding grounds, during their fall migration, and throughout winter using the Argos satellite
tracking system. The tracking effort yielded 74,041 useable locations over 49,350 tag days; median duration of tracking of individual cranes was 352 days
and 73 cranes were tracked >12 months. Genetic sequencing of mitochondrial DNA (mtDNA) from blood samples taken from each of our random
sample of tagged cranes indicated 64% were G. c. canadensis and 34% were Grus canadensis tabida. Tagged cranes during the breeding season settled in
northern temperate, subarctic, and arctic North America (U.S. [23%, n¼35], Canada [57%, n¼87]) and arctic regions of northeast Asia (Russia [20%,
n¼31]). Distribution of tagged cranes by breeding affiliation was as follows: Western Alaska–Siberia (WA–S, 42 4% [SE]), northern Canada–Nunavut
(NC–N, 21 4%), West-central Canada–Alaska (WC–A, 23 4%) and East-central Canada–Minnesota (EC–M, 14 3%). All tagged cranes returned
to the same breeding affiliation used during the previous year with a median distance of 1.60 km (range: 0.08–7.7km, n¼53) separating sites used in year 1
and year 2. Fall staging occurred primarily in central and western Saskatchewan (69%), North Dakota (16%), southwestern Manitoba (10%), and
northwestern Minnesota (3%). Space-use sharing indices showed that except for NC–N and WC–A birds, probability of finding a crane from one
breeding affiliation within the home range of another breeding affiliation was low during fall staging. Tagged cranes from WC–A and EC–M breeding
affiliations, on average, spent 25 and 20 days, respectively, longer on fall staging areas in the northern plains than did WA–S and NC–N birds. Cranes in
the NC–N, WA–S, and WC–A affiliations spent 99%, 74%, and 64%, respectively, of winter in western Texas in Hunting Zone A; EC–M cranes spent
83% of winter along the Texas Gulf Coast in Hunting Zone C. Tagged cranes that settled within the breeding range of the Gulf Coast Subpopulation
spent 28% and 42% of fall staging and winter within the range of the Western Subpopulation, indicating sufficient exchange of birds to potentially limit
effectiveness of MCP harvest management. Harvests of EC–M and WC–A cranes during 1998–2003 were disproportionately high to their estimated
numbers in the MCP, suggesting more conservative harvest strategies may be required for these subpopulations in the future, and for sandhill cranes to
occupy major parts of their historical breeding range in the Prairie Pothole Region. Exceptionally high philopatry of MCP cranes of all 4 subpopulations to
breeding sites coupled with strong linkages between crane breeding distribution, and fall staging areas and wintering grounds, provide managers guidance
for targeting MCP crane harvest to meet management goals. Sufficient temporal or spatial separation exists among the 4 subpopulations on fall staging
areas and wintering grounds to allow harvest to be targeted at the subpopulation level in all states and provinces (and most hunting zones within states and
provinces) when conditions warrant. Knowledge gained from our study provides decision-makers in the United States, Canada, Mexico, and Russia with
improved guidance for developing sound harvest regulations, focusing conservation efforts, and generating collaborative efforts among these nations on
sandhill crane research and management to meet mutually important goals. ß2011 The Wildlife Society.
KEY WORDS breeding affiliation, Central Flyway, Grus canadensis, harvest, Mid-continent Population (MCP), sandhill
crane, satellite telemetry, fall staging areas, subspecies, wintering grounds.
Distribucio
´n Geogra
´fica de la Poblacio
´n Centro-Continental
de la Grulla Canadiense y Aplicacio
´n de Gestiones
Relacionadas
RESUMEN La Poblacio
´n Centro-continental (MCP) de grulla canadiense (Grus canadensis) es cazada ampliamente en Norte Ame
´rica y, para
propo
´sitos de manejo, esta
´dividida en las Subpoblaciones de la Costa del Golfo y Oeste. Una gestio
´n efectiva de la especie requiere un conocimiento
detallado sobre la distribucio
´n de las a
´reas reproductivas de subespecies y subpoblaciones, la cronologı
´a del uso de las a
´reas de escala durante el oton
˜o y de las
a
´reas de invernada, ası
´como la exposicio
´n y presio
´n a la cacerı
´a. Para poder obtener estos conocimientos necesarios, marcamos 153 grullas canadienses con
transmisores PTT (Platform Transmitter Terminal) entre el 22 de febrero y el 12 de abril de 1998 a 2003 en los valles del Central Platte River y North Platte
River, localizados en la zona centro sur de Nebraska. Monitoreamos las grullas marcadas con transmisores PTT (en lo sucesivo grullas marcadas), desdesu
llegada hasta su partida a las zonas de reproduccio
´n, durante su migracio
´n de oton
˜o y durante todo el perı
´odo de invierno, utilizando el sistema de
seguimiento por sate
´lite ARGOS. Dicho seguimiento dio como resultado 74,041 localizaciones u
´tiles de un total de 49,350 dı
´as; la duracio
´n mediana de
seguimiento de individuos fue de 352 dı
´as y 73 grullas fueron rastreadas >12 meses. Las secuencias gene
´ticas de ADN Mitocondrial en sangre, tomadas a
Wildlife Monographs 175:1–38; 2011; DOI: 10.1002/wmon.1
Received: 26 December 2008; Accepted: 29 May 2010.
1
E-mail: gkrapu@usgs.gov
2
Present Address: Environmental Health Sciences, University of Georgia, 140 Environmental Health Sciences, 150 E. Green St., Athens, GA 30602, USA.
Krapu et al. Geographic Distribution of Sandhill Cranes 1
partir de muestras aleatorias de grullas marcadas, indicaron que el 64% eran G. c. canadensis y el 34% eran Grus canadensis tabida. Durante la estacio
´n
reproductiva, las grullas marcadas se establecieron en las regiones templadas, suba
´rticas, y a
´rticas de Norte Ame
´rica (Estados Unidos [23%, n¼35],
Canada
´[57%, n¼87]) ası
´como en las regiones a
´rticas del nordeste de Asia (Rusia [20%, n¼31]). Las grullas marcadas, pertenecientes a diferentes
afiliaciones reproductivas, se distribuyeron de la siguiente forma: Oeste de Alaska–Siberia (WA–S, 42 4% [SE]), norte de Canada
´–Nunavut (NC–N,
21 4%), centro oeste de Canada
´–Alaska (WC–A, 23 4%) y centro este de Canada
´–Minesota (EC–M, 14 3%). Todas las grullas marcadas regresaron
a la misma afiliacio
´n reproductiva del an
˜o anterior con una distancia mediana de 1.60 km (rango: 0.08–0.77 km, n¼53) de separacio
´n entre los sitios
utilizados en el an
˜o 1 y en el an
˜o 2. Las a
´reas de escala oton
˜al se concentraron principalmente en el centro y oeste de Saskatchewan (69%), Dakota del Norte
(16%), sudoeste de Manitota (10%) y noroeste de Minesota (3%). Los ı
´ndices de compartimiento de espacio indicaron que a excepcio
´n de las aves NC–N y
WC–A, la probabilidad de encontrar una grulla perteneciente a una afiliacio
´n reproductiva dentro del territorio de otra afiliacio
´n, era baja durante la escala
oton
˜al. Las aves WC–A y EC–M pasaron respectivamente como promedio 25 y 20 dı
´as ma
´s en las a
´reas de escala oton
˜al de las grandes planicies, que las
aves WA–S y NC–N. Las grullas pertenecientes a las afiliaciones NC–N, WA–S y WC–A pasaron el 99, 74 y 64% del invierno respectivamente, en el oeste
de Texas en la Zona de Caza A; las grullas EC–M pasaron el 83% del invierno a lo largo de la Costa del Golfo de Texas en la Zona de Caza C. Las grullas
marcadas que se asentaron en las a
´reas de reproduccio
´n de la Subpoblacio
´n de la Costa del Golfo, pasaron el 28% de la escala oton
˜al y el 42% del invierno
dentro del territorio de la Subpoblacio
´n Oeste, indicando un intercambio de individuos lo suficiente grande como para potencialmente limitar el manejo
efectivo de la MCP. La caza de aves EC–M y WC–A entre los an
˜os 1998 a 2003 fue desproporcionalmente alta en relacio
´n a los nu
´meros estimados en la
MCP, sugiriendo que se podrı
´an requerir estrategias de manejo ma
´s conservadoras para estas subpoblaciones en un futuro, y para que la grulla canadiense
pueda reocupar gran parte de su a
´rea reproductiva histo
´rica en la regio
´ndelPrairie Pothole. El hecho de que las 4 subpoblaciones de grullas pertenecientes a
la MCP exhibieron una excepcional filopatrı
´a a los sitios de reproduccio
´n y de que existieron fuertes vı
´nculos entre las distribuciones reproductivas de las
grullas, las a
´reas de escala oton
˜al y las a
´reas de invernada, ofrece una orientacio
´n a los manejadores de la caza de grullas de la MCP para que puedan cumplir
con las metas de gestio
´n establecidas. Existe suficiente separacio
´n temporal y/o espacial entre las 4 subpoblaciones en las a
´reas de escala oton
˜al y en las a
´reas
de invernada como para permitir la caza dirigida al nivel subpoblacional en todos los estados y provincias (y en la mayorı
´a de las zonas de caza dentro de los
estados y provincias) siempre y cuando las condiciones lo ameriten. El conocimiento obtenido a partir de este estudio proporciona a los responsables de
decisiones en los Estados Unidos, Canada
´,Me
´xico y Rusia, una mejor direccio
´n para desarrollar cuotas de caza razonables, enfocar los esfuerzos en ma teria
de conservacio
´n y generar colaboraciones entre estas naciones en materia de investigacio
´n y gestio
´n de la grulla canadiense, para poder cumplir metas
importantes y a la vez comunes.
Distribution Ge
´ographique de la Population des Grues du
Canada Dans le Centre du Continent et Les Applications
Relatives a
`Leur Gestion
SOMMAIRE La grue du Canada (Grus canadensis) est largement chasse
´e en Ame
´rique du Nord et elle se se
´pare entre la sous-population co
ˆtie
`re
du Golfe et la sous-population de l’Ouest afin de tel que de
´fini. Pour administrer la population des grues efficacement, il est ne
´cessaire d’avoir une
connaissance de
´taille
´edelare
´partition des colonies, des sous-espe
`ces et des sous-populations. Il faut aussi une table chronologique des divers points de
ravitaillements visite
´s par ces grues et leur exposition aux dangers de la chasse au cours de leur migration. Pour re
´pondre a
`ce besoin, nous avons e
´tiquete
´
cent cinquante-trois (153) grues du Canada avec des bornes de transmission plate-forme (VTP), entre le 22 fe
´vrier et le 12 avril des anne
´es 1998 et 2003, au
centre et au nord de la valle
´e de la Rivie
`re Platte dans le sud-central du Nebraska. Gra
ˆce au syste
`me de repe
´rage satellite ARGOS, nous avons contro
ˆle
´les
bornes de transmission (VTP) des grues du Canada e
´tiquete
´es, de leur arrive
´jusqu’a
`leur de
´part des zones de reproduction, pendant la migration
printanie
`re et tout au long de la pe
´riode hivernale. L’effort de repe
´rage a localise
´74,041 lieux utilise
´s sur une pe
´riode de 49,350 jours. Le suivi des grues
individuelles e
´tait d’une dure
´eme
´diane de 352 jours, dont 73 grues ont e
´te
´s repe
´re
´es sur une pe
´riode de plus de douze mois.
Le se
´quenc¸age ge
´ne
´tique de l’ADNmt a
`partir de certains e
´chantillons de sang pre
´leve
´sur des grues e
´tiquete
´es au hasard a de
´montre
´que 64% e
´taient des
G. c. canadensis et que 34% e
´taient des Grus canadensis tabida. Les grues e
´tiquete
´es durant la saison de reproduction se sont installe
´es majoritairement dans
les re
´gions tempe
´re
´es, subarctiques et arctiques de l’Ame
´rique du Nord (E
´tats-Unis [23%, n¼35], Canada [57%, n¼87]) ainsi que dans les re
´gions de
l’Arctique de l’Asie du Nord (Russie [20%, n¼31]). La re
´partition des grues e
´tiquete
´es lors de la reproduction a de
´montre
´les affiliations suivantes: Ouest
de l’Alaska–Sibe
´rie (OA–S, 42 4% [SE]), Nord du Canada–Nunavut (NC–N, 21 4%), Centre-Ouest du Canada–Alaska (COC–A, 23 4%) et
Centre-Est du Canada–Minnesota (CEC–M, 14 3%). Toutes les grues e
´tiquete
´es sont retourne
´es dans les me
ˆmes groupes de reproduction que les
anne
´es pre
´ce
´dentes, quoiqu’une distance me
´diane de 1.60 km (gamme: 0.08–7.7 km, n¼53) se
´parait les sites entre les deux anne
´es e
´tudie
´es.
Les relais automnaux des grues e
´tudie
´es sont principalement situe
´s dans le centre et l’ouest de la Saskatchewan (69%), au Dakota du Nord (16%), dans le
sud-ouest du Manitoba (10%) et dans le nord-ouest du Minnesota (3%). Les indices de partage d’espace ont de
´montre
´qu’a
`l’exception des grues du NC–N
et du COC–A, les probabilite
´s de trouver une grue dans le domaine vital d’une autre affiliation de reproduction que la sienne est tre
`s faible lors de la saison
d’accouplement. Les grues e
´tiquete
´es provenant des zones de reproduction du COC–A et du CEC–M passent, respectivement, en moyenne 25 et 20 jours
de plus en automne dans les plaines du nord que les oiseaux des re
´gions OA–S et du NC–N. Les grues se situant dans les affiliations du NC–C, de l’OA–A
et du COC–A ont respectivement passe
´99, 74, et 64% de leur hiver dans l’ouest du Texas, zone de chasse A. Les grues du CEC–M ont passe
´83% de leur
hiver le long de la re
´gion de la Co
ˆte du Golfe du Texas, dans la zone de chasse C. Les grues e
´tiquete
´es qui se sont installe
´es dans la zone d’accouplement de
la sous-population co
ˆtie
`re du Golfe, ont passe
´entre 28 et 42% de la saison d’accouplement et de la saison hivernale dans la re
´gion de la sous-population de
l’Ouest. Cela indique qu’un e
´change suffisant entre les deux populations de grues peut potentiellement limiter l’effet de la gestion de la reproduction.
L’e
´tude des grues du CEC–M et du COC–A en 1998 et 2003 a de
´montre
´une disproportion vis-a
`-vis du nombre estime
´de la PGCC, sugge
´rant que de
nouvelles strate
´gies conservatrices sont ne
´cessaire pour que les sous-populations de grue du Canada re
´occupent la majorite
´de leur territoire historique de la
re
´gion des cuvettes des prairies. Les quatre sous-populations de grues de la PGCC ressentent une philopatrie exceptionnelle entre leurs sites de
reproduction respectifs, leurs sites d’e
´levage, leurs lieux de rassemblement automnaux et leurs relais hivernaux. Cette philopatrie offre aux chasseurs une
cible incroyable pour ge
´rer et atteindre leurs objectifs quand vient la saison de la chasse. Parmi les quatre sous-populations, une se
´paration temporelle et
spatiale suffisante existe entre les aires de repos automnales et hivernales. Quand les conditions le permettent, cette se
´paration permet aux grues des sous-
populations d’e
ˆtre la cible des chasseurs durant la saison de chasse dans tous les e
´tats et toutes les provinces (et dans la plupart des zones de chasses). Les
connaissances acquises lors de cette e
´tude permettent aux E
´tats-Unis, au Canada, au Mexique et a
`la Russie de pouvoir ame
´liorer leurs lois en ce qui a
`trait a
`
2 Wildlife Monographs 175
la saison de la reproduction. Elle aide aussi ces pays a
`diriger leurs efforts sur la conservation des grues et de leur territoire. Finalement, elle ge
´ne
`re une
nouvelle collaboration entre ces quatre nations en matie
`re de recherche et de gestion sur les grues du Canada, pour re
´pondre a
`ces objectifs mutuels.
Krapu et al. Geographic Distribution of Sandhill Cranes 3
Contents
INTRODUCTION ................................................................................. 4
STUDY AREA ......................................................................................... 5
METHODS............................................................................................. 6
Trapping and Transmitter Deployment ................................................. 6
Data Processing and Analyses................................................................ 8
RESULTS.............................................................................................. 10
Geographic Distribution During Breeding Season ............................... 11
Philopatry and Chronology of Use of Breeding Grounds ...................... 12
Fall Migration Routes ......................................................................... 12
Chronology of Migration to Fall Staging Areas .................................... 13
Use of Fall Staging Areas..................................................................... 16
Chronology of Migration From Fall Staging Areas to Wintering
Grounds.............................................................................................. 19
Use of Wintering Grounds .................................................................. 20
Seasonal Exchange Between Gulf Coast and Western Subpopulations.... 21
Exposure to Hunting Seasons .............................................................. 22
MCP Harvest by Breeding Affiliation .................................................. 26
DISCUSSION ....................................................................................... 26
Extent of Breeding Distribution .......................................................... 26
Factors Influencing Breeding Distribution and Abundance .................. 27
Factors Influencing Fall Staging Distribution ...................................... 28
Factors Influencing Winter Distribution.............................................. 28
MCP Use of the Gulf Coast Management Unit .................................... 29
Range Overlap With Other Crane Populations .................................... 29
Factors Influencing Composition of Harvest by Breeding Affiliation .... 30
Major Risks to MCP From Habitat Change......................................... 30
MANAGEMENT IMPLICATIONS .................................................... 31
Targeting Harvest at Subpopulation Level........................................... 32
Opportunities for Increased International Collaboration...................... 33
SUMMARY ........................................................................................... 33
ACKNOWLEDGMENTS..................................................................... 33
LITERATURE CITED......................................................................... 34
APPENDICES ...................................................................................... 36
INTRODUCTION
Sport hunting of the Mid-continent Population (MCP; see
Appendix A for an index of all acronyms and abbreviations) of
sandhill cranes occurs in 11 states in the United States, 2 prov-
inces in Canada, and 3 states in Mexico (Sharp et al. 2007).
Hunting and harvest of the MCP is centered in the plains states
of North Dakota, Kansas, Oklahoma, Texas, and the prairie
provinces of Saskatchewan and Manitoba. In the 1960s, sport
hunting of the MCP was introduced, in part, to alleviate crop
depredation problems, particularly in the northern plains
(Lewis 1977). Over the past 5 decades, hunting seasons for
sandhill cranes have been implemented in 9 of 10 states that
form the Central Flyway. Along with a growing interest and
expansion of sandhill crane hunting, season lengths, number of
crane hunters, and harvest have increased (Kruse et al. 2008).
By 2000–2005, total harvest (including birds shot in Canada
and the estimated unretrieved kill) averaged approximately
33,000 birds annually (Sharp et al. 2007). Rapid growth in
annual harvest has resulted in a concern that insufficient
information is available to prevent overharvest of key segments
of the population, ultimately leading to diminished hunting
opportunities.
State and federal wildlife agencies in the Central Flyway have
taken a lead in seeking detailed quantitative information on the
MCP to provide diverse recreational opportunities consistent
with the welfare of the MCP, international treaties, and
socio-economic constraints while maintaining MCP crane abun-
dance at 1982–2005 levels (Central Flyway Webless Migratory
Game Bird Technical Committee 2006, Sharp et al. 2007).
Because sandhill cranes have the lowest annual recruitment rate
of game birds in North America (Drewien et al. 1995), a high
potential exists for overharvest. Therefore, greater precision is
required when making decisions concerning where and when to
target harvest than is required for most species. The MCP is
separated into the Gulf Coast and Western subpopulations for
management purposes (Tacha and Vohs 1984, Tacha et al. 1994).
Ranges of these subpopulations were delineated based upon
limited information on where MCP sandhill cranes of known
breeding origins staged during fall or spent winter. To ensure that
harvest strategies for the MCP are conservative, detailed insight
is required regarding the breeding origins of MCP sandhill cranes
using specific fall staging areas and wintering grounds, chronol-
ogy of use of these areas, knowledge of approximate numbers of
cranes in each subpopulation, and distribution and size of the
harvest by subpopulation.
Three subspecies are recognized in the MCP based on differ-
ences in crane morphometry: greater sandhill crane (Grus can-
adensis tabida), Canadian sandhill crane (G. c. rowani), and lesser
sandhill crane (G. c. canadensis; Tacha et al. 1994). Four recent
studies analyzing mitochondrial DNA (mtDNA) of these 3
subspecies concluded that genetic variation was insufficient to
warrant classifying the Canadian sandhill crane as a separate
subspecies from the greater sandhill crane (Rhymer et al. 2001,
Glenn et al. 2002, Peterson et al. 2003, Jones et al. 2005).
However, results of microsatellite and mtDNA lineage data
for greater sandhill crane also eliminated the possibility of these
birds being simply an alternate body form of greater sandhill
crane (Jones et al. 2005). Rather, results suggested that the
Canadian sandhill crane morphotype is intermediate in morph-
ometry, geography, and genetics, with a gradation in morpho-
metric and nuclear DNA variation from arctic-nesting lesser
sandhill crane to non-arctic greater sandhill crane. To allow
researchers and managers to relate our results to existing liter-
ature but also gain insight into the implications of current
genetic research, we present our results, where possible, in
relation to taxonomy based on crane morphometry and genetic
analyses.
Developing sound hunting regulations that will ensure a long-
term maximum sustainable harvest requires knowledge of
4 Wildlife Monographs 175
subpopulation sizes and a balance of annual mortality with annual
recruitment for each subpopulation. Detailed knowledge of
distribution of MCP sandhill cranes of each breeding affiliation
by date, along with information on their turnover rates and length
of stay on fall staging areas and wintering grounds is required to
evaluate distribution of harvest by subpopulation in states and
provinces of the Central Flyway where the species is hunted.
Currently, distribution of harvest by subspecies and subpopu-
lation is not known (Kendall et al. 1997). Also, information is
needed on extent of overlap of the MCP range with other
populations (i.e., the Eastern Population [EP] and Rocky
Mountain Population [RMP]) of greater sandhill cranes and
the Pacific Flyway Population (PFP) of lesser sandhill cranes.
Knowledge of extent of overlap with other populations is most
needed in regions where cranes from the MCP are hunted
because the EP, RMP, and PFP are smaller in size with either
no sport hunting permitted (EP) or limited sport hunting
allowed (RMP, PFP).
