Download - the Spanish European Eel population recruitment ... · The volume of catches of European Eel has suffered a severe decline as it is reflected in all available historical series (Dekker,

Transcript
Page 1: the Spanish European Eel population recruitment ... · The volume of catches of European Eel has suffered a severe decline as it is reflected in all available historical series (Dekker,

European Geosciences Union, General Assembly 2016

Vienna | Austria | 17 – 22 April 2016

Influence of the North Atlantic oceanograghic and climatic parameters onthe Spanish European Eel population recruitment: relationships in the

past and for a future climate changeJaime Ribalaygua1, Javier Pórtoles1, Robert Monjo1, Estíbaliz Díaz2, María Korta2 and Guillem Chust2

¹ Climate Research Foundation (FIC, http://www.ficlima.org/), [email protected] AZTI-Tecnalia, Basque Country, Spain

Introduction The European Eel

Methodology and Results

Conclusions on the Causes of Decline

Acknowledgements

The status of the European eel population is critical; the annual recruitment of glass eel to European waters in 2015 is 1.2% ofthe 1960–1979 level in the „North Sea‟ area, and 8.4% in the rest of Europe (ICES 2015) . There are a number of anthropogenicimpacts potentially affecting eel population including commercial exploitation, habitat loss, dam and weir construction,hydropower, pumping stations and surface water abstractions. Furthermore, the first eel stages and larval migration and marinesurvival are heavily influenced by oceanic and climatic factors since the species breeds in the Sargasso Sea and migrates to thecontinental shelf of the Atlantic coast of Europe and North Africa. Therefore, the study of the relations between recruitment andoceanic conditions may allow to study the potential effect of climatic change on the future eel recruitment and therefore stock.

In the present study, the relation between glass eel recruitment and oceanic and climatic factors has been studied. Historic glasseel catches data beginning in the 50s from two Mediterranean and two Atlantic estuaries have been used as a proxy ofrecruitment. The relation of catches with the main oceanographic and climatic factors identified in the literature was establishedusing an ocean reanalysis, the Simple Ocean Data Assimilation (SODA) and determined which variables are significantly relatedto the number of catches. The analysis shows significant relationships between catches and oceanic (Surface Downward Stress,Sea Water Temperature and Sea Water Velocity) and atmospheric (NAO Index, AMO Index) variables. Subsequently, we appliedthe results of three climate models (GFDL-ESM2M, CanESM2 and CNRM-CM5), associated with the Coupled ModelIntercomparison Project Phase 5 (CMIP5) under two simulations of climate change (RCP4.5 and RCP8.5), both associated withthe 5th Assessment Report of the IPCC, for possible future influences on the eel.

Data and Study Area

Local recruiment data

For Atlantic analysis

Oceanic variables from the OceanicReanalysis SODA* :

T AU (tx, ty)≡ Surface downward stress

U, V (0 – 300 m) ≡ Oceanic currentsT (0 – 300 m) ≡ Oceanic temperature

Indexes:AMO* * & winter NAO* * * (Enfield et al, 2001; Cropper et al, 2015)

* SODA (Simple Ocean Data Assimilation –http://www.atmos.umd.edu/~ocean/) (Carton et al, 2008). ** AMO (Atlantic Multidecadal Oscillation): Surface temperature anomalies (not tend) of theNorth Atlantic (0 to 70º N) (Enfield et al, 2001) *** NAO: (North Atlantic Oscillation) Ponta Delgada (Azores, Islands Portugal) and Reykjavik(Iceland) (Cropper et al, 2015).

Due to: Oceanographic changes Over-fishing Alteration of rivers

Recruitment time series: stationarity

Before studying theirs relationships with meteorological series, it is necessary to treat the recruitment series of eels for making them stationary and avoid false relationships between series as a result of:

1. The negative trend in the volume of catches observedfrom the 80's.

2. Abrupt jumps and heterogeneities.

For avoiding these problems,

We have used a log transformation.

We have work with the residues, ie, its has been performed a trend extraction.

Autocorrelation analysis (ACF Autocorrelation Function) series before and after its treatment.

Region EXTENSIÓN NÚCLEOSARGASSO

ALLIANCELONGITUD

Number of sig. Correl.

