Table of Contents Abstract Introduction Data and Methods Results References Cited Appendix
Northeast Fisheries Science Center Reference Document 12-20
Description of the 2011 Oceanographic Conditions on the Northeast U.S. Continental Shelfby Paula S. Fratantoni, Tamara Holzwarth-Davis, Cristina Bascuñán, and Maureen H. Taylor
National Oceanic Atmospheric Administration, National Marine Fisheries Service, Northeast Fisheries Science Center, 166 Water Street, Woods Hole, MA, 02543 USA
Web version posted September 14, 2012Citation: Fratantoni PS, Holzwarth-Davis T, Bascuñán C, Taylor MH. 2012. Description of the 2011 Oceanographic Conditions on the Northeast U.S. Continental Shelf. US Dept Commer, Northeast Fish Sci Cent Ref Doc. 12-20; 35 p. Available from: National Marine Fisheries Service, 166 Water Street, Woods Hole, MA 02543-1026, or online at http://www.nefsc.noaa.gov/nefsc/publications/
Information Quality Act Compliance: In accordance with section 515 of Public Law 106-554, the Northeast Fisheries Science Center completed both technical and policy reviews for this report. These predissemination reviews are on file at the NEFSC Editorial Office.
Hydrographic observations from eleven surveys spanning the Northeast U.S. Continental Shelf are combined into a descriptive overview of the broad-scale oceanographic conditions that were observed during 2011. Temperature and salinity observations are combined into six 2-month time periods in order to maximize both the spatial coverage of the data and its temporal resolution during the year. Maps of near-surface and near-bottom property distributions are presented for each bi-monthly period and time series of regional average properties are discussed for five geographic regions spanning the shelf: western Gulf of Maine (GOMW), eastern Gulf of Maine (GOME), Georges Bank (GBNK), and northern and southern Middle Atlantic Bight (MABN and MABS, respectively). Surface conditions along the entire Northeast U.S. Continental Shelf were generally warm and very fresh in 2011 relative to the reference period (1977-1987). Surface warming was largest in the spring in the southern Middle Atlantic Bight and in late summer elsewhere. Freshening was greatest in the surface waters near the coast in the western Gulf of Maine and southern MAB during periods of maximum spring discharge, suggesting some influence from local fresh water sources in these regions. A later period of enhanced freshening also appears to align with a second large peak in precipitation over the Northeastern United States. The surface freshening observed over Georges Bank and throughout the MAB penetrates to the bottom, while bottom waters in the Northeast Channel and deep basins of the GoM were saltier than normal reflective of slope water influences in these regions. A particularly large Gulf Stream meander was responsible for extraordinarily warm and salty conditions observed throughout the water column on the New England Shelf between October-December 2011.
The Northeast Fisheries Science Center (NEFSC) conducts multiple surveys on the Northeast U.S. continental shelf each year in support of its ongoing mission to monitor the shelf ecosystem and assess how its components influence the distribution, abundance, and productivity of living marine resources. In support of this mission, the Oceanography Branch provides conductivity, temperature and depth (CTD) instruments to all NEFSC cruises for the measurement of water column profiles of temperature and salinity. In addition to providing oceanographic context to specific field programs, these data contribute to a growing database of historical measurements that are used to monitor seasonal and interannual variability in the water properties on the northeast continental shelf.
Broad-scale surveys, sampling the shelf from Cape Hatteras, North Carolina, into the Gulf of Maine, are conducted up to six times per year during shelf-wide spring and fall bottom trawl surveys and typically on four dedicated seasonal Ecosystem Monitoring (EcoMon) surveys. Profiles of conductivity, temperature and depth are collected at each station on these shelf-wide surveys. Observations are also collected on other more regionally focused NEFSC surveys, where station coverage varies depending on the objectives of the particular field program. During 2011, hydrographic data were collected on 11 individual NEFSC cruises, amounting to 1,840 profiles of temperature and salinity (Table 1). Here we present an annual summary of these observations, including surface and bottom distributions of temperature and salinity, as well as their anomalies relative to a consistent reference period. In addition, regional average values of temperature and salinity and their anomalies are computed for five different regions of the shelf during six bi-monthly periods. Finally, the volume and properties of the shelf water are examined for the Middle Atlantic Bight region.
The Oceanography Branch provides CTD instrumentation and support to all NEFSC programs requesting this service. Training in instrument maintenance and operation, including deployment, data acquisition, recovery and preliminary processing, is provided as needed prior to sailing. On NEFSC surveys, CTD instruments are typically deployed in one of two modes: (1) during a bongo net tow, the CTD instrument is mounted on the conducting wire above the bongo frame and data are collected as a double oblique profile with the ship steaming at approximately 2 knots, (2) during a non-net tow, the CTD is mounted vertically on the wire and the sensors are soaked for one minute at the surface prior to descent. The sensors are not soaked at the surface prior to descent during bongo tows, rendering the upper 30 meters or more of the downcast unreliable. For this reason, the up-cast profile data are routinely processed as the primary data for each station.
