Predictability of the Galápagos Archipelago Upwelling Plume (P-GUP)

Sea surface temperature anomaly - December 2017

Galápagos Islands © NOAA

Alberto Naveira Garabato(1) Alexander Forryan(1), Clément Vic(1) & G. Nurser(2)

(1)University of Southampton (2)National Oceanography Centre, Southampton The Challenge § Galápagos upwelling sustains an iconic biological hot spot, which underpins:

§ one of the world’s largest UNESCO World Heritage Sites & Marine Reserves

Predictability of the Galápagos Archipelago Upwelling Plume (P-GUP) The Challenge The waters surrounding the Galápagos Archipelago (GA), located ~600 miles to the west of mainland Ecuador, are an iconic and globally significant biological hot spot (Fig. 1a). The region features an exceptionally productive and diverse ecosystem that is sustained by upwelling events – upward surges of deep, cold, nutrient-rich waters (collectively referred to as the GA upwelling plume) that fuel the growth of the phytoplankton biomass upon which the entire ecosystem thrivesThe1. ChallengeThe GA upwelling plume is of critical strategic importance to Ecuador on several grounds. First, it is a primary supply of food for the extensive stocks of tuna and shrimp in the eastern and central Equatorial § GaláPacific,pagos upwelling which supportsustains ana majoriconic biological fisheries hot industry spot, which responsible underpins: for the third highest source (approximately $1 billion) of national export revenue2. Second, it is key to sustaining a range of coastal communities (including >50,000 fishers) in § one of the world’s largest UNESCO World Heritage Sites & Marine Reserves the GA and mainland Ecuador whose economy and way of life is founded on artisanal fishing in waters fertilised by 3 § thea major upwelling (US$1 billion) plume tuna. And and thirdshrimp, it fisheries underpins industry one of the largest UNESCO World Heritage Sites and Marine Reserves in the world, universally recognised as having ‘outstanding national and international importance for conservation’4. Modified from Palacios et al. (2006)

FigureSatellite-measured 1. (a) Avera chlorophyllge (shaded) concentration and locations of of satellite Ecuadorian - industrialmeasured tuna chlorophyll (shaded) and locations of Ecuadorian industrial tuna fishingfishing (dots). (dots) White in dotted 2006 line: -Marine 2009 Reserve in the boundary. eastern Equatorial Pacific. The boundary of the GA Marine Reserve is indicated by the white dotted line. (b) Satellite-derived time series of the area and chlorophyll biomass of the GA upwelling plume1. A foremost challenge in sustainably managing the extraordinary natural resources of the GA upwelling plume against the mounting pressures of human exploitation and development is the acute episodicity and climate sensitivity of upwelling events (Fig. 1b), the causes of which are unknown1. Variations in the frequency and intensity of upwelling events have in the past brought about abrupt fluctuations in the region’s ecosystem state and fisheries yields, leading to a call for generating a prediction system to anticipate and mitigate the potentially severe conservation and socio- economic impacts of changes in upwelling5. The pressing need for developing predictability and assessing resilience of the GA upwelling plume is underscored by projections of regional climate change over the 21st century, which indicate that the frequency and severity of Equatorial Pacific climate variability (largely associated with the El Niño – Southern Oscillation, ENSO) will increase as Earth warms6. In this proposal, we will respond to this need by laying the foundations of a GA upwelling plume prediction system. We will do this by fulfilling two objectives. Objectives 1. To define, quantify and provide mechanistic understanding of the key processes controlling the rate of GA upwelling. 2. To design and implement an ocean circulation model with a cost-effective representation of these key processes – to support the future development of a GA upwelling plume prediction system. Research Approach and Timeline A first step in developing an operational GA upwelling plume prediction system is to identify the mechanisms that control how local upwelling responds to perturbations in external forcings, so that those mechanisms can be effectively included in the prediction system. Observational studies1,7-9 indicate that forcings of GA upwelling can be grouped into two classes: (i) changes in atmospheric conditions (e.g., winds) over the GA; and (ii) large-scale changes in the depth of the thermocline in the eastern Equatorial Pacific (triggered by e.g., ENSO events or tropical instability waves). The sporadicity and short duration of upwelling events (Fig. 1b) suggest that these often occur as a simultaneous, interacting response to the two types of forcing1,7. Thus, our research approach must consider how GA upwelling reacts to each isolated forcing, and to both forcings at once. To address objective (1), we will construct a numerical model of the GA upwelling plume region, subject it to idealised representations of each and both forcings, and examine the modelled ocean’s dynamical response with a focus on upwelling. Our approach will capitalise on state-of-the-art dynamical expertise and modelling techniques developed within our group as part of the NERC-funded OSMOSIS consortium (see Applicant Career Summary). The model will be configured using the MITgcm within the domain in Fig. 1a, with a horizontal resolution of 0.5 km that permits the development of the submesoscale instabilities and topographic interactions underpinning vertical motion. Simulations will be initialised with a density field mimicking a ‘shallow’, ‘average’, or ‘deep’ thermocline state of the GA region, defined from observations and sustained via boundary conditions. A baseline of 9 experiments is planned, in which the climatological average and two representative extremes of atmospheric forcing (derived from atmospheric reanalysis data) will be alternately applied to the three thermocline configurations of the model. Supplementary simulations will be conducted to deepen dynamical insight into the generation of upwelling events in the baseline experiments. A major scientific outcome of this research will be pioneering understanding of how 1

