Presentations

Harrison, J., S. Van Parijs, S. Moore, L. Alter, J. Barlow, B. Best, M.F. Baumgartner, D. Cholewiak, J. Cleary, M. Ferguson, , K. Forney, L. Garrison, P. Halpin, T. Haverland, A. Kumar, D. Palacios, and J. Redfern. 2011. The NOAA Cetacean Density and Distribution Mapping Working Group: Developing Comprehensive Geospatial Tools to Assist Management in Impact Analyses of Cetaceans in US EEZ Waters. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.

Esch, H.C., M.F. Baumgartner, F.W. Wenzel, and S.M. Van Parijs. 2007. Automated detection of stereotyped baleen whale vocalizations: advantages over traditional methods. 17th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Cape Town, South Africa. November 29 – December 3, 2007.

Schick, R.S., J.S. Clark, B.R. Mate, M.F. Baumgartner, C.S. Slay, S.D. Kraus, P.N. Halpin, A.J. Read, D.L. Urban, F.E. Werner, and B.D. Best. 2007. Should I stay or should I go? Estimating movement forces in N.W. Atlantic right whales with hierarchical bayes. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 7-8, 2007.

Patrician, M.R., I.S. Biedron, H.C. Esch, F.W. Wenzel, L.A. Hall, A.H. Glass, M.F. Baumgartner. 2007. Evidence of the first known western North Atlantic right whale calf (Eubalaena glacialis) born outside of the southeastern U.S. calving grounds. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 7-8, 2007.

Fucile, P.D., R.C. Singer, M.F. Baumgartner, and K. Ball. 2006. A self contained recorder for acoustic observations from AUV's. MTS/IEEE Oceans '06, Boston, MA, September 18-21, 2006.

Baumgartner, M.F. 2005. Oceanographic data sources, data management, and software tools for marine mammal modeling. Workshop on the application of GIS and spatial/temporal prediction models for marine mammal scientists and management. The Society for Marine Mammalogy. San Diego, California. December 11, 2005.

Baumgartner, M.F. and F. Wenzel. 2005. Observations of near-bottom foraging by right whales in the Great South Channel. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 2-3, 2005.

Baumgartner, M.F. and B.R. Mate. 2003. North Atlantic right whale habitat inferred from satellite telemetry. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 4 - 5, 2003.

Baumgartner, M.F. and B.R. Mate. 2002. The foraging ecology of North Atlantic right whales (Eubalaena glacialis) in the lower Bay of Fundy. EOS, Transactions, AGU 83(4), Ocean Sciences Meeting Supplement, Abstract OS41J-10, 2002. Ocean Sciences Meeting. American Geophysical Union and American Society of Limnology and Oceanography. Honolulu, Hawaii. February 11 - 15, 2002.

Baumgartner, M.F. and B.R. Mate. 2000. Right whale diving and foraging behavior. Right Whale Consortium Meeting. Boston, Massachusetts. October 26 - 27, 2000.

Baumgartner, M.F. and K.D. Mullin. 1999. Cetacean habitat in the northern Gulf of Mexico. 13th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Wailea, Maui, Hawaii. November 28 - December 3, 1999.

Baumgartner, M.F. 1999. Characterizing right whale habitat on the Scotian shelf and in the lower Bay of Fundy. Right Whale Consortium Meeting. Boston, Massachusetts. October 21 - 22, 1999.

Trask, R., W. Paul, M.F. Baumgartner. 1998. Design characteristics and field experience with a near surface horizontal moored array. Buoy Workshop. Office of Naval Research and Marine Technology Society. San Diego, CA. April 2, 1998.

Hurst, T.P., M. Johnson, M.F. Baumgartner, D.M. Fratantoni, and K.A. Shorter. 2011. DMON as the basis of a low power, low noise passive acoustic detection and monitoring system. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
The DMON is a low power, low noise acoustic recorder with the capability of running on-board classification and detection algorithms. It’s a compact instrument capable of executing multiple user defined functions, from detection and classification to data compression, and is low power enough to not compromise mission parameters making it ideal for PAM. To meet multiple platform and mission-specific requirements the DMON can be configured for continuous recording or for record-on-detect operation. Continuous recordings are stored to on-board flash memory while detections can be recorded and sent in real time via serial link. We have developed and implemented a beaked whale detector; a pre-whitened CFAR matched filter detector with parametric classification. The detector classifies clicks into three categories (probable, possible, unclassified transient) using four parameters (processing gain, duration, noise level, energy/noise ratio). The settings for the whitening and matched filters were determined using over 300,000 beaked whale clicks from DMON data collected in the Canary Islands in Oct. 2009 and May 2010. APEX profilers and a Slocum glider were deployed at the SCORE range in January 2011 performing 12 hour/1000 meter dives and 40 min/200 meter dives respectively. The profilers reported over 4500 probable beaked whale detections over 50 hours and the glider reported 5600. Post-analysis of the results of one profiler revealed 86% of the probable detections were beaked whales, 11% dolphins and 3% unspecified. Further analysis of the results are needed (going over all of the SCORE data as well as comparisons with other designs) to completely characterize DMON based PAM systems. However, these results show that the DMON, running the real-time detector and classifier, provides a robust first cut of possible beaked whale detections and given its flexible API and 100mWatt power consumption is ideally suited to be the basis of many PAM systems.

Mussoline, S.E., M.F. Baumgartner, D. Risch, and S.M. Van Parijs. 2011. Comparative analysis of the occurrence and seasonality patterns of North Atlantic right whale call types in the Northwest Atlantic. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
Passive acoustic monitoring has the potential to reduce ship strikes, a major cause of right whale mortality, through real time reporting of right whale presence. Although passive acoustic monitoring has focused on right whale upcalls to indicate species presence, it has not yet been verified whether this call type is the most appropriate to use in all seasons and habitats across this species’ range. In addition to the upcall, right whales produce two other call types: gunshots and tonal calls. We examined the relationship between upcalls and gunshots using persistent passive acoustic recordings collected in the Stellwagen Bank National Marine Sanctuary (SBNMS). From December 2007 through April 2008 an array of 10 marine acoustic recording units was deployed in SBNMS. An automated detector (ISRAT, Urazghildiiev et al. 2008, Canadian Acoustics 36: 111-117) was used to determine seasonal occurrence of upcalls, while other right whale call types were identified through manual analysis. Gunshots and upcalls occurred throughout the winter months, with gunshots peaking in January, and upcalls in December. In the early spring, few gunshots were heard compared to near continuous recordings of upcalls. In the wintertime, gunshots occurred primarily between 16:00 and 04:00, while upcalls primarily occurred between 11:00 and 18:00. The average wintertime gunshot rate was 2.15 calls/min, while the average upcall rate was 3.45 calls/min. These analyses demonstrate that the upcall is a suitable call type for identifying right whale presence in SBNMS during both winter and spring. Future research should consider detecting gunshots, however, as these may be of behavioral significance and are useful to corroborate right whale upcalls in the presence of humpback whales, which produce very similar upcalls.

Esch, H.C., P. Tyack, J. Lynch, and M.F. Baumgartner. 2011. Modeling probability of detection of North Pacific right whale calls: implications for abundance estimation using passive acoustics. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
Recent efforts to estimate North Pacific right whale (NPRW) abundance using acoustic cue counting techniques rely on knowledge of the probability of acoustically detecting a particular call. Detection probabilities often incorporate point estimates for input parameters (e.g., mean call source level); however the resulting probabilities do not reflect uncertainty or variability in these parameters. The goal of the current study is to develop a stochastic model of the probability of detecting NPRW calls based on observed distributions of 1) received call amplitude as a function of range and frequency, 2) source level by call type, 3) ambient noise, 4) signal-to-noise ratio (SNR), and 5) the effect of the environment on call propagation. We focus here on NPRW calling behavior in the southeast Bering Sea (SEBS) critical habitat during the late summer because we have collected relevant acoustic and oceanographic data near NPRWs in this region, and recent acoustically-derived abundance estimates have been generated for NPRWs in this area. Preliminary acoustic propagation models illustrate the substantial influence of soundspeed profile structure and caller depth on transmission loss. Received levels (relative to range) are higher for the 300-500Hz than the 1100-1300Hz frequency bandwidth. However, for a given range and frequency bandwidth, received levels vary up to 25dB (rms re 1µPa), reflecting amplitude modulation in call production and/or variation in transmission loss. Source levels reported for North Atlantic right whale calls are reported to vary by 28dB (p-p re 1µPa), reflecting similar flexibility in the amplitude of call production. Ambient noise varied by 17dB (rms re 1µPa) over the 51-hour study period, influencing SNR. This study demonstrates that the probability of detecting NPRW calls depends on many contextual variables. Abundance estimates of NPRWs will be influenced by the extent to which these parameters are characterized within a given region and season.

Silva, M.A., R. Prieto, I. Cascão, C. Oliveira, J. Vaz, M.O. Lammers, M.F. Baumgartner, and R.S. Santos. 2011. Fin whale migration in the North Atlantic: placing the Azores into the big picture. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
North Atlantic fin whales are thought to migrate between low latitude winter mating/ calving grounds and high latitude summer feeding areas. Although there is some information on fin whale distribution and behavior at the feeding grounds, location of the breeding areas remains unknown and migratory patterns are poorly understood. Fin whales cross the Azores every year during the migration to northern latitudes, providing a great opportunity to investigate the migratory behavior of this species. We analyzed information from long-term sighting surveys, 3-years of passive acoustic monitoring and satellite tagging. Fin whales were frequently sighted from May to September, with higher encounter rates in May-June. Few acoustic detections were made during this period; instead, fin whale acoustic activity was more intense during late fall and winter, especially from October to December. Six fin whales that were tagged in spring remained in the Azores for 2-16 days, whereas one whale tagged in September left the area shortly after being tagged. Track sinuosity suggests whales were foraging close to the islands and nearby seamounts. Three tags stopped transmitting while the whales were within 500km of the Azores. All the whales that were tracked to higher-latitudes, including the one tagged in September, followed a nearly straight line trajectory heading to the area between Greenland and Iceland. Our work shows that 1) fin whales are present in the Azores nearly year round, indicating that some individuals may remain at mid-latitudes or that migration is desynchronized; 2) fin whales suspend their migration to forage in the Azores taking advantage of the increased productivity in the area in spring and early summer; and 3) the seasonal peak in singing activity matches the known fin whale mating season, suggesting that whales begin displaying before reaching the mating grounds, or mating may occur close to the Azores.

