Oxygen-nutrient relationships in the Pacific Otean ’

In the northeastern Pacific Otean the oxygen-phosphatc and oxygen-nitratc relationships vary systematically with latitude. The variation above the oxygen minimum zone is much more pronounced than below it. The slopes of these relationships in general increase with increasing latitude. In the entire water column these slopes vary with depth. A seasonnl effect on the slopes of the oxygen-phosphate and oxygen-nitrate relationships, similar to that with decreasing latitude, is observed; the winter slopes are higher than the summcr slopes. Potential temperature versus preformed phosphate diagrams were constructcd for different stations in the Pacific and Indian Oceans to study their water masses. Postma (1964) plotted dissolved oxygen concentrations, O,, versus phosphate concentrations, PO.,, from stntions nt different latitudes (55”N to 62’s) in the Pacific Otean, and found n variable AO, : APO, ratio with depth. Pytkowicz (1964) nnd Park (19670) found a linear relationship of 02 versus PO4 nbove the oxygen minimum zone off the Oregon coast. AlvarezBorrego et al. (1972, 1975) tested Redfield’s model for OY-nutrient relationships and found it consistent with the field data, showing that the variation of the 02-nutrient relationship slopes with depth, latitude, and time of the year is due to mising between different water types with different preformed portions of Oand nutrients. Diagranls of 02 versus nutrient may thus be used to characterize water masses and preformed nntrient (Redfield 1942) versus temperature diagrams may be used to detect and trace water mnsses thnt are not very well identified by T-S dingrams. The purpose of this work js to study the 1 This work \vas supported by National Science Foundation grnnts GX-381F7 and GA-12113 and Office of Naval Research contract NO0014Gi’-A0369-0007 undcr project NR-083-102. The “Conseio Nacional de Ciencia v Tccnoloeia” of Mexico snpported the work of S. Alvarez-Borrego. This papcr compriscs part of n dissertation submitted by S. Alvarrz-Borrego in partial fulfillmcnt of the requirements for the Ph.D. at Oregon State University. O---PO, and O,-NO:, relationships for tlle northeastern Pacific Ocenn and soutlieastcm Bering Sea, nnd to trace Pacific Intermedinte \Vnters using prcformed nutrient versus potentinl temperature, 8, tliagrains.


In the northeastern
Pacific Otean the oxygen-phosphatc and oxygen-nitratc relationships vary systematically with latitude.The variation above the oxygen minimum zone is much more pronounced than below it.The slopes of these relationships in general increase with increasing latitude.
In the entire water column these slopes vary with depth.
A seasonnl effect on the slopes of the oxygen-phosphate and oxygen-nitrate relationships, similar to that with decreasing latitude, is observed; the winter slopes are higher than the summcr slopes.
Potential temperature versus preformed phosphate diagrams were constructcd for different stations in the Pacific and Indian Oceans to study their water masses.
Postma (1964) plotted dissolved oxygen concentrations, O,, versus phosphate concentrations, PO.,, from stntions nt different latitudes (55"N to 62's) in the Pacific Otean, and found n variable AO, : APO, ratio with depth.Pytkowicz (1964) nnd Park (19670) found a linear relationship of 02 versus PO4 nbove the oxygen minimum zone off the Oregon coast.Alvarez-Borrego et al. (1972, 1975) tested Redfield's model for OY-nutrient relationships and found it consistent with the field data, showing that the variation of the 02-nutrient relationship slopes with depth, latitude, and time of the year is due to mising between different water types with different preformed portions of O-and nutrients.Diagranls of 02 versus nutrient may thus be used to characterize water masses and preformed nntrient (Redfield 1942) versus temperature diagrams may be used to detect and trace water mnsses thnt are not very well identified by T-S dingrams.
The purpose of this work js to study the O---PO, and O,-NO:, relationships for tlle northeastern Pacific Ocenn and soutlieastcm Bering Sea, nnd to trace Pacific Intermedinte \Vnters using prcformed nutrient versus potentinl temperature, 8, tliagrains.The positions of these stations are shown in Fig. 1.

