Available via license: CC BY 4.0
Content may be subject to copyright.
1
Natural Hazards and Extreme Events in the Baltic Sea region
1
Anna Rutgersson1,2, Erik Kjellström3,4, Jari Haapala5, Martin Stendel6, Irina Danilovich7, Martin
2
Drews8, Kirsti Jylhä5, Pentti Kujala9, Xiaoli Guo Larsén10, Kirsten Halsnæs8, Ilari Lehtonen5, Anna
3
Luomaranta5, Erik Nilsson1,2, Taru Olsson5, Jani Särkkä5, Laura Tuomi5, Norbert Wasmund11
4
Affiliations:
5
1Department of Earth Sciences, Uppsala University, Uppsala, Sweden
6
2Centre of Natural Hazards and Disaster Science, Uppsala University, Uppsala, Sweden
7
3Swedish Meteorological and Hydrological Institute, Norrköping, Sweden
8
4Department of Meteorology and the Bolin Centre, Stockholm University, Stockholm, Sweden
9
5Finnish Meteorological Institute, Helsinki, Finland
10
6Danish Meteorological Institute, Copenhagen, Denmark
11
7Institute for Nature Management, National Academy of Sciences, Minsk, Belarus
12
8Department of Technology, Management and Economics, Technical University of Denmark, Kgs. Lyngby,
13
Denmark
14
9Aalto University, Espoo, Finland
15
10Wind Energy Department, Technical University of Denmark, Roskilde, Denmark
16
11Leibniz Institute for Baltic Sea Research, Warnemünde, Germany
17
18
Correspondence to: Anna Rutgersson (anna.rutgersson@met.uu.se)
19
20
Abstract.
21
A natural hazard is a naturally occurring extreme event with a negative effect on people and society or
22
the environment. Natural hazards may have severe implications for human life and they can potentially
23
generate economic losses and damage ecosystems. A better understanding of their major causes,
24
probability of occurrence, and consequences enables society to be better prepared and to save human
25
lives and to invest in adaptation options. Natural Hazards related to climate change are identified as one
26
of the Grand Challenges in the Baltic Sea region. We here summarise existing knowledge of extreme
27
events in the Baltic Sea region with the focus on past 200 years, as well as future climate scenarios. The
28
events considered here are the major hydro-meteorological events in the region and include wind storms,
29
extreme waves, high and low sea level, ice ridging, heavy precipitation, sea-effect snowfall, river floods,
30
heat waves, ice seasons, and drought. We also address some ecological extremes and implications of
31
extreme events for society (phytoplankton blooms, forest fires, coastal flooding, offshore
32
infrastructures, and shipping). Significant knowledge gaps are identified, including the response of large
33
scale atmospheric circulation to climate change, but also concerning specific events, for example,
34
occurrences of marine heat waves and small-scale variability of precipitation. Suggestions for future
35
research includes further development of high-resolution Earth System models, and the potential use of
36
methodologies for data analysis (statistical methods and machine learning). With respect to expected
37
impact of climate change, changes are expected for sea-level, extreme precipitation, heat waves and
38
phytoplankton blooms (increase) and cold spells and severe ice winters (decrease). For some extremes
39
(drying, river flooding and extreme waves) the change depends on the area and time period studies.
40
1 Introduction
41
Natural hazards and extreme events may have severe implications on society including threat to human
42
life, economic losses and damaged ecosystems. A better understanding of their major causes and
43
implications enables society to be better prepared, to save human lives and mitigate economic losses.
44
Many natural hazards are of hydro-meteorological origins (storms, storm surges, flooding, droughts)
45
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
2
and impacts can sometimes be due to a mixture of several factors (e.g. a storm surge in combination
46
with heavy precipitation and river discharge).
47
In Europe in 2018, four severe storms caused almost 8bn$ losses (Munich Re, 2018), while a heatwave
48
and drought caused roughly 3.9bn$ losses. According to the EEA (European Environment Agency),
49
increase in frequency and/or magnitude of extreme events such as floods, droughts, windstorms or heat-
50
waves will be among the most important consequences of climate change (EEA 2010). Despite climate
51
change having received considerable scientific attention the knowledge on changing extremes and their
52
impacts is still to some extent fragmented, in particular when it comes to compound events (Zscheischler
53
et al., 2018). The confidence level of the knowledge of relation between global warming and hot
54
extremes is high while confidence is only medium for heavy precipitation/drought (IPCC, 2018).
55
Furthermore, the confidence level reduces when approaching the local scale (IPCC, 2014). Significant
56
advances have occurred, but the understanding of mechanistic drivers of extremes and how they may
57
change under anthropogenic forcing is still incomplete.
58
What is defined by “extreme” depends on the parameter and the application in relation thresholds for
59
generating extreme consequences in society or ecosystems. A large amount of the available scientific
60
literature is based on extreme indices, which are either based on the probability of occurrence of given
61
quantities or on threshold exceedances. Typical indices include the number, percentage, or fraction of
62
days of occurrence below the 1st, 5th, or 10th percentile, or above the 90th, 95th, or 99th percentile,
63
generally defined for given time frames (days, month, season, annual) with respect to the 1961-1990
64
reference time period (Seneviratne et al., 2012). Using predefined extreme indices allow for
65
comparability across modelling and observational studies and across regions. Peterson and Manton
66
(2008) discuss collaborative international monitoring efforts employing extreme indices. Extreme
67
indices often reflect relatively moderate extremes, for example, events occurring during 5 or 10% of the
68
time. For more rare extremes Extreme Value Theory (EVT) is often used due to sampling issues. EVT
69
(e.g., Coles, 2001), aims at deriving a probability distribution of events from the upper or lower tail of
70
a probability distribution (typically occurring less frequently than once per year or per period of
71
interest). Some literature has used other approaches for evaluating characteristics of extremes or changes
72
in extremes, for instance, analyzing trends in record events or investigating whether records in observed
73
time series are being set more or less frequently than would be expected in an unperturbed climate
74
(Benestad, 2003, 2006; Zorita et al., 2008; Meehl et al., 2009c; Trewin and Vermont, 2010). Besides
75
the actual magnitude of extremes (quantified in terms of probability/return frequency or absolute
76
threshold), other relevant aspects from an impact perspective include the duration, the spatial area
77
affected, timing, frequency, onset date and continuity (i.e., whether there are ‘breaks’ within a spell).
78
There is thus no precise definition of an extreme (e.g. Stephenson et al., 2008). In particular, there are
79
limitations in the definition of both probability-based or threshold-based extremes and their relations to
80
impacts. In the reviewed literature a range of definitions are used.
81
The Baltic Sea watershed drains nearly 20% of European land areas (see Figure 1). It ranges from the
82
highly populated south, with a temperate climate and intensive agriculture and industry, to the north,
83
where the landscape is boreal and rural. Changes in the recent climate as well as probable future climate
84
change of mean parameters in the Baltic Sea region are relatively well described (e.g. BACC I, 2008;
85
BACC II 2015; Rutgersson et al., 2014), but the uncertainty is larger for extreme events due to larger
86
statistical uncertainties for rare event. Natural hazards and extreme events have been identified as one
87
of the grand scientific challenges for the Baltic Sea research community (Meier et al., 2014).
88
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
3
Changes in extreme events can be caused by a combination of changes in local/regional conditions with
89
changes of the larger scale; atmospheric circulation patterns are thus of crucial importance. Extreme
90
events occur over a wide range of scales in time and space; short term events range from sub daily to a
91
few days (basically meso-scale and synoptic scale events) while long-lasting events range from a few
92
days to several months. There is no clear separation between short term and long term events and
93
sometimes the presence of a long-term event may intensify the impact of a short-term one. We here
94
summarise existing knowledge of extreme events in the Baltic Sea region. We focus on past and present
95
state, as well as future climate scenarios and expected changes when possible. The events considered
96
here include wind storms, extreme waves, high and low sea level, ice ridging, heavy precipitation, sea-
97
effect snowfall, river floods, heat waves, ice seasons, and drought. We also address some ecological
98
extremes and implications of extreme events for society (phytoplankton blooms, forest fires, coastal
99
flooding, offshore infrastructures, and shipping). The text focuses on the current base of knowledge, but
100
also identifies knowledge gaps and research needs.
101
102
Since almost three decades, the knowledge on the Baltic Sea ecosystem has also been systematically
103
assessed, initially by BALTEX and since 2013 by its successor Baltic Earth. As a result, two
104
comprehensive assessment reports have been released: BACC I (2008) and BACC II (2015). The
105
present study is one of the thematic Baltic Earth Assessment Reports (BEARs), which comprises a
106
series of review papers that summarize and assess the available published scientific knowledge on
107
climatic, environmental and human-induced changes in the Baltic Sea region (including its
108
catchment). As such, the series of BEARs constitutes a follow-up of the previous BACC assessments.
