Introduction

Wind-driven coastal upwelling results from the action of winds along a coast thatgenerate an Ekman drift directed offshore. This causes pumping of cool and nutrient-richwater towards the sea surface along a narrow region close to the coast, which enhancesprimary production. Thus, coastal upwelling systems are among the most productive marineregions in the world’s oceans.

Due to the ecological and economic importance of these regions, changes in upwellingstrength and timing have attracted considerable scientific interest in recent decades.In several studies, researchers have primarily focussed on analysing trends in windstrength due to climate variability and the resultant changes in the coastal upwellingprocess. In 1990, Bakun1 reported the strengthening of upwellingintensity along the major coastal upwelling systems of the world from 1945 to 1985. Heproposed that this increase was due to global warming, which would create anintensification of the land-sea thermal contrast. This intensification would bereflected in increased land-sea pressure gradients, which in turn would cause thestrengthening of upwelling-favourable winds that would result in cooling of the oceansurface.

Studies focused on different upwelling regions have been conducted to investigate theBakun hypothesis through an analysis of available wind data. However, these studiesreported contradictory results, which indicate that wind estimates from differentdatabases can differ in trends and variability2. In addition,existing time-series data are limited in duration, quality and spatial extent, andresults obtained from different data products in the same area may vary because they arehighly dependent on the length of the time series. Recently, Sydeman et al.3 conducted an analysis of the literature on upwelling-favourablewinds along the major eastern boundary current systems to test the Bakun hypothesis.They synthesized results from more than 20 studies published between 1990 and 2012 basedon time series ranging in duration from 17 to 61 years. Most of published data supportgeneral wind intensification in the California, Benguela and Humboldt upwelling systemsand weakening in the Iberian system. This study highlighted the dependence of theresults on the length of the time series and season and it revealed contradictoryresults between observational data and model-data reanalysis. In addition, numerousavailable time series present a spatial resolution that is too coarse to accuratelyresolve conditions at the scale of coastal upwelling in intense and localized upwellingzones. Thus, higher-resolution temporal scales and greater spatial-resolution studiesare needed.

The aim of this study was to identify the temporal and spatial trends in coastalupwelling regimes worldwide using wind stress data from the National Centers forEnvironmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR)4 database. This database provides high spatial resolution (approximately0.3°) with data available from 1982 to 2010. The length of this databaseallows detailed estimation of upwelling trends over the recent period of strong globalwarming and the same database can be used for all areas of interest.

Methods

Wind data were acquired from the NCEP CFSR database at http://rda.ucar.edu/pub/cfsr.html developed by the National Oceanicand Atmospheric Administration (NOAA). Data were retrieved from the NOAA NationalOperational Model Archive and Distribution System, which is maintained by the NOAANational Climatic Data Center. Detailed information about the CFSR database can befound in Saha et al.4. This worldwide database has aspatial resolution of approximately0.3 × 0.3° and a temporal resolutionof 6 hours from January 1982 to December 2010. The reference height of the wind datais 10 m. Coastal upwelling analysis requires the use of pixels as closeto shore as possible to represent coastal processes. To avoid land contamination,only coastal pixels with less than 25% of land were used.

The process used to identify upwelling areas and to calculate trends is summarizedbelow:

  1. 1

    Wind data were initially averaged at a daily scale to smooth the effect ofhigh frequency events such as breezes. In addition, upwelling eventstypically last 3–14 days5, so a daily scaleseems accurate to describe them.

  2. 2

    Alongshore wind stress was calculatedusing the equation , where is the air density, is the drag coefficient and is the upwelling-favourable wind. can be calculated as+, where is thelatitude, is the zonal windcomponent, is the meridional windcomponent, is the angle defined byan unitary vector normal to the shoreline and pointing seaward andabs means absolute value. Alongshore wind stress has beenpreviously used to estimate variations in upwelling intensity1,6,7. This variable provides information aboutupwelling intensity spreading both in time and in space without gaps, evenin regions close to the Equator were other variables such as Ekman transportdiverge. and were stored at the daily scale for eachcoastal pixel over the period 1982–2010.

  3. 3

    and were averaged at a monthly scale to obtain and because the study was not focused on particular eventsbut rather on identifying regions that show well-developed upwellingconditions lasting for long periods.