158 208 151 222

Capture area

ALBUFERA EBRO MIÑO NALÓN

OceanicRegion

EXTENSION 1 50 45 62

NUCLEO 27 48 76 57

SALLIANCE 20 43 34 54

LONGITUD 37 53 63 69

Evolution of identified relations under future scenarios of climate change

Variable TAU X TAU Y T U V

Nº rel. signif 39 16 139 414 131

Delays in years 0 1 2 3

Nº sig, relations 158 208 151 222

Climatic data

This project has been funded by the Spanish Ministry of Agriculture, Food and Environment

Direct anthropogenic factors:

Over-fishingObstacles in rivers: dams, pipelines, etc.

Natural variability factors:

NAO: changes in currents and phytoplanktonGulf Stream: changes in the intensity and extentLocal rivers flow: too low levels

Factors due to climate change (natural + anthropogenic):

Changes in the extent and durability of large circulation systems (polar storms, Azores anticyclone): effect on the NAO.

Changes in the temperature of the North Atlantic and in the salinity due to the melting of Geoland: effect on Gulf Stream (- intensity and + subsidence).

Changes in local rainfall: level of critical runoff

- 4 local capture areas (Valencia, Ebro, Miño and Nervión).

- 4 atlantic regions (Sargasso Alliance, Longitud, Núcleo and Extensión)

- 5 variables: 2 (tx, ty) + 3 (T, u, v) × 17 levels (de 0 a 300m).

- 4 possible delays: 0, -1, -2 and -3 years

- 2 statistical aggregates for each area: mean and median

Oceanic relationships.

6785 statistic inspections 739 significant correlations

1. Very sharp decreases in different periods.2. Heterogeneity of series.

The European eel (Anguilla anguilla) has a long and complex life cycle that isstill not well enough known. The species breeds in the Sargasso Sea(approximately between 22 and 30° North, and between 48 and 65° West) and,from there, the larvae are carried by ocean currents to the continental shelf ofthe Atlantic coast of Europe and North Africa, where they transform intoelvers, a migration that can last more than 2 years. From here, the elversmigrate upstream, passing in the rivers most of his adult life to reach thestage of silver eel, which is when it reaches maturity and must migrate backto the Sargasso Sea to spawn and die. This complex life cycle, which includesdiverse ecosystems and long migrations, causes eels are exposed to variousenvironmental conditions which strongly influence their possible survival.

The volume of catches ofEuropean Eel has suffered asevere decline as it is reflectedin all available historical series(Dekker, 2003).

Fig.2 Decline in Eel catches for Europe(recruitment index) as shown in the report of theInternational Council for the Exploration of theSea, 2015 (ICES - International Council for theExploration of the Sea, http://www.ices.dk).

Climatic ModelSpatial

resolutionTemporalresolution

Avaiblescenarios

Research Center

GFDL-ESM2M 2ºx2,5º daily

HistoricalRCP2.6RCP4.5RCP8.5RCP6.0

National Oceanic and Atmospheric Administration

(NOAA), E.E.U.U.

CanESM2 2,8ºx2,8º daily

HistoricalRCP2.6RCP4.5RCP8.5

Canadian Centre for Climate Modeling and Analysis (CC-

CMA), Canadá

CNRM-CM5 1,4ºx1,4º daily

HistoricalRCP2.6RCP4.5RCP8.5

Centre National de Recherches

Meteorologiques (CNRM), Meteo-France, Francia

The study has taken into account that the determinant area for the reproduction of eels and the birth of new larvae is theSargasso Sea: any possible change on the Sargasso Sea, which is given by climatic conditions, will impact significantly on the eelpopulation since directly impact on its reproduction.The Sargasso Sea is a geographical area of imprecise boundaries since its size fluctuates according to prevailing currents thatcreate this particular space. It is a sea of calm waters not limited by any land that work as a system of surface waters relativelywarm which slowly rotate over the deep ocean waters, much colder and dense. This temperature difference has importantecological consequences.Due to the difficulty of defining this Sea, it has been considered four possible oceanic areas we could give as determining to theSargasso Sea and therefore the ocean area that affects the population of European Eel.

