Sales

Propagation of temperature signals from the northwest Atlantic continental shelf edge into the Laurentian Channel

Description
Not to be cited without prior reference to the author ICES CM 2004/N:07 Propagation of temperature signals from the northwest Atlantic continental shelf edge into the Laurentian Channel Denis Gilbert,
Categories
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
Not to be cited without prior reference to the author ICES CM 2004/N:07 Propagation of temperature signals from the northwest Atlantic continental shelf edge into the Laurentian Channel Denis Gilbert, Maurice Lamontagne Institute, Dept. of Fisheries and Oceans Canada, P.O. Box 1000, Mont-Joli, Québec, Canada, G5H 3Z4, Phone : ; Fax : ; Web: Abstract The Labrador Current carries cold, low-salinity, oxygen-rich waters as far south as the tail of the Grand Banks of Newfoundland, where a major fraction of its transport gets entrained eastward in the North Atlantic Current (NAC). A time-varying fraction of the Labrador Current transport continues to flow with the 200 m isobath on its right, turning around the tail of the Grand Banks in the northwestward direction, towards the mouth of the Laurentian Channel. This Labrador Current Water (LCW) mixes with North Atlantic Central Water (NACW), and the result of this mixing becomes the Slope Water mass. The Slope Water enters the Laurentian Channel at the continental shelf edge, and travels some 1240 km landward, carrying time-varying water temperature, salinity and oxygen signals. The analysis of temperature time series from four polygons along the Laurentian Channel yields a cross-channel average propagation speed of 1 cm s -1 at 250 m depth, twice faster than previously estimated. We briefly discuss the mechanisms that might be responsible for the observed interdecadal ocean climate variability, focusing on the northward migrations of the Gulf Stream. Introduction The dominant bathymetric feature in the Gulf of St. Lawrence is the Laurentian Channel, a submarine valley with depths everywhere exceeding 250 m. The Laurentian Channel extends some 1240 km landward from the continental shelf edge to the Lower St. Lawrence Estuary (Fig. 1). As part of the estuarine circulation of the Gulf of St. Lawrence, water flows seaward in the surface layer and landward in the deeper layers (Saucier et al. 2003). The cross-channel average landward flow of the 200 m to 300 m deep layer was roughly estimated at 0.5 cm s -1 (Bugden 1991). Using this estimate, a parcel of water that enters the mouth of the Laurentian Channel at 250 m depth would reach the head of the Laurentian Channel in about 7 years. Estimating the residence time of the deep waters of the Laurentian Channel as accurately as possible is important because it is one of the key parameters in the construction of oxygen, nutrient and carbon budgets of the Estuary and Gulf of St. Lawrence. Below a depth of ~ m, the waters of the Laurentian Channel take their origin in the northwest Atlantic Slope Water region (Fig. 2). The temperature-salinityoxygen properties of the Slope Water result from the mixing of Labrador Current Water (LCW) and North Atlantic Central Water (NACW) in time- and space-varying proportions. Depending on the relative proportions of LCW and NACW in the Slope Water, we may distinguish between warm Slope Water and Labrador Slope Water (Gatien, 1976). When warmer than normal Slope Water is present at the mouth of the Laurentian Channel, this signal will propagate into the Laurentian Channel in such a way that we may expect the arrival of these warmer waters at various locations along the Channel a few years later. The main objective of this paper is to determine this time lag more precisely. Materials and methods All temperature and salinity data are from the publicly available CLIMATE database developed and maintained at the Bedford Institute of Oceanography (http://www.mar.dfo-mpo.gc.ca/science/ocean/database/data_query.html). The general study area together with the polygons for which we derived the temperature time series are shown in Fig. 1. Missing values in the annual mean time series of temperature were filled by linear interpolation (e.g. Cabot Strait in 1963, northwest Gulf in 1960). To determine the statistical significance of the cross-correlation functions as a function of the time lags, we used equation of von Storch and Zwiers (1999). Results and discussion Annual mean temperature time series at 250 m depth for the Cabot Strait and northwest Gulf of St. Lawrence polygons are presented in Fig. 3. The lagged correlation between the temperature time series at Cabot Strait (leading) and in the Northwest Gulf of St. Lawrence (lagging) is shown in Fig. 4. Our results indicate that on average, a temperature signal at Cabot Strait will reach the northwest Gulf of St. Lawrence 1 to 3 years later. The highest correlation is found at a lag of 2 years. Given the ~550 km distance separating the centers of the Cabot Strait and northwest Gulf polygons, this yields an average propagation speed of 0.9 cm s -1. This