Lysiak, N.S., J.C. George, and M.F. Baumgartner. 2011. Correlating shifting baselines in Arctic food webs to long-term bowhead whale isotope records. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
Long-term stable isotope records derived from marine mammals present a unique retrospective glimpse at historical conditions, as they record the sources and flow of nutrients within food webs. These isotope records also provide a baseline from which to judge recent changes in the environment, and the subsequent impacts on the ecology of higher consumers. We assessed geochemical changes in the Arctic ecosystem by examining long-term trends in carbon (δ13C) and nitrogen (δ15N) stable isotope ratios of baleen plates from zooplanktivorous bowhead whales (Balaena mysticetus), and by comparing these data to large-scale environmental variability in the Arctic and Bering Sea ecosystems. Incremental carbon and nitrogen stable isotope samples were collected from the baleen plates of bowhead whales harvested during subsistence hunts off of the North Slope of Alaska from 2004-2010. These data were added to an existing archive of bowhead whale baleen stable isotope ratios (from 1974-2003) in order to examine long-term biogeochemical trends. Furthermore, this study performs novel hypothesis testing using atmospheric (carbon dioxide concentrations), oceanic (sea surface temperature and sea ice index), and climatological (El Niño Southern Oscillation and Pacific Decadal Oscillation indexes) data from the North Pacific and Arctic regions to highlight the primary sources of the observed long-term isotopic decrease and inter-annual variability in these multi-decadal δ13C and δ15N bowhead whale stable isotope records. The comparison between biological and physical datasets facilitates the description of the linkages between environmental conditions and bowhead whale ecology, thus using a top predator to monitor the rapidly changing Arctic ecosystem.

Horstmann-Dehn, L., C. George, G. Sheffield, and M.F. Baumgartner. 2011. Bowhead whale feeding efficiency - Making a living in the Arctic. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
Bowhead whales (Balaena mysticetus) are a subsistence resource of cultural significance to Arctic Native communities. Prey density is of paramount importance to filter feeding cetaceans to maintain energy balance, yet little is known about bowhead metabolic demands and digestive efficiency of their common zooplankton prey. Samples of fresh euphausiids and digestive contents were taken along the alimentary tract of subsistence-harvested bowheads during fall 2009 and 2010 from the forestomach, fundic and pyloric chambers, duodenum, and colon (n=10). We used proximate analyzes (%water, %lipid, %protein, %ash) and bomb calorimetry to assess changes in energy density and composition of digesta. Digestive efficiencies were calculated based on "start" composition of euphausiids to "end" composition of colon contents. Protein digestion occurred in the forestomach, consistent with chitinolytic, microbial fermentation leading to lipid release from prey. Efficiency of bowhead protein digestion was ~90%. Lipids were not taken up until the duodenum (consistent with typical mammalian digestion) with an efficiency of ~50%. Due to the high caloric density of lipids, this trend was repeated in total caloric contents of different stomach compartments and dropped from 22.4kJ/g in euphausiids to 10.8kJ/g in colon contents. Using respiratory frequency of whales tagged near Barrow in 2009 and 2010 and lung volume estimates, we determined field metabolic rate (MR) of feeding adults (~9m length) as ~20kW. MR estimates for migrating whales were ~7kW (1.7x Kleiber). Preliminary estimates indicate that feeding whales may expend as much energy acquiring food as is gained and suggests tradeoffs between prey intake and digestibility. This emphasizes the importance of finding high density prey patches and minimizing searching. Fat reserves stored in bowhead blubber far exceed thermoregulatory requirements; we estimate that a 9m bowhead would require 260 days (migratory MR) to 1.2 years (basal MR) of fasting to completely deplete its blubber stores.

Harrison, J., S. Van Parijs, S. Moore, L. Alter, J. Barlow, B. Best, M.F. Baumgartner, D. Cholewiak, J. Cleary, M. Ferguson, , K. Forney, L. Garrison, P. Halpin, T. Haverland, A. Kumar, D. Palacios, and J. Redfern. 2011. The NOAA Cetacean Density and Distribution Mapping Working Group: Developing Comprehensive Geospatial Tools to Assist Management in Impact Analyses of Cetaceans in US EEZ Waters. 19th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Tampa, Florida, USA. November 27 – December 2, 2011.
In 2011, the U.S. National Oceanic and Atmospheric Administration (NOAA) initiated several efforts to improve methods to manage cumulative impacts of human activities on marine mammals, including convening a working group to develop tools to map cetacean density and distribution within U.S. waters. The specific objective of the Cetacean Density and Distribution Mapping Group (CetMap) is to create regional cetacean density and distribution maps that are time- and species-specific, using survey data and models that estimate density using predictive environmental factors. The CetMap identified a hierarchy of methodologies to apply, based on available science, and is producing and/or geospatially depicting one of the following (in order of preference) for all areas, periods, and cetacean species within the U.S. Exclusive Economic Zone (EEZ): 1) habitat-based density estimates; 2) stratified density estimates; 3) habitat affinity indicators; 4) presence only information, or; 5) an indicator that no data are available. New comprehensive habitat-based density modeling, in some cases using new data, is being completed for the U.S. East Coast, the Gulf of Mexico, and the Alaskan Arctic. For all areas, satellite tagging and passive acoustic data are being used to develop presence maps indicating where and when certain species have been tracked or detected, but for which there are not sufficient data to support a density estimate. Additionally, two case studies, using North Atlantic right whales along the U.S. East Coast and sperm whales in the Gulf of Mexico, are being developed to illustrate how satellite tagging and passive acoustic data can be used to augment or refine density estimates to better reflect finer-scale cetacean movement and behavior. These maps and tools will support and strengthen future analyses of anthropogenic impacts on cetaceans and inform human-use mapping efforts integral to emerging U.S. Ocean Policy (e.g., comprehensive Coastal and Marine Spatial Planning).

van der Hoop, J.M. and M.F. Baumgartner. 2010. Estimating swimming paths of tagged baleen whales using an acoustic tracking system. The American Cetacean Society's 12th International Conference. Seaside, California. November 12-14, 2010.
Continuous small-scale tracking of baleen whales allows for the understanding and quantification of multiple behavioural states that cannot be directly observed. The three-dimensional movements of humpback whales were measured by the real-time acoustic tracking system (RATS), an array of floating buoys equipped to detect and localize 36-kHz pings emitted from an animal-borne acoustic transmitter using time differences of arrival. The system provides real-time location estimates of the tagged animal at <5s time intervals with positional accuracies of around 10m. By employing simple filtering techniques, the swimming path of the tagged animal can be resolved over time scales of tens of seconds, and behavioural parameters such as swimming speed and turning angle can be estimated and compared between tagged individuals. This approach will allow individual movement behaviour to be characterized, and for parameters to be established and used in movement models.

Mussoline, S., A. Guerreiro da Silva, D. Risch, M.F. Baumgartner, and S. Van Parijs. 2010. Comparative analysis of the occurrence and seasonal patterns of North Atlantic right whale call types in the Northwest Atlantic. North Atlantic Right Whale Consortium Meeting. New Bedford, Massachusetts. November 3-4, 2010.
Passive acoustic monitoring has the potential to reduce ship strikes, a major cause of right whale mortality, through real time reporting of right whale presence. Although passive acoustic monitoring has focused on right whale upcalls as a metric of species presence, it has not been yet verified whether this call type is the most appropriate to use in all seasons and habitats across this species’ range. In addition to the upcall, right whales produce two other call types: gunshots and tonal calls. We examined the relationship between upcalls and gunshots using persistent passive acoustic recordings collected in the Stellwagen Bank National Marine Sanctuary (SBNMS). From December 2007 through April 2008 an array of 10 marine acoustic recording units was deployed in SBNMS. An automated detector (ISRAT, Urazghildiiev et al. 2008, Canadian Acoustics 36:111-117) was used to determine seasonal occurrence of upcalls, while other right whale call types were identified through manual analysis. Gunshots and upcalls occurred throughout the winter months, with gunshots peaking in January, and upcalls in December. In the early spring, few gunshots were heard compared to near continuous recordings of upcalls. In the wintertime, gunshots occurred primarily between 16:00 and 04:00, while upcalls primarily occurred between 11:00 and 18:00. Average wintertime gunshot activity was 2.15 calls/min, while average upcall activity was 3.45 calls/min. Further analyses will examine the patterns in springtime calling behavior, as well as the distribution of tonal calls in relation to upcalls and gunshots. Although not yet finalized, these analyses demonstrate that the upcall is a suitable call type for identifying right whale presence in SBNMS during both winter and spring. Future research should consider detecting gunshots, however, as these may be of behavioral significance and are useful to corroborate right whale upcalls in the presence of humpback whales.

Baumgartner, M.F., N. Lysiak, C. Schuman, J. Urban-Rich, and F.W. Wenzel. 2010. The influence of copepod diel vertical migration on competition between right whales and zooplanktivorous fish. North Atlantic Right Whale Consortium Meeting. New Bedford, Massachusetts. November 3-4, 2010.
In 1986, Payne et al. (1990, Fishery Bulletin 88:687-696) observed an anomalous reduction in sandlance abundance on Stellwagen Bank and an equally anomalous increase in both right and sei whale abundance. They reasoned that right whales compete for copepods with zooplanktivorous fish, and suggested that the recovery of the right whale may be inhibited in the northwest Atlantic Ocean by competition for its prey by herring, mackerel, and sandlance. Our recent research in the Great South Channel on the diel vertical migration of Calanus finmarchicus – the primary copepod prey of right whales, sei whales, and adult zooplanktivorous fish in this region – suggests that the relationship among these copepod predators may be more complicated than originally envisioned by Payne et al. Diel vertical migration (DVM) is a strategy used by active late-stage C. finmarchicus to avoid predation by visual hunters (e.g., zooplanktivorous fish); copepods remain at depth during the day when surface waters are illuminated, but ascend to the surface at night to feed on phytoplankton. During anchor station studies in 2005-2007, we found that sei whales were absent during periods of strong C. finmarchicus DVM, whereas right whale abundance was unaffected by DVM behavior. Sei whales appear to be restricted to feeding only on surface aggregations of copepods, but tagging studies suggest that right whales can feed on C. finmarchicus both at the surface and at depth during day or night. In habitats with a deepwater refuge, C. finmarchicus is continuously available to right whales, but DVM behavior may significantly reduce predation by zooplantivorous fish. Over banks where C. finmarchicus cannot escape, zooplanktivorous fish likely have much greater control over local C. finmarchicus populations and can outcompete right whales. This hypothesis is consistent with Payne et al. and may explain why right whales are seldom observed directly over banks.

