tWo decadeS of pelagic ecology of the WeSterN aNtarctic peNiNSula

USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: info@tos.org or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA. Oceanography THE OffICIAl MAGAzINE Of THE OCEANOGRAPHY SOCIETY

). The shelf is cut by deep channels and troughs that affect the hydrography and, in turn, help structure the region's biological community. As in other Antarctic coastal marine ecosystems, the production and life cycles of organisms here are intimately tied to the annual cycle and interannual variations in sea ice cover. The WAP differs from the rest of the Antarctic Continent, however, in that it is experiencing the most rapid warming of any marine ecosystem on Earth, with a 3°C increase in annual mean air temperature and a 6°C rise in mean winter temperature over the last six decades (Vaughan et al., 2003). The productive food web, unique geographical setting, and interactions between climate warming and the marine ecosystem aBStr act. Significant strides in our understanding of the marine pelagic ecosystem of the Western Antarctic Peninsula (WAP) region have been made over the past two decades, resulting from research conducted aboard ARSV Laurence M.
Gould and RVIB Nathaniel B. Palmer. These advances range from an understanding of the physical forcing on biology, to food web ecology (from microbes to top predators), to biogeochemical cycling, often in the larger context of rapid climate warming in the region. The proximity of the WAP to the Antarctic Circumpolar Current and WAP continental shelf bathymetry affects the hydrography and helps structure the biological community. Seasonal, spatial, and interannual variability at all levels of the food web, as well as the mechanisms supporting their production, are now more clearly understood. New tools and technologies employed in the region were critical for making this research possible. As a result, our knowledge of the WAP pelagic ecosystem during a time of rapid climate change has vastly improved. roughly 300 km in width. The domain is naturally divided into three subregionscontinental slope, shelf, and coastal regions-consistent with the bathymetry, ocean dynamics, water mass, and biological distributions ; Figure 1). Circulation in the WAP is characterized by a large meandering cyclonic (clockwise) flow that sometimes breaks into northern and southern cyclones with a small anticyclone separating them midway up the peninsula (Hofmann et al., 1996;Martinson et al., 2008).

The ACC delivers Upper
Climate change is directly affecting the WAP marine system. Its most notable impact is on sea ice, whose temporal and spatial distribution plays a major role in upper ocean stratification and in the production, distribution, and life cycles of WAP species at all trophic levels (described below). Specifically, in the WAP and Bellingshausen Sea region, sea ice advances two months later and retreats more than one month earlier than elsewhere in the Antarctic, resulting in a greater than three month shorter ice season. In an area extending from the southern Bellingshausen Sea to the eastern Amundsen Sea, along and just offshore of the continental shelf break, the figure 1. Western antarctic peninsula (Wap) and proximity to the antarctic circumpolar current (acc). (left panel) Map shows the sampling region of the palmer long term ecological research (pal-lter) project (yellow rectangle) and encompasses the Southern ocean global ocean ecosystems dynamics sampling region (approximately in the middle of the pal-lter sampling region, encompassing Marguerite Bay (MB). The orange line separates the coastal from the shelf region, and the blue line separates shelf from deeper (> 750 m) slope waters. The red dot locates palmer Station (right inset) on anvers island (ai); adelaide island (adi) and charcot island (ci) are also indicated. arSV Laurence M. Gould is shown in the left insert. (right panel) location of the acc (in red) in relation to the Wap (yellow rectangle). The acc flows clockwise around antarctica; west of the antarctic peninsula, its southern extent consistently hugs the shelf-slope break and supplies warm, nutrient-laden upper circumpolar deep Water to shelf waters. Location of ACC adapted from Orsi et al., (1995); Martinson and McKee (2012) seasonal sea ice season from 1979-2006 shortened by 83 days (updated from Stammerjohn et al., 2008a), and nearly all of the perennial (permanent) ice in this area has been lost (Stammerjohn et al., 2008b). Repeated sampling in the WAP by the Gould since its launch in 1997 has helped to document the dramatic warming of UCDW . The UCDW is the major source of heat responsible for melting the bottom of the ice streams draining the West Antarctic Ice Sheet as they enter the Amundsen Sea Embayment, and is the likely source of heat dominating winter atmospheric warming in the WAP.

priMary produc tioN aNd MicroBial ecology
Investigations of WAP primary production commenced soon after the Palmer was commissioned and entered Antarctic service in 1992. One of the last RACER cruises, an early LTER cruise, and others used the Palmer from 1992 onward, supplementing earlier work aboard the Polar Duke. The major product of this pioneering research was to establish an understanding of the seasonal variability and mean levels of primary production, and to suggest the mechanisms supporting this production. In particular, localto regional-scale diatom blooms over the shelf were found associated with intrusions of warm, nutrient-rich UCDW onto the shelf (Prézelin et al., 2000). The strong relationship between PP and sea ice, coupled with the sharp decline in the duration of sea ice cover since 1978 (Stammerjohn et al., 2008b) suggest PP rates could be declining as well. Extracting a temporal trend required the greater temporal and spatial resolution of remotely sensed ocean color observations. Analyzing coastal zone color scanner (CZCS) and Sea-viewing Wide Field-of-view Sensor (SeaWiFS) imagery obtained over a nearly 30-year period (1978-1986 and 1998-2006) (Karl et al., 1996). The role of temperature in regulating bacterial activity is a classic problem in microbial ecology. The governing paradigm is that temperature interacts with organic matter availability, and that low temperature (i.e., near the annual minimum) inhibits bacterial activity when organic matter is low (Pomeroy and Deibel, 1986). However, recent observations along the WAP showed no universal relationship between BP and temperature.
Historically, Antarctic marine ecosystems were believed to be simple, linear food chains where primary production by diatoms was conveyed via krill to large consumers such as penguins, seals, and whales. We now know these systems harbor more complicated food webs made up of a microbial assemblage that includes microzooplankton (zooplankton < 200 μm, the majority of which are protozoa). There is a diverse assemblage of microzooplankton in the WAP that can ingest a major fraction of the daily primary and bacterial production, more so than can the macrozooplankton described below, at least in summer (Lori Price, Virginia Institute of Marine Science, pers. comm., July 2012).

