REINTRODUCTION TECHNIQUES: POST-RELEASE PERFORMANCE OF SANDHILL CRANES (1) RELEASED INTO WILD FLOCKS AND (2) LED ON MIGRATION BY ULTRALIGHT AIRCRAFT

Transcription

1 REINTRODUCTION TECHNIQUES: POST-RELEASE PERFORMANCE OF SANDHILL CRANES (1) RELEASED INTO WILD FLOCKS AND (2) LED ON MIGRATION BY ULTRALIGHT AIRCRAFT RICHARD P. URBANEK, U.S. Fish and Wildlife Service and International Crane Foundation, Necedah National Wildlife Refuge, W th Street West, Necedah, WI 54646, USA JOSEPH W. DUFF, Operation Migration, P.O. Box 280, Blackstock, ON L0B 1B0, Canada SCOTT R. SWENGEL, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53913, USA LARA E. A. FONDOW, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53913, USA Abstract: Two methods of reintroducing a migratory population of whooping cranes (Grus americana) were tested with costume/ isolation-reared juvenile greater sandhill cranes (G. canadensis tabida): (1) release into wild flocks during autumn staging and (2) leading on autumn migration by ultralight aircraft. Birds in the first group were released singly, and all integrated quickly into the wild flocks and adopted similar behavioral patterns. Birds in the second group were led to winter on an inland site on the Gulf Coast of Florida. Most of the birds led by ultralight aircraft remained in their juvenile cohort through the following summer and wandered more extensively than the birds released into autumn staging flocks. Both groups demonstrated adequate survival, return rates to Central Wisconsin (8/8 for autumn release, 9/11 for ultralight-led migration), foraging, roosting, social association, and human avoidance behaviors. Both of these techniques could be effective in reintroducing a migratory population of whooping cranes. PROCEEDINGS NORTH AMERICAN CRANE WORKSHOP 9: Key words: reintroduction, techniques, sandhill crane, Grus canadensis, whooping crane, migratory population, autumn release, ultralight aircraft. As part of an effort to develop reintroduction techniques for the endangered whooping crane (Grus americana), greater sandhill cranes (Grus canadensis) were costume/isolationreared and released in 2000 according to 2 different methods. One method involved releasing captive-reared sandhill cranes into wild sandhill crane flocks that were staging on the rearing area prior to autumn migration. These releases were made according to the one-by-one method developed by Ellis et al. (2001). The other method involved leading sandhill cranes with ultralight aircraft to a predetermined wintering area (Duff et al. 2001). We describe the survival, movements, and general behavior of these 2 groups of birds. This paper is a contribution of the Whooping Crane Eastern Partnership, a consortium of federal and state agencies and non-profit organizations committed to reestablishment of a migratory population of whooping cranes in eastern North America. STUDY AREAS Central Wisconsin Reintroduction Area The selected core whooping crane reintroduction area consists of a large shallow wetland complex in watersheds in Juneau, Wood, Jackson, Monroe, Clark, and Adams Counties. Approximately 20,170 ha of marsh occur in federal or state ownership on Necedah National Wildlife Refuge (NWR), Necedah Wildlife Management Area (Meadow Valley State Wildlife Area [SWA]), Sandhill SWA, and Wood County SWA. At least as much shallow wetland is present on other lands, including cranberry properties, within this core area. The landscape is an interspersion of shallow wetlands, forests, and farmlands on poorly drained, sandy soils of low relief. Corn is a major crop. The specific site of the reintroduction, Necedah NWR, contains approximately 7,725 ha of suitable crane habitat in marshland (6,860 ha) or pools with water-control structures (865 ha) (Trick 2001). Dominant plants include sedges (Carex spp.), cattails (Typha spp.), willows (Salix spp.), and reed canary grass (Phalaris arundinacea). Refuge lands also include 695 ha of scrub-shrub, 8,530 ha of forest, and 686 ha of grasslands. The dominant forest type is Hill s oak (Quercus ellipsoidalis) along with red pine (Pinus resinosa), jack pine (P. banksiana), and quaking aspen (Populus tremuloides). Migration Route The local population of greater sandhill cranes migrates from Central Wisconsin to a major extended stopover at Jasper- Pulaski State Fish and Wildlife Area (SFWA) in northwestern Indiana and then on to wintering areas in southern Georgia and peninsular Florida (Toepfer and Crete 1979, McMillen 1988, McMillen et al. 1991). As a result of changes in management practices, i.e., planting of corn as food for cranes, Hiwassee Wildlife Refuge in eastern Tennessee, equidistant between Jasper-Pulaski SFWA and the traditional wintering areas, had also developed into a major migration stopover and wintering area by the early 1990 s. The route used by ultralight aircraft to lead juveniles on their first migration deviated significantly from the sandhill route by avoiding Chicago. The aircraft route took a 203

2 204 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. Proc. North Am. Crane Workshop 9:2005 wide berth to the west to avoid flight over the congested airspace of that large metropolis. The route also passed south of and did not include Jasper-Pulaski SFWA. Wintering Areas The traditional wintering areas of the eastern population of greater sandhill cranes include numerous sites extending from Okefenokee NWR and Banks Lake NWR/Grand Bay Wildlife Management Area in southern Georgia southward through peninsular Florida to the northern edge of Lake Okeechobee (Urbanek et al. 1988). Without parents to lead them to specific, predetermined wintering areas, released captive-reared cranes tend to remain within the wild flocks in the northern part of this wintering range (Urbanek and Bookhout 1994). Traditional wintering habitats have consisted of wet prairies, muck farms, and cattle ranchland (Urbanek et al. 1988). During the extensive droughts of the 1990 s that continued through this study, shallows and mudflats of partially dewatered lakes also served as wintering habitat. The release site used by ultralight-led birds during their first winter consisted of grassland/live oak (Quercus virginiana) savanna with some small wetlands on Crystal River State Buffer Preserve (SBP), St. Martins Marsh Aquatic Preserve Complex, Citrus County, Florida. The site was pastureland before state ownership, and an active private cattle pasture adjoined the property. This site did not contain wintering greater sandhill cranes and was deliberately so chosen to provide a more representative test of spring migration of reintroduced whooping cranes. The latter would be wintering in saltmarsh and not be migrating with wild sandhill cranes during spring (Urbanek et al. 2005). METHODS Eggs were collected from nests of wild sandhill cranes within the proposed core reintroduction area for whooping cranes in Juneau, Wood, Jackson, and Monroe Counties, westcentral Wisconsin, between 2-5 May 2001, and then assigned to 1 of 2 groups: Nine eggs were retained for hatching and costume/isolation-rearing of chicks at a captive-rearing facility at a remote field site on Necedah NWR. These juveniles would eventually be released into flocks of wild sandhill cranes staging in Central Wisconsin during autumn The remaining 20 eggs were transferred to Patuxent Wildlife Research Center (Patuxent), Laurel, Md., where chicks were hatched and trained to follow ultralight trike aircraft (Cosmos, Dijon, France) according to techniques developed by Operation Migration (Lishman et al. 1997, Duff et al. 2001). These juveniles would complete autumn migration 2001 by following ultralight aircraft. Cranes Released into Autumn Staging Wild Flocks Eight sandhill cranes (6 males, 2 females) were hatched (3-7 May) and costume/isolation-reared (Horwich 1989, Urbanek and Bookhout 1992) at the rearing facility on Necedah NWR. The ninth chick, hatched on 11 May, was removed from the study and remanded to permanent captivity after developing an eye infection. When the chicks were days old, they were also used in another study to assess the necessity of early training for juveniles to be led by ultralight aircraft (Urbanek 2000). That study was, however, discontinued after aircraft landing and other logistical problems developed. Chicks were usually led behind a costumed parent in natural habitats for at least an hour twice daily until after fledging. After fledging, cranes were allowed to roam freely in habitats in the rearing area during the day and were kept in a 650-m 2 top-netted pen at night. As in previous studies (Urbanek and Bookhout 1992), a costumed dummy was successfully used to attract and hold fledged juveniles at a desired roost point. Releases occurred each day from 15 to 23 October at 4 major sandhill crane staging areas in Central Wisconsin: Gallagher Marshes, Sandhill SWA (Fig. 1: H); Sprague-Mather Pool (Fig. 1: S) and Rynearson Pools (Fig. 1: R), Necedah NWR; and Quincy Bluff Preserve, Nature Conservancy (Fig. 1: Q). Three of these sites, i.e., those on Necedah NWR and Sandhill SWA, were centrally-located within the core reintroduction area for the planned whooping crane reintroduction. All sites contained hundreds, or at Sandhill SWA thousands, of staging cranes. All birds were released into the wild sandhill crane flocks singly at roost sites near dusk except for 1 bird released at a feeding site in mid-afternoon. Cranes Trained to Follow Ultralight Aircraft From 20 eggs allotted to this part of the study, 18 hatched 7-19 May. Two eggs did not hatch due to death of late embryos and 4 chicks either died (1 from collision with aircraft) or were euthanized at Patuxent because of developmental problems. After initial training to follow wingless aircraft at Patuxent, the remaining 14 cranes were transported to Necedah NWR on 29 June at days of age and transferred to large outdoor pens with adjacent aircraft training areas. Each pen included a large, fenced wet portion and a smaller, panel-enclosed dry portion. Only the dry portion was top-netted. Juveniles had access to both portions during morning through afternoon, but they were locked in the dry pen overnight to protect them from predators. Training to follow aircraft continued throughout the summer. One juvenile was euthanized after a mid-air collision with aircraft at Necedah; 1 juvenile broke away on the first day of ultralight-led migration, joined wild sandhill cranes, and was not recaptured; and 1 juvenile died from a collision in the overnight holding pen during migration. Thus 11 cranes (5 males, 6 females) reached the wintering area in Florida and remained for study after release. The ultralight-led group left Necedah NWR on 3 October and arrived at their predetermined wintering site on Crystal River SBP on 11 November (32 days). The birds were placed