In 2011, hydrographic data were collected aboard the NOAA ships Delaware II and HB Bigelow, and the R/V HR Sharp using a combination of Seabird Electronics SBE-19 and SBE-19+ SEACAT profilers and SBE 9/11 CTD units (Table 1). All raw CTD profile data were processed ashore, using standard Seabird Electronics software to produce 1-decibar averaged profiles in ascii-formatted files. Water samples were collected twice daily at sea during vertical casts. Following each cruise, these samples were analyzed using a Guildline AutoSal laboratory salinometer to provide quality control for the CTD salinity data. A salinity offset was applied to instrument data if the mean difference between the reference Autosal readings and the CTD values exceeded +/- 0.01 (a threshold chosen based on the expected instrument accuracy.) Vertical density profiles were examined for inversions due to bad conductivity or temperature readings and/or sensor misalignment. Egregious cases were replaced with a flag value. The processed hydrographic data were loaded into ORACLE database tables and made publically available via anonymous ftp (ftp://ftp.nefsc.noaa.gov/pub/hydro). Cruise reports have been prepared for each survey listed in Table 1 and are available online (http://www.nefsc.noaa.gov/epd/ocean/MainPage). Readers are referred to the individual cruise reports for notes, property maps and aggregate data specific to a particular survey.
Here, we aim to provide a descriptive overview of the hydrographic sampling that was conducted in 2011 and to characterize the broad-scale oceanographic conditions that were observed. In order to maximize both the spatial coverage of the data and its temporal resolution, the processed 2011 CTD data have been sorted into six 2-month time bins. Maps of near-surface and near-bottom temperature and salinity have been produced from profile data falling within each 2-month period. Surface fields include the shallowest observed temperature/salinity at each station that is also in the upper 5 meters of the water column, while bottom maps include the deepest observation at each station that also falls within 10 meters of the reported water depth. In order to examine the spatial and temporal variability over broader areas of the shelf, average values have been computed from the data within five sub-regions spanning the Northeast U.S. Continental Shelf (Figure 1). Regional averages have been computed for the bi-monthly binned fields (Table 2 and Table 3) and for individual cruises (Appendix Tables 1-5).
In order to characterize variability that is not related to seasonal forcing, anomalies have been calculated at each station relative to a standard reference period (1977-1987). During this period the NMFS Marine Resources Monitoring and Prediction (MARMAP) program repeatedly occupied stations spanning the entire Northeast U.S. Shelf so that an annual cycle could be constructed for water properties across all regions of the northeast shelf (Mountain et al., 2004; Mountain and Holzwarth, 1989). The anomalies presented here are defined as the difference between the observed 2011 value at individual stations and the expected value for each location and time of year based on this reference period. Similarly, regional anomalies are the area-weighted average of these anomalies within a given domain. The methods used and an explanation of uncertainties is presented in Holzwarth and Mountain (1990).
Finally, we calculate the temperature, salinity and volume of the shelf water in the Middle Atlantic Bight during 2011 and relate this to the conditions observed during the MARMAP reference period. Following Mountain (2003), the shelf water mass is defined as water within the upper 100 meters having salinity less than 34. For each survey in 2011, the area of a sub-region was apportioned among its stations by an inverse distance squared weighting. The shelf water volume at a given station is the thickness of the shelf water at the station multiplied by its apportioned area, and the total shelf water volume within the sub region is the sum of these products for all stations within the region. Similarly, the average temperature and salinity was calculated in the shelf water layer at each station and multiplied by the total shelf water volume for that station. The sum of these products over all stations within a given sub-region, divided by the total shelf water volume for the region, determines the volume-weighted average temperature and salinity. Anomalies in the property and volume of the shelf water mass are calculated relative to like variables derived from MARMAP hydrographic data, as described above. Hence, here regional anomalies are computed as the mathematical difference between regional averages, not an average of the anomalies computed for a given sub-region.
Table 1 provides a listing of the NEFSC cruises that collected hydrographic data in 2011. In total, 1840 profiles of temperature and salinity were collected, processed and archived during the year. Combining the hydrographic data from multiple cruises into bi-monthly bins improves the spatial coverage compared with that of individual surveys, enabling us to examine the spatial and temporal patterns in hydrography over the region. Nonetheless, there are still significant gaps in several of the bi-monthly distribution maps shown in Figure 2. For instance, the lack of sampling over the Northeast Shelf during July-August was a consequence of the cancellation of the August Ecosystem Monitoring (EcoMon) cruise. There were also significant gaps in station coverage in portions of the Gulf of Maine during the March-April and September-October periods. These gaps result from a misalignment between the bi-monthly periods and the longer bottom trawl surveys that work from south to north along the shelf. These periods encompass all but the final portion of the spring and fall ground fish surveys, when sampling was focused in the Gulf of Maine. Large gaps in station coverage preclude the calculation of a representative regional average surface/bottom temperature and salinity value during July/August, particularly in the Gulf of Maine (Table 2, Table 3, Figure 3, and Figure 4). While station coverage is substantially better during all other periods, not all of the average properties reported are true area-weighted averages representative of the entire region. Those cases are flagged in Table 2 and Table 3 and the reader should keep this in mind when interpreting results.