Baumgartner, M.F. and D.M. Fratantoni. 2010. Monitoring right whale distribution and habitat in the Outer Fall region west of Jordan Basin during late autumn with autonomous vehicles. North Atlantic Right Whale Consortium Meeting. New Bedford, Massachusetts. November 3-4, 2010.
Aerial surveys conducted by the NOAA Northeast Fisheries Science Center in recent years have identified a seasonal high-use area for right whales west of Jordan Basin in a region of the Gulf of Maine known as the Outer Fall. Cole et al. (2009, 18th Biennial Conference on the Biology of Marine Mammals) presented evidence to suggest that this late fall and early winter habitat is a mating ground, the first ever discovered for North Atlantic right whales. We seek to understand why this region is important to right whales, but its remote location and frequent stormy conditions make traditional shipboard habitat studies extremely difficult. To overcome these challenges, we deployed three autonomous vehicles (two profiling floats and an ocean glider) to survey the region for one month from mid-November to mid-December 2009. All of the vehicles were equipped with a WHOI-designed DMON passive acoustic recorder. Each profiling float carried a temperature sensor, and the ocean glider carried sensors to measure acoustic backscatter (an indicator of zooplankton abundance), chlorophyll fluorescence (an indicator of phytoplankton abundance), temperature, salinity, and turbidity. The profiling floats surveyed over a broad region of the northwestern Gulf of Maine, while the ocean glider survey focused specifically on the Outer Fall region. Right whale upcalls and gunshots were abundant in the Outer Fall region throughout the deployment period, as were calls from humpback and fin whales. Sound speed profiles from the glider temperature and salinity observations suggest the presence of a surface duct that may dramatically increase the range over which right whales can communicate at this time of year. Future work in this region will focus on identifying the primary prey of right whales (likely diapausing Calanus finmarchicus) and enabling near real-time automated detections of baleen whale calls from gliders and profiling floats.

Baumgartner, M.F. and T. Hammar. 2010. Using a new short-term dermal attachment tag to study bowhead whale foraging ecology in the western Beaufort Sea. AGU/ASLO Ocean Sciences Meeting. Portland, Oregon. February 22-26, 2010.
Bowhead whales visit the waters off of Barrow, Alaska in the late summer during their migration from the eastern Beaufort Sea summering grounds to the northern Bering Sea wintering grounds. Stomach contents from harvested whales indicate the whales feed primarily on euphausiids in this region, but little is known about the whales’ foraging behavior. Attempts to deploy suction-cup attached archival tags on the whales to observe their diving and feeding behavior were unsuccessful during September 2008 owing to evasive behavior by the whales and rough uneven skin that was not amenable to suction cup attachment. A new projectile, short-term, dermal attachment tag consisting of a small implantable anchor and a trailing instrument housing was developed (1) to increase the distance at which a tag can be deployed and (2) to ensure attachment regardless of skin condition. The new tag was attached to 3 bowhead whales in September 2009 for periods of 0.5-5 hours. Oceanographic and prey sampling was conducted in proximity to the tagged whales using a vertical profiling instrument package consisting of a conductivity-temperature-depth instrument, fluorometer, optical plankton counter, and video plankton recorder. These simultaneous observations of diving behavior and the environment indicated that the tagged whales were feeding several meters off of the sea floor on layers of euphausiids. Contrary to observations of foraging in other baleen whales, our preliminary results suggest that the tagged whales did not engage in area-restricted foraging behavior, but instead moved long distances while feeding.

Esch, H.C., M.F. Baumgartner, C. Berchok, and A.N. Zerbini. 2009. Fine-scale temporal variability of North Pacific right whale call production. 18th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Québec, Canada. October 12-16, 2009.
The North Pacific right whale (NPRW) is a critically endangered species, and is currently the focus of intensive monitoring efforts in the southeastern Bering Sea (SEBS). Although passive acoustic monitoring has proven to be a useful technique for detecting right whales in this region, little is known about fine-scale temporal patterns in call production. In addition, coincident acoustic and oceanographic data are rarely reported, although water column structure directly influences the acoustic ecology of right whales. In August 2008, a multidisciplinary research effort was conducted to investigate the distribution, movement and ecology of right whales in the SEBS. To evaluate the relationship between right whale call production and environmental conditions, we deployed a free-floating geo-referenced passive acoustic listening array in NPRW critical habitat (18:00 6Aug – 10:00 7Aug). After deploying the 2 km radius array, we positioned our vessel in the center of the array and collected profiles of temperature, salinity, fluorescence, and zooplankton abundance every half-hour. We analyzed 16 hours of acoustic recordings, yielding 12,086 calls (gunshots, moans, upcalls). In general, call rates (calls/hour) increased over time. Gunshots comprised over 80% of calls produced during the first 2 hours (evening); however, gunshot production decreased to 29% in hour 3, and no gunshots were detected throughout the remainder of the recordings. Conversely, moans comprised 3% of calls produced in hour 1, increased in hours 2 and 3 (15% and 61% respectively), and dominated throughout the night and following morning (~90% in hours 4-16). Upcalls were produced at low levels in all hours. Variability in calling behavior did not appear to be related to changes in oceanographic conditions, including the distribution of the right whale’s primary copepod prey. Future research efforts will focus on extending the sampling period to 24+ hours to further investigate periodicity in call production and environmental conditions.

Baumgartner, M.F., H.C. Esch, and A.N. Zerbini. 2009. Association between North Pacific right whales and a subsurface front in the southeastern Bering Sea. 18th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Québec, Canada. October 12-16, 2009.
The population of eastern North Pacific right whales (Eubalaena japonica) likely numbers in the tens of animals and is therefore seriously endangered, yet very little is known about their distribution and ecology. During the summer of 2008, associations between North Pacific right whales, their primary copepod prey, and oceanographic conditions were studied in the southeastern Bering Sea. Two adaptive cross-isobath transects (37 and 56 km long) were conducted on August 6 and 12 through areas occupied by right whales to examine gradients in oceanographic conditions in proximity to the whales. Every 4.6 km along the transects, a vertical profiling instrument package was deployed to measure temperature, salinity, chlorophyll fluorescence, and zooplankton abundance and community composition throughout the water column. Strong vertical stratification was observed throughout the region with a sharp pycnocline separating warm fresh waters in the upper 20 m from cold salty waters at depth. A subsurface front was observed in the same area as the right whales. Compared to the shallower waters to the northeast of this front, the deeper waters to the southwest were characterized by colder temperatures, higher salinity, and higher phytoplankton and zooplankton biomass. This water mass is termed the "cold pool," the remnants of nutrient-rich winter water, and we hypothesize that it may support increased phytoplankton production, which in turn, fuels higher production of Calanus marshallae, the primary prey of right whales on the Bering Sea shelf. We further hypothesize that right whales are largely found within the confines of this water mass during the summer in the southeastern Bering Sea.

Newhall, A.E., Y-T. Lin, J.F. Lynch, and M.F. Baumgartner. 2009. Sei whale localization and vocalization frequency sweep rate estimation during the New Jersey Shallow Water 2006 experiment. 157th Meeting of the Acoustical Society of America. Portland, Oregon. May 18-22, 2009.
The Shallow Water 2006 (SW06) experiment was conducted in the Mid-Atlantic continental shelf off the New Jersey coast. A fast-sampling, 48-channel hydrophone array system recorded a number of sei whale (Balaenoptera borealis) vocalizations during this time. This system had 16 hydrophones on the vertical line array (VLA) component covering the water column from 13-m depth to the bottom (80 m) and 32 hydrophones on the 480-m horizontal line array (HLA) component that lay on the bottom. The sei whale receptions on the arrays are low-frequency (less then 100 Hz) downsweep chirps and have the typical acoustic modal arrival pattern seen in shallow-water, low-frequency sound propagation. Due to acoustic modal dispersion, the frequency sweep durations observed from the receptions are found to be longer than the original signal. A horizontal beamforming technique using the HLA component is implemented to determine the bearings to whales. The whale location along the determined bearing and the frequency sweep rate of original whale sound are simultaneously estimated using the VLA component with a multiple-parameter optimization technique. This optimization technique is based on acoustic normal mode theory and is designed to compensate for the effect of acoustic modal dispersion.

Johnson, C.L., A.W. Leising, J.A. Runge, A.M. Tarrant, M.F. Baumgartner, E.J. Head, P. Pepin, and S. Plourde. 2008. The lipid accumulation hypothesis for dormancy control in oceanic copepods. Summer Meeting. American Society of Limnology and Oceanography. St. Johns, Newfoundland, Canada. June 8-13, 2008.
Demographic time-series of Calanus finmarchicus abundance from four stations in the northwest Atlantic Ocean were analyzed to estimate the timing of entry into and emergence from dormancy and to evaluate hypothesized environmental cues for dormancy. No single environmental cue (photoperiod, temperature, or chlorophyll concentration) explained entry or emergence dates across all stations. Among hypotheses to explain dormancy in Calanus, only the lipid accumulation window hypothesis for onset of dormancy and a lipid-modulated endogenous timer controlling dormancy duration could not be eliminated. According to these hypotheses, individuals can only enter dormancy if their food and temperature history allows them to accumulate sufficient lipid to endure overwintering, molt, and undergo early gonad maturation. Individual-based modeling of C. finmarchicus population growth, including lipid-based dormancy control, and recent research on differential gene expression in active and dormant C. finmarchicus both support a key role for storage lipid accumulation during preparation for dormancy. While lipid accumulation probably does not control dormancy, it may be either a downstream result of a shift in metabolic pathways or a precondition for endocrine control of dormancy.

