North Pacific Working Group Summary

We seek to identify dominant uncertainties in our knowledge of the variability of carbon fluxes in the Pacific Ocean on seasonal, decadal and centennial time scales, and to determine what effort will yield the greatest reduction in these uncertainties. The most clearly defined time scales of physical variability in the Pacific basin are ENSO (3 - 7 y) and PDO (~30 y). The goal is to understand how these natural variations and those associated with anthropogenic effects influence the structure and dynamics of biogeochemical cycles and carbon fluxes between the ocean and atmosphere.

The enormous increase in our knowledge of oceanographic processes in the Pacific Ocean over the past decade has documented large temporal variations in the carbon cycle. For example, chlorophyll concentration in the Equatorial Pacific underwent factor of 20 changes associated with the 1997-1998 El Niño and subsequent recovery to La Nina conditions. The ecosystem in this region is now believed to shift states abruptly depending on the availability of iron. Studies at the subtropical Pacific time-series station suggest the food web dynamics and productivity in this location shifted to one strongly influenced by nitrogen fixation and phosphorus limitation sometime in the 1980s. SeaWifs ocean color observations from 1997 and 1998 show the presence of meso-scale coccolithophore blooms in the Bearing sea that were never noted in older Coastal Zone Color Scanner data (1978-1986). Forcings that control these changes and the mechanisms that link them to carbon fluxes must be understood and incorporated into basin-scale coupled biological and physical models to accurately predict the response to natural and anthropogenically induced variability.

The questions listed below have all grown out of the synthesis of existing data and summarize our view of the most important unknowns in the processes that control variability in Pacific Ocean carbon fluxes.

• How are ecological dynamics and carbon cycling affected by the basin scale differences between nitrogen fixation and denitrification?

Nitrogen fixation creates reactive nitrogen which is then available to the rest of the life in the sea. Denitrification removes this reactive nitrogen as part of the metabolism of a specific group of micro-organisms. The balance of these two processes controls the total pool of reactive nitrogen, primarily nitrate, in the ocean. The Pacific Ocean contains one of the largest areas of water column and sediment denitrification in the world. Nitrogen fixation is widespread over the oligotrophic surface waters of the Pacific and apparently fluctuates on interannual and decadal time-scales, although the absolute rates over the whole basin are unclear. The creation of reactive nitrogen by diazotrophs will result in the net uptake of atmospheric CO₂, and denitrification will cause the return of CO₂ to the atmosphere, modulated by the residence time of the nitrate in the ocean. If these two rates are exactly in balance, the net exchange of CO₂ with the atmosphere will be zero; imbalances between these two rates will result in a net ocean uptake or outgassing of CO₂.

The nitrogen fixation-denitrification balance has the potential to have an effect on the concentration of carbon in the atmosphere. Current estimates of the global denitrification rate are over 300 Tg/y, and Pacific rates may be greater than half of this total. Estimates of the global rates of nitrogen fixation are only about 100 Tg/y, again with approximately half of this in the Pacific Ocean. Given these estimates, either the ocean is wildly out of balance or one or both of the rates is in error. An imbalance implies a concomitant trend in atmospheric CO₂ change. Ice core data indicate that since about 8000 years before the industrial revolution the atmospheric pCO₂ increased slowly by about 20 ppm. One of the goals of nitrogen cycle studies is to determine its role in the air-sea partitioning and variability in atmospheric CO₂.

• What controls the east-west variation in carbon fluxes in the subarctic Pacific Ocean?

Observations of nitrate draw-down, ¹⁴C primary productivity and ¹⁵N uptake incubation experiments in the Canadian and Japanese JGOFS time-series sites indicate that the rates of carbon export and new production in the western subarctic Pacific are about twice that of the eastern part of this basin. The reasons for this difference are unknown given that both sides of this basin are nutrient replete, year-round. Since the western subarctic Pacific has been identified by global surface CO₂ maps as one of the important regions of net CO₂ flux into the oceans, an investigation into the reasons for this asymmetry is warranted. Key possibilities for the difference are: (1) differences in Fe flux between the eastern and the western basin because of prevailing wind patterns; (2) physical processes of seasonal mixed layer deepening and mechanisms controlling exchange between the euphotic zone and nutrient-enriched upper thermocline; and (3) hydrological cycle differences in the two basins with emphasis on the influence of freshwater input to the upper ocean on physical mixing processes. To understand the effect of natural decadal-scale oscillations and anthropogenically-induced climate changes on the rates of physically and biologically induced CO₂ exchanges, one must know the importance of these mechanisms in the subarctic Pacific ocean.

