Aerial surveys for pack-ice seals along the Ingrid Christensen and Princess Astrid Coasts, East Antarctica

Author Details: R. Suresh Kumar is a scientist in the department of endangered species management at WII and has been involved in field research on threatened and lesser known fauna since 1998. His research interests are in understanding species distributions, their habitat requirements, assessing their status and developing conservation strategies. And, his current research activity includes the study of Olive Ridley Sea Turtle movements and migration along the Odisha and Maharashtra coast of India using satellite telemetry. Dr. J.A. Johnson is a scientist in the department of habitat ecology at WII and mainly works on aquatic resource management. His primary interests are stream fish ecology, resource partitioning, stream community response on impacts, aquatic habitat monitoring and conservation of threatened fish species.


INTRODUCTION
Five of the 18 extant species of true or earless seals of the family Phocidae: the Southern Elephant Seal Mirounga leonina, Weddell Seal Leptonychotes weddellii, Crabeater Seal Lobodon carcinophaga, Leopard Seal Hydrurga leptonyx and Ross Seal Ommatophoca rossii occur exclusively in the southern Hemisphere.They have a circumpolar distribution around the Antarctic Continent and also occur a little further north in the southern Ocean and on sub-antarctic islands (Folkens et al. 2008).Being amongst the dominant top predators in the Antarctic ecosystem they likely play an important role (Bowen 1997), however the role of marine mammals in aquatic ecosystems is poorly understood (Yodzis 1994(Yodzis , 2001;;Bowen 1997;Mackinson et al. 2003;Morissette et al. 2006).
The population status of the Antarctic pack-ice seals except for the Southern Elephant Seal until recently was unknown and, therefore, a continent-wide census under the Antarctic Pack Ice Seals (APIS) program was initiated by the Scientific Committee on Antarctic Research Group of Specialists on Seals (Southwell et al. 2012).In the 19 th and early 20 th centuries, the Southern Elephant Seal was hunted almost to extinction, while the other species were free from commercial exploitation (Shirihai 2008).These other species escaped by virtue of their range being in dangerous and ice-filled seas (Testa & Siniff 1987).All of these five species of seals are currently protected under the Antarctic Treaty and there has been no commercial sealing in Antarctica since the 1950's, though there have been Norwegian and Russian experimental commercial seal harvests since that time.Due to the present widespread occurrence and large population size globally all of the Antarctic pack-ice seals are classified as Least Concern by the IUCN.
In recent times, loss of sea ice due to climate change has been reported or anticipated to have an impact on marine mammals including pinnipeds (see Learmonth et al. 2006).Loss of suitable breeding and resting habitats of pack-ice seals due to loss of sea ice, coupled with changes in food web dynamics may negatively impact seal populations, specifically the Crabeater and Weddell seals (Siniff et al. 2008;Costa et al. 2010;Forcada et al. 2012).Considering the potential threat to Antarctic pack-ice seals it becomes important to monitor their habitats and track population trends.
The National Centre for Antarctic and Ocean Research (NCAOR), Ministry of Earth Sciences, India, has been undertaking annual expeditions to the Antarctic continent since 1981.As part of NCAOR's "Long-term  then flying out over the fast ice and up to the pack ice edge, and then surveying all along the pack ice edge and over fast ice before returning to the ship.Censuses of Weddell Seals in particular are recommended to be carried out after 1430 and before 1700 hr local time, especially in the moulting season, which is from January to March (Lake et al. 1997).While our study partly overlapped this period we could not conduct surveys during the recommended time due to logistic difficulties, and all surveys were made between 0900 and 1100 hrs local time.
Two observers along with the pilot searched for hauled-out seals on either side of the helicopter.The helicopter was flown at a constant speed on average 90 knots and at an altitude of approx.80m.The pilot and the observer seated next to him reported the seal sightings, the species, and group size, while the observer seated behind the pilot recorded the data.The observer in the front also estimated visually the perpendicular distance of the seal from the track-line and marked the location on a hand-held GPS at every sighting.To take counts of seals in a large congregation the helicopter was slowed down.On a few occasions when the seal species could not be identified or the number of seals in a large congregation could not be counted, a photograph of the same was taken.The photographs were later examined in camp to identify and or count the number of seals recorded at that point.Following Erickson et al. (1993) all seals occurring within 20m of each other were considered as a group.

