SOUTHEASTERN NATURALIST Gulf of Mexico Natural History and Oil Spill Impacts Special Series 2012 11(1):G1–G16 Short-Term Response of Brown Pelicans (Pelecanus occidentalis) to Oil Spill Rehabilitation and Translocation Will Selman1,*, Thomas J. Hess, Jr.1, Brac Salyers1, and Carrie Salyers1 Abstract - Translocation of wildlife is a valuable tool for managers to alleviate impacts of human/wildlife conflicts, habitat fragmentation, and small population sizes. Pelecanus occidentalis (Brown Pelican) were previously successfully translocated into Louisiana in 1960–1980s and had rebounded significantly prior to the Deepwater Horizon Oil Spill in April 2010. The Deepwater Horizon Oil Spill had a dramatic impact on southeastern Louisiana, including the people, wildlife, and coastal habitats of the region. We translocated 182 oil-rehabilitated Brown Pelicans from southeastern Louisiana to an area non-impacted by the oil spill in southwestern Louisiana (Rabbit Island, Cameron Parish). Daily surveys were conducted at the island for six weeks and documented mortality, movements, integration with local pelican flocks, persistence of pelicans at the island, and the role of supplemental feeding. We documented no mortality of rehabilitated birds and found that translocated Brown Pelicans readily integrated with local pelican flocks. Supplemental feeding likely contributed to the persistence of pelicans at Rabbit Island from weeks 1 to 4. By weeks 4 to 6, many local and translocated pelicans moved away from the island, likely due to a combination of natural and human-induced factors. After the initial stages of the translocation, we tentatively suggest that the program was a success and propose recommendations for future translocation attempts with Brown Pelicans. Introduction Translocation, or the moving of living organisms from one locality to another, has been widely used by wildlife managers to supplement wildlife populations, re-establish populations, or to establish new populations (Armstrong and Seddon 2008, IUCN 1987). Translocation techniques are becoming increasingly important with rare, threatened, or endangered species that have been extirpated from all or part of their native range (Griffith et al. 1989, Kleiman 1989). However, to determine the effectiveness of a wildlife translocation project, it is critical to quantify the short- and long-term success or failure of the project. Yet, many projects fail to adequately quantify the success or failure of a translocation effort and/or document alternate strategies that may have improved translocation success (Griffith et al. 1989). Consequently, the outcome of many translocation projects performed in the past remains unknown (Dodd and Seigel 1991, Fischer and Lindenmayer 2000). During the 1940s, Pelecanus occidentalis L. (Brown Pelican) were considered a “common bird” to Louisiana, but their populations were “decreasing in numbers” (LDWF 1941). By the early 1960s, Brown Pelicans had ceased 1Rockefeller Refuge, Louisiana Department of Wildlife and Fisheries, 5476 Grand Chenier Highway, Grand Chenier, LA 70643. *Corresponding author - wselman@wlf. la.gov. G2 Southeastern Naturalist Vol. 11, No. 1 nesting in Louisiana (Williams and Martin 1968), with environmental pesticide contamination and subsequent eggshell thinning as a plausible link to declines (Blus et al. 1979). Following the banning of many toxic pesticides (i.e., endrin, DDT), translocation efforts were conceived by Louisiana Department of Wildlife and Fisheries (LDWF) and the Florida Game and Fresh Water Fish Commission to restore Brown Pelicans to Louisiana (Nesbitt et al. 1978) to recipient sites that previously had nesting colonies of the species (i.e., barrier islands of southeastern Louisiana). Between 1968 and 1980, 1276 Brown Pelicans were translocated from Florida to Louisiana (McNease et al. 1992), and these translocations were considered extremely successful (Holm et al. 2003, McNease et al. 1992). Prior to the Deepwater Horizon oil spill off the southeastern coast of Louisiana, these barrier islands had 9 active Brown Pelican nesting colonies, with approximately 10,114 nesting pairs (Hess and Linscombe 2010). Coastal land loss is a major concern to existing pelican-nesting barrier islands in southeastern Louisiana (Holm et al. 2003; S. Walter et al., University of Louisiana-Lafayette, Lafayette, LA, unpubl. data). Further, many of the barrier islands that held the largest nesting colonies of Brown Pelicans, and the islands that were the focus of translocation efforts between the 1960s and 1980s, were some of the most heavily impacted areas following the Deepwater Horizon Oil Spill (LDWF 2010, NOAA Fisheries Service 2010). Ironically, the Brown Pelican was removed from the Endangered Species List by the US Fish and Wildlife Service (USFWS) five months prior to the Deepwater Horizon Oil Spill, primarily due to the recoveries that have been achieved in Louisiana and the northern Gulf of Mexico region (USFWS 2009). The Deepwater Horizon explosion and oil spill was the largest offshore oil spill in US history, releasing an estimated 4.4 million barrels of oil (7.0 x 105 m3 ± 20%; Crone and Tolstoy 2010) into the Gulf of Mexico. Consequently, this had a dramatic impact on the coastal habitat/wildlife of southeastern Louisiana and neighboring Gulf of Mexico states (Restore the Gulf 2010). The response by local, state, federal, non-profit, and academic entities to the oil spill occurred on a massive scale, with many performing wildlife surveys, both on and off shore. During these surveys, many oiled wildlife specimens were found both dead and alive (i.e., birds, sea turtles, marine mammals; Restore the Gulf 2010). Individuals that were found alive were taken to wildlife habilitation facilities, with the ultimate goal of eventually releasing them back into their native habitats. However, many of these habitats in the southeastern part of the state remained impacted for prolonged periods (LDWF 2010, NOAA Fisheries Service 2010) and thus, successfully rehabilitated individuals could not be released back into their original sites of collection. These heavily impacted areas included many of the barrier island chains of southeastern Louisiana utilized by Brown Pelicans for nesting colonies (Hess and Linscombe 2004). In contrast to southeastern Louisiana, the southwestern part of the state was relatively unimpacted by the Deepwater Horizon Oil Spill (LDWF 2010, NOAA Fisheries Service 2010). In southwestern Louisiana, Rabbit Island serves as the only active nesting colony for Brown Pelicans. In 2003, a small nesting colony was discovered on Rabbit Island (4 nests, 7 fledglings; Hess 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G3 and Linscombe 2003), a small shell island in West Calcasieu Lake (≈89 ha [220 acres]; 29º50.882'N, 93º23.014'W). Since the discovery of the Rabbit Island nesting colony, the local Brown Pelican population has increased substantially, with approximately 1000 fledglings per year and a local adult population of at least 3000–5000 individuals (T.J. Hess, Jr., pers. observ.). We determined that Rabbit Island would make a suitable candidate translocation site for oil-spillrehabilitated Brown Pelicans for multiple reasons: 1) Rabbit Island is currently a suitable nesting habitat for local Brown Pelicans, 2) this area was not impacted by the Deepwater Horizon Oil Spill, 3) the resident Brown Pelican population continues to increase in size and only utilizes parts of the island for nesting, 4) there are bountiful foraging opportunities in the region (i.e., Calcasieu Lake and Gulf of Mexico), and 5) prior attempts to translocate Brown Pelicans in Louisiana have been successful. Further, we also presumed that the native population would not suffer any ill effects or genetic “swamping” following supplemental translocations of oil-rehabilitated Brown Pelicans. The latter was not an issue as almost all of the Louisiana Brown Pelicans are derived from previously translocated Florida stock from 1968–1980 (McNease et al. 1984, Nesbitt et al. 1978). Prior to translocation, we proposed to document the efficacy of the translocation effort of oil-rehabilitated Brown Pelicans from native southeastern Louisiana marshes to the existing Rabbit Island colony in southwestern Louisiana. Specifi- cally, we selected several a priori research questions with which we could gauge the success/failure of the translocation: 1) would we document any mortality in translocated birds?, 2) how long would translocated and local Brown Pelicans remain at Rabbit Island?, 3) would translocated Brown Pelicans integrate into native pelican groups?, 4) would supplemental feeding of translocated birds improve retention?, and 5) would we be able to adequately monitor short-term movements of translocated birds? Methods Translocation Out of the 719 total oil-recovered Brown Pelicans from southeastern Louisiana, 182 Brown Pelicans (166 hatch year, 14 after hatch year, 2 unknown age) were rehabilitated and translocated to Rabbit Island (Fig. 1); pelicans were translocated to other states prior to this translocation due to fear of the oil spill impacting the whole coast of Louisiana. Prior to leaving the rehabilitation facility, metal bands and colored auxiliary bands (red or pink bands with large alphanumeric codes) were affixed to the legs of translocated Brown Pelicans. The birds were transported from the rehabilitation facilities to Cameron, LA (Cameron Parish) in six different translocation events (Table 1). Pelicans were transported in an air-conditioned covered trailer and in pet carriers (maximum 2 pelicans per carrier). They were delivered early in the morning to avoid any heat stress that may have occurred during the middle of the day. Upon arrival in Cameron, each individual was inspected by a veterinarian to ensure a positive health status prior to being released by LDWF staff at the northern end of Rabbit G4 Southeastern Naturalist Vol. 11, No. 1 Figure 1. Recovery locations of 183 oiled Brown Pelicans (grey circles) that were rehabilitated and translocated to Rabbit Island (inset). Sixty-eight Brown Pelicans were recovered from Raccoon Island and 67 in the Barataria Bay region. Table 1. Brown Pelican translocation dates to Rabbit Island (Cameron Parish, LA), number of pelicans released, and colored leg-band codes. Translocation # of pelicans date released Colored-band numbers 5 August 2010 15 Pink (A55–A69) 6 August 2010 59 Pink (A70–A84), Red (0Z5–7Z1, except 4) 20 August 2010 33 Pink (A85–A00; C01–C04, C06–C14), Red (2Z1, 6Z4, 6Z8, 7Z1) 28 August 2010 48 Pink (C05, C15–C61) 3 September 2010 19 Pink (C62–C80) 10 September 2010 8 Pink (C81–C88) 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G5 Island. We chose the northern end of the island as it had fewer resident pelican nests, and subsequent hatch-year birds already present, relative to the southern end of the island. Rabbit Island counts Following the initial translocation event (5 and 6 August 2010), we made daily counts for roosting Brown Pelicans on Rabbit Island for 6 weeks (from 7 August 2010 to 18 September 2010). Thirty-seven daily counts were completed, and we did not complete surveys on five days due to severe weather or on days of additional pelican releases. Surveys started at 0715 hrs and lasted approximately 1.5 hours. During each survey, we counted both auxiliary red/ pink-banded pelicans (hereafter “banded”) and unbanded pelicans from the deck of a push barge while it moved slowly around the island, approximately 100–150 m from shore. Other band colors were observed (i.e., yellow, green, white), but these individuals were adults, and therefore, we are confident that those birds observed with red or pink bands were the part of the translocation group of pelicans. The island was divided into a northern and southern section, and pelicans were counted individually using a 20–60x spotting scope (Leica Televid 77) with tripod and/or 12 x 36-mm image-stabilizing binoculars (Canon). We also opportunistically took photographs (Nikon D90) of banded birds in transit between the boat ramp and Rabbit Island, during counts, or during supplemental feedings (discussed further below). Following the 18 September survey, we continued weekly monitoring at Rabbit Island and the surrounding region (i.e., Calcasieu Ship Channel, Cameron Jetties) until 29 October 2010 (6 additional surveys); weekly data will not be used for statistical comparisons but will be discussed. Supplemental feeding Following the first translocation, we provided supplemental feedings of dead fish to translocated Brown Pelicans. This feeding was done in an attempt to increase retention rates of translocated birds (i.e., prevent long-distance movements back toward the oil-spill impacted area) and is similar in concept to other supplemental feedings in previous translocation events with Brown Pelicans (Joanen and McNease 1974, Nesbitt et al. 1978). Frozen Brevoortia patronus Goode (Gulf Menhaden; ≈15–25 cm total length), a locally common fish species and a staple in Brown Pelican diets (Franklin 2007), were thawed to ambient temperatures before feeding to pelicans. Following all visual surveys, feeding was completed by tossing fish into the water from a motorized boat; researchers opportunistically took photographs of banded pelicans during this time as they would be feeding near the boat. Supplemental feeding occurred for 31 days following the initial release of the first pelican group. The schedule for pelican feeding was twice a day (morning and evening; Joanen and McNease 1974) and 133 kg (500 lb) of menhaden per feeding. During the study, habituation to the feeding and to the presence of the feeding boat occurred with both translocated G6 Southeastern Naturalist Vol. 11, No. 1 and local birds. Once it was clear that such habituation was occuring, supplemental feeding was suspended at the recommendation of the authors. Thus, there was no supplemental feeding that occurred for the last 9 survey days of the study (weeks 5 and 6). Statistical analysis Since the number of pelicans counted during the daily counts deviated from normality, we used a Kruskal-Wallis test to determine if daily counts were equal across sample weeks. Also, we standardized the number of banded pelicans observed to percent of banded pelicans observed; this standardization was done due to the changing number of banded pelicans available in our population after every release event. Similarly, the daily percentage of banded pelicans observed was non-normal, and therefore, we used a Kruskal-Wallis test to calculate if percentages were equal across sample weeks. Lastly, we used a Kruskal-Wallis test to determine if the number of banded pelicans relative to the total pelicans observed (in percent) were equal across weeks; this was done in order to determine if a larger number of the banded pelicans remained on the island relative to the local pelicans. To determine the effect of supplemental feeding on the pelicans, we used a Wilcoxon rank sums test (with chi-squared approximation) to determine if 1) pelican counts and 2) percentage of banded pelicans observed were equal during and after supplemental feeding. Results Mortality Throughout our surveys, we did not document any mortality of the 182 translocated pelicans, but we did notice several fresh carcasses of non-translocated Brown Pelicans at Rabbit Island. We