Pacific Ocean Boundary Ecosystems http://www.pacific-ecosystems-climate.org

The goal of this task is to explore the large-scale ocean/atmosphere dynamics of the Pacific associated with modes of low-frequency variability such as the Pacific Decadal Oscillation (PDO) and the North Pacific Gyre Oscillation (NPGO), ENSO and others.

In Di Lorenzo et al. (2009b) we show that decadal dynamics of the Pacific Decadal Oscillation (PDO) and the North Pacific Gyre Oscillation (NPGO) are linked through their relationships to ENSO. The PDO and NPGO are the oceanic expression of the two dominant modes of North Pacific atmospheric variability -- the Aleutian Low (AL) and the North Pacific Oscillation (NPO). We compute the two dominant modes of ocean/atmosphere co-variability in the Pacific sector [40S-62N] and find that the first co-variability mode captures the mature phase of the canonical eastern Pacific warming (EPW) ENSO and its atmospheric teleconnections to the AL, while the second co-variability mode captures the NPGO/NPO tropical expression, which leads the ENSO mode by ~8-12 months. The atmospheric projections of these first two modes are used to extract the AL and NPO forcings related to ENSO. These forcings are then integrated with an AR-1 model and lead to skillful reconstructions of the PDO (R=0.65), NPGO (R=0.60), and Pacific surface temperature decadal variance (R=0.4-0.8). We synthesize these results with to propose a framework for quasi-deterministic decadal oscillations in physical and biological variables of the Pacific (see diagram below).

More recently in Di Lorenzo et al. (2010) these findings have been expanded to account for additional teleconnections from the tropics to the extra-tropics between ENSO and the Pacific decadal modes. We find that dominant decadal fluctuations of the North Pacific sea surface temperatures and gyre-scale circulation of the North Pacific Gyre Oscillation (NPGO), are dynamically linked to the central Pacific warming (CPW) El Niño -- an emerging mode of interannual variability of the tropical Pacific that is different from the canonical EPW ENSO in that the SST maximum anomalies are displaced from the eastern to the central Pacific. We show that the CPW drives low-frequency changes in the extra-tropical atmospheric circulation that are integrated by the ocean to form the decadal NPGO. This new link provides us with an improved conceptual framework of Pacific climate variability that is summarized in the figure below. In this new framework, evidence for more frequent CPW events under greenhouse forcing scenarios is consistent with a more energetic late 20th century NPGO6, implying that NPGO will play an increasingly important role in shaping Pacific climate and marine ecosystems in the 21st century.

In Ceballos et al. (2009) we show that the NPGO variability orginally isolated in the central and eastern North Pacific is also linked to the North Pacific western boundary, another target region of POBEX. This study shows that Rossby waves dynamics excited by the NPO propagate the NPGO signature from the central North Pacific into the Kuroshio-Oyashio Extension (KOE), and trigger changes in strength of the KOE with a lag of 3 years. This suggests that the NPGO index can be used to track changes in the entire northern branch of the North Pacific sub-tropical gyre. These results provide a physical mechanism to explain coherent decadal climate variations and ecosystem changes between the North Pacific eastern and western boundaries.

References:

ENSO and the North Pacific Gyre Oscillation: an integrated view of Pacific decadal dynamics
Di Lorenzo E., N. Schneider, K. M. Cobb, J. Furtado and M. Alexander
Geophysical Research Letters, 2009, submitted. [ PDF ]

North Pacific Gyre Oscillation synchronizes climate fluctuations in the eastern and western North Pacific
Ceballos L., E. Di Lorenzo, N. Schneider, B. Taguchi
Journal of Climate, 2009, DOI: 10.1175/2009JCLI2848.1. [ PDF ]

Central Pacific Warming El Nino and decadal climate change in the North Pacific
Di Lorenzo E., K. M. Cobb, J. Furtado, N. Schneider, B. T. Anderson, A. Bracco, M. Alexander and D. Vimont
Nature Geoscience, 2010, submitted. [ PDF ]

One of the goals of POBEX is to investigate the response of Pacific Boundary physcial and biological systems to future climate. As a first step we have examined the output of several IPCC climate models.

In Furtado et al. (2010) we explore the dynamics of North Pacific decadal variability (NPDV) in 10 coupled climate models used in the Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC) for both the 20th century climate and under a greenhouse warming scenario. We examine the two dominant oceanic modes of NPDV, the Pacific Decadal Oscillation (PDO) and the North Pacific Gyre Oscillation (NPGO), along with their respective atmospheric forcing patterns associated with variability of the Aleutian Low (AL) and the North Pacific Oscillation (NPO). We find that the decadal modes of ocean/atmosphere variability (e.g. PDO/AL, NPGO/NPO) are spatially consistent with modern observations in most models and show no change during climate change. The frequency content of the PDO and NPGO also show no significant change in future climate. However, most climate models do not adequately reproduce the observed temporal variability in the 20th century, with some failing to capture the low frequency nature of the PDO and the broad frequency band of the NPGO. In the 20th century, we also find that while the dynamics of the model’s PDO are consistent with forcing by the AL variability, the NPGO of the models is not always driven by atmospheric variability associated with NPO as suggested from modern observations. Further analysis also reveals that the NPDV of the IPCC models is mostly independent of the tropics and does not exhibit the observed relationships between the North Pacific modes (e.g. PDO/AL, NPGO/NPO) and the El Niño-Southern Oscillation (ENSO). In contrast with modern observations, the PDO of most models appears as a mode independent of ENSO during the 20th century and the Seasonal Footprinting Mechanism (SFM), whereby the midlatitude NPGO/NPO winter expression leads the tropical ENSO peak in fall, is not captured in the IPCC models, except for the GFDL 2.1 and MIROC-HIRES. These inconsistencies between observed and modeled climate variability of the IPCC models are being carefully considered before using the output of the IPCC models to downscale climate change scenarios at the regional and coastal scale to make decadal predictions across the North Pacific.