Overall, surface conditions along the entire Northeast U.S. Continental Shelf were warm and very fresh in 2011 relative to the reference period (Figure 3 and Figure 4). The warmest surface anomalies were observed in the MAB, peaking earlier in the south than the north (Figure 3). Surface temperatures in the MAB began the year colder than reference values, becoming warmer later in the year. This is consistent with long-term trends in the region, indicating that the seasonal temperature range has been increasing throughout the recent decade (Friedland and Hare, 2007). Surface waters were freshest in the southern MAB, measuring over 1.0 unit fresher than the MARMAP reference period during May-October (Figure 4). Bottom waters were warmer than normal in the Gulf of Maine with peak warming in the western region (Figure 3). By contrast, bottom waters on Georges Bank and in the northern MAB were near the reference temperature at the beginning of the year, becoming warmer through spring and fall, while temperatures remained near reference values in the southern MAB. Freshening was observed near the bottom on Georges Bank and in the Middle Atlantic Bight while bottom waters in the Gulf of Maine were not significantly different than the reference salinity at any point in the year. Compared with the reference period, the shelf water volume in the MAB was high in 2011 relative to the MARMAP period (salinity less than 34), with largest volume anomalies observed in the southern MAB. This suggests that the shelf/slope front, typically identified by the 34 isohaline, was consistently located seaward of its position during the reference period (Figure 5). On average, the shelf water was up to 0.7 units fresher over the entire MAB and the temperature was consistently warmer than the reference temperature in the north.
Details related to the temporal trends in Figure 3 and Figure 4 are explored in surface and bottom property distribution maps (Figures 6-11). Maps of surface temperature reveal the seasonal cycle of warming and cooling over the region, with warmest temperatures observed at the surface during late summer (Figures 6-11a). Even though regional averages indicate warming over most of the region relative to the MARMAP reference period, the details of this warming varies from region to region. Cooler anomalies dominate most of the shelf during January/February, although warm anomalies are observed at the surface over the Northeast Channel and Georges Bank (Figure 6b). At the onset of seasonal warming (May/June), warm anomalies dominate all regions of the shelf (Figure 8b). It should be noted that the tongue of cooler surface water extending southwestward across the MAB from Cape Cod to the shelf edge near Maryland is an artifact of combining two surveys that occupied adjacent stations at different times: the colder stations were occupied in mid-May while the surrounding warmer stations were occupied in early June (Figure 8a). This cold feature disappears when contouring temperature distributions observed during a single shelf-wide survey (Figure 8c; Table 1). The May/June anomaly map shows warming over the entire region (Figure 8b), demonstrating that our method is successful at removing temporal variations associated with the seasonal cycle.
Seasonal cooling typically begins in September/October, with colder isotherms gradually pushing southward along the shelf. Cold anomalies observed on the shelf between 38-40N during September/October are suggestive of interannual variations in this progressive cooling of surface waters (Figure 10b). Also of note, surface temperatures were up to 8C warmer (and 2 units saltier) during this period at the edge of the shelf south of Cape Cod (Figure 10b). Similarly, in November/December 2011 the sea surface was anomalously warm over Nantucket Shoals and along the shelf edge south of Cape Cod, establishing a sharp temperature gradient at this location (Figure 11). These warm anomalies were particularly strong near the bottom (Figure 11b) and collocated with the warm/salty anomalies observed in September/October (Figure 10b). The anomalous features were forced by an extraordinarily large meander in the Gulf Stream that brought extremely warm/salty water into contact with the shelf and slope in this region during two periods, early-November and December 2011 (Gawarkiewicz et al., 2012).
Maps of surface salinity show the seasonal influence of freshwater discharge in the nearshore regions, with the freshest waters appearing very near the coast in the eastern Gulf of Maine and MAB in March/April, followed by more wide-spread freshening over these regions in May/June (Figure 7a and Figure 8a). While overall surface waters were fresher than normal in 2011, the largest anomalies were observed near the coast in the vicinity of major freshwater sources (e.g., western Gulf of Maine, Hudson River, Delaware and Chesapeake Bays) during periods of maximum spring discharge (March/April and May/June; Figure 7b and Figure 8b). Curiously, a second period of fresh anomalies was also observed near these freshwater sources in September/October (Figure 10b).