also did not receive any calls from the public on Brown Pelican carcasses with bands in the area. Persistence of pelicans at Rabbit Island The mean number of Brown Pelicans observed on Rabbit Island was 584.6 individuals (SD: ± 422, range: 0–1762), which were primarily hatch-year birds. Survey counts were significantly different across the six weeks of observations (χ2= 27.49, df = 5, P < 0.001; Fig. 2), with lower counts occurring in later survey weeks relative to higher counts during earlier surveys. Throughout the survey effort, we also positively identified 48 of the 182 individuals (26.3% of total released; Table 2) via photograph or spotting scope from all six release groups. Of the 48 individuals, 17 were identified on two different surveys, 7 were identified three different times, and 4 were identified on four different surveys. Of these 48 individuals, the mean time between release and last resighting observation was 16.3 days (SD: ± 10.40, range: 3–40 days), with the longest time between release and last confirmed sighting being 40 days (individual Red 4Z6; 1st resight: 13 days following release, 2nd: 26 days, 3rd: 33 days, and 4th: 40 days). The longest confirmed observation of any translocated pelican was of an 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G7 unknown red-banded individual on 30 September (56 days post-release) at the Cameron jetties. This individual was also the last banded bird observed during the study (2 weeks following the end of daily counts). Banded Brown Pelicans were observed during almost every survey (32 of 37), with all five of the surveys that lacked banded pelicans occurring during week 6 (Fig. 2). The mean number of banded pelicans observed per survey was 6.8 (SD: ± 5.73, range: 0–20), with a maximum of 20 banded pelicans observed four days after the first release (August 9th; 27% of banded birds available) and zero observed on September 11, 13, 14, 16, and 17. There was a significant difference in the percent of banded pelicans observed by week (χ2 = 20.6, df = 2, P = 0.0010; Fig. 3). We observed a higher percentage of banded birds during week one (mean = 11.5), Figure 2. Mean number of pelicans observed roosting at Rabbit Island (grey bars) and the number of rehabilitated pelicans released at Rabbit Island (black line) by week. Supplemental feeding ended in the middle of week 5. Error bars represent one standard error. Figure 3. Percentage of translocated Brown Pelicans observed at Rabbit Island by week following the initial bird release (light grey bars) and percentage of banded pelicans relative to the total number of pelicans observed (dark grey bars). Supplemental feeding ended on September 7, which occurred during the middle of week 5. Error bars represent one standard error. G8 Southeastern Naturalist Vol. 11, No. 1 Table 2. Resighted translocated Brown Pelicans (i.e., positively identified to individual) during surveys of Rabbit Island (7 August to 18 September). Recovery date indicates day that pelican was collected from the wild due to oil contamination and transferred to a rehabilitation center. Individual Release (color Recovery Release to last leg band) date date 1st resight 2nd resight 3rd resight 4th resight sighting A60 7/19 8/5 8/9 4 A62 7/14 8/5 8/26 21 A65 7/14 8/5 9/2 9/8 34 A75 7/15 8/6 8/26 20 A82 7/8 8/6 8/22 16 A84 7/15 8/6 9/2 27 2Z0 6/9 8/6 8/13 9/1 26 3Z1 6/6 8/6 9/2 27 3Z4 6/21 8/6 9/8 33 3Z6 6/6 8/6 8/25 8/26 8/30 9/9 34 4Z0 6/21 8/6 8/27 21 4Z2 6/6 8/6 8/25 8/29 9/1 9/8 33 4Z3 6/6 8/6 8/30 9/3 9/4 9/6 31 4Z4 6/21 8/6 8/30 24 4Z5 6/20 8/6 8/25 9/1 9/3 28 4Z6 6/18 8/6 8/19 9/1 9/8 9/15 40 4Z7 6/20 8/6 8/30 24 4Z8 6/21 8/6 9/2 27 4Z9 6/8 8/6 8/9 9/1 9/2 28 5Z0 6/20 8/6 9/2 27 6Z8 6/13 8/20 8/29 9 A87 6/4 8/20 9/1 9/2 13 A88 7/6 8/20 9/1 12 A95 6/12 8/20 8/24 9/1 12 C02 7/29 8/20 9/3 9/7 18 C04 6/24 8/20 9/2 9/3 14 C15 7/29 8/28 9/7 9/8 11 C24 8/7 8/28 9/8 11 C26 8/6 8/28 9/8 11 C28 8/4 8/28 9/4 9/9 12 C32 8/3 8/28 9/8 11 C34 8/1 8/28 9/1 3 C39 8/4 8/28 9/8 11 C45 8/8 8/28 9/8 11 C49 8/8 8/28 9/8 11 C50 8/8 8/28 9/1 9/2 4 C51 7/26 8/28 9/2 5 C52 8/1 8/28 9/3 6 C62 7/30 8/28 8/29 9/9 9/14 17 C63 8/6 9/3 9/8 5 C64 8/14 9/3 9/12 9 C73 8/7 9/3 9/6 3 C76 7/31 9/3 9/8 5 C77 8/1 9/3 9/12 9 C78 8/23 9/3 9/8 5 C81 8/10 9/10 9/15 5 C83 8/1 9/10 9/14 9/15 5 C85 8/22 9/10 9/15 5 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G9 and the percentage observed gradually decreased throughout the study (week six mean = 0.89). However, the percentage of banded pelicans relative to total pelicans observed increased significantly across weeks (χ2 = 2.66, df = 5, P = 0.041; Fig. 3), with generally a higher percentage observed as the study progressed. Integration Even though Brown Pelicans were translocated to the north end of Rabbit Island, many of these individuals integrated with the larger Brown Pelican flocks on the southern end of the island (1.2 km SE; Fig. 4); some individuals did this in as little as 1 day post release. In total, 264 observations of color-banded Brown Pelicans were made at Rabbit Island, with 186 observations made on the southern end of Rabbit Island and 78 made on the northern