Tarrant, A.M., M.F. Baumgartner, T. Verslycke, and C.L. Johnson. 2008. Molecular characterization of diapause in the marine copepod Calanus finmarchicus. Summer Meeting. American Society of Limnology and Oceanography. St. Johns, Newfoundland, Canada. June 8-13, 2008.
During the last juvenile stage, Calanus finmarchicus undergoes a facultative diapause consisting of delayed development, migration to depth, reduced metabolism, and utilization stored lipids. To facilitate studies of diapause regulation, we characterized gene expression patterns in C. finmarchicus C5 copepodids collected from surface and deep waters of the Gulf of Maine. Morphological and biochemical measurements demonstrated that these copepodids were from active and diapausing populations. Quantitative RT-PCR indicated (1) upregulation of three genes associated with lipid synthesis, transport and storage in active copepods, (2) upregulation of ferritin in diapausing copepods, consistent with a role of ferritin in protecting lipids from oxidative damage, and (3) upregulation of ecdysteroid receptor in diapausing copepods. This study represents the first molecular characterization of gene expression associated with calanoid copepod diapause, and provides a suite of markers that could be used to indicate physiological changes associated with preparation for, progression through, and emergence from diapause.

Baumgartner, M.F. and F.W. Wenzel. 2008. Springtime foraging ecology of North Atlantic right whales. Ocean Sciences Meeting. American Geophysical Union and American Society of Limnology and Oceanography. Orlando, Florida. March 2-7, 2008.
The North Atlantic right whale is a zooplanktivorous whale that feeds on the calanoid copeopod Calanus finmarchicus in the northwestern Atlantic Ocean. The foraging ecology of this species has been studied intensively in summertime habitats where it feeds on deep layers of diapausing C. finmarchicus. We conducted a study of right whale foraging ecology in a springtime habitat where right whales feed primarily on active non-diapausing (often vertically migrating) populations of C. finmarchicus that aggregate at the surface, at the sea floor, and at the pycnocline. These conditions are different from the right whale summertime habitats, and consequently, right whales employ different foraging strategies. We attached tags consisting of time-depth recorders and pitch and roll sensors to right whales, tracked their movements, and sampled both oceanographic conditions and prey distribution in proximity to the tagged whales using a CTD, optical plankton counter, and a video plankton recorder. The results of our study demonstrate the importance of predator behavior and prey life history in the development of oceanic hotspots.

Baumgartner, M.F. 2007. Comparative studies of baleen whale foraging ecology. New Frontiers in Marine Science, Early Career Scientists Conference, North Pacific Marine Science Organization (PICES). Baltimore, Maryland. June 26-29, 2007.
Baleen whales have evolved to feed relatively low on the marine food chain (copepods, euphausiids, small bait fish), and their high metabolism and large body size imply that these whales need to consume enormous quantities of prey. Prior to exploitation, baleen whales were likely important components of the marine ecosystem throughout the temperate, sub-polar, and polar seas. The need to efficiently capture small organisms spurred the evolution of baleen, which acts as a sieve to separate prey from seawater. Yet not all whales use their baleen in the same way: right and bowhead whales continuously strain copepods by simply opening their mouths and swimming forward (much like a plankton net), blue and fin whales engulf entire aggregations of euphausiids in a single mouthful and then slowly strain these prey through their small baleen, and humpback whales engage in elaborate behaviors to corral bait fish into a tight aggregation before engulfing them. These predator behaviors are adapted to the kind of prey the whales are targeting, the behavior of the prey, and the particular relationship between each prey species and its environment. Documenting these behaviors has been challenging, since much of it occurs below the sea surface and out of view. Concomitant advances in tagging and tracking technology and instrumentation to measure prey distribution have enabled novel studies of baleen whale foraging ecology that can improve our understanding of how these whales find and exploit prey resources and what impact, if any, these large predators have on the marine ecosystem. I will discuss the application of these new tools to investigations of right and humpback whale foraging behavior and the relationship between those behaviors and that of their prey.

Baumgartner, M., S. Van Parijs, F. Wenzel, and A. Warde. 2006. Linking right whale acoustic behavior with environmental factors in the Great South Channel. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 8-9, 2006.
Passive acoustic monitoring promises to be an extremely useful means to augment traditional visual surveys for characterizing the distribution, behavior, and ecology of North Atlantic right whales. Accurate assessments of occurrence and distribution from acoustic data require an understanding of temporal variability in right whale calling rates and the environmental factors that influence that variability. We examined right whale acoustic behavior in the context of both prey distribution and oceanographic conditions over diel time scales in the Great South Channel to determine whether calling rates were influenced by potential feeding conditions. On two separate occasions, we deployed four recoverable pop-up buoys in a diamond pattern around a central anchor station. The buoys were each located 3.7 km (2 nmi) from the anchor station, and each was equipped with an acoustic recorder that sampled at 10 kHz. At the anchor station, the NOAA Ship Albatross IV deployed a vertically-profiled instrument package consisting of a video plankton recorder, optical plankton counter, conductivity-temperature-depth profiler, and a fluorometer once every half hour for 24 and 37 hours during the two studies occurring on May 7 and May 23 2006, respectively. During daylight hours, a 15-minute visual census of animals near the ship was also conducted once every half hour. Diel vertical migration of right whale prey, Calanus finmarchicus, was clearly evident in the optical plankton counter data, and right whales were encountered near the anchor station during both studies. Temporal variability in calling rates relative to the diel vertical migration of C. finmarchicus will be discussed.

Fucile, P.D., R.C. Singer, M.F. Baumgartner, and K. Ball. 2006. A self contained recorder for acoustic observations from AUV's. MTS/IEEE Oceans '06, Boston, MA, September 18-21, 2006.
An experiment to acoustically observe baleen whale vocalizations from autonomous underwater vehicles (AUV's) in the southwestern Gulf of Maine was conducted in May of 2005. A digital recorder and hydrophone were installed on three Slocum Electric Gliders, which have the desirable features of a low acoustic signature and the ability to persistently keep station. To adequately identify whale vocalizations, the recorders combined fidelity (a combination of sampling frequency and bit depth) and unit-to-unit clock accuracy, for determining source location by using time of arrival triangulation. A fourth glider was configured with an ADCP. All four gliders were also outfitted with conductivity, temperature, depth, and bio-optical sensors. They were programmed to keep station 5-7 km apart in an approximate equilateral formation. The recorders were installed internally with a bottom-mounted hydrodynamic-shaped hydrophone. The electronics were electrically isolated from the glider and designed to be started and stopped in response to the passage of a magnet near a section of the hull. Each recorder, powered by a 10 'C' cell battery pack and using a 4 GB compact FLASH card, can thus store a single channel linear encoded 16-bit audio signal, sampled at 2048 Hz for 12 days. A module was designed and built to synchronize the ultra stable temperature tracking real time clocks. By using spectrogram cross correlation of the acoustic data, and estimates of the glider location, whale positions were determined with an accuracy of a few hundred meters.

Baumgartner, M.F., D.M. Fratantoni, and C.W. Clark. 2006. Investigating baleen whale ecology with simultaneous oceanographic and acoustic observations from autonomous underwater vehicles. Eos Transactions, American Geophysical Union 87(36), Ocean Sciences Meeting Supplement. Abstract OS24E-05. Honolulu, Hawaii. February 20-24, 2006.
A central goal of marine mammal ecology research is to understand the environmental factors that influence animal distribution. Standard methods use ship-based human observers and in-situ instrumentation to simultaneously characterize marine mammal distribution and oceanographic conditions. Unfortunately, this approach is expensive, labor intensive, and inefficient, since observers can only detect marine mammals in daylight, good visibility, and low sea state. In contrast, autonomous underwater vehicles are capable of continuous operation, even in adverse conditions, while measuring many of the same oceanographic properties observed during ship-based studies. Autonomous gliders, in particular, have characteristics that make them ideal for ecological studies: high endurance (> 60 days at sea), fine horizontal and vertical measurement resolution, and relatively silent operation that allow passive acoustic measurements. To characterize the physical and biological environment near baleen whales over spatial scales of several kilometers, we deployed an array of gliders 75 km southeast of Cape Cod, Massachusetts near an aggregation of right and sei whales. Each of the four gliders continuously measured temperature, conductivity, depth, chlorophyll fluorescence, and optical backscatter while profiling from the surface to 100 m once every 20 minutes. In addition, three gliders were outfitted with custom-built acoustic recorders, and the fourth carried an acoustic Doppler current profiler (ADCP). The gliders were deployed on May 6, 2005, operated flawlessly through a gale during which seas reached 5.2 m (17 ft), and were recovered on May 11. Downsweep whale calls in the 30-90 Hz frequency band exhibited a diel pattern (fewer calls by night, more by day) that corresponded strongly to the diel vertical migration of zooplankton observed in the ADCP acoustic backscatter measurements. These results may indicate a diel pattern in feeding by the whales producing these 30-90 Hz calls (likely fin or sei whales). The number of calls in this frequency band did not diminish during the gale. Acoustic events were accurately time-stamped, thus the moving glider array was also used to determine positions of vocalizing whales. Our pilot study has demonstrated the utility of autonomous platforms for collecting high-resolution oceanographic data while simultaneously monitoring the presence and location of vocalizing marine mammals over time scales of days.

Baumgartner, M.F., D.M. Fratantoni, and C.W. Clark. 2005. Advancing marine mammal ecology research with simultaneous oceanographic and acoustic observations from autonomous underwater vehicles. 16th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. San Diego, California. December 12-16, 2005.
A central goal of marine mammal ecology research is to understand the environmental factors that influence animal distribution. Standard methods use ship-based human observers and in-situ instrumentation to simultaneously characterize marine mammal distribution and oceanographic conditions. Unfortunately, this approach is expensive, labor intensive and inefficient, since observers can only detect marine mammals in daylight, good visibility, and low sea state. In contrast, autonomous underwater vehicles are capable of continuous operation, even in adverse conditions, while measuring many of the same oceanographic properties observed during ship-based studies. Autonomous gliders, in particular, have characteristics that make them ideal for ecology studies: high endurance (> 60 days at sea), fine horizontal and vertical measurement resolution, and relatively silent operation that permits passive acoustic measurements.