We used satellite telemetry as the primary tool to address
identified information needs. We attached Platform
Transmitting Terminals (PTTs) to a representative sample of
MCP cranes and systematically monitored their locations from
arrival on their breeding grounds to departure, during fall
migration, and throughout winter. We monitored tagged
MCP cranes to establish their breeding affiliations and to link
cranes of known breeding origin to their fall staging areas and
wintering grounds. We coupled information collected on
chronology of use by tagged MCP cranes of known breeding
affiliation at sites where they were exposed to hunting with
knowledge of hunting frameworks for states and provinces where
recreational hunting occurs, thus providing a sound basis for
estimating annual harvest by breeding affiliation.
STUDY AREA
Our study area encompassed the breeding grounds, fall migration
corridors, fall staging areas, and wintering grounds of the
MCP as determined by systematically monitoring locations
and movements of tagged sandhill cranes using satellite tele-
metry. Breeding grounds of the MCP encompassed most of the
central and western arctic of Canada eastward to Hudson Bay
and westward along the Arctic Ocean to Yukon Territory,
Alaska, and northeastern Russia (Tacha et al. 1994). The
southern edge of the breeding range extended from northwestern
Minnesota through southeastern and central Manitoba, central
Saskatchewan, central Alberta, northeastern British Columbia,
and central and southwestern Alaska. This vast area included
temperate grassland, mixed coniferous and deciduous forest,
parkland, boreal forest, and tundra. The winter range included
parts of the south-central and southwestern United States and
northern Mexico. We trapped and tagged all sandhill cranes we
monitored with PTTs in the Central Platte River Valley (CPRV)
and North Platte River Valley (NPRV) of south-central
Nebraska during 22 February to April 1998–2003 (Fig. 1).
Virtually the entire MCP staged in the CPRV and NPRV during
Figure 1. Locations (indicated by triangles) in (A) North Platte River Valley, and (B) Central Platte River Valley Nebraska where we captured and tagged 153 sandhill
cranes of the Mid-continent Population with Platform Transmitting Terminals (PTTs) during February–April, 1998–2003.
Krapu et al. Geographic Distribution of Sandhill Cranes 5
early spring, staying an average of 26 days (G. Krapu, U.S.
Geological Survey, unpublished data). Detailed descriptions of
the CPRV (U.S. Fish and Wildlife Service 1981, Krapu et al.
1982) and NPRV (Krapu et al. 1987, Iverson et al. 1987) have
been presented previously.
METHODS
Trapping and Transmitter Deployment
It was important from both a scientific and management perspect-
ive to capture and monitor a sample of cranes that would allow us to
make inferences concerning the entire MCP. We took severalsteps
to help ensure that our sample of tagged cranes was representative
of the MCP with regard to geographic distribution and chronology
of use of fall staging areas and wintering grounds, including sites
where most hunting occurred. Sandhill cranes begin arriving in the
CPRV and NPRV in mid-February; a rapid buildup follows from
early to mid-March, and the population generally peaks in the last
week of March (U.S. Fish and Wildlife Service 1981). As a result,
we scheduled our trapping efforts accordingly by capturing birds
throughout the period they were arriving in Nebraska. Magnitude
of trapping effort increased as the percentage of the MCP present
increased. Trapping effort and numbers of birds captured and
tagged with PTTs were distributed approximately proportional
to the number of birds using each section of the river.
We conducted trapping and tagging at numerous sites in the
CPRV and NPRV (Fig. 1), specifically in the Chapman to
Lexington reach of the CPRV, and in the Hershey area of the
NPRV, from late February to early April 1998–2003. Trapping
sites generally were located in grasslands (pastures) and hay lands in
the CPRV and NPRV, often in areas that functioned as secondary
roosts (i.e., where birds congregate in the morning after leaving the
nocturnal roosts, where they return in mid-day immediately after
feeding in adjacent agricultural fields, or where they occur before
moving on to nocturnal roosts). We captured birds by positioning
taxidermy-mounted sandhill cranes as small flocks in areas pre-
viously used and where we could fire well-hidden rocket-propelled
nets over cranes drawn within the capture zone of the net (Wheeler
and Lewis 1972). We set nets before the expected arrival of cranes,
accounting for areas of concentrated crane activity. We concealed
net setups by raking surrounding vegetation and covering all
components, taking care to provide total visual concealment and
minimal vertical obstruction. We fired rocket nets remotely (Fig. 2)
with radio-controlled detonators held by field personnel concealed
within 300 m of the net.
We removed captured cranes immediately and placed them into
burlap bags to restrain movement during processing. Upon cap-
ture, we took the following linear measurements (mm) of all
captured cranes: post-nares culmen (Fig. 3A), tarsus length
(Fig. 3B), and flattened wing chord (Fig. 3C). Based upon these
measurements, we later categorized birds by morphometry as
greater sandhill crane, Canadian sandhill crane, or lesser sandhill
crane using the discriminant methods of Johnson and Stewart
(1973). We selected adults for PTT attachment, marking 2
cranes from each throw of the net based on social status (e.g.,
family groups or pairs) or location under the net (e.g., pairs at
opposite ends). We established these criteria to limit the chance
of sampling related birds. Also, minimizing capture group size
shortened the stress period associated with capture and handling.
We sought a sample consisting of primarily breeding adults, so
we avoided sub-adults by tagging cranes with plumage charac-
teristic of adults. Adult characteristics included a bare, brightly
colored reddish forehead, lores, and crowns. We avoided tagging
cranes with tawny feathers on the crown, occiput, nape, or tawny
plumage on the body coverts and wing coverts, which are plu-
mage traits characteristic of juveniles (Lewis 1979).
We drew a blood sample from each crane selected to receive a
PTT from the metatarsal vein just below the tibio-tarsus joint of
the right leg (Fig. 3D) and placed it into a storage lysis buffer
(0.1 M Tris, 0.1 M EDTA, 5% SDS, 0.01 M NaCl; Longmire
et al. 1991) for later extraction to determine sex and mtDNA
lineage. We followed protocols for DNA isolation from the
Promega Wizard Genomic DNA Purification Kit (Promega
Corp., Madison, WI). We determined sex using polymerase-
chain-reaction (PCR) methods of Duan and Fuerst (2001). We
diagnosed mtDNA lineage using a modified PCR restriction-
fragment-length-polymorphism (RFLP) methodology of Glenn
et al. (2002). We substituted MSE I for HAE III, as it better
discriminated between the 2 mtDNA lineages.
We attached a PTT (Microwave Telemetry Inc., Columbia,
MD; North Star Science and Technology LLC, Baltimore, MD)
to the left leg of each crane using a 2-piece leg band (Fig. 4). Leg
bands consisted of a pair of 7.62-cm, semi-circular, flanged halves
of color-coated polyvinyl chloride (PVC) (Haggie Engraving,
Crumpton, MD) of which one half had 2 superimposed 2.5-cm
numeric or character codes engraved in it (Fig. 4). These 2 pieces
together formed a band with an inside diameter approximately
equivalent to U.S. Geological Survey Bird Banding Laboratory
band sizes 8 and 9. Manufacturers attached PTTs to the blank
band half with various methods. Band halves were lined with
1-mm-thick closed-cell neoprene to prevent abrasion and for
insulation from ambient cold and heat conduction from the
package. We mounted PTT leg bands on the left tibia above
the tibio-tarsus with the antenna pointing down. Preliminary
results indicated these PTTs arrangements resulted in acceptable
levels of signal reception and less stress to birds than did back-
pack harnesses (Ellis et al. 2001).
Figure 2. Capture of sandhill cranes for Platform Transmitting Terminal (PTT)-
tagging in the Central Platte River Valley, Nebraska during March 2002 using
concealed rocket-propelled nets fired remotely with radio-controlled detonators.
We used crane taxidermy mounts to attract groups of cranes to capture sites.
6 Wildlife Monographs 175
We released most captured birds simultaneously within 30 min
(range 15–60 min) of capture to maintain potential group and
family bonds. We released cranes captured in the evening under
enough ambient light to enable visual navigation to river roosts,
usually before sunset. All capture and marking procedures con-
formed to recommendations of the American Ornithologists’
Union (1997) and followed the protocol contained in Study
Plan 169.02 which was approved on 13 July 1998 by the
Chairman of the Animal Care and Use Committee at
Northern Prairie Wildlife Research Center.
We programmed PTTs at the time of manufacture to follow 1
of 4 duty cycles (Table 1). We structured duty cycles to give
more-frequent locations during migration and less-frequent
locations during periods when cranes were more sedentary
(i.e., while on breeding and wintering grounds). Improved tech-
nology allowed us to increase frequency of locations as the study
progressed, but the overall strategy of transmission frequency
relative to the annual crane cycle remained consistent. We sim-
ultaneously activated PTTs prior to deployment to ensure
Figure 3. We took the following linear measurements (mm) on each captured sandhill crane to be tagged with a Platform Transmitting Terminal (PTT): (A) post-nares
culmen length, (B), flattened wing chord and (C) tarsus length. We later used these measurements to categorize birds to subspecies based on their morphometry using the
discriminant methods of Johnson and Stewart (1973). (D) We drew a blood sample from the metatarsal vein just below the tibio-tarsus joint of the right leg of each
captured bird to determine sex and assign subspecies based on mtDNA lineage. We measured and tagged cranes in the Central Platte River Valley and North Platte River
Valley of south-central Nebraska during February–April, 1998–2003.
Figure 4. Platform Transmitting Terminal (PTT)-tagged sandhill crane with
plastic leg band on the left tarsus above the tibio-tarsus. The PTT is fused to
one half of a 2-part leg band and is located on the underside in this photo;numeric
code on leg band is to allow visual identification of each individual in the field. We
tagged MCP cranes in the Central Platte River Valley and North Platte River
Valley of south-central Nebraska during February–April 1998–2003.
Krapu et al. Geographic Distribution of Sandhill Cranes 7
synchronization of duty cycles and transmissions. Life span of
PTTs was projected to be 16 months to enable an evaluation of
level of philopatry to breeding sites used the previous year (we
programmed North Star PTTs to shut off during early June after
reaching the breeding grounds for the second year).
We used the Argos satellite system (Service Argos 2008) to
determine locations of tagged cranes throughout the annual cycle.
The Argos system consists of ultra-high-frequency (UHF)
receivers carried on 5 polar-orbiting National Oceanic and
Atmospheric Administration (NOAA) weather satellites that
receive PTT transmissions within their field of view.
Locations are calculated from the Doppler shift in the received
frequency as the satellite passes over the transmitter. Information
is transferred to Earth-based processing centers that make the
data available to users through personal computers within a few
hours of acquisition. Along with location, data from sensors built
into the PTTs that provided temperature of the PTT, battery
voltage, activity, and current duty cycle were also resolved during
transmission and relayed to processing centers. We received data
from Service Argos via daily E-mail. Fancy et al. (1988) and
Harris et al. (1990) provided a more detailed description of the
Argos system and its application in tracking wildlife.
Data Processing and Analyses
Argos assigns a quality code (Location Class [LC]) for each
location denoting its relative accuracy based upon satellite tele-
metry and transmitter geometry during the satellite pass, number
of messages received during the pass, and transmitter frequency
stability. Quality assessment by Argos has shown that assuming
isotropic error, the accuracy defined by each LC code represents
one standard deviation around the true PTT latitude and longi-
tude. Argos states accuracy by LC as follows: LC-3 ¼<250-m
radius; LC-2 ¼250–500-m radius; LC-1 ¼500–1,500-m radius;
LC-0 ¼>1,500-m radius; LC-A, LC-B, or LC-Z ¼no
accuracy assessment (Service Argos 2008, Section 3.4).
Service Argos estimates accuracy of their standard locations using
high-power PTTs under ideal ambient conditions. Empirical tests
of low-power wildlife PTTs under a variety of environmental
settings have reported slightly poorer accuracies for the standard
location classes (Harris et al. 1990, Vincent et al. 2002). Keating
et al. (1991) and Clark (1989) showed that a 68% distribution (1
SD) does not always hold true for smaller PTTs used in wildlife
tracking. Although Argos gives no accuracy ratings for LC-0, LC-
A, or LC-B, the calculated location was shown to be within
11.5 km, 6.8 km, and 98.5 km, respectively, of the true location
68% of the time for PTTs similar to those we used (Britten et al.
1999). In our opinion, magnitudes of errors like these are accept-
able for interpreting continental-scale migration.
Filtering methods can improve robustness of the Argos
auxiliary locations, but most published algorithms are for marine
and terrestrial species that are far less mobile than birds
(McConnell et al. 1992, Keating 1994, Hays et al. 2001,
Austin et al. 2003). We used the Douglas Argos-Filter
Algorithm version 6.5 (Alaska Science Center 2010) developed
by D. Douglas (U.S. Geological Survey). The Douglas filter
extracts locations from the Argos diagnostic format files and
provides output as a number of filtered data sets including all
locations (no filtering), minimum-redundant-distance (MRD)
filtered locations, distance-angle-rate (DAR) filtered locations,
and a hybrid of both. The MRD algorithm allows the user to set a
maximum distance from each location beyond which all other
locations are rejected within a set time frame. The DAR algor-
ithm determines whether to accept a location by determining the
angle of divergence away from the path created by connecting 3
consecutive locations with the location under evaluation. The
hybrid filter uses both the MRD and DAR algorithms. User
inputs include a limit on the plausible rate of travel and the
acceptable angle of divergence. The filter default does not remove
LC-3 locations but this limit may be set lower. We used the
hybrid filter with the LC limit set at 1 (LC-1, LC-2, and LC-3
always retained), and we set the maximum rate of travel 100 km/
hr, maximum redundancy value to 30 km, and the angle of
divergence parameter to ignore all angles >1258. For more
detailed information on how the filter works, see Alaska
Science Center (2010).
We ran all resolved satellite locations through the filter, which
marks records for deletion that do not meet the user-defined
criteria for inclusion. We then manually scrutinized the resulting
output as a geographic information system (GIS) layer and made
subjective decisions as to the legitimacy of removing each indi-
vidually marked as well as any unmarked locations that appeared
unrealistic or improbable. This processing resulted in a final
dataset of crane locations that contained only those locations
that passed through this stepwise sequential geospatial algorithm
and subjective manual review.
We initially imported the final filtered data into MapInfo
Professional software (TETRAD Computer Applications Inc.,
Bellingham, WA) for viewing and summary. We attached annual
cycle attributes (Platte River, Breeding Area, Fall Migration, and
Wintering Area) post hoc to each location based upon manual
inspection of arrival and departure to and from appropriate
discrete geographic areas. We designated Breeding Area to
locations from arrival at the geographic terminus of spring
migration through a major departure from that area. We assigned
Fall Migration to locations from departure from breeding
grounds through arrival at Wintering Area. Wintering Area
was any location in a state (U.S. or Mexico) where a tagged bird
generally terminated its fall migration in any year. During the
breeding season (Jun-Aug), most cranes were confined to a small
area, suggesting a breeding status. We created one breeding-
Table 1. Programmed transmission schedules by annual cycle for Platform Transmitting Terminals (PTTs) we deployed on sandhill cranes captured in the Central
Platte River Valley and North Platte River Valley, Nebraska, 1998–2003. Numbers represent transmission to the satellites every nth day.
Program nSpring staging Spring migration Breeding grounds Fall migration Winter grounds Second spring migration
A614 4 7 4 10 5
B 9 5 2 8 4 10 4
C175 1 8 2 9 4
D 66 1.5 1 4 1.5 4 3
8 Wildlife Monographs 175
season location for each monitored crane by calculating a
weighted mean of all breeding ground locations. We weighted
each observation by a numeric representation of the Argos
Location Class (LC-3 ¼4, LC-2 ¼3, LC-1 ¼2, LC-0 ¼1,
LC-A ¼1, LC-B ¼1, LC-Z ¼1) to reflect the decreasing
accuracy of these LCs and to theoretically arrive at a more precise
estimate of the activity center.
We assigned the MCP to 4 breeding affiliations based upon
breeding-ground locations and locations of cranes’ fall staging
areas in the northern plains (Fig. 5). The 4 breeding affiliations
were East-central Canada–Minnesota (EC–M), West-central
Canada–Alaska (WC–A), Western Alaska–Siberia (WA–S),
and northern Canada–Nunavut (NC–N). To assess whether
our tagged sample reliably represented the distribution of cranes
by breeding affiliation in the CPRV, we compared the percent
distribution of our tagged sample of sandhill cranes by bridge
section with estimates of percent crane distribution by bridge
section obtained from aerial infrared videography while cranes
were gathered on nocturnal roosts. We conducted aerial infrared
surveys of cranes on or near the fourth Tuesday of March 2000–
2003 (Kinzel et al. 2006). We assumed that we sampled breeding
affiliations of cranes in proportion to their abundance in each
bridge section. We next adjusted numbers of PTT-tagged cranes
by bridge section to reflect the proportion of roosting cranes of
each breeding affiliation in each bridge section using estimates
available from infrared videography.
We investigated areas that cranes of each breeding affiliation
occupied on fall staging areas in the northern plains and on the
wintering grounds in Texas to gain insight into spatial relation-
ships among the 4 breeding affiliations. Given the large spatial
scale of our coverage, we converted all locations to meters using a
cylindrical equal area map projection to minimize area distortion
(Yildirim and Kaya 2008). Because multiple locations with min-
imal temporal separation are often resolved during one PTT
transmission event, we calculated an average location, weighted
by LC, for each individual bird and satellite overpass using all
resolved locations in the final database. We created kernel density
estimates using ArcMap software. We set the grid-cell parameter
to the mean error estimate from all locations in our final dataset
(12.231 km) based upon NQ (0–3) observed by Keating et al.
(1991) and LC-A and LC-B reported in Soutullo et al. (2006).
We used an interactive average-nearest-neighbor approach in
setting the bandwidths (k) for the analyses, integrating aver-
age-nearest-neighbor values for kuntil nominal coverage and
actual coverage of 50%, 75%, and 95% kernel density estimates
(KDEs) were similar. We found that the 5th average-nearest-
Figure 5. Locations where Platform Transmitting Terminal (PTT)-tagged sandhill cranes (n¼153) of the Mid-continent Population settled on breeding grounds
in the United States, Canada, and Russia, 1998–2004. Locations are by breeding affiliation, that is, Western Alaska–Siberia (WA–S), Northern Canada–Nunavut
(NC–N), West-central Canada–Alaska (WC–A), and East-central Canada–Minnesota (EC–N). Abbreviations of states, provinces, and territories are as follows:
AK ¼Alaska, AB ¼Alberta, BC ¼British Columbia, MB ¼Manitoba, MN ¼Minnesota, NU ¼Nunavut, NWT ¼Northwest Territories, ON ¼Ontario,
QB ¼Quebec, SK ¼Saskatchewan, YT ¼Yukon Territory.
Krapu et al. Geographic Distribution of Sandhill Cranes 9
neighbor distance (9,687 m) fit the fall staging data well and the
11th average-nearest-neighbor distance (10,296 m) fit the win-
tering area data well.
We used grid-cell counts based on the average size of our errors
to summarize the area used by each breeding affiliation and to
estimate the utilization distribution (UD) for each. We com-
puted 5 overlap indices (Fieberg and Kochanny 2005) for each
pair of breeding affiliations. We used 3 of these metrics, 2 for
interpreting overlap via probabilities of co-occurrence (PHR
i,j
and PHR
j,i
) and 1 for quantifying space-use sharing (i.e., the UD
overlap index [UDOI]). Probability of breeding affiliation jbeing
located in an area used by breeding affiliation i(PHR
i,j
) accounts
for how much a cell was used. We computed PHR
i,j
by summing
the probability of use by breeding affiliation jacross cells used by
both breeding affiliations iand j. We needed to compute PHR
i,j
separately for the overlap of breeding affiliation iwith jand the
overlap of breeding affiliation jwith i, which resulted in 2 values
for each overlap index for each pair of breeding affiliations. The
UDOI is non-directional and we only needed to compute it
once for each pair of breeding affiliations. The utilization distri-
bution overlap index equals zero for 2 breeding affiliations
with no overlap and equals 1 for 2 breeding affiliations if both
UDs are uniformly distributed and have complete overlap; how-
ever, UDOI can also exceed 1 if the UDs of 2 breeding affiliations
are non-uniformly distributed and have a high degree of overlap.
To estimate crane exposure to hunting seasons, we eliminated
all locations from the final dataset except for those dates where
hunting was allowed under the Federal framework for sandhill
crane hunting seasons in accordance with the Migratory Bird
Treaty Act (1 Sep–10 Mar). For each tagged crane, we summar-
ized the dates when it occurred in a state or a zone within a state
for that time frame. Some cranes were located in multiple states
or zones within a state on the same day and we assigned exposure
to each. We then compared specific season dates where crane
hunting was allowed within each state and any hunting zones
within a state for each year to the defined exposure period for each
crane. If that date or span of dates fell within the defined hunting
season, we assigned exposure to hunting. If a tagged crane was
located in a state or hunting zone within a state that allowed
hunting but no season was in effect for that period, or if a crane
was located in a state that did not have a crane hunting season, we
assigned exposure to the non-hunting category. We summarized
these data for descriptive presentation as mean exposure to
hunting and non-hunting by state and province, by hunting unit
within state, and by breeding affiliation.
We estimated harvest of the MCP by breeding affiliation by
assigning harvest in proportion to the exposure of tagged cranes of
each breeding affiliation within each state or province. For
Saskatchewan, Manitoba, North Dakota, and Texas, where nearly
80% of the annual crane harvest occurred, we used smaller-scale
calculations (county level for states, and 18latitude by 28longitude
for provinces) to estimate use and harvest. We then multiplied
proportional use by each breeding affiliation by the average annual
harvest during the same years as our study (1998–2003) to arrive at
an estimated harvest by breeding affiliation.
RESULTS
We attached PTTs to 153 cranes in the CPRV and NPRV of
Nebraska and monitored the birds via receivers on NOAA
weather satellites for 7 years between 1 April 1998 and 30
May 2004. We tagged 131 sandhill cranes (86%) with PTTs
in the CPRV and 22 (14%) in the NPRV. We excluded the 2003
sample of tagged cranes from analyses that required a random
sampling of the entire population because in that year, we focused
on capturing and tagging a representative sample of greater
sandhill cranes to gain more insight into life history of this less
plentiful but important component of the MCP. We tracked an
average of 35 cranes (range ¼4–56) annually (Table 2); median
duration of tracking of individual cranes was 352 days, and we
tracked 73 cranes >12 months. Overall, tracking effort yielded
74,041 Argos-determined locations over 49,350 tag-days
(Table 2). We used these locations for analyses after we applied
the filter algorithm and manually scrutinized the location data.
Number of cranes monitored each month and number of days
they carried functioning PTTs peaked in spring shortly after
tagging and declined through late winter (Table 3).