end. These numbers are total banded pelicans observed and likely include recounts of individuals in subsequent surveys. Supplemental feeding Within two weeks of initiating the supplemental feeding regimen, both translocated and native pelicans became “habituated” to the feeding regimen and vessels. After we ceased supplemental feeding on 7 September, we observed fewer total pelicans at Rabbit Island (χ2 = 13.5, df = 1, P = 0.0002), as well as lower percentages of banded pelicans (χ2 = 10.3, df = 1, P = 0.0014). Movements Following the initial translocation, we also observed translocated pelicans short distances away from Rabbit Island. On August 11th (5–6 days following Figure 4. Translocated Brown Pelican (pink C26) observed during supplemental feedings with resident pelicans. Photograph © W. Selman (9/8/10). G10 Southeastern Naturalist Vol. 11, No. 1 the initial release), one banded pelican was confirmed moving a short distance away from Rabbit Island to the northern point of St. John Island (3.4 km to ESE). On many subsequent surveys, we noted banded pelicans on a channel marker near St. John’s Island (September; 3.7 km ESE), flying down the Calcasieu Ship Channel (1, 2, 8, 12, 14, and 15 September; 7.8 km SE), loafing at the Cameron Jetties Park beach (12 September; 11.0 km SSE), and perched on the Cameron jetties (30 September; 12.0 km SSE). Discussion Mortality It is often difficult to determine the level of mortality in animal populations and/or during a translocation event unless tracking devices are affixed to a subsample of individuals prior to release. Under this study’s circumstances (short time frame, ≈5 days between notification of translocation and first bird arrival, and request by rehabilitators), no transmitters or other tracking devices could be affixed to determine mortality. We therefore, had to use the detection of pelican carcasses as an indicator of mortality in this study. The ability to detect carcasses is often quite difficult (Wobeser and Wobeser 1992), and searches are often imprecise and/or inaccurate (Stutzenbaker 1986), primarily due to the persistence time of carcasses on the landscape (Wobeser and Wobeser 1992). One benefit to carcass searching for juvenile pelicans is that they are large, readily identifi- able birds due to their large bills and white breasts. Their carcasses also seem to persist for lengthy periods in water, since carcasses often float, or on land (W. Selman, pers. observ.). Additionally, many pelicans from this population frequent places that the public uses for recreational activities (i.e., Cameron Jetties Park, Calcasieu Ship Channel, Calcasieu Lake), and dead pelicans are often reported by the public to LDWF. During this time period, the public reported on many dead wildlife, and we feel that it was likely that a citizen would have reported a color-banded dead pelican. Even though we did not observe or receive public information on translocated pelican carcasses, we presume that there likely was some low level of mortality due to the inability to accurately survey carcasses in all areas the pelicans may have been. McNease et al. (1984) documented an 89.5% survival rate of translocated Brown Pelicans two weeks after reintroduction. Along with initial mortality rates, it is possible that translocated pelicans, particularly juveniles, more likely succumbed to winter-associated mortality (McNease et al. 1992). Winter mortalities are primarily driven by a lack of food supplies and/or heavy parasite loads during sub-freezing temperatures that occurred five months following the translocation (January and February 2011). During the winter, we observed many native juvenile pelican carcasses, and upon necropsy, all pelicans were found to have empty stomachs and high intestinal parasite loads (W. Selman and J. LaCour, pers. observ.). This scenario is not uncommon, as another study documented a large die-off of wintering Pelicanus 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G11 crispus Bruch (Dalmatian Pelican) in Greece, with all carcasses examined having similar pathologies to those observed in Louisiana Brown Pelicans (Pyrovetsi and Papazahariadou 1995). Persistence of pelicans at Rabbit Island We suspect that four variables worked in concert and prompted the pelican movements observed across the latter weeks at Rabbit Island, including: 1) the natural building of flight muscles leading to emigration from Rabbit Island, 2) unusually high tides at the end of week four that persisted into week five, 3) the beginning of Anas discors L. (Blue-wing Teal) season and the presence of hunters on the island during week six, and 4) the ceasing of supplemental feeding during week five (discussed in later section). Prior to this study, we documented Brown Pelican nesting on Rabbit Island in early May 2010, with 5–7-week-old juveniles present on the island by mid-June (T.J. Hess, Jr., pers. observ.). Brown Pelicans usually are capable of flight by 11–12 weeks and can provide for themselves by 3 months (Shields 2002); almost all translocated pelicans released at Rabbit Island were capable of flight, and therefore, both local and