• How much does the interannual and decadal variability in oceanic ventilation and regional heat storage change the uptake/partitioning of ocean/atmosphere carbon?

Many lines of evidence exist that show past variability in the ventilation of the upper ocean on interannual to decadal time scales. For example decadal variability of the climate mean state shift occurred abruptly in 1976. After this time, more frequent, severe El Niño events occurred. These changes in the climate mean state are associated with large variability of the wind field and regional heat storage in the upper ocean. This significant variation from year to year in the ventilation of the Pacific waters is large and likely affects carbon fluxes in profound ways. Local manifestations of the decadal shifts likely include changes in mixing, convection, gas exchange and upwelling. All would contribute to the changes in the solubility pump. Further, each has an impact on nutrient supply and productivity leading to changes in the biological pump. Given the grand size of the North Pacific basin, interannual (ENSO) and decadal (PDO) scale variability in ventilation is likely important in upper ocean control of atmospheric pCO₂. Whether recent changes are a result of global warming or natural variability due to atmospheric forcing or interactions between circulation in the tropical and extratropical Pacific can be resolved with a combination of paleoceanographic and instrumental data. A boost in the number and location of moorings and/or the use of drifters or ships of opportunity are strategies to be considered for increasing spatial and temporal instrumental data coverage.

• How important are margins to C pathways on basin to global-scale, at annual, decadal, and centennial time scales?

Margin processes are not marginal in the context of global carbon cycling on decadal to centennial time scales. Continental margins provide Fe both to the overlying water and to the open ocean, are sites of a major portion of global denitrification, and are the gateway for terrestrial carbon input to the ocean. Primary productivity measurements, from S. California, central California, and the Washington margin indicate strong seasonality in primary productivity, and average values are >100 mmolC/m²d. New production is on the order of 50 mmolC/m²d which is 10 times greater than new production at BATS, HOT and Sta. P and 7 times greater than new production in the Equatorial Pacific (JGOFS EqPac). Margins represent approximately 10% of the area of ocean basins, hence global new production should be about evenly divided between the margins and the open ocean.

Important reasons to study continental margins in carbon cycle research are mainly their role in carbon export and preservation and their susceptibility to anthropogenic perturbations. Variability in the recycling of exported carbon and its remineralization length scales should affect global CO₂ budgets on decadal to centennial time scales. Carbon burial in margin environments is a much more important removal term than burial in the deep ocean because of the high flux and relatively shallow water column. Burial rates on river-dominated and narrow margins are not well documented. Organic mater preservation rates in margin areas are subject to large variations in time and space because of physical (floods, coastal erosion) and anthropogenic (eutrophication) forcing. Margins are particularly sensitive to anthropogenic perturbations (trawling, pollution, coastal building) and many elements of coastal ecosystems have been documented to track changes in physical forcing (PDO, ENSO).

• How are ecosystem dynamics and carbon cycling in the North and South Pacific affected by the source of Fe from above and below (dust, upwelling, margins, hydrothermal sources)?

Iron is now considered to play a crucial role in oceanic nutrient cycles. It may control the biological activity in high-nutrient, low chlorophyll areas like the North Pacific, Equatorial Pacific and the Southern Ocean. It may also control the rate of nitrogen fixation in the oligotrophic gyres. Both of these processes may influence the air-sea partitioning of CO₂, and they determine the ecosystem structure of these basins. The sources of iron are much more diverse than previously thought, and an understanding of each is required to determine the overall controls on biogeochemistry. In open ocean areas, atmospheric deposition of dust is the primary source of iron. These fluxes are closely tied to terrestrial sources and their dynamics and tend to decline with distance from the source. Upwelling of subsurface, regenerated iron is important in areas where the primary macronutrients are brought up from below. This cycle operates in rough concert with the macronutrient cycles. Near land, margin sediments are a third source of iron, both as a dissolved flux and a particulate, resuspended flux. As iron is mobilized from sediments, it spreads into the interior and becomes available through upwelling. Finally, active volcanoes emit large amounts of iron and phosphate and could be local airborne sources.