Analysis of mtDNA genotypes for 129 tagged cranes (we did
not perform mtDNA analysis on 4 birds tagged in 1998) indi-
Table 2. Satellite-tracking effort by year, as measured by the number of Platform Transmitter Terminal (PTT)-tagged sandhill cranes and number of days that sandhill
cranes were carrying functioning PTTs from March 1998 to June 2004, in mid-continent and northwestern North America, and northeastern Asia.
Year of study
1998 1999 2000 2001 2002 2003 2004
Tagged cranes (n) 4 18 41 54 56 50 20
Tag-days (n) 1,036 4,654 6,914 11,019 12,749 10,970 2,008
Table 3. Satellite-tracking effort by month, as measured by the number of individual Platform Transmitter Terminal (PTT)-tagged sandhill cranes that transmitted
successfully during each month of each calendar year from deployment to last successful transmission while in mid-continent and northwestern North America, and
northeastern Asia. Tag-days represent the total number of days that sandhill cranes were carrying functioning PTT tags each month from March 1998 to June 2004,
inclusive.
Month of study
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Tagged cranes (n) 86 89 196 221 211 193 166 138 129 114 98 92
Tag-days (n) 2,718 2,361 4,189 6,484 6,374 5,489 4,792 4,168 3,679 3,393 2,872 2,831
% of total days 5.5 4.8 8.5 13.1 12.9 11.1 9.7 8.5 7.5 6.9 5.8 5.7
10 Wildlife Monographs 175
cated 83 (64%) were lesser sandhill crane and 46 (36%) were
greater sandhill crane. Based on morphometry, 56%, 38%, and
5% of tagged cranes (n¼133) were lesser sandhill crane, the
Canadian sandhill crane morph, and greater sandhill crane. The
tagged sample of 20 sandhill cranes from 2003 were all G. c.
tabida based on mtDNA lineage and 30% greater sandhill crane
and 70% Canadian sandhill crane based on morphometry.
Geographic Distribution During Breeding Season
MCP-tagged adult cranes captured in the CPRV (n¼131) and
NPRV (n¼22) settled on breeding grounds on 2 continents
(North America, Asia) and 3 nations (U.S., 23% [n¼35];
Canada, 57% [n¼87]; and Russia, 20% [n¼31]; Fig. 5). Of
tagged cranes, 97% spent the breeding season on the continental
land masses of North America (n¼117) and Asia (n¼31), and
3% (n¼5) spent the breeding season on arctic islands off the
coast of continental North America on Richards Island (n¼1),
Banks Island (n¼2), and Victoria Island (n¼1) in Canada and
on Nunivak Island (n¼1) in Alaska. In the United States, tagged
cranes settled in Minnesota (3%) and Alaska (20%); in Canada
they settled in Quebec (<1%), Ontario (7%), Manitoba (9%),
Saskatchewan (3%), Alberta (7%), Nunavut (15%), British
Columbia (<1%), Northwest Territories (14%), and Yukon
Territory (<1%); in Russia, they settled in northeastern
Siberia (20%; Fig. 5). Annual movements and geographic distri-
bution of MCP sandhill cranes that breed in northeastern Russia,
their ecology, estimated harvest, and factors influencing recent
breeding range expansion will be addressed in a separate publi-
cation (G. L. Krapu, unpublished data).
Tagged cranes of the EC–M breeding affiliation settled prim-
arily in the Hudson Bay Lowlands near James Bay in northeast-
ern Manitoba, northern Ontario and western Quebec, the
Interlake region of central Manitoba, and northwestern
Minnesota and adjacent parts of southeastern Manitoba
(Fig. 5). Of 30 tagged cranes that settled in East-central
Canada and northwestern Minnesota during spring, 93%
(n¼28) were greater sandhill crane, 3% (n¼1) were lesser sand-
hill crane, and 3% (n¼1) were unclassified based on results from
mtDNA analyses, whereas 70% (n¼20) were Canadian sandhill
crane and 30% (n¼10) were greater sandhill crane based on
morphometry. From the smaller, randomly trapped sample of 19
tagged EC–M cranes (excluding 2003 cranes), which represented
an estimated 14 3% of the MCP, 42% settled in central and
southeastern Manitoba, central Ontario, and northwestern
Minnesota, and 58% settled in northeastern Manitoba, northern
Ontario, and western Quebec (Fig. 5).
Sandhill cranes of the WC–A breeding affiliation settled in
central Saskatchewan, across central and northern Alberta,
northeastern British Columbia, the Great Slave Plains in the
Northwest Territories, and in the Yukon Flats of east-central
Alaska (Fig. 5). Of 39 WC–A tagged cranes, 85% (n¼33) were
greater sandhill crane and 15% (n¼6) were lesser sandhill crane
based on mtDNA, whereas 90% (n¼35) were Canadian sandhill
crane, 8% (n¼3) were greater sandhill crane, and 2% (n¼1)
were lesser sandhill crane based on morphometry. Of 30 ran-
domly tagged cranes, representing an estimated 23 4% of
the MCP that settled in west-central Canada and interior
Alaska, 66%, mostly Canadian sandhill cranes, settled from
interior Alaska eastward to the Northwest Territories, and
33%, also mostly Canadian sandhill cranes, settled primarily in
Saskatchewan and Alberta (Fig. 5).
Cranes of the WA–S breeding affiliation settled in western
Alaska from the Yukon-Kuskokwim Delta northward to the
Seward Peninsula and in northeastern Russia. Of the 56 tagged
sandhill cranes that settled in western Alaska and northeastern
Russia, 92% (n¼52) were lesser sandhill crane, 4% (n¼2) were
greater sandhill crane, and 4% (n¼2) were unclassified based on
results from mtDNA analyses. Based on morphometry, 88%
(n¼49) were lesser sandhill crane and 12% (n¼7) were
Canadian sandhill crane. This group of tagged cranes represented
an estimated 42 4% (SE) of the MCP. Of tagged cranes
that stayed in western Alaska, 64% settled on the Yukon-
Kuskokwim Delta; remaining tagged WA–S cranes in Alaska
settled northeast near Nikolai, along the inner reaches of the
Yukon River and northward to the Seward Peninsula and Selawik
National Wildlife Refuge (NWR) near the Kobuk River Delta
(Fig. 5).
Sandhill cranes of the NC–N breeding affiliation were distrib-
uted from near the Arctic Ocean in the Yukon Territory eastward
to the Boothia Peninsula, in parts of the Canadian Archipelago
(i.e., Richards Island, Banks Island, and Victoria Island), and on
the northwest side of Hudson Bay (Fig. 5). Based on their
mtDNA, 96% of PTT-tagged NC–N cranes (n¼27) were lesser
sandhill crane and 4% (n¼1) were unclassified based on results
from mtDNA analyses; 83% (n¼25) were lesser sandhill crane
and 17% (n¼3) were Canadian sandhill crane, based on morph-
ometry. Our NC–N sample represented an estimated 21 4% of
the MCP, and 50% settled along the western and central Arctic
coast and high Arctic islands versus 50% along the northwest side
of Hudson Bay (Fig. 5). In the western Canadian Arctic, tagged
cranes settled principally on or near the MacKenzie Delta and on
Banks Island. In the central Canadian Arctic, PTT-tagged cranes
settled near the Arctic Ocean eastward from near Bathhurst Inlet
to the southwest edge of Boothia Peninsula (Fig. 5). Along the
northwest side of Hudson Bay, cranes were distributed across
approximately 50,000 km
2
from near Rankin Inlet southward to
near the Manitoba border at 608N, a distance of approximately
260 km (Fig. 5).
We found only marginal differences between the percentage of
tagged cranes of each breeding affiliation by bridge section in the
CPRV based on our adjusted tagged sample and the percentage
obtained after adjusting for any disparities in distribution based
on crane estimates from aerial infrared videography (see Totals by
breeding affiliation, Table 4). The largest difference between the
2 estimates was for EC–M, which averaged 1.0% less after
adjusting for population distribution based on results from infra-
red videography (Table 4). A potential cause for this difference,
based on knowledge gained from related work, is that EC–M
cranes, on average, leave the CPRV 6, 8, and 10 days earlier than
cranes from WC–A, NC–N, and WA–S (G. Krapu, unpublished
data) so a higher proportion of cranes from this breeding affili-
ation may have left by the dates of the surveys. Of the 22 cranes
we captured and tagged in the NPRV, 90% were WA–S cranes.
When we added these birds to tagged cranes from the CPRV,
WA–S became the most numerous breeding affiliation (56 of
133), accounting for 42.1% of the MCP (Table 4).
Krapu et al. Geographic Distribution of Sandhill Cranes 11
Philopatry and Chronology of Use of Breeding Grounds
All 53 tagged sandhill cranes returning to breeding grounds in
the second year after tagging settled at sites near those occupied
during the previous year (Table 5), with 38% at sites averaging
<1 km from the previously occupied locations. All returning
birds were located <8 km from the activity center occupied
during the previous breeding season (median distance ¼1.6 km).
No tagged birds switched breeding affiliation between years.
Mean arrival dates for cranes of the 4 breeding affiliations on
their breeding grounds varied by up to 32 days, with the EC–M
affiliation being the first to arrive followed by WC–A, WA–S,
and NC–N affiliations (Table 6). Cranes of the EC–M affiliation
spent the most time on their breeding grounds followed by cranes
in the WC–A, WA–S, and NC–N affiliations (Table 6). In
autumn, cranes from the WC–A affiliation were the first to
depart from the breeding grounds, followed by cranes from
EC–M, WA–S, and NC–N affiliations (Table 6).
Fall Migration Routes
Cranes of WC–A affiliation breeding in northern Alberta, north-
eastern British Columbia, the Northwest Territories, Yukon
Territory, and Alaska migrated southeastward on flight paths
to their fall staging areas in central Saskatchewan (Fig. 6A).
Cranes of WC–A affiliation breeding in Saskatchewan required
short flights to reach their fall staging areas in central
Saskatchewan. When departing from their fall staging areas,
WC–A cranes migrated southeast across southeastern
Saskatchewan and through central and western North Dakota,
South Dakota, Nebraska, Kansas, and Oklahoma, en route to
wintering grounds located primarily in western, central, and
coastal Texas (Fig. 6A). Only 5 of 29 cranes (17%) stopped in
North Dakota for any period of time (range 3–7 days).
All EC–M cranes breeding in Manitoba, Ontario, and Quebec
(n¼22) migrated in September to fall staging areas in south-
western Manitoba and North Dakota (Fig. 6B). Of those, 36%
(n¼8) went first to southwestern Manitoba then continued on to
North Dakota, whereas 27% (n¼6) staged exclusively in
Manitoba and 36% (n¼8) proceeded directly to North
Dakota. Four EC–M cranes breeding east of the Inter-lakes
Region of central Manitoba and 1 crane that spent the breeding
season in the Interlake Region staged 9–21 days between Lake
Winnipeg and Lake Manitoba before continuing to staging areas
in southwestern Manitoba and eastern North Dakota (Fig. 6B).
All EC–M cranes breeding in northwestern Minnesota (n¼4)
staged during fall in northwestern Minnesota and migrated
southwestward to their wintering grounds without stopping in
North Dakota or southwestern Manitoba. One EC–M crane
breeding in southeastern Manitoba near the Minnesota border
staged in eastern North Dakota.
When departing from fall staging areas, EC–M cranes migrated
southward within a narrow migration corridor that approxi-
mately followed the route of U.S. Highway 281 through the
plains states (Fig. 6B) and mostly east of WC–A, which migrated
across a wider corridor that extended from the central to near the
western borders of the plains states (Fig. 6A). About 48% of EC–
M cranes (n¼11) utilized areas in Kansas (especially Kirwin
NWR, Quivira NWR, and Cheyenne Bottoms Wildlife Area)
and Oklahoma (especially along the Red River on the Texas–
Oklahoma border) before continuing on to more southerly desti-
nations. Of EC–M cranes, 22% (n¼5) terminated fall migration
Table 4. Distribution of sample of 133 Platform Transmitter Terminal (PTT)-tagged sandhill cranes by river section and breeding affiliation in the Central Platte
Valley and North Platte Valley, Nebraska, compared with distribution of cranes on nocturnal roosts obtained by infrared videography (IR) during 2000–2003 (Kinzel et
al. 2006). Data in IR columns represent the estimated number of cranes by breeding affiliation that would have been PTT-tagged if capture proportions by river segment
had been equivalent to crane breeding affiliation composition based on IR distribution.
Breeding affiliation
a
River segment Totals
Lexington to Kearney Kearney to Shelton Shelton to Chapman North Platte
b
PTTs IR
PTTs IR PTTs IR PTTs IR PTTs IR No. % No. %
WA–S 9 6.0 13 16.9 14 13.0 20 20 56 42.1 56.0 42.1
NC–N 6 4.0 12 15.6 8 7.4 2 2 28 21.1 29.1 21.9
WC–A 1 0.7 7 9.1 22 20.5 0 0 30 22.6 30.3 22.7
EC–M 0 0 0 0 19 17.7 0 0 19 14.3 17.7 13.3
Combined 16 10.7 32 41.7 63 58.6 22 22 133 133
a
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
b
Estimates are not available on percentage of MCP cranes that occupied the North Platte section from IR. However, the 16.5% (22/133) of our PTT-tagged sample
that came from the North Platte section would represent an estimated 99,000 cranes in an MCP numbering 600,000 birds. This estimate is only marginally higher than
the 93,225 cranes estimated for the North Platte section based on aerial surveys conducted by the U.S. Fish and Wildlife Service and Nebraska Game and Parks
Commission (J. Solberg, U.S. Fish and Wildlife Service, personal communication; M. Vrtiska, Nebraska Game and Parks Commission, personal communication).
Table 5. Median and mean distances in kilometers separating locations on the breeding grounds during successive years for Platform Transmitting Terminal (PTT)-
tagged sandhill cranes of the Western Alaska–Siberia, Northern Canada–Nunavut, West-central Canada–Alaska and the East-central Canada–Minnesota breeding
affiliations, 1998–2004. We did not include birds with <5 locations on the breeding grounds in a year.
Breeding affiliation nMedian xSE
Western Alaska–Siberia 13 1.81 3.16 0.76
Northern Canada–Nunavut 4 0.99 1.64 0.91
West-central Canada–Alaska 18 1.13 1.97 0.50
East-central Canada–Minnesota 18 1.78 2.06 0.47
12 Wildlife Monographs 175
in areas north of or along the Red River on the Texas–Oklahoma
border. By mid-December, EC–M cranes had largely arrived at
their final winter destination either on the Texas Gulf Coast,
western Texas, or east-central Texas or remained at sites in
Oklahoma or Kansas.
Cranes of WA–S affiliation departing from the Yukon-
Kuskokwim Delta breeding grounds migrated northeastward
in early September and converged in central Alaska with cranes
coming from Russia (Fig. 6C). After migration paths of cranes
from western Alaska and Russia merged, the birds flew southeast
within a narrow corridor (30–50 km wide), mostly following the
Tanana River. In southeastern Alaska, the birds crossed over to
the Pelly River Valley and continued southeastward across Yukon
Territory, northeastern British Columbia, and northwestern,
central, and southeastern Alberta to their primary fall staging
area in southwestern Saskatchewan, adjacent to the western
reaches of the South Saskatchewan River (Fig. 6C).
Cranes of WA–S affiliation departed in mid-October from
western Saskatchewan after staging for approximately 1 month,
migrating southeast across central and eastern Montana, eastern
Wyoming, eastern Colorado, and northeastern New Mexico en
route to their primary wintering grounds in western Texas. The
eastern edge of the fall migration corridor of WA–S cranes
extended into western South Dakota, Nebraska, Kansas, and
the panhandle of Oklahoma (Fig. 6C). When WA–S cranes
within the primary migration corridor reached northeastern
New Mexico, 76% continued to western Texas, whereas 24%
took direct routes to wintering areas in central New Mexico,
southeastern Arizona, and northern Mexico (Fig. 6C). Three
cranes (9%) initially utilized wintering areas in western Texas but
ultimately migrated into northern Mexico. Two cranes initially
migrated to northern Mexico and then moved east back into
western Texas; one ultimately settled in Coahuila (Fig. 6C).
Tagged WA–S cranes made no major stops during migration
from fall staging areas to wintering grounds.
Tagged NC–N cranes during the breeding season were dis-
tributed from near the coast in Yukon Territory eastward across
the Arctic to near Churchill, Manitoba on Hudson Bay
(Fig. 6D). As a result of this wide breeding distribution, several
migration routes were followed when moving from major breed-
ing grounds to primary fall staging areas in central and east-
central Saskatchewan, and northwestern and central North
Dakota. Tagged cranes breeding on Banks Island migrated
southeast through the central part of the Northwest
Territories and continued on a southeasterly course across north-
eastern Alberta en route to fall staging areas in central
Saskatchewan (Fig. 6D). Cranes breeding west of Hudson Bay
in Nunavut flew southwest across northwestern Manitoba en
route to staging areas. Three of 12 (25%) cranes from this
breeding area continued on a southwest course until they reached
the Quill Lakes area of Saskatchewan, whereas the remaining 9
(75%) birds crossed into Saskatchewan stopping only briefly
before moving on to fall staging areas in central North
Dakota. The 3 NC–N cranes from the western Hudson Bay
breeding area that staged primarily in Saskatchewan migrated
southeast into North Dakota, where 2 birds were known to
have stopped for a short period. From North Dakota, all NC–
N cranes migrated south through central South Dakota and in
central Nebraska turned southwest and crossed western Kansas
and the panhandle of Oklahoma en route to their wintering
grounds in western Texas (Fig. 6D); NC–N cranes made no
major stops between their fall staging areas and wintering
grounds.
Chronology of Migration to Fall Staging Areas
We used median weekly locations to represent overall movement
of individuals from each breeding affiliation. Fall migrations of
WC–A and EC–M breeding affiliations began during the third
and fourth weeks of August (weeks [WKs] 33 and 34; Fig. 7),
with mean departure dates from breeding areas of 20 August and
30 August, respectively (Table 6). By WK 35 (27 Aug–2 Sep),
WC–A and EC–M cranes had moved 1,245 km and 175 km,
respectively, from sites occupied during the breeding season
(Fig. 7). Cranes of WC–A affiliation migrated more rapidly than
EC–M cranes and reached their fall staging areas, on average, 6
days earlier (Table 7). After EC–M cranes arrived on their fall
staging areas, the center of the EC–M crane distribution gradu-
ally moved from southern Manitoba into eastern North Dakota
during 10 September–21 October (WKs 37 to 42; Fig. 7). As
determined by median location, WC–A and EC–M cranes trav-
eled approximately 1,400 km and 700 km from breeding grounds
to fall staging areas in central Saskatchewan and southwestern
Manitoba, respectively. Distribution of WC–A cranes changed
only slightly during 3 September–14 October (WKs 36–41),
whereas the median location of the EC–M crane distribution
moved south 93 km by 1–7 October (WK 39 to WK 40). By 15–
21 October (WK 42), EC–M distribution had moved another
123 km, with cranes becoming widely dispersed in eastern and
central North Dakota (Fig. 7). From median arrival to median
departure on their fall staging areas in Saskatchewan and
Manitoba/North Dakota, WC–A and EC–M cranes moved a
weekly average of 34 km and 61 km, respectively.
Table 6. Mean arrival date, departure date, and length of stay of Platform Transmitting Terminal (PTT)-tagged sandhill cranes by breeding affiliation on the breeding
grounds. We captured PTT-tagged sandhill cranes while they were on spring staging areas in the Central Platte River Valley and North Platte River Valley of Nebraska,
1998–2003. We calculated dates using mid-points of appropriate previous and successive locations.
Breeding affiliation
Arrival date
a
Departure date
a
Length of stay (days)
nxSE (days) nxSE (days) nxSE
Western Alaska–Siberia 69 17 May 1.0 40 3 Sep 1.6 40 108 2.1
Northern Canada–Nunavut 32 25 May 1.8 20 8 Sep 1.5 19 107 3.1
West-central Canada–Alaska 57 30 Apr 1.5 35 20 Aug 2.8 35 113 3.9
East-central Canada–Minnesota 47 23 Apr 1.8 25 30 Aug 2.9 25 127 4.0
a
We did not include cranes with location intervals >10 days between previous and successive locations.
Krapu et al. Geographic Distribution of Sandhill Cranes 13
The WA–S breeding affiliation initiated fall migration during
3–9 September (WK 36) with a mean departure date of
3 September (Table 6). During WKs 36–38, the center of the
WA–S crane distribution moved 500 km, 808 km, and 2,436 km,
respectively, southeast from the Chutkotka Peninsula which lies
adjacent to the Bering Strait separating Alaska from Russia
(Fig. 8). Cranes of WA–S affiliation staged primarily near
Cabri and other areas near the South Saskatchewan River south
Figure 6. Fall migration routes of Platform Transmitting Terminal (PTT)-tagged MCP sandhill cranes of the 4 breeding affiliations: (A) West-central Canada–Alaska
[WC–A]; (B) East-central Canada–Minnesota [EC–M]; (C) Western Alaska–Siberia [WA–S]; (D) Northern Canada–Nunavut [NC–N] based on their PTT-
locations during fall 1998–2003. Closed circles with white rings represent breeding locations for cranes of each breeding affiliation.
14 Wildlife Monographs 175
of Eston in western Saskatchewan during 24 September–
14 October (WKs 39–41), with limited movement of cranes
occurring during that period (Fig. 8).
Cranes of NC–N breeding affiliation began moving southward
from their breeding grounds during 3–9 September (WK 36;
Fig. 8) and moved 500 km by the end of the following week
(WK 37). By 24–30 September (WK 39), PTT-tagged NC–N
cranes had traveled approximately 1,900 km from their median
breeding ground locations and most had reached their fall staging
sites in east-central Saskatchewan and North Dakota (Fig. 8).
During 24 September to 14 October (WKs 39–41), the center of
fall distribution of NC–N cranes moved progressively southward
across central North Dakota, averaging 119 km per week (Fig. 8).
About a third (35%) of NC–N cranes headed directly to east-
central Saskatchewan where they spent most of the staging
period. Ten of 17 (59%) NC–N cranes continued on to North
Dakota, spending less than one week in east-central
Saskatchewan, whereas only 1 of 17 birds split the staging interval
Table 7. Mean arrival date, departure date, and length of stay of Platform Transmitting Terminal (PTT)-tagged sandhill cranes by breeding affiliation on fall staging
areas and wintering grounds. We captured the PTT-tagged sample of sandhill cranes during springs 1998–2003 while cranes were on spring staging areas in the Central
Platte River Valley and North Platte River Valley in Nebraska. We calculated dates using mid-points of appropriate previous and successive locations.