109
BEARs are constructed around the Grand Challenges and scientific topics Baltic Earth
110
(baltic.earth/grandchallenges) with a general summary (Meier et el., 2021).
111
1.1 Methods, past and present conditions
112
For the past and present conditions, we focus on time periods covering up to the last 200 years, to rely
113
on robust in situ measurements only (not proxy data). The Baltic Sea area is relatively unique in terms
114
of long-term data, with a dense observational network (compared to most regions) covering an extended
115
time period, although many national (sub-) daily observations still await digitization and
116
homogenization. The network of stations with continuous and relatively accurate measurements has
117
been developed since the middle of the 19th century (few stations were established in the middle of the
118
18th century). The period since about 1950 is relatively well covered by observational data. For some
119
applications (i.e. heavy precipitation) the relatively low frequency of sampling is a limitation; this was
120
improved with the establishment of automatic stations at the end of the 20th century. In spite of the
121
relatively good observational coverage over a long time, lack of observations is a major obstacle for
122
assessing long-term trends and past extreme events and for climate model evaluation. The density of the
123
observational network is high compared to many regions, but still low compared to the resolution
124
required for evaluation of today’s most fine-scale climate models. Despite shortcomings, a number of
125
high-resolution gridded data sets derived from point-based observations exist at resolutions as high as a
126
few km for parts of the Baltic Sea region.
127
The inclusion of satellite data since 1979 added to the spatial information, particularly over data-sparse
128
regions. However, data that spans extended periods cannot be expected to be homogeneous in time. This
129
is particularly important for the increasing number of re-analyses products that are available for the
130
region. In a reanalysis, all available observations are integrated as increments into a numerical model
131
by means of data assimilation in space and time. This works fine if the overall structure of the observing
132
system does not change dramatically over time; however, when completely new observing systems (for
133
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
4
example observations from satellites) are introduced, this structure changes. Making use of all available
134
observations, a frozen scheme for the data assimilation of observations into state-of-the-art climate
135
models is used to minimise inhomogeneities caused by changes in the observational record over time.
136
However, studies indicate that these inhomogeneities cannot be fully eliminated (e.g., Stendel et al.,
137
2016). In addition, systematic differences between the underlying forecast models, such as due to their
138
different spatial resolutions (Trigo 2006; Raible et al. 2008) and differences in detection and tracking
139
algorithms (Xia et al. 2012) may affect parameters like cyclone statistics (for example changes in their
140
intensity, number and position). Reanalysis products includes NCEP/NCAR (from 1948 onwards;
141
Kalnay et al. 1996; Kistler et al. 2001), ERA-Interim, starting in 1979 (Dee et al. 2011) or, more recently,
142
ERA5 (Hersbach et al. 2020). Other reanalyses use a limited data assimilation scheme to go further back
143
in time, like the 20th Century Reanalysis 20CR (from 1871 onwards; Compo et al. 2011) or CERRA
144
(Schimanke et al., 2019). On the regional scale, detailed regional reanalysis with higher resolution
145
models and more observations have been developed (e.g. Dahlgren et al., 2016; Kaspar et al., 2020).
146
1.2 Methods, future scenarios
147
The development of general circulation models (GCMs) has created a useful tool for projecting how
148
climate may change in the future. Such models describe the climate at a set of grid points, regularly
149
distributed in space and time. In some cases, also dynamical downscaling with regional models or
150
empirical-statistical downscaling using statistical models are used. A large multi-model coordinated
151
climate model experiment, CMIP Project Phase was initiated, currently version 5 (CMIP5, Taylor et al.,
152
2012) is the main source of information while the next phase CMIP6 (Eyring et al., 2016) is increasingly
153
being used. Coordinated downscaling activities including regional climate models (RCMs) include
154
those of the European research projects PRUDENCE (Déqué et al. 2007) and ENSEMBLES (Kjellström
155
et al. 2013) as well as the WCRP supported international CORDEX project with its European branch
156
EURO-CORDEX (Jacob et al., 2014).
157
Projections of climate change depend inherently on scenario assumptions of future human activities.
158
Widely used are the Representative Concentration Pathways (RCPs) (van Vuuren et a., 2011). An RCP
159
represents a climate forcing scenario (e.g. including changes in greenhouse gas emissions, aerosols,
160
land use etc.) trajectory adopted by the IPCC for its Fifth Assessment Report (AR5) in 2014. RCPs
161
describe different climate futures, all of which are considered possible depending on how strong the
162
forcing of the climate system is. The four RCPs used for AR5, namely RCP2.6, RCP4.5, RCP6, and
163
RCP8.5, are labelled after their associated radiative forcing values in the year 2100 (2.6, 4.5, 6.0, and
164
8.5 W/m2, respectively, (Moss et al, 2008; Weyant et al., 2009) relative to that in pre-industrial times,
165
e.g. 1750. RCP4.5 is used in many studies assuming increasing carbon dioxide emissions until 2040 and
166
after that decreasing. RCP8.5 assumes a continuously growing population and rapidly increasing carbon
167
dioxide and methane emissions and is increasingly seen as an unlikely worst-case scenario (Hausfather
168
and Peters, 2020). Prior to the RCPs, scenarios from the Special Report on Emission Scenarios
169
(Nakicenovic et al., 2000) were widely used. The main scenario families included were: A1,
170
representing an integrated world with rapid economic growth; A2, a more divided world with regional
171
and local focus; B1, representing an integrated and more ecologically friendly world; B2 of a divided
172
but more ecologically friendly world.
173
2 Current state of knowledge
174
2.1 Changes in circulation patterns
175
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
5
The atmospheric circulation in the European/ Atlantic sector plays an important role for the regional
176
climate of the Baltic Sea basin and the surrounding areas (e.g., Hurrell 1995; Slonosky et al. 2000,
177
2001). Large-scale flow characteristics is one of the main drivers of the connection between local
178
processes and global variability and change. It is therefore essential to investigate the changes in large
179
scale flow. The main driver is the NAO (Hurrell et al. 2003), with quasi-stationary centers of action, the
180
Icelandic Low and the Azores High, it is a measure of the zonality of the atmospheric flow. The
181
dominant flow is westerly, but due to the large variability also other wind directions are frequently
182
observed.
183
The strength of the westerlies is controlled by the pressure difference between the Azores High and the
184
Icelandic Low (Wanner et al. 2001; Hurrell et al. 2003; Budikova 2009) and is expressed by the NAO
185
index, which is the normalized pressure difference between these two regions. The NAO index varies
186
from days to decades. The long-term (1899-2018) temporal behavior of the NAO (Fig. 2) is essentially
187
irregular, and there is large interannual to interdecadal variability, reflecting interactions with and
188
changes in surface properties, including sea surface temperature (SST) and sea ice content (SIC). While
189
it is not clear whether there is a trend in the NAO, for the past five decades, specific periods are apparent.
190
Beginning in the mid-1960s, a positive trend towards more zonal circulation with mild and wet winters
191
and increased storminess in central and northern Europe, including the Baltic Sea area has been observed
192
(Hurrell et al. 2003, Gillett et al., 2013). After the mid-1990s, however, there was a tendency towards
193
more negative NAO indices, in other words a more meridional circulation and more cold spells in
194
winter, which can only occur with winds from an easterly or a northerly direction (see section 2.2.7).
195
Other studies (e.g., Deser et al., 2017; Marshall et al., 2020) do not find a significant trend. It has been
196
speculated that these changes are due to a shift of the Atlantic Multidecadal Variability (AMO) into the
197
warm phase (Gastineau and Frankignoul, 2015).
198
Most of the state-of-the-art climate models reproduce the structure and magnitude of the NAO
199
reasonably well (e.g. Davini and Cagnazzo, 2014; Ning and Bradley, 2016; Deser et al., 2017; Gong et
200
al., 2017).
201
There is no consensus on how large a fraction of the interannual NAO variability is forced externally
202
(Stephenson et al. 2000; Feldstein 2002; Rennert and Wallace 2009). Several such external forcing
203
mechanisms have been proposed, including SST (Rodwell et al. 1999; Marshall et al. 2001), volcanoes
204
(Fischer et al. 2007), solar activity (Shindell et al. 2001; Spangehl et al. 2010; Ineson et al. 2011), and
205
stratospheric influences (Blessing et al. 2005; Scaife et al. 2005), including the quasi-biennial oscillation
206
(Marshall and Scaife 2009) and stratospheric water vapour trends (Joshi et al. 2006). Remote SST
207
forcing of the NAO originating from as far as the Indian Ocean was proposed by Hoerling et al. (2001)
208
and Kucharski et al. (2006), while Cassou (2008) proposed an influence of the Madden-Julian
209
Oscillation. In addition, Blackport and Screen (2020) showed that recent observations suggest that the
210
observed correlation between surface temperature gradients and circulation anomalies in the middle
211
troposphere have changed in recent years.
