Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CLIMATE SCIENCE (oxfordre.com/climatescience). (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 13 August 2020

Projected Oceanographical Changes in the Baltic Sea until 2100

Summary and Keywords

In this article, the concepts and background of regional climate modeling of the future Baltic Sea are summarized and state-of-the-art projections, climate change impact studies, and challenges are discussed. The focus is on projected oceanographic changes in future climate. However, as these changes may have a significant impact on biogeochemical cycling, nutrient load scenario simulations in future climates are briefly discussed as well. The Baltic Sea is special compared to other coastal seas as it is a tideless, semi-enclosed sea with large freshwater and nutrient supply from a partly heavily populated catchment area and a long response time of about 30 years, and as it is, in the early 21st century, warming faster than any other coastal sea in the world. Hence, policymakers request the development of nutrient load abatement strategies in future climate. For this purpose, large ensembles of coupled climate–environmental scenario simulations based upon high-resolution circulation models were developed to estimate changes in water temperature, salinity, sea-ice cover, sea level, oxygen, nutrient, and phytoplankton concentrations, and water transparency, together with uncertainty ranges. Uncertainties in scenario simulations of the Baltic Sea are considerable. Sources of uncertainties are global and regional climate model biases, natural variability, and unknown greenhouse gas emission and nutrient load scenarios. Unknown early 21st-century and future bioavailable nutrient loads from land and atmosphere and the experimental setup of the dynamical downscaling technique are perhaps the largest sources of uncertainties for marine biogeochemistry projections. The high uncertainties might potentially be reducible through investments in new multi-model ensemble simulations that are built on better experimental setups, improved models, and more plausible nutrient loads. The development of community models for the Baltic Sea region with improved performance and common coordinated experiments of scenario simulations is recommended.

Keywords: Baltic Sea, climate change, regional climate modeling, dynamical downscaling, future projections, uncertainties, scenarios, natural variability, model deficiencies, ensemble simulations

Introduction

The Baltic Sea is a semi-enclosed sea in northern Europe with limited water exchange with the world ocean (Figure 1). It stretches from about 54°N to 66°N latitude and from about 10°E to 30°E longitude and is surrounded by a large catchment area that is about four times as large as the surface of the sea itself.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 1. Bottom Topography of the Baltic Sea (depth in m).

Source: H. E. Markus Meier, IOW and Swedish Meteorological and Hydrological Institute (SMHI).

Note. The Baltic proper comprises the Arkona Basin, Bornholm Basin, and Gotland Basin. The border of the analyzed domain of the models is shown as black line in the northern Kattegat. In addition, the monitoring station Gotland Deep (BY15) is shown.

The climate of the Baltic Sea region varies considerably due to maritime and continental weather regimes. For the period 1980 to 2009, the annual mean sea surface temperature (SST) amounts to about 8.3°C (Figure 2). The seasonal cycle of the SST is pronounced and the northern Baltic Sea is ice covered every winter (not shown). Due to the large spatial extension, the Baltic Sea is characterized during all seasons by a distinct SST difference between colder northern and warmer southern sub-basins (Figure 2).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 2. Seasonal Mean SST (in °C) during 1980–2009 in Hindcast Simulations Driven with Atmospheric Reanalysis Data.

Source: H. E. Markus Meier, IOW and SMHI.

Note. From the upper left to the lower right panel: (a) winter (December–February), (b) spring (March–May), (c) summer (June–August), (d) autumn (September–November) and (e) annual mean SSTs are shown.

In the southern Baltic Sea, there is also a pronounced west–east temperature gradient, mainly during summer and autumn. The annual mean sea surface salinity (SSS) distribution shows a large north–south gradient mirroring the input of fresh water from rivers, mostly located in the northern catchment area, and salt water inflows from the North Sea (Figure 3).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 3. Annual Mean SSS and Bottom Salinity and Winter Mean Sea Level, 1980–2009.

Source: H. E. Markus Meier, IOW and SMHI.

Note. (a) Annual mean SSS and (b) bottom salinity (in g kg-1) and (c) winter (December–February) mean sea level (in cm) during 1980–2009 in hindcast simulations driven with atmospheric reanalysis data (from left to right). Note that the model results of the sea level are given in the Nordic height system 1960 (NH60) by Ekman and Mäkinen (1996).

As the water exchange between the Baltic Sea and North Sea through the Danish straits is limited, the SSS drops from about 20 g kg-1 in Kattegat to about 8 g kg-1 in the southwestern Baltic Sea. For the period 1980 to 2009, the annual mean SSS of the Baltic Sea including Kattegat amounts to about 6.6 g kg-1. Occasionally big inflows of heavy salt water from Kattegat ventilate the bottom water of the Baltic Sea, filling its deeper regions (Mohrholz, 2018; see also Figure 3). Due to almost absent tides, mixing is limited and the water column is characterized by a pronounced vertical gradient in salinity, and consequently also in density, between the sea surface and the bottom. The freshwater-induced seasonal cycle in SSS is relatively small compared to the spatial salinity gradients (not shown). Due to the seasonal cycle in wind speed, with wind directions predominantly from southwest, the sea level in the Baltic Sea varies considerably throughout the year, with highest sea levels of about 40 cm, relative to Kattegat at the northern coasts in Bothnian Bay and at the eastern coasts in the Gulf of Finland (Figure 3).

For the period 1980 to 2009, the annual mean sea level amounts to about 16 cm, with a horizontal north–south difference of about 35 cm (not shown). This slope in sea level is explained by the lighter brackish water in the northeastern Baltic Sea compared to the Kattegat and by the mean wind from southwesterly directions which pushes the water to the north and to the east (Meier, Broman, & Kjellström, 2004a).

Due to its location and physical characteristics, the Baltic Sea is vulnerable to external pressures such as eutrophication, pollution, or global warming (e.g., Jutterström, Andersson, Omstedt, & Malmaeus, 2014). The reason for the relatively high vulnerability can be found in the slow response compared to changes in the forcing. The volume of the Baltic Sea amounts to 21,700 km3 (Sjöberg, 1992) and consequently the turnover time of the total freshwater supply of about 16,000 m3 s-1 (Meier & Kauker, 2003a) is about 40 years. The time scale of the salinity response to changes in atmospheric and hydrological forcing is estimated at about 20 years (Meier, 2006). If age is defined as the time elapsed since a water particle left the sea surface, median ages of the bottom water will range between one year in the Bornholm Basin and seven years in the northwestern Gotland Basin (Meier, 2005). Hence, regardless of the considered pressure the Baltic Sea is a relatively slow coastal system.

In the early 21st century, about 85 million people, in 14 countries, are living in the catchment area, representing a considerable anthropogenic pressure for the marine ecosystem (HELCOM, 2018). Insufficiently treated wastewater, emissions of pollutants, overfishing, habitat degradation, and intensive marine traffic such as oil transports put a heavy burden on the ecosystem of the Baltic Sea. One example is the oxygen depletion of the Baltic Sea deep water, with the consequence of dead sea bottoms lacking higher forms of life (Carstensen, Andersen, Gustafsson, & Conley, 2014; Meier, Eilola, et al., 2018; see also Figure 4). In 2016, the area of the dead bottoms reached the size of the Republic of Ireland, with an area of about 69,000 km2, which is about one sixth of the sea surface of the Baltic Sea.1 Bottom oxygen of the deeper parts of the Baltic is depleted because of the limited ventilation of the deep water and because of the accelerated oxygen consumption due to the remineralization of organic matter (Meier, Väli, Naumann, Eilola, & Frauen, 2018; see also Figure 4). Since the 1940s, nutrient loads into the Baltic have increased due to population growth and intensified fertilizer use in agriculture (Gustafsson et al., 2012). Nutrient loads reached their peak in the 1980s and declined thereafter until the early 21st century as a consequence of the implementation of abatement strategies. Nutrients (i.e., phosphorus and nitrogen) in the surface layer during winter are a good indicator for the intensity of the following spring bloom. Highest sea surface concentrations of winter mean phosphate and nitrate are found in the coastal zone, in particular close to the mouths of the large rivers in the southern Baltic Sea that transport elevated loads of nutrients into the sea (Figure 4). During the period 1980 to 2009, high concentrations of phytoplankton blooms are confined to the coastal zone, the area with the highest nutrient concentrations (Figure 5). Water transparency, measured with the so-called Secchi disk, is low in the Baltic Sea compared to the open ocean (Fleming-Lehtinen & Laamanen, 2012).2 For the period 1980 to 2009, the annual mean Secchi depth averaged for the entire Baltic Sea, including Kattegat, amounts to about 6.5 m only. In the coastal zone the Secchi depth is much smaller than in the open Baltic Sea (Figure 5). Also in the northern Baltic Sea, Secchi depth is smaller than in the Gotland Basin due to yellow substances originating from land (Fleming-Lehtinen & Laamanen, 2012).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 4. Summer Mean Bottom Oxygen Concentrations, Winter Mean Surface Phosphate Concentrations, and Winter Mean Surface Nitrate Concentrations, 1980–2009.

Source: H. E. Markus Meier, IOW and SMHI.

Note. (a) Summer (June–August) mean bottom oxygen concentrations (in mL L-1), (b) winter (December–February) mean surface phosphate concentrations (in mmol P m-3), and (c) winter (December–February) mean surface nitrate concentrations (in mmol N m-3) during 1980–2009 in hindcast simulations driven with atmospheric reanalysis data (from left to right). Nutrient concentrations are vertically averaged for the upper 10 m.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 5. Annual Mean Phytoplankton Concentrations and Annual Mean Secchi Depth, 1980–2009.

Source: H. E. Markus Meier, IOW and SMHI.

Note. (a) Annual mean phytoplankton concentrations (in mg Chl m-3) and (b) annual mean Secchi depth (in cm) during 1980–2009 in hindcast simulations driven with atmospheric reanalysis data. Phytoplankton concentrations are vertically averaged for the upper 10 m. As only one value of yellow substances per sub-basin used as background value for the calculation of Secchi depth are available, artificial borders between sub-basins are visible.

In addition to environmental challenges, climate is changing. Although climate, that is, the statistics of weather both in the atmosphere and in the ocean, has changed all the time since the emergence of the Baltic Sea after the last glaciation around 10,000 years ago, changes in air temperature and sea ice during1980 to 2009 are unprecedented during the era of instrumental measurements (BACC II Author Team, 2015). As the Baltic Sea is located in high latitudes (Figure 1), Arctic amplification (e.g., Serreze & Francis, 2006) impacts its water temperature trends considerably (Kniebusch, Meier, & Neumann, 2019). Reinforced by the land-locked location (under global warming the land is warming faster than the ocean, e.g., IPCC, 2013) and the impact of the North Atlantic variability such as the Atlantic Multidecadal Oscillation (AMO), the Baltic Sea is currently warming faster than any other coastal sea worldwide (Belkin, 2009; Kniebusch et al., 2019).

In addition, a systematic increase in sea level is observed, which, at least in the southern Baltic Sea, is not counteracted by land uplift (Figure 6). Land uplift is at its maximum in the northern Baltic Sea sub-basins as a consequence of the land response to the melting of glacial ice sheets after the last glaciation. Other ongoing changes are observed as well, although they may not be statistically significant on long time scales compared to the natural variability of the system, such as changes in wind speed (BACC II Author Team, 2015). The large natural variability on time scales of years and longer makes it difficult to detect anthropogenic changes in the marine environment and to distinguish natural and anthropogenic changes. For the Baltic Sea region, in particular, two modes of long-term variability have been observed and analyzed in detail: first, the so-called North-Atlantic Oscillation (Hurrell, 1995) affects the Baltic Sea region by variations in westerly winds with alternating a warm and humid (maritime) climate during strong west wind situations and a cold (continental) climate on time scales of about six to eight years (e.g., Omstedt & Chen, 2001); and, second, the AMO on time scales of about 60–90 years (e.g., Börgel, Frauen, Neumann, Schimanke, & Meier, 2018). Hence, the anthropogenic signals since preindustrial time are difficult to detect compared to the “noise” of the system (e.g., Kniebusch et al., 2019).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 6. Annual Mean Sea Level Changes in Centimeters for 14 Swedish Mareographs since 1886.

Source: SMHI.

Note. The data are corrected for land uplift. The grey line shows a smoothed curve.

To disentangle the various drivers of the Baltic Sea climate variability, coupled environmental–climate models that explicitly resolve both the circulation in the ocean and the marine ecosystem, including carbon and nutrient cycles, have been developed (see Meier, Edman, et al., 2018, and references therein). Understanding the response to drivers is a prerequisite for the development of regional mitigation and adaptation strategies and, consequently, highly requested from policymakers and stakeholders (e.g., HELCOM, 2013). In this context, scenario simulations of the marine ecosystem have been performed, taking into account environmental pressures under changing climate (e.g., Meier, Andersson, et al., 2011; Meier, Edman, et al., 2018; Omstedt et al., 2012).

In this article, concepts of regional climate modeling for the future Baltic Sea are presented. The development and current state of the research and its applications are described, and the challenges are discussed. The focus is on projected oceanographic changes in a future climate. Examples of ensembles of scenario simulations are presented and discussed.

Regional Climate Modeling of the Future Baltic Sea

Motivation

Society and decision-makers request information in order to adapt to climate change, especially when long-term investments are concerned. For instance, icebreakers have a lifespan of about 30 years and harbors and bridges have an even longer lifespan. Very often owners of holiday houses in the Swedish archipelago request information about the rising sea level and whether their properties will be flooded in a future climate.

