In order to separate the temporal changes of LSW from the changes set by en route mixing and resulting in spatial transformation of this water, we have calculated LSW anomalies along AR7. The anomalies will better reflect the spreading nature of the LSW events than the original property values. To compute the LSW anomalies presented in the distance-time coordinates in Fig. 24.11, a special function of distance reflecting the multiyear (20 years for LSW1994 and 7 years for LSW2000) average state of LSW was subtracted from all LSW properties, which in their turn were based on individual hydrographic profiles. Such a "special function of distance" used as the reference in our anomaly calculations was constructed piece by piece for each deep basin as follows: (1) all individual property values from the

Fig. 24.10 Scatter plots of (a) potential temperature, (b) salinity, and (c) thickness (in meters) of LSW1987 1994 "extracted" from individual hydrographie profiles along AR7 (Fig. 24.1). The circles indicate medians over stations grouped in 40 km spatial (distance) bins. The gray lines are polynomial fits of the temperature and salinity medians as functions of distance; these curves provide a continuous norm used to calculate the LSW anomalies shown in Fig. 24.11

Fig. 24.10 Scatter plots of (a) potential temperature, (b) salinity, and (c) thickness (in meters) of LSW1987 1994 "extracted" from individual hydrographie profiles along AR7 (Fig. 24.1). The circles indicate medians over stations grouped in 40 km spatial (distance) bins. The gray lines are polynomial fits of the temperature and salinity medians as functions of distance; these curves provide a continuous norm used to calculate the LSW anomalies shown in Fig. 24.11

Distance kin

Distance kin

Fig. 24.11 Anomalies of potential temperature (left column) and salinity (right column) averaged over time- and distance-dependent vertical density ranges identifying LSW (lowerrow) and LSW2000 (upper row), 1987-2005 (inclusive). The distance indicates the position along the composite AR7 section, with its origin in the central Labrador Sea. The eastward (right) pointing arrows indicate how record cold and fresh classes of LSW spread across the ocean, reaching first the Irminger Sea and then the Iceland Basin. The westward (left) pointing arrows indicate how a bulk of anomalously warm and saline water appeared near the western flank of the Reykjanes Ridge in 2000 and over subsequent years spread west along AR7, to the central and then western parts of the Irminger and Labrador basins. The anomalies were computed for hydrographic profiles indicated by dark dots. A distance-dependent long-term mean reference state used to compute these anomalies was comprised of three segments, derived for each basin separately to prevent any influence of adjacent basins (Fig. 24.10). Each basin-wide norm was constructed by grouping all individual property values in 40 km distance bins; calculating medians in each group; and then fitting a polynomial function of distance to these medians to achieve continuity of the reference state in each basin (Reproduced from Yashayaev et al. 2007a)

Fig. 24.11 Anomalies of potential temperature (left column) and salinity (right column) averaged over time- and distance-dependent vertical density ranges identifying LSW (lowerrow) and LSW2000 (upper row), 1987-2005 (inclusive). The distance indicates the position along the composite AR7 section, with its origin in the central Labrador Sea. The eastward (right) pointing arrows indicate how record cold and fresh classes of LSW spread across the ocean, reaching first the Irminger Sea and then the Iceland Basin. The westward (left) pointing arrows indicate how a bulk of anomalously warm and saline water appeared near the western flank of the Reykjanes Ridge in 2000 and over subsequent years spread west along AR7, to the central and then western parts of the Irminger and Labrador basins. The anomalies were computed for hydrographic profiles indicated by dark dots. A distance-dependent long-term mean reference state used to compute these anomalies was comprised of three segments, derived for each basin separately to prevent any influence of adjacent basins (Fig. 24.10). Each basin-wide norm was constructed by grouping all individual property values in 40 km distance bins; calculating medians in each group; and then fitting a polynomial function of distance to these medians to achieve continuity of the reference state in each basin (Reproduced from Yashayaev et al. 2007a)

same basin were used to populate spatial bins (size: 40km, overlap: 50%); (2) all values from each bin were reduced to their median value; (3) the medians were approximated by a polynomial function of distance; and finally; (4) the polynomial functions from all basins of study were lined up along the same distance axis to construct the sought reference or long-term norm. The latter was used in the calculations of LSW anomalies. (Notes: (1) the reason for splitting the basins was to prevent the influence of neighboring seas; (2) polynomial fitting was employed because it provides continuity of a norm within each basin).

Was this article helpful?

## Post a comment