212
Regarding sea ice, many authors have found an effect of sea ice decline on the NAO (Strong and
213
Magnusdottir 2011; Peings and Magnusdottir, 2016; Kim et al., 2014; Nakamura et al., 2015), while
214
others (Screen et al., 2013; Sun et al., 2016; Boland et al., 2017) do not identify any dependence on
215
changing sea ice extent. Furthermore, the interaction of changes in the Arctic on midlatitude dynamics
216
are still under debate (Dethloff et al. 2006; Francis and Vavrus, 2012; Barnes, 2013; Cattiaux and
217
Cassou, 2013; Vihma, 2017).
218
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
6
Atmospheric blocking refers to persistent, quasi-stationary weather patterns characterized by a high-
219
pressure (anticyclonic) anomaly that interrupts the westerly flow in the mid-latitudes. By redirecting the
220
pathways of mid-latitude cyclones, blockings lead to negative precipitation anomalies in the region of
221
the blocking anticyclone and positive anomalies in the surrounding areas (Sousa et al., 2017). In this
222
way, blockings can also be associated with extreme events such as heavy precipitation (Lenggenhager
223
et al., 2018) and drought (Schubert et al., 2014).
224
A weakening of the zonal wind, eddy kinetic energy and amplitude of Rossby waves in summer
225
(Coumou et al., 2015) as well as an increased waviness of the jet stream associated with Arctic warming
226
(Francis and Vavrus, 2015) in winter have been identified, which may be linked to an increase in
227
blocking frequencies. Blackport and Screen (2020) argue that observed correlations between surface
228
temperature gradients and the amplitude of Rossby waves have broken in recent years. Therefore,
229
previously observed correlations may simply have been internal variability. On the other hand, it has
230
been shown that observed trends in blocking are sensitive to the choice of the blocking index, and that
231
there is a huge natural variability that complicates the detection of forced trends (Woollings et al., 2018),
232
compromising the robustness of observed changes in blocking. A review by Overland et al. (2015)
233
concluded that mechanisms remain uncertain as there are many dynamical processes involved, and
234
considerable internal variability masks any signals in the observation record. There is weak evidence
235
that stationary wave amplitude has increased over the north Atlantic region (Overland et al., 2015),
236
possibly as a result of weakening of the north Atlantic storm track and transfer of energy to the mean
237
flow and stationary waves (Wang et al., 2017).
238
The decrease of the poleward temperature gradient will lead to a weakening of westerlies and increase
239
the likelihood for blockings. On the other hand, maximum warming (compared to other tropospheric
240
levels) will occur just below the tropical tropopause due to the enhanced release of latent heat, which
241
tends to increase the poleward gradient, strengthen upper-level westerlies and affect the vertical
242
stability, thus altering the vertical shear in mid-latitudes. It is not clear which of these two factors has
243
the largest effect on the jet streams (Stendel et al., 2020).
244
State-of-the-art models are generally able to capture the general characteristics of extratropical cyclones
245
and storm tracks, although many of them underestimate cyclone intensity and still exhibit comparatively
246
large biases in the Atlantic/European sector (Davini and d’Andrea 2016, Mitchell et al., 2017). It was
247
already stated by IPCC (2013) that this is resolution-related (IPCC 2013; Zappa et al., 2013). In addition,
248
there is evidence for a correlation of the quality of simulations of cyclones and of blockings (Zappa et
249
al. 2014).
250
There is significant natural variability of the atmospheric circulation over Europe on decadal time scales
251
(Dong et al., 2017; Ravestein et al., 2018). Drivers for circulation changes have been proposed,
252
including polar and tropical amplification, stratospheric dynamics and the Atlantic Meridional
253
Overturning Circulation (AMOC) (Haarsma et al., 2015; Shepherd et al., 2018; Zappa and Shepherd,
254
2017). For more local changes, the attribution is more straightforward, where one example is the soil
255
moisture feedback, for which an enhancement of heat waves due to a lack of soil moisture has been
256
demonstrated (Seneviratne et al., 2013; Teuling, 2018; Whan et al., 2015).
257
Räisänen (2019) find only a small impact of circulation changes on the observed annual mean
258
temperature trends in Finland, but circulation changes have considerably modified the trends in
259
individual months. In particular, changes in circulation explain the lack of observed warming in June,
260
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
7
the very modest warming in October in southern Finland, and about a half of the very large warming in
261
December.
262
On a more global scale, CMIP5 simulations suggest enhanced drying and consequently an increase of
263
summer temperatures due to more meridional circulation which would result in extra drying, in
264
particular in spring. If that is the case, the summer soil moisture feedback would be enhanced (van der
265
Linden et al., 2019; van Haren et al., 2015). Soil drying, e.g. under extended blocking situations, would
266
lead to nonlinear interactions between atmosphere and land resulting in further temperature increase
267
(Douville et al., 2016; Douville and Plazzotta, 2017; Seneviratne et al., 2013; Teuling, 2018; van den
268
Hurk et al., 2016).
269
2.2 Extreme conditions (current knowledge, now and potential future change)
270
2.2.1 Winds storms
271
In situ observations allow direct analysis of winds, in particular over sea (e.g., Woodruff et al. 2011).
272
However, in situ information, especially over land, is often locally influenced, and inhomogeneities
273
make the straightforward use of these data difficult, even for recent decades. Examples include an
274
increase in roughness length over time due to growing vegetation or building activities, inhomogeneous
275
wind data over the German Bight from 1952 onwards (Lindenberg et al. 2012) or ‘atmospheric stilling’
276
in continental surface wind speeds due to widespread changes in land use (Vautard et al. 2010). Many
277
studies turn down direct wind observations and instead rely on reanalysis products (see section 1.1).
278
However, analysis of storm-track activity for longer periods using reanalysis data suffers from
279
uncertainties associated with changing data assimilation and observations before and after the
280
introduction of satellites, resulting in large variations across assessments of storm-track changes (Chang
281
and Yau, 2016; Wang et al., 2016).
282
Owing to the large climate variability in the Baltic Sea region, it is unclear whether there is a trend in
283
wind speed. Results regarding changes or trends in the wind climate are thus strongly dependent on the
284
period and region considered (Feser et al. 2015a,b). Through the strong link to large-scale atmospheric
285
variability over the North Atlantic, conclusions about changes over the Baltic Sea region are best
286
understood in a wider spatial context, considering the NAO. The positioning of the jet stream and storm
287
tracks and the strength of the north–south pressure gradient in the North Atlantic can largely explain the
288
decadal changes in 10-m wind speeds in Northern Europe, with low windiness in winters of the 1980s
289
and 2010s and high windiness of the 1990s (Laurila et al., 2021).
290
Recent trend estimates of the total number of cyclones over the NH extratropics during 1979-2010 reveal
291
a large spread across the reanalysis product, strong seasonal differences, as well as decadal-scale
292
variability (Tilinina et al., 2013; Wang et al., 2016; Chang et al., 2016; Matthews et al., 2016; Gregow
293
et al., 2020). Common to all reanalysis datasets is a weak upward trend in the number of moderately
294
deep and shallow cyclones (7 to 11% per decade for both winter and summer), but a decrease in the
295
number of deep cyclones in particular for the period 1989-2010. Chang et al. (2016a) have reported a
296
minor reduction in cyclone activity in Northern Hemisphere summer due to a decrease in baroclinic
297
instability as a consequence of Arctic temperatures rising faster than at low latitudes. Chang et al.
298
(2016b) also notice that state-of-the art models (CMIP5) generally underestimate this trend. In Northern
299
Hemisphere winter, recent studies claim an increase in storm track activity related to Arctic warming.
300
Recent research (Feser et al. 2021) reveals no clear trend, but an increasing similarity over time in
301
reanalyses, observations and dynamically downscaled model data.
302
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
8
Despite large decadal variations, there is still a positive trend in the number of deep cyclones over the
303
last six decades, which is consistent with results based on NCEP reanalyses between 1958 and 2009
304
over the northern North Atlantic Ocean (Lehmann et al. 2011). Using an analogue-based field
305
reconstruction of daily pressure fields over central to northern Europe (Schenk and Zorita 2012), the
306
increase in deep lows over the region might be unprecedented since 1850 (Schenk 2015). For limited
307
areas the conclusions are rather uncertain. Past trends in homogenized wind speed time series (1959-
308
2015), in both mean and maximum, have been generally negative in Finland (Laapas et al., 2017).