Results
In the northenstern Pacific Otean the O--PO, and Oz-NO.? relationships for the regioxi of the water column above and below the O-minimum zone vary systematically with latitude (Figs. 2 and 3).The vnriation above the O-minimum zone is much more pronounced thnn below it.The slopes of these relationships in general increase with increasing latitude; in thr entire wnter column, they vary with depth.The slopes are steeper in winter thnn in summcr (Figs. 4 and 5).
These variations of the OZ-P04 and Oz-NO:, slopes were estimnted by applying the lenst squnres method to the O-, P04, nnd NOn field data.The confidente intervals were estimated at the 95% confidente level.fornia, to -0.404 * 0.022 in the southeastern Bering Sea.
Above the 02 minimum zone, near 45"N, 180", the Oa-POa slope varies from -3.38 rt: 0.08 in March to -2.74 f 0.10 in June (Fig. 4); the Oz-NO:< slope varies from   Figure 8 shows that the Antarctic Intermediate Water is not chnracterized by a Pa(,) minimum in the South Pncific.Thus, the PO, WI minimum present in the North Pacific Otean and southeastern Bering Sea is not formed at the Pacific sector of the Antarctic Convergence.
Figure 8a shows that for stations 71 and 72 (off Chile) of the SCORPIO expedition, the Antarctic mediate Water is not characterized by a well defined PO, (,,, maximum. Redfield ( 1942) found this POac,,, maximum from the Antarctic Convergence to about 10"N in the Atlantic Ocenn.For stations 29 nnd 30 of the SCORPIO espedition (off New Zealand) the PO 4,,1) masimum is not apparent.These four stations are at the.same latitude (Fig. 1).Although Fig. Sa does not show the @-PO, ,,), diagrams for stations If the P04(,, minimum found in the North Pacific Otean is not formed at the Antarctic Convergence, the only possibility left is that it is formed in the eastern Atlantic Otean and western Indian Otean sectors.It may be transported through the Indian Otean and the Celebes Sea, north of Australia, into the Pacific Otean.This is a very remote possibility indeed, but to include it, we plotted 0-P04(,, diagrams for severa1 stations of Anton Bruun cruise 2. Figure 9 shows that there is no P04(,, minimum or maximum in the Indian Otean between 35"s and OO"; The values of deep waters decrease northward, possibly because of mixing with surface water with low PO4 (r).Thus, the P04(,, minimum at intermediate depths in the North Pacific Otean and southeastern Bering Sea is not formed at any sea surface source.
Figure 10 shows &PO4(,, diagrams for two stations of cruise 26 of Vityaz in the Central Pacific Otean (near the equator).Unfortunately there are only a few data points for these two stations, and the near surface data points with 0 > 20°C are not plotted.Nevertheless, no significant mini-  There is a trend for sigma-t values corresponding to the P04(,, minimum to increase as latitude increases.At HAH28 the PO4 (P) minimum is at the 27.11 sigma-t surface while at HAH50 it is at 27.44.North of HAH.50 (Fig. 1) the sigma-t values corresponding to the P04(,, minimum do not change significantly. The PO~(,I values for near bottom waters of the stations studied (Figs. 6,7,8) differ in some cases from station to station, and from cruise to cruise, by as much as 0.2 $vl.This is probably because there is not yct an international standard for PO, analyses.
Standards are often prepared manually for each station.However, the main problem with PO, analysis is not Xcuracy but precision.Steps should be taken to improve the precision of the methods before an international standard is accepted.The data from GEOSECS GOGO-1 are the most precise used here.

Summary
In the northeastern Pacific Otean the O?-PO* and O?-NO3 relationships for the region of the water column above and below the OZ minimum zone vary systemntically with latitude.The variation above the OZ minimum zone is much more pronounced than below it.The slopes of these relationships in general increase with increasing latitude.In the cntire \vater column these slopes vary with depth.An effect on the slopes of the O--PO4 and 02-NO,, relationships, similar to that obscrved with decreasing latitude, is scen in comparing seasonal data: winter slopes are steeper.
Fig. 3. Oxygen-nitrate diagram.Data presented as in Fig. 2. The nmitinn nf the ctatinn nnt aivm -. __._ Fig. 4. Oxygen-phosphate diagram.Comparison between winter and summer data.The numbers by the data points represent depth in meters.The position of the station not given in the legend to Fig. 3 is: ll6 (44"51.O'N, 174"57.6'E).
POU,> diagrams.a-Stations 29, 30, 71, and 72; b-95 and 144 of SCORPIO expedition.Station positions are: 29 '(43"15.O'S, 169"5O.O'W), 30 (43"15.0'S,169'04.5'W),71 (43" 14.7'S, 80"02.O'W), 72 (43"19.(YS,79"01S'W), 95 (28"15.7'S,79"073'W), and 144 (28"15.7'S,170" 14.8'W).AIW means Antarctic Intermediate Water.The Subarctic Intermediate Water has been classically characterized by a salinity, S, minimum.According to Reid (1965) this S minimum is recognizable as far South as lO"N, along 16O"W.Park's (19670) section also shows the deeper POd(,, minimum.The PO*(,, minimum (Figs.6c,d and 7) coincides with the Or minimum.Alvarez-Borrego et al. (1972) found the PO.* Il>) minimum in the southeastern Bering Sea and thought it possibly the core of a water mass which when at the sea surface had undergone intense photosynthesis, with depletion of PO, to very low values accompanied by equilibration of the dissolved O-with the atmosphere.Off Baja California and near Hawaii (stations GOGO-1 and HAH22) (Figs. 1 and Ga,b) neither the POdI,,> maximum nor the minimum is apparent.At station HAH30 the POi(,, minimum (Fig. 6d) almost coincides with the sigma-t surface of 27.28 that Reid (1965) used to study the Antarctic Intermediate Water.According to Reid this sigma-t surface lies more than 350 m below the sea surfnce everywhere in the North Pacific.If the explnnntion given by Alvarez-Borrego et al. (1972) for the PO4 (I>) minimum is correct, the source of this water should not be in the North Pacific Otean.
Figure lla illustrates diagrammatically how the PO1(,, maximum characterizing the Antarctic Intermediate Water in the South Pacific is eroded as it moves northward by mixing with shallower waters with lower P04(,, values.The higher PO1(,, values of the Subarctic Intermediate Water and of the surface waters of the subarctic region cause the PO((,, values at intermediate depths to be minimum.Thus, the P04(,, minimum observed at high latitudes in the North Pacific is the result of high PO4(,, values in the surface waters of this region of the otean.There is a trend for sigma-t values corresponding to the P04(,, minimum to increase as latitude increases.At HAH28 the