3 Proc. North Am. Crane Workshop 9:2005 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. 205 Fig. 1. Locations of experimental hatch-year 2000 sandhill cranes in Central Wisconsin, Juneau County: (N) Lemonweir and Wisconsin Rivers, (M) Mill Bluff State Park and Lemonweir River, (O) Meadow Valley Flowage, (R) Rynearson Pools and southeastern Necedah NWR, (S) Sprague-Mather Pool, Necedah NWR, and Finley, Adams County: (I) Ship Rock, (L) Leola Grasslands, (Q) Quincy Bluff, (T) Petenwell Dam and upper Castle Rock Lake. Wood County: (C) Rocky Run tributary, (H) Sandhill SWA, (V) Vesper, (W) Wood County SWA. Clark County: (P) Pray Avenue, (Y) Upper and South Branch Yellow River. Jackson County: (F) Bear Bluff. Monroe County: (E) Shennington, Wyeville, and Oakdale. Marquette County: (B) Briggsville and Neenah Creek, (D) Duffy s Marsh and Thompson Lakes, (G) Gale Road, (K) Klawitter Creek. in a small top-netted holding pen within the larger 1.4-ha open release pen. On 13 November the flock was led behind aircraft to a pen on a small island along Haulover Creek in the Gulf Coastal marshes of Hernando County on southern Chassahowitzka NWR. They were dropped at this site, i.e., the aircraft descended over costumed caretakers on the ground, the birds landed with the caretakers, and then the aircraft ascended quickly and left the area. The sole purpose of this latter flight was to simulate completion of an actual whooping crane migration to the saltmarsh. However, because this type of habitat is not suitable for sandhill cranes, the birds spent 2 nights at this site, then were picked up on 15 November, i.e., by reverse of the drop procedure, and led back to the pen site on Crystal River SBP, where they wintered inland. On 16 November the temporary holding pen was removed and the cranes were released as free-flying birds into the open pen from which they could roam at will. A costumed dummy, used successfully in previous studies with sandhill cranes to control roost site location (Urbanek and Bookhout 1992, Ellis et al. 2001) was positioned in the center of a pond within the enclosure to hold released juveniles at a desired roost site. Monitoring Before being led on migration, birds were individually marked with colored legbands and equipped with leg-bandmounted VHF ( MHz) lithium battery (Advanced Telemetry Systems, Isanti, Minn.) or solar/nicad (Telemetry Systems, Mequon, Wis.) transmitters. After they were released, cranes were tracked by conventional (VHF) telemetry with scanner receivers (Advanced Telemetry Systems, Isanti, Minn.; Telonics, Mesa, Ariz.). Most of this tracking was done from vehicles on the ground, although Cessna aircraft were sometimes used, especially during mi-

4 206 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. Proc. North Am. Crane Workshop 9:2005 gration and to search for missing birds. Each ground tracking vehicle was equipped with a through-the-roof, 7-element yagi antenna (Cushcraft Corporation, Manchester, N.H.). RESULTS Cranes Released into Autumn Staging Wild Flocks Autumn Staging All birds integrated completely into wild sandhill crane flocks within a few days. Five males did so immediately and remained in the wild flocks into which they were released. Another male (no. 5), released at a feeding site 16 km east of the rearing area, returned and landed alone at the rearing area that evening when the wild flock flew over en route to the roost. He was subsequently re-released at Sprague- Mather Pool and remained with the wild flock after this second release attempt. One female (no. 6) separated from the wild flock at the roost the morning after release on Sprague-Mather Pool. She flew alone about 18 km eastward and then sought cover in the edge of some woods. She was retrieved by costumed parents and re-released at the same pool but returned to the rearing area 3 days later. She was then released again, this time on Sandhill SWA, and on this attempt she remained with the wild flock. A second released female (no. 1) spent her first day after release with a solitary wild pair but thereafter joined the wild flock. Cranes numbers 2, 4, 6, and 8, released at Gallagher Marshes on Sandhill SWA, continued to roost on that area until migration. They foraged daily in farm fields 5-29 km north or northeast of the roost site. None of these individuals re-associated with each other after becoming members of the large flock at Sandhill SWA. Nos. 1 and 5, released at Sprague-Mather Pool, Necedah NWR, continued to roost on that area. They foraged daily in farm fields 8-11 km northeast, where they also often loafed in adjacent cranberry beds. No. 3, released at Quincy Bluff, remained to roost in that area until migration and foraged daily in farm fields 5-10 km south or southwest along the Wisconsin River. No. 7, released on East Rynearson Pool, Necedah NWR, continued to use that roost as well as a roost 16 km southwest at Mill Bluff State Park. While roosting at the latter site, he foraged daily in farm fields 2-7 km away. The primary farm field type used by all staging cranes was harvested corn. Autumn Migration All eight birds migrated successfully from Central Wisconsin approximately November. Four males (nos. 2, 5, 7, and 8) were recorded during autumn migration at Jasper-Pulaski SFWA, Indiana. Monitoring effort there was minimal, especially during the first week of migration. First Winter (2000/01). - Six of the 8 cranes were confirmed wintering within large wild flocks on the following major sandhill crane wintering areas (Table 1, Fig. 2): Grand Prairie, Ware County, Okefenokee NWR, Georgia (no. 5, male) Hixtown Swamp, Madison Co., Florida (no. 6, female) Ashley Prairie and Smith Lake, Putnam Co., Florida (no. 2, male) Fish Prairie, Marion Co.; and Paynes Prairie and Lake Kanapaha, Alachua Co.; Florida (no. 4, male) Hiwassee Wildlife Refuge, Meigs Co., Tennessee (nos. 7 and 8, males) One female (no. 1) and one male (no. 3) with malfuncional transmitters were not found during the winter. Return and Summer All 8 of the cranes returned to Central Wisconsin in spring 2001 as integral members of the wild population. No. 7 was the earliest documented return on 10 April. He remained in the East Rynearson Pool area (Fig. 1: R) for the remainder of the summer and also staged there. No. 2 also spent the summer at Rynearson Pools but then staged primarily in the Petenwell Dam area in nearby Adams County (Fig. 1: T). No. 4 also summered and staged at Rynearson Pools after spending most of spring in southwestern Clark County (Fig. 1: P). No. 8 summered in Bear Bluff, Jackson County; Wood County SWA, Wood County; and Meadow Valley SWA, Juneau County (Fig. 1: F, W, and O, respectively) and then staged on Sprague-Mather Pool (Fig. 1: S) and Rynearson Pools, Necedah NWR, and at Mill Bluff State Park (Fig. 1: M). No. 6, a female, wandered among several sites in spring (Fig. 1: S, V, L, R, Y, and H) before summering in west-central Wood County (Fig. 1: C) and staging at Sandhill SWA (Fig. 1: H). One male (no. 5) returned to his release site at Sprague-Mather Pool in spring, summered in Leola Grasslands (Fig. 1: L), Adams County, and later staged at Sandhill SWA. Another male (no. 3), which appeared at Rynearson Pools in late spring, and a female (no. 1), observed 6 times in April near Briggsville (Fig. 1: B), 54 km southeast of the rearing site, could not be monitored further because of nonfunctional transmitters. The 6 birds still being monitored in late autumn (5 remaining with functioning transmitters but broken antennae) began their second autumn migration either on or shortly before 19 November Three were found at Jasper-Pulaski SFWA on that evening and another at Hiwassee State Wildlife Refuge in eastern Tennessee on 25 November. Second Winter (2001/02). - During winter 2001/02, the 5 birds with functional transmitters were found in wild flocks at Jasper-Pulaski SFWA (1), Hiwassee Wildlife Refuge (2), and in southern Georgia or northern Florida (2). Three of these birds used the same areas (nos. 4 and 8) or areas within 25 km (no. 5) as during the previous winter (Table 1). Winter 2001/02 was extremely mild at Jasper-Pulaski SFWA with no snow cover for most of the season. For the first time on record, 10,000-12,000 cranes overwintered at that location in the northern part of the migration route. Some southbound migration apparently occurred in mid-january when some of these cranes moved only as far south as Hiwassee Wildlife Refuge in southeastern Tennessee before returning to Jasper-Pulaski SFWA only 2-3 weeks later. No. 2 and possibly no. 6 apparently shortstopped in response to this mild weather. Spring 2002 through Spring Seven of 8 birds from

5 Proc. North Am. Crane Workshop 9:2005 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. 207 Table 1. Locations of experimental hatch-year 2000 sandhill cranes during winters 2000/01 and 2001/02. St. Martins = Crystal River State Buffer Preserve, St. Martins Marsh Aquatic Preserve Complex, Citrus County, Florida. Hiwassee = Hiwassee State Wildlife Refuge, Meigs County, Tennessee. J-P = Jasper-Pulaski SFWA, Jasper, Pulaski, and Stark Counties, Indiana. nt = nonfunctional transmitter. (Numbers in parentheses refer to locations in Fig. 2). Crane no. Winter 2000/01 Winter 2001/02 Cranes Led by Ultralight Aircraft 1 St. Martins nt 2 Deltona, Volusia Co., Fla. (5) Clermont, Lake Co., Fla. (8) 3 St. Martins Hiwassee (November), nt 4 St. Martins J-P, Hiwassee 5 St. Martins Lake Tsala Apopka, Citrus Co., Fla. (7) 6 St. Martins nt 8 St. Martins Hiwassee, J-P 9 St. Martins J-P, nt 11 St. Martins J-P, Hiwassee 12 St. Martins J-P, Hiwassee 14 St. Martins J-P, nt Cranes Released into Autumn Staging Wild Flocks 1 nt nt 2 Putnam Co., Fla. (1) J-P 3 nt nt 4 Gainesville area, Alachua Co., Fla. (2) Gainesville area, Alachua Co., Fla. (2) Floyds Island Prairie, Charlton County, Okefenokee NWR, Ga. (6) 5 Grand Prairie, Ware County, Okefenokee NWR, Ga. (3) 6 Hixtown Swamp, Madison Co., Fla. (4) Hiwassee (January), unknown 7 Hiwassee nt 8 Hiwassee Hiwassee the original release were confirmed back in the core Central Wisconsin study area in spring 2002 (Table 1). Nos. 2, 4, and 7 were paired with wild females in the Rynearson Pools area or areas near the Yellow River to the east. No. 8, also paired, had apparently established a territory on Meadow Valley SWA, but died during the summer. His remains were recovered by a hunter during autumn. No. 6 summered at Bear Bluff. Nos. 3 and 5 were observed at East Rynearson and Sprague-Mather Pools, respectively. By spring 2003 transmitters were largely nonfunctional and no birds could be consistently tracked. However, nos. 2 and 7 were observed back in the East Rynearson Pool area, and female no. 6 summered at Bear Bluff. [After this paper was written, no. 7 and his mate successfully hatched a chick in 2004 on their territory at East Rynearson Pool. The chick was within a few weeks of fledging when it disappeared, probably taken by a predator.] Cranes Led by Ultralight Aircraft First Winter (2000/01).- All 11 sandhill cranes that were led behind ultralight aircraft to the west-central Gulf Coast of Florida successfully overwintered on Crystal River SBP. The preserve had been chosen as a wintering site for logistical reasons and absence of wintering migratory sandhill cranes, and habitat there was limited. With the exception of 1 bird on 1 night in December, all 11 birds roosted in the pond in the pen each night. They foraged during the day in or near the pen and in a cattle pasture on adjacent property. On 25 February, the 5 males (nos. 1, 4, 8, 11, and 14) and 5 of the females (nos. 3, 5, 6, 9, and 12) departed on spring migration. On 17 March the remaining female (no. 13) departed alone.