309
The role of differential temperature trends on storm tracks has been recently addressed, both in terms of
310
upper tropospheric tropical warming (Zappa and Shepherd, 2017) and lower tropospheric Arctic
311
amplification (Wang et al., 2017), including the direct role of Arctic sea ice loss (Zappa et al., 2018),
312
and a possible interaction of these factors (Shaw et al., 2016). The remote and local SST influence has
313
been further examined by Ciasto et al. (2016), who further confirmed sensitivity of the storm tracks to
314
the SST trends generated by the models and suggested that the primary greenhouse gas influence on
315
storm track changes was indirect, acting through the greenhouse gas influence on SSTs. The importance
316
of the stratospheric polar vortex in storm track changes has received more attention (Zappa and
317
Shepherd, 2017). In an aqua-planet simulation, Sinclair et al. (2020) find a decrease in the number of
318
extratropical cyclones and a poleward and downstream displacement due to an increase in diabatic
319
heating.
320
A projection of future behavior of extratropical cyclones is impeded by the fact that several drivers of
321
change interact in opposing ways. With global warming, the temperature gradient between low and high
322
latitudes in the lower troposphere decreases due to polar amplification. Near the tropopause and in the
323
lower stratosphere, the opposite is true, thus implying changes in baroclinicity (Grise and Polvani 2014,
324
Shaw et al 2016, Stendel et al., 2020). An increase in water vapour enhances diabatic heating and tends
325
to increase the intensity of extratropical cyclones (Willison et al. 2015, Shaw et al. 2016) and contribute
326
to a propagation further poleward (Tamarin and Kaspi 2017). The opposite is also true in parts of the
327
North Atlantic region, e.g. south of Greenland. For this region the N-S gradient is rather increasing as
328
the weakest warming in the entire NH is over ocean areas south of Greenland. North of this local minima
329
the opposite is true. The increase in the N-S gradient over the N. Atlantic may be responsible for some
330
GCMs showing an intensification of the low pressure activity and thereby high wind speed over a region
331
from the British Isles and through parts of north-central Europe (Leckebusch and Ulbrich, 2004).
332
So, in summary, there is no clear consensus in climate change projections in how far changes in
333
frequency and/or intensity of extratropical cyclones have an effect on the Baltic Sea region.
334
Wind storms can also be accompanied by wind gusts (downbursts), potentially causing severe damage.
335
Wind gusts driven by convective downdrafts or turbulent mixing can also occur during larger-scale
336
windstorms, like Mauri in 1982 (Laurila et al., 2019). There is limited information concerning past or
337
future trends concerning occurrence of wind gusts.
338
2.2.2 Extreme waves
339
Vertical motions on the ocean surface consists of an extensive spectrum of frequencies and periods
340
(Munk 1950, Holthuijsen 2007). Here we focus on the wind generated waves and mainly on the
341
significant wave height representing the average height of the highest third of the waves. Significant
342
wave height serves as an indicator when discussing extreme waves, however, the highest individual
343
wave in a wave record is 1.6-2.0 times higher than significant wave height (Björkqvist et al. 2018,
344
Pettersson et al. 2018). Some ambiguity exists when it comes to which sea states can be called extreme
345
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
9
(Hansom et al. 2015) because locally higher wave heights in not particularly stormy conditions can lead
346
to damages and fatalities and may become labeled in the media as extreme, giant, freak, monster or
347
rogue waves. Rogue waves are typically defined as a maximum wave height of more than two times the
348
significant wave height. The horizontal resolution of wave modeling hindcast studies for the Baltic Sea
349
have varied from about 1.1-1.85 km to about 22 km (Nilsson et al., 2019; Björkqvist et al., 2018; Jönsson
350
et al. 2003). The small-scale spatial and time variations are often missed by the models and coarse
351
resolution (6-11 km) may not provide sufficient accuracy to study extremes (Larsén et al. 2015;
352
Björkqvist et al. 2018).
353
On 12 January 2017, an intensive low-pressure system generated a wave in the northern Baltic Sea
354
referred to in the media as a monster wave above 14 m, equaling or exceeding the previous record from
355
December 22nd 2004 (EUMETSAT 2017, Björkqvist et al. 2018). Significant wave heights measured
356
around 8 m according to the Finnish Meteorological Institute (FMI). Even higher waves with significant
357
wave heights up to 9.5 m have been estimated to occur in the northern Baltic proper during the wind
358
storm Gudrun in January 2005 (Soomere et al. 2008, Björkqvist et al. 2018). A high-resolution
359
numerical model study for the time period 1965 to 2005 (Björkqvist et al. 2018) showed a 99.9th
360
percentile for significant wave height in the Baltic Sea of 6.9 m. They found 45 unique extreme wave
361
events with modeled significant wave height above 7 m during the 41 year-simulation. Twelve of which
362
had a maximum above 8 m, six events exceeded 9 m, and one event showed significant wave height
363
over 10 m. Extreme waves in the Baltic Sea can have a significant impact on sea level dynamics and
364
coastal erosion also discussed further in Weisse et al. (2021).
365
Many studies have been conducted to characterize the variations in the wave fields using measurements
366
(e.g. Kahma et al. 2003, Pettersson and Jönsson 2005, Broman et al. 2006) and using modeling (e.g.
367
Jönsson et al. 2003, Räämet and Soomere, 2010, Björkqvist et al. 2018) describing also the seasonal
368
dependence (e.g. Soomere 2008, Räämet and Soomere, 2010). Björkqvist et al. (2018) showed that 84%
369
of wave events with significant wave heights above 7 m occurred during the months November until
370
January. The areas of highest significant wave heights are found in the southern and eastern Baltic
371
Proper (Björkqvist et al., 2018). This is consistent with the typical synoptic weather pattern of middle
372
latitudes but modulated by bathymetry and fetch conditions, as well as meso-scale weather effects
373
(Soomere 2003, Nilsson et al. 2019). The pattern of 100-year return value estimates of significant wave
374
height, based on 10 km resolution simulations for 1958-2009, is represented here by the 99.9th percentile
375
significant wave height in Figure 3 (in agreement with Aarnes et al., 2012; Björkqvist et al., 2018;
376
Nilsson et al., 2020). The northern basins typically experience reduced wave heights, both due to the
377
shorter fetch conditions, as well as the occurrence of sea-ice limiting the wave growth during the season
378
when the highest waves otherwise can be expected to occur (e.g. Tuomi et al. 2019, Nilsson et al. 2019).
379
Some studies have also been conducted on near-shore extreme waves; e.g. Gayer et al. 1995, Paprota et
380
al. 2003, Sulisz et al. 2016 discussed the formation of extreme waves and wave events along Polish and
381
German coasts and reported a large number of freak-type waves. Although significant progress in
382
understanding and prediction of ocean extremes and freak waves (e.g. Cavaleri et al. 2017, Janssen et
383
al. 2019) have been achieved, a practical definition using usually more well-predicted parameters, such
384
as significnat wave height, is presently used in warnings (Björkqvist et al. 2018).
385
From long-term in-situ observations and modeling results trends in wave climate are inconclusive and
386
possibly site-specific (e.g. Soomere and Räämet 2011b). From reviewing multiple studies discussing
387
changes and trends in significant wave heights at Baltic Sea sites across time periods of more than 30
388
years there is often no clear trend in severe wave heights or the trends are small and explained by the
389
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
10
large natural variability in the wind climate (section 2.1 and 2.2.1) (e.g. Räämet et al. 2010, Soomere et
390
al. 2012; Soomere and Räämet 2011a). Trends in mean wave height are small but statistically significant
391
(0.005 m/year for 1993-2015) from satellite altimetry (Kudryatseva and Soomere, 2017) but higher
392
quantiles behaved less predictable. A spatial pattern with an increase in the central and western parts of
393
the sea and a decrease in the east was observed.
394
For the wave field in a future climate Mentaschi et al. (2017) reported an increase of extreme wave
395
energy flux (on average 20%, with maxima up to 30%). They used a global wave model (approximately
396
1.5-degree resolution) driven by an ensemble of global coupled models from the CMIP5 under the high
397
emission RCP scenario 8.5. They suggest the changes are caused by changes in the NAO index. Groll
398
et al. (2017) analysed wave conditions under two IPCC AR4 emission scenarios (A1B and B1) by
399
running a more high-resolution wave model and implementing effects of sea-ice through ice-covered
400
grid cells if ice thickness was larger than 5 cm. They found higher significant wave height in the future
401
for most regions and simulations. Median wave results showed temporal and spatially consistent
402
changes (sometimes larger than 5% and 10%), whereas extreme waves (99th percentile) showed more
403
variability in space and among the simulations and these changes were smaller (mostly less than 5% or
404
10%), and more uncertain. The changes reported were attributed to higher wind speeds, and also from
405
a shift to more westerly winds. The sea-ice was clearly reduced in the Bothnian Sea, Bothnian Bay and
406
Gulf of Finland in the simulations but changes in the 30-year mean of annual wind speed maximum
407
showed a decrease in the northern Baltic Sea. Multi-decadal and the inter-simulation variability
408
illustrated the uncertainty in the estimation of a climate change signal (Dreier et al., 2015; Groll et al.