We deployed an array of gliders off Cape Cod, Massachusetts to characterize the physical and biological environment near baleen whales. Each of the four gliders continuously measured temperature, conductivity, depth, chlorophyll fluorescence, and optical backscatter while profiling from the surface to 100 m once every 20 minutes. In addition, three gliders were outfitted with custom-built acoustic recorders, and the fourth carried an acoustic Doppler current profiler (ADCP). The gliders were deployed on May 6, 2005, operated flawlessly through a gale during which seas reached 5.2 m (17 ft), and were recovered on May 11. Downsweep whale calls in the 16-160 Hz frequency band exhibited a diel pattern (fewer calls by night, more by day) that corresponded strongly to the diel vertical migration of zooplankton observed in the ADCP acoustic backscatter measurements. The number of calls in this frequency band did not diminish during the gale. Acoustic events were accurately time-stamped, thus the moving glider array was used to determine the position of vocalizing whales. Our novel observations demonstrate the utility of autonomous platforms in marine mammal ecology studies.

Angell, C.M., W.L. Perryman, P.B. Best, D. Reeb. H.M. Pettis, A.R. Knowlton, M.F. Baumgartner, and M.J. Moore. 2005. Understanding right whale body fat condition. 16th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. San Diego, California. December 12-16, 2005.
Investigating biological and environmental factors contributing to high variability in reproductive performance of northwest Atlantic right whales is critical to the conservation of this endangered population. Blubber thickness and body shape (a series of body widths) were measured on free-swimming right whales (Eubalaena glacialis and Eubalaena australis) using ultrasound and aerial photogrammetry to determine whether body fat condition varied in response to changes in reproductive condition and periods of reduced prey availability. Blubber thickness of females decreased during lactation, increased after weaning, and was thickest in females measured a few months prior to the start of pregnancy, indicating that females draw on blubber for energetic support for reproduction. Moreover, body shape caudal to the blowholes of lactating females was significantly thinner than that of non-lactating females; most of this reduction occurred during the fasting portion of lactation. Juveniles and adult males measured during a year of low food abundance had significantly thinner blubber than those measured during years of greater food abundance. Taken together, these results suggest that blubber thickness and dorsal body shape are indicative of right whale energy balance; therefore, the marked fluctuations in northwest Atlantic right whale reproduction have a nutritional component. Additionally, in the absence of assessing the capacities of these indices to predict actual body fat condition, the relationships between the indices were explored and compared with subjective assessments of relative fatness. While subjective assessments appeared to capture general trends in body fat condition, they lacked the precision of the quantitative indices. Maximum body width and blubber thickness were correlated in juveniles, but not in lactating females, suggesting differences in rates of lipid catabolism among various lipid reserves. Such comparisons indicate assessing body fat condition with a combination of indices is important to future monitoring of the condition of the northwest Atlantic right whale population.

Baumgartner, M.F. 2005. Inferring top predator distribution from primary production: Lessons from the North Atlantic right whale. Tagging of Pacific Pelagics, Oceanographic Hot Spots Workshop. Pacific Grove, California. December 7-8, 2005.
It is tempting to look at remotely-sensed ocean color data, pick out the "green" patches in the ocean, and say, "Look at all that phytoplankton biomass; surely this is a hotspot." If hotspots are defined as areas that attract top predators in the ocean, then this statement assumes that high phytoplankton biomass attracts top predators. Is this really the case? Clearly in some environments, notably the eastern boundary current upwelling systems (e.g., off California and Peru), persistent phytoplankton production causes high secondary production and attracts fish, piscivores, and eventually top predators. However, in other environments primary production may not be as persistent, and trophic interactions can be quite complex.

The North Atlantic right whale is an unusual top predator that feeds on herbivorous calanoid copepods. Because it sits atop a very short food chain (phytoplankton - zooplankton - whales), the right whale's distribution might be closely tied to primary production. Three studies of right whale ecology were conducted at different spatial scales to examine the oceanographic processes that influence right whale distribution - a foraging ecology study involving short-term tagging (hundreds of meters to kilometers over hours), simultaneous visual and oceanographic vessel surveys (tens of kilometers over days), and a satellite-tracking study (hundreds of kilometers over weeks). These studies indicated that phytoplankton biomass had little to do with right whale distribution during the summer. Instead, the life history of a major prey species and physical processes influencing both horizontal and vertical aggregation of prey were paramount.

Baumgartner, M.F. and F. Wenzel. 2005. Observations of near-bottom foraging by right whales in the Great South Channel. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 2-3, 2005.
Floating ground line in pot or trap fishing gear may pose a significant entanglement risk to baleen whales. To document this risk, a better understanding of how whales use the water column, particularly near the sea floor, is required. We began a study of North Atlantic right whale diving and foraging behavior in the western Gulf of Maine during the spring of 2005 aboard the NOAA Ship Albatross IV. Archival tags were attached to right whales in the Great South Channel, and we sampled zooplankton and phytoplankton distribution and the physical structure of the water column near the tagged whales. Sea floor depths were also measured near the tagged whales using both a ship-mounted echo-sounder and a profiled instrument package that included a depth sensor and an acoustic altimeter. Several tagged right whales made presumed foraging excursions to the bottom, remaining there for up to eight minutes at a time. Sampling in proximity to the tagged whales with an optical plankton counter (OPC) indicated that high abundances of Calanus finmarchicus occurred near the bottom. Additional sampling at a fixed station for 28 hours with an OPC indicated that some C. finmarchicus vertically migrated from the surface to the bottom (100 m) between midnight and 0600, remained at the bottom until 1400, and returned to the surface by sunset. These preliminary results suggest that bottom feeding in the Great South Channel may only occur during daylight hours.

Herman, A.W., M.F. Baumgartner, and D.M. Checkley. 2005. Utilization of biomass size spectra in monitoring zooplankton community structure and identifying species. Summer Meeting. American Society of Limnology and Oceanography. Santiago de Compostela, Spain. June 19-24, 2005.
Normalized biomass size spectra (NBSS) and log-based size distributions have evolved from the allometric representations of Sheldon (1977) and Platt and Denman (1979) and are powerful analytical tools in describing zooplankton community structure. The size data sampled by Optical Plankton Counters (LOPCs) can be collected by LOPC rapidly over broad regions and lend themselves to NBSS analyses by describing a broad size range of zooplankton. NBSS based on LOPC data are being utilized for developing monitoring indices in the Atlantic Zone Monitoring Program of the Bedford Institute. An analytical method has been developed using NBSS 'residuals', that is, the difference between the measured NBSS and a minimum NBSS base, and is used to track zooplankton community shifts as a function of 1) size, 2) dispersion (broadness of species composition), and 3) biomass production. Other analytical properties emerging from the NBSS analyses are the logarithmic compression of the LOPC size scales which isolates peaks of individual zooplankton species otherwise unobservable. Examples are Calanus finmarchius V from the Gulf of Maine, Paracalanus spp. from the Scotian Shelf and species groups from the CALCOFI region.

Fratantoni, D.M. and M.F. Baumgartner. 2005. AUV-based physical, biological, and acoustic observations in support of marine mammal ecology studies. International Ocean Research Conference. The Oceanography Society. Paris, France. June 6-10, 2005.
A central goal of marine mammal ecology research is to understand the environmental factors that influence animal distribution. Current protocols use ship-based human observers to characterize marine mammal distributions by visually detecting and counting individual animals. A variety of relevant oceanographic properties (both physical and biological) are simultaneously measured using shipboard instruments. These two datasets are examined together to identify associations between the animals' distribution and oceanographic features or conditions. Unfortunately, this approach is expensive, labor intensive, and inefficient, since observers can only detect marine mammals with daylight, good visibility and low sea state. Furthermore, the vertical and horizontal resolution of the in-situ oceanographic survey is typically coarse relative to the expected scale of prey distributions.

Autonomous underwater vehicles are capable of working continuously in adverse conditions while measuring many of the same oceanographic properties observed during ship-based ecological studies. Here we present results from a recent field study during which several high-endurance autonomous gliders were used to characterize the physical and biological environment near actively feeding North Atlantic right whales (Eubalaena glacialis). This application exploits intrinsic characteristics of gliders: high endurance, fine vertical measurement resolution, and relatively silent operation. The slow-moving (0.5 m/s) gliders collected several days of high-resolution physical and bio-optical data in proximity to whales observed east of Cape Cod, Massachusetts. Passive acoustic recorders deployed on the array of gliders enabled detection of whale vocalizations and rudimentary localization of individual animals. This experiment demonstrates the utility and efficiency of autonomous platforms for long-term studies of marine mammal ecology.

Baumgartner, M.F. and B.R. Mate. 2003. The foraging ecology of North Atlantic right whales and its potential energetic implications. 15th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Greensboro, North Carolina. December 15 - 19, 2003.
The foraging behavior of the endangered North Atlantic right whale (Eubalaena glacialis) was investigated in the lower Bay of Fundy and the southwestern Scotian Shelf during the summers of 2000 and 2001 by monitoring the diving behavior of whales equipped with suction-cup mounted archival tags. The vertical distribution of the right whaleís primary prey, the older life stages of the copepod Calanus finmarchicus, was measured at resurfacing locations of the tagged whale using an optical plankton counter. Right whale feeding dives were characterized by rapid descent from the surface to a particular depth between 80 and 175 m, remarkable fidelity to that depth for 5-14 min and then rapid ascent back to the surface. Calanus finmarchicus aggregated in discrete layers above a turbulent bottom mixed layer, and a strong correlation was observed between the average right whale feeding dive depth and the average depth of peak C. finmarchicus abundance (r = 0.902, n = 31 individuals, p < 0.0001). We estimated maintenance metabolic rates for the tagged whales using a simple model based on the Kleiber curve, and compared these to estimated ingestion rates computed from observations or literature values of gape area, swim speed, time at depth, and the abundance and energy content of C. finmarchicus. Of the 22 whales tagged for over an hour, 8 (36%) did not acquire enough energy at the estimated ingestion rates to meet the modeled daily maintenance needs. Biases in our methods (e.g., short attachment durations, limited seasonal and geographic sampling, uncertainty in metabolic rates) make it impossible to determine from these observations if right whale reproduction is limited by nutritional uptake. However, this study has demonstrated a method for producing meaningful estimates of prey consumption by planktivorous whales over time scales of several hours.