Another prominent example is the rebuilding of the locks (Slussen) in Stockholm between Lake Mälaren and the Baltic Sea.3 Lake Mälaren is the drinking water reservoir for about 2 million people living in the valley of the lake. Since the 1940s, the water level of Lake Mälaren has been regulated through channels and gaps in Stockholm and Södertälje, a city located south of Stockholm. To prevent saltwater intrusions, the water level of the lake is regulated by opening or closing the locks when the water level drops or rises to a certain level. The flood risks in Mälaren’s valley are increasingly severe due to rising sea levels in the Baltic Sea. The possibility of enough draining water from Lake Mälaren is too low to prevent floods from heavy water flows to the lake. The new locks will be built to last for about 100 years, which is normal for this type of construction. The locks will be adapted to cope with the future sea level rise that may occur throughout its lifespan. The costs of the project are estimated to be about 12.2 billion Swedish crowns or about 2.5% of the gross domestic product (GDP) of Sweden in 2017 (Andersson, 2018).

Thus, there are many reasons why detailed information about the effects of climate change in the Baltic Sea can be important. Since the 1990s, more climate information focusing on regional topics has been requested by decision-makers and by the public than ever before.

Climate Information

Worldwide, the perception of climate change is characterized by extreme weather events such as breaking temperature records, Arctic sea-ice decline, and increasing sea level. To provide unbiased and objective information about climate change, the Intergovernmental Panel on Climate Change (IPCC) has performed regular assessments about the scientific knowledge on past and future climate change. However, these assessments focus on large-scale changes of the Earth system and do not provide regional details. Therefore, regional assessments were carried out as a complement to the IPCC reports, such as the Assessment of Climate Change for the Baltic Sea Region (BACC, see BACC Author Team, 2008; BACC II Author Team, 2015) and the North Sea Region Climate Change Assessment (NOSSCA, see Quante & Colijn, 2016).

In addition, within the Global Framework for Climate Services (GFCS) climate services were developed that ‘provide climate information to help individuals and organizations make climate smart decisions’ (WMO, 2014). To inform decision-makers at all levels about climate variability and future climate change, in many countries climate service centers have been established.

Climate Models

Numerical models of the climate system have been developed to obtain projections, that is, to simulate the future climate. However, projections are not the only motivation. Further reasons for the development of climate models are the better understanding of the behavior of the climate system and the exploration of the causes of past climate change. A global climate model is based on a general circulation model (GCM) that describes the circulation in the atmosphere and in the ocean. In a GCM, the entire Earth is covered by grid boxes in which transports of mass, momentum, energy, and matter go in and out. Typical horizontal and vertical resolutions of the grid boxes in the atmosphere and ocean are about 80 km and 40–100 levels, respectively. The equations that are solved in GCMs are based on first principles, that is, the conservation of mass, momentum, energy, and matter. This set of partial differential equations is transformed into a set of coupled equations of finite differences and solved by forward integration in time. Running a GCM is based on the principle that the integration is starting from a given state of the climate system. Then the time tendencies of state variables are calculated and added to the state of the system. During the next time step new tendencies are derived and added to the state and so on. Vilhelm Bjerknes (1862–1951) was the first to propose this numerical method to solve the underlying equations of numerical weather prediction models (Bjerknes, 1904).

Limitations of Climate Models

Although GCMs are supposed to describe all relevant processes of the climate system, not all processes can explicitly be resolved because of limited computer resources that prevent very high resolution, that is, the grid boxes are not small enough. In addition, approximations for, for example, ocean eddies, sea ice, turbulence, clouds, and precipitation are needed. These approximations are called parameterizations and they express small-scale phenomena in large-scale parameters. Hence, GCMs compromise between detailed descriptions and high computational speed. Further, GCMs comprise the physical components of the Earth system such as atmosphere, sea ice, ocean, and land surface. State-of-the-art climate models are models that explicitly include the carbon cycle within and between model components, so-called Earth system models (ESMs) (Flato, 2011).

Dynamical Downscaling

Today’s GCMs reproduce large parts of the observed climate, in terms of long-term averages, variability, and extreme conditions. However, some weaknesses are that GCMs only represent large-scale (> 100 km) phenomena explicitly and that the knowledge of how relevant small-scale processes with impact on the large scale can be described in the models (e.g., mixing) is not fully available. Hence, regional climate models (RCMs) have been developed to refine the information from global GCMs to regional and local scales. To consider the large-scale flow, limited-area RCMs are driven by data from GCMs at the lateral boundaries (Figure 7). The first generation of RCMs consisted of regional atmosphere models and were not coupled to other components of the Earth system such as the ocean or sea ice (Giorgi, 1990). Regional climate atmosphere models have an added value compared to global climate models with respect to the representation of height relief details, the land–sea mask, sea surface boundary conditions (SST and sea ice), more detailed vegetation and soil characteristics, and extremes such as cyclones (e.g., Feser, Rockel, Storch, Winterfeldt, & Zahn, 2011; Rockel, 2015; Rummukainen, 2010). An example of the added value of RCMs was presented by Gómez-Navarro et al. (2011). They showed that model results over the Iberian Peninsula for the last millennium improved considerably when the horizontal resolution of the atmospheric model was refined compared to the driving global model and when reanalysis data (a combination of model results and observations) were provided at the lateral boundaries.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 7. Model Hierarchy Used for Dynamical Downscaling.

Source: H. E. Markus Meier, IOW and SMHI.

Note. Consists of a global GCM or ESM, an RCM, a land surface model (LSM), a BSM, scenarios for radiative forcing according to RCPs or the SRES and nutrient load scenarios into the Baltic Sea.

Emission Scenarios

The various scenario simulations for the Baltic Sea that were developed over time are not comparable because they follow several different assessment reports of the IPCC (IPCC, 2001, 2007, 2013). For the regional projections, global model results of the Coupled Model Intercomparison Projects CMIP2, CMIP3, and CMIP5 were utilized. Not only have the GCMs been improved, but the underlying greenhouse gas (GHG) emission scenarios have also changed. While the first projections just accepted a scenario with 150% increased equivalent CO2 concentration in the atmosphere in the future compared to the historical time slice (Räisänen, Rummukainen, & Ullerstig, 2001; Rummukainen et al., 2001), more elaborate GHG emission scenarios have been developed for the Third and Fourth Assessment Reports following the Special Report on Emission Scenarios, SRES (Nakićenović et al., 2000). These scenarios were derived from differing assumptions on worldwide, socioeconomic development leading to differing GHG emissions, and assume either regionalization or globalization and either emphasis on human wealth or emphasis on sustainability and equity. In the Fifth Assessment Report (IPCC, 2013), representative concentration pathways (RCPs) were introduced, prescribing the future evolution of GHG concentrations in the atmosphere. These concentrations were chosen such that for the RCPs 2.6, 4.5, 6.0, and 8.5, the corresponding radiative forces at the end of the century amount to 2.6, 4.5, 6.0, and 8.5 Wm-2, respectively (Moss et al., 2010). Hence, scenario simulations driven by SRES (CMIP3) or RCPs (CMIP5) are not comparable because GHG concentrations differ (Knutti & Sedláček, 2013; their Fig. 1).

Nutrient Load Scenarios

Projections of future marine biogeochemical cycling require the definition of nutrient load scenarios. Many studies utilize best-case, reference, and worst-case scenarios (e.g., Meier, Andersson, et al., 2011) that significantly differ among themselves (Meier, Edman, et al., 2018). Saraiva et al. (2019a, 2019b) used global shared socioeconomic pathways (SSPs; O’Neill et al., 2014) that were downscaled to the Baltic Sea (Zandersen et al., 2018) and assigned to nutrient load abatement strategies following the so-called Baltic Sea Action Plan (BSAP, see HELCOM, 2013), reference conditions (REF) representing the average loads of the 2010–2012 period, and a business-as-usual (BAU) or worst-case scenario.

The BSAP scenario assumes the implementation of countrywide maximum allowable input until 2021 (HELCOM, 2013). After 2020, nutrient loads will remain constant until the end of the 21st century. Under REF it is assumed that no socioeconomic changes will occur, that is, land and fertilizer usage, soil properties, and sewage water treatment in each sub-basin will not change over time. Only the impacts of the changing climate on air temperature and precipitation and consequently on river runoff are considered. Under REF, atmospheric deposition is also assumed to be constant in time. Finally, the BAU or worst-case scenario assumes changes caused by a “fossil-fuelled development” scenario following SSP5 (O’Neill et al., 2014) coupled to increasing river runoff. Changes in nitrogen and phosphorus loads were calculated from the regional assumptions, for example, on population growth, changes in agricultural practices such as land and fertilizer use, and expansion of sewage water treatment plants (Zandersen et al., 2018).

Projections of biogeochemical cycles for the 21st century considerably depend on the nutrient load scenario (Meier, Edman, et al., 2018).

Coupled Regional Modeling in BALTEX and Baltic Earth

The development of RCMs and in particular regional coupled atmosphere–sea ice–ocean models of the Baltic Sea region was fostered by BALTEX (the Baltic Sea Experiment; Raschke et al., 2001) and its successor program Baltic Earth (Earth System Science for the Baltic Sea Region; Meier, Rutgersson, & Reckermann, 2014).

In 1992, BALTEX was launched as a continental-scale experiment (CSE) of the Global Energy and Water Cycle Experiment (today Global Energy and Water Exchanges, GEWEX) within the World Climate Research Program (WCRP). BALTEX research focused on the water and energy cycle in the Baltic Sea region, assessing the fluxes between atmosphere, ocean, sea ice, and land surface, the so-called BALTEX box (e.g., Omstedt, Elken, et al., 2014; Reckermann et al., 2011). Within BALTEX, the first regional, coupled atmosphere–sea-ice–ocean models were developed to improve short-range weather forecasting (e.g., Gustafsson, Nyberg, & Omstedt, 1998) or to study processes and the impact of coupling on air–sea exchange (e.g., Döscher et al., 2002; Hagedorn, Lehmann, & Jacob, 2000; Schrum, Hübner, Jacob, & Podzun, 2003).

In the first simulation of a fully coupled model without flux corrections, Hagedorn et al. (2000) showed that simulated SSTs agreed well with satellite observations during a six-month period. Simulated SSTs were at least as good as the previously used SSTs from operational analyses and in some cases even better. Hagedorn et al. (2000) found that vertical air–sea fluxes are superimposed by horizontal advection and that only under specific conditions do atmospheric variables show a significant response to the air–sea coupling. Kjellström, Döscher, and Meier (2005) showed in multi-year simulations that during summer the impact of the lower boundary condition on near surface atmospheric fields and atmosphere–ocean fluxes is larger when horizontal advection is smaller, for example during years with a negative North Atlantic Oscillation (NAO) index, that is, with a weaker large-scale flow over the Baltic Sea region from more northerly directions.

In 2013, after about 20 years of successful scientific networking, BALTEX was terminated and a new program, Baltic Earth, was launched, focusing on Earth system science for the Baltic Sea region (Meier, Rutgersson, et al., 2014). In contrast to the old program, Baltic Earth considers the entire regional Earth system, including carbon, nutrient and pollutant cycling in the atmosphere, land surface and ocean components, and the anthroposphere. In this context, within Baltic Earth a working group was formed to foster regional climate system models (RCSMs) that take (as a vision) all relevant components of the Earth system and the fluxes, interactions, and feedbacks between the components into account. The working group co-organized inter alia two international workshops in Rome (2015) and in Palma de Mallorca (2018). Together with scientists from HyMeX (the Mediterranean counterpart to BALTEX) and Med-CORDEX (a subproject of Coordinated Regional Climate Downscaling Experiment (CORDEX) focusing on Mediterranean climate variability and trends) current issues in regional climate system modeling, in particular the coupling between the atmosphere and the ocean, were discussed.

Although the possibilities of such a working group on regional climate system modeling are of course limited due to a lack of funding, the research community followed the science plan of BALTEX and Baltic Earth. In recent years, coupled atmosphere–sea-ice–ocean models have been further elaborated by adding sub-models to the RCSMs. For instance, sub-models for surface waves (e.g., Rutgersson, Nilsson, & Kumar, 2012), land vegetation (e.g., Smith, Samuelsson, Wramneby, & Rummukainen, 2011), hydrology and land biochemistry (e.g., Arheimer, Dahné, & Donnelly, 2012; Meier, Müller-Karulis, et al., 2012), marine biogeochemistry (e.g., Meier, Andersson, et al., 2011; Meier, Hordoir, et al., 2012; Neumann, 2010; Omstedt et al., 2012; Saraiva et al., 2019a, 2019b), the marine carbon cycle (e.g., Omstedt et al., 2012), and food web modeling (e.g., Bauer et al., 2018; Niiranen et al., 2013) were added to the RCSMs. The aim of many of these studies was to develop RCSMs that represent regional dynamical feedback mechanisms such as the ice–albedo feedback (e.g., Meier, Höglund, et al., 2011), by including interactive coupling between the regional climate system components (i.e., atmosphere, ocean, sea ice, hydrology, land vegetation, marine biogeochemistry).

Recently, several high-resolution fully coupled atmosphere–sea-ice–ocean–land-surface models for the Baltic Sea region, which allow for consideration and resolution of local feedbacks, have been developed and applied in regional climate studies (e.g., Dieterich et al., 2013; Dieterich et al., 2019; Gröger, Dieterich, Meier, & Schimanke, 2015; Gröger et al., 2019; Ho-Hagemann et al., 2017; Sein et al., 2015; Tian et al., 2013; Van Pham, Brauch, Dieterich, Frueh, & Ahrens, 2014; Wang et al., 2015).

National Regional Climate Modeling Programs

Whereas BALTEX and Baltic Earth focus on process understanding, at the end of the 1990s national climate modeling programs in Denmark (Christensen, Christensen, Machenhauer, & Botzet, 1998), Germany (Jacob & Podzun, 1997), Norway (Bjørge & Haugen, 1998), and Sweden (Rummukainen et al., 2000) were launched to perform scenario simulations with high-resolution RCMs inter alia to better describe processes related to the Scandinavian mountains and the Baltic Sea. These programs aimed to produce climate information for the society and fostered the development of RCSMs.