Fig. 3-6. Geographical definition (red dots) of each studied region. All points (red and blue) correspond to the grid of the SODA reanalysis. The truncated ellipse is the study area of the Sargasso Sea Alliance. The Sargasso Sea Commission

(http://www.sargassoseacommission.org/) is an international scientific organization focused on the study and conservation of the Sargasso Sea.

The working method consists in looking for possible relationships between recruitment, using the volume capture of elvers as an indicator of the same, and ocean data, measured in each of these four regions, seeking to empirically find which of these possible "definitions" of Sea of Sargasso shows greater relationship with our data collected of eels capture.

Figs 8-14 annual increases expectedfor the XXI century, represented asmoving averages 30 years, accordingto the RCPs represented (4.5 and8.5) compared to the average for theperiod 1975 to 2005 (taken asreference). The dashed line marksthe end of the experiment Historicaland the beginning of the RCPs. Thelines show the median of all values;the shadows cover from the 10thpercentile to 90.

Variable Capture Area Rule in the past Simulated futureFuture

captures

NAO Index All

More catches the

more negative the

winter index (less

the more positive)

Winter NAO Index

more positive

AMO Index All

More catches the

more negative the

index (less the more

positive). Non-lineal

relationship

AMO Index more

positive

X-Component

(longitudinal) of the

Surface Downward

Stress

All

More catches the

larger the component

(positive values)

Lower values of the

component

Oceanic

temperature

between 120-300

meter of deepness

All

Less catches the

higher the

temperature

Higher temperatures

U-Component

(longitudinal) of the

current velocity

All

More catches the

larger the component

(positive values)

Lower values of the

component

Fig.7. (Top) Original captures data set of the

Miño river. (Bottom) Series built from residues of the

regression line of the original data.

Simulation of the eel population evolution under future climatic and oceanic projections

Based on the studies conducted and considering: 1) Relations between populations and oceanic (or index variables), for different Atlantic regions andfor different delays of the year, with statistical significance; 2) The relations between local populations and climatic variables for different annualaggregations of weather and for different delays of the year with statistical significance variables; and 3) Future simulations of the variables ofinterest (statistically significant) and conclusions on possible future developments.

Fig 3. Area 1: "Sargasso Alliance" area defined by

the Sargasso Sea Commission.

Fig 4. Area 2. "Length" aims to

collect the catchment area studied in

Pacariz et al (2013).

Fig. 5. Area 3, “Núcleo”. The core of the region studied by

the Sargasso Sea Commission, seeking

to find possible relations with the central area of the

Sargasso Sea.

Fig.6. Area 4, "Extension".

Extension of the main area of the Sargasso

Alliance without reaching the outer

limits of the original area.

Table 1. Number of significant correlations for each area of study.

Table 2. Number of significant relations for each capture area and oceanic region.

Table 3. Number of significant relations for each variable.

Table 4. Number of significant relations for different delays.

The most decisive oceanic regionof the studied is the one called“Length”. The “Núcleo” region isalso of interest, althoughsignificant relationships found inthis region are mainly those witha delay of three years.

The x components (longitudinal)of vector fields are much moreimportant (have more meaningfulrelationships) than the ycomponents (latitudinal). Thisresult has sense because the xcomponent of the currents is themain responsible for the transferof the larvae of eels from theSargasso Sea to Europe.

Fig 1. Sargasso Sea Area defined by the Sargasso Sea Commission (http://www.sargassoseacommission.org/).

Fig.8. Theta at -125m Fig.9. Theta at -315m Fig.10. Taux Fig.11. AMO Index

Fig.12. U-component at -315m

for “ Longitud “Fig.13. U-component at -315m

for “Núcleo”Fig.14. Winter NAO Index

The conclusions based on thesignificant correlations foundthanks to the past data(reanalysis or observations)for different capture zones torelate these correlations withfuture simulated for each ofthe variables or interestrates. It can be expected afuture decline in the eelpopulation based onsimulations of future climatechange, at least as far asweather conditions areresponsible for the size of theeel population.

These results belong to the“Scientific research on future population of Eels in Spain under the influence of climate change“ project

Data from 3 Earth System Models belonging to the CMIP5 (Coupled Model IntercomparisonProject Phase 5).