References:

North Pacific Decadal Variability and Climate Change in the IPCC AR4 Models
Furtado J., E. Di Lorenzo, N. Schneider, N. Bond and J. Overland
Journal of Climate, 2010, submitted. [ PDF ]

An important aspect of the project is to separate the large-scale variability from the regional variability independent of the large-scale dynamics. So far we have taken two approaches, which we are testing using SST, SSH, and CHL-a data from both reanalysis and satellite.

Using the decadal framework explained in Di Lorenzo et al. (2009b) (see figure below) we quantify how much variance is connected to the large-scale ocean/atmosphere modes ENSO (EPW/CPW), PDO/AL and NPGO/NPO. Previous studies have shown that the decadal variations of the PDO and NPGO emerge from integrating the atmospheric forcing of the AL [Schneider and Cornuelle, 2006] and NPO (Chhak et al., 2009). We use the atmospheric SLPa projections of the first two co-variability modes over the Pacific basin (see figure below), referred to as AL_ENSO and NPO_ENSO, as proxies of the AL and NPO atmospheric forcings related to ENSO. These atmospheric projections track closely the AL and NPO indices on different frequency bands.

We can now quantify the large-scale forced decadal dynamics of a variable of choice (e.g. the Pacific SSTa) that are associated with the PDO/AL and NPGO/NPO components connected to ENSO by considering the AR-1 model (see figure below) forced by both the AL_ENSO and NPO_ENSO:

This model is integrated at each location and is therefore independent of the spatial domain chosen for the analysis. The coefficients are determined at each location using a least-square fit. The AR-1 reconstruction has high skill in capturing the decadal (timescales > 7 years) variance of Pacific SSTa (R=0.4-0.8) (Figure below). The skill of the reconstructed SSTa in the ENSO frequencies (defined here as 2-7 years) is uniformly high, extending into the tropical Indian and Atlantic oceans (Figure 3d), while timescales < 2 years exhibit reduced skill. We will apply this AR-1 modeling approach to quantify the fraction of variance in satellite CHLa and in other biological timeseries available to POBEX, that is linked to the large-scale forcing decadal dynamics. The residual will then be investigated in the context of local dynamics in the various boundary regions. This work is still in progress.

Analysis of Satellite Sea Surface Height Anomalies (SSHa)

In order to make a connection between basin-scale and coastal circulation, we have concentrated on extending the alongtrack altimeter Sea Level Anomaly (SLA) data closer to the coast along the eastern boundaries of the Pacific Ocean. Support from GLOBEC has been leveraged by combining efforts with a NASA-funded project, as part of the Ocean Surface Topography Science Team (OST-ST). To retrieve altimeter data from coastal regions, we must examine and address the problems that affect the altimeter data when the tracks are within approximately 50 km of the land. It has become clear that the primary fields or calculations that need improvement in the coastal ocean are the (1) wet troposphere correction (within 40-50 km of the coast); (2) the actual range calculation through “retracking” (within 10-15 km of the coast); and (3) tidal corrections, which are usually less important over the narrow continental shelves of Eastern Boundary Currents than over the wider shelves of Western Boundary Currents. These methods are being developed within scientific analyses of the alongtrack SLA fields next to the coasts of the U.S. West Coast (USWC, our GLOBEC test area).Off the USWC, we have formed a 15-year, monthly climatology of alongtrack SLA off Oregon and Washington. The along-track integral of the monthly 15-year climatology of SLA provides a proxy for the average density of the water (low heights mean greater density). Changes (month-to-month) of these integrated SLA signals have been compared to the Coastal Upwelling Index (surface Ekman transports, as calculated from the alongshore winds from a NAVY model). The hypothesis is that the upwelling or downwelling will move denser or lighter water over the shelf, resulting in month-to-month changes in the SLA integral over several hundred kilometres of the coast. The results of this analysis were presented by Strub et al. (2010) at the AGU Ocean Sciences meeting in February, 2010. Figure below, below, presents the results. More work is in progress.

References:

ENSO and the North Pacific Gyre Oscillation: an integrated view of Pacific decadal dynamics
Di Lorenzo E., N. Schneider, K. M. Cobb, J. Furtado and M. Alexander
Geophysical Research Letters, 2009, submitted. [ PDF ]

Forcing of low-frequency ocean variability in the Northeast Pacific
Chhak, K., E. Di Lorenzo, N. Schneider and P. Cummins
Journal of Climate, 2009, 22(5), 1255-1276, DOI: 10.1175/2008JCLI2639.1 . [ PDF ]

The goal of this task is to isolate coherent variations in ecosystems across different boundary systems and separate the variability connected to large-scale Pacific dynamics from that one associated with local processes. Our first study examines the two Pacific eastern boundary current systems.