409
2017).
410
Simulations of sea-ice variations in a warmer climate may be one of the most factors determining the
411
future wave field. If significant reduction of ice in the northern Baltic Sea basin occurs, changes to the
412
wave field are likely unless compensated for by changing wind patterns (Groll et al. 2017). Zaitseva-
413
Pärnaste and Soomere (2013) showed significant correlation between energy flux and ice season.
414
Comparing ice-free and ice-time included statistics, ice-free conditions increased significant wave
415
heights on the order of about 0.3 m both for mean values and 99th percentile values (Tuomi et al. 2011,
416
Björkqvist et al. 2018). Fairly small anthropogenic effects for the wave fields are expected for the next
417
century but results are uncertain and depend on both changes in wind climate (section 2.1 and 2.2.1)
418
and ice conditions (section 2.2.10 and 2.2.11).
419
2.2.3 Sea level
420
The rising global mean sea level poses a major hazard for the population living in the vicinity of the
421
coast and will compound the risk of coastal floods. The effects of climate change on wind climate and
422
tidal extremes may lead to further increases in the frequency and intensity of extreme sea levels on top
423
of the mean sea level rise. Even if the sea level extremes only last a limited time, they are capable of
424
causing large damage to the coastal infrastructure and endanger human lives. Likewise, extreme sea
425
levels are a major threat to coastal areas along the Baltic Sea coast due to flooding and erosion. Hence,
426
sand dunes may experience large deformations during a single storm.
427
In the Baltic Sea, extreme sea levels are caused by wind, air pressure (inverse barometric effect) and
428
seiche. The Danish straits prevent the entrance of tidal waves into the Baltic Sea, and the amplitude of
429
the internal tides is only a few centimeters. Only exceptions are the southwestern Baltic Sea and the
430
eastern Gulf of Finland, where tides can reach 20 cm (Medvedev et al., 2016). The water exchange
431
between the North Sea and the Baltic Sea causes about maximum 1 m variation in monthly mean sea
432
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
11
levels (Leppäranta and Myrberg 2009). Due to the shape of the Baltic Sea, the highest and lowest sea
433
levels are found in the end of the bays, as in the eastern end of the Gulf of Finland, northern end of the
434
Gulf of Bothnia, and in the Gulf of Riga, whereas the amplitude of variation is smallest in the central
435
Baltic Sea. The Baltic Sea areas with the largest sea level variations, based on tide gauge data 1960-
436
2010, are shown in Fig. 4 (from Wolski et al. 2014).
437
The observed maxima and minima on the Baltic Sea coast along with 100-year return levels based on
438
interpolated coastal tide gauge observations 1960-2010 were studied by Wolski et al. (2014). They
439
observed an increase in the yearly number of storm surges (defined as sea levels 70 cm above zero level
440
of the European Vertical Reference Frame or local mean sea level in Finland and Sweden). The increase
441
was largest in the Gulf of Finland (Hamina and Narva) and in the Gulf of Riga (Pärnu). Ribeiro et al.
442
(2014) investigated the changes in extreme sea levels in 1916-2005 from detrended daily tide gauge
443
records of seven stations in Denmark and Sweden on the Baltic Sea coast, using GEV (Generalized
444
Extreme Value) and quantile regression methods. They observed a statistically significant trend in
445
annual sea level maxima in the Gulf of Bothnia (1.9 mm/year for Ratan and 2.6 mm/year for
446
Furuögrund). For other locations, the maxima could be considered stationary. Marcos and Woodworth
447
(2017) studied the tide gauge data concluding that the changes in the 100-year return levels after 1960
448
in the Baltic Sea were explained by the mean sea level rise. Projected extreme sea levels for the Baltic
449
Sea coast in 2100 were calculated by Vousdoukas et al. (2016) considering only the effect of the
450
atmosphere on the sea level (storm surges) while omitting global mean sea level rise and land uplift.
451
The Delft3D sea level model was forced with 8 global climate models from CMIP5 database, and the
452
projected changes were calculated from ensemble means of model simulations. In 2100, the present-day
453
100-year storm surge was projected to take place every 72 years under RCP4.5 and every 44 years under
454
RCP8.5. The ensemble means of storm surges (return periods from 5 to 100 years) increase along the
455
northern Baltic Sea coast with time for both RCPs. The increase is largest in the Bothnian Bay and in
456
the Gulf of Finland, reaching about 0.5 m. Along the southern Baltic Sea coast, there is smaller or no
457
increase in most scenarios. When the storm surges are averaged over the Baltic Sea coast, the increase
458
in the storm surges of return periods from 5 to 500 years is only 10-20 cm for different scenarios. By
459
2100, the inter-annual variation in the seasonal maxima, indicated by the standard deviation, increased
460
by 6 per cent in RCP4.5 and by 15 per cent in RCP 8.5. This indicated that the variations in the maxima
461
might increase more than the 30-year mean, suggesting that the maxima could have a higher increasing
462
trend than the mean sea level has. The extreme sea levels along Europe’s coasts, caused by the combined
463
effect of mean sea level, tides, waves and storm surges were studied by Vousdoukas et al. (2017). In the
464
Baltic Sea, the 100-year sea level due to waves and storm surges was projected to rise 35 cm (average
465
over the Baltic coast) by 2100 in RCP8.5. The rise is largest in the eastern coast of the Baltic Sea, and
466
the intra-model variation of the 100-year level increases up to 0.6 meters in 2100. The large variation
467
between the models causes a large uncertainty in the evaluation of the change in extreme sea levels
468
during the present century. These sea level estimates should be considered preliminary. To increase the
469
confidence in the future projections of storm surges in the Baltic Sea, we must rely on future research
470
where a larger set of regional and global climate models are used with refined sea level models. The
471
dependence between extreme sea levels and wind waves has to be assessed when the joint effect of
472
storm surge and wave setup on the coast is studied. For the Baltic Sea, this dependence should be
473
included when joint probabilities of compound events of high sea levels and waves are calculated, as is
474
done in Kudryavtseva et al. (2020). An extended discussion on sea-levels is seen in Weisse et al. (2021).
475
2.2.4 Precipitation
476
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
12
Precipitation extremes in the Baltic Sea region are mainly related to i) synoptic-scale mid-latitude low
477
pressure systems and ii) convective precipitation events associated with meso-scale convective systems
478
or resulting from single intense cloudbursts. Additionally, sea-effect snow fall events can generate large
479
amounts of snow in coastal areas downstream of the Baltic Sea (section 2.2.5). Climatologically,
480
summer is the season with the strongest convective activity and this is also the season with the strongest
481
cloudbursts. Precipitation extremes associated with low pressure systems are most frequent in fall and
482
winter when the large-scale atmospheric circulation is favorable for bringing low-pressure systems
483
towards northern Europe.
484
High-resolution gridded data sets that may be used for evaluation of climate model performance for
485
precipitation includes: PTHBV covering Sweden at 4 km grid (Johansson and Chen, 2005); the Finnish
486
data set at 1 km and10 km grid by Aalto et al. (2016) ; the REGNIE data set at 1 km grid covering
487
Germany (Rauthe et al., 2013); CPLFD-GDPT5 for Poland at 5 km (Berezowski et al., 2016) and
488
seNorge2 for Norway at 1 km grid (Lussana et al., 2018). Another recent data set is the joint product
489
consisting of PTHBV data in combination with precipitation estimates from radar data over Sweden
490
resulting in the 4x4 km, one hourly resolution HIPRAD (HIgh-resolution Precipitation from gauge-
491
adjusted weather RADar) data set covering 2009-2014 (Berg et al., 2016). Finally, it is noted that these
492
national data sets are derived with slightly different methods implying that they cannot directly be
493
compiled and used as one high-resolution data set for the entire Baltic Sea region.
494
Representing the strong spatial and temporal variability of precipitation constitutes a true challenge for
495
climate models and careful evaluation against observations is key before the models can be applied.
496
Typically, large-scale features such as the total precipitation volume over the Baltic Sea region are
497
relatively well captured by climate models even at coarser resolution as shown for a regional climate
498
model at 50 km resolution by Lind and Kjellström (2009). However, such coarse-scale climate models
499
are limited in their ability of reproducing fine-scale details of the observed precipitation climate. Higher
500
resolution, for instance in the EURO-CORDEX ensemble (12.5km grid spacing) improves this (e.g.
501
Prein et al., 2016) but spatial details are still too coarsely represented to adequately address precipitation
502
over complex topography (e.g. Pontoppidan et al., 2017). In addition to spatial details also the simulation
503
of the diurnal cycle is often flawed in coarse-scale models (e.g. Walther et al., 2013). With even higher
504
horizontal resolution, so-called convection permitting models with grid spacing of a few km, are found
505
to improve the simulation of both spatial and temporal features of precipitation (e.g. Belušić et al, 2020).