translocated pelicans could have started to move away from the island at the beginning of our study (7 August). Additionally, astronomically high tides (2.5 to 3.75 feet above mean lower low water levels) for seven consecutive days during weeks four and five limited the amount of dry ground available on the island, with coastal flood advisories in effect during this period. Water completely covered the island on several days, and the few pelicans that were observed on the island were all standing in water. Thereafter during week six, we twice observed morning teal hunters on the island when it was closed to waterfowl hunting. On these two days, we observed 6 and 7 total pelicans, likely due to the disturbance of the hunters, which has been widely documented (for review, see Madsen and Fox 1995). However, we did not hear any shots fired by the hunters or see any pelicans leaving the island; thus, we can only surmise that “human activity” on/near the island altered pelican numbers those days. One behavioral aspect of the translocated birds that deviated from local birds was that the percentage of banded pelicans relative to the total number of pelicans observed was significantly higher at the end of the study relative to the beginning. This finding indicates that a larger proportion of translocated birds generally remained at the island slightly longer than native birds. It is unknown why this occurred, but it could be linked to the extended supplemental feedings. However, this pattern did not continue for long, as all birds had vacated the island by the beginning of October (following week 6 of the study). Integration The integration of translocated individuals into native populations is an important factor that should be considered in any translocation study. Individuals may readily integrate into the population or be driven away by native G12 Southeastern Naturalist Vol. 11, No. 1 individuals (Kleiman 1989), with the worst scenario including severe harassment and even death from local individuals (Borner 1985). During this study, many observations were made indicating that rehabilitated Brown Pelicans integrated successfully into local pelican populations. On several occasions, we observed the integration of translocated pelicans with resident pelicans (i.e., loafing, roosting, or flying with native pelicans) within one day following their release. We did not see evidence of translocated pelicans being harassed by resident pelicans (i.e., chased away from island) and/or social groups that were composed of only translocated individuals. Presumably, this translocation provided a similar scenario to the original successful translocation of Brown Pelicans to Louisiana starting in the 1960s (Nesbitt et al. 1978). In future nesting seasons, we will continue to monitor to determine if translocated individuals preferentially form pair bonds with resident or translocated pelicans to further quantify integration into the local colony. Supplemental feeding It appears that supplemental feeding may have had an impact on the persistence of pelicans in the vicinity of Rabbit Island until it ceased in week 5. Cessation of feeding coincided with the other aforementioned events, but there were significantly fewer pelicans observed at the island, as well as the percentage of translocated pelicans observed during surveys, following the termination of supplemental feeding. Even though these feedings provided easy forage for rehabilitated pelicans, they were also heavily exploited by resident pelicans. For future translocations of Brown Pelicans, we suggest that short durations of supplemental feedings should occur, while also considering alternative feeding strategies, including using blinds or remote feeders due to the easy habituation of pelicans to humans. Movements It was difficult to determine if there were long-range movements of translocated Brown Pelicans away from Rabbit Island. This limitation is primarily due to the protocol of our surveys, which did not extend much outside of the Rabbit Island colony survey area, and the inability to equip translocated/resident hatch-year pelicans with satellite transmitters. The latter would have allowed us to accurately document the movements of translocated individuals, as well as allow us to quantify differences/similarities in movements of translocated pelicans relative to local pelicans. This electronic tracking capability, however, could not be achieved due to the immediacy of the response to the oil spill and the short notification time (≈5 days) that we received before the first translocation event. Some translocated pelicans may have dispersed along the coast, but we suspect that most individuals remained in the area but just outside of our survey route; a previous study found that most translocated Brown Pelicans in southeastern Louisiana remained within 32 km of the release site (McNease et al. 1984). Presumably, the low percentage of banded pelicans observed during our surveys, 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G13 especially toward the end of the survey period, is likely due to a combination of these two