Within the European Union-funded PRUDENCE project, the first coordinated multi-RCM intercomparison of scenario simulations for the Baltic Sea region was performed (Christensen & Christensen, 2007). Later, similar activities were continued within the ENSEMBLES (van der Linden & Mitchell, 2009) and EURO-CORDEX initiatives (Jacob et al., 2014). In the latter project, an ensemble of very high-resolution (12.5 km) RCM simulations for Europe was produced.

Within the Swedish Regional Climate Modeling Program (SWECLIM), multi-year coupled regional atmosphere–sea-ice–hydrology model simulations were performed (Döscher et al., 2002) and a mini-ensemble of scenario simulations for the end of the century was developed (Räisänen et al., 2004; see also Table 1). Both regional climate modeling projects, PRUDENCE and SWECLIM, were very important for gaining knowledge about climate variability in the Baltic Sea region because it was learned that projections of Baltic Sea SSTs need coupled atmosphere–ocean models and that multi-annual simulations with coupled models could be performed without artificial drift or other technical problems. In fact, the SWECLIM (1997–2003) was particularly innovative because of strong, complementing oceanographic and hydrological research components in addition to well-established atmospheric research leading to the development of an advanced, coupled atmosphere–—sea-ice–ocean–hydrology RCSM (Döscher et al., 2002; Rummukainen, Bergström, Persson, Rodhe, & Tjernström, 2004). In this program the dynamical downscaling approach was further developed. Whereas the dynamical downscaling approach was first applied to limited-area atmospheric models only (Giorgi & Mearns, 1991), the SWECLIM coupled model included either a process-oriented (Rummukainen et al., 2001) or an ocean circulation model for the Baltic Sea (Räisänen et al., 2004).

Table 1. Selected Ensembles of Scenario Simulations for the Baltic Sea Carried Out in International Projects

Project

Swedish Regional Climate Modeling Program

Advanced Modeling Tool for Scenarios of the Baltic Sea ECOsystem to SUPPORT Decision-Making

Holocene Saline Water Inflow Changes into the Baltic Sea, Ecosystem Responses, and Future Scenarios

Building Predictive Capability Regarding the Baltic Sea Organic/Inorganic Carbon and Oxygen Systems

Well-being from the Baltic Sea—Applications Combining Natural Science and Economics

Impacts of Climate Change on Waterways and Navigation

Acronym

SWECLIM

ECOSUPPORT

INFLOW

Baltic-C

BalticAPP

KLIWAS

Duration

1997–2003

2009–2011

2009–2011

2009–2011

2015–2017

2012–2013

Project summaries

Rummukainen et al. (2004)

Meier, Andersson, et al. (2014)

Kotilainen et al. (2014)

Omstedt, Humborg, et al. (2014)

Saraiva et al. (2019a)

Bülow et al. (2014)

GCMs

AR3

AR4

AR4

AR4

AR5

AR4/AR5

RCSM

RCAO

RCAO

RCAO

RCA

RCA4-NEMO

REMO-MPIOM, REMO-HAMSOM, RCA4-NEMO

Horizontal resolution atmosphere/ocean

50 km/10.8 km

25 km/3.6 km

25 km/3.6 km and 50 km/3.6 km for paleoclimate

25 km/horizontally integrated

25 km/3.6 km

varying

Period(s)

1961–1990 and 2071–2100

1961–2099

1961–2099 and 950–1800

1960–2100

1976–2100, improved initial conditions

1961–2099

Ocean model

One physical BSM

Three physical-biogeochemical BSMs

See ECOSUPPORT

One physical-biogeochemical BSM including the carbon cycle

One physical-biogeochemical BSM

Two physical regional models with focus on the Baltic Sea and North Sea regions and one physical-biogeochemical ocean model

References

Döscher and Meier (2004), Meier et al. (2004a, b)

Meier, Höglund, et al. (2011); Neumann et al. (2012)

See ECOSUPPORT, Schimanke and Meier (2016)

Omstedt et al. (2012)

Saraiva et al. (2019a, 2019b)

e.g., Dieterich et al. (2019), Gröger et al. (2019)

Note. AR = IPCC Assessment Report, GCM = General Circulation Model, RCSM = Regional Climate System Model, RCAO = Rossby Centre Atmosphere Ocean model, RCA = Rossby Centre Atmosphere model, RCA4-NEMO = Rossby Centre Atmosphere model—Nucleus for European Modelling of the Ocean.

The first SWECLIM scenario simulations were carried out for selected time slices in present and future climates (Haapala, Markus Meier, & Rinne, 2001; Meier, 2002a, 2002b; Omstedt, Gustafsson, Rodhe, & Walin, 2000; Rummukainen et al., 2001). However, the simulations by Omstedt et al. (2000), Haapala et al. (2001), Rummukainen et al. (2001), and Meier (2002a, 2002b) were based on a single GCM and a single GHG concentration scenario (150% increase in equivalent CO2 concentration in the atmosphere in future climate compared to historical climate) and only covered 10-year time slices.

After these very first attempts, a mini-ensemble of scenario simulations with the Rossby Centre Atmosphere Ocean (RCAO) model (Döscher et al., 2002), driven by two GCMs and two GHG emission scenarios based upon IPCC’s Third Assessment Report (IPCC, 2001), were performed (Döscher & Meier, 2004; Meier, Döscher, & Halkka, 2004; Meier, Broman, et al., 2004a; Räisänen et al., 2004; see also Table 1). To improve the statistical significance level, the length of the time slices was extended to 30 years. However, for salinity in the Baltic Sea these periods are still too short (see the discussion by Meier, 2002a). As salinity conditions at the beginning of the time slices are unknown, any chosen initial salinity profile would adapt to the specific atmospheric and hydrological forcing on time scales of about 20–40 years (see “Introduction”), which is as long as the length of the time slices. This artificial drift would overshadow the physically based results of the simulations.

To overcome this limitation of the time-slice approach, for the control experiment the ocean model was forced with observed river flow and reconstructed atmospheric surface fields for the period 1903 to 1998 (Kauker & Meier, 2003; Meier & Kauker, 2003a, 2003b). In the scenario simulations, the same initial conditions and forcing functions as in the control simulation were used as a baseline. In addition, 30-year climatological monthly mean changes of projected atmospheric surface fields and runoff from the earlier performed time-slice scenario simulations by Räisänen et al. (2004) were added to the atmospheric and hydrological forcing, respectively (Meier, 2006; Meier, Kjellström, & Graham, 2006; Meier, Höglund, et al., 2011). Hence, the latter studies assumed that only the seasonal cycle may change in the future climate compared to present climate but that the inter-annual variability will remain unchanged compared to the baseline.

State-of-the-art projections are not based on the time-slice approach anymore. Instead, continuously integrated transient simulations from present to future climates are performed, even including biogeochemical modules (Eilola, Mårtensson, & Meier, 2013; Friedland, Neumann, & Schernewski, 2012; Meier, Andersson, et al., 2011, 2012; Meier, Eilola, & Almroth, 2011; Meier, Hordoir, et al., 2012; Meier, Müller-Karulis, et al., 2012; Neumann, 2010; Neumann et al., 2012; Ryabchenko et al., 2016; Saraiva et al., 2019a, 2019b; Skogen et al., 2014).

Since the early 21st century, transient simulations for the period 1960–2100 using regional ocean (Holt et al., 2016; Pushpadas, Schrum, & Daewel, 2015) and regional coupled atmosphere–ocean models (Bülow et al., 2014; Dieterich et al., 2019; Gröger et al., 2019) have also been available for the combined Baltic Sea and North Sea system.

Ensemble Modeling

To address uncertainties caused by global and regional models’ biases, unknown GHG emission scenarios, and natural variability, ensembles of scenario simulations were carried out and analyzed with respect to their signal-to-noise ratios. The latter are the ratios between the ensemble mean changes between future and present climates and the ensemble spreads (or standard deviations) of the changes of all ensemble members. For instance, Meier, Andersson, et al. (2011, 2012), Meier, Müller-Karulis, et al. (2012), and Neumann et al. (2012) investigated projected climate change calculated with three different Baltic Sea ecosystem models, two GCMs, two realizations of global climate simulations, and two GHG emission scenarios following Nakićenović et al. (2000). Within the ECOSUPPORT project (Meier, Andersson, et al., 2014) uncertainty ranges in regional projections for the Baltic Sea ecosystem were introduced (Table 1). A similar ensemble approach, taking even more GCMs into account, was chosen by Omstedt et al. (2012) and Saraiva et al. (2019b). Within the BalticAPP project (e.g., Saraiva et al., 2019a, 2019b) the ensemble approach for regional projections was further refined by using updated versions of GCMs and RCMs, RCPs instead of SRES scenarios, nutrient load scenarios following regionalized SSPs, and an improved experimental setup (Table 1). A relatively big ensemble of scenario simulations using a RCSM driven by eight GCMs and three RCPs was carried out by Dieterich et al. (2019) (Table 1).

Model Performance

For the quantification of the performance of Baltic Sea models (BSMs), hindcast and climate simulations of the historical period need to be distinguished. The first type of simulations are forced with atmospheric reanalysis or forecast data that might be regarded as “perfect” atmospheric forcing. Hence, biases compared to observations would solely be caused by the lack of understanding of unresolved ocean processes and by shortcomings in the numerical implementation of the ocean model. The results of hindcast simulations are close to observations (Placke et al., 2018; see also Figure 8). However, the impact of atmospheric forcing biases is big. Figure 8 illustrates the differences in physical and biogeochemical ocean variables between simulations with the same ocean model but forced with various atmospheric data sets of different quality.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 8. Simulated Temperature, Salinity, Nitrate Concentration, Oxygen Concentration, and Phosphate Concentration at Gotland Deep, 1980–2009.

Source: Sofia Saraiva, SMHI and MARETEC.

Note. Simulated temperature (in °C), salinity (in g kg-1), nitrate concentration (in mmol N m-3), oxygen concentration (in mL L-1), and phosphate concentration (in mmol P m-3) for the period 1980 and 2009 at the monitoring station Gotland Deep (BY15), at the surface (left panels) and in 200 m depth (right panels). In addition, observations from the long-term monitoring programs (red dots) are depicted. Three simulations driven with various state-of-the-art atmospheric datasets are shown: HiResAFF (black lines), ERA40 (green lines), and EURO4M (orange lines). As the EURO4M atmospheric dataset is closest to observations, the results shown in Figures 2 to 5 are taken from this simulation. The location of BY15 is shown in Figure 1. The figure illustrates ocean model biases and how sensitive the ocean variables are to the atmospheric forcing.

The second type of simulations are climate simulations. The performance of ocean model simulations forced with regionalized climate models compared to observations is less good because of shortcomings of the forcing data and by natural variability that cannot be reproduced. Meier, Edman, et al. (2018) found that compared to monitoring observations the performance of model simulations during the historical climate differs considerably. Although they focused on uncertainties of simulated biogeochemical cycles, the skills of water temperature and salinity in historical climate and the spread of projections in future climate were assessed as well. Model results were regarded as acceptable when the mean biases during the control period are less than two standard deviations of the observations on average. Overall, temperature profiles at monitoring stations in the sub-basin centers were acceptably simulated, although the quality in the northern sub-basins is lower compared to the quality in the Baltic proper, probably due to the amplifying effect of the ice–albedo feedback on air temperature biases and other feedback mechanisms.

In general, BSMs have relatively bigger problems with the simulation of salinity, probably because of insufficient horizontal and vertical resolutions causing spurious numerical diffusion across the strong horizontal and vertical gradients in salinity that are typical for the Baltic Sea (Rennau & Burchard, 2009). During the historical period, several of the BSMs investigated by Meier, Edman, et al. (2018) overestimated SSS and underestimated the depth of the halocline in the Baltic proper. The halocline is a vertical layer with a pronounced gradient in salinity separating the brackish surface layer from the more saline deep water. In the northern Baltic Sea, all models tended to overestimate the stratification and some models were too saline on average. These biases are very likely shortcomings of the ocean models and not of the atmospheric or hydrological forcing because even under perfect atmospheric boundary and river runoff conditions, that is, under hindcast simulations driven by regionalized atmospheric reanalysis data, similar results were found (Placke et al., 2018). Compared to the standard deviation of the observations, salinity biases of the investigated models increased systematically from south to north (Placke et al., 2018).

Nevertheless, the results by Meier, Edman, et al. (2018) suggested that projected changes in temperature and salinity at the end of the century are larger than the ensemble spread.

Uncertainties in Projections

According to Meier, Edman, et al. (2018, p. 3), “uncertainty is defined as the spread in future projections within an ensemble of scenario simulations expressed by the standard deviation of mean changes.” The uncertainties of state-of-the-art scenario simulations for the Baltic Sea were analyzed by Saraiva et al. (2019b), Meier, Edman, et al. (2018), and Meier et al. (2019), based upon relatively large suites of regional and global climate models. Saraiva et al. (2019b) selected five climate indices, that is, volume-averaged water temperature, salinity, primary production, nitrogen fixation, and hypoxic area, and analyzed the spread of their projections at the end of the century caused by the GCMs, RCPs, nutrient loads, and global mean sea level rise. They found that the largest source of uncertainty in temperature projections is the choice of the GHG concentration scenario. For salinity, the largest sources are the assumed sea level rise of 1 m above normal at the lateral boundary in Kattegat and the biases of the GCMs. A sea level rise of 1 m would significantly increase the salt transport between the North Sea and the Baltic Sea. Consequently, salinity and the vertical stratification in the Baltic Sea would be larger at the end of the 21st century compared to scenarios with no global mean sea level rise (Meier, Höglund, Eilola, & Almroth-Rosell, 2017; Saraiva et al., 2019b). Biases of the GCMs affect air temperature and precipitation changes in the Baltic Sea region as simulated by the RCMs. According to Saraiva et al. (2019b), the projected runoff changes vary between 1 and 21% in RCP 4.5 and between 6 and 20% in RCP 8.5 scenario simulations. These numbers are close to earlier projections by Meier, Müller-Karulis, et al. (2012), who calculated runoff changes simulated by a hydrological model under the GHG emission scenarios A1B and A2. In addition to the uncertainties in the projected changes of the hydrological cycle, the response of the halocline depth to changes in river runoff differs between BSMs (Meier, Höglund, et al., 2011). These model deficiencies contribute, for instance, to the smallest signal-to-noise ratios in bottom oxygen concentration changes at the depth of the halocline (Meier, Andersson, et al., 2011). Oxygen concentrations decrease below the halocline because of the reduced ventilation of the Baltic Sea deep water. A projected deepening of the halocline would lead to an increase in bottom oxygen concentrations in areas that in present climate are located below the halocline and that will be located above the halocline in future.

