Synthesis of Pacific Ocean Climate and Ecosystem Dynamics

Author(s): Di Lorenzo, E; Combes, V; Keister, JE; Strub, PT; Thomas, AC; Franks, PJS; Ohman, MD; Furtado, JC; Bracco, A; Bograd, SJ; Peterson, WT; Schwing, FB; Chiba, S; Taguchi, B; Hormazabal, S; Parada, C | Abstract: The goal of the Pacific Ocean Boundary Ecosystem and Climate Study (POBEX) was to diagnose the large-scale climate controls on regional transport dynamics and lower trophic marine ecosystem variability in Pacific Ocean boundary systems. An international team of collaborators shared observational and eddyresolving modeling data sets collected in the Northeast Pacific, including the Gulf of Alaska (GOA) and the California Current System (CCS), the Humboldt or Peru-Chile Current System (PCCS), and the Kuroshio-Oyashio Extension (KOE) region. POBEX investigators found that a dominant fraction of decadal variability in basin and regional-scale salinity, nutrients, chlorophyll, and zooplankton taxa is explained by a newly discovered pattern of ocean-climate variability dubbed the North Pacific Gyre Oscillation (NPGO) and the Pacific Decadal Oscillation (PDO). NPGO dynamics are driven by atmospheric variability in the North Pacific and capture the decadal expression of Central Pacific El Ninos in the extratropics, much as the PDO captures the low-frequency expression of eastern Pacific El Ninos. By combining hindcasts of eddy-resolving ocean models over the period 1950-2008 with model passive tracers and long-term observations (e.g., CalCOFI, Line-P, Newport Hydrographic Line, Odate Collection), POBEX showed that the PDO and the NPGO combine to control low-frequency upwelling and alongshore transport dynamics in the North Pacific sector, while the eastern Pacific El Nino dominates in the South Pacific. Although different climate modes have different regional expressions, changes in vertical transport (e.g., upwelling) were found to explain the dominant nutrient and phytoplankton variability in the CCS, GOA, and PCCS, while changes in alongshore transport forced much of the observed long-term change in zooplankton species composition in the KOE as well as in the northern and southern CCS. In contrast, cross-shelf transport dynamics were linked to mesoscale eddy activity, driven by regional-scale dynamics that are largely decoupled from variations associated with the large-scale climate modes. Preliminary findings suggest that mesoscale eddies play a key role in offshore transport of zooplankton and impact the life cycles of higher trophic levels (e.g., fish) in the CCS, PCCS, and GOA. Looking forward, POBEX results may guide the development of new modeling and observational strategies to establish mechanistic links among climate forcing, mesoscale circulation, and marine population dynamics. © 2013 by The Oceanography Society. All rights reserved.

through Rossby waves that propagate westward from the central and eastern North Pacific and modulate the KOE upon arrival.AL/PDO Rossby waves drive changes in the axis of the KOE (Miller and Schneider, 2000;Qiu et al., 2007;Taguchi et al., 2007), while NPO/ NPGO Rossby waves modulate the speed and strength of the KOE (Ceballos et al., 2009).A similar Rossby wave connection has been also isolated in the South Pacific (Holbrook et al., 2011).

Eastern Pacific ENSO and the Pacific Decadal Oscillation
In its positive phase, the "traditional"   and temperature that are captured in the PDO pattern (Figure 3b).These dynamical links allow the PDO to be modeled as a simple integrator of the AL variability and the EP-ENSO teleconnection (Newman et al., 2003;Schneider and Cornuelle, 2005;Vimont, 2005).The EP-ENSO projection index captures both the EP-ENSO and the AL stochastic contributions to North Pacific SLPa variability.A simple integration of the EP-ENSO projection index with an autoregressive model of order-1 (AR1) with a six-month memory time scale leads to a skillful reconstruction of the PDO index (Figure 3b; see Chhak et al., 2009, for details on the AR-1 approach).Niiler et al., 2003) showing how the NPGO impacts the mean gyre circulation and captures changes in the strength of the major gyre-scale currents, the North Pacific Current (NPC) and the Kuroshio-Oyashio Extension (KOE).The indices of NPC and KOE strength were computed from the Ocean General Circulation Model for the Earth Simulator (OFES) model (see Di Lorenzo et al., 2009).The NPGO tracks long-term changes in salinity, nutrients, and Chl-a observed in the California Current and the Gulf of Alaska (see Di Lorenzo et al., 2009, for a detailed explanation of the data).