  4. 4

    and climatology ( and) was calculated for thewhole period. Different conditions were imposed on these variables toidentify upwelling areas. First, must be higher than 5.4 ms–1, whichcorresponds to the transition from a gentle to a moderate breeze on theBeaufort scale. García-Reyes et al.5defined upwelling events as periods of time with alongshore winds stronger5 ms–1 following Cury andRoy8. Sensibility tests have shown that results areindependent of the particular value of the threshold within the range 5 to5.5 ms–1. This condition must befulfilled for at least three consecutive months. Second, the area isconsidered to be an upwelling region only if at least 10 consecutive points(about three degrees) fulfil the previous condition. This second conditiondiscards the appearance of small local areas.

  5. 5

    Trends were calculated using only the months under strong upwellingconditions. Those months were selected from the climatology data considering values higher than the 50%percentile. As a consequence, the number of months per year to be used inthe analysis differed from zone to zone. This number varied from five toseven months, dependent on the area, with six months being the mostconsistently found value.

  6. 6

    Trends were calculated at each pixel as the slope of the linear regressionof the alongshore monthly wind stress anomalies versus time. Monthlyanomalies were calculated by subtracting from the alongshore wind stress ofa certain month () the meanalongshore wind stress of that month over the period1982–20109,10. All trends werecalculated using raw data without any filter or running mean. The Spearmanrank correlation coefficient was used to analyse the significance of trendsdue to its robustness to deviations from linearity and its resistance to theinfluence of outliers9,10. The significance level ofeach pixel is shown in the figures for those points that exceed 90% (circle)or 95% (square) of significance.

  7. 7

    All figures herein were generated using Matlab.

Results and discussion

The present study investigated trends in coastal upwelling for coastal regions of theworld using the CFSR wind database with high spatial resolution (approximately0.3°). This database facilitates the analysis of wind behaviour at asmall scale, which is a key factor when considering coastal mesoscale effects asupwelling. Recent studies have compared this database and different wind productswith wind measured by several buoys along the Iberian Peninsula coast11,12,13. Statistical results confirmed that datasets withfiner spatial resolution, such as CFSR, gave better results, especially near thecoast.

In this study, 3025 coastal points worldwide were analysed to select the majorupwelling regions and evaluate upwelling trends from 1982 to 2010. Under our termsof selection, ten upwelling systems were evaluated: Benguela, Canary, the southernCaribbean Sea, Chile, Peru, California (north and south), West Australia, Java,North Kenya and Somalia–Oman. These systems were grouped together intoseveral macroscopic zones, namely the eastern and western coasts of the Atlantic,Pacific and Indian Oceans. No upwelling systems along the western Pacific Oceanwere assessed. Alongshore wind stress was calculated over more than 600 points alongthis coast and no region fulfilled the conditions required to be considered acoastal upwelling area.

Eastern Atlantic Ocean

Two upwelling systems were analysed along this coast from south to north:Benguela (South 30.13–23.26°S; North20.76–16.39°S) and Canary(18.89–32.63°N) (Fig. 1). Alongthe Benguela upwelling system, which is one of the major upwelling regions ofthe world, two different areas (south and north) were assessed based on theresults of the calculated alongshore wind stress (Fig. 1).Previous studies were focused mainly on the southern area from the southern tipof Africa to about 20°S. Nevertheless, several studies evaluated theentire area from south of 15°S. The results of the present study forthe southern and northern areas will be described below and then a comparisonwith results from previous studies will be presented.

Figure 1
figure 1

Wind stress along the Eastern Atlantic Ocean.

Annual cycle of alongshore wind stress (Nm–2) forthe period 1982–2010. Red points on the map show the coastalpoints studied. Black lines indicate those regions considered to be coastalupwelling areas. This figure has been performed using Matlab.

Benguela

At the southern coast of the Benguela system, alongshore wind stress had positivevalues (upwelling-favourable conditions) throughout the year, but the highestvalues occurred during the austral spring and summer months (Fig.1, points 16–38). The upwelling season was defined asSeptember to March based on the annual cycle of alongshore wind stressmeridionally averaged over these points. trends were calculated during the upwelling season (Fig.2a). Non-significant trends were identified south of 24°S and only the three northernmost grid boxes (–23.3 to24°S) displayed a significant positive trend. These significantpositive trends continued throughout the Northern Benguela Zone (Fig. 3), although the defined upwelling season differed for thiszone (from July to November). A significant positive trend was observed for theentire region, with values ranging from 410−3 Nm–2 dec–1in the northern area to 10−3 Nm–2 dec–1in the southern one.

Figure 2
figure 2

Upwelling trends in the selected areas along the Eastern AtlanticOcean.