6 208 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. Proc. North Am. Crane Workshop 9:2005 Fig. 2. Wintering areas of experimental hatch-year 2000 sandhill cranes in Tennessee, Georgia, and Florida, winters 2000/01 and 2001/02 (an additional wintering area, Jasper-Pulaski SFWA, is not shown): (1) Ashley Prairie and Smith Lake, Putnam County, (2) Gainesville area including Fish Prairie, Marion County, and Paynes Prairie and Lake Kanapaha, Alachua County, (3) Grand Prairie, Ware County, Okefenokee NWR, (4) Hixtown Swamp, Madison County, (5) Big Lake and Savannah Marsh, Deltona, Volusia County, (6) Floyds Island Prairie, Charlton County, Okefenokee NWR, (7) Lake Tsala Apopka, Citrus County, (8) Clermont wetlands, Lake County. The male (no. 2) who had disassociated from the ultralightled flock on the first day of autumn migration wintered with both migratory and nonmigratory sandhills at Deltona, Volusia County, northeastern Florida (Fig. 2). In Wisconsin, he had initially dropped into a staging flock near Mauston, but he then returned to Necedah NWR. He remained in local staging flocks and usually roosted on East Rynearson Pool while foraging in farm fields km east in Adams County, or he roosted in Mill Bluff State Park and foraged in fields 2-7 km from that roost. Summer Nine of the group of 10 cranes returned as a group (nos. 1, 3, 4, 5, 6, 9, 11, 12, and 14) to their rearing area at Rynearson Pools, Necedah NWR (Fig. 1: R) for an overnight stop on 27 April before leaving the next day. Eight of the birds again returned as a group to Rynearson Pools on 18 May, while 1 bird (no. 5) remained behind with wild sandhills in Marquette County (Fig. 1: G). On 19 May the group of 8 birds returned to Marquette County, where they remained at Duffy s Marsh (Fig. 1: D) until 29 or 30 May and then moved 24 km northwest to the Klawitter Creek bottoms (Fig. 1: K). They returned to the refuge again on 18 June. The group remained on the refuge 34 days, and then all but 1 male (no. 14) departed on 23 July. The group of 7 birds then inhabited sites in eastern Adams County (Fig.1: mainly I and L) until 21 September. On that date the group moved to western Adams County (Fig. 1: T), just east of Necedah NWR. On 6 October they returned to the refuge, where the group of 7 rejoined with no. 14 and again became a group of 8. The group then spent most of its time in Monroe County (Fig.1: M and E), 21 km westsouthwest of the original rearing area, with frequent visits to the refuge, until it departed from the refuge among more than 1,000 wild, staging cranes on 19 November. The 5 birds with functional transmitters were tracked to Jasper-Pulaski SFWA in northwestern Indiana on that day, where individuals of the group separated and dispersed among the more than 20,000 wild migrating cranes present. One male (no. 8) of the 12 cranes that wintered in Florida (1 of the group of 10 that departed on 25 February) was not found on the Central Wisconsin summering area or elsewhere. The female (no. 13) that began spring migration alone on 17 March was reported near Old Fort, McDowell County, western North Carolina, on 21 June, having not completed the migration and apparently without fear of humans. She was captured by a local resident and transferred to local animal health personnel. She was transported back to Wisconsin on 16 July and given a medical examination. Because of a swollen head and other symptoms of a chronic but undetermined illness, she was removed from the study and later euthanized. The female (no. 5) that remained behind in Marquette County on 18 June after separating from the larger group spent most of the summer with wild cranes in southern Adams or Marquette Counties. She returned to the refuge on 17 November and migrated to Jasper- Pulaski SFWA on 19 November. The male (no. 2) that joined wild sandhill cranes on the first day of autumn 2000 migration spent most of the summer with wild cranes in the vicinity of Necedah NWR. Second Winter (2001/02). - As noted previously, winter 2001/02 was extremely mild with 10,000 or more cranes overwintering at Jasper-Pulaski SFWA. At least 6 birds (nos. 4, 8, 9, 11, 12, and 14) wintered at Jasper-Pulaski SFWA or Hiwassee Wildlife Refuge, Tennessee, instead of proceeding farther south (Table 1). The female (no. 5) that had disassociated from the main group during the previous spring wintered with a wild flock near Inverness, Citrus County, Florida. This site was 40 km east of where the flock had been led and spent their first

7 Proc. North Am. Crane Workshop 9:2005 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. 209 winter (Fig. 2). She was the only bird found to winter in the general area of the Central Gulf Coast and the only ultralightled bird found returning to winter in Florida. The male (no. 2) that had joined sandhill flocks on the first day of autumn 2000 migration wintered in Florida at a site 62 km southwest of his previous winter location (Table 1). Another male (no. 8), who did not apparently return to Central Wisconsin in spring 2001, was found at Hiwassee Wildlife Refuge and at Jasper-Pulaski SFWA during the winter. Spring 2002 through Spring Seven individuals (nos. 2, 4, 5, 8, 11, 12, and 14) of the original release were known to have returned to Central Wisconsin in spring The males were found on or near Necedah NWR. Female no. 5 spent the spring near Mauston (Fig. 1: N); her transmitter then expired, but later observations indicated that she summered in this area or in nearby Adams County. Female no. 12 wandered among several sites in spring including Rynearson Pools, Leola Grasslands, and Bear Bluff. No. 8 was again not recorded during the summer but was found with a functional transmitter at Jasper-Pulaski SFWA in autumn By spring 2003 all transmitters were nonfunctional and birds could no longer be tracked. However, nos. 1, 6, 11, and 14 were observed back in the Necedah NWR area. Males nos. 1 and 14 were paired with wild-hatched females, and no. 1 had established a territory at northeastern Sprague-Mather Pool. Female no. 6 was observed missing her left foot. DISCUSSION Comparison of One-by-One and Other Methods of Release into Wild Flocks Gentle releases of captive-reared cranes have generally been much more successful than abrupt releases (Ellis et al. 1992, 2000; Nagendran et al. 1996; Horwich 2001). Previous successful studies have often employed a large acclimation pen to effect the gentle release (Valentine and Logan 1987, Urbanek and Bookhout 1992). The one-by-one release method (Ellis et al. 2001) used in this study had in effect, because the birds were reared in natural habitats in the field, maintained sufficient aspects of gentle release while at the same time adding the economy and logistical flexibility of no release pen. This method may thus provide an efficient and effective alternative to the more conventional release procedure. As in previous studies (Urbanek and Bookhout 1992), some released individuals did not immediately integrate into the flock into which they were introduced. However, as in past work, these individuals could easily be retrieved by costumed parents and re-released with different birds or under different circumstances. In all cases, the desired associations occurred after only at most a few trials. Results of the one-by-one release at Necedah NWR were consistent with those of 3 cohorts of sandhill cranes reared and released at Seney NWR, Upper Michigan, (Urbanek and Bookhout 1992, 1994). In those releases, the majority of males returned and established residence on the immediate rearing/release area and then went on to successfully pair, establish territories, nest, and produce young (Duan 1994). All of these birds were costume/isolation-reared from hatching on the experimental reintroduction area and then released locally within wild sandhill flocks. This rearing and release procedure resulted in optimal release candidates who developed strong philopatry toward their natal area. For a reintroduction of whooping cranes to be successful, such philopatry by the initial core group of captive-reared birds will be critical. Comparison of Autumn Release and Ultralight Aircraft-led Migration Survival. - Post-release survival, as demonstrated by birds which could be monitored (i.e., those with functional radiotransmitters or to about 2 years after release), was high in both groups. One individual in the ultralight group had health problems, did not complete spring migration, and was removed from the study. All other birds in both groups survived at least through their first complete migration. Social Behavior. - Members of both groups integrated fully into the wild population. The birds released in autumn did so immediately or shortly after release. This process was delayed, however, for the ultralight-led group by maintenance of the original juvenile cohort during the first year after release. Based on limited observations, pairing behavior appeared normal for both groups. Human Avoidance. - Both methods produced birds which adequately avoided humans. Although the autumn release with wild cranes greatly and initially facilitated this behavior in that group, eventual association of the ultralight-led birds with wild cranes resulted in similar human avoidance behavior by 1 year after release. Migration and Wintering. - Both groups of cranes migrated and wintered in suitable habitat as free birds within wild flocks. However, the ultralight-led group demonstrated poor homing to their previous wintering area. This may have been related to the relative unsuitability of that site as habitat for wintering migratory cranes. The record mild winter of 2001/02 also may have influenced normal wintering patterns. Ultralight-led whooping cranes have demonstrated good homing ability to the wintering area on Chassahowitzka NWR (Urbanek et al. 2005), although most did not remain because of preference for habitat inland. Natal Site Fidelity and Summer Range. - The autumn released group demonstrated better homing to the core study area and less wandering during their first year after release. However, this group had a higher proportion of males than the ultralight-led group, and males home more effectively than females (Urbanek and Bookhout 1994). Also, persistence of the large juvenile cohort during the first summer after release, an occurrence that would not be possible in a natural population, may have influenced movement patterns of the ultralight-led group.

8 210 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al. Proc. North Am. Crane Workshop 9:2005 The interaction of this large group with the local wild population may also have affected movements. The juvenile group dissolved during autumn migration By the second year after release, based on limited data, natal homing was similar for both groups. Management after Release. - Except for some expected retrievals and re-releases, the autumn released birds required no further assistance after release. Once the ultralight-led birds began spring migration from the winter release site, they likewise required no further assistance from monitoring personnel to ensure survival or improve behavior. Minor concerns involving potential exposure to human activity occurred mainly during spring wandering and were largely resolved after the ultralight-led birds returned to the refuge and associated with greater numbers of wild sandhill cranes. Advantages. - The main advantage of reintroducing cranes by ultralight-led migration is that the birds can be led to a predetermined area on which they will winter and to which they will return in subsequent winters if habitat conditions are suitable. Although this premise was not proven in the current study with ultralight-led sandhills, it was supported by subsequent work with whooping cranes (Urbanek et al. 2005). The main advantage of an autumn release is that the major expenses and logistical outlay associated with training birds to follow aircraft, the aircraft-led autumn migration, and care of birds at the release site through the first winter are not incurred. CONCLUSIONS AND MANAGEMENT IMPLICATIONS The post-release survival, return rates, and foraging, roosting, social association, and human avoidance behaviors demonstrated after (1) autumn release of cranes singly into wild flocks and (2) leading cranes on autumn migration by ultralight aircraft indicate that both of these techniques could be effective in reintroducing a migratory population of whooping cranes. Both methods depend on a rigorous costume/isolation-rearing protocol. Cost and winter distribution may differ significantly between the 2 methods. Both methods could also be used to complement each other in the same reintroduction effort. Although autumn release into wild sandhill crane flocks was originally conceived as a standalone technique for reintroducing whooping cranes (Urbanek and Bookhout 1992), juvenile whooping cranes could be released into groups of older whooping cranes that had been previously introduced by ultralight-led migration. This cost-effective augmentation of the population could reduce dispersal and ultimately promote pair formations on the wintering grounds. ACKNOWLEDGMENTS This paper is a product of the Whooping Crane Eastern Partnership (WCEP), which was established in 1999 to reintroduce a migratory population of whooping cranes to eastern North America. The nine founding members are the Canada- U.S. Whooping Crane Recovery Team, U.S. Fish and Wildlife Service (USFWS), USGS Patuxent Wildlife Research Center, USGS National Wildlife Health Center, Wisconsin Department of Natural Resources (DNR), Operation Migration, Inc., International Crane Foundation (ICF), National Fish and Wildlife Foundation, and Natural Resources Foundation of Wisconsin. Many additional organizations and individuals have played an important role in the reintroduction, and the efforts of all participants are acknowledged as vital to its success. We especially thank the following individuals, most of whom worked directly with the birds. Their diligent efforts and conscientious attention to detail made these projects successful: Chick rearing and training to follow ultralight aircraft: D. Sprague, G. Gee, C. Caldwell, G. Olsen, and D. Ellis (Patuxent) and D. Clark (Operation Migration). Ultralight aircraft-led migration: D. Clark, R. Van Heuvelen, W. Lishman, R. Cohen-Pardo, G. Lee, and D. and P. Lounsbury (Operation Migration). Chick rearing and monitoring for autumn release: K. Brunette, M. Knoch, and A. Nelson (USFWS), M. Wellington, K. Maguire, S. Zimorski, T. Kaldenberg, and B. Hartup (ICF), L. Moore (USFWS and ICF), and M. McDaniel and W. Hall (Wisconsin DNR). We are especially grateful to L. Wargowsky and staff (Necedah NWR), S. Blitch and staff (St. Martins Marsh Aquatic Preserve Complex), and J. Kraus and staff (Chassahowitzka NWR) for their support and provision of facilities. We thank the many additional cooperators and state personnel who assisted in monitoring, especially J. Bergens (Jasper-Pulaski SFWA), and J. W. Akins (Hiwassee Wildlife Refuge). T. Kohler (Windway Capital Corporation) graciously provided tracking aircraft and financial support. J. Harris (ICF) and J. Christian (USFWS) were instrumental in facilitating logistic support. LITERATURE CITED Duan, W Social and reproductive behavior of isolationreared, released sandhill cranes (Grus canadensis). Thesis, Ohio State University, Columbus, Ohio, USA. Duff, J. W., W. A. Lishman, D. A. Clark, G. F. Gee, and D. H. Ellis Results of the first ultralight-led sandhill crane migration in eastern North America. Proceedings of the North American Crane Workshop 8: Ellis, D. H., D. P. Mummert, R. P. Urbanek, M. Kinloch, C. Mellon, T. Dolbeare, and D. P. Ossi The one-byone method for releasing cranes. Proceedings of the North American Crane Workshop 8:225 (abstract)., G. H. Olsen, G. F. Gee, J. M. Nicolich, K. E. O Malley, M. Nagendran, S. G. Hereford, P. Range, W. T. Harper, R. P. Ingram, and D. G. Smith Techniques for rearing and releasing nonmigratory cranes: lessons from the Mississippi sandhill crane program. Proceedings of the North American Crane Workshop 6: Horwich, R. H Use of surrogate parental models and age periods in a successful release of hand-reared sandhill cranes. Zoo Biology 8:

9 Proc. North Am. Crane Workshop 9:2005 REINTRO. TECHNIQUES FOR MIGRATORY CRANES Urbanek et al Developing a migratory whooping crane flock. Proceedings of the North American Crane Workshop 8: Lishman, W. A., T. L. Teets, J. W. Duff, W. J. L. Sladen, G. G. Shire, K. M. Goolsby, W. A. Bezner Kerr, and R. P. Urbanek A reintroduction technique for migratory birds: leading Canada geese and isolation-reared sandhill cranes with ultralight aircraft. Proceedings of the North American Crane Workshop 7: McMillen, J. L Productivity and movements of the greater sandhill crane population at Seney National Wildlife Refuge: potential for an introduction of whooping cranes. Dissertation, Ohio State University, Columbus, Ohio, USA., R. P. Urbanek, and T. A. Bookhout Autumn migration of greater sandhill cranes from Upper Michigan, U.S.A. Pages in J. Harris, ed. Proceedings of the 1987 International Crane Workshop, International Crane Foundation, Baraboo, Wisconsin, USA. Nagendran, M., R. P. Urbanek, and D. H. Ellis Special techniques, part D: reintroduction techniques. Pages in D. H. Ellis, G. F. Gee, and C. M. Mirande, eds. Cranes: their biology, husbandry, and conservation. National Biological Service, Washington, D.C., and International Crane Foundation, Baraboo, Wisconsin, USA. Toepfer, J. E., and R. A. Crete Migration of radiotagged greater sandhill cranes from Minnesota and Wisconsin. Pages in J. C. Lewis, ed. Proceedings of the 1978 crane workshop. Colorado State University Printing Service, Fort Collins, Colorado, USA. Trick, J Final environmental assessment: proposed reintroduction of a migratory flock of whooping cranes in the eastern United States. U.S. Fish and Wildlife Service, Green Bay, Wisconsin, USA. Urbanek, R. P Effect of delayed training on following ultralight aircraft by costume isolation-reared sandhill cranes. Unison Call 12(2):8., and T. A. Bookhout Development of an isolation-rearing/gentle release procedure for reintroducing migratory cranes. Proceedings of the North American Crane Workshop 6: , and Performance of captive-reared cranes released into a migration route in eastern North America. Pages in H. Higuchi and J. Minton, eds. The future of cranes and wetlands. Wild Bird Society of Japan, Tokyo, Japan., L. E. A. Fondow, C. D. Satyshur, A. E. Lacy, S. E. Zimorski, and M. Wellington First cohort of migratory whooping cranes reintroduced to eastern North A- merica: the first year after release. Proceedings of the North America Crane Workshop 9: , S. A. Nesbitt, J. L. McMillen, and T. A. Bookhout Winter distribution and site fidelity of greater sandhill cranes from Upper Michigan. Ohio Cooperative Fish and Wildlife Research Unit, Columbus, Ohio, USA. Valentine, J. M., and T. J. Logan The captive release program for the Mississippi sandhill crane. Pages in J. Harris, ed. Proceedings of the 1987 International Crane Workshop, International Crane Foundation, Baraboo, Wisconsin, USA.

10 212 Proc. North Am. Crane Workshop 9:2005

11 FIRST COHORT OF MIGRATORY WHOOPING CRANES REINTRODUCED TO EASTERN NORTH AMERICA: THE FIRST YEAR AFTER RELEASE RICHARD P. URBANEK, International Crane Foundation and U.S. Fish and Wildlife Service, Necedah National Wildlife Refuge, W th Street West, Necedah, WI 54646, USA LARA E. A. FONDOW, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53919, USA COLLEEN D. SATYSHUR, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53919, USA ANNE E. LACY, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53919, USA SARA E. ZIMORSKI, International Crane Foundation, E Shady Lane Road, Baraboo, WI 3919, USA MARIANNE WELLINGTON, International Crane Foundation, E Shady Lane Road, Baraboo, WI 53919, USA Abstract: We describe the post-release movements and survival of the first cohort in the eastern migratory whooping crane (Grus americana) reintroduction from release the first winter through return the second winter. Six cranes were led behind ultralight aircraft from Necedah National Wildlife Refuge (NWR), Central Wisconsin, to Chassahowitzka NWR, Gulf Coast of Florida. After release in Florida, 1 of these cranes and another transported there by truck were killed by bobcats (Lynx rufus). The winter management protocol was modified and no further predation occurred. The 5 remaining cranes migrated unassisted back to Necedah NWR in spring, left the refuge during a spring wandering period, and then 4 returned to Necedah NWR to spend the summer. All 5 birds migrated back to Florida to winter, and 4 returned at least initially to Chassahowitzka NWR or adjacent salt marsh before 3 dispersed to suitable habitat inland. One yearling remained at Chassahowitzka NWR with the newly released juveniles from the second year s release. Of the other 4 birds, 2 wintered separately with sandhill cranes (G. canadensis) in northern Florida, and 2 wintered as a pair on ranchland 62 km from the original winter release site. After the bobcat predation problem was solved, the subsequent survival, migration, summering, and wintering of these reintroduced whooping cranes were favorable for a successful reintroduction. PROCEEDINGS NORTH AMERICAN CRANE WORKSHOP 9: Key words: whooping crane, Grus americana, reintroduction, migratory population, ultralight aircraft, migration, Wisconsin, Florida. Walkinshaw (1978) was one of the first to propose the reintroduction of an eastern migratory population of whooping cranes (Grus americana), and he recommended the Upper Peninsula of Michigan to be the site of this reintroduction. During most of his crane research, which began in the early 1930 s, there were only a small number of whooping cranes surviving in the single viable population that migrated between Wood Buffalo National Park in the Northwest Territories of Canada and Aransas National Wildlife Refuge on the Gulf Coast of Texas. There was no captive propagation program, and the whooping crane hovered on the brink of extinction. In 1966, a captive propagation program began at Patuxent Wildlife Research Center (Patuxent) in Laurel, Maryland. In 1975, the first reintroduction was attempted at Grays Lake National Wildlife Refuge, Idaho. This was to be a migratory flock wintering in New Mexico. In 1983, Bookhout (McMillen 1988) initiated a series of studies involving sandhill cranes in the Upper Peninsula of Michigan in preparation for reintroduction of an eastern migratory flock. However, by 1988 cross-fostering, the primary technique used in the Rocky Mountain reintroduction, had proven to be ineffective, survival in that migratory flock was low, and that effort would eventually be discontinued. Initiation of an eastern migratory whooping flock was dropped from consideration in favor of a non-migratory flock on the Kissimmee Prairie of Central Florida. In 1993 this second reintroduction attempt was begun. That effort to establish a non-migratory flock has continued, but, survival has been low (Nesbitt et al. 1997, 2001). In 2001, after an 18-year effort of proposals, development of new techniques using sandhill cranes (Horwich 1989, 2001; Urbanek and Bookhout 1992, 1994; Lishman et al. 1997; Ellis et al. 2000, 2001, 2003; Duff et al. 2001), planning, and persistence, an initial cohort of reintroduced migratory whooping cranes was led behind ultralight aircraft from Necedah NWR, Central Wisconsin, to Chassahowitzka NWR on the Central Gulf Coast of Florida. Reintroduction of an eastern migratory population of whooping cranes was finally underway. This report documents the survival, movements, and general behavior of the first cohort of whooping cranes in the eastern migratory whooping crane reintroduction. The period covered begins with release of these birds in Florida for their first winter and continues through spring migration, summering in Wisconsin, fall migration, and their second winter in Florida. This paper is a contribution of the Whooping Crane Eastern Partnership, a consortium of federal and state agencies and nonprofit organizations committed to reestablishment of a migratory population of whooping cranes in eastern North America. 213