506
Importantly, this involves also the representation of extreme events as they are much more capable of
507
representing high-intensity rainfall than their coarser-scale counterparts (e.g. Kendon et al., 2012;
508
Lenderink et al., 2019; Lind et al., 2020). For an example see Figure 5 showing how a convection-
509
permitting model improves the representation of precipitation over Sweden.
510
According to BACC I (2008) and BACC II (2015) precipitation trends in the Baltic Sea basin over the
511
past 100 years have varied in time and space. Examples exist of both increasing and decreasing trends
512
in different areas for different periods and seasons. Positive trends were detected for the cold part of the
513
year for Fennoscandia by Benestad et al. (2007) and Estonia, Latvia and Lithuania by Jaagus et al.
514
(2018). Along with warming it is also noted that the fraction of snowfall to total precipitation is
515
decreasing with time (Hynčica and Huth, 2019; Luomaranta et al. 2019).
516
Increasing intensity of precipitation events resulting from the larger water-holding capacity of a warmer
517
atmosphere is an expected impact of climate change (Bengtsson, 2010). Based on European E-OBS
518
data, Fischer and Knutti (2016) demonstrate that heavy daily precipitation, defined as the 99.9th
519
percentile that roughly corresponds to a one in three years event, has become 45% more frequent
520
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
13
comparing the last thirty years with the preceding 30 years. For even more extreme precipitation events
521
like one in ten, twenty or even fifty years the large variability makes it difficult to draw any firm
522
conclusions about changes especially for small areas with only few observational stations. For example,
523
Olsson et al (2017a) found no significant trend in annual maxima based on Swedish gauge data from
524
1880 to 2017, even when gauges were pooled across the whole country. For less intense events like the
525
90th, 95th and 99th percentiles of daily precipitation or the total number of days with more than 10 mm
526
of precipitation a number of studies have reported on increasing trends in Europe (e.g. Donat et al. 2016)
527
or parts of the Baltic Sea region for different seasons (e.g. BACC I (2008) and BACC II (2015) and
528
references therein).
529
Climate projections of future climate show increasing precipitation in northern Europe including the
530
Baltic Sea region (IPCC, 2013; BACC I; BACC II, 2015). Southern Europe, on the other hand, is
531
projected to receive less precipitation and as the border line between increasing and decreasing
532
precipitation moves from the south in winter to the north in summer there are some models that project
533
less precipitation in parts of the Baltic Sea region in summer (Christensen and Kjellström, 2018). In
534
addition to changes in mean precipitation projections show a similar north-south pattern of changes in
535
wet‐day frequency with increases in the north and decreases in the south (Rajczak et. al., 2013).
536
Regardless of sign of change in seasonal mean precipitation, heavy rainfall is projected to increase in
537
intensity for most of Europe including the Baltic Sea region (Nikulin et al., 2011; Rajczak et. al., 2013;
538
Christensen and Kjellström, 2018) as illustrated in Fig. 6. Snowfall is projected to decrease on an annual
539
mean basis but in winter daily snowfall amounts and extreme events may experience increases (Danco
540
et al. 2016). Precipitation intensities are projected to increase at durations ranging from sub-daily to
541
weekly. Martel et al. (2020), based on three large ensembles including one with a high-resolution
542
regional climate model, concludes that increases in 100-year return values of annual maximum
543
precipitation are strongest at sub-daily time scales than for 1-day or 5-day events. Newly developed
544
convective permitting regional climate models have been shown to sometimes yield different climate
545
change signals for extreme precipitation events compared to coarser scale models (> 10 km grid
546
spacing). For instance, Kendon et al. (2014) showed stronger increase in summertime intense
547
precipitation in a 1.5 km model compared to a 12 km on for the southern UK. Similarly, Lenderink et
548
al. (2019) showed stronger increase for intense precipitation in a number of summer months when
549
applying a synthetic warming signal of 2°C to the large-scale boundary conditions. Until now, such
550
models have not been applied for climate change studies to the Baltic Sea region and it is not clear what
551
the response to warming would be.
552
Stronger precipitation extremes associated with a warmer climate can have strong impacts on society.
553
Large amounts of precipitation are strongly associated with flooding which is common in the Baltic Sea
554
region. More intense cloud bursts are strongly associated with urban flooding but also with adverse
555
effects on agriculture and infrastructure in rural areas. Stronger climate change signals in recently
556
developed convective permitting models compared to previous state-of-the-art models can have strong
557
impacts for the provision of climate services and as advice in the context of climate change adaptation.
558
2.2.5 Sea-effect snowfall
559
Sea-effect (lake- or bay-effect) snowstorms may disrupt several sectors of society and can cause millions
560
of euros of damage (Juga et al. 2014). Intense and prolonged sea-effect snow events can produce tens
561
of centimeters of snow accumulation and last for days. In Northern Europe the transport systems are
562
most impacted by winter extremes, such as snowfall, cold spells and winter storms by increasing the
563
number of vehicle accidents, injuries and other damage, as well as leading to highly increased travel
564
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
14
times (Vajda et al., 2014; Groenemeijer et al. 2016). Critical infrastructures are affected by disturbances
565
in the emergency and rescue services as well as roof and tree damages and failures in power transmission
566
due to heavy snow loading. Road maintenance and snow transportation to disposal sites is costly if there
567
is not enough space for snow storage along the streets (Keskinen 2012).
568
The impacts of a sea-effect snowfall event depend on its intensity and duration as well as on the location.
569
In Stockholm (November 2016, ~40 cm of snow accumulation) and Gävle (December 1998, ~100 cm)
570
in Sweden, the public transport; busses, trains and flights, were late or cancelled and cars were trapped
571
on roads. Also the Danish island of Bornholm was overwhelmed by ~140 cm deep snowdrift in
572
December 2010. As the snowfall lasted for several days, the island ran out of places to move the snow.
573
A sea-effect snowfall in the Helsinki metropolitan area, Finland, in February 2012 (Juga et al. 2014)
574
caused severe pile-ups on the main roads, with hundreds of car accidents and tens of injured persons.
575
On the other hand, no damages or accidents were reported due to a much larger snowfall accumulation,
576
73 cm of new snow in less than 24 hours, in a small municipality of Merikarvia, western coast of Finland
577
in January 2016 (Fig. 7, Olsson et al. 2017b, 2018).
578
Our current knowledge is mainly based on studies from the Great Lakes in North America (Wright et
579
al. 2013, Cordeira and Laird 2008, Laird et al. 2009, 2003, Niziol et al. 1995, Hjelmfelt 1990). For the
580
Baltic Sea there is an increasing number of studies concerning the formation (Olsson et al. 2017b,
581
Mazon et al. 2015, Savijärvi 2015, Savijärvi 2012, Andersson and Nilsson 1990, Gustafsson et al. 1998)
582
and statistical analysis (Jeworrek et al. 2017, Olsson et al. 2020) of sea-effect snowfalls, as well as
583
effects of excess snowfall to society (Juga et al. 2014, Vajda et al. 2014).
584
The sea-effect snowfall is typically generated in the early winter when thick cold air masses flow over
585
the relatively warm open water basin. The warm water heats the cold air above the water and acts as a
586
constant source of heat and moisture leading to convection. The rising air generates bands of clouds,
587
which quickly grow into snow clouds. Snowfall is enhanced when the moving air mass is uplifted by
588
the orographic effect on the shores or by the convergence of air near the coast as it packs air and forces
589
it to rise inflating convection (Savijärvi 2012). The highest precipitation occurs over the sea close to the
590
coast (Andersson and Nilsson 1990). With suitable wind direction, these snow bands can bring heavy
591
snowfalls to the coastal land area.
592
The sea-effect snowfall is very sensitive to the wind direction because a long fetch over the water body
593
is required (Laird et al. 2003). On the Baltic Sea the most favorable wind directions vary from north to
594
northeast (Jeworrek et al. 2017) due to the cold air outbreaks from the northeastern continent.
595
Nevertheless, for the two major bays (the Gulf of Bothnia and the Gulf of Finland), the sea-effect
596
snowfall can occur on any coast with cold air outbreaks. Favorable conditions for the development of
597
convective snow-bands include an optimum strong wind, large air-sea temperature difference, low
598
vertical wind shear, high atmospheric boundary layer height and favorable wind directions (Jeworrek et
599
al., 2017; Olsson et al. 2020).
600
Using simulations conducted with the regional climate model RCA4 for the period 2000-2010, annually
601
4 to 7 days was seen to be favorable for snowband formation in the western Baltic Sea area and 3 days
602
per year in the eastern Baltic Sea area (Jeworrek et al., 2017; Olsson et al. 2020). A good physical
603
understanding is essential if we want to assess potential changes in frequency and intensity in the future.