Predictability of the Galápagos Archipelago Upwelling Plume (P-GUP) The Challenge

The waters surrounding the Galápagos Archipelago (GA), located ~600 miles to the west of mainland Ecuador, are Predictability of the Galápagos Archipelago Upwelling Plume (P-GUP) an iconic and globally significant biological hot spotThe Challenge(Fig. 1a). The region features an exceptionally productive and The waters surrounding the Galápagos Archipelago (GA), located ~600 miles to the west of mainland Ecuador, are diverse ecosystem that is sustained by upwellingan iconic events and globally – significant upward biological hotsurges spot (Fig. 1 a) of. The deep,region features cold an exception, nutrientally productive-rich and waters diverse ecosystem that is sustained by upwellingThe events Challenge – upward surges of deep, cold, nutrient-rich waters (collectively referred to as the GA upwelling plume)(collectively that fuel referred the to as thegrowth GA upwelling of plume the) that phytoplankton fuel the growth of the phytoplankton biomass biomass upon upon which the which the 1 entire ecosystem thrives1. The GA upwelling plume is of critical strategic importance to Ecuador on several grounds. entire ecosystem thrives . The GA upwelling plume§ ManagementFirst, is it ofis a primarycritical supply of stof the foodrategic for tension the extensive import between stocksance of tuna conservationand to shrimp Ecuador in the eastern and onand central industrialseveral Equatorial grounds. exploitation Pacific, which support a major fisheries industry responsible for the third highest source (approximately $1 billion) of First, it is a primary supply of food for the extensivemadenational stocks export extremely revenue of 2tuna. Second, challenging itand is key toshrimp sustaining by thea range in acute ofthe coastal eastern variabilitycommunities (includi and andng >central 50,000 climate fishers ) Equatorialin sensitivity of the GA and mainland Ecuador whose economy and way of life is founded on artisanal fishing in waters fertilised by Pacific, which support a major fisheries industry responsibleGalthe áupwellingpagos plume upwelling 3for. And thirdthe, it underpins–third the causes onehighest of the largest of whichsource UNESCO areWorld ( approximately unknownHeritage Sites and Marine Reserves $1 billion ) of in the world, universally recognised as having ‘outstanding national and international importance for conservation’4. national export revenue2. Second, it is key to sustaining a range of coastal communities (including >50,000 fishers) in the GA and mainland Ecuador whose economy and way of life is founded on artisanal fishing in waters fertilised by the upwelling plume3. And third, it underpins one of the largest UNESCO World Heritage Sites and Marine Reserves in the world, universally recognised as having ‘outstanding national and international importance for conservation’4.