12 214 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. Proc. North Am. Crane Workshop 9:2005 STUDY AREAS Central Wisconsin Reintroduction Area The core reintroduction area consists of a large shallow wetland complex in watersheds in Juneau, Wood, Jackson, Monroe, Clark, and Adams Counties. Approximately 20,170 ha of marsh occur in federal or state ownership on Necedah NWR, Necedah Wildlife Management Area (Meadow Valley State Wildlife Area [SWA]), Sandhill SWA, and Wood County SWA. At least as much shallow wetland is present on other lands, including cranberry properties, within this core area. The landscape is an interspersion of shallow wetlands, forests, and farmlands on poorly drained, sandy soils of low relief. Corn is a major crop. The specific site of the reintroduction, Necedah NWR, contains approximately 7,725 ha of suitable crane habitat in marshland (6,860 ha) or pools with water-control structures (865 ha) (Trick 2001). Dominant plants include sedges (Carex spp.), cattails (Typha spp.), willows (Salix spp.), and reed canary grass (Phalaris arundinacea). Refuge lands also include 695 ha of scrub-shrub, 8,530 ha of forest, and 686 ha of grasslands. The dominant forest type is Hill s oak (Quercus ellipsoidalis) along with red pine (Pinus resinosa), jack pine (Pinus banksiana), and quaking aspen (Populus tremuloides). Migration Route The reintroduced whooping cranes would share the migration route used by the wild sandhill crane population (Toepfer and Crete 1979, McMillen 1988, Urbanek 1988). The latter route extended from Central Wisconsin to stopovers at Jasper- Pulaski State Fish and Wildlife Area (Jasper-Pulaski) in northwestern Indiana and Hiwassee Wildlife Refuge (Hiwassee) in eastern Tennessee to wintering areas in southern Georgia and peninsular Florida. The route used by ultralight aircraft to lead juveniles on their first migration deviated significantly from the sandhill route by avoiding Chicago. The aircraft took a wide berth to the west to avoid flying through congested airspace of that large metropolis. Also, the ultralight aircraft did not fly near Jasper-Pulaski, a major sandhill crane stopover site. Central Gulf Coast of Florida Wintering Area The release pen on Chassahowitzka NWR was located in an area representative of the surrounding salt marsh. Typical salt/brackish marsh habitats characteristic of the Central Gulf Coast were located within 2.4 km of the pen. Dominant vegetation consisted of extensive monotypic stands of black needlerush (Juncus roemerianus) with scattered islands of cabbage palm (Sabal palmetto). Surface access was by airboat, and airboat operation by the public on this part of the refuge was prohibited. The pen was expanded from 0.6 ha in winter 2001/02 to 1.6 ha in winter 2002/03. The expanded portions included (1) a deeper pool, (2) an artificially constructed oyster bar (on top of an existing natural oyster bar), and (3) an area of salt grass (Distichlis spicata), originally just outside the southeast boundary of the old pen, that was a favorite loafing area for the hatch-year 2001 birds during their first winter and on which they frequently attempted to roost. With the improvements, water of suitable roosting depth was present somewhere in the pen at almost any tidal level in winter 2002/03. Prescribed burned areas were also present adjacent to and in the vicinity of the pen. These amounted to several hundred ha in winter 2001/02 but were less extensive in winter 2002/03. These burns constituted most of the area outside of the pen that was usable by the cranes until early March. By that time needlerush had regrown and rendered much of the burned areas again unusable. METHODS Whooping cranes were hatched (7-24 May 2001) at Patuxent and then trained from shortly after hatching to follow ultralight trike aircraft (Cosmos, Dijon, France) according to techniques developed by Operation Migration, Blackstock, Ontario (Lishman et al. 1997, Duff et al. 2001). On 10 July at days of age, chicks were transferred to large, top-netted outdoor pens with adjacent aircraft training areas on Necedah NWR, Juneau County, Wisconsin. Each pen included separate wet and dry portions, which the juveniles had access to during morning through afternoon. Juveniles were, however, locked in the dry pen overnight as a safety precaution against predators. On 11 September birds were individually marked with colored legbands and equipped with legband-mounted VHF ( MHz) lithium battery (Advanced Telemetry Systems, Isanti, Minn.) or solar/nicad (Telemetry Systems, Mequon, Wis.) transmitters. The initial (2001) cohort left Necedah NWR on 17 October, arrived at an inland site near Chassahowitzka NWR on 3 December, and then arrived at the salt marsh release pen on 5 December (50 days) (J. W. Duff, personal communication). Six birds had completed the migration behind ultralight aircraft. Another bird with early wing problems never successfully completed flight training and was transported by truck for the entire migration. They were enclosed in a small top-netted holding pen within the larger 0.6-ha pen. On 6 December 3 birds, including the male transported by truck, were each equipped with a leg band-mounted satellite transmitter (PTT) (Microwave Telemetry, Columbia, Md.). On the following day the holding pen was removed, and the cranes were released in the larger pen as free-flying birds. They were allowed to roam during the day at will. A costumed dummy, used successfully in previous studies with sandhill cranes to control roost site location (Urbanek and Bookhout 1992, Urbanek et al. 2005) was positioned in the center of the pool within the pen. Cranes were initially also allowed to roost at will until mortalities necessitated a protective roost strategy. After they were released, cranes were tracked by conven-

13 Proc. North Am. Crane Workshop 9:2005 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. 215 tional (VHF) telemetry with scanner receivers (Advanced Telemetry Systems, Isanti, Minn.; Telonics, Mesa, Ariz.). Most of this tracking was done from vehicles on the ground, although Cessna aircraft were sometimes used, especially during migration and to search for missing birds. Each ground tracking vehicle was equipped with a through-the-roof, 7-element yagi antenna (Cushcraft Corporation, Manchester, NH). PTT s were used to identify distant search locations in areas not routinely covered by VHF tracking. RESULTS First Winter Seven juvenile whooping cranes were released into the remote, open-topped pen on Chassahowitzka NWR, Central Gulf Coast of Florida, on 7 December Six of these birds (males nos. 1, 5, and 6 and females nos. 2, 7, and 10) were led by ultralight aircraft from Necedah NWR in Central Wisconsin and reached Chassahowitzka NWR on 5 December. Another bird (male no. 4), transported in a box by truck during migration, had arrived at Chassahowitzka NWR the previous day. Predation. - The male transported by truck was killed just outside the pen perimeter by a bobcat on 17 December. Female no. 10 was killed while attempting to roost in a narrow tidal creek on the night of 9 January. The offending bobcats were trapped and removed from the area, and a continuous trapping effort was implemented. Although bobcat sign occasionally reappeared near the pen site, no additional bobcats were captured. After the second mortality, more rigorous overnight protection measures (i.e., ensuring that the birds roosted either within the predator-proofed pen or in water more than 6 m from shore) were implemented, and no further mortalities occurred. Roosting. - During the 2001/02 winter the released juveniles occupied the original 0.6-ha release pen. Based on behavior at roost time, cranes initiated roosting in water as deep as 23 cm but optimal depth was 15 cm or less. They showed little or no attraction to the costumed dummy. They did not roost on exposed mudflat. Data on water depth in the pool in the pen are available for each night (n = 81) at roosting time during the period 18 January-8 April During this period, water depth at dusk was optimal for roosting on 17 nights (21%), marginally suitable, i.e., adequate but somewhat high on 8 nights (10%), too low on 4 nights (5%), too high (but within banks) on 43 nights (53%), and the entire surrounding landscape was flooded, i.e., tidal creeks and pools exceeded their banks, on 9 nights (11%). The adjoining part of the pool southwest of the original fenced part was deeper; therefore, this larger part of the pool, unfenced in winter 2001/02, was rarely usable by the cranes at roosting time. On the majority of nights, there also appeared to be few or no safe places to initiate roosting in the surrounding tidal landscape because water was too deep (i.e., the bays, creeks, and pools to the west were even deeper than the pool at the release pen). The cranes frequently flew out of the pen at roosting time, landed on the adjacent salt grass area that was unsuitable for roosting, and then had to be led back into the pen by a costumed caretaker as darkness fell. During the period of consistent data collection, 18 January-8 April 2002, birds were led back into the pen on 37% of nights. Except on 3 of these nights, all 5 birds needed to be led into the pen. Birds occasionally roosted safely outside of the pen. This roosting occurred on 17% of nights. These nights were usually characterized by extremely low tides, when birds usually roosted in a tidal pool 0.24 ha east of the pen, or extremely high tides, when birds were allowed to roost on the flooded salt grass loafing area adjacent to the pen. Cranes went to roost in water a safe distance from shore on 34% of nights. Otherwise, they roosted on land. Salinity. - Whooping cranes will drink water with salinity less than 23 parts per thousand (ppt) (Allen 1952, Hunt 1987). Salinities near the pen site were too high to provide a good source of drinking water: January (19-21), February (17-24), March (19-23), April (23-25). Salinity usually decreased briefly only after heavy rains. Cranes were largely dependent on fresh water artificially provided in a drinking receptacle. Foraging and Movements. - In winter 2001/02 cranes ranged km from the pen. The small size of the pen and large amount of burned habitat in the vicinity of the pen contributed to their movement. However, by mid-february needlerush had regrown on the burn, rendering most of this habitat unusable. In late winter until migration, cranes focused their away-from-pen movements on 2 barrens (i.e., open dry land with sparse vegetative ground cover). These areas were 1.6 km east near Rose Creek and 0.8 km south at Pumpkin Creek Impoundment. Cranes were frequently observed foraging on natural foods, including blue crabs (Callinectes sapidus). However, most feeding was on commercial pellets provided at a feeding station within the pen. Spring Migration The 5 cranes that survived the winter (nos. 1, 2, 5, 6, and 7) began migration as a single flock on 9 April and flew to Wilcox County in southcentral Georgia (Fig. 1, Table 1). After being grounded for 2 days with rain, they made a short flight to Henry County, Georgia, just south of Atlanta on 12 April. After another day of rain, they resumed migration on 14 April, when a female (no. 7) separated in flight from the other 4 birds over northern Georgia. She landed in McMinn County, southeastern Tennessee, while the main group of 4 proceeded to Fentress County, northeastern Tennessee. On 15 April the group of 4 flew to Johnson County, southcentral Indiana. The following day they migrated through northern Indiana and being pushed eastward by a strong west wind, the group encountered Lake Michigan at Indiana Dunes. After circling the shoreline for 2 hours (they had never before encountered a large obstacle in their flightpath), they correctly flew westward. They landed to roost in a pool in a gravel pit that was closed to public access in a forest preserve

14 216 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. Proc. North Am. Crane Workshop 9:2005 Fig. 1. Migration route of hatch-year 2001 whooping cranes from Chassahowitzka NWR, Florida, to Necedah NWR, Wisconsin, spring Stopover sites are identified in Table 1.