Central Pacific ENSO and North Pacific Gyre Oscillation
for potential positive feedback between tropics and extratropics (Figure 1).While it is well known that ENSO variability can be triggered and energized by stochastic atmospheric variability in the tropics (e.g., westerly wind bursts,
Higher modes in Chl-a distributions reflect regional expressions of both largescale climate modes and local forcing, and need further investigation.
Changes in the timing and amplitude of the seasonal cycle likely exert dominant control on primary productivity.
Along the eastern boundary system, POBEX researchers described large, latitudinally dependent, low-frequency changes in the timing, duration, and intensity of coastal upwelling in the California Current (Bograd et al., 2009).
An important fraction of these changes in the southern CCS was linked to the NPGO (Chenillat et al., 2012).In the PCCS, changes in seasonal timing are still not well understood but are likely connected to changes in the seasonal cycle of the ENSO system.These phenological changes could be important, as many organisms have life histories that are closely adapted to the strong seasonal cycle (Barth et al., 2007).

Changes in Ocean Horizontal Transport Explain Observed Zooplankton Variability
Large-scale climate forcing was also found to affect regional-scale horizon- Fisheries Science Center off Newport, Oregon.There, zooplankton communities cycle in the relative dominance of cold-water or warm-water species (e.g., Peterson and Keister, 2003)