• How do carbon fluxes and ecosystem structure in the equatorial Pacific respond to physical variability on ENSO and PDO timescales?

Upwelling of waters enriched in nutrients and carbon dioxide creates a tongue of cold surface water along the equator, from the coast of South America to the international dateline in the equatorial Pacific Ocean. The vast area involved makes this region the largest natural oceanic source of atmospheric CO₂. Physical processes and biological production determine the strength of this source. During normal or cool conditions, high concentrations of nitrate in the equatorial Pacific lead to unexpectedly small increases in chlorophyll. This low productivity contributes directly to the loss of upwelled CO₂ to the atmosphere. Small-scale iron fertilization experiments have shown that the low productivity can result from iron limitation. The upwelled water originates from the EUC, which flows eastward across the basin at a depth of 20 to 200m and is enriched in iron.

Every 3 to 7 years the central and eastern equatorial Pacific warms dramatically as El Niño develops. On decadal scales the Pacific Decadal Oscillation (PDO) is an important source of variability. Early conceptual models suggested that during El Niño, Kelvin waves were the primary agents affecting biological productivity, since upwelling-favorable winds were maintained in the eastern Pacific. The waters upwelled were low in nutrients and degassing of CO₂ to the atmosphere, and productivity was hence reduced. During weak to moderate events, zonal wind anomalies are restricted to the western Pacific. However, during the strong 1997-98 event the zonal wind anomalies extended into the eastern equatorial Pacific, shutting down local upwelling and degassing of CO₂ to the atmosphere in the central equatorial Pacific for several months. After recovery from the 1997-98 El Niño the equatorial Pacific again became a source of CO₂ to the atmosphere and an unprecedented large bloom was observed, presumably the result of an enhanced supply of iron. While most think that the equatorial Pacific air-sea flux of CO₂ is physically driven, the role of the biological pump has yet to be adequately resolved for both particulate and dissolved organic carbon and particulate inorganic carbon. Further, the available field data is from the warm phase of the PDO. How the equatorial Pacific will respond to the projected cool phase of the PDO or to global warming is uncertain.

• What controls the fate of material that leaves the zone of net carbon production?

Organic matter respiration below the sun-lit surface ocean is a dominant factor controlling chemical transformations and biological communities of the subsurface ocean. The rate of respiration and downward transport of organic material (both particulate and dissolved) influence the depth scales of the chemical and biological changes, which determines the response time of sequestration of carbon in the ocean by the biological pump and ocean circulation. Tracers in the euphotic zone (including ¹³C, the three oxygen isotopes, and N isotopes) can make a big impact on our understanding of production rates and hence also export processes. Rates of respiration in the sub-euphotic zone have been determined by modeling apparent oxygen utilization and transient and short-lived tracers, because they are too slow to measure directly in most cases. A more complete understanding of the rates of respiration and transport processes will help oceanographers determine the response time of the carbon cycle to climate change, should help close the mass balance inferred from upper ocean tracers, and bear on questions of spatial variability of inputs. A better understanding of these processes will also aid in our attempt to use tracers of metabolic processes as indicators of decadal scale changes in the strength of the carbon pump. Specific areas that are ripe for improved understanding are: (1) the rates and mechanisms of particulate and dissolved organic matter degradation, including the differing rates of remineralization of micronutrients, P, N, and C; (2) the rates and mechanisms of particulate inorganic carbon (PIC) production and dissolution; (3) rates of mixing and subduction in the upper ocean, and (4) the influence of different amounts and types of organic mater ballast (in the form of calcium carbonate and opal) on its sinking and degradation characteristics.