Analysis
A general summarisation of the seal sightings as encounter rates was made since the survey effort was low and not allocated in a manner that would support design-based estimates of density and distribution.The seal sighting data were segregated according to the sighting location such as shelf-ice/edge, fact ice and pack ice.

Seal survey
A total of six aerial surveys were flown during this study, totalling a length of approx.1200km, comprising four surveys in the Prydz Bay region totalling 800km, and two surveys along the Fimbul Ice Shelf totalling 400km.Except for the Southern Elephant Seal and Ross Seal all the other species of pack-ice seals were recorded during the aerial surveys (Image 2).

Encounter rate
A total of 3645 seals that included Weddell, Crabeater and Leopard seals were counted during the six aerial surveys; of these in the Prydz Bay region alone 3273 seals were counted (see Table 1a & 1b).A total of 41 seals in the Prydz Bay and three in the Fimbul Ice Shelf sites could not be distinguished as to whether they were Weddell or Crabeater Seals.Weddell Seals were the most commonly sighted species in both the areas surveyed (97.0%), and had an encounter rate of 2.9 seals/km.This was followed by the sightings of Crabeater Seals (1.7%) and a single record of a Leopard Seal (0.03%).Ross Seals were not encountered during the aerial surveys, however a pair was seen once from the ship en route to the Fimbul Ice Shelf, hauled out on fast ice.
The encounter rate of Weddell Seals in Prydz Bay (4.0 seals/km) was much higher compared to that of the Fimbul Ice Shelf site (0.8 seals/km).Within Prydz Bay the encounter rate of Weddell Seals differed between the months of survey; there 2.7 seals/km were recorded during December while in January 6.1 seals/km were encountered.Also, Weddell Seal encounters for the two months combined were greater in the western shelf (5.4 seals/km) than in the eastern shelf (2.2 seals/km).The encounter rate of Weddell Seals also differed with respect to location and sea ice conditions.Maximum sightings occurred along the shelf-ice/edge with an encounter rate of 6.8 seals/km, while over the fast ice and pack ice 2.3 and 0.1 seals/km were recorded respectively.

Group size
A total of 668 Weddell Seal groups with sizes ranging from solitary to maximum of 42 individuals were sighted during the surveys in the Prydz Bay site.Maximum numbers of Weddell Seals were observed solitary (n = 237) and accounted for 35.5 % of all groups sighted.More solitary individuals were sighted during surveys in January (40.8 %) than in December (27.7 %) in the Prydz Bay (Fig. 1).A higher number of groups with >5 individuals was seen in January (n = 103) as compared to December (n = 86).