Baumgartner, M.F. 2003. The two faces of habitat research: understanding ecology vs. making predictions. Workshop on the application of GIS and spatial/temporal prediction models for marine mammal scientists and management. The Society for Marine Mammalogy. Greensboro, North Carolina. December 13, 2003.
The development of models to predict marine mammal distribution will improve conservation efforts by enabling better management of critical habitats. However, robust prediction models should be based on a fundamental understanding of marine mammal ecology and regional oceanography, not on correlations alone. Given our poor understanding of the environmental and behavioral factors that influence distribution at nearly all spatial and temporal scales, more emphasis must be placed on studies and methods that elucidate habitat characteristics and the frequently complex links between marine mammals, their prey and the ecosystem. Only with this information can reasonable prediction models be built, validated and ultimately used for conservation purposes.

The North Atlantic right whale is an endangered species that sits atop a seemingly simple, 3-tiered trophic chain consisting of whales, copepods and phytoplankton. Compared to other species, efforts to understand the factors that influence the distribution of right whales are arguably the easiest (owing to these "simple" trophic relationships), and implementation of a prediction model would yield substantial conservation benefits by helping to reduce significant sources of human-caused mortality. With the support of several colleagues, I conducted a number of studies to better understand right whale ecology on their summer foraging grounds in the northwestern Atlantic Ocean from 1999-2001. In particular, I examined the environmental factors that were associated with spatial variability in right whale occurrence using logistic regression. Since my modeling approach was not motivated by the desire to predict right whale distribution, I did not build these models using traditional stepwise or "best subsets" methods. Instead, I constructed the models to test hypotheses about associations between right whales and specific environmental variables after accounting for factors such as sea state (affecting detectability) and interannual variability in right whale occurrence. The results suggest that right whale distribution in this region is not associated with easily-measured environmental variables (e.g., remotely-sensed products), but is instead linked to particular circulation and hydrographic features that control the availability of their prey.

Baumgartner, M.F. and B.R. Mate. 2003. North Atlantic right whale habitat inferred from satellite telemetry. Right Whale Consortium Meeting. New Bedford, Massachusetts. November 4 - 5, 2003.
Satellite-monitored radio tags were attached to North Atlantic right whales (Eubalaena glacialis) in Grand Manan Basin of the lower Bay of Fundy during the summer and early fall seasons of 1989-1991 and 2000. Monte Carlo tests were used to examine the distribution of the tagged whales in space and with respect to a variety of environmental variables to characterize right whale habitat on their northern feeding grounds. These environmental variables included depth, depth gradient, climatological surface and bottom hydrographic properties and remotely-sensed surface temperature, chlorophyll and their respective horizontal gradients. When the tagged animals left the Bay of Fundy, they did not frequently visit the deep basins of the Gulf of Maine and Scotian Shelf where high abundances of their primary copepod prey, Calanus finmarchicus, are thought to exist. Instead, right whales visited areas characterized by low bottom water temperatures, low surface salinity and high surface stratification. No evidence was found to suggest that the tagged right whales associated with ocean fronts or regions with high standing stocks of phytoplankton.

Wiebe, P.H., M.C. Benfield, C.H. Greene, A.C. Lavery, M.F. Baumgartner, N. Copley, D. Mountain, and G.L. Lawson. 2003. Spatial and temporal variation in the hydrography and plankton distributions in the Gulf of Maine during autumns of 1997, 1998, and 1999. International Council for the Exploration of the Sea. Annual Science Conference. Tallinn, Estonia. September 24 - 27, 2003.
A series of cruises were conducted in the autumns of 1997, 1998, and 1999 to survey diapausing populations of Calanus finmarchicus and their predators in Wilkinson, Jordan, and Georges Basins (Gulf of Maine) as part of the U.S. GLOBEC Georges Bank program. The sampling was done with the BIO-Optical Multi-frequency Acoustical and Physical Environmental Recorder or BIOMAPER-II, a towed system consisting of a multi-frequency sonar, a Video Plankton Recorder (VPR), a bio-optical sensor package, and an environmental sensor package (CTD). It was towed along survey track-lines in each of the basins to collect acoustic data, video images, and environmental data between the surface and bottom. In addition, a 1-m² MOCNESS was towed obliquely from near-bottom to the surface at stations in each basin, sampling eight depths intervals for biomass, taxonomic analyses, and species counts.

During the three years, a dramatic change in the hydrography took place in the Gulf of Maine that lagged by about two years the precipitous drop in the NAO index that occurred in 1996. Colder and fresher water of Labrador Sea origin was present throughout much of 1998. Coincidently, the autumn diapausing C. finmarchicus abundance was much lower and there were also extraordinary numbers of a large predatory siphonophore. Despite the lower Calanus abundance in 1998, overall zooplankton biomass levels were comparable between the three years. This comprehensive data set may enable the issue of far-field (principally NAO driven changes in hydrography) and near-field effects (predator/prey relationships) on the plankton community in the Gulf of Maine basins to be evaluated.

Baumgartner, M.F. 2003. Comparisons of Calanus finmarchicus fifth copepodite abundance estimates from nets and an optical plankton counter. 3rd International Zooplankton Production Symposium. Gijon, Spain. May 20 - 23, 2003.
The response of an optical plankton counter (OPC) to concentrations of Calanus finmarchicus fifth copepodites (C5) ranging from 2 to 1621 copepods m^-3 was examined during the summers of 1999-2001 over the continental shelf of the northwest Atlantic Ocean. Net tows from either a bongo or MOCNESS were collocated with vertical OPC casts to provide comparable data. OPC-derived particle abundance in the 1.5 to 2.0 mm equivalent circular diameter range was strongly correlated with net-derived abundance of C. finmarchicus C5 (r² = 0.655 and 0.726 for comparisons in two independent datasets). Particle abundance in this size range increased with increases in the descent speed of the vertically profiled OPC, which indicated avoidance of the small sampling aperture by C. finmarchicus C5. A regression model was developed to relate OPC particle abundance in the 1.5-2.0 mm size range to the abundance of C. finmarchicus C5 and the descent speed of the OPC. The model fit the data well (r² = 0.684) and the inverted model was used as a calibration equation to predict C. finmarchicus C5 abundances from OPC measurements in an independent comparison to net abundances. In that case, the calibration equation underestimated net abundance by an average factor of 2. However, anomalously low OPC particle abundances for some casts suggest spatial heterogeneity (patchiness) can confound such comparisons.

Baumgartner, M.F., T.V.N. Cole, P.J. Clapham and B.R. Mate. 2002. Right whale habitat in the lower Bay of Fundy and Roseway Basin: Why might a right whale be in this particular place at this particular time? Right Whale Consortium Meeting. New Bedford, Massachusetts. October 29 - 30, 2002.
Simultaneous oceanographic and visual surveys were conducted aboard the NOAA Ship Delaware II during the summers of 1999-2001 in the lower Bay of Fundy and Roseway Basin to investigate environmental factors associated with right whale occurrence. Logistic regression analysis was used to examine right whale sighting probability with respect to depth, depth gradient, hydrographic conditions, remotely-sensed sea surface temperature and surface chlorophyll and the distribution and abundance of their primary prey, the fifth copepodite stage of Calanus finmarchicus. Right whales frequented areas with greater depths and thick bottom mixed layers, and there was some evidence that right whales visited ocean fronts in Roseway Basin. We suggest that closed circulation in the lower Bay of Fundy aggregates C. finmarchicus over the deepest portions of Grand Manan Basin, while a thick bottom mixed layer causes the C. finmarchicus that aggregate above it to occur shallower in the water column. Right whales foraging on these shallower layers of C. finmarchicus can do so longer than on deeper layers, and thus ingestion rates can be increased. Ocean fronts in Roseway Basin may provide an aggregative mechanism similar to the closed circulation found in the lower Bay of Fundy.

Durbin, E., G. Teegarden, R. Campbell, A. Cembella and M. Baumgartner. 2002. North Atlantic right whales, Eubalaena glacialis, exposed to Alexandrium spp. PSP toxins via a zooplankton vector, Calanus finmarchicus. 10th International Conference on Harmful Algae. St. Petersburg, Florida, USA. October 21 - 25, 2002.
The PSP toxin-producing dinoflagellate, Alexandrium spp., regularly blooms each summer in the Grand Manan Basin of the lower Bay of Fundy. The dominant copepod in this region is the large copepod Calanus finmarchicus. During a study the summer of 2001, a large deep-resting population of C. finmarchicus was present, along with a smaller actively feeding surface population. The surface population showed a day-night feeding periodicity that was reflected in a 2-3 fold higher level of PSP toxins in the copepods at night. Despite the fact that the deeper-resting population was not feeding, these copepods still retained toxins at about half the daytime concentrations found in the surface population. On a wet weight basis, total PSP toxin concentrations in copepods at depth were about one-third of the maximum limit considered safe for human consumption when present in shellfish (800 µg saxitoxin equivalents per kg). Approximately half of the western north Atlantic right whale population feeds each summer in the Grand Manan Basin. Mature stages of C. finmarchicus are a preferred food of the right whales. During our summer study, the right whales were feeding at depth on the resting C. finmarchicus population. Via this vector mechanism, there may be both acute and chronic effects of PSP toxicity on the behavior and health of right whales.

Durbin, E.G., G.J. Teegarden, R.G. Campbell, A.D. Cembella and M.F. Baumgartner. 2002. North Atlantic right whales, Eubalaena glacialis, exposed to Alexandrium spp. PSP toxins via a zooplankton vector, Calanus finmarchicus. ASLO Summer Meeting. American Society of Limnology and Oceanography. Victoria, British Columbia, Canada. June 10 - 14, 2002.
The large copepod, Calanus finmarchicus, feeds upon PSP toxin producing dinoflagellates, Alexandrium spp., in both the laboratory and in the field. Although efficiency of toxin retention is low, total levels of toxin body burden accumulated can be high. A summer field study in the Grand Manan Basin, lower Bay of Fundy, where Alexandrium spp. regularly bloom, has shown a Calanus spp. population with a split vertical distribution, comprising a small surface population and a larger deep resting population. In the vicinity of feeding right whales, copepods from both surface and deep pools were collected by MOCNESS tows, and were analyzed for PSP toxins with HPLC-FD. Both surface and deep samples contained significant levels of toxin. A significant proportion of the western North Atlantic right whale population feeds each summer in the Grand Manan Basin. Older stages of Calanus are a preferred food source of the right whales. Estimates of exposure of right whales to toxin will be presented.