Zoopl aNKtoN
The central role that the Antarctic krill, and defecation rates of salps can exceed that of krill (Phillips et al., 2009). Thus, changes in community structure from krill to salps could affect particle export in the WAP (Gleiber et al., in press).
Physical processes play a major role in structuring the WAP zooplankton community. WAP circulation acts both as a means of transport (the WAP is considered to be a source of krill for regions to the north and east) and as a retention mechanism (gyres in regions such as Marguerite Bay could retain krill locally) (Hofmann et al., 1996;Ashjian et al., 2004). While recent studies did not find with endemic neritic species (Donnelly and Torres, 2008;Cullins et al., 2011).
The degree to which oceanic or neritic species dominate the WAP assemblage is dependent upon the subsurface temperature and salinity maxima, with colder, less-saline waters favoring neritic fauna (Donnelly and Torres, 2008). The dominant neritic fishes are the notothenioids, particularly Pleuragramma antarcticum, the Antarctic Silverfish (Figure 3), and they are able to survive in cold shelf waters because of biological antifreezes in their blood (Cullins et al., 2011;and see Detrich et al., 2012, in this issue).
Fronts are known to aggregate prey, and seabirds were most frequently congregated during the winter near the ice edge, the Antarctic Polar Front, and the Shelf Break Front (Chapman et al., 2004). During summer, when both penguins and flying seabirds are constrained by their need to return to their breeding colonies, they were most associated with the Shelf Break Front and the southern boundary of the ACC .
During winter, a region of particularly high abundance of top predators, a biological "hotspot, " was identified within Marguerite Bay (Friedlaender et al., 2011;Figure 5). Marguerite Bay has a persistent polynya and a deep trough that facilitates the intrusion of UCDW onto the shelf and into the bay (Dinniman et al., 2011). Although forage  (McDonald et al., 2008). A surprise finding based on stable isotope signatures from whiskers and scat analysis was that in addition to their primary prey, Antarctic krill, crabeater seals also consume fish (Hückstädt et al., 2012).
The movement patterns and foraging behavior of a variety of birds and mammals have been examined using satellite telemetry. Adélie penguins forage in shallow (< 200 m) waters near land and in the mixed layer (200-500 m) near the edges of deep troughs that cut across the continental shelf, even in winter (Erdmann et al., 2011). In the WAP, Weddell seals appear to be sedentary, remaining in the fjords, whereas crabeater seals move extensively along the continental shelf (as much as 664 km to northeast, 1,147 km to southwest), staying deep within the pack ice throughout the winter and closer to shore in regions where the change in bathymetry was greatest (Burns et al., 2004). Elephant seals moved along the outer margins of the continental shelf and considerable distances offshore into pelagic waters .   Costa et al. (2010) lower range of temperature and salinity typically found on the inner shelf . The CTD-tagged seals have also provided oceanographic data in regions rarely visited by ships and can operate throughout the Antarctic winter ( Figure 6; Costa et al., 2008Costa et al., , 2010. Oceanographic data collected by with peak flux in summer following the annual sea ice retreat and phytoplankton bloom (Ducklow et al., 2008).
Zooplankton fecal pellets, mostly from krill, dominate particulate organic carbon (POC) flux over the WAP continental shelf; thus, climate-induced changes in plankton community structure could alter POC export (Gleiber et al., in press), as noted above. The connection between water column POC flux and benthic processes was explored in the FOODBANCS program (Smith and DeMaster, 2008).
Interestingly, the seasonal response of the WAP-shelf benthic ecosystem is mild compared to the extreme seasonal variability in POC flux. Labile organic carbon from these seasonal flux pulses accumulates in the WAP sediments, resulting in a "food bank" that sustains benthic detritivores through the winter, with processes such as benthic feeding and sediment-community oxygen consumption relatively uninterrupted (Smith and DeMaster, 2008).

coNcluSioN
Clearly, the era of research supported by the Palmer and the Gould has resulted in significant increases in our understanding of the WAP pelagic ecosystem. Prior to this era, we had some understanding of the hydrography, the major "players" in the WAP food web and their life histories, and enough of a foundation to begin to answer: What drives production and distribution of WAP organisms over time and space? The effects of climate warming on the WAP region were just beginning to be recognized, but the effects, if any, on most organisms was unknown. The major contributions of the Palmer and the Gould era have been to make important connections between hydrography and biology, to enhance understanding of how community structure can affect biogeochemical cycling, and to show the unprecedented effects of climate warming on ecosystem structure. The interdisciplinary, cooperative nature of many of the major programs supported by these ships was key to our success, as it will be in the future.