DISCUSSION
The high number of Weddell Seal encounters during this expedition in both Prydz bay and the Fimbul Ice Shelf site is because the survey tracks there were almost entirely near the coast of Antarctica, where the species   is known to largely inhabit (Lugg 1966;Stirling 1969a;Kooyman 1975;Testa & Siniff 1987).The presence of fast ice does not necessarily limit Weddell Seal occurrence as they can maintain breathing holes by abrading the sea ice with their canine teeth (Stirling 1969a).Moreover, in inshore fast ice areas, Weddell Seals use ice breaks or cracks caused by tidal action and glacial movement for breathing, hauling out and pupping (Stirling 1969a).This was typical of the surveyed site in the Prydz Bay, which is located far inshore and with smaller bays sheltered by the rocky slopes of the Larsemann Hills, and is possibly the reason why more Weddell seals were encountered there.In the western Ross Sea counts of Weddell Seals in a square mile of pack ice were reported to be low when compared to a similar track length in fast ice where more than 100 seals were recorded (Stirling 1969), consistent with the species' preference for fast ice during the summer.Within the Prydz Bay site the higher encounters of Weddell Seals in the western shelf than in the east during this study appeared to be as a result of availability of fast ice in the area.Much of the fast ice cover along the eastern shelf was reduced to open water earlier than the western parts.In a span of a month the fast ice cover in the entire area observed in mid-December of approx 2500km 2 reduced to approx.600km 2 in January, much of which remaining in the western parts.Along the western parts the break-away ice sheets from the Amery Ice Shelf provided hauling sites for seals, which likely lead to the crowding of Weddell Seals there and thus resulted in higher encounters.Along the Princess Martha Coast seal densities were reported to increase as the amount of pack ice diminishes with the advance of summer (Bester et al. 1995).
Weddell seals are known to form large aggregations in breeding colonies (Stirling 1969a), but females are reported to be solitary after giving birth; the pupping season being late September to early November (Smith 1965;Stirling 1969a).By the end of November, Weddell Seal females are known to wean their pups (Lake et al. 1997), which may be a reason for the high number of solitary Weddell Seals seen here.Further, during January, the progressing austral summer in Prydz Bay resulted in sea ice melt and innumerable cracks in the fast ice became available for seals to haul-out along, leading to higher number of solitary Weddell Seal sightings.Large aggregations of up to 42 individuals were found in the Prydz Bay area when there was a single tidal crack along a vast stretch of thick fast ice (Image 3).These groups may not represent any sociality and are probably a result of limited number of natural openings in the fast ice.
The low occurrence of other species such as Crabeater and Leopard Seal particularly in the Prydz Bay area during this study is possibly because these seals are inhabitants of the pack-ice edge.Their absence may also be related to their seasonal movements Image 3. Large aggregations of Weddell Seal were only observed when there were a few tidal cracks available for hauling out along a vast stretch of thick fast ice and may not represent any sociality.associated with breeding and feeding.Crabeater Seals are predominantly krill feeders and use pack ice to haul out upon (Gilbert & Erickson 1977;Kooyman 1981;Nordøy et al. 1995;Bengtson & Cameron 2004).Krill is typically abundant near pack-ice and occurs in proximity to the 1000m isobaths around the Antarctic coast during summer (Ichii 1990;Hosie et al. 2000).Consistent with this Southwell et al. (2005), using a presence-ocean depth model predicted Crabeater Seal distribution to be in a band extending northwards by 1.5-5.00 latitude from the continental shelf-break.The Prydz Bay area is located much further south from the predicted distribution range of Crabeater Seals and thus, may have resulted in our low encounters of the species.Similarly, Leopard Seals are believed to breed on the outer fringes of the pack ice in summer, when they are solitary and sparsely distributed (Erickson et al. 1971;Sniff & Stone 1985).In the adjoining Vestfold Hills in the Prydz Bay encounter rates of Leopard Seals during austral spring and summer surveys in 1992 and 1993 were only 0.06 and 0.14, and seals were reported confined to the very outer edge of the fast ice (Rogers & Bryden 1997).
Since our survey effort was only minimal no conclusions could be made, though it appears that seal abundance particularly within the Prydz Bay is related to the extent of fast ice cover available in the area.It is possible that more hauled-out seals may have been encountered had the aerial surveys been conducted mid-day as recommended by Lake et al. (1997), and if more frequent surveys were conducted within and between seasons.We, therefore, recommend that future surveys in the area take into account the time of the aerial surveys and adopt standardized, quantitative methods for design or model-based estimates such as those used by Southwell (2005), Conn et al. (2013), and ver Hoef et al. (2014).

Figure 1 .
Figure 1.Map of Antarctica showing the locations of the two Indian research stations Maitri, in the Princess Astrid Coast and Bharti located in the Larsemann Hills, Ingrid Christenson Coast.The aerial survey routes and the Weddell Seal sighting locations in the Prydz Bay region during December 2009 and January 2010 are shown in the figures below.The fast ice cover observed during the two survey months is only an approximate rendition, though drastic reduction in the spread of fast ice was observed in January.

Figure 1 .
Figure 1.The percentage distribution of different seal groups observed in the Prydz Bay during December 2009 and January 2010.

Table 1a . Details of the aerial survey for seals carried out in the Prydz bay during December 2009 and January 2010. Region & Date Survey stretch Distance (km) Duration (hr:min) Weddel (seal/ km) Crabeater (seal/ km) Leopard (seal/ km) Unknown (seal/ km) Fimbul Eastern Shelf
* Of a total of 205km surveyed only 80km was on fast ice while the rest was open water conditions # Of a total of 125km surveyed only 107km was on fast ice while the rest 18km was open water conditions