Mesoscale Eddies Control Cross-Shelf Exchanges and Impact Fish Habitats
While large-scale climate modes strongly controlled upwelling and alongshore transport variability, the cross-shelf exchanges diagnosed from passive tracers (Combes et al., 2009(Combes et al., , 2013) ) were not as coherent across the POBEX regions.Indeed, they were mostly independent of large-scale climate forcing.Mesoscale features (e.g., eddies and filaments), rather than wind-driven Ekman transport, dominate cross-shelf transport variability along the eastern boundary (Combes et al., 2009(Combes et al., , 2013)).These structures drive the offshore export of surface waters and subsurface waters of the eastern boundary undercurrents (Combes et al., 2013, for the CCS;Hormazabal et al., 2013, for the PCCS).Although a large fraction of mesoscale variance is internal to the ocean and unpredictable, regionalscale forcings were found to control the statistics of eddies in the Gulf of Alaska and the CCS, especially the anticyclones, which showed stronger low-frequency variability than the cyclones (Combes and Di Lorenzo, 2007;Combes et al., 2009;Davis and Di Lorenzo, in press).
There is growing evidence that mesoscale circulation features strongly impact ecosystem dynamics (e.g., primary productivity; McGillicuddy et al., 2007).
POBEX did not fully explore their influence on marine ecosystems due to a lack of adequate long-term observations that resolve eddy-scale processes.Still, studies in the NEP and PCCS suggest strong links between mescoscale circulation and the distribution of zooplankton and higher tropic levels (e.g., fish).Upwelling filaments transport significant portions of coastal zooplankton populations offshore (Keister et al., 2008; recent work of author Keister and Stephen Pierce, Oregon State University), resulting in offshore "hotspots" of upper trophic activity.
In the PCCS, a bio-physical modeling study highlighted the potential impact of mesoscale eddies on retention of jack mackerel (Trachurus murphyi, Nichols) along the Challenger Plateau and the East Pacific Rise, more than 3,500 km from historically known coastal nursery grounds and ocean spawning regions (Parada et al., in press).Retention for at least four months in anticyclonic eddies with their associated environmental conditions, such as sea surface temperature, Chl-a, wind, and turbulence levels, suggests strong recruitment in these features.
Evidence of jack mackerel juveniles in the region for over 20 years, obtained from Russian research vessel logbooks, support the hypothesis (Parada et al., in press).Lorenzo and Ohman (2013) skill (R = 0.6, Figure 8b).This integration by the ocean filters out the highfrequency variability of the white noise atmospheric forcing and enhances the decadal and lower-frequency variability (Figure 8b).Because zooplankton and other marine populations are sensitive to changes in ocean conditions, the ecosystem will integrate the white noise atmospheric forcing a second time-first integration from atmosphere to ocean, and second integration from ocean to marine populations.This double integration allows us to filter away even more high-frequency atmospheric variability and leads to time series of marine ecosystem variability that are dominated by low-frequency fluctuations (Figure 8c) As US GLOBEC has ended, it will be critical to find new sources of funding to support science networks like POBEX.
Synthesis of Pacific Ocean Climate and Ecosystem Dynamics B Y E M A N U E L E D I L O R E N Z O , V I N C E N T C O M B E S , J U L I E E .K E I S T E R , P. T E D S T R U B , A N D R E W C .T H O M A S , P E T E R J .S .F R A N K S , M A R K D .O H M A N , J A S O N C .F U R TA D O , A N N A L I S A B R A C C O , S T E V E N J .B O G R A D , W I L L I A M T. P E T E R S O N , F R A N K L I N B .S C H W I N G , S A N A E C H I B A , B U N M E I TA G U C H I , S A M U E L H O R M A Z A B A L , A N D C A R O L I N A PA R A D A S P E C I A L I S S U E O N U S G L O B E C : U N D E R S TA N D I N G C L I M AT E I M PA C T S O N M A R I N E E C O S Y S T E M S The main objectives of POBEX were to: (1) understand and quantify how large-scale climate variability drives the regional-scale physical variability that is coherent along the Pacific boundary, and (2) use regional-scale dynamics in combination with existing long-term ecological observations to interpret marine ecosystem processes.Specifically, POBEX quantified how changes in regional ocean processes (e.g., upwelling, transport dynamics, mixing, mesoscale structure) in each Pacific Ocean boundary region control phytoplankton and zooplankton dynamics and the extent to which large-scale climate modes such as the Pacific Decadal Oscillation (PDO; Mantua et al., 1997), the El Niño-Southern Oscillation (ENSO), and the recently discovered North Pacific Gyre Oscillation (NPGO; Di Lorenzo et al., 2008) drive these regional ocean dynamics.The underlying hypothesis of POBEX was that large-scale Pacific climate forcing drives changes in transport dynamics that exert dominant and coherent bottom-up control on coastal ocean ecosystems.To explore how interannual-todecadal variations in upwelling and horizontal transport affect the lower trophic levels of the Pacific boundary marine ecosystems, POBEX combined a series of historical (1950-present) eddy-resolving (10 km spatial resolution) ocean simulations at both global and regional scales with model passive tracers to generate temporal indices for transport processes such as upwelling and horizontal advection.The transport indices were then used to test the extent to which observed changes in phytoplankton are connected to modeled changes in the strength, structure, and timing of upwelling and to explore specific hypotheses concerning links between changes in modeled horizontal transport and changes in zooplankton abundance and species diversity.By exploring regional-scale dynamics, POBEX identified the important role of recently identified patterns of climate variability (e.g., central Pacific El Niños and NPGO) and clarified their largescale and regional-scale dynamics.This has brought improved understanding of the mechanisms of large-scale Pacific climate variability and their regionalscale impacts on the coastal ocean and marine ecosystems.

Figure 1 .
Figure 1.Synthesis of Pacific climate dynamics and teleconnections.The Pacific Decadal Oscillation (PDO; red path) and North Pacific Gyre Oscillation (NPGO; blue path) outline teleconnections at low-frequency time scales.The gray path shows how sources of high-frequency stochastic variability in the atmosphere energize the Aleutian Low (AL), North Pacific Oscillation (NPO), and El Niño-Southern Oscillation (ENSO) systems.In the schematic, NPO low-frequency variability is linked to Central Pacific (CP)-El Niño; however, processes internal to the North Pacific atmosphere appear to drive its high-frequency variability (gray path).