Meier et al. (2019) claimed that unknown current and future bioavailable nutrient loads from land and atmosphere and the experimental setup of the dynamical downscaling technique are the largest sources of uncertainties for projections of the Baltic Sea biogeochemical cycling. The latter potential source of uncertainty in scenario simulations was not considered until then. Most of the available scenario simulations started with initial conditions in the 1960s or 1970s. In some cases, long spin-up simulations (starting 1850) with the same BSM, which was also used for the future projections, were performed to take the long turnover time scale of the water column and sediment in the Baltic Sea into account (e.g., Saraiva et al., 2019a). However, regionalized and homogenized atmospheric forcing is not available for the entire 1850–2100 period. Hence, the switch in the 1960s or 1970s from one atmospheric dataset to another one may cause artificial drifts for at least 30 years that will overshadow climate-induced changes of the Baltic Sea system.

Idealized Sensitivity Experiments

As the dynamical downscaling approach is rather complicated, idealized sensitivity studies have been performed to get an idea of the system response in future climate (e.g., Hordoir & Meier, 2012; Hordoir, Höglund, Pemberton, & Schimanke, 2018; Meier, 2005; Meier et al., 2017; Schrum, 2001). These studies are very useful for understanding the governing processes that may explain potential changes under climate change. However, sensitivity experiments are not projections because in the latter several drivers may change in a rather complex way while in sensitivity experiments simplified changes are investigated in just one driver at a time.

Results of Recent Projections

According to the BACC II Author Team (2015), the number of available scenario simulations for the Baltic Sea has increased since the first assessment (BACC Author Team, 2008), and more detailed information from multi-model ensemble studies is now available. However, the number of regional studies is still limited and much smaller than the number of global climate studies, restricting this overview article mainly to the work of the authors’ group.

The two recent ensembles of scenario simulations by Saraiva et al. (2019a, 2019b), driven by CMIP5 GCMs and RCP 4.5 and RCP 8.5 GHG concentration scenarios (henceforth called BalticAPP scenario simulations), are compared with earlier projections by Meier, Andersson, et al. (2011) and Meier, Müller-Karulis, et al. (2012), driven by CMIP3 GCMs (henceforth called ECOSUPPORT scenario simulations) (Table 1). Results of the latter simulations were summarized and assessed by the BACC II Author Team (2015). The BACC II report included only climate studies published until 2012. The BalticAPP scenario simulations by Saraiva et al. (2019a, 2019b) are based upon: (1) improved initial conditions in 1976; (2) avoidance of bias corrections except for wind speed and mean runoff; (3) revised, more plausible nutrient load scenarios; (4) GHG concentration scenarios RCP 4.5 and RCP 8.5; and (5) four driving GCMs of CMIP5. Hence, uncertainty analyses estimated from the ensemble spread are possible and were performed by Saraiva et al. (2019b) and Meier, Edman, et al. (2018). Another novel aspect of the BalticAPP projections was that nutrient load scenario simulations in present climate were carried out and compared with scenario simulations in future climate (Saraiva et al., 2019a).

Water Temperature

The BACC II Author Team (2015) concluded that water temperature would significantly increase at the end of the 21st century compared to the historical climate defined as a 30-year period in the 1980s to 2000s (Figure 9). The amount of the warming would depend on the GHG emission scenario. The increase in SST is projected to be largest in the northern Baltic Sea during early summer, very likely due to the ice–albedo feedback causing earlier warming during the melting season (Figure 9). No changes in SST would occur in regions that remain on average ice-covered also in future climate and where the SST is equal to the freezing point temperature in both present and future climates. The surface layer would warm more than the deep layer (not shown). Hence, the thermally induced, vertical stratification in the surface layer (i.e., the so-called thermocline) would increase and might lead to reduced vertical nutrient fluxes from the bottom into the biologically productive surface mixed layer (Gröger et al., 2019). In addition to the ice–albedo feedback, a decrease in thermal convection (Hordoir & Meier, 2012) might also contribute to the earlier warming in the northern Baltic Sea. However, the latter was not identified as the dominating driver of amplified northern warming.

Summer SST changes in ECOSUPPORT and BalticAPP scenario simulations show similar patterns (Figure 9). A robust feature among all projections is that the greatest warming occurs in the entire northern Baltic Sea (Bothnian Sea and Bothnian Bay) during summer while the smallest warming takes place in the coastal zone of the Bothnian Bay during winter (not shown). The ice cover may explain this seasonality. The warming increases in the following order: BalticAPP RCP 4.5, ECOSPPORT, and BalticAPP RCP 8.5 (Table 2). This order is explained by the different GHG emission scenarios used in CMIP3 and CMIP5. The GHG emission scenario A1B (used in ECOSUPPORT) results in a regional warming in between the GHG concentration scenarios RCP 4.5 and RCP 8.5 (used in BalticAPP). Hence, the discrepancy in regional projections is explained by the differing assumptions used in the Fourth and Fifth IPCC Assessment Reports (IPCC, 2007, 2013). Probably due to the proximity to the Arctic with amplified warming and due to the semi-enclosed location, the Baltic Sea is warming more than the North Sea (Holt et al., 2016; Pushpadas et al., 2015).

Table 2. Ensemble Mean Changes in SST (in °C)

Δ‎ SST

DJF

MAM

JJA

SON

Annual Mean

ECOSPPORT

SRES A1B

2.5

2.8

2.8

2.5

2.6

BalticAPP RCP 4.5

1.7

1.9

2.0

1.8

1.8

BalticAPP RCP 8.5

2.9

3.2

3.3

3.0

3.1

Note. Shown are changes in ECOSUPPORT, BalticAPP RCP 4.5, and BalticAPP RCP 8.5 scenario simulations averaged for the Baltic Sea, including the Kattegat. DJF = December, January, February, MAM = March, April, May, JJA = June, July, August, SON = September, October, November.

Salinity

Salinity, both at the sea surface and bottom, is projected to decrease rather uniformly in the entire Baltic Sea basin, slightly more in the stratified southern sub-basins and less in the north (Meier, 2015; see also Figure 9). The reason for the decrease in salinity is mainly the projected increased river runoff. However, the spread in projections of the hydrological cycle is substantial, although most of the projections point toward increased total river flow (Meier, Müller-Karulis, et al., 2012). Nevertheless, the BACC II Author Team (2015) concluded that it is unclear whether salinity in the Baltic Sea will increase or decrease.

In all ECOSUPPORT and BalticAPP scenario simulations, SSS changes are spatially rather uniform (Figure 9), with a small seasonal cycle (not shown). The freshening in the ECOSUPPORT scenario simulations is in terms of annual mean SSS changes (Figure 9) averaged over the Baltic Sea more than twice as large as in the BalticAPP scenario simulations (Table 3) because of a larger change in river runoff (cf. Meier, Müller-Karulis, et al., 2012; Saraiva et al., 2019b). Bottom salinity changes show similar patterns to the SSS changes and are about as large as the SSS changes (Table 3).

Table 3. Ensemble Mean Changes in Annual Mean SSS (in g kg-1), Annual Mean Bottom Salinity (BS) (in g kg-1), and Winter Mean Sea Level (SL) (in cm)

Annual Changes

ECOSUPPORT

BalticAPP RCP 4.5

BalticAPP RCP 8.5

Δ‎ SSS

-1.5

-0.7

-0.6

Δ‎ BS

-1.6

-0.6

-0.6

Δ‎ SL

5.5

0.4

3.7

Note. Shown are changes in ECOSUPPORT, BalticAPP RCP 4.5, and BalticAPP RCP 8.5 scenario simulations averaged for the Baltic Sea, including the Kattegat.

Vertical Density Gradients

A reduced overall salinity would cause a reduced stratification between surface and deep layer in the Baltic Sea, causing an improved ventilation of the bottom layer and, in some sub-basins such as the Gulf of Finland, even an increase in bottom oxygen concentration (Meier, Andersson, et al., 2011; see also Figure 9). Changes in the strength of the low-frequency component of the wind over the Baltic Sea region, preferably blowing from westerly directions, may cause changes in salinity as well (Meier & Kauker, 2003a; Meier et al., 2006). However, systematic changes in wind between future and present climates are statistically insignificant except over regions with declining sea-ice cover (Meier, Höglund, et al., 2011). The latter changes are caused by reduced stratification in the planetary boundary layer due to the warmer sea surface in future climate.

Salt Water Inflows

Neglecting the impact of a global mean sea level rise, there is no evidence of changes in future salt water inflows into the Baltic Sea (Meier, Hordoir, et al., 2012), just as there are no indications of a systematic change in the past inflow frequency (Mohrholz, 2018). Between 1887 and 2017, a pronounced multi-decadal variability of large inflows with a main period of 25–30 years prevents the detection of any centennial trend (Mohrholz, 2018). However, short-term sea level pressure variations favorable for salt water inflows may increase slightly in the future (Schimanke, Dieterich, & Meier, 2014). In future high-end global mean sea level projections, reinforced salt water inflows lead to higher salinity and increased vertical stratification compared to 20th-century conditions. Meier et al. (2017, p. 163) found that, “contrary to intuition, reinforced ventilation of the deep water does not lead to overall improved oxygen conditions but causes instead expanded dead bottom areas accompanied with increased internal phosphorus loads from the sediments and increased risk for cyanobacteria blooms.”

Sea Ice

Changes in sea-ice extent and sea-ice thickness would mainly follow the changes in winter air temperature over the Baltic Sea. All available scenario simulations suggested a drastic decrease in sea-ice cover in the future (Meier, 2015). However, even the high-end scenarios suggested that, at the end of the century, there would still be sea ice in the northern Baltic Sea. The increased wind speed over sea areas with disappearing sea-ice cover might, however, increase wave heights, mixing, and resuspension, and the improved light conditions may trigger an earlier onset of the spring bloom (Eilola et al., 2013).

Sea Level

Low-frequency changes in the Baltic sea level will approximately follow the corresponding changes in global mean sea level, albeit somewhat slower because of land uplift (Grinsted, 2015; see also Figure 9). Global, and thus regional, sea levels are rising because of the thermal expansion of sea water and melting ice sheets and glaciers on a global scale (Church & White, 2006; see also Figure 6). This effect is called eustatic sea level rise. From global scenario simulation results, it is expected that global mean sea level rise very likely will be accelerated in future. For the 2081–2100 period relative to 1986–2005, the rise was projected to be in the range between 26 and 82 cm (IPCC, 2013). In medium (RCP 4.5) and high-end scenarios (RCP 8.5), the mean sea level is, at the end of the century, 0.47 and 0.63 m higher than during the reference period, respectively. In the Baltic Sea, land uplift, particularly in the northern Baltic Sea, would partly counteract the projected eustatic sea level rise due to the glacial isostatic adjustment after the last glaciation of Fennoscandia (Grinsted, 2015). Today the maximum land uplift is found in Bothnian Bay close to the Swedish city Luleå and amounts to about 1 m per century (e.g., Harff, Furmańczyk, & Von Storch, 2017; Hill, Davis, Tamisiea, & Lidberg, 2010). Both global and regional future sea level projections are rather uncertain due to the unknown eustatic sea level rise (IPCC, 2013).

In addition to the global sea level rise, some climate models suggested that the predominantly southwesterly wind in the Baltic Sea region would increase, causing elevated sea levels in the Bothnian Bay and Gulf of Finland during winter, especially in the northern and eastern coastal zone of these sub-basins, respectively (Meier, 2015). Further, some scenario simulations showed that the seasonal water level changes are actually greatest in spring (March to May), probably because of the melting sea ice and the related increase in sea level due to the Archimedes’ principle (Meier, 2015, figure 13.7). Overall, the global sea level rise is believed to be a greater risk for increasing sea level extremes than increasing wind speed (BACC II Author Team, 2015).

In BalticAPP scenario simulations, sea level changes are, compared to the other seasons, largest in winter and increase toward the northern and eastern Baltic Sea (Figure 9). On the other hand, sea level changes in ECOSUPPORT scenario simulations are largest in spring, because one member of the multi-model ensemble considered Archimedes’ principle. Sea levels increase in the following order: BalticAPP RCP 4.5, BalticAPP RCP 8.5, and ECOSPPORT (Table 3). Note that the global mean sea level rise and land uplift are not included here and have to be added (Meier, 2006; Meier, Broman, et al., 2004a). The sea level changes depicted in Figure 9 consider only changing river runoff, changing wind, and melting sea ice affecting the sea level via Archimedes’ principle (only in the ECOSUPPORT ensemble).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 9. Changes of Summer Mean SST, Annual Mean SSS, Annual Mean Bottom Salinity, and Winter Mean Sea Level, 1978–2007 and 2069–2098.

Source: H. E. Markus Meier, IOW and SMHI.