Breeding affiliation
Arrival date
a
Departure date
a
Length of stay (days)
nxSE (days) n x SE (days) n x SE
Location
Fall staging areas
Western Alaska–Siberia 35 19 Sep 1.0 32 13 Oct 1.3 32 24 1.2
Northern Canada–Nunavut 18 17 Sep 1.1 17 15 Oct 2.1 17 29 2.5
West-central Canada–Alaska 35 29 Aug 2.5 31 16 Oct 0.9 31 49 3.0
East-central Canada–Minnesota 24 4 Sep 3.2 24 24 Oct 2.2 22 49 2.2
Wintering grounds
Western Alaska–Siberia 32 17 Oct 1.3 17 6 Mar 2.1 17 140 2.5
Northern Canada–Nunavut 15 18 Oct 1.8 7 9 Mar 3.1 7 144 5.1
West-central Canada–Alaska 31 21 Oct 1.3 20 9 Mar 2.1 20 142 1.8
East-central Canada–Minnesota 25 27 Oct 1.9 19 2 Mar 2.4 19 128 3.2
a
We did not include cranes with location intervals >10 days between previous and successive locations.
Figure 7. Median weekly locations (e.g., 1–7 Jan ¼week 1, etc.) of Platform Transmitting Terminal (PTT)-tagged sandhill cranes of the West-central Canada–Alaska
(WC–A), and East-central Canada–Minnesota (EC–M) breeding affiliations between departure from the breeding grounds and arrival on the wintering grounds, 1998–
2003. BG ¼breeding ground, WG ¼wintering ground. Numbers within symbols indicate week(s) birds were located at a site.
Krapu et al. Geographic Distribution of Sandhill Cranes 15
between North Dakota and Saskatchewan. Of the 35% (6 of 17)
of NC–N cranes that staged primarily in east-central
Saskatchewan, the mean duration of stay was 32 days. For those
birds from NC–N that stayed primarily in North Dakota, mean
length of stay was 19 days. Overall, NC–N cranes spent 29 days
on fall staging areas (Table 7).
Use of Fall Staging Areas
Fall staging was centered at traditionally used sites in central and
western Saskatchewan (69%); northwestern, central, and eastern
North Dakota (16%); southwestern Manitoba (10%); and north-
western Minnesota (3%). Among WA–S and WC–A cranes,
97% of use occurred in Saskatchewan (Table 8; Fig. 9). Cranes of
WC–A affiliation were distributed across 2 major areas in
Saskatchewan, that is, from the Quill Lakes region, through
the Quill plains and Kutawagan basin south to Last Mountain
Lake, and adjacent to the South Saskatchewan River from Luck
Lake to Saskatoon (Fig. 10). Comparatively, 47% and 52% of fall
staging use by NC–N cranes was in North Dakota and
Saskatchewan, respectively (Table 8). Sites most used by NC–N
cranes in Saskatchewan were near the South Saskatchewan River
in the Outlook and the Quill-Kutawagan-Last Mountain Lake
areas (Fig. 10). The 16% of MCP fall use that occurred in North
Dakota was split between EC–M (49%), NC–N (47%), and
WC–A (4%; Table 8) affiliations, with NC–N use centered in
Burke, Ward, Kidder, and Emmons counties. Of the 10% of
MCP use occurring in Manitoba and the 3% of MCP use
occurring in Minnesota during fall staging, all but one crane
Figure 8. Median weekly locations (e.g., 1–7 Jan ¼week 1, etc.) of Platform Transmitting Terminal (PTT)-tagged MCP sandhill cranes of the Western Alaska–Siberia
(WA–S) and Northern Canada–Nunavut (NC–N) breeding affiliations between departure from the breeding grounds and arrival on the wintering grounds, 1998–2003.
BG ¼breeding ground, WG ¼wintering ground. Numbers within symbols indicate week(s) birds were located at a site.
Table 8. Distribution of known hunting-exposure days for Platform Transmitting Terminal (PTT)-tagged sandhill cranes from the Mid-continent Population in each
state and province by breeding affiliation during fall staging in the northern plains, 1998–2003.
Breeding affiliation n
No. of exposure days by state and province
ND % MN % SD % SK % MB % AB % MT %
Western Alaska–Siberia 31 622 96.7 18 2.8 3 0.5
Northern Canada–Nunavut 16 176 47.4 3 0.8 192 51.8
West-central Canada–Alaska 29 27 2.0 1,287 96.5 3 0.2 16 1.2
East-central Canada–Minnesota 24 408 38.3 154 14.4 20 1.9 1 0.1 483 45.3
Totals
a
495 16.3 92 3.0 16 0.5 2,099 69.2 290 9.6 37 1.2 4 0.1
a
MCP representative sample: Adjusted to represent correct breeding affiliation proportions.
16 Wildlife Monographs 175
(a WC–A greater sandhill crane) were of the EC–M breeding
affiliation (Fig. 5). Overall, about 38% and 45% of fall locations of
EC–M cranes were in North Dakota and Manitoba, respectively
(Table 8). Major staging locations of EC–M cranes in North
Dakota included Bottineau, Eddy, Sheridan, Stutsman, Wells,
and Kidder counties and in Manitoba, the Whitewater–Oak
Lake area and Souris River Valley.
During their fall stay in the northern plains EC–M cranes
occurred over the largest area of the 4 breeding affiliations
(Table 9); no overlap occurred with WA–S cranes and minimal
overlap occurred with WC–A cranes (Table 10). Overlap
between EC–M and NC–N cranes was limited to North
Dakota, mostly in Kidder County (Fig. 10); overlap of 95%
KDEs in the state accounted for <3% of their combined use
areas, and probability of an individual from one breeding
affiliation occurring in the other’s area of use was 0.07
(Table 10).
Cranes of WA–S affiliation staged over the smallest area of the 4
breeding affiliations (Table 9). Overlap with NC–N and WC–A
affiliations occurred only in west-central Saskatchewan (Fig. 10).
Probability of an individual from WA–S affiliation occurring in
the fall use area of NC–N cranes was extremely low
(PHR
NC–N,WA–S
¼0.07). Cranes of WC–A and WA–S breeding
affiliations overlapped approximately 5% of their combined 95%
KDEs, with probability of a crane of the WA–S breeding affili-
ation occurring in a WC–A crane fall use area being less than the
opposite (PHR
WC–A,WA–S
¼0.09 vs. PHR
WA–S,WC–A
¼0.16)
reflecting WA–S cranes having a smaller fall use area centered
more to the west (Table 9; Fig. 10). Cranes of NC–N and WC–A
affiliations overlapped broadly in east-central and central
Figure 9. Distributions of Platform Transmitting Terminal (PTT)-tagged sandhill cranes of the Mid-continent Population by breeding affiliation (Western Alaska–
Siberia [WA–S], Northern Canada–Nunavut [NC–N], West-central Canada–Alaska [WC–A], and East-central Canada–Minnesota [EC–M]) during fall staging in
south-central Canada and the north-central United States, 1998–2003. We color-coded PTT-locations by breeding affiliation.
Table 9. Areas (km
2
) occupied by Platform Transmitting Terminal (PTT)-tagged sandhill cranes of the 4 breeding affiliations during fall staging in the northern plains
and during winter in Texas, 1998–2004.
Breeding affiliation
Kernel-density estimate of use area (km
2
)
Fall staging Winter
95% 75% 50% n95% 75% 50% n
Western Alaska–Siberia 9,306 3,483 1,158 34 19,255 7,999 2,138 24
Northern Canada–Nunavut 14,820 6,637 1,601 17 25,397 11,932 3,631 16
West-central Canada–Alaska 16,788 4,844 1,499 34 33,424 11,535 3,601 28
East-central Canada–Minnesota 24,002 9,034 2,498 25 22,497 6,532 2,184 22
Krapu et al. Geographic Distribution of Sandhill Cranes 17
Table 10. Measures of magnitude of overlap in areas occupied by sandhill cranes of the 4 breeding affiliations during fall staging in the northern plains and during winter
in Texas 1998–2004.
Breeding affiliations
a
Fall staging Winter
KDE overlap
b
Home range overlap
c
KDE overlap
b
Home range overlap
c
95% 75% 50% PHR
i
,
j
PHR
j
,
i
UDOI 95% 75% 50% PHR
i
,
j
PHR
j
,
i
UDOI
WA–S:NC–N 3.1 0.8 0.0 0.13 0.07 0.01 23.3 20.8 14.7 0.42 0.71 0.55
WA–S:WC–A 5.1 3.9 0.6 0.16 0.09 0.02 11.1 4.5 <0.1 0.19 0.49 0.08
WA–S:EC–M 0.0 0.0 0.0 0.00 0.00 0.00 1.0 0.0 0.0 0.02 0.03 0.00
NC–N:WC–A 9.5 8.6 8.0 0.24 0.32 0.10 20.7 14.5 6.2 0.38 0.54 0.22
NC–N:EC–M 2.9 2.0 <0.1 0.05 0.07 0.00 3.2 0.9 0.0 0.04 0.06 0.00
WC–A:EC–M 0.2 0.0 0.0 0.01 0.00 0.00 10.0 4.8 0.2 0.21 0.30 0.06
a
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
b
Percentage of overlap area divided by combined kernel-density estimate (KDE) areas.
c
PHR
i
,
j
¼Probability of finding a crane from breeding affiliation jin breeding affiliation i’s home range. PHR
j
,
i
¼Probability of finding a crane from breeding
affiliation iin breeding affiliation j’s home range. UDOI ¼Space-use sharing index (0 ¼no overlap, 1¼complete overlap) for 2 home ranges (Fieberg and Kochanny
2005).
Figure 10. Spatial relationships among MCP sandhill cranes from the 4 breeding affiliations (Western Alaska–Siberia [WA–S], Northern Canada–Nunavut (NC–N],
West-central Canada–Alaska [WC–A], and East-central Canada–Minnesota [EC–M]), while on fall staging areas in the northern plains, 1998–2003. We based spatial
relationships among the 4 breeding affiliations on 75% kernel density estimates (KDE).
18 Wildlife Monographs 175
Saskatchewan (Fig. 10) with each affiliation occurring in similar
proportions over each other’s range (Table 10).
Distances between arithmetic mean locations of each of the 4
breeding affiliations while on fall staging areas in the northern
plains ranged from 313 km between WA–S and WC–A affili-
ations to 1,022 km between WA–S and EC–M affiliations
(Table 11). Distances separating WA–S from NC–N affiliations
(604 km) was nearly twice the distance separating WA–S from
WC–A affiliations (313 km). Cranes of WC–A and EC–M
affiliations ranked second highest in distance separation
(731 km) associated with WC–A cranes staging largely in
Saskatchewan and EC–M cranes staging only in Manitoba,
Minnesota, and North Dakota (Fig. 9).
The space-use sharing indices for the 4 breeding affiliations on
fall staging areas indicated patterns similar to other measure-
ments of overlap. Indices of space-use sharing among breeding
affiliations showed no overlap between WA–S and EC–M affili-
ations, WC–A and EC–M affiliations, or NC–N and EC–M
affiliations and only minimal sharing between WA–S and NC–N
or WC–A affiliations. Except for NC–N and WC–A affiliations,
probability of finding a crane from one breeding affiliation within
the home range of a crane from another breeding affiliation was
extremely low (P¼0–0.16) during fall staging (Table 10).
Chronology of Migration From Fall Staging Areas to
Wintering Grounds
Length of stay in the northern plains by cranes from WA–S and
NC–N and from EC–M and WC–A averaged about 4 weeks and
7 weeks, respectively, with WC–A and EC–M affiliations spend-
ing 25 days and 20 days longer than WA–S and NC–N affili-
ations, respectively (Table 7). Cranes in all 4 breeding affiliations
moved quickly to their wintering grounds, averaging 5 days en
route (Table 7).
Cranes of WC–A affiliation began their migration to wintering
grounds in WK 42, when their median location was centered
in northwestern Kansas, about 1,304 km southeast of their
median location the previous week (Fig. 7). They arrived on
their wintering grounds in western Texas in WK 43 (Fig. 7).
Cranes of EC–M affiliation departed fall staging areas centered
in North Dakota on 24 October and arrived on wintering
grounds on 27 October. However, EC–M and WC–A cranes
together spent an average of 10.9 days and 14.0 days in Kansas
and Oklahoma (considered part of their winter range), respect-
ively, before continuing on to wintering grounds in Texas.
During 22 October–11 November (WKs 43, 44, and 45), weekly
median locations of EC–M cranes moved 626 km, 486 km,
and 68 km as the center of their distribution moved through
Nebraska, to southern Kansas, to northern Oklahoma, respect-
ively (Fig. 7). These movements indicated a slowing of migration
as some birds stopped in areas such as the Cheyenne Bottoms
Wildlife Management Area (WMA), Quivira NWR, and
near the Red River along the border of Texas and Oklahoma.
On average, WC–A and EC–M cranes flew 1,986 km and
2,025 km from their primary fall staging areas to arrival at winter-
ing areas.
Cranes of WA–S affiliation departed their staging areas, on
average, by 13 October (Table 7) and by 15–21 October (WK
42); their median distribution was centered in eastern New
Mexico (Fig. 8). For NC–N cranes, the median weekly location
moved slowly southward from 17 September to 13 October
remaining in the northern plains (WKs 38–41; Fig. 8). When
departing from Saskatchewan for wintering grounds, some NC–
N cranes stopped briefly in North Dakota, whereas others con-
tinued to the wintering grounds. Cranes of NC–N affiliation
began migrating from fall staging areas on 15 October and moved
quickly, averaging 3 days to areas defined as wintering grounds
Table 11. Distance (in km) between centers of fall staging areas in the northern plains and wintering grounds in Texas by the 4 breeding affiliations based on distribution
of PTT locations 1998–2004. Center of use represents the arithmetic mean of all locations for a breeding affiliation.
Breeding affiliation comparison
a
Fall staging areas Texas wintering grounds
WA–S:NC–N 604 79
WA–S:WC–A 313 240
WA–S:EC–M 1,022 626
NC–N:WC–A 324 167
NC–N:EC–M 418 552
WC–A:EC–M 731 386
a
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
Table 12. Distribution of known hunting-exposure days for Platform Transmitting Terminal (PTT)-tagged sandhill cranes from the Mid-continent Population by
breeding affiliation while settled on the wintering grounds in each state in south-central and southwestern United States and Mexico, 1998–2004. We calculated
exposure days only from tagged birds that we monitored from arrival at the wintering grounds through departure on spring migration after their first winter following
tagging.
Breeding affiliation n
Exposure days by state (U.S. and Mexico)
TX % NM % OK % KS % AZ % COA % CHH % TAM %
Western Alaska–Siberia 20 1,894 72.5 28 1.1 111 4.2 86 3.3 493 18.9
Northern Canada–Nunavut 11 1,444 9.9 1 <0.1 1 <0.1
West-central Canada–Alaska 24 2,830 88.6 1 <0.1 101 3.2 57 1.8 108 3.4 96 3.0
East-central Canada–Minnesota 21 1,808 73.8 325 13.3 268 10.9 48 1.9
Totals
a
8,087 82.3 45 0.5 242 2.5 181 1.8 178 1.8 215 2.2 789 8.0 93 0.9
a
Projected MCP exposure: Sample adjusted to represent correct breeding affiliation proportions.
Krapu et al. Geographic Distribution of Sandhill Cranes 19
in western Texas (Table 7). Cranes of WA–S and NC–N affili-
ations flew 1,900 km and 1,460 km, respectively, from their last
median locations on fall staging areas to arrival on or near their
wintering areas.
Use of Wintering Grounds
We estimated that 82% of crane winter use occurred in Texas and
remaining use was split between Oklahoma, Kansas, eastern and
west-central New Mexico, southeastern Arizona, and the states
of Chihuahua, Coahuila, and Tamaulipas in northern Mexico
(Table 12; Fig. 11). Cranes of NC–N, WA–S, WC–A, and EC–
M affiliations spent 99%, 73%, 64%, and 7%, respectively, of
winter in western Texas in Hunting Zone A (Table 13; see Texas
Parks and Wildlife Department Program [2010] for locations of
sandhill crane hunting zones in TX). Cranes from WC–A win-
tered mostly in Texas (89%; Table 12) with 81% of use occurring
in Hunting Zones A and B (Table 13). Remaining use was in
Zone C (18%) and the area closed to hunting (<1%). Cranes
from EC–M wintered primarily in Texas (74%), followed by
Oklahoma (13%), Kansas (11%), and Tamaulipas (2%; Table 12).
Within Texas, EC–M crane use occurred primarily along the
Gulf Coast in Zone C (68%), and the Texas High Plains and
Rolling Plains in Zones A and B (17%); remaining use occurred
in the area closed to hunting (14%; Table 13).
Figure 11. Distribution of Platform Transmitting Terminal (PTT)-tagged sandhill cranes of the Mid-continent Population by breeding affiliation (Western Alaska–
Siberia [WA–S], Northern Canada–Nunavut [NC–N], West-central Canada–Alaska [WC–A], and East-central Canada–Minnesota [EC–M]) while on wintering
grounds in the south-central and southwestern United States and northern Mexico, 1998–2004. Boundaries of Hunting Zones A, B, and C are included for Texas. We
color-coded PTT-locations by breeding affiliation.
Table 13. Distribution of hunting-exposure days for Platform Transmitting Terminal (PTT)-tagged sandhill cranes of the 4 breeding affiliations by hunting zone in
Texas through 10 March, 1998–2004. We include only birds monitored from arrival on wintering grounds through spring departure.
Breeding affiliation
Hunting zone
a
A%
b
B%
b
C%
b
Closed %
b
Western Alaska–Siberia 2,082 67.5 0 0 0
Northern Canada–Nunavut 1,414 69.7 1 100 0 0
West-central Canada–Alaska 2,053 69.6 266 69.5 527 21.4 11 0
East-central Canada–Minnesota 162 61.1 148 20.9 1,224 27.2 259 0
a
Zone A ¼western Texas panhandle region, zone B ¼central Texas region, zone C ¼southeastern Texas, coastal region, Closed ¼northeastern Texas closed region
and small coastal portion of zone C. See Texas Parks and Wildlife Department Program (2010) for more detailed information.
b
Percentage of stay that birds were susceptible to harvest.
20 Wildlife Monographs 175
Mean distances between centers of distribution of each of the 4
breeding affiliations on Texas wintering grounds varied widely,
reflecting differing degrees of spatial separation among breeding
affiliations (Table 11). Distribution of the WA–S affiliation was
centered farthest west of the 4 breeding affiliations in Hunting
Zone A in western Texas (Fig. 12A). Overlap of NC–N and
WA–S affiliations (75% KDE) was centered in the west-central
part of Hunting Zone A (Fig. 12A) and declined eastward where
NC–N and WC–A affiliations overlapped broadly (Fig. 12B). In
Zone B, few WA–S and NC–N cranes occurred, with overlap
restricted to WC–A and EC–M affiliations (Table 13). In Texas,
WC–A cranes occupied the largest area; NC–N, WA–S, and
EC–M cranes ranged over 24%, 42%, and 33% less area than did
WC–A (Table 9). The greatest overlap in area of use occurred
between the NC–N and WC–A affiliations (12,199 km
2
), but the
greatest percentage of overlap was between WA–S and NC–N
affiliations (Table 10). The difference between overlap of use
areas versus magnitude of geographic overlap is better illustrated
by the space-use sharing index, which indicated that NC–N and
WA–S affiliations had an overlap value >2 times that of NC–N
and WC–A affiliations (Table 10). Probability of finding an
individual from the NC–N breeding affiliation in WA–S crane
home range was nearly equivalent to the probability of finding the
bird in WC–A crane home range (PHR
WA–S,NC–N
¼0.42 vs.
PHR
WC–A,NC–N
¼0.54), whereas probability of an individual
WA–S crane occurring in the home range of the NC–N affili-
ation was 1.9 times greater than a WC–A crane occurring there
(PHR
NC–N,WA–S
¼0.71 vs. PHR
NC–N,WC–A
¼0.38; Table 10).
Distribution of the EC–M breeding affiliation overlapped least
with the other 3 breeding affiliations in Texas (Table 10) prim-
arily because most EC–M cranes wintered along the Gulf Coast
where few NC–N and WA–S cranes occurred (Fig. 11). The
EC–M crane distribution during winter was concentrated along
the upper Texas Gulf Coast, whereas WC–A cranes occurred
primarily along the lower Texas Gulf Coast (Fig. 11). In Hunting
Zone B, WC–A and EC–M cranes accounted for 64% and 36%
of use, respectively (Table 13). Outside of Texas, there was little
overlap among breeding affiliations on wintering grounds as
WA–S cranes accounted for nearly all exposure days in New
Mexico, Arizona, Chihuahua (Table 12), and presumably other
states in Mexico where some MCP lesser sandhill cranes occur
(Durango and Zacatecas; Tacha et al. 1994) but no tagged birds
settled. Cranes of EC–M affiliation accounted for most winter
locations in Oklahoma and Kansas (Table 12).
Seasonal Exchange Between Gulf Coast and Western
Subpopulations
The MCP was managed as 2 subpopulations, that is, the Gulf
Coast Subpopulation and the Western Subpopulation, with each
having ranges as defined by Tacha et al. (1994:85; Fig. 13).
Seventy-one and 100% of tagged NC–N migrated from breeding
grounds of the Gulf Coast Subpopulation to fall staging areas
(central ND and Saskatchewan) and wintering grounds (western
TX) of the Western Subpopulation (Table 14), respectively. For
EC–M, 18% and 21% migrated from breeding grounds of the
Gulf Coast Subpopulation to fall staging areas (central ND) and
Figure 12. Spatial relationships between (A) sandhill cranes from Northern Canada–Nunavut (NC–N) and Western Alaska–Siberia (WA–S) breeding affiliations
while on wintering grounds in Hunting Zone A in Western Texas and (B) sandhill cranes from NC–N and West-central Canada–Alaska (WC–A) breeding affiliations
while on wintering grounds in Hunting Zones A and B in Western Texas, 1998–2004. Contour lines represent 95% kernel density estimates (KDE) and shaded areas
represent areas where 75% KDEs of breeding affiliations intersect each other.
Krapu et al. Geographic Distribution of Sandhill Cranes 21
wintering grounds (western TX) of the Western Subpopulation,
respectively.
Tagged cranes from the Western Subpopulation were less
prone to move between subpopulation ranges (Table 14). No
WA–S cranes switched to fall staging areas or wintering grounds
of the Gulf Coast Population as they migrated through Montana,
Wyoming, and Colorado (Fig. 6C), states that lie to the west of
the fall staging areas and wintering grounds of the Gulf Coast
Subpopulation. In winter, WA–S cranes stayed in western Texas,
New Mexico, Arizona, and Mexico (Fig. 11). Of WC–A cranes,
29% moved onto the winter range of the Gulf Coast
Subpopulation on the Texas Gulf Coast (Table 14).