604
Based on simple physical reasoning the probability of the events might increase or decrease due to
605
climate change. The ice-cover season is becoming shorter in different parts of the Baltic Sea and also
606
the annual maximum ice extent is projected to decrease (BACC II, 2015; Luomaranta et al., 2014,
607
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
15
Höglund et al., 2017; see also Sec. 2.2.10), extending the time period when convective snowbands can
608
form. Besides, wintertime precipitation amounts are increasing (Sec. 2.2.4). On the other hand, on an
609
annual mean basis, conditions might become less favorable for sea-effect snowfall due to a shorter
610
thermal winter (Ruosteenoja et al., 2020) and a smaller share of snowfall compared to rain in the
611
warming climate (Sec. 2.2.4).
612
The sea-effect snowfall events typically have temporal and spatial scales smaller than what is covered
613
by the observational network and resolved by climate models. The high resolution ERA5 data was used
614
in a case study for January 2016. The preliminary results were promising towards the use of re-analysis
615
data over sea but the data cannot produce intensive enough convective snowfall over land (Olsson et al.
616
2018). Newly developed convective permitting regional climate models (see Sec. 2.2.4), in turn, open
617
up new possibilities to assess the future evolution of the probability of the occurrence.
618
2.2.6 River floods
619
A detailed assessment of climate change of river floods for northern Europe was provided in BACC I
620
(2008) and BACC II (2015). The regional features in the Baltic Sea Basin during last decades according
621
to Stahl et al. (2010) consist in positive trends with increasing streamflow in winter months in most
622
catchments of the Basin, while in spring and summer months, strong negative trends were found
623
(decreasing streamflow, shift towards drier conditions).
624
After the last BACC publication in 2015 there are only few studies devoted to the past hydrological
625
regime changes. Arheimer (2015) concluded that the observed anomalies in annual maximum daily flow
626
for Sweden were normally within 30% deviation from the mean of the reference period. There were no
627
obvious trends in the magnitude of high flows events over the past 100 years. There was a slight decrease
628
in flood frequency, although in a shorter perspective it seems that autumn floods increased over the last
629
30 years. The flood decreasing is connected with seasonality change in the study region. Changes in
630
flood time occurrence in Europe were also established by Bloschl et al (2017). In the Baltic Sea region
631
they detected the floods shifting from late March to February due the earlier snow-melting, driven by
632
temperature increases in the region and a decreasing frequency of arctic air mass advection (see section
633
2.1).
634
The number of severe floods has increased significantly since the 1980s in the Nemunas River Delta.
635
The floods occur often in spring and winter but the lifetime of individual floods has become shorter
636
(Valiuškevičius et.al., 2018). No significant long-term trends in annual streamflow have been found in
637
northwest Russia (Nasonova et al., 2018; Frolova et al., 2017) or Belarus (Partasenok, 2014).
638
Meanwhile, the intra-annual distribution of runoff has changed significantly during the last decades. In
639
particular, runoff during winter low-flow periods has increased significantly while spring runoff and
640
floods during snow-melt were decreasing due to the exhausted water supply in snow before spring.
641
However, the general pattern of described changes in water regime varies from year to year due to the
642
increasing and decreasing frequency of extreme flow events.
643
For future climate a decrease of annual mean (Latvia, Lithuania and Poland) and seasonal streamflow
644
according to the SRES scenario A1B, A2 and B2 was projected for the rivers in Norway and Finland,
645
and (Beldring et al., 2008; Veijalainen et al., 2010b; Apsīte et al., 2011; Kriaučiūnienė et al., 2008;
646
Szwed et al., 2010). An annual streamflow increase by 9-34% has been projected for Denmark (Thodsen
647
et al., 2008, Jeppesen et al., 2009). Large uncertainties in the future hydrological regime were reported
648
for Sweden (Yang et al., 2010, Olsson et al., 2011).
649
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
16
Alfieri et al. (2015) showed positive changes in mean flow in northern and eastern Europe. Significant
650
negative changes in maximum flow are mainly located in north-eastern Europe, including the Baltic
651
countries, Scandinavia and north-western Russia. According to Olsson et. al. (2015) moderate changes
652
in annual mean flow by 2051–2090 are expected in Finland. Winter, summer discharges and early spring
653
discharge peaks will decrease more notably, the autumn mean flow will increase in northern Finland
654
and decrease in catchments with high lake percentage in southern Finland. A significant decrease in
655
magnitude of spring floods and a significant increase in autumn floods are shown for Sweden (Arheimer
656
at al. 2015). For spring floods, the trend obtained using two climate projections (Hadley and Echam)
657
indicates a 10–20% reduction by the end of the century compared to the 1970s. For autumn floods, the
658
trend was in the opposite direction, with 10–20% higher magnitudes by the end of the century. There
659
are slight increases in some parts of Sweden and Norway, north-eastern Europe, according to Donelly
660
et. al. (2017). High runoff levels are found to increase over large parts of continental Europe, increasing
661
in intensity, robustness and spatial extent with increasing warming. Roudier et al. (2016) established the
662
relatively strong decrease in flood magnitude in parts of Finland, NW Russia and North of Sweden with
663
the exception of southern Sweden and some coastal areas in Norway where increases in floods are
664
projected. Northern streams in Finland predicted to lose much of the seasonality of their flow regimes
665
by 2070 to 2100 that is explained by projected air temperature increase and maximal flow decrease.
666
(Mustonen et.al., 2018). The increase of winter runoff and peak discharges was projected by (Kasvi et
667
al., 2019), the most significant changes are expected in wintertime – by 20-40% to 2050-2079 in
668
Southwestern Finland. Almost everywhere the increase in 100-year floods (QRP100) is stronger than
669
the 10-year floods (QPR10). The continuation of current changes in hydrological regime observing
670
within the territory of Belarus in recent decades (increase of winter and decrease of spring streamflow)
671
has been projected for 4 main river basins in the country (Western Dvina, Neman, Dnepr and Pripyat`
672
rivers) by Volchek et al. (2017).
673
According to Thober et al (2018) in northern Europe floods decrease by up to −5% under 3 °C global
674
warming and high flows increase up to 12%. A decrease of floods in this region has been projected in
675
several studies (Arheimer et al., 2015, Alfieri et al., 2015, Roudier et al., 2016). The streamflow in the
676
east of the Baltic Sea Basin (the Western Dvina River within Russia and Belarus) will be characterized
677
by mostly decrease of mean streamflow in the upper stream and increase in the lower part of the river
678
basin. The projected maximal streamflow is expected to decrease with largest changes in the lower part
679
of the river basin up to 25 %.
680
2.2.7 Warm and cold spells in the atmosphere
681
A significant surface air temperature increase in the Baltic Sea region during the last century has been
682
detected, with the largest warming trends in spring (and winter south of 60°N) and the smallest in
683
summer (BACC I, 2008; BACC II 2015; Rutgersson et al., 2014; Meier et al., 2021 and references
684
therein). More recent studies conducted e.g., for Poland (Owczarek and Filipiak, 2016), the three Baltic
685
States (Jaagus et al., 2014, 2017), Finland (Irannezhad et al., 2015; Aalto et al. 2016), Sweden (SMHI,
686
2019) and the whole Baltic Sea drainage basin (Räisänen, 2017), indicate that mean temperatures
687
continue to rise in the region and that the increases are larger than the global average warming.
688
Extreme events related to air temperature include individual high (or low) temperatures, but what is
689
often more influential is extended periods with high (or low) temperatures. The Baltic Sea area is
690
generally less exposed to severe heat waves compared to, for example southern parts of Europe. During
691
the recent decade, however, record breaking heat waves have hit the region, namely those in 2010, 2014
692
and 2018 (Sinclair et al., 2019; Liu et al., 2020; Baker-Austin et al., 2016; Wilcke et al., 2020). Because
693
of adaptation of people living in the Baltic Sea region to relatively cool climate, high summer-time
694
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
17
temperatures pose a significant risk to health also in the current-day climate (e.g., Kollanus and Lanki
695
2014; Åström et al., 2016; Ruuhela et al., 2018).
696
In general, the magnitude, temporal and spatial extent, and frequency of heat waves depend on large-
697
scale fluctuations in atmospheric circulation (see section 2.1), particularly on the occurrence of
698
blockings and other circulation patterns (Horton et al., 2015; Brunner et al., 2017) but other factors,
699
such as local soil moisture feedbacks (Brulebois et al., 2015; Miralles et al., 2014; Whan et al., 2015;
700
Cahynová and Huth, 2014; see also Sec 2.2.9) and solar radiation are of importance. For example,
701
Tomczyk and Bednorz (2014) showed a clear link between heat waves along the southern coast of the
702
Baltic Sea and the circulation patterns. Furthermore, the heat wave in 2018 in Finland was strongly
703
affected by abundant incoming short-wave radiation due to unusually clear skies (Sinclair et al., 2019;
704
Liu et al., 2020). Regarding the local/regional amplitude of a heat wave, land cover use may also play a
705
role. For example, the record high temperature in Finland in 2010 (37.2 °C) was likely contributed by
706
several factors in addition to the very warm air mass (Saku et al., 2011), and in a recent simulation study
707
it was found that replacing a dense urban layout by a suburban type land use resulted in small but
708
systematic decreases in air temperatures in July (Saranko et al., 2020).