Note. From left to right changes of summer (June, July, and August) mean SST (°C), annual mean SSS (g kg-1), annual mean bottom salinity (g kg-1), and winter (December, January, and February) mean sea level (cm) between 1978 and 2007 and 2069 and 2098 are shown. From top to bottom, results of the ensembles ECOSUPPORT (white background), BalticAPP RCP 4.5 (grey background), and BalticAPP RCP 8.5 (grey background) are depicted.

Marine Biogeochemistry

The projected changes in physical variables would have big impacts on the marine biogeochemistry because of enhanced nutrient cycling and reduced air–sea fluxes of oxygen due to the warming (Schneider, Eilola, Lukkari, Müller-Karulis, & Neumann, 2015). However, the dominating impact is expected from changes in nutrient loads from the air and from the catchment (Meier, Edman, et al., 2018; Meier et al., 2019; Saraiva et al., 2019a, 2019b). Although uncertainties of available scenario simulations are large, the implementation of nutrient load abatement scenarios such as the BSAP might result in improved environmental conditions in the Baltic Sea despite the counteracting impact of changing climate (Meier, Edman, et al., 2018). The impact of climate change is larger in case of higher nutrient loads (Saraiva et al., 2019a).

Projected oceanographic changes might have impacts on phytoplankton blooms, water transparency, and oxygen concentrations in the Baltic Sea deep water (e.g., Meier, Müller-Karulis, et al., 2012). In the euphotic zone (the uppermost layer that is exposed to intense sunlight), phytoplankton growth is controlled by light conditions and by the depth of the mixed layer (Sverdrup, 1953). Increasing turbulence mixes nutrients from the deep water into the surface layer, enabling increased biological production. However, in greater depth light conditions are getting worse, limiting phytoplankton growth. Hence, net production in the mixed layer will exceed losses if the mixed layer is shallower than a critical depth defined by the light levels, phytoplankton growth, and loss. The latter is called the Sverdrup’s critical depth hypothesis (Sverdrup, 1953).

Further, higher water temperatures will reduce the air–sea fluxes of oxygen and speed up phytoplankton growth and remineralization, causing an intensified nutrient cycling in the water column (e.g., Meier, Andersson, et al., 2011). Higher temperatures may also increase the phosphorus flows from the sediment into the water column (e.g., Eilola, Almroth-Rosell, & Meier, 2014; Eilola et al., 2012). Enlarged remineralization of precipitated dead organic material may increase oxygen consumption and reinforce phosphorus fluxes from the sediments in case of anoxic conditions. Hence, the accumulated organic material in the sediments may form an internal source of phosphorus, amplifying eutrophication.

Projected enlarged river runoff might increase the external nutrient loads from land, reinforcing future eutrophication (e.g., Meier, Andersson, et al., 2011). However, in the northern Baltic Sea phytoplankton production may actually be reduced in future climate due to the inflow of increased river-borne organic matter (Andersson et al., 2015). The growth of heterotrophic bacteria will be favored by river-borne organic matter increases compared to phytoplankton production. Observed changes may support this future scenario. Increased water temperatures and increased organic matter supply caused the observed decline in oxygen concentrations in the Bothnian Sea during the 2000s (Ahlgren, Grimvall, Omstedt, Rolff, & Wikner, 2017). The projected sea-ice decline might improve light conditions, causing an earlier onset of the spring bloom (Eilola et al., 2013).

Using again the three ensembles ECOSUPPORT, BalticAPP RCP 4.5, and BalticAPP RCP 8.5, the combined impacts of future oceanographic changes and nutrient load scenarios on selected biogeochemical variables are compared. These biogeochemical variables are oxygen, nutrient and phytoplankton concentrations, and Secchi depth. Finally, based upon existing literature, changes in pH (a measure of acidification) are assessed.

Oxygen Concentrations

Bottom oxygen concentration changes differ considerably between ECOSUPPORT and BalticAPP scenario simulations as illustrated for summer (Figure 10). These differences mainly reflect the differences in experimental setup of the simulations and nutrient load scenarios (Meier, Edman, et al., 2018). While in the shallow regions future bottom oxygen concentrations decrease in all scenario simulations due to the reduced air–sea fluxes of oxygen (because the saturation concentration of oxygen is lower in warmer water), in the deeper offshore regions changes in bottom oxygen concentration depend largely on the applied nutrient load scenario (Figure 10). In the ECOSUPPORT scenario simulations, future bottom oxygen concentration decreases in all scenarios significantly except under BSAP where the bottom oxygen concentrations in the deeper regions are only slightly reduced on average (cf. Meier, Andersson, et al., 2011). By contrast, in the BalticAPP projections, bottom oxygen concentrations under BSAP increase in the deeper regions considerably regardless of the degree of warming (cf. Saraiva et al., 2019b). Under RCP 4.5, bottom oxygen concentrations increase even under REF and BAU nutrient loads whereas under RCP 8.5 slight reductions in the Bothnian Sea and southwestern Baltic Sea, in particular under BAU, are found. Area averaged changes are summarized in Table 4.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 10. Ensemble Mean Summer Bottom Oxygen Concentration Changes, 1978–2007 and 2069–2098.

Source: H. E. Markus Meier, IOW and SMHI.

Note. Ensemble mean summer (June, July, and August) bottom oxygen concentration changes (mL L-1) between 1978 and 2007 and 2069 and 2098. From left to right, results of the nutrient load scenarios BSAP, REF, and BAU are shown. From top to bottom, results of the ensembles ECOSUPPORT (white background), BalticAPP RCP 4.5 (grey background), and BalticAPP RCP 8.5 (grey background) are depicted.

Table 4. Ensemble Mean Changes in Summer Mean Bottom Oxygen Concentration (in mL L-1)

Summer Changes

ECOSUPPORT

BalticAPP RCP 4.5

BalticAPP RCP 8.5

BSAP

-0.1

0.6

0.5

REF

-0.6

0.1

-0.2

BAU

-1.1

-0.1

-0.5

Note. Shown are changes in ECOSUPPORT, BalticAPP RCP 4.5, and BalticAPP RCP 8.5 scenario simulations averaged for the Baltic Sea, including the Kattegat. The project changes depend on the nutrient load scenario BSAP, REF, and BAU.

Nutrient Concentrations

Changes in winter mean surface phosphate and nitrate concentrations are shown in Figures 11 and 12, respectively. While in ECOSUPPORT scenario simulations winter surface phosphate concentrations increase under all three nutrient load scenarios (except in the Gulf of Finland in BSAP), in BalticAPP projections winter surface phosphate concentrations decrease almost everywhere (except in the Odra Bight in REF and BAU) (Figure 11).

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 11. Winter Mean Surface Phosphate Concentration Changes.

Source: H. E. Markus Meier, IOW and SMHI.

Note. As Figure 10 but for winter (December, January, and February) mean surface phosphate concentration changes (mmol P m-3). Concentrations are vertically averaged for the upper 10 m.

In contrast to spatial patterns of surface phosphate concentration changes, larger nitrate concentration changes are usually confined to the coastal zone, showing varying signs of the changes. In ECOSUPPORT projections, winter surface nitrate concentrations increase in particular in the Gulf of Riga, eastern Gulf of Finland, and along the eastern coasts of the Baltic proper in REF and BAU (Figure 12). In BalticAPP projections, in REF and BAU winter surface nitrate concentrations increase in particular in Bothnian Bay and Odra Bight while concentrations decrease in the Gulf of Riga and Vistula lagoon. Overall, the differences in surface nutrient concentrations between the two ensembles are considerable.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 12. Winter Mean Surface Nitrate Concentration Changes.

Source: H. E. Markus Meier, IOW and SMHI.

Note. As Figure 10 but for winter (December, January, and February) mean surface nitrate concentration changes (mmol N m-3). Concentrations are vertically averaged for the upper 10 m.

Phytoplankton Concentration and Secchi Depth

Annual mean surface phytoplankton concentration changes follow the changes in nutrient concentrations and are confined to the productive zone along the coasts (Figure 13). Secchi depth (Sd) is a measure of water transparency and is calculated from Sd = 1.7/k(PAR), where k(PAR) is the coefficient of underwater attenuation of the photosynthetically available radiation (Kratzer, Håkansson, & Sahlin, 2003). Factors controlling k(PAR) in the models are the concentrations of phytoplankton and detritus. In addition, salinity is used in one of the models as a proxy of the spatio-temporal dynamics of yellow substances.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 13. Annual Mean Surface Phytoplankton Concentration Changes.

Source: H. E. Markus Meier, IOW and SMHI.

Note. As Figure 10 but for annual mean surface phytoplankton concentration changes (mg Chl m-3). Concentrations are vertically averaged for the upper 10 m.

In ECOSUPPORT projections, annual mean Secchi depths are decreasing in all scenario simulations (see Figure 14 and Table 5). On the other hand, in BalticAPP projections, area averaged Secchi depths generally increase, except in the combination of RCP 8.5 and BAU scenarios (Table 5), indicating an improvement of the water quality in future compared to present climate. The most striking changes occur in the BSAP scenario, showing Secchi depth increases of up to 2 m in the coastal zone of the eastern Baltic proper. In summary, Secchi depth projections are characterized by contradictory signs in the changes when EOCUPPORT and BalticAPP scenario simulations are compared.

Projected Oceanographical Changes in the Baltic Sea until 2100

Figure 14. Annual Mean Secchi Depth Changes.

Source: H. E. Markus Meier, IOW and SMHI.

Note. As Figure 10 but for annual mean Secchi depth changes (m).

Table 5. Ensemble Mean Changes in Annual Secchi Depth (in m)

Annual Changes

ECOSUPPORT

BalticAPP RCP 4.5

BalticAPP RCP 8.5

BSAP

-0.3

0.6

0.6

REF

-0.6

0.2

0.1

BAU

-0.8

0.1

-0.1

Note. Shown are changes in ECOSUPPORT, BalticAPP RCP 4.5, and BalticAPP RCP 8.5 scenario simulations averaged for the Baltic Sea, including the Kattegat. The project changes depend on the nutrient load scenario BSAP, REF, and BAU.

pH

Rising atmospheric CO2 concentrations mainly control future changes in pH in the surface water of the Baltic Sea, while eutrophication would mainly enhance the seasonal cycle of pH (Schneider et al., 2015). Available GHG emission scenarios suggested that the pH would significantly decrease by the end of this century (Omstedt et al., 2012). However, increasing alkalinity may counteract future acidification (Müller, Schneider, & Rehder, 2016).

Conclusions

Since about the beginning of the 2000s, regional climate ocean circulation models became available and were used for climate studies and future projections. At the same time, scenario simulations were successively improved. About 20 years later (in 2019), more reliable information about the consequences of climate change for the marine environment was available than ever before since the beginning of RCM development, including not only the physical variables of the Baltic Sea but also the marine ecosystem. The reliability of projections has obviously been improved because GCMs and RCMs were refined, and for the dynamical downscaling approach, experimental setups were improved and bias corrections were less often applied. In addition, more plausible nutrient load scenarios and larger multi-model ensembles were employed, enabling the estimation of uncertainties.

The latest available projections (BalticAPP) suggest that water temperatures will increase and sea-ice cover and salinity will decrease in the future. For instance, the ensemble mean SST increase between the periods 1978–2007 and 2069–2098 amounts to 1.8 and 3.1°C in medium (RCP 4.5) and high-end (RCP 8.5) scenarios, respectively.

Due to the increasing total annual river discharge, salinity is projected to decrease. During winter, runoff from the northern Baltic Sea catchment area will increase, while during summer, runoff from the southern regions will decrease. However, all hydrological scenario simulations suggest an increase in annual mean runoff for the entire Baltic Sea catchment. As a consequence, SSS and bottom salinity are projected to decrease by about 0.6 g kg-1 in both medium and high-end scenarios.

Vertical density gradients across the seasonal thermocline (layer between the atmospherically influenced surface layer and the deep water) may increase due to warming whereas the stratification across the perennial halocline (layer between the brackish surface layer and the more saline deep water) may decrease due to the increasing runoff.

According to IPCC (2013), the mean sea level at the end of the century will be 0.47 and 0.63 m higher than during the reference period in medium and high-end scenarios, respectively. Mean sea level rise in the Baltic Sea will follow the global mean sea level rise with an additional, wind-induced small increase at the end of the 21st century of about 4 cm in the high-end scenario. However, in the northern Baltic Sea land uplift may counteract future sea level rise, resulting in a decrease of the mean sea level relative to the bedrock.

Projections of bottom oxygen concentrations mainly depend on the nutrient load scenarios. Under the best case scenario (BSAP), summer bottom oxygen concentrations will improve at the end of the 21st century by about 0.6 and 0.5 mL L-1 on average compared to the reference period 1978–2007 in medium and high-end climate scenarios, respectively. However, under the worst-case scenario (BAU), summer bottom oxygen concentrations will correspondingly decline by about 0.1 and 0.5 mL L-1.

Similarly, future winter surface nutrients and phytoplankton concentrations will crucially depend on the nutrient load scenarios. As the available nutrient concentrations during winter affect spring and summer blooms and, consequently, water transparency, future Secchi depth changes depend more on the nutrient load than on the climate scenario when nutrient loads and GHG concentrations vary within the ranges between best (BSAP) and worst (BAU) cases and between medium (RCP 4.5) and high-end (RCP 8.5) climate scenarios, respectively. Under the best case nutrient load scenario, the annual mean Secchi depth will increase at the end of the 21st century by 0.6 m on average regardless of the climate scenario. Under the worst-case nutrient load scenario, corresponding changes are 0.1 and -0.1 m in medium and high-end climate scenarios, respectively.

Hence, the implementation of the BSAP are expected to result in considerably improved environmental conditions in the Baltic Sea, indicated by bottom oxygen concentrations and water transparency, despite the counteracting impact of changing climate. However, the impact of climate change is more important under high nutrient loads. Thus, an incomplete implementation of the BSAP may jeopardize the overall goal of achieving good environmental status in the Baltic Sea.