Adapted from Palacios et al. (2006) Figure 1. (a) Average concentration of satellite-measured chlorophyll (shaded) and locations of Ecuadorian industrial tuna fishing (dots) in 2006 - 2009 in the eastern Equatorial Pacific. The boundary of the GA Marine Reserve is indicated by the white dotted line. (b) Satellite-derived time series of the area and chlorophyll biomass of the GA upwelling plume1. A foremost challenge in sustainably managing the extraordinary natural resources of the GA upwelling plume against the mounting pressures of human exploitation and development is the acute episodicity and climate sensitivity of upwelling events (Fig. 1b), the causes of which are unknown1. Variations in the frequency and intensity of upwelling

events have in the past broughtElNiño about abrupt fluctuations in the region’s ecosystem state and fisheries yields, leading to a call for generating a prediction system to anticipate and mitigate the potentially severe conservation and socio- economic impacts of changes in upwelling5. The pressing need for developing predictability and assessing resilience of the GA upwelling plume is underscored by projections of regional climate change over the 21st century, which indicate that the frequency and severity of Equatorial Pacific climate variability (largely associated with the El Niño – Southern Oscillation, ENSO) will increase as Earth warms6. In this proposal, we will respond to this need by laying the foundations of a GA upwelling plume prediction system. We will do this by fulfilling two objectives. Objectives 1. To define, quantify and provide mechanistic understanding of the key processes controlling the rate of GA upwelling. 2. To design and implement an ocean circulation model with a cost-effective representation of these key processes – to support the future development of a GA upwelling plume prediction system. ResearchSatellite-measured Approach and timeTimeline series of the area and chlorophyll biomass of the Galápagos upwelling plume Figure 1. (a) Average concentration of satellite-measuredA first step chlorophyll in developing an operational (shaded) GA upwelling and plume locations prediction system of is Ecuadorianto identify the mechanisms industrial that tuna control how local upwelling responds to perturbations in external forcings, so that those mechanisms can be fishing (dots) in 2006 - 2009 in the eastern Equatorialeff Pacifectively includedic. The in the predictionboundary system. Observationalof the GA studies Marine1,7-9 indicate that Reserve forcings of GA isupwelling indicated can be by the grouped into two classes: (i) changes in atmospheric conditions (e.g., winds) over the GA; and (ii) large-scale1 white dotted line. (b) Satellite-derived time series of thechan areages in the and depth ofchlorophyll the thermocline in the biomass eastern Equatoria ofl Pacificthe (triggeredGA upwelling by e.g., ENSO events plume or tropical. instability waves). The sporadicity and short duration of upwelling events (Fig. 1b) suggest that these often occur as a simultaneous, interacting response to the two types of forcing1,7. Thus, our research approach must consider how A foremost challenge in sustainably managing the extraordinaryGA upwelling reacts to each natural isolated forcing resources, and to both forcings of at once the. GA upwelling plume against To address objective (1), we will construct a numerical model of the GA upwelling plume region, subject it to the mounting pressures of human exploitation andidealised development representations of each is and the both forcings, acute and examineepisodicity the modelled ocean’s and dynamical climate response sensitivity with a of focus on upwelling. Our1 approach will capitalise on state-of-the-art dynamical expertise and modelling techniques upwelling events (Fig. 1b), the causes of which aredeveloped unknown within our group. Variations as part of the NERC in-funded the OS frequencyMOSIS consortium (andsee Applicant intensity Career Summary of ).upwelling The model will be configured using the MITgcm within the domain in Fig. 1a, with a horizontal resolution of 0.5 km events have in the past brought about abrupt fluctuationsthat permits thein developmentthe region’s of the submesoscale ecosy instabilitiesstem and state topographic and interactions fisheries underpinning yields, vertical leading motion. Simulations will be initialised with a density field mimicking a ‘shallow’, ‘average’, or ‘deep’ thermocline state to a call for generating a prediction system to anticipateof the GA region and, defined mitigate from observations the and potentially sustained via boundary severe conditions. conservationA baseline of 9 experiments and is socio- 5 planned, in which the climatological average and two representative extremes of atmospheric forcing (derived from economic impacts of changes in upwelling . The pressingatmospheric reanalysisneed data for) will