Baumgartner, M.F. and B.R. Mate. 2002. The foraging ecology of North Atlantic right whales (Eubalaena glacialis) in the Lower Bay of Fundy. EOS, Transactions, AGU 83(4), Ocean Sciences Meeting Supplement, Abstract OS41J-10, 2002. Ocean Sciences Meeting. American Geophysical Union and American Society of Limnology and Oceanography. Honolulu, Hawaii. February 11-15, 2002.
The diving behavior of right whales (Eubalaena glacialis) was monitored with recoverable, suction-cup mounted, time-depth recorders (TDRs) in the lower Bay of Fundy during the summers of 2000 and 2001. The whales were tracked and the TDRs were recovered with the aid of radio and acoustic transmitters incorporated in the tag. Vertical profiles of temperature, salinity and particle abundance were collected with a conductivity-temperature-depth (CTD) instrument and an optical plankton counter (OPC) at the location of each surfacing of a tagged whale. The lower Bay of Fundy contains low zooplankton diversity and the vertical distribution of late stage Calanus finmarchicus, the primary prey of right whales in this area, is readily observable with the OPC. In 2000, 28 tags were deployed and of the 26 that were subsequently recovered, 31% (n = 8) remained attached for over 30 minutes and 23% (n = 6) remained attached for over one hour. In 2001, 25 tags were deployed and recovered of which 76% (n = 19) remained attached for over 30 minutes and 64% (n = 16) remained attached for over one hour. Dive characteristics and collocated environmental measurements were averaged over entire TDR deployments to achieve independent statistics for each individual right whale. Foraging right whales descended quickly (1.5-2.2 m s^-1) from the surface to a particular depth between 95 and 145 m and stayed within 2-8 m of that depth for 7-11 min before returning rapidly (1.7-2.4 m s^-1) to the surface. Average foraging dive depths were strongly correlated (n = 20, r = 0.918, p < 0.0001) and coincident (p > 0.05 for Ho: slope = 1 and Ho: intercept = 0) with the average depth of maximum OPC-detected Calanus finmarchicus abundance. Average foraging dive depths were also strongly correlated with the average depth of the bottom mixed layer (n = 20, r = 0.895, p < 0.0001), however dive depths were typically shallower than the bottom mixed layer by an average of 10 m. These results suggest that right whales forage just above the bottom mixed layer where the maximum abundance of late stage Calanus finmarchicus is consistently found. By directly and simultaneously measuring the physical environment, availability of prey and predator behavior, we have gained unique insights about the ecological conditions that facilitate the location and exploitation of prey by a marine apex predator.

Baumgartner, M.F. and B.R. Mate. 2001. Understanding the relationship between North Atlantic right whale movements and habitat characteristics from satellite-monitored radio tag data: A novel approach. 14th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Vancouver, British Columbia, Canada. November 29 - December 3, 2001.
Quantifying movements for animals tracked with satellite-monitored radio tags has been hampered by location errors and uneven sampling in time. We have developed a method to estimate site fidelity from satellite tracks that accounts for these difficulties. Assuming an animalís maximum swimming speed is known or can be reasonably estimated, movement between two satellite-acquired locations can be bounded by an elliptical perimeter beyond which the animal can not swim. The proportion of overlap between two successive boundary ellipses provides an estimate of site fidelity. This proportion is termed the site fidelity index (SFI) and it ranges from 0 (low site fidelity or "traveling" behavior) to 1 (high site fidelity or large-scale "milling" behavior). Errors in the SFI are due primarily to disparate time intervals between successive movements, but the satellite track can often be subsampled in time to minimize these errors. The scales of movement for cetaceans swimming at maximum speeds over time scales of 12 or more hours will be sufficiently large to minimize errors in the SFI due to location uncertainties. North Atlantic right whales (Eubalaena glacialis) were tagged with satellite-monitored radio tags in 1989, 1990, 1991 and 2000 and their movements over time scales of 18 to 30 hours were examined with respect to physiographic and oceanographic conditions. Strong site fidelity behavior was observed in association with a particular water mass feature found in a regional, summertime climatology of bottom temperature and salinity. This hydrographic feature presumably affects the distribution and abundance of right whale prey (principally the copepod Calanus finmarchicus), and it occurs in the lower Bay of Fundy, southwestern Scotian shelf, Maine Coastal Current and the Great South Channel but is absent from the deep basins of the Gulf of Maine and Scotian shelf.

Mate, B.R. and M.F. Baumgartner. 2001. Summer feeding season movements and fall migration of North Atlantic right whales from satellite-monitored radio tags. 14th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Vancouver, British Columbia, Canada. November 29 - December 3, 2001.
During July/August 2000, we instrumented 16 North Atlantic right whales in the Bay of Fundy with Argos (satellite-monitored) radio tags coated with a long-dispersant antibiotic. Our immediate objective was to identify oceanographic correlates of right whale summer feeding habitats, and ultimately, fall migration routes to as yet undiscovered wintering areas. We heard from 12 whales (75%), 9 (56%) of which provided locations for >5 days (mean = 43.4 d), accounting for a total of at least 16,132 km of travel (mean = 1,792 km). The longest track (6,505 km in 126 days) was of an adult female which migrated 1,928 km in 23 days (mean = 3.5 km h^-1) from 40 km west of Brown's Bank to Georgia, where she was sighted by other researchers using Argos location data during an aerial survey. All whales tracked for more than 23 days left the Bay of Fundy at least once, traveling broadly over the central to southwestern Scotian Shelf and/or throughout the Gulf of Maine. Average speed was 1.8 km h^-1. Most locations were over shallow shelf waters (88% in <200 m depth) and 75% were <30 km from land (including islands). Examination of recent photos taken by other researchers for inclusion in the Right Whale Consortium catalog revealed that one whale had lost its tag, and 6 of 7 others had either a broken antenna or saltwater switch. Analysis and re-design of those elements have been completed. Tagged right whales have been resighted with tags still attached for up to eight months, suggesting that long-term tracking is feasible. Despite our antenna and saltwater switch problems, tags lasting >5 days comprised a higher percentage of tags applied in 2000 (56%) than for earlier (1989-91) surface-mounted tags (47%), and transmitter operation averaged 226% longer than earlier designs.

Baumgartner, M.F. 2001. The utility of GIS in characterizing marine mammal distribution and habitat from survey data. Workshop on the Use of Geomatic Technologies for Marine Mammal Scientists. The Society for Marine Mammalogy. Vancouver, British Columbia, Canada. November 28, 2001.
Geographic information systems (GIS) are relatively low cost, easy-to-use software packages that are being used widely in both terrestrial and marine ecology. Because of the increasing popularity of GIS, it seems prudent to critically evaluate its capabilities in the context of the challenges inherent in characterizing marine mammal distribution and habitat. Modern GIS provides several capabilities that may be particularly useful in distribution and habitat studies, including (1) data management, (2) data processing, (3) dynamic mapping, (4) data exploration and visualization, (5) "hypothesis testing" or "what if" scenarios and (6) modeling and prediction. Large studies or long-term programs can certainly benefit from the data management and dynamic mapping capabilities of GIS. Incorporation of data into a GIS is rarely a trivial task, but these programs can invest in developing rigorous data ingestion procedures and protocols so that their large databases can be easily updated, managed, accessed and displayed using GIS software. Dynamic mapping and visualization are clearly the most powerful capabilities of GIS and environmental data and model results can be easily explored using GIS.

GIS also provides very useful tools for processing spatially referenced, two-dimensional data (e.g., slope, distance and area computations, projection transformations), however, today's GIS software cannot handle three- or four-dimensional (i.e., latitude, longitude, depth, time) oceanographic datasets. While it is tempting to reduce these multi-dimensional datasets to conform to a two-dimensional GIS, habitat studies must be guided by an understanding of the relevant spatial and temporal scales of variability in the environment. Therefore, reducing the dimensionality of some datasets cannot be a matter of convenience, but instead must be justified on oceanographic or ecological grounds. Dynamic mapping, data exploration/visualization and "hypothesis testing" are of limited value when working with survey data because the distribution of sightings (or acoustic contacts) is biased by the distribution of effort expended to obtain those sightings. Raster grids of sightings per unit effort are frequently used to overcome this challenge, but a sensible representation of relative abundance using this technique requires a substantial amount of survey effort that is evenly distributed over the study area. GIS packages provide few statistical tools that are useful for the analysis of survey data, making exportation of the data to a separate statistical or analysis package necessary.

There are excellent alternatives to GIS that can be equally or more effective with survey data. These include analytical software packages such as Matlab, IDL or PV-Wave. For data storage, scientific data formats such as NetCDF are capable of handling multidimensional oceanographic data easily, have excellent metadata capabilities, are platform independent and can be used with public domain mapping software such as Generic Mapping Tools. For an experienced user, these software packages and data storage formats can replicate the functionality of a GIS and can be used in all phases of research, including data processing and management, visualization, quantitative and statistical analysis and creation of publication-quality graphics. Although these tools require some programming skills and their learning curve is typically longer than for GIS, investment in them will pay dividends by providing the flexibility to design custom analyses of the complex datasets that are typical of marine mammal distribution and habitat studies.<

Baumgartner, M.F., R.G. Campbell, G.J. Teegarden and T.V.N. Cole. 2001. Right whale nighttime feeding behavior in the lower Bay of Fundy: Inferences from a study of Calanus finmarchicus diel vertical migration. Right Whale Consortium Meeting. Boston, Massachusetts. October 25 - 26, 2001.
Recent observations of right whale diving and foraging behavior in the lower Bay of Fundy suggest that daytime feeding occurs on relatively deep layers of Calanus finmarchicus (100-200 m). Migration to the surface at night by these layers of C. finmarchicus and subsequent surface feeding by right whales may (1) increase the risk of collisions between right whales and ships and (2) increase right whales' exposure to natural saxitoxins that may be accumulating in C. finmarchicus which are actively feeding in surface waters where toxic dinoflagellates (Alexandrium spp.) are abundant. We examined the diel vertical migration of C. finmarchicus from the NOAA Ship Albatross IV at two oceanographic stations in Grand Manan Basin beginning on July 29 and July 31, 2001. The vertical distribution of C. finmarchicus was measured with an optical plankton counter (OPC) every half hour over a 28 hour period and depth-stratified plankton samples were collected with a multiple opening-closing net (MOCNESS) every 6 or 12 hours. The OPC data suggest that the highest abundances of C. finmarchicus occurred below 100 m in layers that did not migrate to the surface at night. However, the OPC detected a layer of C. finmarchicus that resided at approximately 50 m by day, migrated to the surface at sunset and returned to depth again at sunrise. Preliminary observations from MOCNESS samples, egg production experiments and feeding experiments indicate that this nighttime surface layer was an actively feeding and reproducing population consisting of nauplii, all copepodite stages, adult males and adult females. In contrast to the surface layer, the abundant deeper layers consisted predominantly of less active stage 5 copepodites with well-developed oil sacs. Despite the accessibility of the C. finmarchicus surface layer at night, we think it is likely that right whales continue to feed in the deeper layers throughout both the day and night because these layers contain much higher concentrations of the most lipid-rich copepodite stage of C. finmarchicus.