Although the information from regional climate modeling is already frequently used for climate services, including climate change adaptation management with large economic consequences, there are still big challenges to overcome. The few studies that exist and which address the range and the sources of uncertainties suggest that uncertainties in scenario simulations of the Baltic Sea cannot be neglected. In particular, projections of the marine biogeochemical cycles still suffer from uncertainties due to GCM and RCM biases, natural variability, and unknown GHG emission and nutrient load scenarios. In this study, the comparison of ensembles of scenario simulations driven by CMIP3 and CMIP5 GCMs illustrates these uncertainties. However, the assumptions underlying the newer projections are regarded as more plausible than those underlying the older projections.

Unknown current and future bioavailable nutrient loads from land and atmosphere and the experimental setup of the dynamical downscaling technique were perhaps the largest sources of uncertainties for projections of the Baltic Sea biogeochemical cycling. Further uncertainties were: (1) the shortcomings in the projections of the GCMs and RCMs, in particular with respect to global sea level rise, regional water cycle, and ice–ocean interactions such as the consideration of Archimedes’ principle (which is not the case in all models); (2) shortcomings in the simulated response of biogeochemical cycles to long-term changes in external nutrient loads and to changing climate in the Baltic Sea region; and (3) unknown future GHG emissions.

However, there is hope. If the conclusions about the sources of uncertainties are correct, the large ensemble spread will potentially be reducible through investments in new multi-model ensemble simulations that are built on better experimental setups, more reliable estimates of nutrients’ bioavailability, and the coastal filter capacity, by including the relevant processes of the coastal zone and improved models, for example, for the regional water cycle, global sea level rise, and ice–ocean interaction.

As for the Baltic Sea region, the international capacities in regional Earth system science are limited so we recommend developing community models with improved performance and common coordinated experiments of scenario simulations to allow the analysis of big multi-model ensembles.

Acknowledgments

This work forms part of the Baltic Earth program (Earth System Science for the Baltic Sea Region) and was funded by the BONUS BalticAPP (Well-Being from the Baltic Sea: Applications Combining Natural Science and Economics) project, which has received funding from BONUS, the joint Baltic Sea research and development program (Art 185), funded jointly from the European Union’s Seventh Programme for research, technological development, and demonstration and from the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS, Grant No. 942-2015-23). Additional support by FORMAS within the project “Cyanobacteria Life Cycles and Nitrogen Fixation in Historical Reconstructions and Future Climate Scenarios (1850–2100) of the Baltic Sea” (Grant No. 214-2013-1449) is acknowledged. Further, technical support for the preparation of the figures by Berit Recklebe (Leibniz Institute for. Baltic Sea Research, Warnemünde; IOW) is very much acknowledged.

References

Ahlgren, J., Grimvall, A., Omstedt, A., Rolff, C., & Wikner, J. (2017). Temperature, DOC level and basin interactions explain the declining oxygen concentrations in the Bothnian Sea. Journal of Marine Systems, 170, 22–30.Find this resource:

    Andersson, A., Meier, H. E. M., Ripzsam, M., Rowe, O., Wikner, J., Haglund, P., . . . Elmgren, R. (2015). Future climate change scenarios for the Baltic Sea ecosystem and impacts for management. AMBIO: A Journal of the Human Environment, 44, Supplement 3, S345–S356.Find this resource:

      Andersson, T. (2018). För Stockholms överlevnad: Ett nytt Slussen varje århundrade. Dagens infrastructure, February 14.Find this resource:

        Arheimer, B., Dahné, J., & Donnelly, C. (2012). Climate change impact on riverine nutrient load and land-based remedial measures of the Baltic Sea Action Plan. AMBIO: A Journal of the Human Environment, 41(6), 600–612.Find this resource:

          BACC Author Team. (2008). Assessment of climate change for the Baltic Sea basin. Berlin, Germany: Springer Science & Business Media.Find this resource:

            BACC II Author Team. (2015). Second assessment of climate change for the Baltic Sea Basin. Springer Regional Climate Studies. Cham, Switzerland: Springer.Find this resource:

              Bauer, B., Meier, H., Casini, M., Hoff, A., Margoński, P., Orio, A., . . . Tomczak, M. T. (2018). Reducing eutrophication increases spatial extent of communities supporting commercial fisheries: A model case study. ICES Journal of Marine Science, 75(4), 1306–1317.Find this resource:

                Belkin, I. M. (2009). Rapid warming of large marine ecosystems. Progress in Oceanography, 81(1–4), 207–213.Find this resource:

                  Bjerknes, V. (1904). Das Problem der Wettervorhersage, betrachtet vom Standpunkte der Mechanik und der Physik. Meteorologische Zeitschrift, 21, 1–7.Find this resource:

                    Bjørge, D., & Haugen, J. E. (1998). Simulation of present-day climate in HIRLAM using “perfect” boundaries, Vol. 1. RegClim General Technical Report.Find this resource:

                      Börgel, F., Frauen, C., Neumann, T., Schimanke, S., & Meier, H. E. M. (2018). Impact of the Atlantic Multidecadal Oscillation on Baltic Sea variability. Geophysical Research Letters, 45(18), 9880–9888.Find this resource:

                        Bülow, K., Dieterich, C., Elizalde, A., Gröger, M., Heinrich, H., Hüttl-Kabus, S., . . . Su, J. (2014). Comparison of three regional coupled ocean atmosphere models for the North Sea under today’s and future climate conditions. KLIWAS Zeitschriftenreihe 27/2014. Koblenz, Germany: Bundesanstalt für Gewässerkunde.Find this resource:

                          Carstensen, J., Andersen, J. H., Gustafsson, B. G., & Conley, D. J. (2014). Deoxygenation of the Baltic Sea during the last century. Proceedings of the National Academy of Sciences, 111, 5628–5633.Find this resource:

                            Christensen, J. H., & Christensen, O. B. (2007). A summary of the PRUDENCE model projections of changes in European climate by the end of this century. Climatic Change, 81(1), 7–30.Find this resource:

                              Christensen, O. B., Christensen, J. H., Machenhauer, B., & Botzet, M. (1998). Very high-resolution regional climate simulations over Scandinavia: Present climate. Journal of Climate, 11(12), 3204–3229.Find this resource:

                                Church, J. A., & White, N. J. (2006). A 20th century acceleration in global sea-level rise. Geophysical Research Letters, 33(1), L01602.Find this resource:

                                  Dieterich, C., Schimanke, S., Wang, S., Väli, G., Liu, Y., Hordoir, R., . . . Meier, H. E. M. (2013). Evaluation of the SMHI coupled atmosphere-ice-ocean model RCA4-NEMO. Report Oceanography (RO) No. 47, SMHI, Norrköping, Sweden.Find this resource:

                                    Dieterich, C., Wang, S., Schimanke, S., Gröger, M., Klein, B., Hordoir, R., Samuelsson, P., Liu, Y., Axell, L., Höglund, A., & Meier, H. E. M. (2019). Surface Heat Budget over the North Sea in Climate Change Simulations. Atmosphere, 10.Find this resource:

                                      Döscher, R., & Meier, H. E. M. (2004). Simulated sea surface temperature and heat fluxes in different climates of the Baltic Sea. AMBIO: A Journal of the Human Environment, 33(4), 242–248.Find this resource:

                                        Döscher, R., Willén, U., Jones, C., Rutgersson, A., Meier, H. E. M., Hansson, U., & Graham, L. P. (2002). The development of the regional coupled ocean-atmosphere model RCAO. Boreal Environment Research, 7(3), 183–192.Find this resource:

                                          Eilola, K., Almroth-Rosell, E., Dieterich, C., Fransner, F., Höglund, A., & Meier, H. E. M. (2012). Modeling nutrient transports and exchanges of nutrients between shallow regions and the open Baltic Sea in present and future climate. AMBIO: A Journal of the Human Environment, 41(6), 586–599.Find this resource:

                                            Eilola, K., Almroth-Rosell, E., & Meier, H. E. M. (2014). Impact of saltwater inflows on phosphorus cycling and eutrophication in the Baltic Sea: A 3D model study. Tellus A: Dynamic Meteorology and Oceanography, 66, 23985.Find this resource:

                                              Eilola, K., Gustafsson, B. G., Kuznetsov, I., Meier, H. E. M., Neumann, T., & Savchuk, O. P. (2011). Evaluation of biogeochemical cycles in an ensemble of three state-of-the-art numerical models of the Baltic Sea. Journal of Marine Systems, 88(2), 267–284.Find this resource:

                                                Eilola, K., Mårtensson, S., & Meier, H. E. M. (2013). Modeling the impact of reduced sea ice cover in future climate on the Baltic Sea biogeochemistry. Geophysical Research Letters, 40(1), 149–154.Find this resource:

                                                  Eilola, K., Meier, H. E. M., & Almroth, E. (2009). On the dynamics of oxygen, phosphorus and cyanobacteria in the Baltic Sea: A model study. Journal of Marine Systems, 75(1–2), 163–184.Find this resource:

                                                    Ekman, M., & Mäkinen, J. (1996). Mean sea surface topography in the Baltic Sea and its transition area to the North Sea: A geodetic solution and comparisons with oceanographic models. Journal of Geophysical Research: Oceans, 101(C5), 11993–11999.Find this resource:

                                                      Feser, F., Rockel, B., Storch, H. von, Winterfeldt, J., & Zahn, M. (2011). Regional climate models add value to global model data: A review and selected examples. Bulletin of the American Meteorological Society, 92(9), 1181–1192.Find this resource:

                                                        Flato, G. M. (2011). Earth system models: An overview. Wiley Interdisciplinary Reviews: Climate Change, 2(6), 783–800.Find this resource:

                                                          Fleming-Lehtinen, V., & Laamanen, M. (2012). Long-term changes in Secchi depth and the role of phytoplankton in explaining light attenuation in the Baltic Sea. Estuarine, Coastal and Shelf Science, 102, 1–10.Find this resource:

                                                            Friedland, R., Neumann, T., & Schernewski, G. (2012). Climate change and the Baltic Sea action plan: Model simulations on the future of the western Baltic Sea. Journal of Marine Systems, 105–108, 175–186.Find this resource:

                                                              Giorgi, F. (1990). Simulation of regional climate using a limited area model nested in a general circulation model. Journal of Climate, 3(9), 941–963.Find this resource:

                                                                Giorgi, F., & Mearns, L. O. (1991). Approaches to the simulation of regional climate change: A review. Reviews of Geophysics, 29(2), 191–216.Find this resource:

                                                                  Gómez-Navarro, J. J., Montávez, J. P., Jerez, S., Jiménez-Guerrero, P., Lorente-Plazas, R., González-Rouco, J. F., & Zorita, E. (2011). A regional climate simulation over the Iberian Peninsula for the last millennium. Climate of the Past, 7(2), 451–472.Find this resource:

                                                                    Grinsted, A. (2015). Projected change: Sea level. In BACC II Author Team, Second assessment of climate change for the Baltic Sea Basin. Springer Regional Climate Studies (pp. 253–263). Cham, Switzerland: Springer.Find this resource:

                                                                      Gröger, M., Dieterich, C., Meier, H. E. M., & Schimanke, S. (2015). Thermal air–sea coupling in hindcast simulations for the North Sea and Baltic Sea on the NW European shelf. Tellus A: Dynamic Meteorology and Oceanography, 67(1), 26911.Find this resource:

                                                                        Gröger, M., Arneborg, L., Dieterich, C., Höglund, A., & Meier, H. E. M. (2019). Hydrographic changes in the North Sea and Baltic Sea projected in an ensemble of climate scenarios downscaled with a coupled regional ocean-sea ice-atmosphere model. Climate Dynamics (published online 1 August 2019).Find this resource:

                                                                          Gustafsson, B. G. (2003). A time-dependent coupled-basin model for the Baltic Sea. Report C47, Earth Sciences Centre, University of Gothenburg, Sweden.Find this resource:

                                                                            Gustafsson, B. G., Schenk, F., Blenckner, T., Eilola, K., Meier, H. E. M., Müller-Karulis, B., . . . Zorita, E. (2012). Reconstructing the development of Baltic Sea eutrophication 1850–2006. AMBIO: A Journal of the Human Environment, 41(6), 534–548.Find this resource:

                                                                              Gustafsson, N., Nyberg, L., & Omstedt, A. (1998). Coupling of a high-resolution atmospheric model and an ocean model for the Baltic Sea. Monthly Weather Review, 126(11), 2822–2846.Find this resource:

                                                                                Haapala, J., Markus Meier, H. E., & Rinne, J. (2001). Numerical investigations of future ice conditions in the Baltic Sea. AMBIO: A Journal of the Human Environment, 30(4), 237–244.Find this resource:

                                                                                  Hagedorn, R., Lehmann, A., & Jacob, D. (2000). A coupled high resolution atmosphere–ocean model for the BALTEX region. Meteorologische Zeitschrift, 9, 7–20.Find this resource:

                                                                                    Harff, J., Furmańczyk, K., & Von Storch, H. (Eds.). (2017). Coastline changes of the Baltic sea from south to east: Past and future projection, Vol. 19. Cham, Switzerland: Springer.Find this resource:

                                                                                      HELCOM (Helsinki Commission). (2013). Summary report on the development of revised maximum allowable inputs (MAI) and updated country allocated reduction targets (CART) of the Baltic Sea Action Plan. Background Document for the 2013 HELCOM Ministerial Meeting.Find this resource:

                                                                                        HELCOM (Helsinki Commission). (2018). State of the Baltic Sea: Second HELCOM holistic assessment 2011–2016. Baltic Sea Environment Proceedings 155.Find this resource:

                                                                                          Hill, E. M., Davis, J. L., Tamisiea, M. E., & Lidberg, M. (2010). Combination of geodetic observations and models for glacial isostatic adjustment fields in Fennoscandia. Journal of Geophysical Research: Solid Earth, 115, B07403, doi:10.1029/2009JB006967Find this resource:

                                                                                            Ho-Hagemann, H. T. M., Gröger, M., Rockel, B., Zahn, M., Geyer, B., & Meier, H. E. M. (2017). Effects of air–sea coupling over the North Sea and the Baltic Sea on simulated summer precipitation over Central Europe. Climate Dynamics, 49(11), 3851–3876.Find this resource:

                                                                                              Holt, J., Schrum, C., Cannaby, H., Daewel, U., Allen, I., Artioli, Y., . . . Wakelin, S. (2016). Potential impacts of climate change on the primary production of regional seas: A comparative analysis of five European seas. Progress in Oceanography, 140, 91–115.Find this resource:

                                                                                                Hordoir, R., Höglund, A., Pemberton, P., & Schimanke, S. (2018). Sensitivity of the overturning circulation of the Baltic Sea to climate change, a numerical experiment. Climate Dynamics, 50(3), 1425–1437.Find this resource:

                                                                                                  Hordoir, R., & Meier, H. E. M. (2012). Effect of climate change on the thermal stratification of the Baltic Sea: A sensitivity experiment. Climate Dynamics, 38(9), 1703–1713.Find this resource:

                                                                                                    Hurrell, J. W. (1995). Decadal trends in the North Atlantic Oscillation: Regional temperatures and precipitation. Science-AAAS-Weekly Paper Edition, 269(5224), 676–679.Find this resource:

                                                                                                      IPCC (Intergovernmental Panel on Climate Change). (2001). The Third Assessment Report of the Intergovernmental Panel on Climate Change. Geneva, Switzerland.Find this resource:

                                                                                                        IPCC (Intergovernmental Panel on Climate Change). (2007). The Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva, Switzerland.Find this resource:

                                                                                                          IPCC (Intergovernmental Panel on Climate Change). (2013). Climate change 2013: The physical science basis. Working Group I contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University, Cambridge, U.K.Find this resource:

                                                                                                            Jacob, D., Petersen, J., Eggert, B., Alias, A., Christensen, O. B., Bouwer, L. M., . . . Yiou. P. (2014). EURO-CORDEX: New high-resolution climate change projections for European impact research. Regional Environmental Change, 14, 563–578.Find this resource:

                                                                                                              Jacob, D., & Podzun, R. (1997). Sensitivity studies with the regional climate model REMO. Meteorology and Atmospheric Physics, 63(1), 119–129.Find this resource:

                                                                                                                Jutterström, S., Andersson, H. C., Omstedt, A., & Malmaeus, J. M. (2014). Multiple stressors threatening the future of the Baltic Sea–Kattegat marine ecosystem: Implications for policy and management actions. Marine Pollution Bulletin, 86(1–2), 468–480.Find this resource:

                                                                                                                  Kauker, F., & Meier, H. E. M. (2003). Modeling decadal variability of the Baltic Sea: 1. Reconstructing atmospheric surface data for the period 1902–1998. Journal of Geophysical Research: Oceans, 108(C8), 3267.Find this resource:

                                                                                                                    Kjellström, E., Döscher, R., & Meier, H. E. M. (2005). Atmospheric response to different sea surface temperatures in the Baltic Sea: Coupled versus uncoupled regional climate model experiments. Hydrology Research, 36(4–5), 397–409.Find this resource:

                                                                                                                      Kniebusch, M., Meier, H. E. M., & Neumann, T. (2019). Temperature variability of the Baltic Sea since 1850 and attribution to atmospheric forcing variables. Journal of Geophysical Research: Oceans, 124(6), 4168–4187.Find this resource:

                                                                                                                        Knutti, R., & Sedláček, J. (2013). Robustness and uncertainties in the new CMIP5 climate model projections. Nature Climate Change, 3(4), 369–373.Find this resource:

                                                                                                                          Kotilainen, A. T., Arppe, L., Dobosz, S., Jansen, E., Kabel, K., Karhu, J., . . . Zhamoida, V. (2014). Echoes from the past: A healthy Baltic Sea requires more effort. AMBIO: A Journal of the Human Environment, 43, 60–68.Find this resource:

                                                                                                                            Kratzer, S., Håkansson, B., & Sahlin, C. (2003). Assessing Secchi and photic zone depth in the Baltic Sea from satellite data. AMBIO: A Journal of the Human Environment, 32(8), 577–585.Find this resource:

                                                                                                                              Meier, H. E. M. (2002a). Regional ocean climate simulations with a 3D ice-ocean model for the Baltic Sea. Part 1: Model experiments and results for temperature and salinity. Climate Dynamics, 19(3–4), 237–253.Find this resource:

                                                                                                                                Meier, H. E. M. (2002b). Regional ocean climate simulations with a 3D ice–ocean model for the Baltic Sea. Part 2: Results for sea ice. Climate Dynamics, 19(3), 255–266.Find this resource:

                                                                                                                                  Meier, H. E. M. (2005). Modeling the age of Baltic seawater masses: Quantification and steady state sensitivity experiments. Journal of Geophysical Research: Oceans, 110, C02006, doi:10.1029/2004JC002607.Find this resource:

                                                                                                                                    Meier, H. E. M. (2006). Baltic Sea climate in the late twenty-first century: A dynamical downscaling approach using two global models and two emission scenarios. Climate Dynamics, 27(1), 39–68.Find this resource:

                                                                                                                                      Meier, H. E. M. (2007). Modeling the pathways and ages of inflowing salt- and freshwater in the Baltic Sea. Estuarine, Coastal and Shelf Science, 74(4), 610–627.Find this resource:

                                                                                                                                        Meier, H. E. M. (2015). Projected change: Marine physics. In BACC II Author Team, Second assessment of climate change for the Baltic Sea Basin. Springer Regional Climate Studies (pp. 243–252). Cham, Switzerland: Springer.Find this resource:

                                                                                                                                          Meier, H. E. M., Andersson, H. C., Arheimer, B., Blenckner, T., Chubarenko, B., Donnelly, C., . . . Zorita, E. (2012). Comparing reconstructed past variations and future projections of the Baltic Sea ecosystem: First results from multi-model ensemble simulations. Environmental Research Letters, 7(3), 34005.Find this resource:

                                                                                                                                            Meier, H. E. M., Andersson, H. C., Arheimer, B., Donnelly, C., Eilola, K., Gustafsson, B. G., . . . Zorita, E. (2014). Ensemble modeling of the Baltic Sea ecosystem to provide scenarios for management. AMBIO: A Journal of the Human Environment, 43(1), 37–48.Find this resource:

                                                                                                                                              Meier, H. E. M., Andersson, H. C., Eilola, K., Gustafsson, B. G., Kuznetsov, I., Müller-Karulis, B., . . . Savchuk, O. P. (2011). Hypoxia in future climates: A model ensemble study for the Baltic Sea. Geophysical Research Letters, 38, L24608, doi:10.1029/2011GL049929.Find this resource:

                                                                                                                                                Meier, H. E. M., Broman, B., & Kjellström, E. (2004a). Simulated sea level in past and future climates of the Baltic Sea. Climate Research, 27(1), 59–75.Find this resource:

                                                                                                                                                  Meier, H. E. M., Döscher, R., & Faxén, T. (2003). A multiprocessor coupled ice–ocean model for the Baltic Sea: Application to salt inflow. Journal of Geophysical Research: Oceans, 108(C8), 3273, doi:10.1029/2000JC000521.Find this resource:

                                                                                                                                                    Meier, H. E. M., Döscher, R., & Halkka, A. (2004b). Simulated distributions of Baltic sea-ice in warming climate and consequences for the winter habitat of the Baltic ringed seal. AMBIO: A Journal of the Human Environment, 33(4), 249–256.Find this resource:

                                                                                                                                                      Meier, H. E. M., Edman, M., Eilola, K., Placke, M., Neumann, T., Andersson, H. C., . . . Savchuk, O. P. (2018). Assessment of eutrophication abatement scenarios for the Baltic Sea by multi-model ensemble simulations. Frontiers in Marine Science, 5, 440.Find this resource:

                                                                                                                                                        Meier, H. E. M., Edman, M., Eilola, K., Placke, M., Neumann, T., Andersson, H. C., . . .Savchuk, O. P. (2019). Assessment of uncertainties in scenario simulations of biogeochemical cycles in the Baltic Sea. Frontiers in Marine Science, 6, 46.Find this resource:

                                                                                                                                                          Meier, H. E. M., Eilola, K., & Almroth, E. (2011). Climate-related changes in marine ecosystems simulated with a 3-dimensional coupled physical-biogeochemical model of the Baltic Sea. Climate Research, 48(1), 31–55.Find this resource:

                                                                                                                                                            Meier, H. E. M., Eilola, K., Almroth-Rosell, E., Schimanke, S., Kniebusch, M., Höglund, A., . . . Saraiva, S. (2018). Disentangling the impact of nutrient load and climate changes on Baltic Sea hypoxia and eutrophication since 1850. Climate Dynamics, 53(1–2), 1145–1166.Find this resource:

                                                                                                                                                              Meier, H. E. M., Höglund, A., Döscher, R., Andersson, H., Löptien, U., & Kjellström, E. (2011). Quality assessment of atmospheric surface fields over the Baltic Sea from an ensemble of regional climate model simulations with respect to ocean dynamics. Oceanologia, 53(1), 193–227.Find this resource:

                                                                                                                                                                Meier, H. E. M., Höglund, A., Eilola, K., & Almroth-Rosell, E. (2017). Impact of accelerated future global mean sea level rise on hypoxia in the Baltic Sea. Climate Dynamics, 49(1–2), 163–172.Find this resource:

                                                                                                                                                                  Meier, H. E. M., Hordoir, R., Andersson, H. C., Dieterich, C., Eilola, K., Gustafsson, B. G., . . . Schimanke, S. (2012). Modeling the combined impact of changing climate and changing nutrient loads on the Baltic Sea environment in an ensemble of transient simulations for 1961–2099. Climate Dynamics, 39(9), 2421–2441.Find this resource:

                                                                                                                                                                    Meier, H. E. M., & Kauker, F. (2003a). Modeling decadal variability of the Baltic Sea: 2. Role of freshwater inflow and large-scale atmospheric circulation for salinity. Journal of Geophysical Research: Oceans, 108(C11), 3368, doi:10.1029/2003JC001799.Find this resource:

                                                                                                                                                                      Meier, H. E. M., & Kauker, F. (2003b). Sensitivity of the Baltic Sea salinity to the freshwater supply. Climate Research, 24(3), 231–242.Find this resource:

                                                                                                                                                                        Meier, H. E. M., Kjellström, E., & Graham, L. P. (2006). Estimating uncertainties of projected Baltic Sea salinity in the late 21st century. Geophysical Research Letters, 33, L15705, doi:10.1029/2006GL026488.Find this resource:

                                                                                                                                                                          Meier, H. E. M., Müller-Karulis, B., Andersson, H. C., Dieterich, C., Eilola, K., Gustafsson, B. G., . . . Schimanke, S. (2012). Impact of climate change on ecological quality indicators and biogeochemical fluxes in the Baltic Sea: A multi-model ensemble study. AMBIO: A Journal of the Human Environment, 41(6), 558–573.Find this resource:

                                                                                                                                                                            Meier, H. E. M., Rutgersson, A., & Reckermann, M. (2014). An Earth System Science Program for the Baltic Sea Region. Eos, Transactions American Geophysical Union, 95(13), 109–110.Find this resource:

                                                                                                                                                                              Meier, H. E. M., Väli, G., Naumann, M., Eilola, K., & Frauen, C. (2018). Recently accelerated oxygen consumption rates amplify deoxygenation in the Baltic Sea. Journal of Geophysical Research: Oceans, 123(5), 3227–3240.Find this resource:

                                                                                                                                                                                Mohrholz, V. (2018). Major Baltic inflow statistics: Revised. Frontiers in Marine Science, 5, 384.Find this resource:

                                                                                                                                                                                  Moss, R. H., Edmonds, J. A., Hibbard, K. A., Manning, M. R., Rose, S. K., Van Vuuren, D. P., . . . Wilbanks, T. J. (2010). The next generation of scenarios for climate change research and assessment. Nature, 463, 747–756.Find this resource:

                                                                                                                                                                                    Müller, J. D., Schneider, B., & Rehder, G. (2016). Long-term alkalinity trends in the Baltic Sea and their implications for CO2-induced acidification. Limnology and Oceanography, 61(6), 1984–2002.Find this resource:

                                                                                                                                                                                      Nakićenović, N., Alcamo, J., Grubler, A., Riahi, K., Roehrl, R. A., Rogner, H.-H., & Victor, N. (2000). Special report on emissions scenarios (SRES), a special report of Working Group III of the intergovernmental panel on climate change. IPCC Special Report.Find this resource:

                                                                                                                                                                                        Neumann, T. (2010). Climate-change effects on the Baltic Sea ecosystem: A model study. Journal of Marine Systems, 81(3), 213–224.Find this resource:

                                                                                                                                                                                          Neumann, T., Eilola, K., Gustafsson, B., Müller-Karulis, B., Kuznetsov, I., Meier, H. E. M., & Savchuk, O. P. (2012). Extremes of temperature, oxygen and blooms in the Baltic Sea in a changing climate. AMBIO: A Journal of the Human Environment, 41(6), 574–585.Find this resource:

                                                                                                                                                                                            Niiranen, S., Yletyinen, J., Tomczak, M. M., Blenckner, T., Hjerne, O., MacKenzie, B., & Meier, H. E. M. (2013). Combined effects of global climate change and regional ecosystem drivers on an exploited marine food web. Global Change Biology, 19(11), 3327–3342.Find this resource:

                                                                                                                                                                                              Omstedt, A. (1990). Modelling the Baltic Sea as thirteen sub-basins with vertical resolution. Tellus A: Dynamic Meteorology and Oceanography, 42(2), 286–301.Find this resource:

                                                                                                                                                                                                Omstedt, A., & Axell, L. B. (1998). Modeling the seasonal, interannual, and long-term variations of salinity and temperature in the Baltic proper. Tellus A: Dynamic Meteorology and Oceanography, 50(5), 637–652.Find this resource:

                                                                                                                                                                                                  Omstedt, A., & Axell, L. B. (2003). Modeling the variations of salinity and temperature in the large Gulfs of the Baltic Sea. Continental Shelf Research, 23(3–4), 265–294.Find this resource:

                                                                                                                                                                                                    Omstedt, A., & Chen, D. (2001). Influence of atmospheric circulation on the maximum ice extent in the Baltic Sea. Journal of Geophysical Research: Oceans, 106(C3), 4493–4500.Find this resource:

                                                                                                                                                                                                      Omstedt, A., Edman, M., Claremar, B., Frodin, P., Gustafsson, E., Humborg, C., . . . Yurova, A. (2012). Future changes in the Baltic Sea acid-base (pH) and oxygen balances. Tellus B: Chemical and Physical Meteorology, 64(1), 19586.Find this resource:

                                                                                                                                                                                                        Omstedt, A., Edman, M., Claremar, B., & Rutgersson, A. (2015). Modelling the contributions to marine acidification from deposited SOx, NOx, and NHx in the Baltic Sea: Past and present situations. Continental Shelf Research, 111, 234–249.Find this resource:

                                                                                                                                                                                                          Omstedt, A., Elken, J., Lehmann, A., Leppäranta, M., Meier, H. E. M., Myrberg, K., & Rutgersson, A. (2014). Progress in physical oceanography of the Baltic Sea during the 2003–2014 period. Progress in Oceanography, 128, 139–171.Find this resource:

                                                                                                                                                                                                            Omstedt, A., Elken, J., Lehmann, A., & Piechura, J. (2004). Knowledge of the Baltic Sea physics gained during the BALTEX and related programmes. Progress in Oceanography, 63(1–2), 1–28.Find this resource:

                                                                                                                                                                                                              Omstedt, A., Gustafsson, B., Rodhe, J., & Walin, G. (2000). Use of Baltic Sea modelling to investigate the water cycle and the heat balance in GCM and regional climate models. Climate Research, 15(2), 95–108.Find this resource:

                                                                                                                                                                                                                Omstedt, A., Humborg, C., Pempkowiak, J., Perttilä, M., Rutgersson, A., Schneider, B., & Smith, B. (2014). Biogeochemical control of the coupled CO2–O2 system of the Baltic Sea: A review of the results of Baltic-C. AMBIO: A Journal of the Human Environment, 43(1), 49–59.Find this resource:

                                                                                                                                                                                                                  O’Neill, B. C., Kriegler, E., Riahi, K., Ebi, K. L., Hallegatte, S., Carter, T. R., . . . van Vuuren, D. P. (2014). A new scenario framework for climate change research: The concept of shared socioeconomic pathways. Climatic Change, 122(3), 387–400.Find this resource:

                                                                                                                                                                                                                    Placke, M., Meier, H. E. M., Gräwe, U., Neumann, T., Frauen, C., & Liu, Y. (2018). Long-term mean circulation of the Baltic Sea as represented by various ocean circulation models. Frontiers in Marine Science, 5(287), 1–20.Find this resource:

                                                                                                                                                                                                                      Pushpadas, D., Schrum, C., & Daewel, U. (2015). Projected climate change impacts on North Sea and Baltic Sea: CMIP3 and CMIP5 model based scenarios. Biogeosciences Discussions, 12(15), 12229–12279.Find this resource:

                                                                                                                                                                                                                        Quante, M., & Colijn, F. (Eds.). (2016). North Sea region climate change assessment. Switzerland: SpringerOpen.Find this resource:

                                                                                                                                                                                                                          Räisänen, J., Hansson, U., Ullerstig, A., Döscher, R., Graham, L. P., Jones, C., . . . Willén, U. (2004). European climate in the late twenty-first century: Regional simulations with two driving global models and two forcing scenarios. Climate Dynamics, 22(1), 13–31.Find this resource:

                                                                                                                                                                                                                            Räisänen, J., Rummukainen, M., & Ullerstig, A. (2001). Downscaling of greenhouse gas induced climate change in two GCMs with the Rossby Centre regional climate model for northern Europe. Tellus A: Dynamic Meteorology and Oceanography, 53(2), 168–191.Find this resource:

                                                                                                                                                                                                                              Raschke, E., Meywerk, J., Warrach, K., Andrea, U., Bergström, S., Beyrich, F., . . . Stigebrandt, A. (2001). The Baltic Sea Experiment (BALTEX): A European contribution to the investigation of the energy and water cycle over a large drainage basin. Bulletin of the American Meteorological Society, 82(11), 2389–2413.Find this resource:

                                                                                                                                                                                                                                Reckermann, M., Langner, J., Omstedt, A., von Storch, H., Keevallik, S., Schneider, B., . . . Hünicke, B. (2011). BALTEX: An interdisciplinary research network for the Baltic Sea region. Environmental Research Letters, 6, 045205.Find this resource:

                                                                                                                                                                                                                                  Rennau, H., & Burchard, H. (2009). Quantitative analysis of numerically induced mixing in a coastal model application. Ocean Dynamics, 59(5), 671.Find this resource:

                                                                                                                                                                                                                                    Rockel, B. (2015). The regional downscaling approach: A brief history and recent advances. Current Climate Change Reports, 1(1), 22–29.Find this resource:

                                                                                                                                                                                                                                      Rummukainen, M. (2010). State-of-the-art with regional climate models. Wiley Interdisciplinary Reviews: Climate Change, 1(1), 82–96.Find this resource:

                                                                                                                                                                                                                                        Rummukainen, M., Bergström, S., Källén, E., Moen, L., Rodhe, J., & Tjernström, M. (2000). SWECLIM: The first three years. Norrköping, Sweden: Swedish Meteorological and Hydrological Institute.Find this resource:

                                                                                                                                                                                                                                          Rummukainen, M., Bergström, S., Persson, G., Rodhe, J., & Tjernström, M. (2004). The Swedish regional climate modelling programme, SWECLIM: A review. AMBIO: A Journal of the Human Environment, 33(4), 176–182.Find this resource:

                                                                                                                                                                                                                                            Rummukainen, M., Räisänen, J., Bringfelt, B., Ullerstig, A., Omstedt, A., Willén, U., . . . Jones, C. (2001). A regional climate model for northern Europe: Model description and results from the downscaling of two GCM control simulations. Climate Dynamics, 17(5–6), 339–359.Find this resource:

                                                                                                                                                                                                                                              Rutgersson, A., Nilsson, E. O., & Kumar, R. (2012). Introducing surface waves in a coupled wave-atmosphere regional climate model: Impact on atmospheric mixing length. Journal of Geophysical Research: Oceans, 117, C00J15, doi:10.1029/2012JC007940.Find this resource:

                                                                                                                                                                                                                                                Ryabchenko, V. A., Karlin, L. N., Isaev, A. V., Vankevich, R. E., Eremina, T. R., Molchanov, M. S., & Savchuk, O. P. (2016). Model estimates of the eutrophication of the Baltic Sea in the contemporary and future climate. Oceanology, 56(1), 36–45.Find this resource:

                                                                                                                                                                                                                                                  Saraiva, S., Meier, H. E. M., Andersson, H. C., Höglund, A., Dieterich, C., Gröger, M., . . . & Eilola, K. (2019a). Baltic Sea ecosystem response to various nutrient load scenarios in present and future climates. Climate Dynamics, 52, 3369.Find this resource:

                                                                                                                                                                                                                                                    Saraiva S., Meier, H. E. M., Andersson, H. C., Höglund, A., Dieterich, C., Gröger, M., . . . & Eilola, K. (2019b). Uncertainties in projections of the Baltic Sea ecosystem driven by an ensemble of global climate models. Frontiers in Earth Science, 6, 244.Find this resource:

                                                                                                                                                                                                                                                      Schimanke, S., Dieterich, C., & Meier, H. E. M. (2014). An algorithm based on sea-level pressure fluctuations to identify major Baltic inflow events. Tellus A: Dynamic Meteorology and Oceanography, 66(S2), 23452.Find this resource:

                                                                                                                                                                                                                                                        Schimanke, S., & Meier, H. E. M. (2016). Decadal-to-centennial variability of salinity in the Baltic Sea. Journal of Climate, 29(20), 7173–7188.Find this resource:

                                                                                                                                                                                                                                                          Schneider, B., Eilola, K., Lukkari, K., Müller-Karulis, B., & Neumann, T. (2015). Environmental impacts: Marine biogeochemistry. In BACC II Author Team, Second assessment of climate change for the Baltic Sea Basin. Springer Regional Climate Studies (pp. 337–361). Cham, Switzerland: Springer.Find this resource:

                                                                                                                                                                                                                                                            Schrum, C. (2001). Regionalization of climate change for the North Sea and Baltic Sea. Climate Research, 18(1–2), 31–37.Find this resource:

                                                                                                                                                                                                                                                              Schrum, C., Hübner, U., Jacob, D., & Podzun, R. (2003). A coupled atmosphere/ice/ocean model for the North Sea and the Baltic Sea. Climate Dynamics, 21(2), 131–151.Find this resource:

                                                                                                                                                                                                                                                                Sein, D. V., Mikolajewicz, U., Gröger, M., Fast, I., Cabos, W., Pinto, J. G., . . . Jacob, D. (2015). Regionally coupled atmosphere–ocean–sea ice–marine biogeochemistry model ROM: 1. Description and validation. Journal of Advances in Modeling Earth Systems, 7(1), 268–304.Find this resource:

                                                                                                                                                                                                                                                                  Serreze, M. C., & Francis, J. A. (2006). The Arctic amplification debate. Climatic Change, 76(3–4), 241–264.Find this resource:

                                                                                                                                                                                                                                                                    Sjöberg B. (ed) (1992). Sea and coast. The National Atlas of Sweden (p. 128). Stockholm: Almqvist and Wiksell International.Find this resource:

                                                                                                                                                                                                                                                                      Skogen, M. D., Eilola, K., Hansen, J. L. S., Meier, H. E. M., Molchanov, M. S., & Ryabchenko, V. A. (2014). Eutrophication status of the North Sea, Skagerrak, Kattegat and the Baltic Sea in present and future climates: A model study. Journal of Marine Systems, 132, 174–184.Find this resource:

                                                                                                                                                                                                                                                                        Smith, B., Samuelsson, P., Wramneby, A., & Rummukainen, M. (2011). A model of the coupled dynamics of climate, vegetation and terrestrial ecosystem biogeochemistry for regional applications. Tellus A: Dynamic Meteorology and Oceanography, 63(1), 87–106.Find this resource:

                                                                                                                                                                                                                                                                          Stigebrandt, A. (1983). A model for the exchange of water and salt between the Baltic and the Skagerrak. Journal of Physical Oceanography, 13(3), 411–427.Find this resource:

                                                                                                                                                                                                                                                                            Sverdrup, H. (1953). On conditions for the vernal blooming of phytoplankton. Journal du Conseil/Conseil Permanent International pour l’Exploration de la Mer, 18, 287–295.Find this resource:

                                                                                                                                                                                                                                                                              Tian, T., Boberg, F., Christensen, O. B., Christensen, J. H., She, J., & Vihma, T. (2013). Resolved complex coastlines and land–sea contrasts in a high-resolution regional climate model: A comparative study using prescribed and modelled SSTs. Tellus A: Dynamic Meteorology and Oceanography, 65(1), 19951.Find this resource:

                                                                                                                                                                                                                                                                                Van der Linden, P., & Mitchell, J. F. B. (2009). ENSEMBLES: Climate change and its impacts. Summary of research and results from the ENSEMBLES project.Find this resource:

                                                                                                                                                                                                                                                                                  Van Pham, T., Brauch, J., Dieterich, C., Frueh, B., & Ahrens, B. (2014). New coupled atmosphere–ocean–ice system COSMO-CLM/NEMO: Assessing air temperature sensitivity over the North and Baltic Seas. Oceanologia, 56(2), 167–189.Find this resource:

                                                                                                                                                                                                                                                                                    Wang, S., Dieterich, C., Döscher, R., Höglund, A., Hordoir, R., Meier, H. E. M., . . . Schimanke, S. (2015). Development and evaluation of a new regional coupled atmosphere–ocean model in the North Sea and Baltic Sea. Tellus A: Dynamic Meteorology and Oceanography, 67(1), 24284.Find this resource:

                                                                                                                                                                                                                                                                                      WMO (World Meteorological Organization). (2014). Implementation plan of the Global Framework for Climate Services (GFCS). WMO, Geneva, Switzerland.Find this resource:

                                                                                                                                                                                                                                                                                        Zandersen, M., Hyytiainen, K., Meier, H. E. M., Tomzcak, M., Bauer, B., Haapasaari, P., . . . Van Vuuren, D. P. (2018). Extending shared socioeconomic pathways for the Baltic Sea region for use in studying regional environmental problems. Regional Environmental Change, 19, 1073–1086.Find this resource:

                                                                                                                                                                                                                                                                                          Notes:

                                                                                                                                                                                                                                                                                          (2.) The Secchi disk is a plain white, circular disk of 30 cm in diameter mounted on a pole or line. It is lowered slowly down in the water. The Secchi depth is the depth when the disk disappears.