El
Niño is characterized by pronounced warming of the tropical eastern Pacific (e.g., EP-ENSO), weakening trade winds, and positive (negative) atmospherically forced sea level pressure anomalies (SLPa) over the western (eastern) tropical Pacific (Figure 2a).These changes in the tropical atmospheric circulation modify the large-scale Hadley Cell and drive an important fraction of the extratropical variability of the AL through the ENSO atmospheric teleconnection pattern (Figure 2a; Alexander, 1992; Alexander et al., 2002).This teleconnection between EP-ENSO and the AL is evident if we develop an index of the variability of the projection of the EP-ENSO pattern in SLPa in the North Pacific (Figure 2a, black box).This EP-ENSO projection index has significant correlation (R = 0.8) with an index of the AL variability defined as the first principal component of North Pacific SLPa (Figure 3a).The ocean integrates these changes in atmospheric forcing, acting as a filter that enhances the decadal energy of the AL atmospheric forcing and of the ENSO teleconnection.The EP-ENSO-derived variability of the AL drives changes in ocean circulation Emanuele Di Lorenzo (edl@gatech.edu) is Professor of Ocean Sciences, School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA, USA.
Figure 2. The flavors of ENSO and their teleconnections.Correlation patterns of NOAA sea surface temperature (SST) and National Centers for Environmental Prediction (NCEP) sea level pressure (SLP) monthly anomalies with (a) the eastern Pacific or canonical El Niño index (positive phase of ENSO) and (b) the central Pacific El Niño index.The ENSO indices are from the NOAA climate diagnostic center.Black rectangles show the atmospheric projections of a positive ENSO onto the North Pacific atmosphere, also referred to as the ENSO teleconnection.The purple rectangles shows the geographical domains targeted by the Pacific Ocean Boundary Ecosystem Climate Study (POBEX).
While many ecosystem fluctuations in the Pacific can be explained within the physical framework ofENSO and   PDO variability (e.g., Mantua et al.,   1997;Hare and Mantua, 2000), longterm observational time series from the California Cooperative Oceanic Fisheries Investigations (CalCOFI) and the Line-P program in the Gulf of Alaska show decadal-scale fluctuations that are not connected to the PDO.Using a set of historical ocean simulations with coupled physical-biological models, POBEX identified a new pattern of North Pacific decadal variability associated with changes in the strength of the subtropical and subpolar gyres-the North Pacific Gyre Oscillation (NPGO; Di Lorenzo et al., 2008; Figure 4).Defined as the second dominant mode of SSHa variability in the Northeast Pacific (180°W-110°W, 25°N-62°N), the NPGO explains the dominant decadal fluctuations of salinity, nutrient upwelling, and chlorophyll-a (CHL-a) in the NEP region (Figure 4;DiLorenzo et al., 2008Lorenzo et al., , 2009) ) as well as important state transitions in marine ecosystems (e.g., fish;Sydeman and Thompson, 2010;Cloern et al., 2010).The NPGO signature in SSTa tracks the second dominant mode of North Pacific SSTa(Bond et al., 2003).Modeling studies conducted within POBEX(Di Lorenzo et al., 2008;Chhak et al., 2009) revealed that the NPGO is the oceanic response to the North Pacific Oscillation (NPO; Figure3b), a well-known pattern of atmospheric variability.The NPO affects weather patterns, particularly storm tracks,

Figure 3 .
Figure 3.The forcing of the two dominant modes of North Pacific variability.The first (panel a, AL) and second (panel c, NPO) principal components of North Pacific SLP anomalies (SLPa) are compared to indices that track the ENSO atmospheric projections (black line) onto the North Pacific.The ENSO projection indices are derived by projecting the North Pacific SLPa pattern of the EP and CP-ENSO onto the SLPa field.The projection indices capture contributions to SLPa from both ENSO and internal variability in the North Pacific.The spatial correlation of the principal components with SLPa are shown in panels a and c.The AL and NPO drive the oceanic PDO and NPGO (panels b and d show the spatial correlation patterns in the SSTa).The time series of the PDO and NPGO (b and d) can be modeled with a simple auto-regressive model of order-1 forced by the two ENSO projection indices.

Figure 4 .
Figure 4. North Pacific Gyre Oscillation.Correlation map between the NPGO index and AVSIO satellite SSHa data (panel a, black contours are the mean dynamic topography;Niiler et al., 2003) showing how the NPGO impacts the mean gyre circulation and captures changes in the strength of the major gyre-scale currents, the North Pacific Current (NPC) and the Kuroshio-Oyashio Extension (KOE).The indices of NPC and KOE strength were computed from the Ocean General Circulation Model for the Earth Simulator (OFES) model (see DiLorenzo et al., 2009).The NPGO tracks long-term changes in salinity, nutrients, and Chl-a observed in the California Current and the Gulf of Alaska (see DiLorenzo et al., 2009, for a detailed explanation of the data).