In Thomas et al. (2009) we use SeaWiFS data to obtain the first systematic comparison of 10 years (1997–2007) of chlorophyll interannual variability over the California (CCS) and Humboldt (HCS) Current Systems. Dominant signals are adjacent to the coast in the wind-driven upwelling zone. Maximum anomalies in both systems are negative signals during the 1997–1998 El Niño that persist into 1999 at most latitudes. Thereafter, anomalies primarily appear to be associated with shifts in phenology, with those in the CCS stronger than those of the HSC. Prominent signals in the CCS are positive anomalies in 2001–2002 at latitudes >35°N and <30°N, and in 2005–2006 from 30 to 45°N that persist at latitudes >40°N into 2007. In the HCS, latitudinally exten- sive positive events occur in austral summers of 2002–2003, 2003–2004. Relationships of chlorophyll anomalies to forcing are explored through correlations to local upwelling anomalies and three indices of Pacific Ocean basin-scale variability, the Multivariate El Niño Index (MEI), the Pacific Decadal Oscilla- tion (PDO) and the North Pacific Gyre Oscillation (NPGO). These show that each system has strong lati- tudinal regionality in linkage to forcing. At higher latitudes, correlations follow expected relationships of increased (decreased) chlorophyll with positive upwelling and NPGO (MEI and PDO). At specific latitudes, notably the Southern California Bight and off Peru, where circulation and/or chlorophyll phenology differ from canonical EBUS patterns, correlations weaken or oppose those expected. Correlations excluding the El Niño period remain similar in the CCS but substantially changed in the HCS, indicating much stronger domination of El Niño conditions on HCS anomaly relationships over this 10-year period.

We are now in the process of applying State Space Models to compare the two eastern boundary current systems CCS and PCCS. Given the complexity of applying these models we have so far to developed and tested these algorithm for the California Current region with our NOAA collaborators Dr. Roy Mendelssohn and Dr. Frank Schwing. Preliminary results discussed in Thomas et al. (2010) are promising and enable us to separate the components of low-frequency variability associated with changes in phenology vs. large-scale changes in the atmosphere/ocean forcing.

References:

Interannual variability in chlorophyll concentrations in the Humboldt and California Current Systems
Thomas, A.C., Brickley P. and Weatherbee R.
Progress in Oceanography, 2009, DOI: 10.1016/j.pocean.2009.07.020. [ PDF ]

Time and space patterns of satellite CHL-a phenoloogy in the California Current
Thomas A., R. Weatherbee, T. Strub, R. Mendelssohn and C. James
Progress in Oceanography, 2010, in prep.. [ PDF ]

This task focuses on diagnosing how large-scale Pacific dynamics affect variations of upper ocean physical and biological variables of the Northeast Pacific (NEP). This region includes two GLOBEC study areas -- the California Current System (CCS) and the Gulf of Alaska (GoA). Below is the summary of our studies:

In Chaak et al. (2009) we use an ocean model to examine the forcing mechanisms and underlying ocean dynamics of the PDO and NPGO in the northeast Pacific (NEP). It is found that the PDO and NPGO modes are each tied to a specific atmospheric forcing pattern. The PDO is related to the overlying Aleutian low, while the NPGO is forced by the North Pacific Oscillation. The above-mentioned climate modes captured in the model hindcast are reflected in satellite altimeter data. A budget reconstruction is used to study how the atmospheric forcing drives the SST and SSH anomalies. Results show that the basinwide SST and SSS anomaly patterns associated with each mode are shaped primarily by anomalous horizontal advection of mean surface temperature and salinity gradients via anomalous surface Ekman currents. This suggests a direct link of these modes with atmospheric forcing and the mean ocean circulation.

In Capotondi et al. (2009), a study the same eddy-permitting ocean model of the northeast Pacific is used to examine the ocean adjustment to changing wind forcing in the Gulf of Alaska (GOA) at interannual-to-decadal timescales. It is found that the adjustment of the ocean model in the presence of mesoscale eddies is similar to that obtained with coarse-resolution models. Local Ekman pumping plays a key role in forcing pycnocline depth variability and, to a lesser degree, sea surface height (SSH) variability in the center of the Alaska gyre and in some areas of the eastern and northern GOA. Westward Rossby wave propagation is evident in the SSH field along some latitudes but is less noticeable in the pycnocline depth field. Differences between SSH and pycnocline depth are also found when considering their relationship with the local forcing and leading modes of climate variability in the northeast Pacific. In the central GOA pycnocline depth variations are more clearly related to changes in the local Ekman pumping than SSH. While SSH is marginally correlated with both Pacific Decadal Oscillation (PDO) and North Pacific Gyre Oscillation (NPGO) indices, the pycnocline depth evolution is primarily related to NPGO variability. The intensity of the mesoscale eddy field increases with increasing circulation strength. The eddy field is generally more energetic after the 1976 – 1977 climate regime shift, when the gyre circulation intensified. In the western basin, where eddies primarily originate from intrinsic instabilities of the flow, variations in eddy kinetic energy are statistically significant correlated with the PDO index, indicating that eddy statistics may be inferred, to some degree, from the characteristics of the large-scale flow.

References:

Forcing of low-frequency ocean variability in the Northeast Pacific
Chhak, K., E. Di Lorenzo, N. Schneider and P. Cummins
Journal of Climate, 2009, 22(5), 1255-1276, DOI: 10.1175/2008JCLI2639.1 . [ PDF ]

Low-Frequency Variability in the Gulf of Alaska from coarse and eddy-permitting ocean models
Capotondi, A., V. Combes, M. A. Alexander, E. Di Lorenzo and A. J. Miller
Journal of Geophysical Research, 2009, doi:10.1029/2008JC004983. [ PDF ]

This task attempts to quantify the nutrient transport pathways and dynamics using forward and inverse passive tracers experiments in high-resolution ocean modeling frameworks.

To date we have completed a study for the Gulf of Alaska -- a region characterized by high-nutrient and low-chlorophyll-a concentration. Recent observational studies suggest that advection of iron-rich coastal water is the primary mechanism controlling open ocean produc- tivity. Specifically, there is evidence that mesoscale eddies along the coastal GOA entrain iron-rich coastal waters into the ocean interior. In Combes et al. (2009) we quantify the cross-shelf transport statistics in the GOA using an eddy-resolving ocean model over the period 1965–2004. The statistics of coastal water transport are computed using a model passive tracer, which is continuously released at the coast. The passive tracer can thus be considered a proxy for coastal biogeochemical quantities such as silicate, nitrate, iron, or oxygen, which are critical for explaining the GOA ecosystem dynamics. On average along the Alaska Current, it has been shown that at the surface while the advection of tracers by the average flow is directed toward the coast consistent with the dominant downwelling regime of the GOA, it is the mean eddy fluxes that contribute to offshore advection into the gyre interior. South of the Alaskan Peninsula, both the advection of tracers by the average flow and the mean eddy fluxes contribute to the mean offshore advection. On interannual and longer time scales, the offshore transport of the passive tracer in the Alaskan Stream does not correlate with large-scale atmospheric forcing, nor with local winds. In contrast in the Alaska Current region, stronger offshore transport of the passive tracer coincides with periods of stronger downwelling (in particular during positive phases of the Pacific decadal oscillation), which trigger the development of stronger eddies.

In Combes et al. (2010) we have extended our study are to investigate the low-frequency variability in the transport (cross-shore and alongshore) of coastal water masses in the California Current System (CCS) over the period 1950-2008. The approach consists of an ensemble of passive tracers released in the ROMS ocean model to characterize the effects of linear (Ekman upwelling) and non-linear (eddy activity) circulation regimes on the statistic of low frequency advection of coastal waters. By looking at the passive tracer concentration distribution, we find that the low-frequency upwelling and the surface offshore transport of the upwelled nutrient-rich coastal water are strongly correlated with the alongshore wind stress, and is coherent between the central and southern CCS. However, the offshore transport of the surface coastal water is not anymore coherent between those two regions, and has been found to be associated with mesoscale eddy activity, where both surface and subsurface waters propagate offshore, trapped in the eddy core. Our results also show that the poleward California Under-Current, at about 200m depth, affects the alongshore transport and provides rich waters to the Central California dominant upwelling cell. Therefore the passive tracer experiments, performed in this study, could provide a dynamical framework to understand the dynamics of the upwelling and offshore transport of nutrient rich coastal water and to interpret how it responds to atmospheric forcing. This also could reinforce our interpretation (and therefore predictions) in the changes in vertical and offshore advection of other important biogeochemical quantities, essential in understanding the Californian ecosystem variability.

References:

Interannual and decadal variations in cross-shore mixing in the Gulf of Alaska
Combes, V., E. Di Lorenzo and E. Curchister
Journal of Physical Oceanography, 2009, 39(4), 1050-1059, DOI: 10.1175/2008JPO4014.1. [ PDF ]

Cross-shore transport varaibility in the California Current: Ekman upwelling vs. eddy regime
Combes V., F. Chenillat, P. Riviere, E. Di Lorenzo, M. Ohman and S. J. Bograd
Journal of Physical Oceanography, 2010, submitted. [ PDF ]

The goal of this task is to explore the links between zooplankton variability and changes in transport dynamics associated with both the mesoscale eddy field and the large-scale circulation.

In Keister et al. (2009) we explored the effect of mesoscale circulation features on zooplankton distributions in the coastal upwelling ecosystem of the northern California Current. We found that mesoscale circulation can create disparate regions in which zooplankton populations are retained over the shelf and biomass can accumulate or, alternatively, in which high biomass is advected offshore to the oligotrophic deep sea.

We also explored in Keister et al. (2010) the dynamics underlying the observed correlation between zooplankton community in the Northern California Current and the Pacific Decadal Oscillation (PDO) index. A “warm-water” copepod species group is more abundant during warm (positive) phases of the PDO and less abundant during cold (negative) phases; the reverse occurs for a “cold-water” species group. This led to the hypothesis that the relative dominance of warm/cold copepod communities in the Northern California Current is associated with changes in the horizontal advection of surface water over different phases of the PDO. The Regional Ocean Modeling System (ROMS) and passive tracer simulations were used to investigate variation in surface water advection to coastal regions of the northern California Current over the period of 1950 to 2007. The results of the simulations were compared to observations of the species composition of zooplankton collected off Newport, Oregon (44.7°N 124.2°W) since 1996. Horizontal surface water movements change with the phase of the PDO and were related to regional changes in copepod species dominance. The surface current anomalies during the positive phase of the PDO showed average downwelling conditions (west to east transport) and stronger northward currents at the coast. These anomalies tended to transport warmer waters and the associated copepod species into the region. During the negative phase of the PDO, an increase in surface currents originating from the north enters the study area, and intrusion from the south and west declines. Overall, the modeled transport anomalies explained a significant fraction of the variance in copepod communities. Elucidating the mechanisms of variability in copepod community composition furthers our understanding of the physical controls on ecosystem processes in the Northern California Current.

References:

Zooplankton distribution and cross-shelf transfer of carbon in an area of complex mesoscale circulation in the northern California Current
Keister, JE, WT Peterson, and SD Pierce
Deep-Sea Research I, 2009, 56: 212-231. [ PDF ]

Copepod species composition linked to ocean transport in the Northern California Current
Keister, J.E., N. Mariani, C.A. Morgan, E. Di Lorenzo, V. Combes and W.T. Peterson
Global Change Biology, 2010, submitted. [ PDF ]

The goal of this task is to synthesis our findings in the NEP in the context of the large-scale ecosystem dynamics.

Our first study in this direction is presented in Di Lorenzo et al. (2009a) where we show that long-term timeseries of upper ocean salinity and nutrients collected in the Alaskan Gyre along Line P exhibit significant decadal variations that that are in phase with variations recorded in the Southern California Current System by the California Cooperative Oceanic Fisheries Investigation (CalCOFI). We present evidence that these variations are linked to the North Pacific Gyre Oscillation (NPGO) -- a climate mode of variability that tracks changes in strength of the central and eastern branches of the North Pacific gyres and of the Kuroshio-Oyashio Extension (KOE). The NPGO emerges as the leading mode of low-frequency variability for salinity and nutrients. We reconstruct the spatial expressions of the salinity and nutrient modes over the northeast Pacific using a regional ocean model hindcast from 1963-2004. These modes exhibit a large-scale coherent pattern that adequately predicts the in-phase relationship between the Alaskan Gyre and California Current timeseries. The fact that large-amplitude, low-frequency fluctuations in salinity and nutrients are spatially phase-locked and correlated with a measurable climate index (the NPGO) open new avenues for exploring and predicting the effects of long-term climate change on marine ecosystem dynamics.

References:

Nutrient and Salinity Decadal Variations in the central and eastern North Pacific
Di Lorenzo E., Fiechter J., Schneider N., A. Bracco, Miller A. J., Franks P. J. S., Bograd S. J., Moore A. M., Thomas A., Crawford W. and Pena and Herman A.
Geophysical Research Letters, 2009, doi:10.1029/2009GL038261. [ PDF ]

This task attempts to quantify the nutrient transport pathways and dynamics using forward and inverse passive tracers experiments in high-resolution ocean modeling frameworks.

In Combes et al. (2010) we investigate the low frequency transport and upwelling dynamics in the Peru-Chile Current System (PCCS) using an eddy resolving ocean model combined with a passive tracer advection-diffusion equation. The PCCS is the one of the world’s most productive regions in fish landings, providing 18 to 20% of the world marine catches despite covering less than 1% of the world’s ocean surface. This high productivity results principally from the upwelling of nutrient-rich water in the photic zone. The upwelling variability is examined using a model passive tracer, continuously released at the coast in the subsurface between 150 and 250 m depth. The concentration of that tracer observed at the surface can therefore be considered as an index of coastal upwelling. Different model experiments are conducted to explore the sensitivity of the PCCS upwelling to different air-sea fluxes of momentum and atmospheric and oceanic teleconnections to ENSO. On average, we find that the model runs forced by ECMWF and QSCAT wind stress compare better with observations than the experiment forced by the NCEP wind stress, with strong upwelling off Peru (5°S-16°S) and central Chile (27°S-34°S) and significantly weaker upwelling in the North Chile Region (18°S-24°S), consistent with the strength of coastal upwelling-favorable winds. Temporal variability is nevertheless well reproduced by NCEP, when comparing model and in situ coastal sea surface height data. There is evidence, in this region, that both changes in surface wind and coastally trapped Kelvin waves controlled the variability of the coastal upwelling. The effect of ocean remote forcing, assessed by comparing the output of two model simulations which does and does not include the presence of waves in their boundaries. The passive tracer approach indicates that, off the coast of Peru, the Southern Oscillation strongly modulates the strength of coastal upwelling principally due to the propagation of downwelling equatorial Kelvin waves (particularly strong during El Niño years) rather than changes in local wind stress. Off central Chile, we find that the first mode of coastal upwelling variability is also strongly correlated with the Southern Oscillation with a decadal variability signal that seems to respond to the second mode of sea level pressure in this region. Our results indicate that the central Chile region is also very sensitive to Kelvin wave generated at the equator, reducing for example considerably the upwelling during strong El Niño events, therefore impacting the ecosystem variability.

References:

Modeling interannual and decadal variability in the Humboldt Current upwelling system
Combes V., E. Di Lorenzo, F. Gomez, S. Hormazabal, T. P. Strub and D. Putrasahan
Journal of Physical Oceanogrpahy, 2010, submitted. [ PDF ]

This task is led by international South American collaborators C. Parada, S. Núñez, A. Sepúlveda, and S. Hormazábal. This research, which begun in late 2009 with a collaboration and student exchange between the POBEX US team (V. Combes, E. Di Lorenzo) and South American team, has now received funding for the period 2010-2013 from the National Committee of Scientific and technologic Research of Chile. The goal is to use the transport modeling framework developed during the first phase of POBEX to study the dynamics of fish, in particular the Jack Mack, and their impacts on fishing. Chilean jack mackerel (CHJM), Trachurus murphyi Nichols, constitutes the most important fishery for Chile. CHJM is a migrating pelagic species which inhabits the Southern Pacific Ocean, and presents a wide distribution characterized by a fairly broad distribution band from Chile to New Zealand and Tasmania (Bailey 1989, Elizarov et al. 1992, Gretchina 1998). Single self-sustained population along the Chilean waters, which includes the oceanic fraction off central-south Chile (Serra, 1991).

The objectives of this task are (1) to implement an individual-based model (IBM) of Chilean Jack Mackerel (CHJM) in the South Pacific coupling to the POBEX ROMS PCCS transport model and (2) to assess the conceptual model of this resource in the South Pacific through the understanding of the spatial and temporal dynamic of the life history of CHJM. A schematic of the IBM model is illustrated below.

Conceptual Model for the Chilean Jack Mackarel CHJM exhibits a strong seasonal migration pattern (Serra 1991, Arcos et al. 2001) showing an onshore migration during the summer linked to coastal food availability. During fall and winter, jack mackerel aggregates in compact schools in coastal and oceanic waters off central Chile, being more available for the Chilean purse-seiner fleet (Arancibia et al. 1995). CHJM develops large offshore migration towards reproductive oceanic habitat in early spring which extend along the Southeastern Pacific Ocean (SPO), but mainly in oceanic waters off central Chile from the 82°W to beyond 90°W (Cubillos et al. 2008). Main spawning peak occurs between October and December, although it can extend from September to February (Grechina et al. 1998) in oceanic waters (at least from 92°W to 75°W) related to the Subtropical front. These hypothesis are tested with the IBM-OFES/ROMS-PCCS coupled biophysical model (see figure below).

Preliminary results form the modeling activities suggest that spawning areas and coastal domains are connected for Lagrangian drifters, but mostly particles are trapped in cyclonic and anticyclonic eddies. There is a significant internanual variability on transport and connectivity between spawning and coastal domains with maximum values in years 1999 and 2005. Mainly successful transport occur to Valparaiso and secondly to Caldera and Concepción domains. This domains are not recorded as real nursery grounds. Successful transport occurs from particles that origin on spawning areas around 82.5W and 36.5-37.5S (oceaS) and between 77.5-78.5W and 32.5-33.5S (oceaN). From these regions oceaN promote successful transport to Caldera region and oceaS mainly to Valparaíso and secondly to Concepción. Mainly successful transport to coastal domains occurs due to advection of meanders on Northeast direction (between eddies) while eddies are transported mainly in West direction.

Nursery areas (coastal domains) that model predict are a bit southern of the ones we have found Juveniles. Using Lagrangian drifters constitute the worst case scenario (no biological movement behavior). Even though this fact, spawning and coastal domains are connected. Larvae of 4-5 months old that is arriving to southern nursery areas can perform some movement behavior to swim northward to achieve realistic northern nursery grounds. The interannual variability on transport show that meanders formation can be promoted in some years, achieving high levels of connectivity under the same initial conditions (only assessing the relevance of transport and physical transport). Next step is develop more realist experiments including initial conditions of egg release associated to spawning events between 1999 to 2008 based on observational data. With this we will be assessing the relevance of spawning patterns on the connectivity as well. There is an open question what happened with the portion of particles that are trapped in cyclonic and anticyclonic eddies? Are oceanic eddies a suitable habitat for CHJM larvae and juveniles to survive?

During 2008-2009, we have collaborated with D. Putrahssan and A. Miller at Scripps to set up and test the coupled ocean-atmosphere model called SCOAR i the PCCS. SCOAR is a combination of the Regional Spectral Model (RSM) and the Regional Ocean Modeling System (ROMS). The PCCS grid covers the same domain used for the oceanic passive tracer experiments (see previous task) from 90W to 68W and 2S to 42S. The atmospheric and ocean model have the same horizontal grid with 20km resolution. However, the atmosphere has 28 vertical layers while the ocean has 30 layers. So far we have conducted an integration that covers the period 2000-2008 so that we can begun comparisons with uncoupled ocean model integration that use the QSCAT winds.Below is a figure from the coupled run

This task attempts to quantify the nutrient transport pathways and dynamics using forward and inverse passive tracers experiments in high-resolution ocean modeling frameworks.

During 2009/2010 we have completed the first high-resolution modeling experiment with KOE-ROMS using boundary conditions from the Japanese OFES model. The forward model was integrated between 1950-2009 and the output is available on the Georgia Tech OpenDAP data server (http://data.eas.gatech.edu). We are now in the process of performing quantitative analysis of the transport dynamics.