(a) Alongshore wind stress trends along the southern Benguela coastcalculated from September to March. (b) Alongshore wind stress trendsalong the northern Benguela coast calculated from July to November.(c) Alongshore wind stress trends along the Canary coastcalculated from April to September. Those points with significance greaterthan 90% are marked with a circle and those greater than 95% are markedwith a square. A negative (positive) trend means a decrease (increase) inupwelling-favourable winds. This figure has been performed using Matlab.

Figure 3
figure 3

Wind stress along the Western Atlantic Ocean.

Annual cycle of alongshore wind stress (Nm–2) forthe period 1982–2010. Red points on the map show the coastalpoints analysed. Black lines indicate those regions considered to be coastalupwelling areas. This figure has been performed using Matlab.

Previous studies reported similar results. For example, Patti et al.14 analysed data from the Comprehensive Ocean-Atmosphere DataSet (COADS) and found an increase in annual wind stress (1010−3 Nm–2 dec–1)for the area extending from 20 to 30°S, indicating upwellingreinforcement from 1958 to 2007. Narayan et al.15described a significant increase (510−3 Nm–2 dec–1)in the COADS and National Center for Environmental Prediction/National Centerfor Atmospheric Research (NCEP/NCAR Reanalysis) meridional wind stress across26–36°S from 1960 to 2001 using annual data. On theother hand, annual wind stress data from European Center for Medium rangeWeather Forecasting (ECMWF) Re-Analysis (ERA-40 Reanalysis) used in the samestudy15 showed a non-significant trend for the sameperiod. Analysis of upwelling trends along the Benguela Upwelling Ecosystem wasalso conducted using a derived upwelling index in terms of Ekman transport torepresent the estimated potential effects of wind stress on the oceansurface9,16. Pardo et al.16 and Santos et al.9 found a general increasein upwelling intensity over the last four decades for the areas20–32°S and 16–30°S,respectively, using the NCEP/NCAR Reanalysis annual data.

Canary

Along the Canary coast, positive values of alongshore wind stress ( were observed throughout the year, with thehighest values occurring in spring and summer months (Fig.1, points 222–266). The upwelling season was consideredto be April to September. trends overthis period were calculated (Fig. 2c) and positive trendswere detected over almost the entire region. However, significant values werefound only around 22–24°N (810−3 Nm–2 dec–1)and 27–29°N (10−3 Nm–2 dec–1).Published results regarding trends in upwelling along the Canary coast arecontroversial. Trends in upwelling can be highly dependent on the length of thetime series, the selected area and the season evaluated in the analysis.Results similar to those found in the present study were reported by Cropperet al.17 using meridional wind speed from CFSRover the period 1981–2012 during summer (June-August). These authorsfound a non-significant increase in upwelling-favourable winds off NorthwestAfrica (11–35°N). On the other hand, different winddatabases such as NCEP/Department Of Energy (NCEP/DOE II), ERA-Interim, 20Century and National Aeronautics and Space Administration-Modern EraRetrospective-Analysis for Research and Applications (NASA-MERRA) used in thesame study for the same period showed a statistically significant increase northof 21°N and a generally significant decrease in upwelling-favourablewinds south of 19°N. Results of previous studies are also inagreement with our finding of increasing trends in upwelling along the Canarycoast. For example, McGregor et al.18 describedincreasing trends in upwelling around 31°N using annual wind stressdata from COADS for the period 1950–1992. Narayan et al.15 and Patti et al.14 also foundsignificant increasing trends (4-510−3 Nm–2 dec–1)using the same database across 24–32°N consideringannual wind stress over the last four decades. Narayan et al.15 reported a significant increase (210−3 Nm–2 dec–1)in the ERA-40 Reanalysis meridional wind stress for the same period. Incontrast, when they used the NCEP/NCAR reanalysis these authors15 identified a reduction in meridional wind stress (–410−3 Nm–2 dec–1)over the last four decades (1960–2001) in the same region(24–32 °N), indicating a reduction in coastal upwelling.More recently, Barton et al.2 conducted an extensivestudy of wind-induced upwelling trends along the whole Canary current upwellingsystem. Using monthly meridional wind data from the Pacific FisheriesEnvironmental Laboratory (PFEL), NCEP/NCAR, ECMWF, ICOADS and Wave andAnemometer-based Sea Surface Wind (WASWind) plus data from coastalmeteorological stations over 40 years (1967–2007), these authorsfound that trends varied among different data products in the same area, as bothnegative and positive trends were evident. In addition, not statisticallysignificant changes in meridional wind components were found. Contradictoryresults were also obtained using Ekman transport data in numerous upwellingtrend studies conducted along the Canary Upwelling Ecosystem. Gomez-Gesteiraet al.19 detected a significant decreasing trendin upwelling strength for all seasons across 20–32°Nfrom 1967 to 2006 using data from the PFEL. Pardo et al.16 also found a general weakening of the upwelling intensity along theIberian/Canary (26–43°N) and NW African(10–24°N) regions from 1970 to 2009 using the NCEP/NCARReanalysis. These trends were clearly observed in winter and autumn for bothregions and a weakening in the upwelling intensity was also detected in summerin the northwest African region. In contrast, Santos et al.10 confirmed a spring-summer increasing trend across22–33°N when they used the same database (NCEP/NCAR)from 1982 to 2010 in accordance with the present study. Opposite resultsobserved in some of the studies described above using the same database(NCEP/NCAR) emphasize that linear trends are strongly dependent on the length ofthe time series and the season evaluated2,20.

Western Atlantic Ocean

Only one region along the coast of the western Atlantic Ocean fits the conditionsrequired to be considered an coastal upwelling area: the southern Caribbean Sea(62.19–76.56°W) (Fig. 3). In thesouthern Caribbean upwelling region, alongshore wind stress had positive valuesthroughout the year, with the highest values during the boreal winter months(Fig. 3, points 261–307). The upwellingseason was considered to occur from December to April and trends were calculated for this period(Fig. 4). The eastern and western regions exhibiteddifferent behaviours. A negative trend was detected east of 71.25°W,with a significance level higher than 90% (– 4 10−3 to– 810−3 Nm–2 dec–1)for almost all points. In the western region (71.25–76.56°W) a non-significant positive trend was observed in most of thearea, with maximum values of around 410−3 Nm–2 dec–1.This region was previously studied mainly in terms of upwelling occurrence usingwind and sea surface temperature data21,22,23,24,25. Asfar as we know, no studies of upwelling trends in terms of wind have beenconducted along the southern Caribbean upwelling system.

Figure 4
figure 4

Upwelling trends in the selected areas along the Western AtlanticOcean.

Alongshore wind stress trends in the southern Caribbean Sea calculated fromDecember to April. Those points with significance greater than 90% aremarked with a circle and those greater than 95% are marked with a square. Anegative (positive) trend means a decrease (increase) inupwelling-favourable winds. This figure has been performed using Matlab.

Eastern Pacific Ocean

Three upwelling systems were evaluated along this coast from south to north:Chile (37.62–28.88 °S), Peru (16.39–10.15°S) and California (South 33.88–36.06 °N;North 37.94–42. 31 °N) (Fig.5).

Figure 5
figure 5

Wind stress along the Eastern Pacific Ocean.

Annual cycle of alongshore wind stress (Nm–2) forthe period 1982–2010. Red points on the map show the coastalpoints analysed. Black lines indicate those regions considered to be coastalupwelling areas. This figure has been performed using Matlab.

Chile

Along the coast of Chile, positive values of alongshore wind stress were observedthroughout the year, with the highest values during the austral spring andsummer months (Fig. 5, points 78–106). Theupwelling season was considered to last from October to April and trends were calculated for these months(Fig. 6a). Significant negative trends were observedsouth of 34 °S and north of 31 °S, with values between–4 10−3 and –810−3 Nm–2 dec–1.Non-significant positive trends were detected at midlatitudes. Previous studiesalong this coast have reported different results. For example, Garreaud andFalvey26 analysed changes in the coastal winds alongthe west coast of subtropical South America (20–55°S;70–85°W) using future climate scenarios. Differentsimulations were performed using the Providing REgional Climate for ImpactStudies (PRECIS) regional climate model and a significant trend in the coastalwind was absent during the late twentieth century (1961–1990).Similar results were observed by Goubanova et al.27,who used a statistical downscaling method to refine the representations ofcoastal winds for a global-coupled general circulation model (Institut PierreSimon Laplace Climate Model (IPSL-CM4)) from 1970 to 1999. More recently, Rahnand Garreaud28 used CFSR over the period1979–2010 to present a synoptic climatology of the coastal windalong the Chile/Peru coast, paying special attention to prominent upwellingregions. Points located along the Chile coast (30 and 36.4°S) showedunclear trend over the last 30 years. Finally, Aravena et al.29 used anomalies of Ekman transport data from 1980 to 2010to assess the interannual evolution of upwelling along the northern-centralcoast of Chile (29–34°S). These Ekman transportanomalies obtained from PFEL showed a positive trend throughout the area. Aspreviously mentioned, trends in upwelling can be highly dependent on factorssuch as the season evaluated in the analysis. This could explain the observedweakening in upwelling in most of the region in the present study, which wecalculated using alongshore wind stress from October to April. When werecalculated trends using annual data,the observed trend was unclear, which is in accordance with most of the studiespreviously conducted along this upwelling system. The dependence of trends onthe season studied was also reported by Sydeman et al.3, who analysed more than 20 studies related to upwelling trends alongthe five major upwelling systems published over the last two decades. They foundthat some of the disagreement in previous studies could be resolved byconsidering winds during only the active upwelling seasons.

Figure 6
figure 6

Upwelling trends in the selected areas along the Eastern PacificOcean.

(a) Alongshore wind stress trends along the Chile coast calculatedfrom October to April. (b) Alongshore wind stress trends along thePeru coast calculated from May to October. (c) Alongshore wind stresstrends along the southern California coast calculated from April toSeptember. (d) Alongshore wind stress trends along the northernCalifornia coast calculated from April to October. Those points withsignificance greater than 90% are marked with a circle and those greaterthan 95% are marked with a square. A negative (positive) trend means adecrease (increase) in upwelling-favourable winds. This figure has beenperformed using Matlab.

Peru

Along the Peru coast, alongshore winds had positive values throughout the year,with maxima from May to October (Fig. 5, points146–166). The upwelling season was considered to occur during thesemonths and trends were calculated forthis time period (Fig. 6b). Small and non-significantpositive trends were observed for almost the entire region except at thesouthernmost point (16.4°S), where a significance level higher than95% was identified (10−3 Nm–2 dec–1).A non-significant negative trend with small values was observed at 14°S. Previous studies have shownthat trends in upwelling along the Peru coast are contradictory. Results similarto those obtained in our study were reported by Rahn and Garreaud28, who analysed annual alongshore winds from CFSR from 1979 to 2010.These authors found a positive trend in alongshore wind from 1995 to 2010( ms–1) at15°S, which is a prominent upwelling region along the Peru coast.Gutierrez et al.30 used ERA-40 Reanalysis in aregion around 14°S and also found an increase inupwelling-favourable winds from 1958 to 2001 for the spring season and usingannual data. Bakun et al.31 reported an increase inupwelling at ~10.5°S using monthly wind stress data fromCOADS from 1948 to 2006. In contrast, Goubanova et al.27 suggested the existence of a weakening of alongshore wind near15°S from 1970 to 1999 based on statistical downscaling of the seasurface wind. Previous studies conducted over a wider region have also showndifferent results. Patti et al.14 described anincrease in annual wind stress using data from COADS for the area extending from6 to 16°S and demonstrated the existence of upwelling reinforcementfrom 1958 to 2007. However, the trend they reported (10−3 Nm–2 dec–1)was much higher than that found in the present study. Narayan et al.15 also analysed the same region from 1960 to 2001 usingdifferent datasets. Annual wind stress data from COADS (510−3 Nm–2 dec–1)and ERA-40 Reanalysis (0.910−3 Nm–2 dec–1)revealed a significant increasing trend in upwelling, in agreement with theresults reported by Patti et al.14, although in thelatter case the trend value was much smaller. The general trend14,15,31 is a similar to that observed in the presentstudy. In contrast, Narayan et al.15 identified astatistically non-significant decrease (–0.710−3 Nm–2 dec–1)at the Peruvian upwelling region (6–16°S) usingmeridional wind stress from the NCEP/NCAR reanalysis. Pardo et al.16 analysed annual Ekman transport using data from theNCEP/NCAR Reanalysis and found a general weakening of the upwelling intensity inthe Peru region (6.7–16.2°S) from 1970 to 2009.

California

For the California upwelling region (Fig. 5), two differentareas were assessed: South (33.88–36.06 °N) and North(37.94–42.31 °N). Almost all published reports abouttrends in upwelling along this system include the entire coast of California(32–42 °N). The results of the present study for thesouthern and northern areas will be described below and then a comparison withresults from previous studies will be presented.

Along the southern California coast, the highest values of alongshore wind stresswere observed during the spring and summer months (Fig. 5,points 350–359). The upwelling season was considered to occur fromApril to September based on the annual cycle of wind stress meridionallyaveraged over these points and trendswere calculated (Fig. 6c). Negative trends were observedfor the entire region, but they were significant (810−3 Nm–2 dec–1)only at the three southernmost points and at the northernmost one. Along thenorthern California coast, alongshore wind stress showed a similar behaviour,although the highest values were mainly observed in summer (Fig.5, points 365–379). Thus, trends were calculated between April and October (Fig. 6d). Non-significant negative trends were detected forthe three southernmost points, with maximum values around –310−3 Nm–2 dec–1.Positive trends were observed north of 38.5 °N with a significancelevel higher than 90% for the northernmost region (4-610−3 Nm–2 dec–1).

Controversial results in relation to the long-term variability in coastalupwelling were also found in the California upwelling system. In terms of windspeed, Mendelssohn and Schwing32 reported trends ofstronger upwelling-favourable winds along 32–40°N basedon April–September COADS data from 1946 to 1990. Patti etal.14 analysed a similar area between 34 and40°N using annual wind stress data from the same database over theperiod 1958 to 2007. They described an increasing trend of around 410−3 Nm–2 dec–1.Narayan et al.15 also found a statisticallysignificant increase in upwelling-favourable winds (310−3 Nm–2 dec–1)using annual wind stress data from the same dataset from 1960 to 2001 for thesame region. Although the southern area evaluated in the present study(33.88–36.06°N) is included in the region studied byMendelssohn and Schwing32, Patti et al.14 and Narayan et al.15, thegeneral trend observed in these three works and the present one iscontradictory. In contrast, Narayan et al.15identified a significant decreasing trend in upwelling from the ERA-40Reanalysis of annual wind stress data (–0.610−3 Nm–2 dec–1)from 34 to 40°N over the last four decades (1960–2001).This decrease is in good agreement with the results of the present study,although the trend value was much higher in our case (810−3 Nm–2 dec–1).Different results were also reported in several studies that covered a widerregion. For example, Garcia-Reyes and Largier33 studied theCalifornia region from 33 to 42°N from 1982 to 2008 using wind speeddata during the upwelling season (March–July) from the National DataBuoy Center (NDBC) buoys. They found significant increasing trends in upwellingwinds north of 35°N and a decreasing trend in the southern region(33–35°N). Similar results were observed when data fromthe NDBC for June-August over the period 1980 to 201034were used. The decreasing trend along the southern coast of California(33–35°N) reported in these two works is in agreementwith the results of our study. On the other hand, positive trends were onlyobserved north of 38.5°N in our case.

Ekman transport data also have been evaluated in different studies conductedalong the California Upwelling Ecosystem. Rykaczewski and Checkley35 found a positive summer trend around 34.5°Nfor the period 1948–2004 using Ekman transport data from theCalifornia Reanalysis Downscaling (CaRD10), which is a dynamically downscaledanalysis of the NCEP/NCAR Reanalysis. Seo et al.34found similar results using the same database over the entire California coast(32–42°N) from 1980 to 2010. Garcia-Reyes andLargier33 detected significant increasing trends inupwelling strength during March-July north of 34.5°N from 1982 to2008 using data from the PFEL. Pardo et al.16 alsostudied the California region from 33 to 45°N using NCEP/NCARReanalysis Ekman transport data from 1970 to 2009 and they reported an unclearannual trend. In contrast, Iles et al.36 found anincreasing trend in Ekman transport annual data from PFEL from 1967 to 2010 overthe same region.

Considering that ENSO (www.esrl.noaa.gov) can be an important source of variabilityalong the Pacific upwelling systems, its influence on the estimated trends of was analysed. No correlationswere found between the variability of ENSO and .

Eastern Indian Ocean

Two upwelling systems were assessed along this coast: West Australia(31.69–21.39 °S) and Java (105.62–116.87°E) (Fig. 7).

Figure 7
figure 7

Wind stress along the Eastern Indian Ocean.

Annual cycle of alongshore wind stress (Nm–2) forthe period 1982–2010. Red points on the map show the coastalpoints analysed. Black lines indicate those regions considered to be coastalupwelling areas. This figure has been performed using Matlab.

West Australia

Along the western Australian coast, maximum values of alongshore wind stress wereobserved from October to March, which corresponds to the austral spring andsummer months (Fig. 7, points 108–142). Theupwelling season was considered to occur from October to March and trends were calculated over these months(Fig. 8a). Positive trends were detected for almostthe whole region, with significant values between 4-810−3 Nm–2 dec–1south of 25.5°S. Non-significant negative trends were found at thenorthernmost latitudes (21.4–22.9°S). As far as we know,no studies regarding upwelling trends in terms of wind have been conducted alongthe western coast of Australia. Previous studies have shown the absence ofpersistent upwelling off this coast despite a prevailing summer wind systemfavouring upwelling. This absence of upwelling has been attributed to thepresence of the Leewin Current (LC), a warm poleward flow transportingnutrient-poor waters from the tropics37,38. Unlike othereastern boundary currents (e.g., the Benguela and Humboldt Currents at similarlatitudes), the poleward flow of the LC suppresses the persistent upwelling ofcool, nutrient-rich, subsurface water onto the western Australia continentalshelf39,40. Thus, large-scale upwelling isincompatible with the poleward flowing LC. Nevertheless, localized seasonalupwelling associated with inner shelf wind-driven currents can appear along someregions, such as the Ningaloo (23–25°S) and CapesCurrents (26–28°S), due to variations in the LC41,42,43,44,45,46,47.

Figure 8
figure 8

Upwelling trends in the selected areas along the Eastern IndianOcean.

(a) Alongshore wind stress trends along the Western Australia coastcalculated from October to March. (b) Alongshore wind stress trendsalong the Java coast calculated from May to October. Those points withsignificance greater than 90% are marked with a circle and those greaterthan 95% are marked with a square. A negative (positive) trend means adecrease (increase) in upwelling-favourable winds.

Java

Along the Java coast, values of alongshore wind stress were higher during theaustral winter (Fig. 8, points 227–263). Theupwelling season was considered to last from May to October and trends were calculated for this time period(Fig. 8b). Significant negative trends were detectedfor almost the entire coast, with values between –410−3 Nm–2 dec–1at the easternmost region and –1010−3 Nm–2 dec–1at the westernmost one. The existence of upwelling along the Java coast and itsbasic features have been documented in previous studies, mainly in terms ofupwelling occurrence determined using wind and SST data, whereas upwellingtrends have not been considered. Different researchers have found that upwellingoccurs between June and November and is mostly forced both locally by thealongshore winds associated with the southeast monsoon and remotely byatmosphere-ocean circulation associated with ENSO48,49,50,51,52,53.

Western Indian Ocean

Two upwelling systems were evaluated along this coast from south to north: NorthKenya (2.03–1.40°N) and Somalia-Oman(1.72–22.01°N) (Fig. 9).

Figure 9
figure 9

Wind stress along the Western Indian Ocean.

Annual cycle of alongshore wind stress (Nm–2) forthe period 1982–2010. Red points on the map show the coastalpoints analysed. Black lines indicate those regions considered to be coastalupwelling areas. This figure has been performed using Matlab.

North Kenya

Along the northern Kenya coast, positive values of alongshore wind stress wereobserved from November to April (Fig. 9, points125–136). The upwelling season was considered to occur from Novemberto April and trends were calculatedover this period (Fig. 10a). Significant negative trendswere detected for the entire region, with values of around –210−3 Nm–2 dec–1.The coast of northern Kenya is characterized by the occurrence of an irregularupwelling linked to the northeast monsoon, which normally develops from Novemberto March. Several researchers have related the occurrence of this upwelling tohigher productivity in the area, although the fact that upwelling events are notregular has attracted relatively little scientific interest about upwellingtrends in terms of wind. This region has been studied mainly in terms of changesin chemical and biological oceanographic parameters related to the occurrence ofthe northeast and southeast monsoons, which lead to important differencesbetween the Kenya and Somalia coasts in terms of upwelling54,55,56,57,58,59,60.

Figure 10
figure 10

Upwelling trends in the selected areas along the Western IndianOcean.

(a) Alongshore wind stress trends along the Northern Kenya coastcalculated from November to April. (b) Alongshore wind stress trendsalong the Somalia-Oman coast calculated from April to October. Those pointswith significance greater than 90% are marked with a circle and thosegreater than 95% are marked with a square. A negative (positive) trend meansa decrease (increase) in upwelling-favourable winds. This figure has beenperformed using Matlab.

Somalia-Oman

Along the Somalia-Oman coast, positive values of alongshore wind stress werefound from April to October (Fig. 9, points136–193). Using the annual cycle of the alongshore wind stress, theupwelling season was considered to last from April to October and trends were calculated for these months(Fig. 10b). Negative trends were observed for almostthe whole region except at the southernmost coast of Somalia(2–6°N) and the northernmost coast of Oman(18–22°N). Significant trends were observed mainly alongthe Somalia coast (6.5–10.5°N), with values between–1.510−3 Nm–2 dec–1and –3.510−3 Nm–2 dec–1.Upwelling is induced by an alongshore current driven by the southwest monsoon insummer. With the onset of the northeast monsoon the circulation patternreverses, causing a cessation of the upwelling. This region has been widelystudied in terms of SST and biodiversity related to the occurrence of theseupwelling events61,62,63,64,65. Nevertheless, fewstudies have focused on upwelling trends in terms of wind. Goes etal.66 reported an interannual escalation in theintensity of summer monsoonal winds accompanied by enhanced upwelling along thecoast of Somalia (47–55°E,5–10°N) using wind data from NCEP-NCAR reanalysis from1997 to 2004. In contrast, using the same database and an advanced coupledatmosphere-ocean general circulation model, Izumo et al.67 detected a decrease in upwelling from 1979 to 2006 caused byanomalously weak southwesterly winds in late spring over the Arabian Sea. Morerecently, Piontkovski et al.68 described a decliningtrend in the zonal component of wind speed over the Sea of Oman(22–25°N) during summer monsoons from the late 1950s to2010.

Results obtained for different upwelling systems around the world illustrate thatit is not possible to describe a homogenous behaviour among them because trendschange substantially, even in regions with similar oceanographic processes. Ofthe five major upwelling regions worldwide, increasing trends in upwelling wereobserved in the coastal areas of Benguela, Peru, Canary and northernCalifornia and the increases were statistically significant only for the twolast systems. A general decrease in upwelling intensity was observed along theChile, southern and central California and central Somalia coasts, withsignificant values in all regions. Thus, no evidence for a generalintensification of upwelling along these systems was observed.

The general trends found in our study were similar to those reported in studiesin which wind stress data were used (Tables 1 and 2). It is important to note that controversial results werealso obtained by different authors using the same variable and even the sameperiod of time2,15, which indicates a dependence ofresults on the database used.

Table 1 Studies in which upwelling trends were analysed in terms of wind along theBenguela and Canary coasts.
Table 2 Studies in which upwelling trends were analysed in terms of wind along theChile, Peru and California coasts.

Different trends were also detected along the less studied upwelling regions. Forexample, significant decreasing trends were observed along the coast of Java andnorthern Kenya, whereas a tendency to an increase in upwelling-favourable windswas detected in western Australia. In the southern Caribbean upwelling region, asignificant negative (positive) trend was found east (west) of71.25°W.

Our analysis covered the last three decades (1982–2010), which is therecent period of strongest global warming, thus it allowed detailed analysis ofthe influence of warming processes on upwelling trends. In another study, Limaand Wethey69 estimated changes in coastal SST by exploringmonthly warming patterns along the world’s coastline at a scale of0.25° for the period 1982–2010. They found that eventhough most coastal areas worldwide have been warming, the magnitude of changehas been highly heterogeneous in both space and season. They described a coastalSST decrease nearly year-round in the areas influenced by the California andHumboldt currents, which could be related to a tendency for intensification ofupwelling following Bakun’s hypothesis. Nevertheless, our resultsrevealed an increasing trend in upwelling along the northern California(38–42°N) and Peru coasts and negative trends along theChile and southern-central California coasts (Table 2).On the other hand, Lima and Wethey69 found a general SSTincrease in the areas of the Canary, Benguela and Somali currents, which couldindicate a decrease of upwelled, cooler waters linked to an upwelling reduction.Our results only revealed a decrease in upwelling-favourable winds along theSomalia coast.

Among the upwelling systems analysed in the present study, Lima and Wethey69 found that coastal temperatures have been warming almosthomogeneously throughout the year along the southern Caribbean upwelling regionand along the Java and northern Kenya coasts. The western Australia coast hasbecome colder from January to September. These results can be linked to theupwelling trends found in the present work. We detected decreasing trends inupwelling along the southern Caribbean, Java and northern Kenya coasts, whereasa general upward trend was found along the western Australia coast.

These results delve into a possible discussion about whether the Bakun hypothesisis being fulfilled or not taking into account the present study and thosementioned in Tables 1 and 2.Even in regions where previously not many studies regarding upwelling trendsexisted, like the Java coast or the southern Caribbean upwelling region,different behaviours can be observed contradicting the general upward trendpredicted by Bakun due to global warning.

Additional Information

How to cite this article: Varela, R. et al. Has upwelling strengthenedalong worldwide coasts over 1982-2010? Sci. Rep.5, 10016; doi: 10.1038/srep10016 (2015).