Baumgartner, M.F. and B.R. Mate. 2000. Right whale diving and foraging behavior. Right Whale Consortium Meeting. Boston, Massachusetts. October 26 - 27, 2000.
The diving behavior of the North Atlantic right whale (Eubalaena glacialis) was examined with respect to the vertical distribution of their copepod prey in the lower Bay of Fundy during July and August, 2000. Suction-cup mounted time-depth recorders (TDR's) were deployed on right whales from a rigid-hulled inflatable boat using a telescoping pole. A VHF radio transmitter was incorporated in the TDR tag design to facilitate tracking the whale and recovering the tag after detachment. An optical plankton counter (OPC) and conductivity/temperature/depth (CTD) instrument were lowered from the NOAA Ship Delaware II at the location where the tagged whale resurfaced after each long dive. While the TDR provided dive profiles, the OPC and CTD furnished data on the prey field and oceanographic conditions through which the tagged animal passed while foraging. The OPC data will be "calibrated" with zooplankton samples collected during bongo net tows to estimate the abundance and vertical distribution of several copepod taxa, including the presumed primary prey of right whales in the lower Bay of Fundy, Calanus finmarchicus. Sixty six OPC/CTD casts were conducted during 28 TDR tagging events. Preliminary results suggest that right whales dive to and forage at the depth of maximum C. finmarchicus abundance during the daytime. Continuing analysis of these data will focus on the relationships between right whale diving behavior, copepod distribution and abundance and physical oceanographic features such as water mass types, vertical density gradients and bottom mixed layer thickness.

Baumgartner, M.F. and K.D. Mullin. Cetacean habitat in the northern Gulf of Mexico. 13th Biennial Conference on the Biology of Marine Mammals. The Society for Marine Mammalogy. Wailea, Maui, Hawaii. November 28 - December 3, 1999.
Marine mammal surveys were conducted by the U.S. National Marine Fisheries Service in the northern Gulf of Mexico during the spring seasons of 1992, 1993 and 1994 to determine the distribution, abundance and habitat of oceanic cetaceans. The distribution of Risso's dolphins (Grampus griseus), Kogia species (pygmy [Kogia breviceps] and dwarf sperm whales [Kogia simus]), pantropical spotted dolphins (Stenella attenuata) and sperm whales (Physeter macrocephalus) were examined with respect to depth, depth gradient, surface temperature, surface temperature variability, the depth of the 15 degree Celcius isotherm, surface chlorophyll concentration and epipelagic zooplankton biomass. Risso's dolphins and the Kogia species were predominantly found over the upper continental slope (200 - 1000 m), but Risso's dolphins were found in regions with high depth gradients (steep sea floor slope) whereas the Kogia species were encountered more frequently in regions with high zooplankton biomass. The pantropical spotted dolphin and sperm whale were both found predominantly in the deep Gulf (> 1000 m), but sperm whales were absent from warm-core oceanographic features characterized by deep occurrences of the 15 degree isotherm. Although the geographic ranges of these species overlap, a multivariate linear discriminant analysis suggests that the distribution of sperm whales with respect to the environmental variables was considerably different than that of the other oceanic cetaceans.

Baumgartner, M.F. 1999. Characterizing right whale habitat on the Scotian shelf and in the lower Bay of Fundy. Right Whale Consortium Meeting. Boston, Massachusetts. October 21 - 22, 1999.
Oceanographic sampling was conducted aboard the NOAA Ship Delaware II during the National Marine Fisheries Service's 1999 large whale assessment cruise from July 26 to September 3. The primary objective of this sampling was to characterize the habitat of the North Atlantic right whale (Eubalaena glacialis) on the Scotian Shelf using both physical and biological oceanographic measurements. Because of the paucity of whale sightings on the Scotian Shelf, a total of 3 days were devoted to surveying and sampling the lower Bay of Fundy where a significant fraction of the North Atlantic right whale population was encountered. Discrete sampling was conducted with a conductivity/temperature/depth (CTD) instrument, an optical plankton counter (OPC) and bongo nets. Continuous measurements were made along the survey transect with an acoustic doppler current profiler (ADCP), scientific echo sounder, thermosalinograph and a flow-through fluorometer. Synoptic observations of sea surface temperature and chlorophyll were also collected during the cruise by the satellite-borne AVHRR and SeaWifs instruments, respectively. Twenty seven oblique bongo tows and 157 vertical casts with the CTD/OPC instrument package were completed during both systematic sampling and localized sampling near right whales. Oceanographic data were collected near 6 groups of right whales in the Bay of Fundy and a single animal in Roseway Basin on the Scotian Shelf.

Trask, R., W. Paul, M.F. Baumgartner. 1998. Design characteristics and field experience with a near surface horizontal moored array. Office of Naval Research and Marine Technology Society 1998 Buoy Workshop. San Diego, CA. April 2, 1998.
Time series observations in three dimensions and on scales of meters to one kilometer are needed in order to understand the structure and variability of the upper ocean and the dynamics that govern the evolution of that structure. Conventional moorings provide platforms on which arrays of sensors can be deployed with good vertical resolution (down to approximately 1 meter). Multiple conventional moorings can be deployed but the closest horizontal spacings used in such arrays is typically the water depth in order to minimize tangling. As a result, measurements with horizontal separations of less than the water depth have been inaccessible to those wishing to field near surface arrays of mooredinstrumentation.

As a step toward developing a three dimensional moored array capability, an instrumented 100 meter long horizontal mooring was successfully deployed during the summer of 1997. Measurements of velocity, temperature, and pressure were made along the horizontal element which was moored at 20 meters depth in 100 meters of water. The design of the horizontal moored array will be presented along with results from several tests conducted on the anchoring system. The technique used to deploy the horizontal mooring will be discussed and the performance of the array will be presented. Plans for a two dimensional array to be deployed in 1998 will also be described.

Baumgartner, M.F. and S.P. Anderson. 1998. Comparison of atmospheric model products with in situ observations during the Coastal Mixing and Optics Experiment. National Centers for Environmental Prediction Seminar. Camp Springs, MD. February 20, 1998.
An important objective of the ONR Coastal Mixing and Optics (CMO) experiment is to identify and understand the oceanic mixing processes that influence the evolution of the vertical temperature stratification on the continental shelf. A moored array of buoys deployed in the Mid-Atlantic Bight from August, 1996 to June, 1997 collected high quality meteorological and oceanographic measurements to characterize the local atmospheric forcing and ocean response. Surface numerical weather prediction (NWP) model products from the Eta (80km and 29km) and Rapid Update Cycle (60km) models were also acquired to characterize the large-scale atmospheric forcing during the experiment. With these data in hand, an evaluation of the models' surface meteorology and air-sea heat, fresh water and momentum fluxes is possible with the goal of developing accurate flux fields to drive an ocean model. The NWP model products are compared to in situ meteorological observations from the CMO site and at six National Data Buoy Center buoys in the Mid-Atlantic Bight. Model air-sea heat fluxes are also compared to estimates derived from the CMO buoy meteorology using a state-of-the-art bulk flux algorithm. Strategies for improving flux fields derived from NWP model meteorology are discussed.

Weller, R.A., M.F. Baumgartner, A.S. Fischer. 1996. Predicting upper ocean variability in the Arabian Sea with a one-dimensional model. American Geophysical Union Fall Meeting. San Francisco, California. December 15-19, 1996.
Local atmospheric forcing and upper ocean velocity, temperature, and salinity were observed from October 1994 to October 1995 in the Arabian Sea at a site (15.5N, 61.5E) near the axis of the Findlater Jet. Winter monsoon winds were typically 5 to 7 m s^-1, while summer monsoon winds were between 10 and 16 m s^-1. Daily mean net heat oceanic heat loss was as strong as 200 W m^-2 during the NE monsoon. There was little heat lost through the sea surface during the SW monsoon; the largest oceanic heat gain of the year, in excess of 200 W m^-2, was observed in August during the later half of the summer monsoon. The mixed layer deepened and sea surface temperature cooled during both the winter and summer monsoons. Surface salinity reached a local maximum in June.

Large changes in temperature and salinity were observed in the upper 150 m, both in and below the base of the mixed layer, that were well correlated with energetic velocity variability over the same depth range. These signals are associated with large eddies or other mesoscale flow features. At times, these velocities were large; in October 1994, before the onset of the NE monsoon, the surface flow exceeded 70 cm s^-1.

A one-dimensional model of the upper ocean is used to examine the locally-forced component of the observed temperature, salinity, and velocity variability. Sensitivity of the model results to the initialization of the temperature and salinity profiles, the role of advection in altering the stability in the upper ocean, the effect of wind-driven vertical flows, and the influence of background shear on local mixing are examined. Closure of the local heat and salt budgets are used to diagnose the need for additional vertical mixing and horizontal advection, and attempts are made to paramaterize these processes in the model.

May, L.N. Jr, T.D. Leming and M.F. Baumgartner. 1996. Remote sensing and Geographic Information System support for the Gulf Cetacean (GULFCET) Project: A description of a potentially useful GIS system for ichthyoplankton studies in the Gulf of Mexico. Meeting of the Ad Hoc GFCM/ICCAT Joint Working Group on Large Pelagic Fish Stocks in the Mediterranean Sea. Genoa, Italy. September 9 - 11, 1996.
Details are given of the Geographic Information System used to integrate marine mammal sightings, oceanographic, bottom depth, ship and aircraft survey tracklines and other collateral data for the Gulf of Mexico Cetacean Project (GULFCET).