15 Proc. North Am. Crane Workshop 9:2005 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. 217 Table 1. First spring migration of reintroduced whooping cranes, Cranes left Chassahowitzka NWR, Citrus Co., Florida on 9 April. Crane no. 7 separated from the other 4 cranes during flight in northern Georgia on 14 April. Flight date Crane nos. Stopover site Site (Fig. 1) Distance (km) Time (hrs) 9 April 1,2,5,6,7 Wilcox Co., Georgia April 1,2,5,6,7 Long Branch Reservoir, Henry Co., Georgia April 1,2,5,6 Cumberland Plateau, Fentress Co., Tennessee April 7 Rodgers Creek Unit, Chickamauga WMA, McMinn Co., Tennessee 15 April 1,2,5,6 Johnson Co., Indiana April 7 north-central Kentucky (site undetermined) 8 16 April 1,2,5,6 Schuth's Grove Forest Preserve, Cook Co., Illinois April 7 Jasper Co., Indiana 9 18 April 1,2,5,6 Shaw Marsh SWA, Dodge Co., Wisconsin April 7 Avon Bottoms SWA, Rock Co., Wisconsin April 1,2,5,6 Necedah NWR, Juneau Co., Wisconsin April 7 Wisconsin River, Wauzeka, Crawford Co., Wisconsin May 7 Necedah NWR, Juneau Co., Wisconsin 124 in the Chicago metropolitan area, Cook County, Illinois. They resumed migration on 18 April and proceeded to Dodge County, southeastern Wisconsin. On 19 April under overcast and a low ceiling, they resumed migration but proceeded northward and stopped to land twice. In mid-afternoon the sun appeared, and the flock abruptly changed course westward and then completed migration to Rynearson Pools (their rearing area of the previous year), Necedah NWR. The entire migration had taken the group 11 days, of which 7 were flight days. The route was roughly direct; distance covered per flight day varied from 150 to 383 km (mean = 274 km). Meanwhile, crane no. 7 stopped in northcentral Kentucky (exact location unknown) and Jasper County, northwestern Indiana, before landing at Avon Bottoms SWA, Rock County, southcentral Wisconsin, on 18 April. She remained at that location until 30 April when she moved to Crawford County in southwestern Wisconsin. She completed migration to Rynearson Pools, Necedah NWR, on 3 May. Spring Wandering The whooping cranes, after returning to Necedah NWR and like previously released experimental sandhill cranes led on fall migration by ultralight aircraft (Urbanek et al., 2005), moved to other locations in Wisconsin, generally south and east of Necedah NWR, during spring After return and 1 night of roosting on Necedah, all of the cranes left the following day. The group of 4 (nos. 1, 2, 5, and 6) moved to several sites south and southeast, spending the largest amounts of time 27 km south near Mauston, southern Juneau County (23 April-ca 5 May) and 174 km southeast near Cold Spring, Jefferson County (ca 10 May-1 June for 3 birds, until 8 June for crane no. 6) (Fig. 2). No. 6 had remained separate from the other birds after he sustained a minor leg injury ca 20 May. The group of 3 returned to Necedah NWR on 2 June, moved back to Mauston on 12 June, then returned to the refuge on 26 June. No. 6 returned to Necedah NWR on 9 June but did not associate with the other whooping cranes. When off refuge, the cranes typically inhabited agricultural lands, feeding on waste corn and roosting in wet areas or ephemeral pools in or near the fields. No. 7 may have spent 5-26 May in southern Wisconsin (exact location unknown, but she was tracked northward from this area), moved to Leola grasslands in Adams County on 27 May, then to Rush Lake, Winnebago/Fond du Lac Counties) ca 29 May, where she remained until ca 22 June. She then moved to Radke Pool, Horicon NWR, in southern Fond du Lac County, 131 km eastsoutheast of Necedah NWR (Fig. 2). Summer Home Range After his return on 9 June, no. 6 settled in the Rynearson Pools area of the refuge and remained there for the summer. The group of 3 also returned to the Rynearson Pools area, and after some interference with training of the current year s juveniles to follow ultralight aircraft, efforts by project personnel to frighten them away from the training site apparently resulted in separation of no. 5 from the group. He remained apart and by 10 July settled for the remainder of the summer about 10 km north at Sprague-Mather Pool on the northern part of the refuge.

16 218 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. Proc. North Am. Crane Workshop 9:2005 Fig. 2. Major spring and summer locations, fall staging areas, fall migration stopovers, and wintering areas of hatch-year 2001 whooping cranes, spring 2002-winter 2002/03. Sites described in text: (1) Mauston, (2) Briggsville, (3) Cold Spring, (4) Walworth, Kenosha, and McHenry Counties, (5) Resaca, (6) Concord, (7) Lake Butler. Nos. 1 and 2, a male and female that remained together, briefly left the refuge to the northwest but returned to Rynearson on 7 July and remained for the summer. All of the whooping cranes associated with sandhill cranes, and nos. 1, 2, and 6 consistently roosted with sandhill cranes from mid-july onward. No. 7 remained on Radke Pool, Horicon NWR, and adjacent areas after arriving there in late June.

17 Proc. North Am. Crane Workshop 9:2005 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. 219 Autumn Staging Nos. 1 and 2 left Rynearson Pools, Necedah NWR, on 8 October and returned to the same area west of Mauston which they had used in late June. That night they roosted in wetlands south of Castle Rock Lake just northeast of Mauston. During the next week they made several trips between the refuge and recently harvested cornfields, usually near Mauston. From 14 October they remained on Rynearson Pools, roosting with either sandhill cranes or near one of their former rearing sites, before returning to the Mauston cornfields on 1 November. Except for trips to a cornfield on 2 and 12 November, they remained on drawn down East Rynearson Pool, where they fed extensively on fish, mainly bullheads (Ictalurus sp.) trapped in the shallows. They migrated on 21 November. Unlike nos. 1 and 2, the 3 single yearling whooping cranes each became integral members of staging sandhill crane flocks: No. 5 left Necedah NWR on 7 October and joined a flock of sandhill cranes staging northeast of Mauston. That flock fed in local cornfields and roosted in wetlands south of Castle Rock Lake. On 2 or 3 November he joined the staging flock at Quincy Bluff, 11 km eastward in Adams County. On 4 November he moved 29 km southeast to a large staging area at Widow Green Marsh, near Briggsville, southwestern Marquette County, where he usually roosted in marsh along Neenah Creek or South Branch (Fig. 2). He migrated on 23 November. No. 6 also left Necedah NWR on 7 October and joined the staging sandhill cranes north of Briggsville. At that time the flock fed in local cornfields and roosted mainly in Widow Green Marsh. By 3 November he was foraging 6 km northeast of Widow Green Marsh and roosting in Endeavor Marsh 6 km east of that feeding area. He migrated on 9 or 10 November. No. 7 remained in the northern Horicon NWR area and by late September had joined large sandhill crane flocks that were roosting in Teal and Luehring Pools (just south of Radke Pool), Dodge County, and feeding in recently harvested cornfields east of the refuge. She migrated on 15 November. Autumn Migration Four of the 5 cranes followed the same general pattern (Fig. 2), i.e., a direct migration consisting of 6 consecutive flight days with 1-night stops at the major crane congregation areas of Jasper-Pulaski, Indiana, and Hiwassee, Tennessee, and 3 opportunistic stops (1 between Jasper-Pulaski and Hiwassee; 2 between Hiwassee and the Central Gulf Coast of Florida). No. 6 followed a different pattern; he spent several days on a staging area at the Wisconsin-Illinois border before passing through Jasper-Pulaski and then spending 1.5 months at Hiwassee. All whooping cranes apparently migrated with sandhills except during the final approach in Florida. Specific itineraries were as follows: Nos. 1 and 2 flew from Necedah NWR to Jasper-Pulaski on 21 November and left Jasper-Pulaski the next morning. The pair arrived at Hiwassee on 23 November, left the next morning, and made overnight stops near Concord, Georgia, and Lake Butler, Florida, before arriving at St. Martins Marsh Aquatic Preserve in early afternoon of 26 November. Crane No. 5 left the Briggsville staging area on 23 November and arrived at Hiwassee on 25 November. He left the next morning and arrived at the pen on Chassahowitzka NWR on 28 November. After leaving Endeavor Marsh on 9 or 10 November, no. 6 moved to a staging area that included parts of Kenosha and Walworth Counties, Wisconsin, and McHenry County, Illinois. He flew to Jasper-Pulaski on 16 November, left the next morning, and arrived at Hiwassee on 18 November. He remained there with wintering sandhill cranes until, in apparent response to depletion of corn on the refuge, he departed southbound with sandhill cranes on 3 January. He roosted that night in Gordon County, Georgia, and resumed migration the next morning. He arrived in the Hixtown Swamp area, Madison County, Florida, after dark on 4 January and joined the group containing whooping crane no. 7 and 50 sandhill cranes on 5 January. Crane No. 7 left Horicon NWR on 15 November and apparently arrived at the pensite on Chassahowitzka on 20 November (observed there the following morning). She remained a few days and then joined wintering sandhill cranes 217 km northnorthwest at Hixtown Swamp, Madison County, Florida (found there on 28 November). Second Winter Cranes Nos. 1 and 2: After arriving at St. Martins Marsh Aquatic Preserve on 26 November and then spending about 3 days exploring the 32 km of St. Martins/Chassahowitzka coastline, they settled on St. Martins Marsh just east of Ozello, 11 km north of the pen site. They roosted in the upper Greenleaf Bay area and typically spent daytimes foraging in openings between the Bay and palm hammock to the east. A favorite area was a large open patch of salt grass with no needlerush, little surface water, and surrounded by brush and palm hammock. They foraged by probing in mud in small wet areas. The pair left St. Martins Marsh on 14 or 15 December. Their departure occurred immediately after high tides which raised water levels in their roosting area in Greenleaf Bay on 13 December. On 17 December they were found inland on a private cattle ranch 63 km south near Land o Lakes, Pasco County, where they stayed for the remainder of the winter. Crane No. 5: After he arrived at the Chassahowitzka pen site on 28 November, the hatch-year 2002 flock of 16 juveniles, led by ultralight aircraft, completed their migration to the pen 2 days later. No. 5 joined the flock in the pen as the dominant bird. Like the juveniles, he usually roosted in the pen, ate from the feeder, and approached the costumed caretaker. The larger pen provided improved conditions at the release site in winter 2002/03, and no. 5 and the 16 new juveniles usually roosted of their own volition within the pen. The birds showed strong