is about twice faster than Bugden s (1991) estimate. When looking at individual peaks and troughs in the time series (Fig. 3), we find that the lag between Cabot Strait and the Northwest Gulf temperature time series is not always equal to 2 years. For some peaks and troughs the lag may appear to be slightly shorter or longer than 2 years, but a 2 year-lag yields the highest correlation between the two time series. A rough attempt to look at whether the mean propagation speed has changed over time is presented in Fig. 5 where we have split our time series in two halves. The cross-correlation functions suggest a slightly slower deep estuarine circulation in the period compared with (Fig. 5), although the difference is not statistically significant. The lagged correlation between temperature time series at the mouth of the Laurentian Channel (leading series) and temperature time series in Cabot Strait, the Northwest Gulf of St. Lawrence and the Lower St. Lawrence Estuary at 250 m depth is shown in Fig. 6. Our results indicate that on average, a temperature signal originating at the mouth of the Laurentian Channel will reach Cabot Strait 1 year later, and will reach the Northwest Gulf after 3 years and the Lower Estuary after 3 to 4 years. Those results also imply a long-term average propagation speed of 0.95 cm s -1 for the temperature signals, taking 900 km as the distance separating the centers of the polygons at the mouth of the Laurentian Channel and in the Northwest Gulf for example. The correlation function between the Laurentian Channel mouth and the Estuary temperature time series quickly rises from 0.02 at zero lag to a broader peak at 3 and 4 year lags, suggesting a 3.5 year mean transit time for water parcels travelling from the mouth of the Laurentian Channel to the Lower St. Lawrence Estuary. What are the driving mechanisms of the observed interdecadal variability in temperature and salinity? One possibility involves large-scale shifts in atmosphere pressure systems such as the North Atlantic Oscillation (NAO, Hurrell 1995). The best correlation between the NAO index and temperature time series on the kg m -3 isopycnal at the four polygons along the Laurentian Channel (Fig. 1) was obtained at Cabot Strait (r = 0.53) with the NAO index leading the Cabot Strait time series by 4 years over the 1949 to 2003 period ( n = 55 years). This correlation is not, however, statistically significant. Both the NAO index and the Cabot Strait time series are strongly auto-correlated, meaning that periods of several consecutive years with values above or below the long-term mean are commonly found in the time series. The values from consecutive years are not independent, thus invalidating one of the assumptions made to test the significance of correlations in standard statistical tables. Using the method proposed by Pyper and Peterman (1998), we find that only 10 of the 55 pairs of observations are effectively independent. With 8 effective degrees of freedom, the correlation of 0.53 between the NAO and the temperature times series on the kg m - 3 at Cabot Strait is only statistically different from zero at the p = 0.12 significance level. The lack of a strong link between the NAO and the Laurentian Channel temperature time series is not surprising, as attempts to link the NAO with temperature indices between 100 m and 300 m depth further west along the edge of the Scotian Shelf have also shown (Pershing et al. 2001). Excursions in the north-south direction of the Shelf/Slope sea surface temperature (SST) front and of the north wall of the Gulf Stream (Fig. 2) may influence water properties in the Laurentian Channel, as suggested by Bugden (1991). Lagged correlations with the temperature time series at the mouth of the Laurentian Channel on the kg m -3 density surface were computed using 1973 to 2001 time series of the latitudinal position of these two SST fronts at intervals of 1 longitude from 75 W to 50 W (Drinkwater et al. 1994). We only present the results obtained with the north wall of the Gulf Stream, because the patterns of correlation are stronger than for the Shelf/Slope front. Restricting our attention to the 0 to 5-year lags, for which we could obtain 5 or more replicates with the 29 year SST front time series, the diagram of lagged correlations as a function of longitude (Fig. 7) reveals a structure that appears to be physically meaningful. For lags of 0 to 4 years and at longitudes between 65 W and 50 W, there is a very broad and consistent pattern of positive correlations with the temperature time series from the mouth of the Laurentian Channel. This is interpreted as meaning that when the Gulf Stream moves further to the north than its normal position, meanders and eddies shed by the Gulf Stream will have a higher likelihood of mixing with the Slope and Shelf Waters to the north (Fig. 2), thus promoting conditions that will generate warmer than usual waters at the mouth of the Laurentian Channel. In an attempt to generate a stronger signal to noise ratio, an average time series from 64 W to 56 W