developing be alternately applied predictability to the three thermocline and configurations assessing of the model. resilience Supplementary simulations will be conducted to deepen dynamical insight into the generation of upwellingst events in of the GA upwelling plume is underscored by projectionsthe baseline experiments. of regional A major scientific climate outcome of change this research will over be pioneering the understanding 21 century, of how which indicate that the frequency and severity of Equatorial Pacific climate variability 1(largely associated with the El Niño – Southern Oscillation, ENSO) will increase as Earth warms6. In this proposal, we will respond to this need by laying the foundations of a GA upwelling plume prediction system. We will do this by fulfilling two objectives. Objectives 1. To define, quantify and provide mechanistic understanding of the key processes controlling the rate of GA upwelling. 2. To design and implement an ocean circulation model with a cost-effective representation of these key processes – to support the future development of a GA upwelling plume prediction system. Research Approach and Timeline A first step in developing an operational GA upwelling plume prediction system is to identify the mechanisms that control how local upwelling responds to perturbations in external forcings, so that those mechanisms can be effectively included in the prediction system. Observational studies1,7-9 indicate that forcings of GA upwelling can be grouped into two classes: (i) changes in atmospheric conditions (e.g., winds) over the GA; and (ii) large-scale changes in the depth of the thermocline in the eastern Equatorial Pacific (triggered by e.g., ENSO events or tropical instability waves). The sporadicity and short duration of upwelling events (Fig. 1b) suggest that these often occur as a simultaneous, interacting response to the two types of forcing1,7. Thus, our research approach must consider how GA upwelling reacts to each isolated forcing, and to both forcings at once. To address objective (1), we will construct a numerical model of the GA upwelling plume region, subject it to idealised representations of each and both forcings, and examine the modelled ocean’s dynamical response with a focus on upwelling. Our approach will capitalise on state-of-the-art dynamical expertise and modelling techniques developed within our group as part of the NERC-funded OSMOSIS consortium (see Applicant Career Summary). The model will be configured using the MITgcm within the domain in Fig. 1a, with a horizontal resolution of 0.5 km that permits the development of the submesoscale instabilities and topographic interactions underpinning vertical motion. Simulations will be initialised with a density field mimicking a ‘shallow’, ‘average’, or ‘deep’ thermocline state of the GA region, defined from observations and sustained via boundary conditions. A baseline of 9 experiments is planned, in which the climatological average and two representative extremes of atmospheric forcing (derived from atmospheric reanalysis data) will be alternately applied to the three thermocline configurations of the model. Supplementary simulations will be conducted to deepen dynamical insight into the generation of upwelling events in the baseline experiments. A major scientific outcome of this research will be pioneering understanding of how 1 P-GUP Objectives § To define, quantify and provide mechanistic understanding of the key processes controlling the rate of Galápagos upwelling

§ To design and implement an ocean circulation model with the key processes – to support the follow-up development of a Galápagos upwelling prediction system Bathymetry of the model domain (technical details available on request)

NERC-funded research on the Ocean Surface Team of Ecuadorian stakeholders Boundary Layer (OSMOSIS) § Representatives of Ecuadorian Government, Navy & National Institute of Fisheries § Advances in upper-ocean physics § Conservation organisations (WWF, Conservation International) § High-resolution (submesoscale) ocean modelling § Ecuadorian biologists and economists at the Universidad de San techniques Francisco de Quito Key result § Galápagos upwelling is controlled by northward winds interacting with sharp ocean fronts at the NW and SW tips of the archipelago

2-d histogram of the number of floats (released at 50-150 m west of 100oW) surfacing in each model grid point © Eric d’Asaro (UW) Outlook § Ongoing translation of new physical understanding into a basic prediction system design

§ Ongoing transfer of model to INOCAR (Oceanographic Branch of Ecuadorian Navy) for development of operational prediction system

§ Knowledge exchange workshop in Puerto Ayora, Galápagos Islands, to discuss policy-relevant outcomes – May 2018

§ Follow-up proposal to Schmidt Foundation to investigate ecosystem response to perturbations in Galápagos upwelling – June 2018