factors. However, our observations confirm that at least some of the 182 translocated Brown Pelicans following rehabilitation remained in the vicinity of Rabbit Island 56 days following release. Problems with marking and resighting techniques Throughout the study, it was difficult to readily identify translocated pelicans using only color leg-band observations. We could not accurately identify all banded pelicans on the island because many of the leg bands remain concealed when pelicans were in the water, behind a group of other pelicans, far away from the observer (especially in interior ponds at Rabbit Island), and/or when they stood in tall grasses. Therefore, we suspect that some translocated pelicans were uncounted in our surveys, lowering our overall percentage of observed pelicans per survey. Further, due to the long distance of observers from the pelicans (≈100–150 meters), the small size of the color bands, and the inability to stabilize optics on a moving boat, positive identification of color-banded individuals was even more difficult. However, we were able to positively identify some birds to individual on many surveys (26% of all translocated pelicans; Table 2, Fig. 4). Additional visual marks (i.e., paint marks, wing streamers) could have been used to better locate translocated pelicans, especially while flying. However, for this study, we were unable to achieve this due to 1) the logistical difficulties between rehabilitation center and researcher location, 2) the short time to prepare before the translocation event (≈5 days), and 3) the decision by rehabilitators to add no additional permanent markings to the birds in order to alleviate additional stress. Conclusions Many of our metrics to determine success/failure in this translocation were difficult to quantify, including confirmation of marked individuals only via color bands, mortality, and long-range movements back to oil spill areas in southeastern Louisiana. However, in the short term, it appears that this translocation was tentatively a success: we were able to document the persistence of some translocated pelicans at Rabbit Island and the surrounding area, document adequate evidence that translocated pelicans readily integrated with native pelican flocks, and that supplemental feeding likely prolonged pelican presence at Rabbit Island. Along with short-term monitoring, we will continue to conduct long-term monitoring of the Rabbit Island pelican colony, including: 1) documenting the number of translocated pelicans that return to nest at the island and 2) determining if translocated pelicans preferentially pair with native or other translocated pelicans. Both of these aspects are critical to further understanding the long-term success or failure of this translocation. For future pelican translocations, we recommend that a subset of individuals be equipped with radio/satellite transmitters to better document movements and mortality, as well as that additional auxiliary markers G14 Southeastern Naturalist Vol. 11, No. 1 be used to better identify translocated individuals through resighting. Coordinated monitoring on a larger scale would also improve the ability of researchers to determine mortality, persistence, and movements. Supplemental feedings should use strategies that will alleviate human/boat habituation such as blinds and remote-controlled feeders. It would also be beneficial to secure Rabbit Island as a bird sanctuary for the Brown Pelican nesting colony and other colonial nesting wading birds. This protection would prevent excessive human disturbance to nesting birds from hunters, recreational fishermen, and commercial crabbers that access interior portions of the island via tidal channels. Acknowledgments We would like to thank Scott Quinn (British Petroleum [BP]) and Matt Bell (Global Pollution Services [GPS]) for their support in coordinating and compiling the resources necessary to complete the translocation and follow-up monitoring. During the translocation, monitoring, and supplemental feeding, we are grateful for the assistance of LDWF Rockefeller staff Brett Baccigalopi, Chance Baccigalopi, Kim Bourriaque, and Ron Hebert and GPS boat captains Sterling Constance, Travis Constance, and Jeremy Robbins. We would also like to thank the veterinary staff and assistants that provided their time and service during the rehabilitation and translocations, including but not limited to: Jim LaCour (LDWF), Rhonda Murgatroyd (Wildlife Response Services), Louise Clemency (USFWS), Luis Padilla (Smithsonian Institute), and Kevin Castle (National Park Service). All banding was performed under the USGS Bird Banding Laboratory Permit #22884. Scott Walter, Susan DeVries, and two anonymous reviewers provided helpful comments that improved the paper. This paper is dedicated to the LDWF employees and all others who spent endless hours in southeastern Louisiana to battle the oil spill and the ill effects it had on Louisiana’s wildlife. Literature Cited Armstrong, D.P., and P.J. Seddon. 2008. Directions in reintroduction biology. Trends in Ecology and Evolution 23:20–25. Blus, L.J., E. Cromartie, L. McNease, and T. Joanen. 