Final Report Summary - PREFACE (Enhancing prediction of tropical Atlantic climate and its impacts)
Executive Summary:
PREFACE has advanced significantly towards its overarching goal to enhance prediction of tropical Atlantic climate and its impacts.
Key oceanic processes in the tropical Atlantic were better understood, by enhancing and analysing equatorial and coastal observations, and by numerical experimentation. We quantified the seasonal surface mixed-layer heat and salt budgets in the Benguela and Gulf of Guinea regions, and the impact of wind-excited near-inertial waves and chlorophyll-induced turbidity on the heat budget. Our measurements identified pronounced semi-annual, annual and interannual variability of the equatorial zonal currents, and quantified for the first time the seasonal cycle of the southward Angola Current. Our model studies indicate that the warm SST bias could be reduced close to the coast by using high ocean model resolution and improved atmospheric forcing. We identified the dynamics of deep ocean jets at the equator and causes of Benguela events, which strongly impact fisheries. Close cooperation between different European and African partner institutions were established or improved resulting in common experiments at sea, data sharing and analysis, publishing of historical observations, as well as application of ocean modelling on regional scale.
PREFACE has identified key causes of systematic error in the tropical Atlantic in climate models. In general, atmospheric model errors are the dominant cause of the large equatorial and southeastern Atlantic SST biases, but the relative roles of wind, cloud-radiative and ocean dynamics differ regionally, seasonally, and among models highlighting the importance of a multi-model approach. Errors in the Southern Ocean radiation budget can also drive tropical biases. Model biases cause an underestimation of dynamical ocean-atmosphere interaction and affect the relation between Atlantic and Pacific climate variability. We have developed a new parameterisation for ocean mixing, high-resolution models, and novel correction techniques to improve the simulation of tropical Atlantic climate and its global impacts.
PREFACE has improved understanding of tropical Atlantic variability and predictability. Tropical Atlantic SST may be predicted based on equatorial waves, deep equatorial jets, and remote forcing from the Pacific and South Atlantic, while Atlantic multidecadal variability modulates their importance. Reducing climate models biases improves skill in predicting Atlantic Niño events, and their impact on continental climate and the Pacific. We showed that Atlantic and Pacific decadal variability and global warming impact African and South American rainfall. Decadal variations in Sahel rainfall can now be predicted based on Atlantic SST. Model biases introduce uncertainty in future evolution of the tropical Atlantic and the West African Monsoon. Lastly, advanced statistical methods were developed to improve SST and rainfall predictions, and to account for model bias.
PREFACE is increasing understanding of the interrelation among marine ecosystems, climate variability and change, and fisheries; and of the socio-economic implications of future climate change. We analysed more than two decades of data from shipboard surveys, carried out with regional partners and by the FAO EAF Nansen project. Climate impacts the distribution of key fish stocks along the African-Atlantic coast and tuna in the tropical Atlantic, with some pelagic species showing clear migration trends and others appearing resilient to current climate change. Fish prey is also affected: over the last 20 years, micronekton distributions have become deeper along this coast. Thus, climate prediction can be useful for fisheries management. We performed an impact assessment for West African fisheries using climate change projections with reduced biases and adapted ecological-economic models. We performed questionnaires to assess the economic and social situation of local fishing communities, and their vulnerability to climate driven changes in the ecosystem.
Project Context and Objectives:
The tropical Atlantic climate recently experienced shifts of great socio-economic importance. The oceanic changes were largest in the eastern boundary upwelling systems, globally important regions for marine productivity and climate. African countries bordering the Atlantic depend upon their ocean for societal development, fisheries, and tourism. They were strongly affected by these changes and face important adaptation challenges associated with global warming. Unfortunately, the tropical Atlantic is a region of key uncertainty in the climate system: state-of-the-art climate models exhibit large systematic errors, prediction of climate variability and climate change are highly uncertain, and how shifts in climate impact marine ecosystems is largely unknown, as are their possible global socio-economic impacts.
To redress this situation, PREFACE joins European and African expertise to provide the first comprehensive assessment of the tropical Atlantic. In four research foci, sophisticated observation systems, extensive field experiments, regional and global scale modelling capabilities, and socio-economic and marine ecosystem models are combined to improve the prediction of tropical Atlantic climate and its impacts.
The first research focus – Role of ocean processes in climate variability – aims to better understand the oceanic processes affecting sea surface temperature (SST) in eastern tropical Atlantic upwelling regions on seasonal to decadal time scales, by using historical and new data as well as medium to very-high resolution forced ocean models. We focus on the mixed layer heat and freshwater budgets, the tropical Atlantic circulation, and the equatorial and coastal wave response. By evaluating the ocean models, we will suggest improvements to reduce biases (systematic errors) in ocean and coupled climate models.
Evaluation of current climate models and reduction of model bias is the second focus, aiming to better understand and subsequently reduce systematic errors in climate models in the tropical Atlantic. Our strategy is to analyse the development of model systematic error in climate predictions and then perform coordinated, multi-model sensitivity experiments to identify the key processes that require improved model representation to mitigate these systematic errors. Additionally, we aim to understand the relation among errors in variability, feedbacks, and atmospheric-links to other regions of the globe.
Our third focus – Climate prediction in the Tropical Atlantic – aims to clarify the mechanisms for climate variability and predictability on seasonal to decadal time scales, by analysing observations and numerical experimentation. We also investigate interactions between the Atlantic and other regions, and both natural and anthropogenic factors. The goal is to improve the prediction of tropical Atlantic climate, and to assess the impact of model systematic error on prediction skill. We are developing a unified statistical prediction framework, and will also use state-of-the-art climate prediction models.
PREFACE’s last research focus – Impacts of climate change on pelagic functional diversity in the tropical Atlantic with effects on western African fisheries economies – aims to disentangle environmental and anthropogenic pressures on pelagic fish stocks in the tropical Atlantic, through exploratory analysis of historical and newly collected data. We also aim to better understand climate change effects on small-scale fisheries and coastal communities, and to assess uncertainties and implications for management. Our approach is to develop socio-economic models, built on knowledge of the marine ecosystem and fishing communities, and to drive them with climate change projections.
PREFACE envisions a closer cooperation between European and West African partner institutions leading to enhancement of research capacities in the fields of ocean physics, ecosystem and fisheries as well as climate research in West Africa.
Project Results:
Towards prediction of tropical Atlantic climate and its impacts.
PREFACE was a bold and ambitious project that targeted one of the marine regions that will be more affected by climate change and its consequences. It provided the first comprehensive, multidisciplinary assessment of the tropical Atlantic climate, from observations to predicting its socio-economic impacts.
Enhanced observations and medium-to-high resolution ocean model experiments
PREFACE has substantially improved the understanding of key oceanic processes affecting sea surface temperature (SST) in the tropical Atlantic on seasonal to decadal time scales, by enhancing and analysing equatorial and coastal observations, and by numerical experimentation.
In particular, PREFACE has drastically improved the ocean and climate observing system in the eastern tropical Atlantic including near-coastal regions, allowing a continuous monitoring of tropical Atlantic variability and climate change. Two new ocean-atmosphere time series stations were established within PREFACE to complement continuous measurements along the equatorial and coastal regions; these were crucial for understanding tropical air-sea interactions and for detecting climate signals propagating from the equatorial Atlantic towards the biologically productive eastern boundary upwelling regions. The first one enhances the deep ocean PIRATA (Prediction and Research Moored Array in the Tropical Atlantic) network in the southeastern tropical Atlantic, and the second one offers similar measurements within the Senegal-Mauritania sector of the Canary coastal upwelling. Due to close cooperation and intense capacity strengthening activities within PREFACE, historical datasets from near-coastal regions off Africa, particularly including hydrographic data from the Nansen and EAF-Nansen programmes, could be made available for scientific analysis thereby drastically enhancing the climate record of the tropical Atlantic. Numerous dedicated shipboard and glider observations carried out within PREFACE complemented the available database.
Medium to very-high resolution forced ocean model simulations complemented observations. Their spatial domain covered the tropical and subtropical Atlantic with high resolution, and very high resolution in the Benguela, Senegal-Mauritania, and northern Gulf of Guinea upwelling regions. These simulations were validated by comparison to available large-scale observations and PREFACE data from moorings and cruises. Sensitivity studies were performed with the models to assess the importance of various processes in the SST and sea surface salinity climatology and variability, and to determine strategies to reduce ocean model biases.
Improved mixed layer heat and freshwater budgets
Observations and models delivered an improved understanding of the seasonal and interannual mixed-layer temperature and salt balances in the equatorial and eastern boundary upwelling regions of the tropical Atlantic. An important project output was an improved mixed-layer temperature and salinity climatology dataset for the tropical Atlantic, PREFCLIM. Unpublished data from African partners were incorporated to close data gaps in coastal regions. On seasonal as well as interannual time scales, vertical mixing within the thermocline is a dominant contributor to the development of the Atlantic Cold Tongue. On seasonal time scales, vertical mixing and horizontal advection dominate the salinity variability whereas on interannual time scales it is dominated by surface freshwater fluxes and winds . For the southeastern tropical Atlantic, we provided a revised understanding of the critical role of advective heat transport by eddies for the mixed layer heat and fresh water balances. In addition, alongshore advection appears to lead to a clear correlation between SST and temperature anomalies in the upper thermocline on interannual to decadal timescales in the Angola-Benguela region . Beyond this we have provided new estimates of the impact of wind-induced near inertial wave energy and chlorophyll concentration on the mixed layer heat content variability.
Ocean circulation, and the equatorial and coastal waves
PREFACE delivered key new findings in three areas. Firstly, our measurements identified pronounced semi-annual, annual and interannual variability of the equatorial zonal currents. We showed that resonant equatorial basin modes can explain much of the enhanced variability on these timescales, and that pronounced deep intraseasonal variability of equatorial zonal currents supplies energy to maintain resonant equatorial basin modes on interannual time scales. Furthermore, we identified the dynamics of deep ocean jets at the equator. Secondly, we quantified for the first time the seasonal cycle of the southward Angola Current, using mooring data from 11°S off Angola together with shipboard hydrographic data (PREFACE and EAF Nansen programme surveys). This allowed a first independent evaluation of different ocean reanalysis products. Lastly, we provided new insights into the causes of Benguela events, which strongly impact fisheries. While most of the anomalous interannual SST events along the southwestern African coast are remotely forced by equatorial waves as could be particularly evidenced for the Benguela Niño in 2011, the warm event in January/February 2016 could not be related to equatorial forcing. Instead local forcing played a dominant role.
Long-term directions to improve ocean and climate modelling
PREFACE assessed the importance of various processes for the realistic simulation of SST and ocean circulation in eastern boundary upwelling regions of the tropical Atlantic. We considered three frequently misrepresented processes: near inertial wave induced mixing, upwelling and poleward heat transports in relation to wind stress patterns, and absorption of solar radiation by chlorophyll. In addition, we investigated the benefit of improving spatial resolution of ocean models and the realism of air-sea forcings. Based on comprehensive experimentation with ocean models we make the following general recommendations for improving ocean model simulations in eastern boundary upwelling regions of the tropical Atlantic: (1) to force the ocean models with a wind stress product having a climatological seasonal mean distribution close to that of scaterometer derived winds; (2) to validate the shortwave radiative forcing against in-situ observations; (3) to use a horizontal resolution in first 500km offshore as close as possible to the Rossby Radius, although 1/10° appear to provide satisfying results; (4) to include the effect of primary production on turbidity of the waters, via a realistic chlorophyll distribution in the equation of penetrating solar radiation; and (5) to consider the importance of transmitting high frequency wind energy into the ocean.
Disentangling the causes of tropical Atlantic climate model biases
PREFACE has deepened the understanding of the ubiquitous model errors (biases) in the tropical Atlantic climatology. The climate model biases are characterised by to warm SST in the eastern-equatorial Atlantic and the southeastern tropical Atlantic. The SST errors are associated with corresponding large precipitation and surface wind errors. We performed coordinate analysis and multi-model experiments to disentangle the causes of tropical Atlantic model biases. A particular novel aspect was to perform idealised prediction experiments. The analysis of coupled model mean biases lead to the following conclusions:
(1) The equatorial warm SST bias, which is a seasonal bias strongly tied to the annual cycle. The spring cooling is largely underestimated in coupled models, leading to strong summer SST biases. There are some indications that errors in the mean state (like a too deep mixed layer in the eastern equatorial Atlantic in early spring) do not favour the cold tongue cooling processes. Improving the surface wind representation in atmospheric models helps in improving the mean state and favours the cold tongue development. The atmospheric wind biases are present in atmospheric models forced by observed SST. This indicates that deficiencies in atmospheric models are the root cause of the warm equatorial Atlantic SST bias, but coupled processes amplify these errors. The surface wind biases are linked to surface pressure gradients errors and in turn to deficiencies in representing the convection. The wind stress biases are thus related to ITCZ errors.
(2) The southeastern warm SST bias, which develops more progressively and independently of the seasonal cycle. In this stratocumulus region, the radiative biases play an important role in explaining large scale SST biases, but other elements also play a role: coastal atmospheric wind jets, vertical ocean mixing, and coastal currents representation. All these processes tend to be better represented when increasing resolution both in atmosphere and in ocean models even if the improvement is model dependent.
(3) Remote errors also contribute to drive tropical Atlantic model biases. In particular, correcting the radiatively caused warm SST biases over the Southern Ocean can modify the stratocumulus-SST feedback and reduce the SST warm bias over the tropical stratocumulus regions. Errors in the simulation of the South Atlantic Anticyclone also drive SST biases in the Angola- Benguela region. In addition, the strength of the simulated Atlantic meridional overturning circulation seems to explain the SST bias pattern in some models.
Improved climate models for the tropical Atlantic
PREFACE proposed three approaches to reduce climate model errors in the equatorial Atlantic. Firstly, we developed a model with higher atmospheric resolution to realistically simulate the tropical Atlantic climatology. Accurate simulation of the surface winds required increased vertical atmospheric model resolution in the lower troposphere. Secondly, we developed new parameterisation for ocean mixing, and one of these accounted for near inertial wave mixing. Lastly, we developed a novel anomaly coupled approach that statistically corrects ocean-atmospheric fluxes.
Reducing mean state errors improves simulated variability
Comprehensive multimodel analysis, and experiments with the PREFACE improved climate models and ocean models have shown that reducing mean state errors improves the simulation of tropical Atlantic climate variability. State-of-the-art climate models show key deficiencies in simulating the two dominant patterns of tropical Atlantic variability – the Atlantic Niño and the Atlantic Meridional Mode – although they capture their strength and spatial structure reasonably well. Regarding the Atlantic Niño, biased models tend to simulate variability that peaks in autumn to winter rather than in boreal summer, as observed. Critically biased models do not reproduce the link between surface and subsurface ocean temperatures, and thus they underestimate this element of the Bjerknes positive feedback. This is because in biased models upwelling is too weak and the thermocline too deep in the eastern equatorial Atlantic. Reducing model biases leads to improvement in the Bjerknes positive feedback and thus to the simulation of Atlantic Niño variability. Regarding the Atlantic Merdional Mode, the Wind Evapourative SST feedback that underlies this variability is too strong in bias coupled models compared to observational analysis. This suggest that in coupled models, a greater portion of SST tendency is controlled by atmospheric heat fluxes than that in the observations; this may suggest that ocean processes are underestimated in biased models.
Relation between Atlantic and Pacific climate variability
We have shown the importance of the tropical Atlantic in driving Pacific El Niño variability, and better understood the mechanisms. We have also assessed the ability of climate models to simulate the relation between tropical Atlantic and Pacific, and shown its potential to enhance seasonal prediction of the El Niño. The following is a brief summary of our key findings achieved through multi-model analysis and by performing sensitivity experiments.
Observations show that Atlantic Niño’s (Niña’s) often precede Pacific La Niña’s (El Niño’s) by about six months. It is now well established that this is because Atlantic Niño variations affect the Pacific trade winds and thus the development of El Niño. This occurs when the Atlantic Inter-tropical Convergence Zone (ITCZ) is anomalously close to the equator, causing a shallower mean thermocline and enhanced eastern equatorial Atlantic SST variability. The larger amplitude and greater westward extent of Atlantic Niño SST variations leads to a larger impact on trade winds in the Pacific, and thus an impact on El Niño. Atlantic ocean multi-decadal variability and global warming can both shift the position of the Atlantic ITCZ. Together these two explain the strengthening of the Atlantic’s influence on the Pacific since the 1970’s that have contributed to stronger El Niño variability.
Climate model biases affect the simulation of the relation between Atlantic and Pacific climate variability. In current climate models the ITCZ is too far south in the Atlantic, and they underestimate dynamical ocean-atmosphere interaction in this region (see above). As a result models underestimate the impact of the Atlantic Niño on the Pacific. Lastly, we have shown that when the connection between Atlantic and Pacific climate variability is properly represented, it can enhance seasonal prediction of El Niño.
Tropical Atlantic climate variability
PREFACE has improved understanding of tropical Atlantic variability and predictability, through analysis of observations and model simulations. The Atlantic Niño was the main focus. We showed that variance in boreal summer SST in the eastern equatorial Atlantic is controlled by ocean dynamics, but thermodynamic processes contribute. Equatorial waves triggered by atmospheric forcing are the dominant process controlling SST warming (cooling) in the central and eastern equatorial during Atlantic Niño (Niña) events. The decay of these events in July is associated with anomalous horizontal advection. The variability of the South Atlantic Anticyclone through its impacts on equatorial winds can explain up to 50% of the equatorial Atlantic SST variance. Another particularly interesting finding is that equatorial SST variability and surface zonal velocity within the North Equatorial Counter Current is influenced by Equatorial Deep Jets (stacked zonal jets along the equator which steadily propagate downwards over time with a time scale of about 4.5 years). Equatorial Deep Jets are driven by deep intraseasonal variability and are associated with upward energy flux. Unfortunately, most current state-of-the-art climate models do not reproduce properly dynamical ocean-atmosphere variability (see above) and thus likely underestimate predictability.
We also better understood long-term climatic change in the tropical Atlantic. We showed for the first time that equatorial Atlantic SST shows decadal variability that is related to the southern subtropics and to the strength of the St. Helena subtropical anti-cyclone. On longer timescales, variability generated internal to the climate system has an important influence on the observed multi-decadal trends in the tropical Atlantic. The amplitude of the observed trend is much bigger than the multi-model estimated trend caused by anthropogenic global warming. Thus, the recent trends in SST over the tropical Atlantic that show stronger warming over Mauritania and Namibia coasts than over South Benguela upwelling region are probably not caused by global warming.
Rainfall variability over adjacent continents
PREFACE has also devoted efforts to better understand the seasonal to interannual variability of rainfall over the continental areas adjacent to the Atlantic. The structure of large- scale circulation over central Africa and rainfall variability over Southern Africa, Ethiopia, West Africa and Brazil have been further analysed and related to SST anomalies over different basins with a special focus on the Pacific El Niño and the stationarity of the links found.
The Pacific El Niño is the main source of variability at interannual timescales. It has been shown to affect rainfall variability in the continental areas adjacent to the Atlantic. Its positive phase causes negative rainfall anomalies over Southern Africa, Ethiopia, Northeast Brazil and West Africa. However, the links of rainfall variability and ocean SST anomalies are not always stationary in time. The impact of the Atlantic Niño on Sahel rainfall changed in the 1970s: before this decade the Atlantic Niño led to negative anomalies of rainfall over the Sahel caused by a more southern location of the Monsoon, while, after the 1970s, the concomitant presence of a Pacific La Niña (see above) counteracted this effect leading to no significant rainfall anomalies over the Sahel. The Pacific El Niño events also had stronger influence on Sahel rainfall after the 1970s. In addition, a link between the Mediterranean and Sahel rainfall was active in recent decades, as well as in the first half of the 20th Century. The influence of Atlantic and Pacific El Niño events on North Eastern South American rainfall also changed after the 1970s, and the combined effect of the Atlantic Niño and the Pacific La Niña amplified the rainfall response. Atlantic multidecadal variability was shown as the leading factor in modulating the link between SST and continental rainfall. These results are important for understanding the predictability of continental climate.
We also improved understanding of the longer-term changes in West African rainfall. In particular, we emphasize the importance of Atlantic multi-decadal variability in driving Sahel rainfall variability. Our analysis of climate model simulations further indicates that global warming induced changes in continental rainfall are very uncertain, and during the historical period not detectable.
Advanced statistical methods
PREFACE has developed advanced statistical methods and bias correction techniques to improve climate predictions. A key achievement of the project was the development of Bayesian hierarchal model framework. It was used to construct different statistical models to quantify systematic climate model errors. The statistical models are (a) Bayesian (to incorporate information from different sources, including expert knowledge) and (b) hierarchical (i.e. it treats errors at data, process and parameter level separately and in a transparent way). Three different models were constructed to focus on a purely spatial, a purely temporal, or a spatio-temporal assessment. They all build on the so-called state-space approach, for which unobservable state variables are used to directly model (statistically) the process of interest generating the observed data. The explicit modelling of the noise at the process level allows for better quantification of uncertainty. The power of the three approaches was demonstrated by (1) identifying forecasts drift and diagnosing its various causes using the purely temporal model, (2) by robustly determining the common tropical Atlantic bias in multi-model ensemble using the purely spatial model; and (3) using the spatio-temporal model to determine time varying model biases. The model codes have been made freely available.
In addition, a statistical prediction model that accounts for non-stationary teleconnections was developed based on Maximum Covariance Analysis. The model was used to predict tropical Pacific SST and Sahel rainfall. The approach was also extended to demonstrate the potential to predict variations in West African fish stock, based on the impact of El Niño on the region.
Improved tropical Atlantic climate predictions
PREFACE has lead to enhanced climate modelling and prediction capabilities, on seasonal to centennial time scales for the tropical Atlantic. Such improved predictions can have major socio-economic benefits, including for ecosystem and fisheries management.
We have demonstrated skilful prediction of equatorial Atlantic SST, and shown that reducing climate models biases improves skill in predicting Atlantic Niño events and their impact on continental climate and the Pacific. Several current operational seasonal climate prediction systems are able to predict Atlantic Niño variations in boreal summer from the beginning May. These models also show skill in predicting rainfall in limited regions affected by the West African monsoon. However, the majority of models are not skilful in predicting eastern equatorial Atlantic, even one month in advance. Analysis of the operational prediction systems shows that there is a relation between model mean state error and prediction skill, with models with reduced SST bias better predicting SST and rainfall. The impact of bias reduction on skill was confirmed using PREFACE models with reduced model error. In particular, we showed the anomaly coupling approach leads to significant skill in predicting equatorial Atlantic SST and that this is because the mechanisms for the Atlantic Niño variability are better represented (see above). This important result highlights the need to improve the mean state of coupled models in the Tropical Atlantic in order to enhance the predictions over the region.
Sahel rainfall remain at similar levels in the near future
Decadal variations in Sahel rainfall can now be predicted. The skill is found to depend on how well Atlantic multidecadal variability is predicted, but also on the response to global warming. Therefore models initialised with concurrent ocean observations are more skilful in predicting Sahel rainfall than historical forced simulations. Furthermore, the skill in predicting decadal variations in Atlantic SST depends on the initialisation method, but at lead-times longer than a few years the impact of global warming also becomes important. Our analysis indicates that there will be little change in Sahel rainfall for the near future (2016-2019) with respect to the previous four-year period (2011-2014).
The equatorial Atlantic Cold Tongue will weaken under global warming.
We have assessed how long-term climate change will impact tropical Atlantic climate variability, and we have shown that model biases introduce uncertainty in the future evolution of the tropical Atlantic and the West African Monsoon. This was achieved by performing climate projections with multiple climate models and with several approaches for model improvement: anomaly coupling, updated model configurations, and new parameterisation for vertical mixing (see above). We showed that improving the simulation of the tropical Atlantic climatology increases the sensitivity of the eastern equatorial Atlantic to the global warming. In particular, a model that realistically simulates the Atlantic Cold Tongue shows greater warming of the SST in the Gulf of Guinea than a model that poorly simulates the Atlantic Cold Tongue. Correspondingly, the tropical Atlantic ITCZ shifts more southward under the global warming in a less biased model. The consistency among our model results and the underlying physical mechanism suggest that the equatorial Atlantic Cold Tongue will weaken under global warming. In addition, we have shown that long-term global warming can influence the dominant patterns of Atlantic variability and the interaction between tropical ocean basins. However, our model experiments do not agree on the extent to which global warming will impact variability in the tropical Atlantic.
Disentangling environmental and anthropogenic pressures
PREFACE has contributed to disentangle environmental and anthropogenic pressures on pelagic fish stocks in the tropical Atlantic. In particular, we have increased understanding of the interrelation among marine ecosystems, climate variability and change, and fisheries, through exploratory analysis of historical and newly collected data. More than two decades of historical data allowed long-term changes and short-term variations to be studied. These include fish monitoring survey data acquired within the Nansen Programme executed by the Food and Agricultural Organization of the United Nations and funded by the Norwegian government. PREFACE cruises contributed to close data gaps and collect unique data, which included Tuna tagging. All these data were collected taking advantage of both local and European expertise, and contributed to training of scientists. With these data we have analysed three different ecosystem levels: top predators, small pelagics, and micronekton.
In terms of top predators, the analysis of Yellowfin tuna (YFT) catches in Cape Verde revealed the importance of climatic factors and fish stock to model local catch dynamics. There is a strong effect of subtropical wind stirring with a subsequent cooling of the surface layer that leads to an increase in abundance. This is evidenced by negative correlations to North Tropical Atlantic sea surface temperature and the North Atlantic Oscillation, which is the dominant mode of large-scale atmospheric variability. A YFT catch model using anomaly corrected climate projection data reveals that improved governance is essential to cope with negative impacts of climate change. In terms of prey fish dynamics, the first long-term and large-scale comparison of mesopelagic fish data in terms of biomass size spectra reveals significant changes in terms of trophic efficiency in oceanic food webs. Together, this highlights the impact of climate on local fisheries with important implications for fisheries economics.
In terms of small pelagics, the analysis of historic time series revealed the poleward expansion of tropical species during the last 25 years on both hemispheres. It also revealed differences in the causes of year-to-year variations in stock size and distribution.
Off northwest Africa, the northern range of the Sardinella aurita stock has shifted by two degrees to the north, to regions inhabited almost exclusively by subtropical sardines during the 1990s. For instance, the distribution of the commercially important S. aurita has shifted a few hundreds of kilometres northwards. The observed distributional shift coincided with the systematic warming characterizing the Canary Current region during the same period. While analysis of the short-term variations indicates that coastal wind and wind-induced upwelling is the key driver of fluctuations in biomass of small pelagic fish, rather than only ocean temperature. In Senegalese waters abundance of sardinella is significantly correlated to wind strength during the upwelling season whereas interannual fluctuations in the proportion of S. maderensis to S. aurita in the catches depends on how early in the year is the onset and how long is duration period of the upwelling season. An early or late upwelling season favours the sedentary S. maderensis or migratory S. aurita, respectively.
Off southwest Africa, the adult sardinella exhibited a strong interannual variation in the seasonal migration cycle, synchronised with the strength of the interannual equatorial climatic events. In the southern range of sardinella distributions a change in fish age structure was observed from adult migratory fish during the 1990s to an increasing proportion of locally born juveniles during the current decade. Overall the principal findings for the southern region include: (1) identifying patterns of correspondence between the strength of seasonal coastally trapped (Kelvin) waves and seasonal migration of sardinella along the Angolan coast, (2) associating the two major sardinella biomass growth periods in Angolan waters, 2006-2007 and 2012-2013, with the interannual climatic scenarios induce by equatorial forcing and (3) identifying the gradual expansion of coastal nursery areas of S. maderensis and S. aurita from the Gabon-Congo coastal regions during the 1990s towards the remotely forced Angolan upwelling area (6°-13°S) during the current decade.
The analysis of micronekton revealed that the three LMEs are highly variable systems, with the Canary Current LME and the Benguela Current LME being particularly impacted by global warming, especially in their lower latitudes. From 1995 to 2015 in the Canary Current LME (CCLME) there was a significant deepening of the micronekton distribution and an increase of the diel vertical migration amplitude. The interannual variability is not significant, showing that the underlying environmental forcing is associated with relatively stable processes. Sea surface temperature was shown to have a minor influence in the north CCLME, but a pronounced effect in the south CCLME. Considering the relative importance of oceanographic factors, stronger ecosystem perturbations are expected in Benguela Current LME (BCLME) than in the CCLME. Overall, oceanographic factors were found to have a significant influence in all LMEs. Therefore, global warming is expected to cause important changes in the three LMEs and thus in the fisheries sector.
Climate change effects on small-scale fisheries and coastal communities, and implications for management
PREFACE has better understood climate change effects on small-scale fisheries and coastal communities, and assessed uncertainties and implications for management. This was achieved by (1) developing ecological-economic models, (2) making an impact assessment by coupling these models to PREFACE climate change projections, and (3) by extensive surveys of coastal fishing communities to acquire information for developing the economic models and providing policy relevant advice.
Single species coupled bio-economic models were developed to understand the threats climate change poses to fisheries with focus on small-scale fisheries. The models simulate the consequences of future climate change and allow a comparison of optimal and realized harvest, and their impact on consumer and producer welfare. PREFACE acquired knowledge on the impacts of climate on fish stock was used to develop the biological component. The economic component of the model was based on available literature and collected through surveys (see below). Due to their importance for local communities and data availability, we focused on the artisanal purse seine fishery for Sardinella aurita in the four major Senegalese fishing regions and the artisanal hand line fishery for YFT in three island regions of the Cabo Verde archipelago.
We selected artisanal fisheries in Senegal for detailed impact assessment, because the effects of climate change on them may be severe, as the dependence on fish as a protein source and on fishing income is strong. We drove a bio-economic model with future climate change projections from two different PREFACE models and different fuel subsidy scenarios. The main effect of climate change indeed is a distributive one. Biomass becomes more vulnerable to Senegalese purse seines over a certain time horizon, because the fish spend more time in the close-to-shore area. In the long term, stock survival requires a stricter response to over-fishing. The cancellation of fuel subsidies entails an increase in total benefits and thus, not only are fuel subsidies expensive for taxpayers, they also reduce welfare in total. All runs reveal that the fisheries contribution to income security will in the future be more important than the food security impact. Our results are informative for policy makers, as they shed light on the effect of climate change and counteracting measures while incorporating endogenous adaptation of actors. We also explicitly show that policy choices dominate socio-economic model outcomes rather than climate model uncertainty.
The socio-economic surveys conducted in Senegal and Cabo Verde shows that impacts of weather changes on the resource seem evident. For Senegal, this is dominantly visible in the perceptions on rainfall pattern effects, while for Cabo Verde, wind changes may prove more important, on the resource as well as on fishing effort in the artisanal sector. The data also shows the sensitivity and adaptive capacity of the fishing sector in the two countries. While in Senegal specialization is at the individual level but on a rather diverse range of target species and gears used, the Cabo Verde fisheries depend to a dangerous degree on only six species, and most importantly tunas. This is true on individual, fishing sector and total-fishery level. In that regard, while only individual sensitivity is high in Senegal and the artisanal sector itself is likely too diverse to be hit by singular species changes, the Cabo Verde fishery is very sensitive to certain species changes.
The economic analysis of the Senegalese fisheries on small pelagic species however, showed that this sector is indeed vulnerable to changes in catchability affected by changes in sea surface temperature. A sea surface temperature shock during summer can increase the catchability by one third. However, long-term effects can be negative due to decrease in population biomass. In addition, economic drivers, like fuel prices also have a significant negative effect on the profitability of the purse seine and encircling net fisheries. For the Senegalese purse seine fishery, operating costs have increased by 90% over the last twenty years, mainly due to an increase in fuel prices.
Enhanced cooperation between European and African researchers
PREFACE has increased cooperation between European and West African partner institutions leading to enhancement of research capacities in the fields of ocean physics, ecosystem and fisheries as well as climate research in West Africa. The close cooperation between different European and African partner institutions resulted in common experiments at sea, data sharing and analysis, publishing of historical observations, as well as application of ocean modelling on regional scale. For example, it led to a new seasonal climatology for the heat and freshwater budget in the mixed layer. The climatology is greatly improved in coastal regions compared to previous climatologies. This is because of the incorporation of unpublished hydrographic data from the EAF Nansen program and other programs of the PREFACE partners that was only possible through close cooperation. Another great success was the dissemination of hydrographic and ship-board velocity data collected during biannual cruises carried out with the RV Dr. Fridtjof Nansen since 1984 within the FAO-NORAD EAF Nansen Programme. Our Angolan partner led this publication of this data.
PREFACE has also helped to train a new generation of researchers, including those from the region. In total there were of 21 PhD theses (15 from African students) and 21 Master theses (18 from Africa students). Its summer schools (two in Africa), targeted workshops, and interdisciplinary meetings have provided education to continue important research on the tropical Atlantic climate and its impacts. The joint organisation of two project meetings with annual tropical Atlantic variability conferences helped integrate young researchers into the international research community. Lastly, we have helped to ensure the long-term sustainability of the tropical Atlantic observing system, by increasing awareness and developing required competence among regional players.
Potential Impact:
Impacts and exploitation
1. Better monitoring of the eastern boundary upwelling
(Deliverables D2.3 D2.8 D2.9 D3.1 D4.4 D12.1 D12.2 D12.3 D12.4 D14.3 and D14.4)
PREFACE has drastically improved the ocean and climate observing system in the eastern Atlantic, allowing a continuous monitoring of tropical Atlantic variability and climate change. PREFACE has installed key moorings in the southern and northern coastal waveguides, extending the PIRATA (Prediction and Research Moored Array in the Atlantic, www.pmel.noaa.gov/pirata/) network toward the eastern tropical South Atlantic, particularly with the new buoy, “Kizomba”, located strategically in an area problematic to ocean-atmosphere coupled climate models, offshore of the Congo River plume and upstream of the Angola Current. Exemplary also is the deployment, in association with several laboratories across West Africa, of an oceanographic and meteorological measuring buoy, Melax, in the heart of the Senegal-Mauritania upwelling.
Along with the increase in new ocean observations, previously unreleased historical oceanographic data acquired in the framework of the FAO EAF Nansen programme is being comprehensively exploited thanks to the close North-South collaboration in PREFACE. This represents the establishment of time series long enough to monitor climate change in the African coastal regions of the Atlantic. This provides valuable input for future ocean monitoring by the EAF Nansen Programme, an initiative to support the implementation of the ecosystem approach in the management of marine fisheries in developing countries, among other programs.
PREFACE has contributed knowledge on the optimal design of the future network to the Tropical Atlantic Observing System review. It is also helping to ensure the long-term sustainability of the network, by making data freely available on international data archives, by increasing awareness and by developing required competence among regional players. The better observing system will be exploited by regional institutes and centres responsible for monitoring and managing their marine resources, and by international climate prediction centres for improving their global models and predictions.
2. New process understanding and parameterisations
(Deliverables D3.2 D3.3 D6.3)
PREFACE has achieved an understanding of the role of near-inertial wave mixing and its role in ocean temperature and momentum budgets, and has developed and tested parameterisation for climate models. PREFACE has also worked on 1-column models to test vertical mixing schemes. We have understood the processes controlling deep equatorial jets and their role in exciting equatorial basin modes and explaining intraseasonal, seasonal, and interannual variability in the equatorial currents and SST. We have improved understanding of the modelling of these processes. We also achieved a deeper understanding of ocean atmosphere interaction. Modelling centres, through the PREFACE network and openly available publications, will exploit these results to improve ocean and climate models.
3. Improved understanding of tropical Atlantic climate and its global impacts
(Deliverables D2.5 D2.8 D2.9 D3.1 D3.4; D4.1-4.3; D5.1 D7.2 D8.2 D9.1 and D9.2)
PREFACE has achieved a better understanding of the mechanisms and predictability of tropical Atlantic climate, and of its global impacts. This is especially the case for both eastern boundary upwelling African-Atlantic regions and the Gulf of Guinea. The following are a few showcase results: We now better understand the importance of remote versus local processes in driving Benguela Niño events, as well as the role of dynamical versus thermodynamical ocean-atmosphere interaction for Atlantic Niño variability. The prominence of the tropical Atlantic in the climate system has been demonstrated. For example, it is now clear that the Atlantic Niño strongly influences the El Niño Southern Oscillation. We have also shown the importance of internal generated climate variability for long-term changes in the tropical Atlantic, and highlighted the uncertainties related to long-term climate change.
International modelling, operational prediction centres, and impacts researchers working on the tropical Atlantic can exploit our findings on the mechanisms for tropical Atlantic variability in order to improve their modelling and prediction capabilities, to provide confidence in predictions and projections of climate, and to develop impact models.
4. Improved climate models
(Deliverables D2.5 D2.8 D2.9 D5.2; D6.1-6.3; D7.1 D7.2 D8.1 and D8.2)
PREFACE has led to enhanced climate modelling and prediction capabilities, on seasonal to centennial time scales. Our targeted ocean model experiments and their comparison with observations provide detailed process-based understanding and long-term directions to improve ocean and climate modelling and prediction capabilities. Our multi-model analysis and coordinated model experiments greatly contributed to the detailed understanding of the causes of model systematic errors and their impact on simulated variability and prediction skill. This detailed understanding has lead to the development of new improved models and alternate approaches to reduce model error.
Global climate modelling and prediction centres will exploit our new approaches to reduce model error and the development of new parameterisations for ocean mixing to improve their modelling systems. Our novel tools to diagnose model errors and their consequences will be exploited by the wider scientific community as they provide new directions for using multi-model ensembles.
5. Enhanced climate prediction
(Deliverables D10.1-10.3; D11.1 and D11.2)
PREFACE has demonstrated the potential to skilfully predict climate on seasonal-to-decadal timescales in the tropical Atlantic, and better understood uncertainties in climate change projections. In particular, we have used improved models and alternate model configurations to reduce model errors and to show improved prediction skill and assess model uncertainties. In addition, PREFACE has developed advanced statistical methods and bias correction techniques to improve climate predictions. We have demonstrated skilful prediction of equatorial Atlantic SST, rainfall in some regions of the West African Monsoon, and decadal shifts in Sahel rainfall.
The improved climate prediction approaches will be exploited by operational centres, and will in the longer term lead to improved operational regional and global predictions. Climate services providers and users can further exploit these. Thus, such improved predictions should have major socio-economic benefits, including sustainable land use and ecosystem-based fisheries management.
6. Deeper understanding of the function of marine ecosystem
(Deliverables D2.6-2.9; and D12.1-D12.4)
PREFACE activities have led to a deeper understanding of the function of marine ecosystem so that climate driven shifts can be better predicted. The following are showcase results. We have shown the poleward expansion of tropical species during the last 25 years on both hemispheres, and related this to environmental changes. The warming of the South Atlantic during the last three decades was associated with threefold increase and southward shift in Sardinella biomass. While the warming off northwest Africa caused the northern range of the Sardinella aurita stock to shift by two degrees to the north, to regions inhabited almost exclusively by subtropical sardines during the 1990s. We have also understood year-to-year variations in Sardinella to Benguela Niño variability, and shifts in tuna distributions to large-scale atmospheric wind and oceanic temperature patterns.
The development of an open-source software, Matecho, for analysing acoustic data on different ecosystem trophic layers (such as zooplankton and small pelagic fish) will enable not only partners in PREFACE to further study and understand ecosystem changes in their marine waters but also other countries in the region. Other new descriptors of micronekton (an essential component of marine food webs) that can serve as a monitoring tool for the health of ecosystems, are being refined and soon to be made available through publishing. Meetings and small workshops are being carried out to further disseminate the existence and use of this software and new descriptors. Importantly, these will enable further comparative analyses for a better monitoring of ecosystems and sustainable management.
Regional fisheries managers and policy makers to ensure long-term sustainability of fish stock can exploit these results. Researchers aiming to develop ecosystem models and predictions can exploit these results. The potential to predict changes in small pelagic fish distributions by coupling bio-climatic models to improved monitoring and climate predictions can produce results to be further exploited by fisheries managers and policy makers.
Another PREFACE finding points to a response pattern in the behaviour of micronekton to long-term climate changes in the water column, which could have an implication on the ocean carbon pump. These results however are still preliminary and currently under scrutiny before international publication.
7. The sustainable management of West African fisheries
(Deliverables D2.6-2.9; D12.1-12.4; D13.1-13.3)
PREFACE can provide science-based support to policy-makers for the sustainable management of West African fisheries in the context of climate change. Knowledge gained through surveys of local fishing communities, better understanding of climatic and anthropogenic influences on marine ecosystems, and improved predictions together lead to improved ecological-economic models and predictions for the region. The following are some key results. Detailed impact assessment for artisanal fisheries in Senegal showed that Sardinella aurita becomes more vulnerable to Senegalese purse seines over a certain time horizon, and the long-term stock survival will require a stricter response to overfishing. Bio-economic modelling further shows that cancelling fuels subsidies for fisheries benefits the entire community. We also highlight that Cape Verde fishery is very sensitive to climate driven species changes, because it depends to a dangerous degree on only six species, and most importantly, tuna. Senegalese fisheries is less specialized and likely more robust to climate driven ecosystem change. Researchers will exploit these results to develop impact models, and also have the tools now to further extend our survey approach, with application in other countries. The application of these models can then benefit fisheries managers and policy makers, who can also further exploit our survey results.
PREFACE reemphasizes the known but seemingly often ignored evidence – albeit because challenging – that sustainable fisheries management requires that fish stock assessments include the systematic monitoring of environmental data, and that management be at a level not restricted by national borders but rather by the migratory behaviour of fish stocks. This is particularly relevant in the face of climate change, as has been shown with the sardinella stocks studied in PREFACE. It is also important that regional fisheries management organizations be given the tools and resources to assess more than a single-species, and in this light, it is crucial that dialogue between involved actors be increased. Importantly, PREFACE stresses that the enhanced dialogue between fishery commissions, scientists, fisherfolks, and higher-level decision bodies can only be efficient if each party is sensitised to the others technical knowledge, that is, there is a need for knowledge brokers. Hopefully, some of the junior scientists educated within PREFACE will become such knowledge-brokers in their local regions.
8. Strengthened capacity in oceanography, climate, marine ecosystem, and socio-economic research
(Deliverables D2.3-2.5 D2.8 D4.4 D10.1-10.3 D12.1-D12.4 D13.1-13.3)
By strengthening existing and fostering new collaborations, PREFACE showcases and further corroborates the significant benefits brought by enhanced cooperation between European and African researchers working on tropical Atlantic climate and its impacts. The project dedicated a substantial amount of resources to capacity strengthening and training of students, postdoctoral fellows and early career scientists of Europe and Africa by means of dedicated bursaries for PhD and Master students, fellowships for postdoctoral fellows, organisation of summer schools and workshops and support for participation to oceanographic surveys (R/V Meteor 2014-2017, PIRATA campaigns 2014-2018, and R/V Thalassa off Senegal in 2015) and international scientific conferences. More specifically, a total of 21 Masters and 21 PhD theses were directly recruited by the project or benefitted from, and contributed to, the project through collaboration. PREFACE also supported, through teaching and supervision, the continuation of the Masters programme: Regional Master 2 SOAC-OA (Sciences de l’Océan, de l’Atmosphère et du Climat : parcours “Océanographie et Applications”) in West Africa, ICMPA/UAC-IRD-UPS-UFPE, Cotonou, Benin. The capacity strengthening activities with ICPC partners were always elaborated to suit the interests and demands of the ICPC partner institution, providing a general scientific education with focus on the objectives inherent to the involved research programs. Additionally, PREFACE has produced a series of open-source software and models (e.g. VEROS, S4CAST v2.0 Matecho, the DLM in described in D10.3) contributing to provide the education and tools required to continue important research on the tropical Atlantic climate and its impacts. PREFACE can thereby be considered as one of many useful actions and initiatives paving the way for the implementation of the Belém Statement on Atlantic Research and Innovation Cooperation and for improved coordination and alignment of programmes/initiatives and projects between South and North Atlantic regions and with the EU and its Member States.
9. The development of climate services
(Deliverables D2.2 D2.3 D2.5-D2.9; D11.1 D11.2; D12.1-D12.4; and D13.1-D13.3)
PREFACE has brought together an interdisciplinary team of African and European scientists to achieve a step-wise improvement in our understanding and modelling of the tropical Atlantic climate and its impacts. Climate models can now provide reliable climate simulations and predictions in the region. When combined with knowledge of environmental-ecosystem interaction, this information can be used to predict changes in the state of marine ecosystems at timescales relevant for anticipatory and more effective policy or management decisions. Furthermore, we have contributed to strengthen capacity by training a new generation of scientists capable of realising and further developing climate services. This includes researchers with expertise in communicating with decision and policy makers. Together these achievements provide the technical and scientific basis for the development of climate services in the region. We recommend to promote and encourage the explotation of the climate models information to evaluate climate changes and to perform useful and reliable predictions of the living marine resources at timescales useful for decision making. Exploitation of climate services by the region will help ensure the sustainable development.
10. Sustainable development
As can be concluded from the above, PREFACE contributes to the United Nations’ 2015 sustainable development agenda to end poverty, protect the planet and ensure prosperity for all. Particularly, PREFACE contributes to the Sustainable Development Goals 13, to “Take correct action to combat climate change” and 14, to “Conserve and sustainably use the oceans, seas and marine resources”, particularly by bringing us closer to reaching targets 14.7 14.4 14.6 14.5 14.2 and 14.3.
Main dissemination activities
(contributing to the exploitation of results and achievement of impacts)
The project website is designed to inform the global community about PREFACE, who we are, our activities and latest results, related programmes and projects, events, and employment and training opportunities. It also contains basic publicly available outreach material, such as the project factsheets, translated into French and Portuguese for a wider outreach on the African continent. Additionally, translated PREFACE outreach material has been prepared and provided upon request, such as for IRD’s PI involvement in political workshops (Fisheries or Environmental Ministries and regional Environmental and/or Fisheries monitoring commissions across the CCLME and GCLME) or after a request from the Norwegian embassy in Angola, in connection with visit of R/V Meteor in Luanda and high-level meeting organised upon the occasion (Milestone 16, Week 4 cruise report, available on website). The website thus aims to not only raise interest in PREFACE, but also to raise awareness in such topics as climate change, improving climate models, the need for a long-term tropical Atlantic observing system, impact of climate-related change on west African coastal communities, the need for ecosystem approaches for the sustainable management of fisheries, and the need for enhanced cooperation across the Atlantic and globally to achieve this.
PREFACE scientific findings have been disseminated through over 120 scientific papers in peer-reviewed international journals. Around 70 new publications are underway and a strong effort is being made to make all openly available. This includes datasets and other tools (such as models or open-source software), all of which are indexed on the website “Output” pages.
The consortium was very active in disseminating project results through communications at scientific events and co-organisation of conferences or conference sessions such as: the annual joint PREFACE-CLIVAR-PIRATA Tropical Atlantic Conferences (South Africa 2015, France 2016 and Brazil, 2017); the annual TACCOVAR colloquia (Benin, 2015-2017); the CLIVAR Open Science Conference “Charting the course for climate and ocean research”, (China, 2016); the ICES Annual Science Conference (2015-2017); the International Workshop on Marine & Atmospheric Sciences in West Africa, organised for the inauguration of the Ocean Science Centre Mindelo (Cape Verde, 2017); the Future Ocean conference on Advances in Integrated Ocean Research towards Sustainable Development, targeting early career researchers (Germany, 2017), several sessions at the European Geosciences Union General Assembly (Austria, 2014-2018) and at the IAPSO-IAMAS-IAGA Assembly (South Africa, 2017).
The project was also active in dissemination activities to the wider public through a series of articles in the popular press and small documentaries or interviews at national level, e.g. short documentaries of PREFACE research in the Canary Current region by beneficiaries UCM and Sorbonne Universités (previously UPMC) in collaboration with other national projects and programmes; TV clip promoting conference on ocean and climate, featuring INDP and the University of Cape Verde; lecture by Bourlès (IRD) on “PIRATA, un observatoire océanographique sous les tropiques”, transmitted on 88.8FM Marseille and www.radiogrenouille.com; participation in several science festivals and Open Door days, among others.
Further dissemination, involving also exploitation of results, consist in the inclusion of PREFACE findings: in courses given by PREFACE researchers, such as university courses in Bergen (Norway) on Climate Dynamics and the Atmospheric General Circulation and several lectures given in schools and summer schools; in teaching activities by UCM and UPMC at UCAD, in coordination with other Spanish, French and Senegalese national programmes and projects; during workshops in Morocco, Senegal, Cape Verde, Ivory Coast, Angola and Namibia, particularly those in Senegal and Cape Verde during which the results from the WP13 interview surveys (Deliverable D13.2) were presented to and discussed with the participating fisherfolk communities; in presentations given during visits to regional research centres and fisheries institutes; and in the Masters course at the UNESCO Chair in Mathematical Physics and Applications at UAC (Benin).
As a consequence of the achievements in PREFACE and wide dissemination, PREFACE scientists have been invited as experts to contribute to higher level scientific and/or political arenas upon several occasions, such as, but not restricted to:
+ The Tropical Atlantic Observing System (TAOS) Review commissioned by CLIVAR and white papers for the upcoming Oceans Observations Conference 2019 (http://www.oceanobs19.net/);
+ The CLIVAR EBUS working group, with several scientists from WP5 invited as members, including as chair;
+ The international “CAS-TWAS-WMO Forum" (CTWF) for Physical-Mathematical Problems Related to Climate Modeling and Prediction (that was jointly founded by the Chinese Academy of Sciences (CAS), the World Academy of Sciences (TWAS) for the advancement of science in developing countries and the World Meteorological Organization (WMO)), with contributions from the progress in the anomaly coupling studies from WP7;
+ The WCRP-JSC/CAS Working Group on Numerical Experimentation (WGNE), for increasing understanding of the nature and cause of errors in models used for weather and climate prediction, with contributions from WP7 and the coordinated experiments of WP6;
+ The seventh IPCC report, with contributions from Dr. Hyacinth Nnamchi (UNN, Nigeria) as a chapter lead author;
+ The German Federal Ministry of Economic Cooperation and Development (BMZ) and Corporation for International Cooperation (GIZ), with advice from CT5 for planning of future cooperation and development activities in Africa;
+ The Senegalese Ministry of Fisheries and Economy, with contributions from Dr. Patrice Brehmer to a discussion on “Technical Validation of the Fisheries Sectorial Policy Review”;
+ The Cran Montana Forum, with contributions from CT5 to the high level panel “Ocean Economy and Fishing industry, a strategic sector for Africa”, specifically for “The blue belt initiative, a South-South integrated regional cooperation and development tool”;
+ The Ministry of the Environment and Sustainable Development of Senegal (MEED/DEEC), with contributions to their workshop on “Operational Planning PAS/PNA – Project of Scientific Support to the National Adaptation to Climate Change in Senegal Plan”.
A selection of other contributions to relevant scientific and/or political events is:
+ The preparation of the Ocean Climate Declaration of Dakar, in collaboration with the AWA project and numerous stakeholders and policy-makers. This is a voluntary initiative of research and higher education institutions, programmes and projects, that aim to bring actions particularly by the Sub-regional Fisheries Commission, the Ministerial Conference on Fisheries Cooperation among African States Bordering the Atlantic Ocean (ATLAFCO), the UNEP Abidjan Convention, the Economic Community of West African States, the Arab Maghreb Union, the Economic and Monetary Union of West Africa, the European Commission and the United Nations, on issues related to climate impacts on the marine environment and fishing. It was presented at both the COP21 (Paris, November 2015) and COP22 (Marrakech, November 2016), besides being widely used in many of the CSRP-SRFC (Senegal) policy meetings across West Africa;
+ An invited lecture on “How can data from a vessel based platform make a contribution in global research, and what types of parameters should be monitored?”, based on experience and results drawn from PREFACE, at the 2nd Annual Hjort Open Science Seminar: Ecosystem research, advice and management - how to put policy into practice?, 9-10th June 2016 in Bergen, Norway; a seminar was related to the science planning for the new vessel Dr Fridtjof Nansen for the FAO EAF Nansen Programme;
+ The showcasing of PREFACE upon a reception organised on-board the R/V Meteor during the 2016 south-eastern boundary current cruise (Milestone 16), supported by INIP and the German Embassy in Angola and attended by the Angolan Minister for Fisheries, the Minister for Science and Technology, and the state secretary for aquaculture, Dr. Carlos Martinó Cordeiro. Correspondingly large was the interest by the media with several interviews given to journalists from press and television. In her speech, the Minister of Fisheries pointed toward the importance of climate change, which represents a grand challenge for the socio-economic development of Angola. The minister was pleased about the good collaboration with German marine research institutes and particularly thanked the German government for the support to investigate the dynamics of the marine ecosystem of Angola. Scientists onboard proceeded to present their research topics and the projects supporting their work, with special emphasis to PREFACE;
+ Showcasing of PREFACE by UCT at a workshop hosted by the ESASTAP (strengthening T,R&I cooperation between Europe and South Africa), on “SOUTH AFRICA (SA)-EUROPEAN UNION (EU) MARINE AND POLAR RESEARCH COLLABORATIONS”;
+ Showcasing of PREFACE by GEOMAR and IRD at the “A New Era of Blue Enlightenment” meeting for the signature of the Belém Statement on Atlantic Research and Innovation Cooperation, at the “Connecting to better observe the Atlantic Ocean” session, co-organised with the EU H2020 AtlantOS project, the ETASTAP session (South Africa - EU cooperation) and the INCOBRA session (Brazil - EU cooperation);
+ Co-organisation of a side-event at the CoP22 (November 2016, Marrakesh, Morocco) Ocean Theme: Oceans, Transport and Energy. The side-event was called “Fisheries and aquaculture in the context of climate change: challenges and opportunities” and jointly organised by the FAO with the projects AWA and PREFACE, showcasing key achievements from these projects and sharing experience and lessons learnt for coordinated action towards a sustainable future;
+ Co-organisation of a side-event at the UN Ocean Conference (June 2017, New York, USA). The side-event was called “Africa in Action for Fisheries and Aquaculture to Face Climate Change: Solutions and Dialogue for the Ocean Economy” and PREFACE contributed specifically with the presentation by Dr. Patrice Brehmer on “AWA and PREFACE experience: foundation for the establishment of an observatory for fisheries and the marine environment, targeting SDG 14 goals in West African waters” (https://preface.w.uib.no/events/events-in-detail/#unoc201706);
+ The PREFACE Science-Policy session (please refer to Deliverables D2.6 and D2.9) April 2018, Spain;
+ The workshop on “Small-scale fisheries in the South” on 25-26 June 2018 in Brest, France, organised by the University of Brest in collaboration with IRD, during which a session was dedicated to the presentation of PREFACE results. 40 researchers, managers, funding agencies, training institutions and consultants were invited to discuss issues for a better integration of research in development programmes and development aid supporting sustainable small-scale fisheries.
Finally, a news piece on PREFACE, by journalist Christian Nielsen assigned by Intrasoft, is being edited for publication in the series of EU-funded R&I actions success stories within International Cooperation (http://ec.europa.eu/research/infocentre/index_en.cfm). The Coordinator is also preparing a news piece on the project, together with the Communications unit of the Bjerknes Centre for Climate Research (Bergen, Norway), to be released in Autumn 2018. It will focus on success stories of PREFACE scientists, such as those of Dr. Abdoulaye Sarré from ISRA-CRODT (Senegal), who through PREFACE grew from electronic engineer to Technical Advisor on Acoustics for the Senegalese Ministry of Fisheries, Dr. Pedro Tchipalanga from INIP (Angola), who shows the value of sharing and collaborating on previously private observational data (Tchipalanga et al., 2018) and Dr. Marta Martin del Rey, who started as a PhD on the project, continued with valuable contributions as a Postdoctoral researcher, and has now earned a prestigious MSCA individual fellowship, thanks to her achievements and networks gained through PREFACE. A final summary article is also in preparation (refer to Deliverable D2.8) as is the further editing for publishing of the policy brief (Deliverable 2.9) which will be disseminated through the networks of the consortium and those who were invited to participate in the Science-Policy session (Deliverable D2.6).
Last but not least, a collection of entertaining illustrations about PREFACE were produced during the PREFACE final meeting by the illustrator Bas Kohler. These are in high demand by project partners and participants of the Science-Policy Session, for reuse in lectures, posters and other science dissemination. They will soon be made freely available to the public. We are also discussing transforming them into a short animation or booklet on the project once activities resume after the summer break. The Project Office counts on the UiB office in Brussels and contact with the DG R&I Communications Team (such as Mr. Robert-Jan Smits or Mrs. Alexandra Ruete) for further promotion of this product, and will attempt to submit any such product to AORA (https://www.atlanticresource.org/) for further dissemination on their News & Events page.
A continually updated detailed list of PREFACE dissemination activities, including scientific publications, is available on CORDIS (Community Research and Development Information Service), the European Commission's primary public repository and portal to disseminate information on all EU-funded research projects and their results in the broadest sense.
List of Websites:
www.preface-project.eu (same as https://preface.w.uib.no/ ).
Project Office (coordinator): prefaceproject(at)uib.no
Project Leader: Noel Keenlyside, Noel.Keenlyside(at)uib.no; Tel. ++47 55 58 20 32
Project Manager: Mahaut de Vareilles, mahaut.vareilles(at)uib.no; Tel. ++47 55 58 37 08
Data Manager: Benjamin Pfeil, benjamin.pfeil(at)uib.no; Tel. ++47 55 58 98 39
Financial Manager: Robert Clarke, Robert.Clarke(at)uib.no; Tel ++47 555 821 26
PREFACE has advanced significantly towards its overarching goal to enhance prediction of tropical Atlantic climate and its impacts.
Key oceanic processes in the tropical Atlantic were better understood, by enhancing and analysing equatorial and coastal observations, and by numerical experimentation. We quantified the seasonal surface mixed-layer heat and salt budgets in the Benguela and Gulf of Guinea regions, and the impact of wind-excited near-inertial waves and chlorophyll-induced turbidity on the heat budget. Our measurements identified pronounced semi-annual, annual and interannual variability of the equatorial zonal currents, and quantified for the first time the seasonal cycle of the southward Angola Current. Our model studies indicate that the warm SST bias could be reduced close to the coast by using high ocean model resolution and improved atmospheric forcing. We identified the dynamics of deep ocean jets at the equator and causes of Benguela events, which strongly impact fisheries. Close cooperation between different European and African partner institutions were established or improved resulting in common experiments at sea, data sharing and analysis, publishing of historical observations, as well as application of ocean modelling on regional scale.
PREFACE has identified key causes of systematic error in the tropical Atlantic in climate models. In general, atmospheric model errors are the dominant cause of the large equatorial and southeastern Atlantic SST biases, but the relative roles of wind, cloud-radiative and ocean dynamics differ regionally, seasonally, and among models highlighting the importance of a multi-model approach. Errors in the Southern Ocean radiation budget can also drive tropical biases. Model biases cause an underestimation of dynamical ocean-atmosphere interaction and affect the relation between Atlantic and Pacific climate variability. We have developed a new parameterisation for ocean mixing, high-resolution models, and novel correction techniques to improve the simulation of tropical Atlantic climate and its global impacts.
PREFACE has improved understanding of tropical Atlantic variability and predictability. Tropical Atlantic SST may be predicted based on equatorial waves, deep equatorial jets, and remote forcing from the Pacific and South Atlantic, while Atlantic multidecadal variability modulates their importance. Reducing climate models biases improves skill in predicting Atlantic Niño events, and their impact on continental climate and the Pacific. We showed that Atlantic and Pacific decadal variability and global warming impact African and South American rainfall. Decadal variations in Sahel rainfall can now be predicted based on Atlantic SST. Model biases introduce uncertainty in future evolution of the tropical Atlantic and the West African Monsoon. Lastly, advanced statistical methods were developed to improve SST and rainfall predictions, and to account for model bias.
PREFACE is increasing understanding of the interrelation among marine ecosystems, climate variability and change, and fisheries; and of the socio-economic implications of future climate change. We analysed more than two decades of data from shipboard surveys, carried out with regional partners and by the FAO EAF Nansen project. Climate impacts the distribution of key fish stocks along the African-Atlantic coast and tuna in the tropical Atlantic, with some pelagic species showing clear migration trends and others appearing resilient to current climate change. Fish prey is also affected: over the last 20 years, micronekton distributions have become deeper along this coast. Thus, climate prediction can be useful for fisheries management. We performed an impact assessment for West African fisheries using climate change projections with reduced biases and adapted ecological-economic models. We performed questionnaires to assess the economic and social situation of local fishing communities, and their vulnerability to climate driven changes in the ecosystem.
Project Context and Objectives:
The tropical Atlantic climate recently experienced shifts of great socio-economic importance. The oceanic changes were largest in the eastern boundary upwelling systems, globally important regions for marine productivity and climate. African countries bordering the Atlantic depend upon their ocean for societal development, fisheries, and tourism. They were strongly affected by these changes and face important adaptation challenges associated with global warming. Unfortunately, the tropical Atlantic is a region of key uncertainty in the climate system: state-of-the-art climate models exhibit large systematic errors, prediction of climate variability and climate change are highly uncertain, and how shifts in climate impact marine ecosystems is largely unknown, as are their possible global socio-economic impacts.
To redress this situation, PREFACE joins European and African expertise to provide the first comprehensive assessment of the tropical Atlantic. In four research foci, sophisticated observation systems, extensive field experiments, regional and global scale modelling capabilities, and socio-economic and marine ecosystem models are combined to improve the prediction of tropical Atlantic climate and its impacts.
The first research focus – Role of ocean processes in climate variability – aims to better understand the oceanic processes affecting sea surface temperature (SST) in eastern tropical Atlantic upwelling regions on seasonal to decadal time scales, by using historical and new data as well as medium to very-high resolution forced ocean models. We focus on the mixed layer heat and freshwater budgets, the tropical Atlantic circulation, and the equatorial and coastal wave response. By evaluating the ocean models, we will suggest improvements to reduce biases (systematic errors) in ocean and coupled climate models.
Evaluation of current climate models and reduction of model bias is the second focus, aiming to better understand and subsequently reduce systematic errors in climate models in the tropical Atlantic. Our strategy is to analyse the development of model systematic error in climate predictions and then perform coordinated, multi-model sensitivity experiments to identify the key processes that require improved model representation to mitigate these systematic errors. Additionally, we aim to understand the relation among errors in variability, feedbacks, and atmospheric-links to other regions of the globe.
Our third focus – Climate prediction in the Tropical Atlantic – aims to clarify the mechanisms for climate variability and predictability on seasonal to decadal time scales, by analysing observations and numerical experimentation. We also investigate interactions between the Atlantic and other regions, and both natural and anthropogenic factors. The goal is to improve the prediction of tropical Atlantic climate, and to assess the impact of model systematic error on prediction skill. We are developing a unified statistical prediction framework, and will also use state-of-the-art climate prediction models.
PREFACE’s last research focus – Impacts of climate change on pelagic functional diversity in the tropical Atlantic with effects on western African fisheries economies – aims to disentangle environmental and anthropogenic pressures on pelagic fish stocks in the tropical Atlantic, through exploratory analysis of historical and newly collected data. We also aim to better understand climate change effects on small-scale fisheries and coastal communities, and to assess uncertainties and implications for management. Our approach is to develop socio-economic models, built on knowledge of the marine ecosystem and fishing communities, and to drive them with climate change projections.
PREFACE envisions a closer cooperation between European and West African partner institutions leading to enhancement of research capacities in the fields of ocean physics, ecosystem and fisheries as well as climate research in West Africa.
Project Results:
Towards prediction of tropical Atlantic climate and its impacts.
PREFACE was a bold and ambitious project that targeted one of the marine regions that will be more affected by climate change and its consequences. It provided the first comprehensive, multidisciplinary assessment of the tropical Atlantic climate, from observations to predicting its socio-economic impacts.
Enhanced observations and medium-to-high resolution ocean model experiments
PREFACE has substantially improved the understanding of key oceanic processes affecting sea surface temperature (SST) in the tropical Atlantic on seasonal to decadal time scales, by enhancing and analysing equatorial and coastal observations, and by numerical experimentation.
In particular, PREFACE has drastically improved the ocean and climate observing system in the eastern tropical Atlantic including near-coastal regions, allowing a continuous monitoring of tropical Atlantic variability and climate change. Two new ocean-atmosphere time series stations were established within PREFACE to complement continuous measurements along the equatorial and coastal regions; these were crucial for understanding tropical air-sea interactions and for detecting climate signals propagating from the equatorial Atlantic towards the biologically productive eastern boundary upwelling regions. The first one enhances the deep ocean PIRATA (Prediction and Research Moored Array in the Tropical Atlantic) network in the southeastern tropical Atlantic, and the second one offers similar measurements within the Senegal-Mauritania sector of the Canary coastal upwelling. Due to close cooperation and intense capacity strengthening activities within PREFACE, historical datasets from near-coastal regions off Africa, particularly including hydrographic data from the Nansen and EAF-Nansen programmes, could be made available for scientific analysis thereby drastically enhancing the climate record of the tropical Atlantic. Numerous dedicated shipboard and glider observations carried out within PREFACE complemented the available database.
Medium to very-high resolution forced ocean model simulations complemented observations. Their spatial domain covered the tropical and subtropical Atlantic with high resolution, and very high resolution in the Benguela, Senegal-Mauritania, and northern Gulf of Guinea upwelling regions. These simulations were validated by comparison to available large-scale observations and PREFACE data from moorings and cruises. Sensitivity studies were performed with the models to assess the importance of various processes in the SST and sea surface salinity climatology and variability, and to determine strategies to reduce ocean model biases.
Improved mixed layer heat and freshwater budgets
Observations and models delivered an improved understanding of the seasonal and interannual mixed-layer temperature and salt balances in the equatorial and eastern boundary upwelling regions of the tropical Atlantic. An important project output was an improved mixed-layer temperature and salinity climatology dataset for the tropical Atlantic, PREFCLIM. Unpublished data from African partners were incorporated to close data gaps in coastal regions. On seasonal as well as interannual time scales, vertical mixing within the thermocline is a dominant contributor to the development of the Atlantic Cold Tongue. On seasonal time scales, vertical mixing and horizontal advection dominate the salinity variability whereas on interannual time scales it is dominated by surface freshwater fluxes and winds . For the southeastern tropical Atlantic, we provided a revised understanding of the critical role of advective heat transport by eddies for the mixed layer heat and fresh water balances. In addition, alongshore advection appears to lead to a clear correlation between SST and temperature anomalies in the upper thermocline on interannual to decadal timescales in the Angola-Benguela region . Beyond this we have provided new estimates of the impact of wind-induced near inertial wave energy and chlorophyll concentration on the mixed layer heat content variability.
Ocean circulation, and the equatorial and coastal waves
PREFACE delivered key new findings in three areas. Firstly, our measurements identified pronounced semi-annual, annual and interannual variability of the equatorial zonal currents. We showed that resonant equatorial basin modes can explain much of the enhanced variability on these timescales, and that pronounced deep intraseasonal variability of equatorial zonal currents supplies energy to maintain resonant equatorial basin modes on interannual time scales. Furthermore, we identified the dynamics of deep ocean jets at the equator. Secondly, we quantified for the first time the seasonal cycle of the southward Angola Current, using mooring data from 11°S off Angola together with shipboard hydrographic data (PREFACE and EAF Nansen programme surveys). This allowed a first independent evaluation of different ocean reanalysis products. Lastly, we provided new insights into the causes of Benguela events, which strongly impact fisheries. While most of the anomalous interannual SST events along the southwestern African coast are remotely forced by equatorial waves as could be particularly evidenced for the Benguela Niño in 2011, the warm event in January/February 2016 could not be related to equatorial forcing. Instead local forcing played a dominant role.
Long-term directions to improve ocean and climate modelling
PREFACE assessed the importance of various processes for the realistic simulation of SST and ocean circulation in eastern boundary upwelling regions of the tropical Atlantic. We considered three frequently misrepresented processes: near inertial wave induced mixing, upwelling and poleward heat transports in relation to wind stress patterns, and absorption of solar radiation by chlorophyll. In addition, we investigated the benefit of improving spatial resolution of ocean models and the realism of air-sea forcings. Based on comprehensive experimentation with ocean models we make the following general recommendations for improving ocean model simulations in eastern boundary upwelling regions of the tropical Atlantic: (1) to force the ocean models with a wind stress product having a climatological seasonal mean distribution close to that of scaterometer derived winds; (2) to validate the shortwave radiative forcing against in-situ observations; (3) to use a horizontal resolution in first 500km offshore as close as possible to the Rossby Radius, although 1/10° appear to provide satisfying results; (4) to include the effect of primary production on turbidity of the waters, via a realistic chlorophyll distribution in the equation of penetrating solar radiation; and (5) to consider the importance of transmitting high frequency wind energy into the ocean.
Disentangling the causes of tropical Atlantic climate model biases
PREFACE has deepened the understanding of the ubiquitous model errors (biases) in the tropical Atlantic climatology. The climate model biases are characterised by to warm SST in the eastern-equatorial Atlantic and the southeastern tropical Atlantic. The SST errors are associated with corresponding large precipitation and surface wind errors. We performed coordinate analysis and multi-model experiments to disentangle the causes of tropical Atlantic model biases. A particular novel aspect was to perform idealised prediction experiments. The analysis of coupled model mean biases lead to the following conclusions:
(1) The equatorial warm SST bias, which is a seasonal bias strongly tied to the annual cycle. The spring cooling is largely underestimated in coupled models, leading to strong summer SST biases. There are some indications that errors in the mean state (like a too deep mixed layer in the eastern equatorial Atlantic in early spring) do not favour the cold tongue cooling processes. Improving the surface wind representation in atmospheric models helps in improving the mean state and favours the cold tongue development. The atmospheric wind biases are present in atmospheric models forced by observed SST. This indicates that deficiencies in atmospheric models are the root cause of the warm equatorial Atlantic SST bias, but coupled processes amplify these errors. The surface wind biases are linked to surface pressure gradients errors and in turn to deficiencies in representing the convection. The wind stress biases are thus related to ITCZ errors.
(2) The southeastern warm SST bias, which develops more progressively and independently of the seasonal cycle. In this stratocumulus region, the radiative biases play an important role in explaining large scale SST biases, but other elements also play a role: coastal atmospheric wind jets, vertical ocean mixing, and coastal currents representation. All these processes tend to be better represented when increasing resolution both in atmosphere and in ocean models even if the improvement is model dependent.
(3) Remote errors also contribute to drive tropical Atlantic model biases. In particular, correcting the radiatively caused warm SST biases over the Southern Ocean can modify the stratocumulus-SST feedback and reduce the SST warm bias over the tropical stratocumulus regions. Errors in the simulation of the South Atlantic Anticyclone also drive SST biases in the Angola- Benguela region. In addition, the strength of the simulated Atlantic meridional overturning circulation seems to explain the SST bias pattern in some models.
Improved climate models for the tropical Atlantic
PREFACE proposed three approaches to reduce climate model errors in the equatorial Atlantic. Firstly, we developed a model with higher atmospheric resolution to realistically simulate the tropical Atlantic climatology. Accurate simulation of the surface winds required increased vertical atmospheric model resolution in the lower troposphere. Secondly, we developed new parameterisation for ocean mixing, and one of these accounted for near inertial wave mixing. Lastly, we developed a novel anomaly coupled approach that statistically corrects ocean-atmospheric fluxes.
Reducing mean state errors improves simulated variability
Comprehensive multimodel analysis, and experiments with the PREFACE improved climate models and ocean models have shown that reducing mean state errors improves the simulation of tropical Atlantic climate variability. State-of-the-art climate models show key deficiencies in simulating the two dominant patterns of tropical Atlantic variability – the Atlantic Niño and the Atlantic Meridional Mode – although they capture their strength and spatial structure reasonably well. Regarding the Atlantic Niño, biased models tend to simulate variability that peaks in autumn to winter rather than in boreal summer, as observed. Critically biased models do not reproduce the link between surface and subsurface ocean temperatures, and thus they underestimate this element of the Bjerknes positive feedback. This is because in biased models upwelling is too weak and the thermocline too deep in the eastern equatorial Atlantic. Reducing model biases leads to improvement in the Bjerknes positive feedback and thus to the simulation of Atlantic Niño variability. Regarding the Atlantic Merdional Mode, the Wind Evapourative SST feedback that underlies this variability is too strong in bias coupled models compared to observational analysis. This suggest that in coupled models, a greater portion of SST tendency is controlled by atmospheric heat fluxes than that in the observations; this may suggest that ocean processes are underestimated in biased models.
Relation between Atlantic and Pacific climate variability
We have shown the importance of the tropical Atlantic in driving Pacific El Niño variability, and better understood the mechanisms. We have also assessed the ability of climate models to simulate the relation between tropical Atlantic and Pacific, and shown its potential to enhance seasonal prediction of the El Niño. The following is a brief summary of our key findings achieved through multi-model analysis and by performing sensitivity experiments.
Observations show that Atlantic Niño’s (Niña’s) often precede Pacific La Niña’s (El Niño’s) by about six months. It is now well established that this is because Atlantic Niño variations affect the Pacific trade winds and thus the development of El Niño. This occurs when the Atlantic Inter-tropical Convergence Zone (ITCZ) is anomalously close to the equator, causing a shallower mean thermocline and enhanced eastern equatorial Atlantic SST variability. The larger amplitude and greater westward extent of Atlantic Niño SST variations leads to a larger impact on trade winds in the Pacific, and thus an impact on El Niño. Atlantic ocean multi-decadal variability and global warming can both shift the position of the Atlantic ITCZ. Together these two explain the strengthening of the Atlantic’s influence on the Pacific since the 1970’s that have contributed to stronger El Niño variability.
Climate model biases affect the simulation of the relation between Atlantic and Pacific climate variability. In current climate models the ITCZ is too far south in the Atlantic, and they underestimate dynamical ocean-atmosphere interaction in this region (see above). As a result models underestimate the impact of the Atlantic Niño on the Pacific. Lastly, we have shown that when the connection between Atlantic and Pacific climate variability is properly represented, it can enhance seasonal prediction of El Niño.
Tropical Atlantic climate variability
PREFACE has improved understanding of tropical Atlantic variability and predictability, through analysis of observations and model simulations. The Atlantic Niño was the main focus. We showed that variance in boreal summer SST in the eastern equatorial Atlantic is controlled by ocean dynamics, but thermodynamic processes contribute. Equatorial waves triggered by atmospheric forcing are the dominant process controlling SST warming (cooling) in the central and eastern equatorial during Atlantic Niño (Niña) events. The decay of these events in July is associated with anomalous horizontal advection. The variability of the South Atlantic Anticyclone through its impacts on equatorial winds can explain up to 50% of the equatorial Atlantic SST variance. Another particularly interesting finding is that equatorial SST variability and surface zonal velocity within the North Equatorial Counter Current is influenced by Equatorial Deep Jets (stacked zonal jets along the equator which steadily propagate downwards over time with a time scale of about 4.5 years). Equatorial Deep Jets are driven by deep intraseasonal variability and are associated with upward energy flux. Unfortunately, most current state-of-the-art climate models do not reproduce properly dynamical ocean-atmosphere variability (see above) and thus likely underestimate predictability.
We also better understood long-term climatic change in the tropical Atlantic. We showed for the first time that equatorial Atlantic SST shows decadal variability that is related to the southern subtropics and to the strength of the St. Helena subtropical anti-cyclone. On longer timescales, variability generated internal to the climate system has an important influence on the observed multi-decadal trends in the tropical Atlantic. The amplitude of the observed trend is much bigger than the multi-model estimated trend caused by anthropogenic global warming. Thus, the recent trends in SST over the tropical Atlantic that show stronger warming over Mauritania and Namibia coasts than over South Benguela upwelling region are probably not caused by global warming.
Rainfall variability over adjacent continents
PREFACE has also devoted efforts to better understand the seasonal to interannual variability of rainfall over the continental areas adjacent to the Atlantic. The structure of large- scale circulation over central Africa and rainfall variability over Southern Africa, Ethiopia, West Africa and Brazil have been further analysed and related to SST anomalies over different basins with a special focus on the Pacific El Niño and the stationarity of the links found.
The Pacific El Niño is the main source of variability at interannual timescales. It has been shown to affect rainfall variability in the continental areas adjacent to the Atlantic. Its positive phase causes negative rainfall anomalies over Southern Africa, Ethiopia, Northeast Brazil and West Africa. However, the links of rainfall variability and ocean SST anomalies are not always stationary in time. The impact of the Atlantic Niño on Sahel rainfall changed in the 1970s: before this decade the Atlantic Niño led to negative anomalies of rainfall over the Sahel caused by a more southern location of the Monsoon, while, after the 1970s, the concomitant presence of a Pacific La Niña (see above) counteracted this effect leading to no significant rainfall anomalies over the Sahel. The Pacific El Niño events also had stronger influence on Sahel rainfall after the 1970s. In addition, a link between the Mediterranean and Sahel rainfall was active in recent decades, as well as in the first half of the 20th Century. The influence of Atlantic and Pacific El Niño events on North Eastern South American rainfall also changed after the 1970s, and the combined effect of the Atlantic Niño and the Pacific La Niña amplified the rainfall response. Atlantic multidecadal variability was shown as the leading factor in modulating the link between SST and continental rainfall. These results are important for understanding the predictability of continental climate.
We also improved understanding of the longer-term changes in West African rainfall. In particular, we emphasize the importance of Atlantic multi-decadal variability in driving Sahel rainfall variability. Our analysis of climate model simulations further indicates that global warming induced changes in continental rainfall are very uncertain, and during the historical period not detectable.
Advanced statistical methods
PREFACE has developed advanced statistical methods and bias correction techniques to improve climate predictions. A key achievement of the project was the development of Bayesian hierarchal model framework. It was used to construct different statistical models to quantify systematic climate model errors. The statistical models are (a) Bayesian (to incorporate information from different sources, including expert knowledge) and (b) hierarchical (i.e. it treats errors at data, process and parameter level separately and in a transparent way). Three different models were constructed to focus on a purely spatial, a purely temporal, or a spatio-temporal assessment. They all build on the so-called state-space approach, for which unobservable state variables are used to directly model (statistically) the process of interest generating the observed data. The explicit modelling of the noise at the process level allows for better quantification of uncertainty. The power of the three approaches was demonstrated by (1) identifying forecasts drift and diagnosing its various causes using the purely temporal model, (2) by robustly determining the common tropical Atlantic bias in multi-model ensemble using the purely spatial model; and (3) using the spatio-temporal model to determine time varying model biases. The model codes have been made freely available.
In addition, a statistical prediction model that accounts for non-stationary teleconnections was developed based on Maximum Covariance Analysis. The model was used to predict tropical Pacific SST and Sahel rainfall. The approach was also extended to demonstrate the potential to predict variations in West African fish stock, based on the impact of El Niño on the region.
Improved tropical Atlantic climate predictions
PREFACE has lead to enhanced climate modelling and prediction capabilities, on seasonal to centennial time scales for the tropical Atlantic. Such improved predictions can have major socio-economic benefits, including for ecosystem and fisheries management.
We have demonstrated skilful prediction of equatorial Atlantic SST, and shown that reducing climate models biases improves skill in predicting Atlantic Niño events and their impact on continental climate and the Pacific. Several current operational seasonal climate prediction systems are able to predict Atlantic Niño variations in boreal summer from the beginning May. These models also show skill in predicting rainfall in limited regions affected by the West African monsoon. However, the majority of models are not skilful in predicting eastern equatorial Atlantic, even one month in advance. Analysis of the operational prediction systems shows that there is a relation between model mean state error and prediction skill, with models with reduced SST bias better predicting SST and rainfall. The impact of bias reduction on skill was confirmed using PREFACE models with reduced model error. In particular, we showed the anomaly coupling approach leads to significant skill in predicting equatorial Atlantic SST and that this is because the mechanisms for the Atlantic Niño variability are better represented (see above). This important result highlights the need to improve the mean state of coupled models in the Tropical Atlantic in order to enhance the predictions over the region.
Sahel rainfall remain at similar levels in the near future
Decadal variations in Sahel rainfall can now be predicted. The skill is found to depend on how well Atlantic multidecadal variability is predicted, but also on the response to global warming. Therefore models initialised with concurrent ocean observations are more skilful in predicting Sahel rainfall than historical forced simulations. Furthermore, the skill in predicting decadal variations in Atlantic SST depends on the initialisation method, but at lead-times longer than a few years the impact of global warming also becomes important. Our analysis indicates that there will be little change in Sahel rainfall for the near future (2016-2019) with respect to the previous four-year period (2011-2014).
The equatorial Atlantic Cold Tongue will weaken under global warming.
We have assessed how long-term climate change will impact tropical Atlantic climate variability, and we have shown that model biases introduce uncertainty in the future evolution of the tropical Atlantic and the West African Monsoon. This was achieved by performing climate projections with multiple climate models and with several approaches for model improvement: anomaly coupling, updated model configurations, and new parameterisation for vertical mixing (see above). We showed that improving the simulation of the tropical Atlantic climatology increases the sensitivity of the eastern equatorial Atlantic to the global warming. In particular, a model that realistically simulates the Atlantic Cold Tongue shows greater warming of the SST in the Gulf of Guinea than a model that poorly simulates the Atlantic Cold Tongue. Correspondingly, the tropical Atlantic ITCZ shifts more southward under the global warming in a less biased model. The consistency among our model results and the underlying physical mechanism suggest that the equatorial Atlantic Cold Tongue will weaken under global warming. In addition, we have shown that long-term global warming can influence the dominant patterns of Atlantic variability and the interaction between tropical ocean basins. However, our model experiments do not agree on the extent to which global warming will impact variability in the tropical Atlantic.
Disentangling environmental and anthropogenic pressures
PREFACE has contributed to disentangle environmental and anthropogenic pressures on pelagic fish stocks in the tropical Atlantic. In particular, we have increased understanding of the interrelation among marine ecosystems, climate variability and change, and fisheries, through exploratory analysis of historical and newly collected data. More than two decades of historical data allowed long-term changes and short-term variations to be studied. These include fish monitoring survey data acquired within the Nansen Programme executed by the Food and Agricultural Organization of the United Nations and funded by the Norwegian government. PREFACE cruises contributed to close data gaps and collect unique data, which included Tuna tagging. All these data were collected taking advantage of both local and European expertise, and contributed to training of scientists. With these data we have analysed three different ecosystem levels: top predators, small pelagics, and micronekton.
In terms of top predators, the analysis of Yellowfin tuna (YFT) catches in Cape Verde revealed the importance of climatic factors and fish stock to model local catch dynamics. There is a strong effect of subtropical wind stirring with a subsequent cooling of the surface layer that leads to an increase in abundance. This is evidenced by negative correlations to North Tropical Atlantic sea surface temperature and the North Atlantic Oscillation, which is the dominant mode of large-scale atmospheric variability. A YFT catch model using anomaly corrected climate projection data reveals that improved governance is essential to cope with negative impacts of climate change. In terms of prey fish dynamics, the first long-term and large-scale comparison of mesopelagic fish data in terms of biomass size spectra reveals significant changes in terms of trophic efficiency in oceanic food webs. Together, this highlights the impact of climate on local fisheries with important implications for fisheries economics.
In terms of small pelagics, the analysis of historic time series revealed the poleward expansion of tropical species during the last 25 years on both hemispheres. It also revealed differences in the causes of year-to-year variations in stock size and distribution.
Off northwest Africa, the northern range of the Sardinella aurita stock has shifted by two degrees to the north, to regions inhabited almost exclusively by subtropical sardines during the 1990s. For instance, the distribution of the commercially important S. aurita has shifted a few hundreds of kilometres northwards. The observed distributional shift coincided with the systematic warming characterizing the Canary Current region during the same period. While analysis of the short-term variations indicates that coastal wind and wind-induced upwelling is the key driver of fluctuations in biomass of small pelagic fish, rather than only ocean temperature. In Senegalese waters abundance of sardinella is significantly correlated to wind strength during the upwelling season whereas interannual fluctuations in the proportion of S. maderensis to S. aurita in the catches depends on how early in the year is the onset and how long is duration period of the upwelling season. An early or late upwelling season favours the sedentary S. maderensis or migratory S. aurita, respectively.
Off southwest Africa, the adult sardinella exhibited a strong interannual variation in the seasonal migration cycle, synchronised with the strength of the interannual equatorial climatic events. In the southern range of sardinella distributions a change in fish age structure was observed from adult migratory fish during the 1990s to an increasing proportion of locally born juveniles during the current decade. Overall the principal findings for the southern region include: (1) identifying patterns of correspondence between the strength of seasonal coastally trapped (Kelvin) waves and seasonal migration of sardinella along the Angolan coast, (2) associating the two major sardinella biomass growth periods in Angolan waters, 2006-2007 and 2012-2013, with the interannual climatic scenarios induce by equatorial forcing and (3) identifying the gradual expansion of coastal nursery areas of S. maderensis and S. aurita from the Gabon-Congo coastal regions during the 1990s towards the remotely forced Angolan upwelling area (6°-13°S) during the current decade.
The analysis of micronekton revealed that the three LMEs are highly variable systems, with the Canary Current LME and the Benguela Current LME being particularly impacted by global warming, especially in their lower latitudes. From 1995 to 2015 in the Canary Current LME (CCLME) there was a significant deepening of the micronekton distribution and an increase of the diel vertical migration amplitude. The interannual variability is not significant, showing that the underlying environmental forcing is associated with relatively stable processes. Sea surface temperature was shown to have a minor influence in the north CCLME, but a pronounced effect in the south CCLME. Considering the relative importance of oceanographic factors, stronger ecosystem perturbations are expected in Benguela Current LME (BCLME) than in the CCLME. Overall, oceanographic factors were found to have a significant influence in all LMEs. Therefore, global warming is expected to cause important changes in the three LMEs and thus in the fisheries sector.
Climate change effects on small-scale fisheries and coastal communities, and implications for management
PREFACE has better understood climate change effects on small-scale fisheries and coastal communities, and assessed uncertainties and implications for management. This was achieved by (1) developing ecological-economic models, (2) making an impact assessment by coupling these models to PREFACE climate change projections, and (3) by extensive surveys of coastal fishing communities to acquire information for developing the economic models and providing policy relevant advice.
Single species coupled bio-economic models were developed to understand the threats climate change poses to fisheries with focus on small-scale fisheries. The models simulate the consequences of future climate change and allow a comparison of optimal and realized harvest, and their impact on consumer and producer welfare. PREFACE acquired knowledge on the impacts of climate on fish stock was used to develop the biological component. The economic component of the model was based on available literature and collected through surveys (see below). Due to their importance for local communities and data availability, we focused on the artisanal purse seine fishery for Sardinella aurita in the four major Senegalese fishing regions and the artisanal hand line fishery for YFT in three island regions of the Cabo Verde archipelago.
We selected artisanal fisheries in Senegal for detailed impact assessment, because the effects of climate change on them may be severe, as the dependence on fish as a protein source and on fishing income is strong. We drove a bio-economic model with future climate change projections from two different PREFACE models and different fuel subsidy scenarios. The main effect of climate change indeed is a distributive one. Biomass becomes more vulnerable to Senegalese purse seines over a certain time horizon, because the fish spend more time in the close-to-shore area. In the long term, stock survival requires a stricter response to over-fishing. The cancellation of fuel subsidies entails an increase in total benefits and thus, not only are fuel subsidies expensive for taxpayers, they also reduce welfare in total. All runs reveal that the fisheries contribution to income security will in the future be more important than the food security impact. Our results are informative for policy makers, as they shed light on the effect of climate change and counteracting measures while incorporating endogenous adaptation of actors. We also explicitly show that policy choices dominate socio-economic model outcomes rather than climate model uncertainty.
The socio-economic surveys conducted in Senegal and Cabo Verde shows that impacts of weather changes on the resource seem evident. For Senegal, this is dominantly visible in the perceptions on rainfall pattern effects, while for Cabo Verde, wind changes may prove more important, on the resource as well as on fishing effort in the artisanal sector. The data also shows the sensitivity and adaptive capacity of the fishing sector in the two countries. While in Senegal specialization is at the individual level but on a rather diverse range of target species and gears used, the Cabo Verde fisheries depend to a dangerous degree on only six species, and most importantly tunas. This is true on individual, fishing sector and total-fishery level. In that regard, while only individual sensitivity is high in Senegal and the artisanal sector itself is likely too diverse to be hit by singular species changes, the Cabo Verde fishery is very sensitive to certain species changes.
The economic analysis of the Senegalese fisheries on small pelagic species however, showed that this sector is indeed vulnerable to changes in catchability affected by changes in sea surface temperature. A sea surface temperature shock during summer can increase the catchability by one third. However, long-term effects can be negative due to decrease in population biomass. In addition, economic drivers, like fuel prices also have a significant negative effect on the profitability of the purse seine and encircling net fisheries. For the Senegalese purse seine fishery, operating costs have increased by 90% over the last twenty years, mainly due to an increase in fuel prices.
Enhanced cooperation between European and African researchers
PREFACE has increased cooperation between European and West African partner institutions leading to enhancement of research capacities in the fields of ocean physics, ecosystem and fisheries as well as climate research in West Africa. The close cooperation between different European and African partner institutions resulted in common experiments at sea, data sharing and analysis, publishing of historical observations, as well as application of ocean modelling on regional scale. For example, it led to a new seasonal climatology for the heat and freshwater budget in the mixed layer. The climatology is greatly improved in coastal regions compared to previous climatologies. This is because of the incorporation of unpublished hydrographic data from the EAF Nansen program and other programs of the PREFACE partners that was only possible through close cooperation. Another great success was the dissemination of hydrographic and ship-board velocity data collected during biannual cruises carried out with the RV Dr. Fridtjof Nansen since 1984 within the FAO-NORAD EAF Nansen Programme. Our Angolan partner led this publication of this data.
PREFACE has also helped to train a new generation of researchers, including those from the region. In total there were of 21 PhD theses (15 from African students) and 21 Master theses (18 from Africa students). Its summer schools (two in Africa), targeted workshops, and interdisciplinary meetings have provided education to continue important research on the tropical Atlantic climate and its impacts. The joint organisation of two project meetings with annual tropical Atlantic variability conferences helped integrate young researchers into the international research community. Lastly, we have helped to ensure the long-term sustainability of the tropical Atlantic observing system, by increasing awareness and developing required competence among regional players.
Potential Impact:
Impacts and exploitation
1. Better monitoring of the eastern boundary upwelling
(Deliverables D2.3 D2.8 D2.9 D3.1 D4.4 D12.1 D12.2 D12.3 D12.4 D14.3 and D14.4)
PREFACE has drastically improved the ocean and climate observing system in the eastern Atlantic, allowing a continuous monitoring of tropical Atlantic variability and climate change. PREFACE has installed key moorings in the southern and northern coastal waveguides, extending the PIRATA (Prediction and Research Moored Array in the Atlantic, www.pmel.noaa.gov/pirata/) network toward the eastern tropical South Atlantic, particularly with the new buoy, “Kizomba”, located strategically in an area problematic to ocean-atmosphere coupled climate models, offshore of the Congo River plume and upstream of the Angola Current. Exemplary also is the deployment, in association with several laboratories across West Africa, of an oceanographic and meteorological measuring buoy, Melax, in the heart of the Senegal-Mauritania upwelling.
Along with the increase in new ocean observations, previously unreleased historical oceanographic data acquired in the framework of the FAO EAF Nansen programme is being comprehensively exploited thanks to the close North-South collaboration in PREFACE. This represents the establishment of time series long enough to monitor climate change in the African coastal regions of the Atlantic. This provides valuable input for future ocean monitoring by the EAF Nansen Programme, an initiative to support the implementation of the ecosystem approach in the management of marine fisheries in developing countries, among other programs.
PREFACE has contributed knowledge on the optimal design of the future network to the Tropical Atlantic Observing System review. It is also helping to ensure the long-term sustainability of the network, by making data freely available on international data archives, by increasing awareness and by developing required competence among regional players. The better observing system will be exploited by regional institutes and centres responsible for monitoring and managing their marine resources, and by international climate prediction centres for improving their global models and predictions.
2. New process understanding and parameterisations
(Deliverables D3.2 D3.3 D6.3)
PREFACE has achieved an understanding of the role of near-inertial wave mixing and its role in ocean temperature and momentum budgets, and has developed and tested parameterisation for climate models. PREFACE has also worked on 1-column models to test vertical mixing schemes. We have understood the processes controlling deep equatorial jets and their role in exciting equatorial basin modes and explaining intraseasonal, seasonal, and interannual variability in the equatorial currents and SST. We have improved understanding of the modelling of these processes. We also achieved a deeper understanding of ocean atmosphere interaction. Modelling centres, through the PREFACE network and openly available publications, will exploit these results to improve ocean and climate models.
3. Improved understanding of tropical Atlantic climate and its global impacts
(Deliverables D2.5 D2.8 D2.9 D3.1 D3.4; D4.1-4.3; D5.1 D7.2 D8.2 D9.1 and D9.2)
PREFACE has achieved a better understanding of the mechanisms and predictability of tropical Atlantic climate, and of its global impacts. This is especially the case for both eastern boundary upwelling African-Atlantic regions and the Gulf of Guinea. The following are a few showcase results: We now better understand the importance of remote versus local processes in driving Benguela Niño events, as well as the role of dynamical versus thermodynamical ocean-atmosphere interaction for Atlantic Niño variability. The prominence of the tropical Atlantic in the climate system has been demonstrated. For example, it is now clear that the Atlantic Niño strongly influences the El Niño Southern Oscillation. We have also shown the importance of internal generated climate variability for long-term changes in the tropical Atlantic, and highlighted the uncertainties related to long-term climate change.
International modelling, operational prediction centres, and impacts researchers working on the tropical Atlantic can exploit our findings on the mechanisms for tropical Atlantic variability in order to improve their modelling and prediction capabilities, to provide confidence in predictions and projections of climate, and to develop impact models.
4. Improved climate models
(Deliverables D2.5 D2.8 D2.9 D5.2; D6.1-6.3; D7.1 D7.2 D8.1 and D8.2)
PREFACE has led to enhanced climate modelling and prediction capabilities, on seasonal to centennial time scales. Our targeted ocean model experiments and their comparison with observations provide detailed process-based understanding and long-term directions to improve ocean and climate modelling and prediction capabilities. Our multi-model analysis and coordinated model experiments greatly contributed to the detailed understanding of the causes of model systematic errors and their impact on simulated variability and prediction skill. This detailed understanding has lead to the development of new improved models and alternate approaches to reduce model error.
Global climate modelling and prediction centres will exploit our new approaches to reduce model error and the development of new parameterisations for ocean mixing to improve their modelling systems. Our novel tools to diagnose model errors and their consequences will be exploited by the wider scientific community as they provide new directions for using multi-model ensembles.
5. Enhanced climate prediction
(Deliverables D10.1-10.3; D11.1 and D11.2)
PREFACE has demonstrated the potential to skilfully predict climate on seasonal-to-decadal timescales in the tropical Atlantic, and better understood uncertainties in climate change projections. In particular, we have used improved models and alternate model configurations to reduce model errors and to show improved prediction skill and assess model uncertainties. In addition, PREFACE has developed advanced statistical methods and bias correction techniques to improve climate predictions. We have demonstrated skilful prediction of equatorial Atlantic SST, rainfall in some regions of the West African Monsoon, and decadal shifts in Sahel rainfall.
The improved climate prediction approaches will be exploited by operational centres, and will in the longer term lead to improved operational regional and global predictions. Climate services providers and users can further exploit these. Thus, such improved predictions should have major socio-economic benefits, including sustainable land use and ecosystem-based fisheries management.
6. Deeper understanding of the function of marine ecosystem
(Deliverables D2.6-2.9; and D12.1-D12.4)
PREFACE activities have led to a deeper understanding of the function of marine ecosystem so that climate driven shifts can be better predicted. The following are showcase results. We have shown the poleward expansion of tropical species during the last 25 years on both hemispheres, and related this to environmental changes. The warming of the South Atlantic during the last three decades was associated with threefold increase and southward shift in Sardinella biomass. While the warming off northwest Africa caused the northern range of the Sardinella aurita stock to shift by two degrees to the north, to regions inhabited almost exclusively by subtropical sardines during the 1990s. We have also understood year-to-year variations in Sardinella to Benguela Niño variability, and shifts in tuna distributions to large-scale atmospheric wind and oceanic temperature patterns.
The development of an open-source software, Matecho, for analysing acoustic data on different ecosystem trophic layers (such as zooplankton and small pelagic fish) will enable not only partners in PREFACE to further study and understand ecosystem changes in their marine waters but also other countries in the region. Other new descriptors of micronekton (an essential component of marine food webs) that can serve as a monitoring tool for the health of ecosystems, are being refined and soon to be made available through publishing. Meetings and small workshops are being carried out to further disseminate the existence and use of this software and new descriptors. Importantly, these will enable further comparative analyses for a better monitoring of ecosystems and sustainable management.
Regional fisheries managers and policy makers to ensure long-term sustainability of fish stock can exploit these results. Researchers aiming to develop ecosystem models and predictions can exploit these results. The potential to predict changes in small pelagic fish distributions by coupling bio-climatic models to improved monitoring and climate predictions can produce results to be further exploited by fisheries managers and policy makers.
Another PREFACE finding points to a response pattern in the behaviour of micronekton to long-term climate changes in the water column, which could have an implication on the ocean carbon pump. These results however are still preliminary and currently under scrutiny before international publication.
7. The sustainable management of West African fisheries
(Deliverables D2.6-2.9; D12.1-12.4; D13.1-13.3)
PREFACE can provide science-based support to policy-makers for the sustainable management of West African fisheries in the context of climate change. Knowledge gained through surveys of local fishing communities, better understanding of climatic and anthropogenic influences on marine ecosystems, and improved predictions together lead to improved ecological-economic models and predictions for the region. The following are some key results. Detailed impact assessment for artisanal fisheries in Senegal showed that Sardinella aurita becomes more vulnerable to Senegalese purse seines over a certain time horizon, and the long-term stock survival will require a stricter response to overfishing. Bio-economic modelling further shows that cancelling fuels subsidies for fisheries benefits the entire community. We also highlight that Cape Verde fishery is very sensitive to climate driven species changes, because it depends to a dangerous degree on only six species, and most importantly, tuna. Senegalese fisheries is less specialized and likely more robust to climate driven ecosystem change. Researchers will exploit these results to develop impact models, and also have the tools now to further extend our survey approach, with application in other countries. The application of these models can then benefit fisheries managers and policy makers, who can also further exploit our survey results.
PREFACE reemphasizes the known but seemingly often ignored evidence – albeit because challenging – that sustainable fisheries management requires that fish stock assessments include the systematic monitoring of environmental data, and that management be at a level not restricted by national borders but rather by the migratory behaviour of fish stocks. This is particularly relevant in the face of climate change, as has been shown with the sardinella stocks studied in PREFACE. It is also important that regional fisheries management organizations be given the tools and resources to assess more than a single-species, and in this light, it is crucial that dialogue between involved actors be increased. Importantly, PREFACE stresses that the enhanced dialogue between fishery commissions, scientists, fisherfolks, and higher-level decision bodies can only be efficient if each party is sensitised to the others technical knowledge, that is, there is a need for knowledge brokers. Hopefully, some of the junior scientists educated within PREFACE will become such knowledge-brokers in their local regions.
8. Strengthened capacity in oceanography, climate, marine ecosystem, and socio-economic research
(Deliverables D2.3-2.5 D2.8 D4.4 D10.1-10.3 D12.1-D12.4 D13.1-13.3)
By strengthening existing and fostering new collaborations, PREFACE showcases and further corroborates the significant benefits brought by enhanced cooperation between European and African researchers working on tropical Atlantic climate and its impacts. The project dedicated a substantial amount of resources to capacity strengthening and training of students, postdoctoral fellows and early career scientists of Europe and Africa by means of dedicated bursaries for PhD and Master students, fellowships for postdoctoral fellows, organisation of summer schools and workshops and support for participation to oceanographic surveys (R/V Meteor 2014-2017, PIRATA campaigns 2014-2018, and R/V Thalassa off Senegal in 2015) and international scientific conferences. More specifically, a total of 21 Masters and 21 PhD theses were directly recruited by the project or benefitted from, and contributed to, the project through collaboration. PREFACE also supported, through teaching and supervision, the continuation of the Masters programme: Regional Master 2 SOAC-OA (Sciences de l’Océan, de l’Atmosphère et du Climat : parcours “Océanographie et Applications”) in West Africa, ICMPA/UAC-IRD-UPS-UFPE, Cotonou, Benin. The capacity strengthening activities with ICPC partners were always elaborated to suit the interests and demands of the ICPC partner institution, providing a general scientific education with focus on the objectives inherent to the involved research programs. Additionally, PREFACE has produced a series of open-source software and models (e.g. VEROS, S4CAST v2.0 Matecho, the DLM in described in D10.3) contributing to provide the education and tools required to continue important research on the tropical Atlantic climate and its impacts. PREFACE can thereby be considered as one of many useful actions and initiatives paving the way for the implementation of the Belém Statement on Atlantic Research and Innovation Cooperation and for improved coordination and alignment of programmes/initiatives and projects between South and North Atlantic regions and with the EU and its Member States.
9. The development of climate services
(Deliverables D2.2 D2.3 D2.5-D2.9; D11.1 D11.2; D12.1-D12.4; and D13.1-D13.3)
PREFACE has brought together an interdisciplinary team of African and European scientists to achieve a step-wise improvement in our understanding and modelling of the tropical Atlantic climate and its impacts. Climate models can now provide reliable climate simulations and predictions in the region. When combined with knowledge of environmental-ecosystem interaction, this information can be used to predict changes in the state of marine ecosystems at timescales relevant for anticipatory and more effective policy or management decisions. Furthermore, we have contributed to strengthen capacity by training a new generation of scientists capable of realising and further developing climate services. This includes researchers with expertise in communicating with decision and policy makers. Together these achievements provide the technical and scientific basis for the development of climate services in the region. We recommend to promote and encourage the explotation of the climate models information to evaluate climate changes and to perform useful and reliable predictions of the living marine resources at timescales useful for decision making. Exploitation of climate services by the region will help ensure the sustainable development.
10. Sustainable development
As can be concluded from the above, PREFACE contributes to the United Nations’ 2015 sustainable development agenda to end poverty, protect the planet and ensure prosperity for all. Particularly, PREFACE contributes to the Sustainable Development Goals 13, to “Take correct action to combat climate change” and 14, to “Conserve and sustainably use the oceans, seas and marine resources”, particularly by bringing us closer to reaching targets 14.7 14.4 14.6 14.5 14.2 and 14.3.
Main dissemination activities
(contributing to the exploitation of results and achievement of impacts)
The project website is designed to inform the global community about PREFACE, who we are, our activities and latest results, related programmes and projects, events, and employment and training opportunities. It also contains basic publicly available outreach material, such as the project factsheets, translated into French and Portuguese for a wider outreach on the African continent. Additionally, translated PREFACE outreach material has been prepared and provided upon request, such as for IRD’s PI involvement in political workshops (Fisheries or Environmental Ministries and regional Environmental and/or Fisheries monitoring commissions across the CCLME and GCLME) or after a request from the Norwegian embassy in Angola, in connection with visit of R/V Meteor in Luanda and high-level meeting organised upon the occasion (Milestone 16, Week 4 cruise report, available on website). The website thus aims to not only raise interest in PREFACE, but also to raise awareness in such topics as climate change, improving climate models, the need for a long-term tropical Atlantic observing system, impact of climate-related change on west African coastal communities, the need for ecosystem approaches for the sustainable management of fisheries, and the need for enhanced cooperation across the Atlantic and globally to achieve this.
PREFACE scientific findings have been disseminated through over 120 scientific papers in peer-reviewed international journals. Around 70 new publications are underway and a strong effort is being made to make all openly available. This includes datasets and other tools (such as models or open-source software), all of which are indexed on the website “Output” pages.
The consortium was very active in disseminating project results through communications at scientific events and co-organisation of conferences or conference sessions such as: the annual joint PREFACE-CLIVAR-PIRATA Tropical Atlantic Conferences (South Africa 2015, France 2016 and Brazil, 2017); the annual TACCOVAR colloquia (Benin, 2015-2017); the CLIVAR Open Science Conference “Charting the course for climate and ocean research”, (China, 2016); the ICES Annual Science Conference (2015-2017); the International Workshop on Marine & Atmospheric Sciences in West Africa, organised for the inauguration of the Ocean Science Centre Mindelo (Cape Verde, 2017); the Future Ocean conference on Advances in Integrated Ocean Research towards Sustainable Development, targeting early career researchers (Germany, 2017), several sessions at the European Geosciences Union General Assembly (Austria, 2014-2018) and at the IAPSO-IAMAS-IAGA Assembly (South Africa, 2017).
The project was also active in dissemination activities to the wider public through a series of articles in the popular press and small documentaries or interviews at national level, e.g. short documentaries of PREFACE research in the Canary Current region by beneficiaries UCM and Sorbonne Universités (previously UPMC) in collaboration with other national projects and programmes; TV clip promoting conference on ocean and climate, featuring INDP and the University of Cape Verde; lecture by Bourlès (IRD) on “PIRATA, un observatoire océanographique sous les tropiques”, transmitted on 88.8FM Marseille and www.radiogrenouille.com; participation in several science festivals and Open Door days, among others.
Further dissemination, involving also exploitation of results, consist in the inclusion of PREFACE findings: in courses given by PREFACE researchers, such as university courses in Bergen (Norway) on Climate Dynamics and the Atmospheric General Circulation and several lectures given in schools and summer schools; in teaching activities by UCM and UPMC at UCAD, in coordination with other Spanish, French and Senegalese national programmes and projects; during workshops in Morocco, Senegal, Cape Verde, Ivory Coast, Angola and Namibia, particularly those in Senegal and Cape Verde during which the results from the WP13 interview surveys (Deliverable D13.2) were presented to and discussed with the participating fisherfolk communities; in presentations given during visits to regional research centres and fisheries institutes; and in the Masters course at the UNESCO Chair in Mathematical Physics and Applications at UAC (Benin).
As a consequence of the achievements in PREFACE and wide dissemination, PREFACE scientists have been invited as experts to contribute to higher level scientific and/or political arenas upon several occasions, such as, but not restricted to:
+ The Tropical Atlantic Observing System (TAOS) Review commissioned by CLIVAR and white papers for the upcoming Oceans Observations Conference 2019 (http://www.oceanobs19.net/);
+ The CLIVAR EBUS working group, with several scientists from WP5 invited as members, including as chair;
+ The international “CAS-TWAS-WMO Forum" (CTWF) for Physical-Mathematical Problems Related to Climate Modeling and Prediction (that was jointly founded by the Chinese Academy of Sciences (CAS), the World Academy of Sciences (TWAS) for the advancement of science in developing countries and the World Meteorological Organization (WMO)), with contributions from the progress in the anomaly coupling studies from WP7;
+ The WCRP-JSC/CAS Working Group on Numerical Experimentation (WGNE), for increasing understanding of the nature and cause of errors in models used for weather and climate prediction, with contributions from WP7 and the coordinated experiments of WP6;
+ The seventh IPCC report, with contributions from Dr. Hyacinth Nnamchi (UNN, Nigeria) as a chapter lead author;
+ The German Federal Ministry of Economic Cooperation and Development (BMZ) and Corporation for International Cooperation (GIZ), with advice from CT5 for planning of future cooperation and development activities in Africa;
+ The Senegalese Ministry of Fisheries and Economy, with contributions from Dr. Patrice Brehmer to a discussion on “Technical Validation of the Fisheries Sectorial Policy Review”;
+ The Cran Montana Forum, with contributions from CT5 to the high level panel “Ocean Economy and Fishing industry, a strategic sector for Africa”, specifically for “The blue belt initiative, a South-South integrated regional cooperation and development tool”;
+ The Ministry of the Environment and Sustainable Development of Senegal (MEED/DEEC), with contributions to their workshop on “Operational Planning PAS/PNA – Project of Scientific Support to the National Adaptation to Climate Change in Senegal Plan”.
A selection of other contributions to relevant scientific and/or political events is:
+ The preparation of the Ocean Climate Declaration of Dakar, in collaboration with the AWA project and numerous stakeholders and policy-makers. This is a voluntary initiative of research and higher education institutions, programmes and projects, that aim to bring actions particularly by the Sub-regional Fisheries Commission, the Ministerial Conference on Fisheries Cooperation among African States Bordering the Atlantic Ocean (ATLAFCO), the UNEP Abidjan Convention, the Economic Community of West African States, the Arab Maghreb Union, the Economic and Monetary Union of West Africa, the European Commission and the United Nations, on issues related to climate impacts on the marine environment and fishing. It was presented at both the COP21 (Paris, November 2015) and COP22 (Marrakech, November 2016), besides being widely used in many of the CSRP-SRFC (Senegal) policy meetings across West Africa;
+ An invited lecture on “How can data from a vessel based platform make a contribution in global research, and what types of parameters should be monitored?”, based on experience and results drawn from PREFACE, at the 2nd Annual Hjort Open Science Seminar: Ecosystem research, advice and management - how to put policy into practice?, 9-10th June 2016 in Bergen, Norway; a seminar was related to the science planning for the new vessel Dr Fridtjof Nansen for the FAO EAF Nansen Programme;
+ The showcasing of PREFACE upon a reception organised on-board the R/V Meteor during the 2016 south-eastern boundary current cruise (Milestone 16), supported by INIP and the German Embassy in Angola and attended by the Angolan Minister for Fisheries, the Minister for Science and Technology, and the state secretary for aquaculture, Dr. Carlos Martinó Cordeiro. Correspondingly large was the interest by the media with several interviews given to journalists from press and television. In her speech, the Minister of Fisheries pointed toward the importance of climate change, which represents a grand challenge for the socio-economic development of Angola. The minister was pleased about the good collaboration with German marine research institutes and particularly thanked the German government for the support to investigate the dynamics of the marine ecosystem of Angola. Scientists onboard proceeded to present their research topics and the projects supporting their work, with special emphasis to PREFACE;
+ Showcasing of PREFACE by UCT at a workshop hosted by the ESASTAP (strengthening T,R&I cooperation between Europe and South Africa), on “SOUTH AFRICA (SA)-EUROPEAN UNION (EU) MARINE AND POLAR RESEARCH COLLABORATIONS”;
+ Showcasing of PREFACE by GEOMAR and IRD at the “A New Era of Blue Enlightenment” meeting for the signature of the Belém Statement on Atlantic Research and Innovation Cooperation, at the “Connecting to better observe the Atlantic Ocean” session, co-organised with the EU H2020 AtlantOS project, the ETASTAP session (South Africa - EU cooperation) and the INCOBRA session (Brazil - EU cooperation);
+ Co-organisation of a side-event at the CoP22 (November 2016, Marrakesh, Morocco) Ocean Theme: Oceans, Transport and Energy. The side-event was called “Fisheries and aquaculture in the context of climate change: challenges and opportunities” and jointly organised by the FAO with the projects AWA and PREFACE, showcasing key achievements from these projects and sharing experience and lessons learnt for coordinated action towards a sustainable future;
+ Co-organisation of a side-event at the UN Ocean Conference (June 2017, New York, USA). The side-event was called “Africa in Action for Fisheries and Aquaculture to Face Climate Change: Solutions and Dialogue for the Ocean Economy” and PREFACE contributed specifically with the presentation by Dr. Patrice Brehmer on “AWA and PREFACE experience: foundation for the establishment of an observatory for fisheries and the marine environment, targeting SDG 14 goals in West African waters” (https://preface.w.uib.no/events/events-in-detail/#unoc201706);
+ The PREFACE Science-Policy session (please refer to Deliverables D2.6 and D2.9) April 2018, Spain;
+ The workshop on “Small-scale fisheries in the South” on 25-26 June 2018 in Brest, France, organised by the University of Brest in collaboration with IRD, during which a session was dedicated to the presentation of PREFACE results. 40 researchers, managers, funding agencies, training institutions and consultants were invited to discuss issues for a better integration of research in development programmes and development aid supporting sustainable small-scale fisheries.
Finally, a news piece on PREFACE, by journalist Christian Nielsen assigned by Intrasoft, is being edited for publication in the series of EU-funded R&I actions success stories within International Cooperation (http://ec.europa.eu/research/infocentre/index_en.cfm). The Coordinator is also preparing a news piece on the project, together with the Communications unit of the Bjerknes Centre for Climate Research (Bergen, Norway), to be released in Autumn 2018. It will focus on success stories of PREFACE scientists, such as those of Dr. Abdoulaye Sarré from ISRA-CRODT (Senegal), who through PREFACE grew from electronic engineer to Technical Advisor on Acoustics for the Senegalese Ministry of Fisheries, Dr. Pedro Tchipalanga from INIP (Angola), who shows the value of sharing and collaborating on previously private observational data (Tchipalanga et al., 2018) and Dr. Marta Martin del Rey, who started as a PhD on the project, continued with valuable contributions as a Postdoctoral researcher, and has now earned a prestigious MSCA individual fellowship, thanks to her achievements and networks gained through PREFACE. A final summary article is also in preparation (refer to Deliverable D2.8) as is the further editing for publishing of the policy brief (Deliverable 2.9) which will be disseminated through the networks of the consortium and those who were invited to participate in the Science-Policy session (Deliverable D2.6).
Last but not least, a collection of entertaining illustrations about PREFACE were produced during the PREFACE final meeting by the illustrator Bas Kohler. These are in high demand by project partners and participants of the Science-Policy Session, for reuse in lectures, posters and other science dissemination. They will soon be made freely available to the public. We are also discussing transforming them into a short animation or booklet on the project once activities resume after the summer break. The Project Office counts on the UiB office in Brussels and contact with the DG R&I Communications Team (such as Mr. Robert-Jan Smits or Mrs. Alexandra Ruete) for further promotion of this product, and will attempt to submit any such product to AORA (https://www.atlanticresource.org/) for further dissemination on their News & Events page.
A continually updated detailed list of PREFACE dissemination activities, including scientific publications, is available on CORDIS (Community Research and Development Information Service), the European Commission's primary public repository and portal to disseminate information on all EU-funded research projects and their results in the broadest sense.
List of Websites:
www.preface-project.eu (same as https://preface.w.uib.no/ ).
Project Office (coordinator): prefaceproject(at)uib.no
Project Leader: Noel Keenlyside, Noel.Keenlyside(at)uib.no; Tel. ++47 55 58 20 32
Project Manager: Mahaut de Vareilles, mahaut.vareilles(at)uib.no; Tel. ++47 55 58 37 08
Data Manager: Benjamin Pfeil, benjamin.pfeil(at)uib.no; Tel. ++47 55 58 98 39
Financial Manager: Robert Clarke, Robert.Clarke(at)uib.no; Tel ++47 555 821 26