18 220 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. Proc. North Am. Crane Workshop 9:2005 preference for roosting on the smooth, firm constructed oyster bar even when it was not or only partially covered with water. Salinities (ppt) near the pen site were generally lower (December [9-15], January [11-20], February [13-20], March [9-15]) than in the previous winter, but still high enough that cranes strongly preferred supplemental fresh water. In winter 2002/03 the cranes only ranged to 0.5 miles from the pen, the farthest points being south to E-Creek and Pumpkin Creek Impoundment. Unlike the previous winter, only a small burned area occurred just northeast of the pen. Crane No. 6: After joining the group containing no. 7 and 50 sandhill cranes at Hixtown Swamp on 5 January, no. 6 remained in the group through the morning of 7 January. By the afternoon of 9 January, he had moved to a different wetland in the same complex but 7 km west. The latter wetland was a major sandhill crane roosting area. No. 7 remained with wintering sandhill cranes in Plant Pond, Hixtown Swamp area. She used a small area and apparently did almost all foraging in the wetland. All whooping cranes remained sedentary on their final selected wintering areas. This behavior was typical of wild migratory sandhill cranes (Urbanek et al. 1988) and whooping cranes (Stehn 1991). DISCUSSION Survival and Management Strategy to Avoid Predation In winter 2001/02, 2 of 7 juvenile whooping cranes were killed by bobcats within 1.5 months of release on the wintering site. Before these mortalities, birds were usually allowed to roost at locations of their own choosing. After the second mortality, security was increased by making sure that the cranes were in the pen or in water at least 6 m from shore at roosting time. This management strategy was successful, and no additional mortalities occurred at the winter release site. Cranes returning to Florida during the following winter had in the interim learned sufficient survival skills to successfully cope with the high threat of predation by bobcats. Habitat Limitations on the Winter Release Area Whooping crane juveniles demonstrated inconsistent water roosting behavior at Chassahowitzka NWR. The main impediment to safe roosting on Chassahowitzka NWR was that consistently usable roosting habitat was not available. Tidal variation both during the night and from night to night often made roosting in water at any single location impossible. The tides in conjunction with poor natural substrate, i.e., jagged oyster rock or extremely soft muck, encouraged whooping cranes at the pensite to often roost on land in winter 2001/02 and on the smooth, firm, artificially constructed oyster bar in winter 2002/03. One reason for inconsistent water roosting may have been that the birds had not been conditioned to consistently roost in water and a safe distance from shore. In previous studies with sandhill cranes (Urbanek and Bookhout 1992; Urbanek et al. 2005), a costumed dummy was successfully used to attract and hold released juveniles at a desired roost site. In winter 2001/02, however, released whooping cranes, unlike sandhill cranes, showed no attraction to a costumed dummy. In summer 2002, chicks being reared at Necedah were provided with water roosting opportunity in their rearing pens and were also given fulltime exposure to a plastic whooping crane decoy. In winter 2002/03 decoys were placed at the end of the newly constructed oyster bar and at the roosting area in the old part of the pen. Some cranes, mainly the same few individuals, showed attraction to the decoy, but the attraction was weak and insufficient to influence roosting behavior of most of the birds. Most whooping cranes did not remain on the Central Gulf Coast during their second winter. Not only the tidal fluctuations, but salinity, unstable or rocky bottom substrates, and general habitat dominance by needlerush also contributed to poor habitat conditions. However, wintering of the population in coastal wetlands, although desirable, is not biologically necessary for success of this reintroduction. Attaining the necessary goals of survival, reproduction, and human avoidance can also be achieved by other wintering strategies. Although Chassahowitzka NWR appeared unsuitable as an ultimate wintering area for reintroduced whooping cranes, it did prove to be an excellent release site. Cranes over-wintering at the pensite could be successfully protected from predators and effectively isolated from human activity, the latter of which could compromise their wildness during the critical period just after release. Occupation of the Chassahowitzka site during their first winter also had no adverse effect on selection of habitat by cranes in their subsequent winter. The returning cranes selected appropriate wintering habitat inland. [Subsequent results after this paper was written indicate an important additional benefit of using Chassahowitzka NWR as a release site. Namely, older cranes completing fall migration have returned to the Chassahowitzka release site, but then, because of the limitations of the habitat, they remained at most a few weeks before moving to nearby quality habitat inland. This pattern allows the same well-built and protected pensite on Chassahowitzka NWR to be used again with minimized interference by older birds harassing the newly arrived juveniles. Domination of the pen site by older birds could be detrimental to this reintroduction by forcing subordinate juveniles outside of the pen where they are susceptible to predation by bobcats. Using the Chassahowitzka pensite as the winter release area for naive juveniles greatly facilitates the strategy of protecting the cranes to ensure survival during their highly vulnerable first winter. After initiating spring migration, they then have 8 months in areas containing few or no bobcats in which to develop the survival skills they will need before returning again to winter in Florida.]

19 Proc. North Am. Crane Workshop 9:2005 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. 221 Spring Migration and Subsequent Spring Wandering These whooping cranes migrated directly back to Wisconsin during their first spring migration. They did not necessarily follow the exact route used during fall migration while following ultralight aircraft. They flew on each day with favorable winds and clear to partly cloudy skies and also on a few days with less than optimal migration conditions. Their migration stops were opportunistic, generally consisting of whatever ponds or other wetlands were present in their flight path at the end of each migration day. This pattern was similar to that of whooping cranes in the natural Aransas-Wood Buffalo population (Howe 1989). After returning to Necedah NWR, all of the yearlings embarked on a spring wandering period that lasted through May and June. This wandering was characterized by frequent and/ or extensive flights that familiarized the cranes with their reintroduction area. Spring wandering, especially of females, may be characteristic of migrating cranes and has been previously noted in sandhill cranes (Urbanek 1990, Urbanek et al. 2005). However, to date this phenomenon has been poorly studied. Roosting Behavior of Yearlings Inconsistent roosting in water or roosting in water too near shoreline was sometimes evident. During spring and early summer these yearling whooping cranes sometimes chose safe roosting sites, but at other times they roosted on land, in small wetlands near shore, or in farm fields with a very limited extent of standing water. This roost site selection may have been related to lack of appropriate conditioning to roost in water at Chassahowitzka NWR during the first winter or at Necedah NWR during the previous summer. During their yearling summer, however, the whooping cranes began associating with wild sandhill cranes. From that time on, they consistently selected safe roosting habitat. Human Avoidance Costume/isolation-reared cranes are sometimes prone to tolerate presence of humans after release. This behavioral corruption can be largely avoided if costume-rearing is done according to rigorous standards and the birds remain isolated from uncostumed humans during the release process and subsequent period of adjustment to the wild. The latter period may be long and of undetermined length depending on the history and psychological disposition of each bird. Although the released whooping cranes were never attracted to uncostumed humans, they occasionally demonstrated inadequate fear of humans and vehicles, especially while on farmlands in spring and early summer. However, after returning to the refuge and/or joining wild sandhill cranes in summer, wildness increased and no significant problem was apparent. Cranes Nos. 1 and 2 demonstrated some attraction to sites subject to limited human activity, especially when food was present, e.g., a duck-banding site baited with corn on Necedah NWR. This opportunity for exposure to humans at the reintroduction site can be effectively or at least partially controlled. More difficult to deter is exposure to human activity on some of the wintering grounds. Nos. 1 and 2, for example, spent their second winter near a human residence on a private cattle ranch that contained non-migratory sandhill cranes that were tolerant of people. That site did little, therefore, to reinforce human avoidance behavior. In addition, development pressures in Pasco County and other areas in west-central Florida could compromise some wildness of these birds as well as result in significant habitat loss. CONCLUSIONS AND MANAGEMENT IMPLICATIONS Leading by Ultralight Aircraft This technique has been successful in leading these and other birds to winter on a protected release area during their first winter (Ellis 2003). A migration route was learned that effectively resulted in released whooping cranes returning to the reintroduction area in Central Wisconsin. When combined with a rigorous costume/isolation-rearing protocol, leading cranes behind ultralight aircraft appears to be a highly effective reintroduction technique for a migratory population. Winter Release Procedure The management strategy of protecting the released birds during their first winter at Chassahowitzka NWR has been highly successful. When employed, this strategy has resulted in 100% survival of naive, newly released birds while they occupied areas with high bobcat densities. Once birds leave on spring migration, they are not subject to high densities of this predator (Anderson and Lovallo 2003; R. P. Urbanek, personal observation). By their return to Florida the following winter, they have adapted to the wild and become more predator wary. Association with sandhill cranes on the summering areas seemingly facilitates this process. Chassahowitzka NWR and the adjacent Gulf Coast of Florida appear to be unsuitable for maintaining a wintering population of whooping cranes. Most of the whooping cranes returned to the salt marsh after the second southward migration and then left shortly thereafter to inhabit nearby areas inland. The most preferred habitat appears to be large cattle ranches containing extensive, large, shallow ponds (described by Nesbitt et al. 1997). Some whooping cranes winter with migratory sandhills and use the same habitats (Urbanek 1988). However, use of a release site for juveniles that is separate from the wintering area of older birds has been highly advantageous by allowing use of the same isolated, well-constructed, and protective pensite for multiple years. This winter release strategy minimizes the problem of dominant older birds interfering with the feeding by the

20 222 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. Proc. North Am. Crane Workshop 9:2005 juveniles and driving them into unprotected roosting habitat. Management during Summer and Migration Once the birds began migration from the winter release site, this cohort required no further assistance from monitoring personnel to ensure survival or improve behavior. Concerns involving exposure to human activity occurred mainly during spring wandering and were largely resolved after the birds returned to the refuge and associated with sandhill cranes. Some interference with training of the new juveniles by the yearlings occurred on the refuge training sites. However, the latter required minimal intervention (i.e., frightening the older birds from sites). Evaluation of Reintroduction after the First Year Survival of the migratory whooping crane flock was 100% after additional measures to protect juveniles from predators at the winter release pen were implemented. The initial cohort completed their unassisted first migration cycle from Florida to Wisconsin and back to Florida with no mortality. Development of adequate foraging, roosting, social association, and human avoidance behaviors raises the probability that a self-sustaining population can be established by these methods. Three requisites are necessary for a reintroduction to be successful: (1) Suitable reintroduction area (particularly the breeding area in a migratory population). (2) Effective reintroduction techniques. (3) Experienced, skilled, and talented personnel who can implement these techniques and understand how to effect the transition from captive-reared to wild birds. These three determinants of success are met in the eastern migratory whooping crane reintroduction. ACKNOWLEDGMENTS This paper is a product of the Whooping Crane Eastern Partnership (WCEP), which was established in 1999 to reintroduce a migratory population of whooping cranes to eastern North America. The nine founding members are the Canada- U.S. Whooping Crane Recovery Team, U.S. Fish and Wildlife Service (USFWS), USGS Patuxent Wildlife Research Center, USGS National Wildlife Health Center, Wisconsin Department of Natural Resources (DNR), Operation Migration, Inc., International Crane Foundation (ICF), National Fish and Wildlife Foundation, and Natural Resources Foundation of Wisconsin. Many additional organizations and individuals have played an important role in the reintroduction, and the efforts of all participants are acknowledged as vital to its success. We especially thank the following individuals, most of whom worked directly with the birds: Chick rearing and training to follow ultralight aircraft: D. Sprague, C. Caldwell, K. Candelora, G. Olsen, G. Gee, J. Chandler, K. O Malley, and D. Ellis (Patuxent) and J. Duff and D. Clark (Operation Migration). Ultralight aircraft-led migration: J. Duff, D. Clark, R. Van Heuvelen, W. Lishman, R. Cohen-Pardo, G. Lee, and D. and P. Lounsbury (Operation Migration), D. Sprague and D. Ossi (Patuxent), and K. Maguire (ICF). We are especially grateful to L. Wargowsky and staff (Necedah NWR), J. Kraus and staff (Chassahowitzka NWR), and S. Blitch and staff (St. Martins Marsh Aquatic Preserve Complex), and for their support and provision of facilities. We thank the many additional cooperators and state personnel who assisted in monitoring, especially J. Bergens (Jasper-Pulaski), and J. W. Akins (Hiwassee). T. Kohler (Windway Capital Corporation) graciously provided tracking aircraft and financial support. J. Harris (ICF) and J. Christian (USFWS) were instrumental in facilitating logistic support. LITERATURE CITED Allen, R. P The whooping crane. National Audubon Society Research Report 3. Anderson, E. M., and M. J. Lovallo. Bobcat and lynx. Pages in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, eds. Wild mammals of North America: biology, management, and conservation. Second edition. John Hopkins University Press, Baltimore, Maryland, USA. Duff, J. W., W. A. Lishman, D. A. Clark, G. F. Gee, and D. H. Ellis Results of the first ultralight-led sandhill crane migration in eastern North America. Proceedings North American Crane Workshop 8: Ellis, D. H., G. F. Gee, S. G. Hereford, G. H. Olsen, T. D. Chisolm, J. M. Nicolich, K. A. Sullivan, N. J. Thomas, M. Nagendran, and J. S. Hatfield. Post-release survival of hand-reared and parent-reared Mississippi sandhill cranes. Condor 102: ,, K. R. Clegg, J. W. Duff, W. A. Lishman, and W. J. L. Sladen Lessons from motorized migrations. Proceedings North American Crane Workshop 8: , W. J. L. Sladen, W. A. Lishman, K. R. Clegg, J. W. Duff, G. F. Gee, and J. C. Lewis Motorized migrations: the future or mere fantasy? Bioscience 53(3): Horwich, R. H Use of surrogate parental models and age periods in a successful release of hand-reared sandhill cranes. Zoo Biology 8: Developing a migratory whooping crane flock. Proceedings of the North American Crane Workshop 8: Howe, M. A Migration of radio-marked whooping cranes from the Aransas-Wood Buffalo population: patterns of habitat use, behavior, and survival. U.S. Fish and Wildlife Service, Fish and Wildlife Technical Report 21. Hunt, H. E The effects of burning and grazing on habitat use by whooping cranes and sandhills cranes on the Aransas National Wildlife Refuge, Texas. Dissertation,