Exposure to Hunting Seasons
Tagged sandhill cranes from WA–S, NC–N, WC–A, and EC–
M breeding affiliations occurred in zones open to hunting during
anestimated68%,69%,68%,and44%,respectively,ofthefall
migration and winter period. Most exposure to hunting seasons
occurred in Texas (54%) and Saskatchewan (23%), with North
Dakota, Chihuahua, and Manitoba combining for an additional
Figure 13. Breeding and wintering ranges of the Mid-continent Population of sandhill cranes in North America as estimated by Tacha et al. (1994:85) and our results.
Areas showing cross-hatching represent parts of the range where the known distribution expanded as a result of monitoring settling patterns of Platform Transmitting
Terminal (PTT)-tagged birds during 1998–2004. We adapted breeding and wintering ranges depicted by gray shading from Tacha et al. (1994) as the black lines that
identify the estimated boundaries of the breeding and wintering ranges of the Western Subpopulation and the Gulf Coast Subpopulation.
22 Wildlife Monographs 175
14% (Table 15). Tagged cranes during fall stopovers in
Manitoba, Saskatchewan, Alaska, Colorado, Montana, North
Dakota, South Dakota, and Chihuahua occurred mostly within
areas open to hunting but length of crane stay varied widely
(Table 15). Closed areas existed in several states that have sand-
hill crane hunting seasons (CO, KS, ND, NM, SD, TX, and
WY) but with the exception of closed areas in Texas, received
limited use.
Sport hunting of the MCP in subarctic and arctic regions was
limited to Alaska. All WA–S cranes migrated across Alaska
(Fig. 6C) during the hunting season and all crane use occurred
in areas open to hunting (Table 16), but access to the birds was
limited by a restricted network of roads. We surveyed the agricul-
tural area near Delta Junction, one of the primary WA–S hunting
areas in Alaska, in early September 2007 and found <500 cranes
using agricultural lands (G. Krapu, unpublished data). Only
Table 14. Use of Gulf Coast and Western fall staging and wintering grounds (Tacha and Vohs 1984) by Platform Transmitting Terminal (PTT)-tagged sandhill cranes
of the 4 breeding affiliations, 1998–2004.
Breeding affiliation
a
% of use
Fall staging areas Wintering grounds
n
Gulf
Coast
range
b
Nos. of sandhill
cranes contributing
locations to each
estimate of % of use
Western
range
b
Nos. of sandhill
cranes contributing
locations to each
estimate of % of use n
Gulf
Coast
range
b
Nos. of sandhill
cranes contributing
locations to each
estimate of % of use
Western
range
b
Nos. of sandhill
cranes contributing
locations to each
estimate of % of use
Subpopulation
c
Gulf Coast
EC–M 26 82% 25 18% 12 25 79% 24 21% 12
NC–N 15 29% 8 71% 13 13 0% 0 100% 13
Total 41 72% 33 28% 25 38 58% 24 42% 25
Western
WC–A 37 1% 2 99% 37 33 29% 13 71% 27
WA–S 36 0% 0 100% 36 33 0% 0 100% 33
NC–N 4 5% 1 95% 4 4 0% 0 100% 4
Total 77 1% 3 99% 77 70 14% 13 86% 64
a
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
b
As defined by subpopulation distribution delineated by Tacha and Vohs (1984).
c
As defined by breeding distribution of subpopulations delineated by Tacha and Vohs (1984) and where our PTT-tagged cranes settled for the breeding season.
Table 15. Relative exposure of Platform Transmitting Terminal (PTT)-tagged MCP sandhill cranes to sport hunting during fall and winter by state and province of the
Central Flyway.
a
We present data as mean number of days and % of exposure to hunting and non-hunting by state and province from 1 September through 10 March
during 1998–2003 (n¼no. of PTT-tagged sandhill crane migrations in sample).
State or province
Mean exposure (days SD)
Hunt Non-hunt Percent of total exposure
xSD n x SD nHunt Non-hunt
Alberta 0.0 0.0 0 3.4 4.7 39 0.0 2.9
Alaska 7.5 6.0 38 0.0 0.0 0 3.3 0.0
Arizona 9.0 8.5 2 60.5 50.2 2 0.2 2.6
British Columbia 0.0 0.0 0 1.3 0.7 15 0.0 0.4
Chihuahua 70.8 45.8 6 12.8 12.0 9 4.9 2.5
Colorado 1.5 1.5 23 1.0 0.0 1 0.4 <0.1
Coahuila 0.0 0.0 0 87.0 11.3 2 0.0 3.8
Kansas 12.9 12.2 10 2.6 2.6 38 1.5 2.1
Manitoba 29.9 16.2 13 3.1 3.5 10 4.5 0.7
Minnesota 0.0 0.0 0 25.0 23.9 4 0.0 2.2
Montana 1.1 0.5 15 1.5 0.7 2 0.2 0.1
North Dakota 9.2 10.6 43 3.9 6.6 8 4.5 0.7
Nebraska 0.0 0.0 0 2.1 2.5 40 0.0 1.8
New Mexico 4.6 5.4 5 7.2 17.2 16 0.3 2.5
Nunavut 0.0 0.0 0 9.5 5.4 17 0.0 3.5
Northwest Territories 0.0 0.0 0 4.1 4.3 9 0.0 0.8
Oklahoma 19.3 24.0 12 5.3 7.8 22 2.7 2.6
Ontario 0.0 0.0 0 9.8 7.5 4 0.0 0.9
South Dakota 1.5 1.0 31 1.3 0.6 3 0.5 0.1
Siberia 0.0 0.0 0 5.7 5.2 27 0.0 3.3
Saskatchewan 24.3 17.5 82 1.3 0.5 4 23.0 0.1
Tamaulipas 0.0 0.0 0 96.0 0.0 1 0.0 2.1
Texas 66.9 33.0 70 38.1 27.9 76 53.9 63.4
Wyoming 1.3 0.5 10 1.0 0.0 1 0.1 <0.1
Yukon Territory 0.0 0.0 0 1.3 0.7 27 0.0 0.8
a
Only birds from representative sample (birds marked in 2003 excluded).
Krapu et al. Geographic Distribution of Sandhill Cranes 23
Table 16. Mean number of days (and %) exposure of Platform Transmitting Terminal (PTT)-tagged sandhill cranes
a
to hunting and non-hunting by breeding
affiliation and hunting unit in states (U.S., Mexico) and provinces (Canada) where hunting seasons took place, from 1 September through 10 March, 1998–2004.
State or province Hunting unit Breeding affiliation
b
Hunting Non-hunting Combined
n
c
Mean exposure days % nMean exposure days % nTotal exposure days
Alaska All Units WA–S 36 7.6 100 0 0.0 0.0 36 272
WC–A 2 6.0 100 0 0.0 0.0 2 12
Arizona Units 30A, 30B, 31, 32 WA–S 2 9.0 12.9 2 60.5 87.1 2 139
Chihuahua
d
WA–S 6 70.8 78.7 9 12.8 21.3 9 540
Colorado Central Flyway Area WA–S 20 1.6 97.0 1 1.0 3.0 20 33
NC–N 1 1.0 100 0 0.0 0.0 1 1
WC–A 2 1.0 100 0 0.0 0.0 2 2
Kansas Closed Area EC–M 0 0.0 0.0 1 2.0 100 1 2
W. of I-35, I-135, US-81 WA–S 0 0.0 0.0 6 1.0 100 6 6
NC–N 1 4.0 22.2 8 1.8 77.8 8 18
WC–A 2 19.0 51.4 15 2.4 48.6 15 74
EC–M 13 13.8 56.5 19 7.3 43.5 21 317
Manitoba Closed Area NC–N 0 0.0 0.0 8 2.4 100 8 19
EC–M 0 0.0 0.0 4 4.0 100 4 16
GBHZ 3 & 4, GHZ 6 & 6A EC–M 20 24.6 100 0 0.0 0.0 20 491
Montana Montana 1 WA–S 14 1.1 100 0 0.0 0.0 14 16
WC–A 1 1.0 100 0 0.0 0.0 1 1
Montana 2 (Sheridan Co.) WC–A 0 0.0 0.0 2 1.5 100 2 3
New Mexico Closed Area WA–S 0 0.0 0.0 7 1.9 100 7 13
NC–N 0 0.0 0.0 1 1.0 100 1 1
Eastern WA–S 3 6.7 50.0 9 2.2 50.0 10 40
WC–A 1 1.0 100 0 0.0 0.0 1 1
EV
e
WA–S 0 0.0 0.0 2 1.5 100 2 3
MRGV
e
WA–S 0 0.0 0.0 1 16.0 100 1 16
SW
e
WA–S 1 2.0 3.1 1 62.0 96.9 1 64
North Dakota Closed Area
f
NC–N 0 0.0 0.0 1 2.0 100 1 2
Zone 1 (W of US-281) NC–N 17 13.8 98.3 2 2.0 1.7 17 239
WC–A 19 2.2 100 0 0.0 0.0 19 41
EC–M 20 13.7 89.3 5 6.6 10.7 20 307
Zone 2 (E of US-281) EC–M 3 4.0 10.0 6 18.0 90.0 8 120
Oklahoma W of I-35 WA–S 1 1.0 25.0 3 1.0 75.0 4 4
NC–N 0 0.0 0.0 3 1.0 100 3 3
WC–A 4 20.8 70.3 9 3.9 29.7 12 118
EC–M 18 13.2 63.9 16 8.4 36.1 23 371
Saskatchewan N & S Game Bird Districts WA–S 38 19.2 100 0 0.0 0.0 38 729
NC–N 15 14.0 98.1 3 1.3 1.9 15 214
WC–A 38 37.1 100 0 0.0 0.0 38 1,408
EC–M 0 0.0 0.0 1 1.0 100 1 1
South Dakota Closed Area EC–M 0 0.0 0.0 5 2.0 100 5 10
W of US-281 WA–S 8 1.0 100 0 0.0 0.0 8 8
NC–N 7 1.3 100 0 0.0 0.0 7 9
WC–A 12 1.4 100 0 0.0 0.0 12 17
EC–M 12 2.3 100 0 0.0 0.0 12 28
Texas Closed Area WC–A 0 0.0 0.0 3 3.7 100 3 11
EC–M 0 0.0 0.0 7 37.0 100 7 259
Zone A WA–S 21 80.4 68.9 26 29.3 31.1 28 2,451
NC–N 16 79.0 70.2 15 35.8 29.8 17 1,801
WC–A 22 69.3 69.3 26 26.0 30.7 26 2,201
EC–M 6 17.7 61.6 3 22.0 38.4 7 172
Zone B NC–N 1 1.0 100 0 0.0 0.0 1 1
WC–A 6 36.0 70.1 15 6.1 29.9 17 308
EC–M 5 19.8 43.6 15 8.5 56.4 17 227
Zone C WC–A 6 25.2 24.4 8 58.5 75.6 8 619
EC–M 13 25.6 27.1 16 56.1 72.9 16 1,230
Wyoming Closed Area WA–S 0 0.0 0.0 1 1.0 100 1 1
Central Flyway (Area 7) WA–S 10 1.3 100 0 0.0 0.0 10 13
a
Includes cranes tagged in all years (1998–2003) for better comparison within breeding affiliations.
b
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
c
n¼no. of PTT-tagged sandhill crane migrations in sample.
d
Data was not available on recent hunting history and season dates for Coahuila and Tamaulipas.
e
Estancia Valley (EV), Middle Rio Grande Valley (MRGV), Southwest NM (SW), all primarily RMP hunting areas.
f
North Dakota east of U.S. 281 was closed prior to 2001.
24 Wildlife Monographs 175
about 2,025 ha of the agricultural area remained in grain pro-
duction in 2007; most of the rest was in the Conservation Reserve
Program, abandoned, or being used for hay production (C. Hadley,
U.S. Department of Agriculture, personal communication).None
of our tagged cranes stopped in the Delta Junction area during fall
migration. Length of stay in Alaska by tagged cranes during the
hunting season averaged about a week (Table 15), with cranes
generally departing the state by mid-September.
During fall staging, most tagged cranes were exposed to hunt-
ing seasons throughout most or all of their stay (Table 15) except
in Minnesota. In Manitoba, EC–M crane use occurred primarily
in areas open to hunting during the hunting period (MB Units 3
and 4; Table 16). The lack of exposure of tagged NC–N cranes in
Manitoba to hunting seasons reflects that the birds flew over the
northwestern part of the province (Fig. 6B), which is closed to
sport hunting of cranes. In Saskatchewan, where virtually the
entire province is open to crane hunting, NC–N, WA–S, and
WC–A breeding affiliations were exposed to hunting seasons
almost 100% of their stay (Table 16). In North Dakota, NC–N
cranes occurred only in Zone 1, whereas about 28% of EC–M
crane use occurred in Zone 2 (Table 16, Fig. 14). Hunting Zone 1
included all lands in the state located west of U.S. Highway 281
and Hunting Zone 2 encompassed the rest of the state. Cranes of
EC–M affiliation were exposed to hunting seasons 89% of their
stay in Zone 1 and 10% of their stay in Zone 2; NC–N cranes
were exposed to hunting seasons about 98% of their stay in North
Dakota. After departing fall staging areas, WA–S and NC–N
crane affiliations encountered limited exposure to hunting en
route to the wintering grounds due to their rapid migration. In
Kansas and Oklahoma, WC–A and EC–M affiliations accounted
for 94% and 99% of all crane use with 56% and 54% occurring
during hunting periods. In Oklahoma, WC–A and EC–M affili-
ations were located in the hunted area (Fig. 14) 70% and 64% of
the hunting period, respectively (Table 16).
Level of exposure of MCP cranes to hunting seasons varied
widely across their winter range. In Texas, the 4 breeding affili-
ations were exposed to hunting seasons 62–70% of their stay
while in Hunting Zone A (Table 16). In Hunting Zone B, EC–
M and WC–A cranes were exposed to hunting seasons 44% and
70% of their stay (Table 16). In Hunting Zone C, 24% and 27%
of use by WC–A and EC–M cranes occurred during the hunting
period (Table 16). In New Mexico, WA–S was the primary MCP
breeding affiliation (Fig. 11), with minimal use occurring in the
closed areas (i.e., an average of 1.9 days; Table 16). Most use by
WA–S cranes in New Mexico occurred in areas also occupied by
the RMP and overall, average WA–S crane exposure was 5.5 days
and 7.6 days during the MCP and RMP hunting and non-
hunting periods, respectively (Table 16, Appendix B). In
Arizona, WA–S crane use occurred primarily during the non-
hunting period and within areas also occupied by the RMP
Figure 14. Distribution of Platform Transmitting Terminal (PTT)-locations of tagged MCP sandhill cranes of the Mid-continent Population of sandhill cranes by
hunting unit and zone during stopovers in: (A) North Dakota; (B) Kansas; (C) Oklahoma; and (D) Texas, 1998–2004. We color-coded PTT-locations by breeding
affiliation. Shaded areas identify parts of states where sport hunting is allowed.
Krapu et al. Geographic Distribution of Sandhill Cranes 25
(Table 16). Use by WA–S cranes in Mexico was highest in
Chihuahua (Table 12; Fig. 14) and occurred primarily during
the hunting season (Table 16, Appendix B).
MCP Harvest by Breeding Affiliation
Sport harvest of MCP cranes during 1998–2003 was concen-
trated in Texas, Saskatchewan, and North Dakota (Table 17).
We estimate that WA–S, NC–N, WC–A, and EC–M affilia-
tions accounted for 30%, 20%, 29%, and 21% of the harvest,
respectively (Table 17), indicating harvest of WC–A and EC–M
affiliations were disproportionately high to their percentages in
the MCP (Table 4). For WA–S cranes, sport harvest was con-
centrated in Saskatchewan, Texas, and possibly Chihuahua, but
detailed information is lacking on size of harvest in Mexico
(Table 17). For WC–A cranes, Saskatchewan accounted for
an estimated 57% of the harvest, followed by Texas at 32%.
We estimate Hunting Zone A in Texas accounted for about
20% of total MCP crane harvest, and 20%, 3%, and 5% of the
MCP harvest of the WC–A affiliation occurred in Hunting
Zones A, B, and C, respectively. In Saskatchewan, harvest of
WC–A cranes was centered near the South Saskatchewan River
by Outlook, and in the Quill Lakes, Kutawagan Lake, and Last
Mountain Lake areas (Fig. 10). For the EC–M breeding affili-
ation, we estimated that North Dakota, Texas, Manitoba, and
Kansas accounted for an estimated 38%, 22%, 19%, and 13% of
the harvest, respectively. In North Dakota, approximately 89% of
the estimated EC–M crane harvest occurred in Hunting Zone 1
(Fig. 14A), along with an estimated 100% of the harvest of NC–
N cranes. Harvest composition in Hunting Zone 1 was estimated
to be: NC–N cranes (60%), EC–M cranes (34%), and WC–A
cranes (6%). In Texas, we estimated about 89% and 16% of EC–
M and WC–A crane harvest occurred in Hunting Zone C,
respectively.
In Canada, we estimated 87% of the MCP harvest occurred in
Saskatchewan, with the remaining 13% in Manitoba
(Table 17). In Saskatchewan, 62% of harvest was of WC–A
cranes (Table 17), whereas all harvest in Manitoba was com-
posed of EC–M cranes (Table 17). In Saskatchewan, harvest
was highest in the eastern region (56%), followed by the west-
ern region (20%) and the central region (17%); harvest of WC–
A cranes was disproportionately high to their percentage in the
MCP population.
DISCUSSION
Extent of Breeding Distribution
Geographic distribution of our tagged sample of MCP sandhill
cranes on the breeding grounds compared favorably with the
distribution as previously reported (Walkinshaw 1949, Johnsgard
1983, Tacha et al. 1994), suggesting our sample was representa-
tive of the geographic distribution of the MCP. Our tagged
sandhill cranes occupied all previously described major breeding
grounds of the MCP in North America and extended the breed-
ing range of the MCP in east-central Canada and the Canadian
sandhill crane morph in northwestern Canada and Alaska.
Settling patterns indicated relative abundance of tagged cranes
followed expected patterns for the MCP across the breeding
range in Canada and Alaska, based on previous work. The
Yukon Delta, for example, has long been recognized as the most
important crane breeding ground in Alaska and contains the
highest density of breeding pairs reported for the MCP
(Conant et al. 1985, Melvin et al. 1990), traits also corroborated
by settling patterns of our tagged sample. The lack of exchange of
cranes between breeding affiliations from 1 yr to the next and the
exceptional level of philopatry to sites used the previous year
suggests that our tagged sample were virtually all breeders or were
approaching breeding age.
The size of the known breeding range of the MCP increased by
about 322,000 km
2
in east-central Canada after accounting for
settling patterns of our tagged sample of cranes (Fig. 13). Much
Table 17. Estimated composition of sandhill crane harvest (1998–2003) by breeding affiliation and by state and province of the Central Flyway. Harvest estimates by
breeding affiliation assume that composition of the harvest in each state and hunting zone is proportional to the spatial and temporal distribution o f exposure by breeding
affiliation during the respective state or provincial season. Harvest estimates for MCP sandhill cranes by state and province during 1998–2003 are from Kruse et al. (2008).
Location 1998–2003 mean harvest
Estimated % harvest by breeding affiliation
a
WA–S NC–N WC–A EC–M
Alaska 830 96.0 0 4.0 0
Colorado 211 91.4 2.9 5.7 0
Kansas 1,075 0 3.1 21.7 75.2
Montana 29 94.1 0 5.9 0
New Mexico 358 95.2 0 4.8 0
North Dakota 5,231 0 48.7 6.3 45.0
Oklahoma 580 0.4 0 30.3 69.3
South Dakota 303 17.8 20.0 33.3 28.9
Texas 8,688 30.6 22.9 31.1 15.4
Wyoming 11 100 0 0 0
Saskatchewan 7,783 24.3 13.4 62.3 0
Manitoba 1,166 0 0 0 100
Mexico
b
2,638 100 0 0 0
Arizona
(RMP)c
146 100 0 0 0
New Mexico
(RMP)c
103 100 0 0 0
Total % 30.4 19.5 29.0 21.1
Total harvest 29,149 8,852 5,681 8,459 6,158
a
WA–S ¼Western Alaska–Siberia, NC–N ¼Northern Canada–Nunavut, WC–A ¼West-central Canada–Alaska, EC–M ¼East-central Canada–Minnesota.
b
Unknown harvests (Mexico) were assumed to be 10% of harvests in the U.S. and Canada.
c
Hunting areas established for harvest of the Rocky Mountain Population of sandhill cranes.
26 Wildlife Monographs 175
of the documented range increase came in the Hudson Bay
Lowlands of northern Ontario, a region where cranes had been
known to breed since the 18th century (Williams and Glover
1969) but previously was thought to be part of the breeding range
of the Eastern Population (EP; Tacha et al. 1994). The large
number of tagged cranes that settled in the Hudson Bay
Lowlands during springs 1998–2003 (Fig. 5) indicates this region
is a major breeding ground of the MCP, with most cranes being
of the Canadian morph. In northwestern Canada, Canadian
sandhill cranes previously were reported nesting in the southern
Mackenzie District of the Northwest Territories (Walkinshaw
1965, Johnsgard 1983); our results indicate Canadian sandhill
cranes now occur northward in the Great Slave Plains (Fig. 5) to
within approximately 350 km of the Arctic Ocean. In Alaska,
Canadian sandhill cranes settled in the forested interior of the
state where lesser sandhill cranes also were present.
We detected no major differences in breeding distribution of
lesser sandhill cranes across arctic North America when we
compared settling patterns of our 87 tagged lesser sandhill cranes
to the distribution reported by Walkinshaw (1981). Tagged lesser
sandhill cranes settled on all major arctic river deltas, along the
arctic mainland coast, the Canadian Archipelago, and along the
northwest side of Hudson Bay. Lesser sandhill cranes occurred
widely but at low breeding densities in the Canadian Archipelago
during the mid-20th century (Parmalee and MacDonald, 1960,
Manning and MacPherson 1961), a pattern also suggested from
our data. An exception is Banks Island, Northwest Territories,
where pair densities were high (i.e., 17 pairs/100 km
2
;
Walkinshaw 1965). Two of 4 tagged cranes that settled in the
Canadian Archipelago were on Banks Island.
Factors Influencing Breeding Distribution and Abundance
Distributions of WA–S and NC–N cranes were centered on
productive river deltas and other sites made fertile by nutrient
transport and deposition. The many tagged cranes breeding in
western Alaska is linked to the vast size and high productivity of
the Yukon Delta. This area had the highest crane nest densities
reported for the MCP (0.54 nests/km
2
in 1975 and 0.78 nests/
km
2
in 1976; Boise 1977). In northern Canada, the largest
concentration of tagged lesser sandhill cranes occurred along
the northwest coast of Hudson Bay within areas where marine
sediments were deposited early in the postglacial period
(Lumsden 1971). Distributions of tagged Canadian sandhill
cranes were centered in the fertile wetland habitats of the
Hudson Bay Lowlands, Interlake area of Manitoba, the western
boreal forest, Great Slave Lake Plains, and Yukon Flats.
The distribution of our tagged sample of cranes during the
breeding season indicates a vast area of central Canada where few
cranes settled extending from near Great Bear Lake in the
Northwest Territories to southwestern Nunavut, across northern
Saskatchewan, northern and southeastern Manitoba, western
Ontario, and northeastern Minnesota (Fig. 5). The geologic land
form of this region, the Laurentian Shield, covers an estimated
3.3 million km
2
(Fig. 5) and has long functioned as a vast natural
barrier separating breeding ranges of Canadian sandhill cranes
and lesser sandhill cranes in subarctic and arctic Canada.
Repeated continental glaciations over the past 2.5 million years
left landscapes of the Laurentian Shield with a thin soil
interspersed with rocky outcrops, resulting in lands not capable
of supporting sandhill cranes except in isolated areas where
nutrient transport and deposition produced fertile sites.
Tagged individuals that settled in the central Arctic of Canada
were all lesser sandhill cranes. However, in northwestern Canada
and interior Alaska where natural barriers comparable to the
Laurentian Shield do not exist, the Canadian sandhill crane
morph breeds near lesser sandhill crane, indicating that gene
flow between greater sandhill crane and lesser sandhill crane has
occurred in this region (Jones et al. 2005). In northern Ontario,
Lumsden (1971:289) noted ‘‘Almost all the summer records
[Canadian sandhill crane] come from the area of post-glacial
marine submergence or from post-glacial lake beds’’ whereas
lesser sandhill crane migrate over the Laurentian Shield
without stopping to breed. The exceptional philopatry to breed-
ing sites in the MCP likely has contributed to keeping Canadian
sandhill crane separated from lesser sandhill crane in the central
Arctic where vast areas of unsuitable habitat separate these sub-
species. Gene exchange also likely is impeded because spring
stopovers of EC–M and NC–N breeding affiliations in the
northern plains are widely spaced, as are migration corridors
from their staging areas to breeding grounds (G. Krapu, unpub-
lished data).
MCP sandhill cranes were extirpated from most of their breed-
ing range in temperate mid-continent North America in the late
19th and early 20th centuries due to habitat loss and uncontrolled
hunting (Walkinshaw 1949, Littlefield and Ryder 1968, Drewien
and Bizeau 1974, Johnson 1976). The area formerly occupied
included virtually all of the Prairie Pothole Region (PPR), a vast
glaciated region within the northern plains containing millions of
fertile basin wetlands of widely varying size. The PPR extends
across most of North Dakota and South Dakota east of the
Missouri River, southern Saskatchewan, and southwestern
Manitoba. Failure of tagged cranes to settle across most of the
PPR during the breeding season reflects the general absence of
breeding by sandhill cranes in most of the region. The scarcity of
recent nesting records by sandhill cranes in most of the PPR
suggests a general lack of pioneering or high mortality of cranes
that attempt to breed in the region, as wetland habitat suitable for
sandhill crane breeding remains widely available. High philopatry
to breeding sites that we documented across the current MCP
breeding range suggests pioneering is limited to young birds that
have not nested.
Failure of the MCP to re-occupy most of its former breeding
range in the PPR also may have been influenced by a harvest
through sport hunting of WC–A and EC–M cranes dispropor-
tionate to their numbers in the population. The disproportionate
harvest of WC–A and EC–M cranes in the northern plains
during 1998–2003 (Table 17) occurred in association with an
early start of the sandhill crane hunting seasons (i.e., 1 Sep in
Saskatchewan and Manitoba and 1–3 weeks later in ND; Kruse
et al. 2008). With most WC–A and EC–M cranes arriving on fall
staging areas before or soon after the start of the fall hunting
seasons (Table 7), WC–A and EC–M affiliations receive higher
exposure to hunting than do NC–N and WA–S affiliations,
which arrive approximately 2 weeks and 3 weeks later, respect-
ively. Early onset of the fall hunting seasons in the northern
plains also may be removing cranes that successfully pioneered
Krapu et al. Geographic Distribution of Sandhill Cranes 27
into the PPR but become vulnerable to hunting early in the
hunting season before joining larger groups of cranes.
Harvest of EC–M and WC–A cranes on fall staging areas likely
has been disproportionately high to their numbers in the MCP
for several decades, as fall hunting seasons began in early
September starting in Saskatchewan and Manitoba in 1964
(Central Flyway Webless Migratory Game Bird Technical
Committee 2006), but significance has grown over time as num-
ber of crane hunters and harvest have increased. Melvin and
Temple (1983) reported that 93% and 62% of VHF radio-tagged
sandhill cranes (EC–M) breeding in the Interlake area of
Manitoba arrived in North Dakota prior to or during the
1978 and 1979 hunting seasons, respectively, which began on
7–11 September. Results of Melvin and Temple (1983) indicated
a first-year hunting mortality rate in North Dakota of 14%, which
exceeds the percentage of fledged young (12%) in the Last
Mountain Lake, Kutawagan Lake, and Quill Lakes region of
Saskatchewan (Buller 1979). After NC–N cranes (virtually all
lesser sandhill crane) arrive in North Dakota during fall, they
become the principal subspecies harvested in areas where they
occur (Kendall et al. 1997), reflecting that composition of the
harvest is linked to exposure to hunting. However, because of
their shorter stay in North Dakota and Saskatchewan, NC–N
cranes accounted for a disproportionately smaller take than EC–
M and WC–A cranes.
Sport hunting of sandhill cranes in Minnesota was prohibited in
1918 with the enactment of the Migratory Treaty Act and did not
re-open until fall 2010. EC–M cranes produced in Minnesota
largely bypass fall staging areas in Manitoba and North Dakota so
experience a lower annual exposure to hunting and harvest than
those that fall stage in Manitoba and North Dakota. This lower
exposure probably has contributed to a higher breeding density of
EC–M cranes in Minnesota than in Manitoba and only
occasional breeding records in North Dakota (Fields et al. 1974).
Factors Influencing Fall Staging Distribution
Cranes stage in the northern plains in landscapes containing
cropland where birds can acquire their energy needs primarily
from waste cereal grains (Madsen 1967, Tacha et al. 1985) and
protein and calcium needs by foraging on soil macro-invert-
ebrates in pastures and hayland (Reinecke and Krapu 1986,
Krapu and Johnson 1990). Stable and plentiful supplies of
high-energy waste grain in cropland, particularly over the past
half century, likely have allowed cranes to develop stronger
staging traditions to specific sites and to stage in larger numbers
than in the past. Prior to agriculture, cranes probably had to be
more opportunistic in their fall staging patterns and winter
distribution given wide annual fluctuations in food resources
in natural prairie ecosystems.
Sandhill cranes are selective in their choice of nocturnal roost
sites, limiting the distribution of fall staging across the northern
plains. Birds roost primarily in association with shallow lakes,
mostly saline, and less frequently in wide, braided river channels
where available (e.g., the Souris River in southwestern Manitoba
and South Saskatchewan River in central Saskatchewan). Cranes
roosting in shallow lakes seek those with sparse vegetation and
open water at some distance from a bare shore, allowing high
visibility (Soine 1982).
Inundation of Missouri River bottomlands in North Dakota
and South Dakota following construction of major dams in the
1960s eliminated several important fall staging areas of sandhill
cranes (Johnson 1963). Crane displacement from some of these
areas was documented by the U.S. Fish and Wildlife Service
during aerial surveys of sandhill crane distribution upstream of
Big Bend and Oahe dams before and during the period the
reservoirs were filling with water (Buller and Boeker 1965).
From fall 1963 to fall 1964 in Hughes County of central
South Dakota upstream from Big Bend Reservoir, the number
of roosting sandhill cranes declined from 3,500 to 9 as waters
inundated roosting areas. Upstream from Oahe Dam in the
Pollock-Mobridge area, aerial surveys revealed 18,000 cranes
roosted in the shallow braided channels of the Missouri River
prior to inundation of their roosts. Northward in south-central
North Dakota, 5,000 cranes roosted on the river in fall 1964 on
river bottomlands that thereafter became part of Oahe Reservoir.
No tagged cranes used the Missouri River Valley of North
Dakota and South Dakota during our study, reflecting avoidance
by sandhill cranes of the river in its altered state. The long-term
reduction of hunting opportunities from loss of the Missouri
River roosts is apparent in South Dakota, where currently only
limited use occurred during fall migration based on information
from tagged cranes during 1998–2003.
Cranes move to other areas when traditionally used wetland
roosts in the northern plains are temporarily made unsuitable for
use because of drought or high water levels. In western
Saskatchewan, where most WA–S cranes stage during spring
in association with saline lakes, tagged cranes during springs 2001
and 2002 returned to find most of their traditionally used lakes to
be dry. Many cranes moved as far as 250 km northwest into
Alberta where they remained for the staging period (Krapu
and Brandt 2008). When suitable habitat conditions re-appeared
in western Saskatchewan staging areas after the drought ended,
most cranes returned to stage there in spring.
Factors Influencing Winter Distribution
Most tagged cranes wintered in parts of the southern and south-
western United States and northern Mexico previously identified
as important wintering sites for the MCP (Drewien and Bizeau
1974, Buller 1982, Drewien et al. 1996, Schmitt and Hale 1997,
Chavez-Ramirez 2005). However, a few tagged cranes, mostly
from the EC–M affiliation, wintered in central Kansas and in
Oklahoma (Table 12, Fig. 11) at sites located north of the
traditional winter range of the MCP (Tacha et al. 1994).
Crane distribution across the winter range occurs near shallow,
mostly saline, water bodies with high visibility for roosting
and near cropland with high-energy waste grains. Wintering
habitats of tagged cranes along the Texas Gulf Coast from near
Houston to the Mexican border typically were large and inac-
cessible bodies of water located in broad tracts of native veg-
etation adjacent to cropland (Aldrich 1979). Along the middle
and lower Texas Gulf Coast, sorghum (Sorghum bicolor) and
rice (Oryza sativa) are the dominant agricultural foods taken
(Ballard and Thompson 2000). In western Texas, cranes roost
largely on saline pluvial lakes. Magnitude of crane use of lakes
is correlated with amount of sorghum stubble surrounding the
lakes (Iverson et al. 1985) reflecting that plentiful supplies of
28 Wildlife Monographs 175
high-energy foods are needed to attract many cranes to a site for
an extended period.
Loss of most high-energy agricultural food on a major winter-
ing ground can dramatically reduce use. The few winter PTT-
locations from eastern New Mexico during 1998–2003 reflects
limited use of this region in marked contrast to 1960–1976, when
concentrations of up to 340,000 cranes moved between eastern
New Mexico and western Texas during winter (Buller 1979).
Distributions of PTT-locations indicate MCP cranes now stay
mostly in western Texas where sorghum remains widely available
from crane arrival to departure. At Bitter Lake NWR in eastern
New Mexico, peak crane numbers during winter fell from 67,000
in 1970 to 5,400 in 1987 concomitant with >90% decreases in
cropland area planted to sorghum, with most being replaced with
alfalfa (Montgomery 1997).
The widespread distribution of WA–S cranes in the arid south-
west, including the Chihuahuan Desert in north-central
Mexico, indicates that these birds have adapted for survival in
highly arid environments where saline wetlands, often reduced to
freshwater springs and salt flats during drought, provide adequate
roosts. Energy needs are supplied from waste grain in cropland
and from grama grasses (Bouteloua spp.) in native grasslands (R.
Drewien, Hornocker Wildlife Institute [Retired], personal
communication).
MCP Use of the Gulf Coast Management Unit
Detailed knowledge of numbers and subspecies composition of
sandhill cranes occupying the Gulf Coast Management Unit
(GCMU) in Texas during late fall and winter is needed to
facilitate sandhill crane management (Ballard et al. 1999).
Satellite telemetry provided a new approach for estimating num-
ber and subspecies composition of cranes present. Our estimated
600,000 cranes in the MCP in spring, assuming a 12% annual
recruitment rate (Buller 1979), would produce a fall population of
672,000 cranes. We based our spring estimate of 600,000 cranes
in the MCP on an estimated average of 510,000 cranes in the
CPRV on the fourth Tuesday of March 2000–2003, when
numbers of cranes on nocturnal roosts were surveyed at the peak
of spring migration using aerial infrared videography (Kinzel
et al. 2006) and from an average ocular estimate of 93,227
MCP cranes in the NPRV during aerial surveys conducted on
the fourth Tuesday of March 2005 and 2006 by the U.S. Fish and
Wildlife Service and the Nebraska Game and Parks Commission
(J. Solberg, U.S. Fish and Wildlife Service, personal communi-
cation; M. Vrtiska, Nebraska Game and Parks Commission,
personal communication). With 17.4% of our random tagged
sample of the MCP wintering within the GCMU, we estimated
an average of 116,928 cranes present during 1998–2002. Our
estimate is nearly 4 times larger than the 30,000 sandhill cranes
estimated by Tacha et al. (1994) and approaches the
121,057 31,521 (SD) estimate made by Ballard et al.
(1999) for the 1997–1998 winter. Our finding that the breeding
area supplying cranes to the GCMU is much larger than pre-
viously thought (includes the Hudson Bay Lowlands of Ontario
and western Quebec and the entire WC–A breeding grounds)
helps explain the many cranes wintering in the GCMU.
Our information showing that cranes using the GCMU are
primarily Canadian sandhill crane and greater sandhill crane is
supported by results from previous studies (Ballard et al. 1999).
Basing subspecies composition of the Gulf Coast Subpopulation
on morphometry of collected birds, Ballard et al. (1999) esti-
mated 62–68% Canadian sandhill crane, 28–32% greater sandhill
crane, and 4–8% lesser sandhill crane. In comparison, we had 77%
Canadian sandhill crane, 23% greater sandhill crane, and no
lesser sandhill crane. The lack of lesser sandhill cranes in our
tagged sample from within the winter range of the Gulf Coast
subpopulation probably reflects lesser sandhill crane occurred in
numbers too small to be represented in our sample.
The GCMU represented the southern terminus of fall
migration for 23% and 77% of WC–A and EC–M cranes,
respectively. Relative proportions of each breeding affiliation
found in the GCMU differed markedly, with EC–M cranes
in the GCMU representing 83% of all wintering EC–M cranes
and with WC–A cranes accounting for 15% of all wintering
WC–A cranes. Evidence of an increase in numbers of sandhill
cranes wintering along the Gulf Coast over the past 4 decades
(see Guthery and Lewis 1979) likely reflects both population
growth and improvements in census techniques.
Range Overlap With Other Crane Populations
Distribution of EC–M cranes on their breeding grounds in
Minnesota approaches and possibly overlaps with the EP.
Greater sandhill crane from the MCP breed widely across north-
western parts of Minnesota based on distribution of our tagged
sample, whereas greater sandhill cranes from the EP breed across
central Minnesota (Henderson 1978, Toepfer and Crete 1979).
Tacha et al. (1994) included most of Ontario in the breeding
range of the EP, which, if accurate, would indicate broad overlap
in distribution with the MCP across the central and northern
parts of the province based on our tagged sample. More likely,
cranes in the Hudson Bay Lowlands are mostly or all from the
MCP, a conclusion supported by the observation that most
cranes from this region have morphological measurements
indicative of Canadian sandhill cranes (Lumsden 1971, our
study). The northern limits of breeding of the EP in Ontario,
Quebec, and, to a lesser extent, Minnesota remain poorly defined
and need further study.
Occasional gene flow occurs between the EC–M affiliation and
EP based on results from microsatellite studies (Jones et al. 2005),
although no tagged cranes moved into the known range of the EP
during our study. Gene flow likely will increase between the EP
and MCP in the future if breeding ranges and wintering grounds
of both populations continue to expand, resulting in greater
overlap in area occupied and likely pairing between cranes of
the 2 populations. Some MCP Canadian sandhill cranes and
greater sandhill cranes (from EC–M) winter in central Louisiana
(S. King, USGS Louisiana Cooperative Fish and Wildlife
Research Unit, unpublished data).
The RMP breeds in Montana, Wyoming, Utah, Idaho, and
Colorado (Drewien and Bizeau 1974), where we found no evi-
dence of breeding by tagged MCP sandhill cranes. However, the
winter range of MCP lesser sandhill cranes overlaps with the
RMP in west-central New Mexico, southeastern Arizona, and
parts of northern Mexico (Drewien and Bizeau 1974, Drewien
et al. 1996, Schmitt and Hale 1997, Krapu and Brandt 2008).
Most MCP lesser sandhill cranes that winter in the same areas as
Krapu et al. Geographic Distribution of Sandhill Cranes 29
the RMP are in the WA–S breeding affiliation, based on winter
distribution of our tagged birds (Fig. 11). Some MCP Canadian
sandhill crane, presumably WC–A cranes, also winter in west-
central New Mexico (Schmitt and Hale 1997) and migrate
through the San Luis Valley of south-central Colorado in spring
migration as do some WA–S lesser sandhill crane (Benning et al.
1997, Krapu and Brandt 2008). No records of pairing have been
reported between MCP lesser sandhill crane and RMP greater
sandhill crane in regions where their spring-fall migration and
winter distributions overlap (R. Drewien, Hornocker Wildlife
Institute [Retired], and W. Brown, U.S. Fish and Wildlife
Service, unpublished data).
The PFP breeds across parts of southwestern Alaska including
the lowlands of the Alaska Peninsula (Mickelson 1987, Petrula
and Rothe 2005) where they occur within approximately 200 km
of the nearest site used by a tagged WA–S crane during the
breeding season (Fig. 13). The PFP winters primarily in the
Central Valley of California (Lewis 1977, Herter 1982,
Littlefield and Thompson 1982) and some may also winter near
the Pacific Coast in Mexico (Lewis 1977, Herter 1982, Littlefield
and Thompson 1982, Drewien et al. 1996). We found no evi-
dence of WA–S mixing with the PFP on their wintering grounds,
staging areas, or breeding grounds, but we suspect that limited
mixing occasionally occurs between these populations. On the
breeding grounds, occasional contact between PFP and WA–S
seems plausible, most likely involving yearlings or subadults,
which move more during the breeding season (Drewien et al.
1999). However, information gained from monitoring summer
movements of a small sample of tagged PFP yearlings did not
indicate movements into areas occupied by WA–S (Petrula and
Rothe 2005).
In spring migration, WA–S breeding in western Alaska may
occasionally stray from their traditional spring migration route in
northeastern British Columbia and take a more direct path to
their breeding grounds, leading to a potential stopover at the
Copper River Delta, a major spring staging area of the PFP
(Herter 1982, Mickelson 1987). Although occasional mixing
between PFP and the MCP has genetic implications, it is
unlikely such infrequent mixing has significance to issues per-
taining to hunting.
Factors Influencing Composition of Harvest by Breeding
Affiliation
Estimated flyway-wide harvest of the WC–A breeding affiliation
was disproportionate to subpopulation size (29% of harvest vs.
23% of MCP) with an estimated 62% of harvest occurring in
Saskatchewan being WC–A cranes. The large and dispropor-
tionate harvest of WC–A cranes in Saskatchewan is linked to an
early start of the hunting season when virtually all cranes present
in the province are of the WC–A affiliation. In addition, an
earlier arrival and longer stay of WC–A cranes than WA–S and
NC–N cranes on fall staging areas in the province, a 5-crane daily
bag limit, harvest concentrations in areas where WC–A domi-
nate, and an increasing number of sandhill crane hunters, prim-
arily non-Canadians from the United States (D. Nieman,
Canadian Wildlife Service, unpublished report) likely contribute
to a disproportionate harvest of WC–A cranes. Most tagged
WC–A cranes were on their fall staging areas at the onset of
the hunting season, arriving an average of 17 days before NC–N
cranes (257 3.8 vs. 240 3.0 Julian date; xSD) resulting in a
2.6-fold higher mean exposure. There were approximately twice
as many WC–A cranes as NC–N cranes, resulting in WC–A
cranes being subject to 5 times as much exposure to hunting as
were NC–N cranes.
The disproportionate take of EC–M cranes relative to sub-
population size (21% of overall harvest but composing an esti-
mated 14% of MCP) resulted from an early onset of hunting
seasons on the primary fall staging areas in southwestern
Manitoba and North Dakota, a slow fall migration that increased
exposure to hunting in Kansas and Oklahoma, and an extended
stay on Gulf Coast wintering grounds in Texas. The dispropor-
tionate harvest of EC–M cranes in North Dakota (Table 17) also
is influenced by the many crane hunters in the state (Kruse et al.
2008). The factor that likely contributed most to disproportion-
ately fewer WA–S and NC–N cranes harvested on fall staging
areas was a much shorter stay than that by WC–A and EC–M
cranes. Also, WA–S cranes benefit from fall staging occurring at a
more remote location than the other 3 breeding affiliations.
Major Risks to MCP From Habitat Change
Habitat loss and degradation from agricultural intensification,
global climate change, and oil and mineral exploration and
development pose potential significant long-term risks to the
MCP. The MCP faces potential major threats from habitat loss
and alteration in parts of its breeding range. Extensive energy
development is underway on a major breeding ground of the
WC–A affiliation in Alberta. Two of the 10 (20%) tagged cranes
that settled in Alberta occurred at sites currently under lease for
extraction of oil from tar sands, one each in the Athabasca Oil
Sands Area and the Cold Lake Oil Sands Area. Under current
plans, an estimated 3,000 km
2
of Alberta’s boreal forest and
associated wetlands will be strip-mined by 2030 to remove bitu-
men to process into oil. Development is expected to spread across
a 149,000-km
2
area (equivalent in size to the state of Florida) and
transform the land into an industrialized landscape fragmented
by a network of steam well pads, roads, pipelines, and other
infrastructure (Woynillowicz et al. 2005). This transformation
could have adverse effects on the WC–A subpopulation if devel-
opment continues as currently planned. Continued northward
expansion of agricultural development in Canada in areas with
fertile soils could cause further fragmentation of crane breeding
habitats and culminate in lower EC–M and WC–A reproductive
success in affected areas. Extensive logging in central Canada also
poses significant threats to the WC–A and EC–M subpopu-
lations if environmental concerns are not adequately addressed.
