Table Of Contents

This Page

Spatial correlation between nitrate and relative vorticity during 2004 in OFESΒΆ

Figs. 1 and 2 show the spatial correlation within each box between nitrate and relative vorticity at 120 m for two different snapshots of 2004. These pictures reveal interesting features. There is a strong correlation within and to the east and west of the Hawaiian Islands (170W-140W, 15N-25N), as well as at the southern edge of the Kuroshio extension (KE; 140E-180E, 25N-35N). It also shows moderate correlation in the central and western part of the subtropical gyre (130E-170W, 15N-25N). These positive correlations suggests horizontal transfer of nutrient by the geostrophic turbulent flow (Klein et al. 1998 explains that relative vorticity should be positively correlated with a tracer that has a horizontal gradient and negatively correlated with a tracer that has a vertical gradient; see also Levy and Klein 2004). The nitrate around the Hawaiian Islands and in the northern edge of the western subtropical gyre may thus be transported horizontally, along isopycnals, from the surrounding regions rich in nitrate, the eutrophic subpolar gyre to the north and the 10N thermocline ridge to the south. The lower positive correlation in the central and western part of the gyre suggest that other processes are important, such as vertical mixing or vertical advection.


Figure 1: Spatial correlation within each box between nitrate and relative vorticity at 120 m on 2004/03/01. The background is the nitrate at 120 m on that day. Computed with RESEARCH/PROJECTS/MARINE_BIOLOGY/SUBMESOSCALE_PROCESSES/OFES/rel_vort_nitrate.m. Notice that the correlation does not vary too much when using different box sizes.


Figure 2: As in Fig. 1 but on 2004/11/01.

The correlation in these regions varies during the year (Fig. 3) but the above conclusions stay valid throughout the year. The mean correlation is 0.38 for the box that includes the main Hawaiian Islands, 0.45 for the box to the east of the islands, 0.30 for the box south of the KE (150E-160E, 15N-25N) and 0.58 for the box at the edge of the KE (150E-160E, 25N-35N). One exception is the region just north of the Hawaiian Islands along the subtropical front (ST; 30N), where positive correlation is found in the first part of the year and negative in the second part (Figs. 1, 2, 3a and 4), suggesting that the region is dominated by vertical advection in Spring and horizontal transfer in Fall. The mean correlation was this box is 0.02.


Figure 3: Time series of spatial correlations between nitrate and relative vorticity at 120 m for 5 of the boxes shown in Fig. 1. Computed with RESEARCH/PROJECTS/MARINE_BIOLOGY/SUBMESOSCALE_PROCESSES/OFES/rel_vort_nitrate.m.


Figure 4: (black) Surface meridional gradient of temperature averaged between 160W and 150W and 25N and 35 N. The green curve is the same as the green curve in Fig. 3a except that its amplitude has been modified for comparison with the curve of temperature gradient.

This is consistent with the seasonal variability of the ST which can be measured, for instance, via its meridional gradient of temperature (Fig. 4). The gradient increases in Fall-Winter and peaks in Spring; the frontogenesis is then at its maximum triggering significant vertical (and horizontal) advection of nutrient which could be consistent with negative correlation between nitrate and relative vorticity. In Spring and Fall, the temperature gradient, frontogenesis and vertical advection all weaken leaving the horizontal advection as the main process. The correlation then changes sign at mid-year and becomes positive in Summer and Fall.

  • How could we strengthen the above argument?
  • To complete the analysis: time-mean spatial correlation over several years + seasonal cycle for some of the boxes.