21 Proc. North Am. Crane Workshop 9:2005 MIGRATORY WHOOPING CRANE REINTRODUCTION Urbanek et al. 223 Texas A&M University, College Station, Texas, USA. Lishman, W. A., T. L. Teets, J. W. Duff, W. J. L. Sladen, G. G. Shire, K. M. Goolsby, W. A. Bezner Kerr, and R.P. Urbanek A reintroduction technique for migratory birds: leading Canada geese and isolation-reared sandhill cranes with ultralight aircraft. Proceedings North American Crane Workshop 7: McMillen, J. L Productivity and movements of the greater sandhill crane population at Seney National Wildlife Refuge: potential for an introduction of whooping cranes. Dissertation, Ohio State University, Columbus, Ohio, USA. Nesbitt, S. A., M. J. Folk, M. G. Spalding, J. A. Schmidt, S. T. Schwikert, J. M. Nicolich, M. Wellington, J. C. Lewis, and T. H. Logan An experimental release of whooping cranes in Florida-the first three years. Proceedings North American Crane Workshop 7:79-85., M. J. Folk, K. A. Sullivan, S. T. Schwikert, and M. G. Spalding An update of the Florida whooping crane release project through June Proceedings North American Crane Workshop 8: Stehn, T. V Unusual movements and behaviors of color-banded whooping cranes during winter. Proceedings North American Crane Workshop 6: Toepfer, J. E., and R. A. Crete Migration of radio-tagged greater sandhill cranes from Minnesota and Wisconsin. Pages in J. C. Lewis, ed. Proceedings of the 1978 crane workshop. Colorado State University Printing Service, Fort Collins, Colorado, USA. Trick, J Final environmental assessment: proposed rein- troduction of a migratory flock of whooping cranes in the eastern United States. U.S. Fish and Wildlife Service, Green Bay, Wisconsin, USA. Urbanek, R. P Migration of sandhill cranes from the north shore of the north channel of Lake Huron, Ontario. Ohio Cooperative Fish and Wildlife Research Unit, Columbus, Ohio, USA Behavior and survival of captive-reared juvenile sandhill cranes introduced by gentle release into a migratory flock of sandhill cranes. Ohio Cooperative Fish and Wildlife Research Unit, Columbus, Ohio, USA., and T. A. Bookhout Development of an isolation-rearing/gentle release procedure for reintroducing migratory cranes. Proceedings North American Crane Workshop 6: , and Performance of captive-reared cranes released into a migration route in eastern North America. Pages in H. Higuchi and J. Minton, eds. The future of cranes and wetlands. Wild Bird Society of Japan, Tokyo, Japan., J. W. Duff, S. R. Swengel, and L. E. A. Fondow Reintroduction techniques: post-release performance of sandhill cranes (1) released into wild flocks and (2) led on migration by ultralight aircraft. Proceedings North American Crane Workshop 9: , S. A. Nesbitt, J. L. McMillen, and T. A. Bookhout Winter distribution and site fidelity of greater sandhill cranes from Upper Michigan. Ohio Cooperative Fish and Wildlife Research Unit, Columbus, Ohio, USA. Walkinshaw, L. H Sandhill crane studies in Michigan s Upper Peninsula. Jack-Pine Warbler 56:

22 224 Proc. North Am. Crane Workshop 9:2005

23 NESTING ECOLOGY AND PRODUCTIVITY OF THE CUBAN SANDHILL CRANE ON THE ISLE OF YOUTH, CUBA XIOMARA GALVEZ AGUILERA, Empresa Nacional para la Protección de la Flora y la Fauna, Ministerio Nacional de Agricultura, La Habana, Cuba. VICENTE BEROVIDES ALVAREZ, Facultad de Biología, Universidad de La Habana, Habana, Cuba FELIPE CHAVEZ-RAMIREZ, Platte River Whooping Crane Maintenance Trust, Inc., 6611 W. Whooping Crane Drive, Wood River, NE 68883, USA Abstract: We studied nesting ecology and productivity of the Cuban sandhill crane in the Isle of Youth Cuba between The nesting season extended from late march through June, but due to variable weather conditions could begin in February or extend through July. Overall, 71.1% of nests located produced eggs, while 84.1% of nests with eggs hatched at least one chick. Mean clutch size was 1.72 eggs. There were significant differences in hatching rates (G = 19.05, P < 0.01) and successful nests between years (G = 9.59, P < 0.10). Chicks/successful nest and rainfall during the breeding period (r = 0.6) were positively correlated. Percent successful nests was negatively correlated with total rainfall during the breeding season (r = -0.50). Causes of egg or nest failure included nest abandonment, predation, flooding, and infertility. In regards to breeding biology we recorded feather painting prior to nesting activity, false nests, and a re-nesting attempt after nest destruction. All nests were built on dry land. Cranes selected four of 11 habitat types present in the study area for nesting: open savannah (SNA) (24.2%), semi-closed natural savannah (SSC) (50.67%), open pine woodland (SPPA) (17.7%), and secondary savannah (SS) (7.7%). Palm density, seedlings, and forbs were lower at nest sites compared to random points, while ground cover of sand and litter was greater at nest sites. There were no significant differences in frequency of plant species among nest sites and random points (G = 3.78, P > 0.05). Tree species richness was less at nest sites, likely due to dominance of Tabebuia lepiodphylla and Byrsonima crassifolia species. Significant differences (G = 15.8, P < 0.01) were found in frequency of palm species, with greater density of Colpothrinax wrightii and Coccothrinax miraguama at nests sites. Shrub density was significantly higher (G = , P < 0.001) at nest sites being dominated by T. lepidophylla. There were no significant differences among nest sites and random points in forb variables (G = 6.67, P > 0.05) with similar frequency at both sites. PROCEEDINGS NORTH AMERICAN CRANE WORKSHOP 9: Key words: Cuba, Sandhill crane, Grus canadensis nesiotes, breeding, productivity, nesting habitat use, Isle of Youth Causes for the population decline of crane species worldwide are varied but habitat destruction is a primary cause. In general, threats faced by Cuban sandhill cranes (Grus canadensis nesiotes) are no different than those occurring elsewhere in the world (Harris 1994). Many freshwater wetlands and swamps have been drained and have been intensely modified by agricultural activity. Dams and deforestation that surrounded low areas have changed the balance and quality of water. Many factors are known to have negative impacts on cranes and they have been well known for decades, however, cranes still continue to be threatened (Meine and Archibald 1996a, Bishop 1988, Nesbitt and William 1990, Cruz 1996, Galvez and Pereira 1995). A management priority for endangered species is to increase population numbers and protecting habitat elements (i.e. breeding areas) is critical (Valentine 1981). For little known species it is important to gather biological and ecological information in order to understand population dynamics and resource requirements. Understanding species-specific habitat requirements can lead to more efficient recovery efforts and more relevant conservation measures to be implemented (Dietzman and Swengel 1994). Restoration of species to their original ranges and population increases are essential steps for the recovery of many threatened and endangered species (Nesbitt and Carpenter 1993). The Cuban sandhill crane is an endemic subspecies to Cuba. It is the largest extant bird in Cuba and the Caribbean and the only crane present in the Caribbean or South American continent. The Cuban Sandhill crane was considered critically endangered at the beginning of this project by IUCN (Meine and Archibald 1996b) due to what were believed to be severely fragmented populations with less than 50 individuals. While the Cuban sandhill crane has been known to science since the mid 1800 s (Gundlach 1875), little has been published in regards to it s biology or specific ecological requirements including breeding habitat, nest site characteristics, or nesting behavior. Walkinshaw (1953), presented the only available information on the breeding biology of this subspecies. The objective of our study was to gather information on breeding ecology and nesting characteristics of the Cuban sandhill crane on the Isle of Youth, Cuba. Specifically, in this work we present information on: a) breeding phenology, nesting success, and estimates of productivity; b) define nest site characteristics and nesting habitat type selection; and c) report on several unique or interesting observations on different aspects of breeding biology and behavior. 225

24 226 CUBAN SANDHILL CRANE Galvez-Aguilera et al. Proc. North Am. Crane Workshop 9:2005 METHODS AND MATERIALS Study Area Our study area was the Los Indios Ecological Reserve and surrounding areas on the Isle of Youth, Cuba. Although field work was conducted outside the reserve boundary, it was limited to the watershed surrounding the reserve (Fig. 1). The Isle of Youth, formerly known as the Isle of Pines, is the largest of Cuba s outlying islands south of western mainland Cuba. The Isle of Youth has a surface area of 2,397 km 2. The dominant vegetation in the reserve and surrounding areas are several varieties of open savannahs and forests. Trees in the savannahs are dominated by bottle palm (Byrsinoma crassifolia) and two pines (Pinus caribea and P. tropicalis). Most savannahs contain a number of different shrub species (Hypericum styphelioides and Tabebuia lepidophylla) and palmettos (Acoelorraphe wrightii) with sparse grass ground cover. The savannah landscape is crisscrossed with riparian vegetation corridors along creeks and arroyos dominated by Royal palm (Roystonea regia), majagua (Hibiscus tiliaceus), and baga (Annona glabra). Several dams on the northern portion of the island control flows in some streams. Along the coastline, on the western portion of our study area, the vegetation is primarily a mangrove (Rhizofora spp., Avicenia spp.) forest. Because the island has been modified extensively in the past for agriculture development, many active and abandoned orchards are present, mostly of grapefruit and coconuts. Because of poor and sandy soils other agriculture crops are not very successful and old abandoned fields are present throughout the area. Some portions of the savannahs have been cleared of woody plants to create pastures for cattle grazing. Nesting and Productivity Measurements Nests of Cuban sandhill cranes were located within the Los Indios watershed by walking systematic transects throughout the area during periods of nest building, egg laying, and incubation. Nest searches were conducted annually from from mid April through July in most years. In 1997, we were only able to monitor 50% of nests found, however, from 1998 through 2002 we monitored 98% of all nests found each year. Havana Cuba Isle of Youth Los Indios watershed boundary Los Indios Ecological Reserve Fig. 1. Location of the Los Indios Ecological Reserve and the Los Indios Watershed boundary on the Isle of Youth, Cuba.