Figure 1 ,
Figure 1, gray path), sources of extratropical stochastic variability associated with the NPO are also important in energizing the ENSO system.Past studies on the Seasonal Footprinting Mechanism (SFM; Anderson, 2003; Vimont et al., 2003) show that boreal wintertime variability in the positive phase of the NPO drives warm SST anomalies in the North Pacific that in turn propagate into the central tropical put of the global eddy-resolving OFES (Ocean General Circulation Model for the Earth Simulator) historical hindcast from 1950-2008 (Masumoto et al., 2004; Sasaki et al., 2008) as boundary conditions.The ROMS model historical simulations were combined with passive tracer experiments (Figure 5) that allowed us to (1) diagnose circulation dynamics, such as upwelling, and alongshore and cross-shelf transport, and (2) explore how changes in transport are linked to ecosystem dynamics (e.g., Chhak et al., 2009; Combes

Figure 5 .Figure 6 .
Figure 5. Regional Ocean Modeling System (ROMS) passive tracer transport experiments.Map showing a February 1998 snapshot of passive tracers released from the coast in the nested eddy-resolving ROMS hindcast from 1950 to 2008.The passive tracers were used to study ocean transport dynamics in each of the POBEX regions.The time series panels show how indices of upwelling defined from the passive tracers are linked to different Pacific climate indices in the different POBEX regions (see Combes et al., 2009 and 2013, for details on definitions).

Figure 7 .
Figure 7. Schematic of ocean transport changes during the PDO (a) and the NPGO (b) and their impacts on zooplankton in the eastern and western boundaries of the North Pacific.The line plots in the Northern (N) and Southern (S) California Currents System (CCS) and the Kuroshio-Oyashio Extension (KOE) compare zooplankton time series (green lines) with zooplankton reconstruction based on indices that capture the changes in local transport (black lines).The local indices of transport in the CCS were connected to the PDO (red lines) and in the KOE to the NPGO (blue line).The definition and the dependence of the transport indices (black lines) on the climate modes are explained in the references placed beside the line plot panels.
in correlation with the PDO-cold-water species vastly dominate the community during cold (negative) phases of the PDO whereas warm-water species are more important during warm (positive) phases.Passive tracer experiments in ROMS showed that surface current variability associated with the PDO is tightly correlated to the observed changes in copepod species composition (Figure 7, NCCS).Stronger equatorward advection over the shelf during the negative phase of the PDO is reversed during the PDO's positive phase, with anomalously strong poleward currents and downwelling conditions.When smoothed over multiyear time scales, these transport fluctuations explain nearly all of the variance of the copepod community (R = -0.96).Similarly, in the SCCS, a simple PDO-transport model captured almost all of the low-frequency variance in abundances of the euphausiid Nyctiphanes simplex (R = 0.82) (Figure 7, SCCS; Di Lorenzo and Ohman 2013).Interestingly, the PDO imprint on the SCCS zooplankton-like the ocean response to atmospheric fluctuations-was best modeled as an integrated response to the physical forcing (e.g., the transport index in Figure 7 for the SCCS is obtained by integrating the PDO; see Di Lorenzo and Ohman, 2013).Note that in Figure 7b, the NPGO mode captures changes in the North Pacific Current and coastal upwelling.However, along the US West Coast, the PDO better captures the strength of the coastal currents.This explains the stronger correlation that exists between zooplankton and PDO in the coastal region.Similarly, changes in the strength of the KOE control variability in the abundance of warm-water copepods in the Kuroshio-Oyashio Transition (KOT) region (Figure 7; Chiba et al. 2013).Rossby waves excited in the central North Pacific by the NPO/NPGO system drive these changes.The waves arrive in the KOE with an approximate lag of 2.5 years following changes in the central and eastern North Pacific (Figure 7, KOE).Passive tracer experiments show that during years of a weak KOE (-NPGO), warm-water species are transported farther north and are retained in the KOT region (Figure 7), leading to the observed zooplankton anomalies (Chiba et al., 2013).Together, these studies provide strong evidence that large-scale climate changes affect marine ecosystems coherently around ocean boundaries through changes in ocean transport.This finding, referred to as the "horizontal advection bottom-up forcing hypothesis, " was shown to be important in driving marine ecosystem variability in all the US GLOBEC regions (Di Lorenzo et al., 2013, in this issue).In the Pacific, the different lags inherent in the processes complicated the discovery of common signals, but they can provide a degree of predictability to ecosystem changes when the mechanisms are sufficiently well understood.

Figure 8 .
Figure 8. Schematic of the double integration hypothesis.(a) Aleutian Low index.(b) Integrated Aleutian Low index (black) and PDO index (blue).(c) Integrated PDO index (blue) and zooplankton time series of Nyctiphanes simplex in the California Current System (red).Reproduced from DiLorenzo and Ohman (2013) with large-amplitude state transitions that can persist for decades (i.e., a super red power spectrum).Long-term time series of zooplankton in the CCS that are sensitive to PDO-related ocean advection were reconstructed with high skill by integrating the PDO index with an AR-1 model (R = 0.82), supporting the double-integration hypothesis (DiLorenzo and Ohman, 2013).These results point to the important need to develop proper null hypotheses of the variability expected in marine populations (e.g., fish) that result from cumulative integration of one or more environmental forcings before interpreting time series of biological or physical variables in terms of (or as indicative of) nonlinear "regime shifts." This need is particularly relevant for correctly assessing the underlying causes and significance of apparent "state transitions" and climate change signatures in marine ecosystems.For example, the double integration of white noise is a linear model that can lead to regime-like variability in ecosystem time series without the need for nonlinear dynamics to be involved.Climate Change Impacts on Marine EcosystemsUnderstanding and modeling the impacts of climate change on marine ecosystems remains an important challenge.Given the important role that Pacific Ocean decadal modes play in driving a large fraction of the low-frequency variability observed in long-term ecological time series (e.g., Mantua and Hare, 2002), it is critical that we understand and constrain the statistics and dynamics of these modes under changing climate conditions.Although International Panel on Climate Change (IPCC)-class climate models provide an important tool for predicting changes in the physical environment, POBEX found that these models are still unable to properly capture the statistics of large-scale Pacific decadal climate modes (Furtado et al., 2011).From modeling point of view, this finding raises additional challenges and questions about how to properly use IPCC-class models to downscale climate scenarios and examine climatechange on marine ecosystems (e.g., PICES Working Groups 20 and 29; http://www.pices.int).It also points to the need for advancing our mechanistic understanding of the links between physical climate and ecosystems (e.g., POBEX) that can lead to hypotheses and reduced-order process models of marine ecosystem responses to climate change.These process models offer an alternative approach for exploring the sensitivity of marine ecosystems to climate POBEX, PICES Working Group 27).In addition to developing proper modeling strategies to address climate change impacts on marine ecosystems, it is also clear that long-term observations of marine ecosystem variability are important.Unfortunately, these longterm observations are rare and often not widely available to researchers because of the nature of the sampling programs that depend on different regional organizations.One path toward accessing and analyzing these data sets is the development of social networks of Pacific scientists from different countries.The success of the POBEX project heavily relied on a prompt exchange of data and methods among the members of the international team.This international collaboration was made successful by prolonged student exchanges (three months to one year) among universities in Japan, Chile, and the United States.In this context, the activities of international organizations such as PICES were invaluable for building the exchange channels that led and maintained POBEX activities.POBEX also leveraged and partnered with the ongoing California Current Ecosystem Long Term Ecological Research (CCE-LTER; http://cce.lternet.edu).This partnership builds on the process studies and modeling activities carried out by CCE-LTER, together with the 65-year CalCOFI time series program.These research collaborations between US GLOBEC and intergovernmental organizations like PICES and US-funded LTERs provide the necessary infrastructure for addressing the scientific and societal issues of climate change impacts on marine resources.While US GLOBEC POBEX was a short-term, four-year project, it was through these partnerships that POBEX scientists were able to conduct a broad range of climate and ecosystem interdisciplinary research (e.g., http://www.pobex.org).