1979. Brown Pelican population status, reproductive success, and organochloride residues in Louisiana, 1971–1976. Bulletin of Environmental Contamination and Toxicology 22:128–135. Borner, M. 1985. The rehabilitated chimpanzees of Rubondo Island. Oryx 19:151–154. Crone, T.J., and M. Tolstoy. 2010. Magnitude of the 2010 Gulf of Mexico oil leak. Science 330:634. Dodd, C.K., and R.A. Seigel. 1991. Relocation, repatriation, and translocation of amphibians and reptiles: Are they conservation strategies that work? Herpetologica 47:336–350. Fischer, J., and D.B. Lindenmayer. 2000. An assessment of the published results of animal relocations. Biological Conservation 96:1–11. Franklin, H.B. 2007. The most important fish in the sea: Menhaden and America. Island Press, Washington, DC. 265 pp. Griffith, B., J.M. Scott, J.W. Carpenter, and C. Reed. 1989. Translocation as a species conservation tool: Status and strategy. Science 245:477–480 2012 W. Selman, T.J. Hess, Jr., B. Salyers, and C. Salyers G15 Hess, T., and J. Linscombe. 2003. Brown Pelican survey and observations: 2003. Unpublished report to the Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. 10 pp. Hess, T. and J. Linscombe. 2004. Brown Pelican survey and observations: 2004. Unpublished report to the Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. 10 pp. Hess, T. and J. Linscombe. 2010. Aerial survey of Brown Pelican nesting colonies along the Louisiana coast. Unpublished report to the Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. 3 pp. Holm, G.O., Jr., T.J. Hess, Jr., D. Justic, L. McNease, R.G. Linscombe, and S.A. Nesbitt. 2003. Population recovery of the Eastern Brown Pelican following its extirpation in Louisiana. The Wilson Bulletin 115:431–437. International Union for Conservation of Nature (IUCN). 1987. IUCN position statement on the translocation of living organisms: Introductions, re-introductions, and re-stocking. Prepared by the Species Survival Commission in collaboration with the Commission on Ecology and the Commission on Environmental Policy, Law and Administration. Available online at http://www.iucnsscrsg.org/policy_guidelines.php. Accessed 7 March 2011. Joanen, T., and L. McNease. 1974. Re-establishment of the Louisiana Brown Pelican. Southern Zoo Workshop, Monroe, LA. 7 pp. Kleiman, D.G. 1989. Reintroduction of captive mammals for conservation: Guidelines for reintroducing endangered species into the wild. Bioscience 39:152–161. Louisiana Department of Wildlife and Fisheries (LDWF). 1941. Common birds of Louisiana. Division of Education and Publicity. New Orleans, LA. 61 pp. LDWF. 2010. Oil spill actions. Available online at http://www.wlf.louisiana.gov/oilspill/ actions. Accessed 19 November 2010. Madsen, J., and A.D. Fox. 1995. Impacts of hunting disturbance on waterbirds: A review. Wildlife Biology 1:193–207. McNease, L., T. Joanen, D. Richard, J. Shepard, and S.A. Nesbitt. 1984. The Brown Pelican restocking program in Louisiana. Proceedings of the Southeastern Association of Game and Fish Commissioners Conference 38:165–173. McNease, L., D. Richard, and T. Joanen. 1992. Reintroduction and colony expansion of the Brown Pelican in Louisiana. Proceedings of the Southeastern Association of Game and Fish Commissioners Conference 46:223–229. Nesbitt, S.A., L.E. Williams, Jr., L. McNease, and T. Joanen. 1978. Brown Pelican restocking efforts in Louisiana. The Wilson Bulletin 90:443–445. NOAA Fisheries Services. 2010. Deepwater Horizon/BP oil spill archives. Available online at http://sero.nmfs.noaa.gov/BPOilSpillArchives.htm. Accessed 19 November 2010. Pyrovetsi, M., and M. Papazahariadou. 1995. Mortality factors of Dalmatian Pelicans (Pelecanus crispus) wintering in Macedonia, Greece. Environmental Conservation 22:345–351. Restore the Gulf. 2010. Deepwater Horizon consolidated fish and wildlife collection report: November 2, 2010. Available online at http://www.restorethegulf.gov/sites/ default/files/documents/pdf/Consolidated%20Wildlife%20Table%20110210.pdf. Accessed 19 November 2010. G16 Southeastern Naturalist Vol. 11, No. 1 Shields, M. 2002. Brown Pelican (Pelecanus occidentalis). In A. Poole (Ed.). The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY. Available online at http://bna.birds.cornell.edu/bna/species/609doi:10.2173/bna.609. Accessed 7 March 2011. Stutzenbaker, C.D., K. Brown, and D. Lobpries. 1986. Special report: An assessment of the accuracy of documenting waterfowl die-offs in a Texas coastal marsh. Pp. 88–95, In J.S. Feierabend and A.B. Russell (Eds.). Lead Poisoning in Wild Waterfowl: A Workshop. National Wildlife Federation, Washington, DC. US Fish and Wildlife Service. 2009. Removal of the Brown Pelican (Pelecanus occidentalis) from the Federal list of endangered and threatened wildlife. Federal Register 74:59443–59471. Williams, L.E., Jr., and L. Martin. 1968. Nesting status of the Brown Pelican in Florida in 1968. Quarterly Journal of the Florida Academy of Sciences 31:130–140. Wobeser, G., and A.G. Wobeser. 1992. Carcass disappearance and estimation of mortality in a simulated die-off of small birds. Journal of Wildlife Diseases 28:548–554.