The primary breeding grounds of WA–S and NC–N in the
Arctic have not been substantially impacted to date by large-scale
energy and mineral development, in part because of the remote-
ness of breeding grounds and severity of the climate. In Alaska,
the most important MCP breeding grounds include the Yukon
Delta NWR and Yukon Flats NWR, which cover >7.3 million
ha and 2.0 million ha, respectively. Strict environmental safe-
guards will be required to protect these and other publicly owned
habitats important to sandhill cranes and other migratory birds,
particularly in those parts not protected by wilderness designation
and thus potentially subject to energy and mineral development.
30 Wildlife Monographs 175
The North Slope in Alaska has undergone extensive oil develop-
ment over the past several decades but the areas most affected by
development to date contained relatively few sandhill cranes
before oil development (Bergman et al. 1977) and none of our
tagged cranes settled there, reflecting low densities across this
region.
Eastward in arctic Canada, the MacKenzie Delta-Beaufort Sea
Basin contains an estimated 10.9 trillion cubic feet of discovered
and 45.8 trillion cubic feet of predicted marketable natural gas
deposits estimated to be worth about $115 billion (Sproule
Associates Ltd., unpublished report). The MacKenzie Delta
region is an important breeding area for sandhill cranes and
waterfowl, so ecologically sound methods of gas extraction would
be prudent if the Delta is to continue to meet the needs of cranes,
waterfowl, and other migratory water birds. Most mineral devel-
opment in the Northwest Territories to date has not involved
major breeding areas of cranes nor has development significantly
impacted crane breeding areas in the central Canadian Arctic,
along the northwest side of Hudson Bay, or in the Hudson Bay
Lowlands. However, because of the myriad of potential threats to
wildlife populations inhabiting subarctic and Arctic regions from
various forms of development, more safeguards may need to be
considered to protect arctic- and subarctic-nesting cranes and
other wildlife before development becomes widespread.
Competition with agriculture for fresh water on wintering
grounds in the United States and Mexico is likely to intensify,
emphasizing a need for effective strategies to ensure protection of
groundwater hydrology responsible for maintaining freshwater
springs associated with saline lakes that support the MCP on the
wintering grounds. Our tagged sample relied primarily on 18
saline lakes in western Texas (Appendix C) to meet roost-site
needs during winter, a number that closely follows the recom-
mendation of Tacha et al. (1994) that conservation of <20 saline
pluvial lakes with freshwater springs in western Texas is essential
to provide sites for roosting and drinking water. Crane use of
saline pluvial lakes in western Texas as roost sites is correlated
with the number of freshwater springs present (Iverson et al.
1985), underscoring the importance of maintaining the shallow
aquifers capable of supplying cranes with fresh drinking water.
Key wintering areas of migratory waterfowl and sandhill cranes
have been lost or degraded in northern Mexico, where compe-
tition for fresh water also is intense (Drewien et al. 2003),
indicating a need for strong conservation measures to protect
important existing migratory waterbird habitat. Greater insight is
needed concerning the distribution of saline lakes used by the
MCP across the winter range so more comprehensive and effec-
tive strategies can be developed to protect important sandhill
crane wintering sites throughout the region.
Maintaining adequate supplies of high-energy food across the
winter range of the MCP also is essential as reflected by the
drastic decline in MCP use of eastern New Mexico over the past
several decades as high-energy agricultural crops declined there.
To help ensure adequate food resources remain available in the
future, it is vital that a widely distributed wetland base capable of
meeting roost-site needs of the MCP be maintained across the
wintering grounds so cranes can continue to shift their distri-
butions as agricultural cropping practices, drought, and other
factors change.
Tagged cranes using fall staging areas in the northern plains
roosted primarily on saline lakes. Expanding development of
center-pivot irrigation in parts of the region where cranes stage
during fall poses a potential major threat to crane habitat in the
future. Pumping of ground water for irrigation, if not adequa-
tely controlled, could eliminate freshwater springs that allow
saline lakes to be important roosts of cranes. With the
Missouri River no longer providing extensive roosting habitat,
a greater urgency exists to ensure availability of saline lakes with
freshwater springs and to maintain the limited braided-river-
channel habitat in the northern plains that remains suitable for
roosting. In general, given the sensitivity of cranes to human
disturbance in the vicinity of their roosts (Krapu et al. 1984),
forms of activity that increase disturbance near nocturnal roost
sites can be expected to displace cranes and alter distribution of
fall staging.
The MCP stages in the CPRV for about 4 weeks during early
spring and acquires a major part of fat reserves that supply energy
used in migration and after arrival on the breeding grounds where
foraging conditions initially are poor, particularly in arctic regions
(Krapu et al. 1985). The ability of cranes to store large quantities
of fat in the CPRV is being threatened by a decline in waste corn
due to increased harvest efficiency (Krapu et al. 2004) and
competition from snow geese (Pearse et al. 2010). Also, reduced
flows in the Platte River and woody vegetation encroachment
into channels have increased crowding at remaining roost sites
over the past half century (Krapu et al. 1982). This change has
forced the birds to roost at fewer sites and fly farther from their
roosts to forage, increasing energy costs (Pearse et al. 2010). As
waste corn has declined in cropland, fat storage has been reduced
(G. Krapu, unpublished data). A Platte River Recovery
Implementation Program has been created in a joint effort
between the United States Department of Interior and the
States of Nebraska, Colorado, and Wyoming to restore key
habitats in the CPRV for whooping cranes (Grus americana)
and other threatened and endangered species (Platte River
Recovery Implementation Program 2010). This program, if suc-
cessful, also has the potential for improving habitat conditions for
sandhill cranes.
MANAGEMENT IMPLICATIONS
Detailed knowledge of breeding affiliations of sandhill cranes
using fall staging areas and wintering grounds has been a long-
standing information need of managers of Central Flyway sand-
hill cranes (Sharp and Cornely 1997). Sandhill cranes from the 4
breeding affiliations we delineated (WA–S, NC–N, WC–A, and
EC–M) exhibited exceptional levels of philopatry from 1 yr to the
next at sites occupied during the breeding season, and we pre-
dicted fall staging and wintering locations by breeding locations
of cranes. These patterns of recurring use of fall staging areas and
wintering grounds by cranes of known breeding origins during
the same time intervals each year, combined with knowledge of
harvest composition by breeding affiliation, provide the necessary
information to target harvest as desired.
We found major exchange of birds between the ranges of the
Western and Gulf Coast subpopulations, complicating efforts to
effectively target harvest where most appropriate. Also, dispro-
portionately greater harvest of temperate and subarctic breeding
Krapu et al. Geographic Distribution of Sandhill Cranes 31
greater sandhill crane and Canadian sandhill crane compared to
arctic-breeding lesser sandhill crane occurred on fall staging areas
due to differences in fall staging distribution and chronology.
Drawing upon our results, MCP sandhill crane managers have
the information necessary on spatial and temporal use of fall
staging areas and wintering grounds by cranes of each of the 4
breeding affiliations to update guidelines as necessary for manag-
ing MCP harvest. We suggest that, in conjunction with updating
the management plan, consideration be given to managing the
MCP as 4 subpopulations (WA–S, NC–N, WC–A, and EC–M)
to help achieve the long-term goal of providing diverse recrea-
tional opportunities consistent with the welfare of the MCP. A
key part of that goal is managing the population in ways that will
maintain a sustainable harvest across subpopulations.
Evidence of a disproportionate harvest of WC–A and EC–M
cranes relative to numbers of individuals in these subpopulations
casts doubt on whether recent harvest levels are sustainable.
Cranes from WC–A and EC–M subpopulations provide a major
part of hunting opportunities in 4 states and 2 provinces that
include key fall staging areas and wintering grounds, indicating
the importance of taking steps necessary to ensure harvest levels
are sustainable in these subpopulations. One of the most con-
spicuous outcomes of the high harvest rates of WC–A and EC–
M cranes suggested by our results has been the failure of MCP
greater sandhill cranes over time to rebuild their numbers to levels
capable of re-occupying their extensive former breeding range in
the PPR. This status contrasts with the EP, which has re-
occupied a major part of their historic range in the Midwest
including parts of Iowa, Illinois, and Indiana (Meine and
Archibald 1996) despite the loss of >90% of the original wetland
habitat in these states (Tiner 1984).
Targeting Harvest at Subpopulation Level
The 4 subpopulations of MCP cranes we identified are spatially
separated to various degrees on their fall staging areas and winter-
ing grounds. In 8 U.S. states, 1 Canadian province, and 1
Mexican state used by tagged cranes, one MCP subpopulation
accounts for all or most use of the zones where cranes are hunted,
permitting harvest to be focused largely on management needs
for that subpopulation (i.e., Alaska [WA–S], Kansas [EC–M],
Oklahoma [EC–M], Colorado [WA–S], Wyoming [WA–S],
Montana [WA–S], New Mexico [WA–S], Arizona [WA–S],
Manitoba [EC–M], and Chihuahua [WA–S]). In parts of New
Mexico, Arizona, and Chihuahua, RMP cranes co-exist on the
same areas as WA–S cranes (Drewien and Bizeau 1974), so RMP
presence must be accounted for when setting hunting regulations.
Because most WA–S cranes take less than a week to migrate from
their fall staging areas in the northern plains to their wintering
grounds, exposure to sport harvest is limited along the migration
route and changes in hunting regulations in those states are not
likely to cause major changes in WA–S harvest.
Harvest of WC–A cranes was disproportionately high to num-
bers present. Saskatchewan is the primary source of harvest of
WC–A cranes so any steps to reduce the disproportionate harvest
of birds from this breeding affiliation would need to include
measures to reduce WC–A harvest in the province. Three poten-
tially effective ways to reduce the take of WC–A in Saskatchewan
would be to delay the onset of sandhill crane hunting in the
province, reduce the bag limit, and establish hunting zones to
allow harvest to be managed more conservatively for WC–A.
In Manitoba, and in North Dakota east of U.S. Highway 281
(Hunting Zone 2), EC–M is the only subpopulation present
(Fig. 14A), so harvest can be managed independently. A later
start date in North Dakota and an expanded Hunting Zone 2
(including fall staging areas where EC–M cranes dominate but
now are located in Hunting Zone 1 [i.e., Bottineau, Pierce,
Wells, Stutsman, and Kidder counties]), would allow harvest
to be targeted at different rates for EC–M and NC–N affiliations.
In Kansas and Oklahoma, harvest is about 75% EC–M cranes
with most of the remainder being WC–A cranes (Table 17). If
reducing harvest of EC–M or WC–A cranes (or both) was
deemed necessary, changing hunting dates or bag limit would
likely be most effective.
Management of cranes in Texas is the most complex, as all
subpopulations winter in the state. Exposure of WC–A cranes to
hunting in Texas was highest in Hunting Zone A, the primary
wintering ground. Our results showed major overlap among
WC–A and NC–N cranes in Hunting Zone A, so more con-
servative hunting regulations in that portion of Zone A where
WC–A distribution is centered would also result in a reduced
harvest of NC–N cranes (Fig. 12B). In Hunting Zone B, WC–A
and EC–M cranes accounted for most use and harvest. Hunting
Zone B (Fig. 14D) was the least used by tagged cranes of the 3
zones in Texas and accounted for only approximately 4% of the
harvest, limiting the potential for altering overall harvest patterns
through management actions undertaken in this zone. Hunting
Zone C (GCMU) comprised the primary and secondary winter-
ing grounds of the EC–M and WC–A subpopulations, respect-
ively, with the EC–M subpopulation occurring primarily in the
Upper Gulf Coast and WC–A occurring principally in the Mid-
to Lower Gulf Coast. This pattern of distribution suggests that it
may be feasible to manage harvest of the EC–M affiliation
differently than the WC–A affiliation by splitting Zone C into
2 sub-units. However, our data suggest that both subpopulations
are being taken disproportionately to their numbers on fall
staging areas and during winter, so it is unclear whether manag-
ing the WC–A affiliation differently than the EC–M affiliation
in Zone C will achieve desired goals under current circumstances.
Any adjustments in harvest in Zone C can be expected to have a
greater effect on the EC–M subpopulation, with 3 times as many
EC–M cranes wintering on the Gulf Coast as WC–A cranes.
Given that few NC–N and WA–S cranes winter along the Gulf
Coast, crane management in Zone C can be directed exclusively
on managing for the needs of the EC–M and WC–A
subpopulations.
We assumed in our estimates of harvest that each subpopulation
is harvested proportional to its presence (exposure) in a hunted
area. In areas of major overlap of WC–A and NC–N cranes (i.e.,
Hunting Zone A in West Texas and east-central and central
Saskatchewan), greater sandhill crane, Canadian sandhill crane,
and lesser sandhill crane commonly occur in mixed flocks and
under these conditions hunters may selectively or subconsciously
target larger cranes (greater sandhill crane and Canadian sandhill
crane). As a result, our estimates of WC–A harvest may be
conservative in areas of major WC–A affiliation overlap with
the NC–N affiliation.
32 Wildlife Monographs 175
Opportunities for Increased International Collaboration
Our results provide natural resource managers in the United
States, Canada, Mexico, and Russia with new insight into where
conservation, research, and management efforts involving specific
subpopulations of the MCP will be most effective. The WA–S
subpopulation is shared by all 4 nations; breeding grounds are
exclusively in western Alaska and northeast Russia, the primary
spring staging area is in Nebraska, the primary fall staging area
along with a major spring staging area are in western
Saskatchewan (Krapu and Brandt 2008), and wintering grounds
are in Texas, New Mexico, Arizona, Chihuahua, Coahuila,
Durango, and Zacatecas. With all 4 nations providing vital
habitat to the WA–S breeding affiliation, close coordination
and collaboration is needed among federal natural resource
agencies to ensure that the habitat base that supports the
WA–S throughout the annual cycle receives adequate protection.
The WC–A, EC–M, and NC–N subpopulations breed primarily
in Canada. All 3 subpopulations winter primarily in the southern
United States, with some use occurring in northern Mexico. As a
result, conservation, research, and management efforts affecting
the WC–A, EC–M, and NC–N subpopulations are the respon-
sibility of natural resource agencies and non-government organ-
izations in the United States, Canada, and Mexico.
SUMMARY
1) We monitored a PTT-tagged random sample of the Mid-
continent Population (MCP) of sandhill cranes from their
arrival on the breeding grounds to the end of winter to
describe chronology of use and breeding affiliation of cranes
using fall staging areas and wintering grounds. Information
gained on temporal and spatial separation of tagged cranes by
breeding affiliation during fall and winter allowed an assess-
ment of exposure to hunting seasons, and distribution and
size of harvest by subpopulation, thereby providing guidance
to Central Flyway crane managers seeking to regulate
crane hunting in ways that will help ensure a sustainable
harvest.
2) We noted a high rate of philopatry by tagged MCP sandhill
cranes to previously used breeding sites and could predict
locations of fall staging areas and wintering grounds by breed-
ing site. These data allowed separation of the MCP into 4
geographically discrete subpopulations for management pur-
poses linked to their breeding affiliation: Western Alaska–
Siberia (WA–S), Northern Canada–Nunavut (NC–N),
West-central Canada–Alaska (WC–A), and East-central
Canada–Minnesota (EC–M).
3) The size of the known breeding area supplying cranes to the
Gulf Coast Management Unit (GCMU) in Texas increased
several-fold with information gained from tagged birds. New
insight into breeding origins and numbers of cranes wintering
in the GCMU together with knowledge gained on wintering
distributions of EC–M and WC–A cranes within the GCMU
provide managers with new insight for managing harvest in
the Unit.
4) Disproportionately high harvests of WC–A and EC–M
cranes relative to their estimated numbers in the MCP have
resulted from high exposure of these subpopulations to
hunting seasons and hunter concentrations, particularly in the
northern plains. Differences in temporal and spatial use of fall
staging areas and wintering grounds by the 4 subpopulations
provide guidance for developing strategies to reduce take of
WC–A and EC–M cranes should a reduction in their harvest
become necessary.
5) The MCP currently is managed as 2 subpopulations, the Gulf
Coast Subpopulation and the Western Subpopulation.
Information from tagged cranes showed 71% and 100% of
NC–N cranes and 18% and 21% of EC–M cranes migrated
from breeding grounds of the Gulf Coast Subpopulation to
fall staging areas and wintering grounds of the Western
Subpopulation, respectively. This magnitude of exchange
of cranes between ranges of the 2 subpopulations reduces
effectiveness of current harvest management strategies. We
recommend that consideration be given to managing the
MCP as 4 subpopulations (WA–S, NC–N, WC–A, and
EC–M) to help maintain a sustainable harvest and current
recreational opportunities consistent with welfare of the
MCP.
6) Texas overwinters approximately 80% of the MCP with
Hunting Zone A providing wintering habitat for an estimated
99%, 74%, and 64% of NC–N, WA–S, and WC–A subpopu-
lations, respectively. The unique role of Texas in providing
wintering habitat sufficient to maintain the MCP at its cur-
rent status of providing diverse recreational and other benefits
emphasizes the need for a comprehensive effort to maintain
habitat resources supporting the MCP in the state.
7) Environmental change, mostly associated with energy and
agricultural development, pose long-term risks to the MCP
at key breeding grounds, migration stopovers, and wintering
areas, suggesting that conditions be monitored at identified
sites and appropriate actions be undertaken where needed to
minimize adverse impacts.
8) Key habitat resources supporting the MCP are located in 4
nations (United States [breeding, wintering, migration],
Canada [breeding, migration], Mexico [wintering], and
Russia [breeding]), indicating a need for international collab-
oration and a holistic approach when addressing conservation
and management issues.
ACKNOWLEDGMENTS
This research was conducted by the Central Platte River Priority
Ecosystem Study, Biological Resources Discipline, USGS. We
thank D. Douglas for providing technical and Argos program
support. We thank D. Sharp, J. Dubovsky, J. Roberson, M.
O’Meilia, T. Mitchusson, J. Hansen, M. Syzmanski, M.
Johnson, J. Gammonley, H. Hands, S. Kohn, D. Nieman,
M. Vrtiska, S. Taylor, S. Vaa, V. Bevill, R. George, L.
Roberts, J. Solberg, D. Dolton, the late J. Gabig, D. Benning,
R. Parker, S. Anschutz, P. Kinzel, J. Cornely, M. Forsberg, and
P. Tebbel for their help in facilitating various aspects of the
study. We thank the late P. Currier and F. Chavez-Ramirez,
Directors of the Whooping Crane Habitat Maintenance Trust,
for allowing use of their facilities near Wood River, Nebraska,
when trapping and tagging sandhill cranes in the Platte River
Valley and North Platte River Valley. We thank R. Kirby, J.
Hestbeck, and J. Powell, Directors, and D. Jorde, Deputy
Krapu et al. Geographic Distribution of Sandhill Cranes 33
Director, at the Northern Prairie Wildlife Research Center for
their support. We thank D. Fronczak, A. Olson, A. Stonsifer, J.
Drahota, W. Jones, T. Buhl, F. Sargeant, J. Fiest, L. Wood, S.
Hawks, D. Grandmaison, M. Westbrock, K. Seginak, B.
Hanson, C. Mettenbrink, R. Knopik, C. Graue, V. Carter, B.
Geaumont, J. McCabe, D. Smith, L. Potter, J. Thibault, B. Toay,
M. Stoley, M. Heiser, and M. Pieron for field assistance. We are
grateful for comments by R. Drewien, D. Sharp, and G. Ivey,
which improved earlier drafts of this manuscript and Ingrid
Barcelo, Audrey Desmarteaux-Houle, and Galina Revina for
translating the abstract into Spanish, French, and Russian.
We thank the many private landowners in Nebraska that
allowed access to their lands to trap and tag cranes. Any mention
of trade, product, or firm names is for descriptive purposes only
and does not imply endorsement by the U.S. Government. This
study was supported by U.S. Geological Survey, Northern Prairie
Wildlife Research Center, U.S. Fish and Wildlife Service, North
Dakota Game and Fish Department, Texas Game and Parks
Commission, Oklahoma Department of Wildlife Conservation,
Nebraska Game and Parks Commission, Kansas Department of
Wildlife and Parks, New Mexico Department of Game and Fish,
South Dakota Game, Fish and Parks Department, Wyoming
Game and Fish Department, Montana Department of Fish,
Wildlife and Parks, International Crane Foundation, Platte
River Whooping Crane Maintenance Trust, and Playa Lakes
Joint Venture. Publication costs were provided by United States
Geological Survey, Science Support Program.
LITERATURE CITED
Alaska Science Center. 2010. Douglas Argos-Filter Algorithm. Archive of
Previous Algorithm Versions. <http://alaska.usgs.gov/science/biology/
spatial/archive.html>. Accessed 20 Jan 2010.
Aldrich, J. W. 1979. Status of the Canadian sandhill crane. Pages 139–148 in J. C.
Lewis, editor. Proceedings 1978 Crane Workshop. Colorado State University
Printing Service, Fort Collins, USA.
American Ornithologist’s Union. 1997. A. S. Gaunt, and L. W. Oring, editors.
Guidelines on the use of wild birds in research. The Ornithological Council,
Washington, D.C., USA.
Austin, D. J., I. McMillan, and W. D. Bowen. 2003. A three-stage algorithm for
filtering erroneous Argos satellite locations. Marine Mammal Science 19:371–
383.
Ballard, B. M., and J. E. Thompson. 2000. Winter diets of sandhill cranes from
central and coastal Texas. Wilson Bulletin 112:263–268.
Ballard, B. M., J. E. Thompson, M. T. Meredino, J. D. Ray, J. A. Robertson,
and T. C. Tacha. 1999. Demographics of the Gulf Coast Subpopulation of
Mid-continent sandhill cranes. Proceedings of the Annual Conference of
the Southeast Association of Fish and Wildlife Agencies 53:449–463.
Benning, D. S., R. C. Drewien, D. H. Johnson, W. M. Brown, and E. L. Boeker.
1997. Spring population estimates of Rocky Mountain greater sandhill cranes in
Colorado. Pages 165–172 in R. P. Urbanek, and D. W. Stalhlecker, editors.
Proceedings of the Seventh North American Crane Workshop, 10–13 Jan 1996,
Biloxi, Mississippi, USA.
Bergman, R. D., R. L. Howard, K. F. Abraham, and M. W. Weller. 1977.
Water birds and their wetland resources in relation to oil development at
Storkersen Point, Alaska. U.S. Department of Interior, U.S. Fish and
Wildlife Service Resource Publication 129, Washington, D.C., USA.
Boise, C. M. 1977. Breeding biology of the lesser sandhill crane, (Grus canadensis
canadensis L.), on the Yukon-Kuskokwim Delta, Alaska. Thesis, University of
Alaska, Fairbanks, USA.