was calculated, and the lagged correlation of this composite time series of the latitudinal position of the north wall of the Gulf Stream with the Laurentian Channel Mouth temperature time series was computed. Correlations of 0.50, 0.53, 0.47 and 0.53 at lags of 1, 2, 3 and 4 years respectively were obtained, with only the 1-year lag correlation being statistically different from zero at the 95% confidence level when taking serial correlation into account (Pyper and Peterman 1998). Such poor predictability is disappointing but not surprising. The Slope Water region from the Tail of the Grand Banks to the Mid-Atlantic Bight is a region densely populated with eddies, and such chaotic motions likely contribute to the weak relationships we found between a large-scale atmospheric forcing index (the NAO) or an oceanic forcing index (position of the north wall of the Gulf Stream) and a time series of temperature at the mouth of the Laurentian Channel. Petrie and Drinkwater (1993) suggested that an index of the westward volume transport of the Labrador Current on the southern edge of the Grand Banks may be the best predictor of temperature and salinity conditions of the Laurentian Channel in the 100 m to 300 m depth range. Future work will involve looking at this possibility, taking into account the recent development of a Labrador Current volume transport index derived from Topex/Poseidon satellite altimetry (Han and Li 2004). References Bugden, G.L Changes in the temperature-salinity characteristics of the deeper waters of the Gulf of St. Lawrence over the past several decades. In J.-C. Therriault (ed.) The Gulf of St. Lawrence: small ocean or big estuary? Can. Spec. Publ. Fish. Aquat. Sci. 113, p Drinkwater, K.F., R.A. Myers, R.G. Pettipas, and T.L. Wright Climatic data for the Northwest Atlantic: The position of the shelf/slope front and the northern boundary of the Gulf Stream between 50 W and 75 W, Can. Data Rept. Fish. Ocean. Sc. No. 125, 103p. Gatien, M.G A study in the slope water region south of Halifax. J. Fish. Res. Board Canada, 33: Han, G., and J. Li Sea surface height and current variability on the Newfoundland slope from TOPEX/Poseidon altimetry. Can. Tech. Rep. Hydrogr. Ocean Sci. No. 234: viii + 40p. Hurrell, J.W Decadal trends in the North Atlantic Oscillation: regional temperatures and precipitation. Science, 269: Pershing, A.J., C.H. Greene, C.G. Hannah, D. Sameoto, E. Head, D.G. Mountain, J.W. Jossi, M.C. Benfield, P.C. Reid, and T.G. Durbin Oceanographic responses to climate in the Northwest Atlantic. Oceanography, 14: Petrie, B., and K. Drinkwater Temperature and salinity variability on the Scotian Shelf and in the Gulf of Maine J. Geophys. Res., 98: Pyper, B.J., and R.M. Peterman Comparison of methods to account for autocorrelation in correlation analyses of fish data. Can. J. Fish. Aquat. Sci., 55: Saucier, F.J., F. Roy, D. Gilbert, P. Pellerin, and H. Ritchie Modeling the formation and circulation processes of water masses and sea ice in the Gulf of St. Lawrence, Canada. J. Geophys. Res., 108(C8): Von Storch, H., and F.W. Zwiers Statistical analysis in climate research. Cambridge University Press, 494p. Fig. 1. Map of the Gulf of St. Lawrence showing its three main deep channels: the Laurentian Channel which begins at the continental shelf break and ends near 70 W, and two smaller channels in the northeast Gulf, the Anticosti and Esquiman Channels. The polygons delineate the four areas for which we have derived time series of temperature (MLC = Mouth of Laurentian Channel, CS = Cabot Strait, NW = Northwest Gulf, LSLE = Lower St. Lawrence Estuary ). Fig. 2. Map of the northwest Atlantic showing the mean position of the northern edge of the Gulf Stream (red line), the Shelf/Slope front (green line), the position of the Labrador Current (blue line), the Tail of the Grand Banks (TGB) and the Slope Water region (yellow). Fig. 3. Interannual time series of temperature at 250 m depth at Cabot Strait (red circles) and the Northwest Gulf (blue squares). The polygons for which we derived these time series are shown in Fig. 1. Fig. 4. Cross-correlation function between the 250 m temperature time series at Cabot Strait and in the Northwest Gulf over the period. The dashed line represents the null hypothesis of zero correlation at the 95% confidence level. Fig. 5. Cross-correlation functions between the 250 m temperature time series at Cabot Strait and in the Northwest Gulf over the period (blue squares), the period (green circles), and the period (red triangles). Fig. 6. Cross-correlation functions between temperature time series at the Laurentian Channel Mouth and three other locations further inland. The dashed line represents the null hypothesis of zero correlation at all lags at the 95% confidence level. Fig. 7. Lagged correlations between time series of water temperature on the kg m -3 isopycnal at the mouth of the Laurentian Channel and time series of the north-south displacements of the north wall of the Gulf Stream between 75 W and 50 W.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks