Skip to main content

North Atlantic Climate: Predictability of the climate in the North Atlantic/European sector related to North Atlantic/Arctic sea surface temperature and sea ice variability and change

Final Report Summary - NACLIM (North Atlantic Climate: Predictability of the climate in the North Atlantic/European sector related to North Atlantic/Arctic sea surface temperature and sea ice variability and change)

Executive Summary:
In the North Atlantic the upper ocean layer strongly communicates with both, the overlying atmosphere and the deep ocean. A large part of the deep waters are created here. The region can be considered as important driver of global climate variability on time scales beyond seasonal. The sea surface state (SST and sea ice) also determines to a large extent the weather in Europe. In NACLIM researchers from 11 different nations have worked together to evaluate uncertainties of climate forecasts and to review and optimize the North Atlantic observation system. The forecasts from the coupled climate models were applied to topics in the marine eco-system and in urban societies.
The predictability of the North Atlantic/Arctic Ocean surface state, its temperature, freshwater content, ice cover and circulation was studied with a suite of state of the art coupled climate models. For the sea surface temperature and upper-ocean heat content in the subpolar North Atlantic we were able to show predictive skill of up to ten years, related to the northward advection of warm and saline subtropical water by the Atlantic meridional overturning circulation. Likewise the decline in Arctic sea ice area and thickness during the last decade could be related to an increased poleward ocean heat transport. Teleconnections between sea-ice cover and the climate over Europe were identified providing some long-term predictability of the atmospheric circulation based on ocean forecasts. The quality of forecasts largely depends on our ability to derive a reliable description of the surface ocean state and to understand the major drivers of its variability. These are still challenging goals as the number of relevant observations in these regions is still poor and the accuracy of the prediction model need to be improved.
NACLIM’s core observation areas were in the North Atlantic subpolar gyre and along the Greenland-Scotland Ridge (GSR). This is where the waters sink into the deep Atlantic Ocean to feed the lower limb of the AMOC. The observations also included the RAPID array, at 26oN, to capture the AMOC farther south. These efforts resulted in time series of basin-wide volume, heat and freshwater fluxes that by now are up to 20 years long and to which NACLIM has contributed data for the past five years. Some of the arrays were redesigned using long term and low cost instrumentation developed in NACLIM. The volume fluxes across the GSR have been remarkably stable during the past decades but the northward heat fluxes have increased significantly due to a warming of the water masses carried. In contrast the volume fluxes in the AMOC at 26 N have declined, leading to a heat divergence in the North Atlantic and an associated slowing of the general warming.
We have assessed potential extensions of the observational system in terms of decadal hindcast skill. We suggest an extension of the global ocean observing system to below 2,000 m, in particular over the Southern Ocean and the North Atlantic. Second, we conclude that salinity observations need to be enhanced, e.g. by using satellite information. We also argue the current RAPID observation is worth continuing particularly to diagnose and understand the variability of the oceanic general circulation, as can be represented by the AMOC, and to verify climate models.
A multivariate data assimilation technique for optimizing model parameters on longer timescales periods and improving the model climate state has been successfully implemented and tested. By using data assimilation based on the adjoint method, we find that the model can be brought into consistency with estimated volume transports from observations. However individual components of the AMOC have a limited skill, linked to the inability of coarse resolution models to simulate overflow across the ridge with the implication that ocean heat transport towards the Arctic is biased.
In NACLIM we also examined the opportunities for forecasting marine ecosystems, exploiting living marine resources in a sustainable and economically efficient manner. We demonstrated predictive skill of biological variables, namely the availability of suitable foraging habitat for Bluefin Tuna and the spatial distribution of suitable spawning habitat for blue whiting. Climate variability and change also affects urban societies. The simulations revealed a nearly tenfold increase in the number of urban heatwave days towards the end of the century, reaching around 30 days with heatwave conditions annually. We had a very close contact with representatives of three target cities, both through bilateral contacts and formal workshops. In the final phase of the project, we organised an international workshop ‘Towards urban climate services’ in Brussels.
Project Context and Objectives:
NACLIM aims at investigating and quantifying the predictability of the climate in the North Atlantic (European) sector related to North Atlantic and Arctic sea surface temperature (SST) and sea ice variability and change on seasonal to decadal time scales.

The overarching goals of NACLIM were
• To quantify the uncertainty of state-of-the-art climate forecasts by evaluating the ability to model the most important oceanic and atmospheric processes in the North Atlantic and Arctic Oceans and by comparing key quantities with observations.
• To optimize the present North Atlantic observation system by evaluating the impact of its components on the quality and quality control of model forecasts, and their value in determining the present ocean state and its past variability.
• To quantify the impact on oceanic ecosystems and on European urban societies of predicted North Atlantic/Arctic Ocean variability.
• To critically assess the use of climate forecast parameters for use by stakeholders in society, politics and industry

To reach these goals NACLIM was organized in ten scientific work packages clustered in 4 scientific core themes. The specific context and objectives of these work packages were as follows:

WP 1.1 – Accessing the predictability of the North Atlantic and Arctic Ocean surface state as well as of key quantities controlling it. These comprise sea surface temperatures, the upper ocean salt content respectively the freshwater content, the changes in Arctic sea Ice coverage and thickness and the subpolar gyre strength. The assessment has been based on existing decadal prediction experiments from the fifth phase of the Coupled Model Intercomparison Project (CMIP5), allowing a multi-model approach and providing more robust and reliable results than previously available. Apart from assessing the predictive skill, WP1.1 has attempted skill attribution, i.e. understanding the mechanisms underlying the predictive skill.

WP 1.2 – To identify the regional pattern and time evolution of the sea surface temperature (SST), surface salinity, and sea ice patterns that strongest influence the atmosphere in the North Atlantic/European sector on seasonal to decadal time scales and quantify their climatic impacts. The novel approach in his WP was the use of an adjoint climate model, known as PLASIM that was developed during the previous FP6 project THOR. The goal was to assess the ability of climate models to reproduce these impacts, identify their potential predictability, and use observations to downscale the model predictions from global to local scales. A second topic in this WP was to quantify the impact of Arctic changes on the generation and activity polar meso-cyclones. The intense low-pressure systems are a potential danger for Arctic shipping.

WP 1.3 – In order to identify the North Atlantic and Arctic surface changes that most affect the atmosphere and to better evaluate their role in the climate predictability, this WP had three main objectives: (1) to characterize the time-space sea surface variability in the Arctic/North Atlantic region, (2) to identify the mechanisms underpinning this variability and link them to indices of variability of the ocean circulation, and (3) to provide information on the respective roles of the atmosphere and the ocean in this variability and identify feedback mechanisms between ocean anomalies and the overlaying atmosphere. Based on the most relevant patterns and feedbacks, anomalous atmospheric fields were constructed and used to force stand-alone ice-ocean simulations in order to evaluate the contribution of the atmosphere to the observed surface variability.

WP 2.1 – In order to verify model-based climate predictions, long-term observations of relevant ocean parameters were necessary. The meridional exchange across the Greenland-Scotland Ridge is of key importance for the North Atlantic climate system and observations of these exchanges are therefore very important in the context of understanding climate variability. The objectives of this WP were thus to provide updated time series of volume and heat transports for all the Atlantic inflow branches to the Nordic Seas and volume and freshwater transports for the most important overflow branches. The variability and trends in these flows were also to be estimated and these data sets served as input to several of the other work packages in NACLIM. It is very costly to maintain these long-term monitoring system, thus in order to make the systems more accurate and sustainable the existing measuring systems were also to be modified within NACLIM.

WP 2.2 - Existing and newly acquired observational data was analysed to derive reference time series suitable for the assessment of the hindcast predictive skill of the CMIP5 models. The focus was on time series of critical variables (volume, heat, and freshwater transport) in focus areas as the warm water inflow region, the deep western boundary current, and the subpolar North Atlantic deep water formation areas. Dedicated observations as well as data available via the GOOS (e.g. satellites, Argo) also have been considered. The options in formatting time series data in a CMOR2 compliant output format have been evaluated. A close cooperation with the US – Canadian – European project OSNAP was envisaged at the beginning of NACLIM and could be started after the funding of this partner project came into place. The WP contributed with a moored array on the Reykjanes Ridge was used to derive transport of volume, heat and freshwater in the northward flowing Irminger Current. Other data sources were considered for the analysis. A major goal was to ensure the sustainability of observations and the linkage to other projects such as the Ocean Observatories Initiative OOI, FixO3, EMSO, RAPID and in particular in the framework of GOOS including Argo.

WP 2.3 – This WP is to large extent a synthesis package. The goal was to develop strategies and corresponding methods for a systematic quantitative comparison of model and observational data. To investigate to what degree models run and analyzed within NACLIM, are capable of reproducing observed variations of hydrographic and kinematic components of the exchanges, across the Greenland-Scotland-Ridge, of the deep western boundary current and of the overturning circulation at 26.5°N. To synthesize observational and model output for the estimate of large-scale and regional budgets we investigated to what extent and to within which accuracy observational data can be used to verify estimates of key ocean parameters and quantities from simulations using the different models employed in NACLIM. We provided, on an annual basis, the observational time series to the NACLIM modelers. The main goal was to close an budgets for the circulation of water masses, both horizontally and vertically, for the North Atlantic and Arctic Oceans.

WP 3.1 – The skill of any dynamical forecast depends critically on the quality of the initial conditions. For climate forecasts these are derived from observations in the ocean and atmosphere that are assimilated in either stand-alone models or in coupled models. Another critical point is the nature of the observations and their distribution in time and space. Rather than being able to freely choose the best data for the initialization, their use is constrained by their availability. This work package aimed to explore the impact of the constraints by working in an ideal model world allowing a free choice of parameters and locations and to find out which components of the observing system are the most important. Detailed objective includes (1) investigating the benefit of the different ocean observing system components for the initialization of decadal climate prediction systems, (2) quantifying the impact of the different observing system components in terms of decadal hindcast skill, and (3) identifying the necessary enhancements and potential reductions of the present observing systems. We investigated whether individual components can be reduced without losing the beneficial effects on the decadal predictive skill as well as potential needs that are not captured by the present ocean observing system in order to enhance decadal predictive skill.

WP 3.2 – Observed changes in the Arctic Ocean have highlighted the role of the Arctic in storage and release of liquid and solid freshwater pools to the North Atlantic on interannual to decadal scales. Redistribution of freshwater constitutes a control on the local air-sea heat exchange and with direct and indirect impact on the regional climate including modification of the Atlantic overturning circulation. Atmospheric tele-connections link Arctic sea-ice and snow cover with climate over the northern hemisphere. The goal in this WP was to develop a broader understanding of the potential predictive skill associated with the initialization of the upper Arctic Ocean, also in order to focus and improve future monitoring systems. This included the development of a new and innovative ice surface temperature remote sensing product that improves the initialization of the Arctic region and to better constrains the heat fluxes associated with the interface. Improvement of model skill, climate and realism was to be achieved through novel developments of systems and techniques that allow systematic parameter optimization - an adjoint assimilation. This system will be trained using climate observations obtained over the Arctic sector. A secondary goal was to evaluate mechanism of the simulated transport variability after the assimilation with those based on observations.

WP 4.1 - Recent years have seen a rapid expansion in the ability of earth system models to describe and predict the state of the ocean: skilful forecasts ranging from seasonal (3 months) to decadal (5-10 years) time scales are now a reality. Such forecasts are potentially of great value in the management of living marine resources and for all of those that are dependent on the ocean for both nutrition and their livelihood. However, translating these forecasts of physical variables (e.g. temperature, salinity) into biological outcomes (e.g. fish distribution and catches) is neither automatic nor straightforward. In this WP we aimed to pioneer the development of so-called “Marine-Ecosystem Climate Services”, whereby the skill of physical forecasts is translated into biological forecasts. We employed three approaches to address this challenge. Firstly, we reviewed the state-of-the-art methods, tools and knowledge for prediction of ecosystem variables in the North Atlantic. We also considered the use of generic biological hypotheses as a basis for making such predictions. We then applied this knowledge to make biological predictions and assess the quality of these forecasts. In the process of this work, we also focused on identifying key gaps in the knowledge and potential areas for future research.

WP 4.2 - While cities occupy only a small fraction of the European continent, the majority of Europeans lives and works in cities. At the same time, urban agglomerations are particularly vulnerable to climate extremes, especially heat wave events. Therefore, it was of particular relevance to investigate the impact of large-scale climate (variability) on urban areas and their populations. The main goal of this WP was apply a deterministic urban climate model to three cities from different areas within Europe, with a rather direct exposure to North Atlantic climate influences (Almada – Portugal, Antwerp – Belgium and Berlin – Germany). This work was to be carried out in close co-operation with the local stakeholders and included a downscaling of the spatially coarse resolution CMIP5 climate predictions to the urban scale, with particular focus on regions where North Atlantic SST variability and changes have a significant influence on the climate. The topic was an investigation of the relation between heat waves and the urban-rural temperature increment, i.e. the urban heat island effect. In an extension of the work these physical parameters were then set into relation to relevant socio-economic data, focusing on health aspects, leading to spatially explicit vulnerability maps (population density, housing quality, age structure ...) and heat risk maps.
Project Results:
Please find the detailed report in the attached pdf file "final report".

Potential Impact:
NACLIM has been characterized by carrying out inter-institutional and interdisciplinary research on different aspects of quantifying the interannual to decadal predictability of the climate in the North Atlantic/European sector related to the North Atlantic/Arctic ocean surface state (WP 1.1). SST and sea ice in the North Atlantic/Arctic region are major factors influencing the atmospheric circulation (WP 1.2) and thus the climate in Europe. In NACLIM we have quantified the uncertainties in predictions of state-of-the-art climate models related to the North Atlantic/Arctic Ocean surface state. This has been done by assessing the multi-model set of initialized decadal prediction ensemble experiments from the CMIP5 project. Based on a wide range of process-oriented modelling studies, we have identified predictable atmospheric circulation patterns related to the North Atlantic/Arctic ocean surface state (WP 1.3) and thus contributed to an increased preparedness to the climatic conditions in the North Atlantic/European sector. The impact of Arctic changes on polar meso-cyclone activity has been quantified (WP 1.2). Polar lows represent a danger for shipping, fishing and off-shore drilling activities due to their strong winds and heavy precipitation. From the modelling studies we have identified the most important feedbacks between the North Atlantic/Arctic Ocean and the atmosphere in the North Atlantic/European sector (WP 1.2) as well as optimal SST and fresh water perturbation patterns that lead to a maximum impact on the atmosphere. These patterns will help to identify where ocean surface state observations will have a maximum impact on improving predictions (WP 3.1). NACLIM has also empirically downscaled atmospheric predictions to local scales of interest for impact studies (WPs 4.2 4.1).
To verify, initialize and improve model-based climate forecasts, monitoring of key oceanic quantities is necessary. NACLIM has extended the monitoring of the exchanges across the Greenland-Scotland-Ridge (WP 2.1) of the deep western boundary currents (WP 2.2) and of the overturning circulation at 26°N (WP 2.2) to duration that by now permits identification of decadal trends. NACLIM has also contributed to filling the existing gap of observing the basin-wide transports in the subpolar North Atlantic (WPs 2.1 2.2). These activities have made an effective contribution to the Global Earth Observing System of Systems (GEOSS) through GEO (Group of Earth Observations). Based on hindcast prediction experiments performed in an ideal model world, NACLIM has identifies potential needs and possible reductions of the existing and future ocean observing system (WP 3.1) in order to enhance the forecast skill in the North Atlantic/European sector. In order to improve the Arctic initialization of climate prediction systems, a new and innovative dataset of Arctic sea (ice) surface temperatures has been constructed (WP 3.2). The project has also identified sources of predictive skill due to initializing the Arctic Ocean and sea ice based on data withholding experiments with a coupled climate model (WPs 3.1 3.2). Based on these studies NACLIM made specific recommendations how to best initialize climate predictions as well as how to optimize the present ocean observing system WPs 2.3 3.1).
Combining the CMIP5 decadal prediction experiments with knowledge about physical – biological links, NACLIM has quantified the reliability of these forecasts for case studies from the North Atlantic marine ecosystem (WP 4.1) including commercially exploited fish stocks, and made recommendations regarding future research requirements necessary to achieve reliable forecasts of high trophic levels of the oceanic ecosystem. The project has down scaled the CMIP5 European climate change predictions to the urban scale, using a deterministic urban climate model in order to match the scale of interest for local stakeholders (WP 4.2). The resulting high resolution urban climate predictions have been coupled to relevant socio-economic data for a number of European cities in order to produce heat risk maps. This latter task has be established by a Belgian SME, partner in the NACLIM consortium. It was also responsible for the dissemination of the urban climate risk results to local stakeholders such as city authorities, the private sector or the industry. To this end, several workshops with stakeholders from all over Europe have been carried out.
NACLIM research focused on the North Atlantic/European sector and the results are relevant for Europe as a whole and not only for a particular country. Furthermore, the project brought together a large number of experts to tackle its interdisciplinary approach of integrating existing model-based climate predictions, process-oriented modelling studies, ocean field and satellite observations as well as reanalysis products, oceanic ecosystem modelling, urban climate modelling and socio-economic data as well as dissemination to climate services. Also NACLIM’s contribution to long-term, basin-wide observations of key oceanic parameters could not have been accomplished on a purely national level, not least due to the considerable costs of conducting field surveys, especially ship time. Benefits and efficiencies are also gained from pooling computing resources across national boundaries.
NACLIM was also heavily engaged in the umbrella project ECOMS, where the expertise of the three FP7 projects SPECS, EUPORIAS and NACCLIM were efficiently combined.

Particular aspects of individual work packages are listed below.

WP 1.1 - The ocean surface state in the subpolar North Atlantic, the Nordic Seas and parts of the Arctic Ocean significantly impacts the climate in Europe through air-sea heat exchanges and changes of the atmospheric circulation in the North Atlantic / European sector. Prior to and within NACLIM we demonstrated in a multi-model approach that surface temperature and salinity in the subpolar North Atlantic and the eastern part of the Nordic Seas are skilfully predictable up to a decade (and maybe beyond) due to the poleward advection of warm and saline subtropical water. Understanding the mechanisms underlying the predictive skill of the North Atlantic / Arctic Ocean surface state and the associated atmospheric predictability (WP1.2) and the differences between models is key to more reliable climate predictions in the North Atlantic / European sector. The work performed within WP1.1 is highly relevant for the predictability of marine ecosystem variability (as addressed in WP4.1). Prior to and within NACLIM, strong co-variability between physical oceanic quantities and abundance and distribution of marine species from various trophic levels, including economically important fish species, has been demonstrated for the north-eastern North Atlantic. Assessing and improving predictability of physical oceanic quantities thus translates into (improved) predictability of abundance and distribution of marine species which is highly beneficial for e.g. fisheries.

WP 1.2 - The socio-economic importance of predicting climate fluctuations and changes on seasonal to decadal time scale is well known, but progress will only be achieved by carefully analysing the observed regional impacts of SST, SIC, snow cover, and ocean circulation variability onto the atmospheric circulation, and by understanding the dominant physics at play, so that the observational evidence can be used to complement, test, and improve the climate models that are used in predictability studies. The most significant dissemination activities have been the publication of research articles, numerous communications and participation in international meetings and workshops, and the holding of seminars in universities and research institutes. These activities have largely contributed to the advancement of our scientific understanding of the ocean influence on climate. In addition, the project has contributed to career development and to training graduate students in atmospheric dynamics, numerical climate modelling, and statistical analysis.

WP 2.1 - One of the main uncertainties in projections of the future climate of Europe is the strength of the AMOC. Most climate models indicate a weakening during the 21st century but they disagree strongly on the magnitude of the weakening. This problem has been highlighted by reports that the weakening has already been initiated. The deep limb of the AMOC is fed from both the western North Atlantic and from the overflow with its entrained water masses but for Europe the most important component is the overflow and the associated warm surface inflow. With the results from NACLIM WP 2.1 it is clear that there has not been any weakening in this component during the last two decades. Another key uncertainty in climate projection is the future fate of Arctic sea ice and the many feedback processes associated with it. This problem has proven to be difficult to model reliably but some studies indicate a significant role for the oceanic heat transport from the Atlantic i.e. the inflows across the Greenland-Scotland Ridge. The increased heat transport documented by NACLIM WP 2.1 may therefore help explain some of the dramatic decreases in Arctic sea ice extent during the last decade. On long time scales a key role of the Greenland-Scotland Ridge exchanges is the removal of carbon dioxide and heat from the atmosphere and transport into the deep ocean. The documented stability of the two main overflow branches shows that this climatically important transport mechanism has remained stable during the last two decades and the documented warming of the Faroe Bank Channel overflow even implies increased heat transport into the deep ocean.

WP 2.2 – The potential impact of this WP is related to the sustained observing efforts in the Sub Polar North Atlantic and elsewhere. It could be shown that Argo floats alone do not provide the required temporal and local (spatial) resolution to substitute the moored observing effort. This is not only related to the fact that Argo does currently not provide data below 2000m but also because of the low (10 days) time resolution. The optimization will be continued for mixed, multiplatform and multidisciplinary observing objective in the currently running H2020 AtlantOS project. The close collaboration with the US/UK OSNAP program and the Canadian VITALS project had scientifically important impacts and related dissemination (conferences, workshops).

WP 2.3 – comprised an integral approach of observational and modelling studies of atmosphere, ocean and climate in the North Atlantic and Arctic oceans. This WP has made an attempt, so summarize the individual components of this effort to provide a comprehensive view on the present knowledge of the ocean circulation and water mass budgets that impact on Europe’s climate. The long time series from the Greenland-Scotland Ridge and the subpolar gyre have provided the backbone that has stimulated new questions and generated additional observational project and theoretical studies of the AMOC. Long time series are not easily obtained. They require time and it is therefore vital that the NACLIM observations are continued. NACLIM and the preceding observation programs have introduced a new generation of sea going oceanographers now ready to continue the observational work.
WP 3.1 - Spatial and temporal coverage of ocean observations have been significantly enhance after introduction of the ARGO floats. However, its main observing areas are limited to the upper ocean, e.g. above 700 m. Results in this work package provide a value of deep ocean observation, e.g. below 2000 m, which in particular can provide invaluable information for better decadal prediction. Also another result pointing the importance of ocean salinity information supports surface ocean surface salinity as a key parameter for proper climate modeling, which strongly support continuous use of satellite information to ocean climate projections.

WP 3.2 - With the DSPE method a robust means for parameter optimization is at hand that enables the automatic tuning of climate models with observed data. As a result a tuned version of the CESAM in medium resolution is now available for climate sensitivity studies. The ice and sea surface temperature data set (IST/SST, AASTI) will be further developed in EU and EUMETSAT R&D projects. The achievements in NACLIM have been the foundation of the operationalization of a level-2 processing chain in the OSI SAF (OSI-205), and subsequently to the Copernicus CMEMS level-4 IST/SST product (CMEMS). Improvements of algorithm and uncertainties are done in the ongoing FP7 ICE-ARC and H2020 EUSTACE projects. Furthermore, the AASTI data set is an important input to the new global surface temperature data set in the EUSTACE project. The continuous work on AASTI heritage has led to highest quality uncertainties (Ghent 2016) and state-of-the-art IST performance (FICE 2017). Production of AASTI version-2, with improved uncertainties, improved cloud mask, various bug-fixes is in progress at DMI and Met Norway and the data set is expected finalizes by summer/autumn 2017. AASTI version-1 in level-3 is available from DMI on ftp from 2000-2009.

WP 4.1 - Taken as a whole, the most important single result from this work is the lesson that forecasting of marine ecosystems in Europe on the seasonal and particularly on the decadal scale is feasible. This result represents a significant advance beyond the state-of-the-art at the start of the project, where just a few marine seasonal forecasts could be found, none-of-which were in Europe. These proof-of-concept results from the NACLIM project can therefore, we believe, lead to a rapid blooming of marine ecological climate services and forecast products in Europe in the near future that will come to play an important role in both the management and exploitation of the oceans, and in the response of human systems to climate variability and climate change. In order to exploit and further develop the possibilities offered by these results, NACLIM WP41 initiated several key activities to raise their profile in the scientific community. The most prominent of these was the creation of a Theme Session dedicated to the topic at the Annual Science conference of ICES (the International Council for the Exploration of the Seas) held in Riga, Latvia, in 2016. Between two and three hundred scientists from Europe and North America attended this session in the course of two days, and the session has already had the desired effect of bringing the potential of these results to the fore.
A further important legacy of NACLIM WP 4.1 has been the creation of a community of researchers that are now working together to develop marine ecological climate services. In 2016, ICES created a scientific working group, the Working Group on Seasonal-to-Decadal Prediction of Marine Ecosystems (WGS2D) to operationalise this knowledge and incorporate it into the routine management of living marine resources. The results of the NACLIM project will be integral to this undertaking and the group will represent one of the main scientific successors to NACLIM WP41. Furthermore, NACLIM has been successful in forming links to groups in other regions of the world that are also attempting to predict ecosystems, notably via several invited talks in the USA. An informal network linking researchers working on ecological prediction in Europe (via ICES WGS2D), the Pacific (via the PICES SG-CEP group) and globally (via IMBER CLIOTOP) has been established, and further joint activities and collaborations are envisaged. NACLIM has therefore played a key role in helping bring disparate groups together into a coherent body, and therefore, effectively establishing a new field of scientific research. The impacts of this project are therefore expected to be felt for well into the future, as this new field attempts to realise the exciting potential of marine ecological forecasting.

WP 4.2 - Scientifically, results obtained with the UrbClim model have already generated impact. Indeed, since the publication of the UrbClim model, requests from many institutes from around the globe have arrived for use of the model. Currently (March 2017), the model is being used, among others, at IC3 (through a collaboration launched from within the ECOMS cluster), and more recently also at the UNESCO-IHE Institute for Water Education in the Netherlands. Moreover, discussions related to installing the model at the Chinese National Climate Centre are ongoing. Most of this scientific impact has been generated through our numerous participations to scientific conferences journal publications. The potential future exploitation of UrbClim in China is largely the result of visits by VITO staff to the National Climate Center in Beijing in early 2016 and early 2017. It should be noted that, even though the UrbClim model itself is to be considered background to the project (i.e. it had been developed largely prior to the start of NACLIM), the results obtained during NACLIM and the subsequent publications have greatly contributed to establishing its reputation and scientific acclaim.
From a societal perspective, the work package ‘impact on urban societies’ has generated impact mostly in the three cities Almada, Antwerp, and Berlin. Representatives of each of these cities have been very actively involved in the project as users, and they have increasingly incorporated NACLIM results into their own policy and climate adaptation planning activities. This impact was achieved, on the one hand, in an informal manner, by regular bilateral contacts between the scientific and the user partners. On the other hand, formal events (user meetings) were organized, twice in Almada (2013 and 2015), and once in each of Antwerp (2013), Berlin (2014), and Brussels (2015). During these events, users were asked to help steer the research work, among others by suggesting the time horizons of the future climate projections and that were relevant for them, and also by suggesting climate adaptation measures to be implemented in the modelling. Users also actively provided local data sets required for the climate simulations, such as local 3-D building cadastre maps, and local in-situ measurement data.
Efforts were also conducted to create societal impact beyond the three formal NACLIM users (target cities). To do so, we organized an international workshop ‘towards urban climate services’ in Brussels, in June 2016. While a part of the this two-day workshop was devoted to communicating NACLIM’s principal urban climate results, another part was dedicated to an interactive panel/audience discussion regarding the way forward in the domain of urban climate services. The workshop was attended by several tens of participants, from academia, cities, and the European Commission.

List of Websites:
Project webpage
www.naclim.eu

Project films
During the project life time four dissemination films were produced and made available in social media. The films can be found i.e. in youtube, please simply search for “NACLIM”.

The North Atlantic Climate (Part 1): What we need to find out - a short introduction to the main challenges NACLIM’s researchers are facing to in their investigations on climate variability in the project. In this video, one of our senior scientist, Mojib Latif of GEOMAR, explains the importance of understanding the mechanisms of climate variability in the North Atlantic area.
https://youtu.be/EAFHBvHmHyw?list=PLoqC0y5-eTV5qxBm_aJERRS0dBy4XIbg2

The North Atlantic Climate (Part 2): Monitoring the Atlantic Ocean - In this EU funded project, scientists from different European institutions take measurements in the ocean from Greenland to the Bahamas with concentration on 3 regions: the Greenland-Scotland Ridge, where the exchange between the North Atlantic Ocean and the Nordic Seas takes place, the Subpolar North Atlantic, and the Subtropical North Atlantic. The continuous observation data will form the input to improve the ability of models, and using these models future climate changes could be predicted.

In this part of the film, our scientists Barbara (Bee) Berx from the Scottish Association for Marine Science (MSS), Laura de Steur from the Royal Netherlands Institute for Sea Research (NIOZ), Gerard McCarthy from the Natural Environment Research Council UK (NERC) and the project coordinator Detlef Quadfasel at University of Hamburg (UHAM) speak about the observational activities carried out in the project and the findings so far.
https://youtu.be/g5qWDk_uDpc?list=PLoqC0y5-eTV5qxBm_aJERRS0dBy4XIbg2

The North Atlantic Climate (Part 3): Adapting to warmer cities - The project aims to better understand the effects of the North Atlantic and Arctic oceans on global climate change. Climate change is predicted to cause more frequent and higher intensity extreme weather events, such as heat waves. Within NACLIM, scientists are trying to find out how to prepare our society for this. In particular European cities which experience the urban heat island effect are being studied.

Three European cities have been involved in the research studies. Griet Lambrechts from the municipality of Antwerp, Sara Dionísio from the city council of Almada, as well as Jörn Welsch from the senate of Berlin explain the needs and expectations of their cities. Scientific clarification is provided by our scientists Dr. Dirk Lauwaet from the Flemish Institution for Technological Research (VITO) and Catherine Stevens from GIM Belgium. Dr. Andrea Tilche from the European Commission summarises the necessary measures to enhance climate science and climate services in Europe.
https://youtu.be/p5WmZsHF5hE?list=PLoqC0y5-eTV5qxBm_aJERRS0dBy4XIbg2

The North Atlantic Climate (Part 4, a summary): the project and what we've learnt - After 4 years of intensive work, the EU funded project NACLIM has come to an end. The project closed on the 31st of January 2017, with 60 reports to the European Commission and around 80 publications. NACLIM (02.2012 – 01.2017) stands for "The North Atlantic Climate". The project has enabled the scientists to gain a deeper understanding of the mechanisms that control the ocean circulation in the North Atlantic and Arctic oceans, how these mechanisms interact with each other and how they affect global climate change. The output of the observations in the North Atlantic and Arctic Oceans is integrated into a surveillance system that can operate as a prediction system for the climate of 15 to 25 years ahead.

Within the project our scientists also tried to understand the phenomenon of more frequent and higher intensity extreme weather events, such as heat waves and in particular the urban heat island effect, so that the end-users (such as European cities) can prepare for this better.
https://youtu.be/9J_Ky-_AUoU?list=PLoqC0y5-eTV5qxBm_aJERRS0dBy4XIbg2

The project was run by 18 research groups with over 60 scientists from 10 European countries. These institutes are:

1. University of Hamburg, Germany
2. Max Planck Institute for Meteorology, Germany
3. University Pierre et Marie Curie – Paris 6, France
4. University of Bergen, Norway
5. UNI Research Bergen, Norway
6. GEOMAR Helmholtz Centre for Ocean Research, Kiel, Germany
7. Danish Meteorological Institute, Denmark
8. Faroe Marine Research Institute, Faroe Islands
9. Finnish Meteorological Institute, Finland
10. Marine Research Institute, Island
11. Royal Netherlands Institute for Sea Research, Netherlands
12. The Scottish Association for Marine Science, UK
13. Natural Environment Research Council, UK
14. Nansen Environmental and Remote Sensing Center, Norway
15. Flemish Institute for Technological Research, Belgium
16. G.I.M. Geographic Information Management, Belgium
17. Technical University of Denmark
18. Marine Scotland, UK

The NACLIM project and all videos are funded by the European Commission, through the 7th Framework Programme for Research, Theme 6 Environment, Grant Agreement 308299.

Publications are available at:
1) http://naclim.zmaw.de/index.php?id=2225 and
2) ZENODO → Community: NACLIM

Contact details:
Prof. Dr. Detlef Quadfasel (University of Hamburg), Tel. +49 40 42838 5756, Email: Detlef.Quadfasel@uni-hamburg.de
Chenbo Guo (Max Planck Institute for Meteorology), Tel. +49 40 41173 285, Email: Chenbo.Guo@mpimet.mpg.de