The goal of this task is to use the extensive zooplankton dataset provided by our Japanese collaborators to explore the links between changes in the physical ocean circulation and various indicators of the zooplankton community (e.g. abundance, diversity, average size). The approach is to use an ensemble of eddy resolving ocean models to explore how changes in advection may have impacted zooplankton.

During 2009/2010 we have been able to deploy our modeling strategy for the KOE. The first high-resolution modeling experiment with KOE-ROMS has been completed using boundary conditions from the Japanese OFES model. The forward model was integrated between 1950-2009. We are now in the process of using passive tracers to test specific hypotheses on the dynamics controlling the spatial distribution of zooplankton abundance and diversity. For this purpose we are using an adjoint model to integrate the passive tracer equations. This approach allows us to assess directly where the water masses found in specific zooplnankton sampling areas have originated from. This task is still ongoing.

The synthesis activity explores the upwelling and cross-shelf transport dynamics along the entire Pacific eastern boundary (GOA, CCS and PCCS) using the high-resolution (10 km) hindcasts of the Regional Ocean Modeling System (ROMS). For each of the eastern boundary regions of the Pacific we generate an ensemble of historical hindcasts covering the period 1950–present to separate the deterministic and intrinsic (mesoscale eddies) component of ocean variability. These model hindcasts available at http://data.eas.gatech.edu. We report results from three model domains illustrated in the Figure below. In each regional model domain a set of passive tracers is injected along the coast, both in the surface and subsurface, to quantify the transport dynamics. Below, we present a short summary of the primary findings on low-frequency transport variability along the eastern Pacific boundary. A synthesis study and comparison between these region is still ongoing.

The Gulf of Alaska
The marine ecosystem of the GOA is very rich despite an open ocean that is characterized by high nutrient and low chlorophyll-a concentration and coastal downwelling. Primary production of the GOA is limited by low-Fe, perhaps due to low aeolian Fe input (Martin and Fitzwater, 1988, Nature, Vol. 331, 341-343), and recent observational studies suggest that advection of iron-rich coastal water may be the primary mechanism controlling open ocean productivity. Specifically, there is evidence that mesoscale eddies along the coastal GOA entrain iron-rich coastal waters into the ocean interior. The statistics of coastal waters transport are computed using a model passive tracer, which is continuously released at the coast at the surface (white hatched box on Fig. 1). On average along the Alaska Current, we find that at the surface, while the advection of tracers by the average flow is directed towards the coast, consistent with the dominant downwelling regime of the GOA, it is the mean eddy fluxes that contribute to offshore advection into the gyre interior. South of the Alaskan Peninsula, both the advection of tracers by the average flow and the mean eddy fluxes contribute to the mean offshore advection. On interannual and longer time scales, the offshore transport of the passive tracer in the Alaskan Stream does not correlate with large scale atmospheric forcing, or with local winds (intrinsic variability). In contrast in the Alaska Current region, stronger offshore transport of the passive tracer coincides with periods of stronger downwelling (forced variability) in particular during positive phases of the PDO, which trigger the development of stronger eddies, known as “Haida” and “Sitka” eddies (Combes et al., 2007, Prog. Oceanogr., Vol. 75 (2), 266-286; Combes et al., 2009, J. Phys. Oceanogr., Vol. 39 (4), 1050-1059).

The California Current System
South of the GOA, the CCS is a typical eastern boundary current coastal upwelling system. To characterize the effects of linear (Ekman upwelling) and non-linear (eddy activity) circulation regimes on the statistic of low-frequency advection of coastal waters, an ensemble of passive tracers is released in the numerical model in the subsurface between 150 and 250-m depth at the coast (white hatched box on Fig. 1). The resultant concentration of tracer at the surface provides an index of coastal upwelling. We find that the low-frequency upwelling and the surface offshore transport of the upwelled nutrient-rich coastal water are strongly correlated with the alongshore wind stress, and with the NPGO (large scale climate variability index). Our results also show that the poleward California Undercurrent, at about 200-m depth, affects the alongshore transport and provides nutrient-rich waters to the Central California dominant upwelling cell. However, if we look only at the surface transport dynamics, the offshore transport of the surface coastal water is associated with mesoscale eddy activity; both surface and subsurface waters propagate offshore mainly through cyclonic eddies (top-right panel on Fig 1, vertical section) (Combes et al, 2010a, J. Phys. Oceanogr., submitted).

The Peru-Chile Current System
In the southern hemisphere, the PCCS is one of the world’s most productive regions in fish landings, providing ~20% of the world marine catches despite covering less than 1% of the world’s ocean surface. This high productivity results principally from the upwelling of nutrient-rich water into the photic zone. Both changes in surface wind and coastally trapped Kelvin waves control the variability of the coastal upwelling. We assessed the effect of ocean remote forcing by comparing the output of two model simulations that do and do not include the presence of waves at their boundaries. Similar to the analyses performed in the CCS, a passive tracer approach is used to characterize the coastal upwelling variability of the PCCS. The evolution of the passive tracers indicates that, off the coast of Peru, the ENSO strongly modulates the efficiency of coastal upwelling, principally due to the propagation of downwelling equatorial Kelvin waves rather than changes in the local wind stress. Our results show that the central Chile upwelling region is also very sensitive to the Kelvin waves generated at the equator, and is considerably reduced during strong El Niño events. Different model experiments have also been conducted to explore the sensitivity of the PCCS upwelling to air-sea fluxes of momentum (ECMWF vs. NCEP surface wind stress forcings). Similar to the CCS region, we find that mesoscale eddies play an important role in the offshore advection of nutrient-rich coastal waters. Both cyclonic and anticyclonic eddies (bottom-left panel on Fig. 1) are able to transport nutrient-rich coastal waters (Combes et al., 2010b, J. Phys. Oceanogr., submitted).
The passive tracer experiments, performed in the GOA, CCS, and PCCS, provide a dynamical framework to understand the dynamics of the upwelling/downwelling and horizontal transport of water masses. Indices derived from the passive tracer can be used as proxies of the vertical and horizontal advection of important biogeochemical quantities, essential in understanding the ecosystem variability along the Pacific eastern boundary.

References:

Interannual and decadal variations in cross-shore mixing in the Gulf of Alaska
Combes, V., E. Di Lorenzo and E. Curchister
Journal of Physical Oceanography, 2009, 39(4), 1050-1059, DOI: 10.1175/2008JPO4014.1. [ PDF ]

Cross-shore transport varaibility in the California Current: Ekman upwelling vs. eddy regime
Combes V., F. Chenillat, P. Riviere, E. Di Lorenzo, M. Ohman and S. J. Bograd
Journal of Physical Oceanography, 2010, submitted. [ PDF ]

Modeling interannual and decadal variability in the Humboldt Current upwelling system
Combes V., E. Di Lorenzo, F. Gomez, S. Hormazabal, T. P. Strub and D. Putrasahan
Journal of Physical Oceanogrpahy, 2010, submitted. [ PDF ]

Copepod species composition linked to ocean transport in the Northern California Current
Keister, J.E., N. Mariani, C.A. Morgan, E. Di Lorenzo, V. Combes and W.T. Peterson
Global Change Biology, 2010, submitted. [ PDF ]

The synthesis activities aims to compare the role of changes in transport dynamics of surface currents in controlling variability in Zooplankton across the boundary region of the KOE, CCS and PCCS. This task is coordinated by Keister and will require working in close collaboration with South American (Prada) and Japanese (Chiba). One important challange is to isolate relevant data sources for the PCCS system, which are generally not available.

So far we have been able to link transport dynamics to zooplankton only in the NEP (Keister et al., 2010; Di Lorenzo and Ohman, in prep)

Copepod species composition linked to ocean transport in the Northern California Current
Keister, J.E., N. Mariani, C.A. Morgan, E. Di Lorenzo, V. Combes and W.T. Peterson
Global Change Biology, 2010, submitted. [ PDF ]

A simple hypothesis for Zooplankton transitions in the NE Pacific
Di Lorenzo E. and M. Ohman
PNAS, 2010, in preparation. [ PDF ]

This task uses the State Space Model methodologies to compare how the seasonal cycle in the CCS and PCCS has changed over the period 1950-present. This task is led by A. Thomas in collaboration with NOAA collaborators. Although studies have been completed that address the phenology in the CCS, we are still in the process of completing the analysis for the PCCS and the comparisons between CCS/PCCS.

Phenology of Upwelling in the California Current
Changes in the amplitude and phasing of seasonal events (phenology) can affect the functioning of marine ecosystems. Phenology plays a particularly critical role in eastern boundary ecosystems, which are driven largely by the seasonal cycle of coastal upwelling. We examined interannual variability in the timing, evolution, intensity, and duration of coastal upwelling in the California Current large marine ecosystem (CCLME), identifying extended periods of high (1970s, 1998-2004) and low seasonally-integrated upwelling (1980-1995) and a trend towards a later and shorter upwelling season in the northern CCLME. El Niño years were characterized by delayed and weak upwelling in the central CCLME. Understanding the causes and ecosystem consequences of phenological changes in coastal upwelling is critical, as climate models project significant variability in the amplitude and phase of coastal upwelling under varying climate change scenarios.

Biological Response to Nonstationary Seasonality in the California Current
As a follow up to the study on upwelling phenology, we further investigated the seasonal structure of California Current upwelling and its impact on the ecology of sentinel species. We found two dominant modes of upwelling variability: the first reflects peak upwelling during the summer months while the second reflects wintertime ocean conditions. The “summer mode” is dominated by low-frequency processes including long-term increases within the central CCLME, while the “winter mode” is dominated by higher-frequency, interannual variability associated with ENSO. Both are biologically relevant to the ecosystem. In particular, we found that seabird reproductive success and rockfish growth were highly correlated with late winter upwelling, implying that anomalous late-winter upwelling can pre-condition the CCLME towards higher productivity. This study illustrates the importance of upwelling seasonality to the physical and biological structure of the CCLME.

References:

The phenology of coastal upwelling in the California Current
Bograd, S.J., I. Schroeder, N. Sarkar, X. Qiu, W.J. Sydeman and F.B. Schwing
Geophysical Research Letters, 2009, 36, L01602, doi:10.1029/2008GL035933. [ PDF ]

Marine ecosystems, climate and phenology: impacts on top predators
Sydeman, W.J., and S.J. Bograd
Marine Ecology Progress Series, 2009, 393, 185-188. [ PDF ]