Maps of near bottom salinity and salinity anomaly suggest that freshening observed at the surface penetrates to the bottom on Georges Bank and throughout the MAB (Figure 4 and Figure 6-11). However, the same is not true in the Gulf of Maine where slope water influences the lower layer properties. Near-bottom temperature and salinity anomalies in Northeast Channel and in the deep basins of the GoM suggest that the slope water was warmer and saltier throughout much of 2011 (Figures 6-11b).
Maps of near-bottom temperature show the seasonal formation of the cold pool in the Middle Atlantic Bight, with coldest temperatures observed during the May/June period (Figure 8a). It must be recognized that the shape and magnitude of the cold pool as seen in this composite map may be influenced by the same temporal discontinuity between adjacent stations that led to artifacts in the surface fields. In fact, the position of the cold pool in Figure 8a lies directly underneath this artificial surface feature. However, it is reassuring that the cold pool feature is also evident in bottom maps constructed from the single June EcoMon Survey (Figure 8c; Table 1). The accompanying maps of near-bottom temperature anomaly suggest that temperatures at the core of the temperature minimum were slightly warmer than normal, while temperatures in the southern tail of the feature were slightly colder than normal (Figure 8b and Figure 8d). By September/October the cold pool had begun to weaken and warm (Figure 10a). Cold anomalies near the core are indicative of interannual variations in the erosion of this feature (Figure 10b).
The erosion of the cold pool in late summer is preceded by warming of bottom waters near the coast in September/October (compare Figure 8a and Figure 10a). This is consistent with historical observations in this region, which indicate that by late summer the effects of seasonal heating extend all the way to the ocean bottom in the near-shore regions (Castelao et al., 2010). The anomaly fields shown in Figure 10b suggest that near-shore bottom waters north of 38N were 2-3 degrees warmer in 2011 than the reference period.
Based on the climate summaries compiled by the Northeast Regional Climate Center (http://www.nrcc.cornell.edu), monthly mean air temperatures over the Northeastern U.S. were warmer than the long term mean throughout most of 2011 (referenced to 1971-2000). On average, annual mean temperatures were almost 1.5 degrees warmer than normal. The annual cycle of heating and cooling in 2011 was aligned with the long-term annual cycle but warming was enhanced during the peak in seasonal heating (3 degrees above normal) and again in late fall (over 5 degrees warmer, Figure 12). By comparison, the largest regional sea surface temperature anomalies were observed in September/October in the north and May/June in the south.
According to Northeast Regional Climate Center records, the annual mean precipitation over the Northeastern U.S. was above normal for 7 months out of 2011 (referenced to 1971-2000). Most notably, precipitation was elevated from March-May (almost 3" above normal) and again in August-October (over 4" above normal). The two periods of enhanced precipitation may account for the anomalously fresh surface conditions observed near the Hudson River, Delaware and Chesapeake Bays and along the coast in the western Gulf of Maine during the May/June and September/October time periods (Figure 8 and Figure 10). However, a more thorough analysis would be needed to discriminate between local freshening due to precipitation and river run-off and freshening due to advection from remote sources.
Castelao R, Glenn S, Schofield O. 2010. Temperature, salinity, and density variability in the central Middle Atlantic Bight. Journal of Geophysical Research, 115, C10005, doi:10.1029/2009JC006082.
Friedland K, Hare J. 2007. Long-term trends and regime shifts in sea surface temperature on the continental shelf of the northeast United States. Continental Shelf Research, 27, 2313-2328.
Gawarkiewicz G., Todd RE, Plueddemann AJ, Andres M, Manning JP, 2012. Direct interaction between the Gulf Stream and the shelfbreak south of New England. Geophysical Research Letters, submitted.
Holzwarth TJ, Mountain DG. 1990. Surface and bottom temperature distributions from the Northeast Fisheries Center spring and fall bottom trawl survey program, 1963-1987. NEFSC Reference Document 90-03, 62 pp.
Mountain DG, Holzwarth TJ. 1989. Surface and bottom temperature distribution for the Northeast continental shelf. NOAA Technical Memorandum NMFS-F/NEC-73, 55 pp. Available from: NOAA Fisheries, 166 Water Street, Woods Hole, MA 02543.
Mountain DG. 2003. Variability in the properties of shelf water in the Middle Atlantic Bight, 1977-1999. Journal of Geophysical Research, 108, 3014, doi:10.1029/2001JC001044, (14-)1-11.
Mountain DG, Taylor MH, Bascuñán C. 2004. Revised procedures for calculating regional average water properties for Northeast Fisheries Science Center cruises. NEFSC Reference Document 04-08, 53pp.
Northeast Regional Climate Center. Seasonal climate summary tables [Internet]. Cornell University. [cited 2011 March 31]. Available from: http://www.nrcc.cornell.edu/.