Britten, M. W., P. L. Kennedy, and S. Ambrose. 1999. Performance and accuracy
evaluation of small satellite transmitters. Journal of Wildlife Management
63:1349–1358.
Buller, R. J. 1979. Lesser and Canadian sandhill crane populations, age structure,
and harvest. U.S. Department of the Interior, Fish and Wildlife Service Special
Scientific Report Wildlife No. 221, Washington, D.C., USA.
Buller, R. J. 1982. Distribution of sandhill cranes wintering in Mexico. Pages 266–
272 in J. C. Lewis, editor. Proceedings of the 1981 Crane Workshop, National
Audubon Society, Tavernier, Florida, USA.
Buller, R. J., and E. L. Boeker. 1965. Coordinated sandhill crane study in the
Central Flyway. North American Wildlife and Natural Resources Conference
30:100–112.
Central Flyway Webless Migratory Game Bird Technical Committee
[Compiler]. 2006. Management guidelines for the mid-continent population
of sandhill cranes. U.S. Fish and Wildlife Service Migratory Bird Management
Office, Golden, Colorado, USA.
Chavez-Ramirez, F. 2005. New locations and range extension of wintering sandhill
cranes in central northern Mexico. Pages173–178 in F. Chavez-Ramirez, editor.
Proceedings of the Ninth North American Crane Workshop, North American
Crane Working Group, Sacramento, California, USA.
Clark, D. D. 1989. Use of Argos for animal tracking in the Rocky Mountain
Region of North America. Pages 129–176 in Proceedings of the North
American Argos users conference and exhibit. Service Argos, Inc.,
Landover, Maryland, USA.
Conant, B., J. G. King, and H. A. Hansen. 1985. Sandhill cranes in Alaska: a
population survey, 1957–1985. American Birds 39:855–858.
Drewien, R. C., and E. G. Bizeau. 1974. Status and distribution of greater
sandhill cranes in the Rocky Mountains. Journal of Wildlife Management
38:720–742.
Drewien, R. C., W. M. Brown, and W. L. Kendall. 1995. Recruitment in Rocky
Mountain greater sandhill cranes and comparison with other crane populations.
Journal of Wildlife Management 59:339–356.
Drewien, R. C., W. M. Brown, and D. S. Benning. 1996. Distribution and
abundance of sandhill cranes in Mexico. Journal of Wildlife Management
60:270–285.
Drewien, R. C., W. M. Brown, J. D. Varley, and D. C. Lockman. 1999. Seasonal
movements of sandhill cranes radiomarked in Yellowstone National Park and
Jackson Hole, Wyoming. Journal of Wildlife Management 63:126–136.
Drewien, R. C., A. L. Terrazas, J. P. Taylor, J. M. O. Barraza, and R. E. Shea.
2003. Status of lesser snow geese and Ross’s geese wintering in the Interior
Highlands of Mexico. Wildlife Society Bulletin 31:417–432.
Duan, W., and O. A. Fuerst. 2001. Isolation of a sex-linked DNA sequence in
cranes. Journal of Heredity 92:392–397.
Ellis, D. H., P. W. Howey, and G. L. Krapu. 2001. Recommendations for the
attachment of satellite transmitters to cranes. Pages 211–212 in D. H. Ellis,
editor. Proceedings of the 8th North American Crane Workshop, 11–14 Jan
2000, Albuquerque, New Mexico, USA.
Fancy, S. G., L. F. Pank, D. C. Douglas, C. H. Curby, G. W. Garner, S. C.
Amstrup, and W. L. Regelin. 1988. Satellite telemetry: a new tool for wildlife
research and management. U.S. Department of Interior, U.S. Fish and Wildlife
Service Resource Publication 172, Washington, D.C., USA.
Fieberg, J., and C. O. Kochanny. 2005. Quantifying home-range overlap: the
importance of the utilization distribution. Journal of Wildlife Management
69:1346–1359.
Fields, R. C., A. K. Trout, and D. T. Walls. 1974. Recent breeding of the sandhill
crane in North Dakota. Wilson Bulletin 86:285–286.
Glenn, T. C., J. E. Thompson, B. M. Ballard, J. A. Robertson, and J. O. French.
2002. Mitochondrial DNA variation among wintering mid-continent gulf coast
sandhill cranes. Journal of Wildlife Management 66:339–348.
Guthery, F. S., and J. C. Lewis. 1979. Sandhill cranes in coastal counties of
southern Texas: taxonomy, distribution, and population. Proceedings 1978
International Crane Workshop 2:121–128.
Harris, R. B., S. G. Fancy, D. C. Douglas, G. W. Gardner, S. C. Amstrup, T. R.
McCabe, and L. F. Pank. 1990. Tracking wildlife by satellite: current systems
and performance. U.S. Fish and Wildlife Service Technical Report 30,
Washington D.C., USA.
Hays, G. C., S. Akesson, B. J. Godley, P. Luschi, and P. Santidrian. 2001. The
implications of location accuracy for the interpretation of satellite tracking data.
Animal Behavior 61:1035–1040.
Henderson, C. L. 1978. Results of an observation card survey for sandhill cranes in
Minnesota for 1977. Loon 50:112–118.
34 Wildlife Monographs 175
Herter, D. R. 1982. Staging of sandhill cranes on the eastern Copper River Delta,
Alaska. Pages 273–280 in J. C. Lewis, editor. Proceedings 1981 Crane
Workshop, National Audubon Society, Tavernier, Florida, USA.
Iverson, G. C., P. A. Vohs, Jr., and T. C. Tacha. 1985. Distribution and
abundance of sandhill cranes wintering in western Texas. Journal of Wildlife
Management 49:250–255.
Iverson, G. C., P. A. Vohs, and T. C. Tacha. 1987. Habitat use by mid-continent
sandhill cranes during spring migration. Journal of Wildlife Management
51:448–458.
Johnsgard, P. A. 1983. Cranes of the world. Indiana University Press,
Bloomington, Indiana, USA.
Johnson, J. W. 1963. The dispossessed sandhill cranes. Audubon Magazine 65(4):
213–217.
Johnson, D. H. 1976. The status of the sandhill crane in North Dakota. Pages 69–
77 in J. C. Lewis, editor. Proceedings of the International Crane Workshop,
International Crane Foundation, Baraboo, Wisconsin, USA.
Johnson, D. H., and R. E. Stewart. 1973. Racial composition of migrant popu-
lations of sandhill cranes in the northern plains states. Wilson Bulletin 85:148–
162.
Jones, K. L., G. L. Krapu, D. A. Brandt, and M. V. Ashely. 2005. Population
genetic structure in migratory sandhill cranes and the role of Pleistocene
glaciations. Molecular Ecology 14:2645–2657.
Keating, K. A. 1994. An alternative index of satellite telemetry location error.
Journal of Wildlife Management 58:414–421.
Keating, K. A., W. G. Brewster, and C. H. Key. 1991. Satellite telemetry:
performance of animal tracking systems. Journal of Wildlife Management
55:160–171.
Kendall, W. L., D. H. Johnson, and S. C. Kohn. 1997. Subspecies composition of
sandhill crane harvest in North Dakota, 1968–94. Pages 201–208 in R. P.
Urbanek, and D. W. Stalhlecker, editors. Proceedings of the Seventh North
American Crane Workshop. 10–13 Jan 1996, Biloxi, Mississippi, USA.
Kinzel, P. J., J. M. Nelson, R. S. Parker, and L. R. Davis. 2006. Spring census of
mid-continent sandhill cranes using aerial infrared videography. Journal of
Wildlife Management 70:70–77.
Krapu, G. L., and D. A. Brandt. 2008. Spring migratory habits and breeding
distribution of lesser sandhill cranes that winter in west-central New Mexico
and Arizona. Pages 43–49 in M. J. Folk, and S. A. Nesbitt, editors. Proceedings
of the Tenth North American Crane Workshop, 7–10 Feb 2006, Zacatecas,
Mexico.
Krapu, G. L., and D. H. Johnson. 1990. Conditioning of sandhill cranes during
fall migration. Journal of Wildlife Management 54:234–238.
Krapu, G. L., K. J. Reinecke, and C. R. Frith. 1982. Sandhill cranes and the Platte
River. Transactions of the North American Wildlife and Natural Resources
Conference 47:542–552.
Krapu, G. L., D. E. Facey, E. K. Fritzell, and D. H. Johnson. 1984. Habitat use by
migrant sandhill cranes in Nebraska. Journal of Wildlife Management 48:407–
417.
Krapu, G. L., G. C. Iverson, K. J. Reinecke, and C. M. Boise. 1985. Fat storage
and usage by arctic-nesting sandhill cranes during spring. Auk 102:362–
368.
Krapu, G. L., D. Facey, B. Gehring, and M. I. Meyer. 1987. Resource inventory
of sandhill crane staging areas in Nebraska. Pages 364–370 in J. C. Lewis,
editor. Proceedings of the 1985 North American Crane Workshop. Platte River
Whooping Crane Habitat Maintenance Trust and U.S. Fish and Wildlife
Service, Grand Island, Nebraska, USA.
Krapu, G. L., D. A. Brandt, and R. R. Cox. Jr., 2004. Less waste corn, more land
in soybeans, and the switch to genetically modified crops, trends with important
implications for wildlife management. Wildlife Society Bulletin 32:127–136.
Kruse, K. L., D. E. Sharp, and J. A. Dubovsky. 2008. Status and harvests of
sandhill cranes: Mid-Continent, Rocky Mountain, and Lower Colorado River
Valley Populations. Administrative Report. U.S. Fish and Wildlife Service,
Denver, Colorado, USA.
Lewis, J. C. (Chairman). 1977. Sandhill crane. Pages 4–43 in G. C. Sanderson,
editor. Management of migratory shore and upland game birds in North
America. International Association of Fish and Wildlife Agencies,
Washington, D.C., USA.
Lewis, J. C. 1979. Field identification of juvenile sandhill cranes. Journal of
Wildlife Management 43:211–214.
Littlefield, C. D., and R. A. Ryder. 1968. Breeding biology of the greater sandhill
crane on Malheur National Wildlife Refuge, Oregon. Transactions North
American Wildlife and Natural Resources Conference 33:444–454.
Littlefield, C. D., and S. P. Thompson. 1982. The Pacific Coast population of
lesser sandhill cranes in the contiguous United States. Pages 288–294 in J. C.
Lewis, editor. Proceedings 1981 Crane Workshop. National Audubon Society,
Tavernier, Florida, USA.
Longmire, J. L., R. E. Ambrose, N. C. Brown, T. J. Cade, T. Maechtle, W. S.
Seegar, F. P. Ward, and C. M. White. 1991. Use of sex-linked mini-satellite
fragments to investigate genetic differentiation and migration of North
American populations of the peregrine falcon (Falco peregrinus). Pages 217–
229 in T. Burke, F. G. Dolf, A. Jefferys, and R. Wolff, editors. DNA finger-
printing: approaches and applications. Birkhauser Press, Basil, Switzerland.
Lumsden, H. G. 1971. The status of the sandhill crane in northern Ontario.
Canadian Field Naturalist 85:285–293.
Madsen, C. R. 1967. Food and habitat selection by fall migrant sandhill cranes in
Kidder County, North Dakota. Thesis, Michigan State University, East
Lansing, USA.
Manning, T. H., and A. H. MacPherson. 1961. A biological investigation of
Prince of Wales Island, N.W.T. Transactions of the Royal Canadian Institute
33(Part 2): 116–239.
McConnell, B. J., C. Chambers, and M. A. Fedak. 1992. Foraging ecology of
southern elephant seals in relation to bathymetry and productivity of the
southern ocean. Antarctic Science 4:393–398.
Meine C. D., and G. W. Archibald. (editors). 1996. The cranes—status survey
and conservation action plan. IUCN, Gland, Switzerland, and Cambridge,
United Kingdom.
Melvin, S. M., and S. A. Temple. 1983. Fall migration and mortality of Interlake,
Manitoba sandhill cranes in North Dakota. Journal of Wildlife Management
47:805–817.
Melvin, S. M., W. J. D. Stephen, and S. A. Temple. 1990. Population estimates,
nesting biology, and habitat preferences of Interlake, Manitoba, sandhill cranes,
Grus canadensis. Canadian Field-Naturalist 104:354–361.
Mickelson, P. G. 1987. Management of lesser sandhill cranes staging in Alaska.
Pages 264–275 in J. C. Lewis, editor. Proceedings of the 1985 Crane
Workshop. Platte River Whooping Crane Habitat Maintenance Trust and
U.S. Fish and Wildlife Service, Grand Island, Nebraska, USA.
Montgomery, J. B. Jr. 1997. Sandhill crane use of the Mid-Pecos Valley of Eastern
New Mexico. Pages 157–164 in R. A. Urbanek, and D. W. Stahlecker, editors.
Proceedings of the Seventh North American Crane Workshop, 10–13 Jan 1996,
Biloxi, Mississippi, USA.
Parmalee, D. F., and S. D. MacDonald. 1960. The birds of west-central Ellsmere
Island and adjacent areas. National Museum of Canada Bulletin 169. Queen’s
Printer, Ottawa, Ontario, Canada.
Pearse, A. T., G. L. Krapu, D. A. Brandt, and P. J. Kinzel. 2010. Changes in
agriculture and abundance of snow geese affect carrying capacity of sandhill
cranes in Nebraska. Journal of Wildlife Management 74:479–488.
Peterson, J. L., R. Bischof, G. L. Krapu, and A. L. Szalanski. 2003. Genetic
variation in the midcontinental population of sandhill cranes, Grus canadensis.
Biochemical Genetics 41:1–12.
Petrula, M. J., and T. C. Rothe. 2005. Migration chronology, routes, and
distribution of Pacific Flyway Population lesser sandhill cranes. Pages 53–67
in F. Chavez-Ramirez, editor. Proceedings of the Ninth North American Crane
Workshop, North American Crane Working Group, 17–20 Jan 2003,
Sacramento, California, USA.
Platte River Recovery Implementation Program. 2010. About the program.
<http://platteriver program.org/aboutus.aspx>. Accessed 21 Jan 2010.
Reinecke, K. J., and G. L. Krapu. 1986. Feeding ecology of sandhill cranes
during spring migration in Nebraska. Journal of Wildlife Management 50:
71–79.
Rhymer, J. M., M. G. Fain, J. E. Austin, D. H. Johnson, and C. Krajewski. 2001.
Mitochondrial phylogeography, subspecific taxonomy, and conservation
genetics of sandhill cranes (Grus canadensis; Aves; Gruidae). Conservation
Genetics 2:203–218.
Schmitt, C. G., and B. Hale. 1997. Sandhill crane hunts in the Rio Grande Valley
and southwest New Mexico. Pages 219–231 in R. P. Urbanek, and D. W.
Stahlecker, editors. Proceedings of the Seventh North American Crane
Workshop, 10–13 Jan 1996, Biloxi, Mississippi, USA.
Krapu et al. Geographic Distribution of Sandhill Cranes 35
Service Argos. 2008. User’s manual. Servos Argos, Inc., Landover, Maryland,
USA.
Sharp, D. E., and J. E. Cornely. 1997. Summary of sandhill crane hunting seasons
in North Dakota, 1968–94. Pages 209–218 in R. P. Urbanek, and D. W.
Stahlecker, editors. Proceedings of the Seventh North American Crane
Workshop, 10–13 Jan 1996, Biloxi, Mississippi, USA.
Sharp, D. E., K. L. Kruse, and J. A. Dubovsky. 2007. Status and harvests
of sandhill cranes: Mid-continent and Rocky Mountain populations.
Admini-strative Report. U.S. Fish and Wildlife Service, Denver, Colorado,
USA.
Soine. J., 1982. Roost habitat selection by sandhill cranes in central North Dakota.
Pages 88–94 in J. C. Lewis, editor. Proceedings of the 1981 Crane Workshop.
National Audubon Society, Tavernier, Florida, USA.
Soutullo, A., L. Cahahia, V. Urios, M. Ferrer, and J. J. Negro. 2006. Accuracy of
lightweight satellite telemetry: a case study in the Iberian Peninsula. Journal of
Wildlife Management 71:1010–1015.
Tacha, T. C., and P. A. Vohs. 1984. Some population parameters of sandhill
cranes from midcontinental North America. Journal of Wildlife Management
48:89–98.
Tacha, T. C., C. Jorgenson, and P. S. Taylor. 1985. Harvest, migration, and
condition of sandhill cranes in Saskatchewan. Journal of Wildlife Management
49:476–480.
Tacha, T. C., S. A. Nesbitt, and P. A. Vohs. 1994. Sandhill crane. Pages 77–93 in
T. C. Tacha, and C. E. Braun, editors. Migratory shore and upland game bird
management in North America. The International Association of Fish and
Wildlife Agencies, Washington, D.C., USA.
Texas Parks and Wildlife Department Program. 2010. Sandhill Crane Hunting
Zones. <http://www.tpwd.state.tx.us/huntwild/season/migratory_game/
crane_zone/>. Accessed 21 Jan 2010.
Tiner, R. W. Jr. 1984. Wetlands of the United States: current status and recent
trends. U.S. Government Printing Office, Washington, D.C., USA.
Toepfer, J. E., and R. A. Crete. 1979. Migration of radio-tagged greater sandhill
cranes from Minnesota and Wisconsin. Pages 159–174 in J. C. Lewis, editor.
Proceedings 1978 Crane Workshop, Rockport, Texas, Colorado State
University Printing Service, Fort Collins, USA.
U.S. Fish and Wildlife Service. 1981. The Platte River ecology study. U.S. Fish
and Wildlife Service Research Report, Jamestown, North Dakota, USA.
<http://www.npwrc.usgs.gov/resource/habitat/plriveco/index.htm>. Accessed
20 Jan 2010.
Vincent, C., B. J. McConnell, V. Ridoux, and M. A. Fedak. 2002. Assessment of
Argos location accuracy from satellite tags deployed on captive gray seals.
Marine Mammal Science 18:156–166.
Walkinshaw, L. H. 1949. The sandhill cranes. Cranbrook Institute of Science
Bulletin 29:1–22.
Walkinshaw, L. H. 1965. Sandhill crane studies on Banks Island, N.W.T. Blue
Jay 23:66–72.
Walkinshaw, L. H. 1981. The sandhill cranes. Pages 151–162 in J. C. Lewis, and
H. Masatomi, editors. Crane Research Around the World: Proceedings of the
International Crane Symposium at Sapporo, Japan in 1980 [21–22 Feb] and
Papers from the World Working Group on Cranes, International Council for
Bird Preservation. International Crane Foundation, Baraboo, Wisconsin.
Wheeler, R. H., and J. C. Lewis. 1972. Trapping techniques for sandhill crane
studies in the Platte River Valley. U.S. Department of Interior, U.S. Fish and
Wildlife Service Resource Publication 107. Washington, D.C., USA.
Williams, G., and R. Glover. 1969. Andrew Graham’s observations on Hudson
Bay 1767–91. Hudson Bay Record Society, London, United Kingdom.
Woynillowicz, D., C. Severson-Baker, and M. Raynolds. 2005. Oil sands fever:
the environmental implications of Canada’s oil sands rush. The Pembina
Institute, Canada. <http://www.oilsandswatch.org/pub/203>. Accessed 21
January 2010.
Yildirim, F., and A. Kaya. 2008. Selecting map projections in minimizing area
distortions in GIS applications. Sensors 8:7809–7817.
Appendix A. List of acronyms and abbreviations.
Acronym Definition
CPRV Central Platte River Valley
DAR Distance Angle Rate
EC–M East-central Canada–Minnesota
EP Eastern Population
GCMU Gulf Coast Management Unit
JD Julian date
KDE Kernel Density Estimate
LC Location
MCP Midcontinent Population
MRD Minimum Redundant Distance
NC–N Northern Canada–Nunavut
NPRV North Platte River Valley
NWR National Wildlife Refuge
PFP Pacific Flyway Population
PTT Platform Transmitter Terminal
PPR Prairie Pothole Region
RMP Rocky Mountain Population
USGS U.S. Geological Survey
WA–S Western Alaska–Siberia
WC–A West-central Canada–Alaska
WK Week
WMA Wildlife Management Area
36 Wildlife Monographs 175
Appendix B: Mean number of days (and %) exposure of Platform Transmitting Terminal (PTT)-tagged MCP sandhill cranes
a
by breeding affiliation to hunting and no-
hunting periods by state (U.S., Mexico) and province (Canada) from 1 September through 10 March, 1998–2004.
State or province Breeding affiliation
b
Hunting Non-hunting
n
c
Mean exposure days % nMean exposure days % Total exposure days
Alberta
d
WA–S 0 0.0 0.0 29 2.1 100.0 60
NC–N 0 0.0 0.0 1 1.0 100.0 1
WC–A 0 0.0 0.0 9 8.0 100.0 72
Alaska WA–S 36 7.6 100.0 0 0.0 0.0 272
WC–A 2 6.0 100.0 0 0.0 0.0 12
Arizona WA–S 2 9.0 12.9 2 60.5 87.1 139
British Columbia
d
WA–S 0 0.0 0.0 15 1.3 100.0 19
Chihuahua WA–S 6 70.8 78.7 9 12.8 21.3 540
Colorado WA–S 20 1.6 97.0 1 1.0 3.0 33
NC–N 1 1.0 100.0 0 0.0 0.0 1
WC–A 2 1.0 100.0 0 0.0 0.0 2
Coahuila WA–S 0 0.0 0.0 1 79.0 100.0 79
WC–A 0 0.0 0.0 1 95.0 100.0 95
Kansas WA–S 0 0.0 0.0 6 1.0 100.0 6
NC–N 1 4.0 22.2 8 1.8 77.8 18
WC–A 2 19.0 51.4 15 2.4 48.6 74
EC–M 13 13.8 56.1 19 7.4 43.9 319
Manitoba NC–N 0 0.0 0.0 8 2.4 100.0 19
EC–M 20 24.6 96.8 4 4.0 3.2 507
Minnesota
d
EC–M 0 0.0 0.0 8 26.6 100.0 213
Montana WA–S 14 1.1 100.0 0 0.0 0.0 16
WC–A 1 1.0 25.0 2 1.5 75.0 4
North Dakota NC–N 17 13.8 97.5 3 2.0 2.5 241
WC–A 19 2.2 100.0 0 0.0 0.0 41
EC–M 20 14.3 67.0 11 12.8 33.0 427
Nebraska
d
WA–S 0 0.0 0.0 11 1.7 100.0 19
NC–N 0 0.0 0.0 7 1.0 100.0 7
WC–A 0 0.0 0.0 20 2.4 100.0 48
EC–M