709
A widely used heat wave indicator is the warm spell duration (WSDI), defined as the annual (or
710
seasonal) count of days with at least 6 consecutive days when the daily maximum temperature exceeds
711
the corresponding 90th percentile. If using the period 1961-1990 as a baseline when calculating the 90th
712
percentiles, as done in Figure 8 (top left), a statistically highly significant increasing trend across the
713
period 1950-2018 can be found in annual WSDI, when averaged over land areas of the Baltic Sea region
714
(with a Theil-Sen's slope of 1.7 per decade). In southern Sweden, the Baltic States and southern and
715
western Finland, 30-year averages of annual WSDI were about 14 days per year or more during for a
716
recent time span (1989-2018) (Figure 8, bottom left), while during the baseline period the annual count
717
there had been about 6-8 on average. Similar results have been obtained by Irannezhad et al. (2019) and
718
Matthes et al. (2015). The former detected statistically significant increases in annual WSDI near the
719
western coast of Finland for the period 1961–2011 but changes of both positive and negative signs in
720
northern and eastern parts of the country and statistically insignificant increases elsewhere. The latter
721
considered WSDI in 1979-2013 separately in winter and summer and reported statistically significant
722
increases in summer at several Swedish and Norwegian weather stations and in winter also at Finnish
723
stations.
724
In the future, heat waves are projected to occur more often and to become longer and more intense.
725
Today’s warm spells tend to become increasingly frequent, but also increasingly ‘normal’, from a
726
statistical point of view (Rey et al. 2020). Accordingly, quantitative estimates of the rates of the future
727
changes strongly depend on the selected definition of the heat wave (Jacob et al., 2014). The mean
728
length (number) of heat waves where the 20 °C daily mean temperature is exceeded has been projected
729
to increase by about 50 % (60 %) in southern Finland under RCP4.5 between the periods 1900-2005
730
and 2006-2100 (Kim et al. 2018). A bias-adjusted median estimate for changes in WSDI in Scandinavia
731
for the period 2071-2100, with respect to 1981-2010, is about 15 days under RCP8.5, with an uncertainty
732
range of about 5-20 days (Dosio, 2016).
733
Accompanied with more frequent and longer warm spells are decreases in the frequency, duration and
734
severity of cold spells, both based on observations (Easterling et al., 2016) and model projections
735
(Sillmann et al., 2013; Jacob et al., 2013). Cold winter weather in the Baltic Sea region is closely
736
associated with a negative phase of NAO and warm conditions in the Greenland region, and this
737
statistical relationship has strengthened during the recent period of rapid Arctic warming (1998–2015),
738
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
18
suggesting that Arctic influences might intensify in the future, perhaps leading to more unusual and
739
persistent weather events (Vihma et al., 2020). On the other hand, northerly winds from the Arctic are
740
milder than before (Screen, 2014). A cold winter, with unusually low temperatures like those in southern
741
parts of the Baltic Sea area in winter of 2009/10, has become less likely because of anthropogenic
742
changes (Christiansen et al., 2018). The role of changes in circulation remains remarkable; they explain
743
about a half of the very large warming in December in Finland during the period in 1979–2018
744
(Räisänen, 2019).
745
Analogously to WSDI, the cold spell duration (CSDI) is defined as the annual or seasonal count of days
746
with at least 6 consecutive days when the daily minimum temperature is below the corresponding 10th
747
percentile. Because of statistically significant decreases in spatially averaged CSDI over land areas of
748
the Baltic Sea region during the period 1950-2018 (with a Theil-Sen's slope of -0.4 per decade), CSDI
749
is nowadays typically clearly smaller than WSDI (Figure 8, right). There are regional and seasonal
750
differences, however. Statistically significant decreases in winter CSDI across the period 1979–2013
751
have been widespread in Norway and Sweden, but less prevalent in eastern Finland, while changes in
752
summer have been small in general (Matthews et al., 2015). It is also worth noting that because of
753
extremely cold weather in January-February 1985 and particularly in January 1987 (Twardosz et al.
754
2016) and owing to cold winters also more recently, results from trend analyses for the occurrence of
755
cold spells can be strongly affected by the selection of a time period.
756
The cold spell duration index in the northern subregion of Europe is projected to decrease in the future
757
with a likely range of from -5 to -8 days per year by 2071-2100 with respect to 1971-2000 (Jacob et al.
758
2014).
759
2.2.8. Marine Heat Waves
760
Marine heat waves are becoming globally more common (Frölicher et al, 2018) and their intensity and
761
occurrence are projected to increase further in the near future (Oliver et al., 2019). A first, documented,
762
marine heat wave event in the Baltic Sea occurred in summer 2018 when the surface mixed layer became
763
extraordinarily warm in many locations. Accompanied with the atmospheric heat wave in summer 2018
764
large parts of the Baltic Sea were anomalously warm from mid-June to August. According to the satellite
765
data, SST at peak of the warming were up to 27°C from the Bornholm Sea to the central eastern and
766
western Gotland Sea, 22-25°C in the Gulf of Bothnia, 23-25°C in the western parts (Naumann et al.,
767
2018). For the entire Baltic Sea May to August was a positive SST anomaly by 4-5 °C.
768
769
In the coastal regions, the exceptional warming extended down to the bottom layer and had a significant
770
impact on marine biogeochemistry (Humborg et al, 2019). According to the long-term measurement at
771
the coastal region of the Gulf Finland, temperature at bottom (31 meters) was higher than 20°C. That
772
was the all-time record since 1926. Humborg et al. showed also that the warming elevated CO2 and CH4
773
concentration at the bottom considerably. After the actual heat wave event, bottom greenhouse-gas rich
774
waters were exposed to the surface due to storm induced upwelling and as a final consequence CO2 and
775
CH4 fluxes from sea to atmosphere were enhanced.
776
777
Knowledge on occurrence and impact of the marine heat waves in the Baltic in the future is limited.
778
Instead of directly analysing changes of marine heat waves, Meier et al. (2019) used climate projections
779
to estimate how number of warm SST days and the record-breaking anomalies of summer mean SST
780
will change in future. According to their study, both of these indicators will become more common in
781
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
19
future but more important findings are that SST extremes exhibit large variability in time scales of
782
decades and the changes are manifested more pronounced in open sea areas than coastal regions.
783
2.2.9 Drought (Irina Danilovich)
784
The Baltic Sea basin is a region that, in general, has sufficient water resources to support natural
785
ecosystems and societal needs. Despite this, dry conditions occur from time to time in the different parts
786
of the region and cause meteorological, soil moisture and hydrological droughts. This section is devoted
787
to the conditions and some consequences caused by long-term precipitation deficit. Drying conditions
788
are frequently connected with extreme temperatures are referred to in section 2.2.7. Change in
789
precipitation during the twentieth century in the Baltic Sea basin has been variable and characterized by
790
extremeness increasing as also reflected in the river flow regime accordingly (see section 2.2.4 and
791
2.2.6).
792
There are some tendencies characterizing changes in dry conditions. Drought frequency has increased
793
since 1950 across southern Europe and most parts of central Europe with a corresponding decrease in
794
low runoff. In many parts of northern Europe drought frequency has decreased, with an increase in
795
winter minimum runoff while in spring and summer months, strong negative trends were found
796
(decreasing streamflow, shift towards drier conditions). (Stahl et al., 2010; 2012; Poljanšek et al., 2017;
797
Gudmundsson et al., 2017). There are local and regional studies generally supporting this broader
798
picture (Valiukas et al., 2011; Przybylak et al., 2007; Stonevičius et al.,2018; Danilovich et. al., 2019).
799
However, Bordi et al. (2009) in an earlier study found a negative trend of droughts since 2000.
800
Future projections show that the number of dry days in the southern and central parts of the Baltic Sea
801
basin increases in summer (Lehtonen et al., 2014a). The time-mean near-surface soil moisture in the
802
Baltic Sea basin during March–May under the RCP8.5 scenario for the period 2070–2099, relative to
803
1971–2000 averaged over 26 GCMs will reduce up to 8% in the north and up to 4% in the south part of
804
the basin (Ruosteenoja et. al., 2018). According to Spinoni et al. (2018) the meteorological droughts are
805
projected to become more frequent and severe by 2041–2070 and 2071–2100 in summer and autumn in
806
the Mediterranean area, western Europe, and Northern Scandinavia according to RCP4.5 and in the
807
whole European continent (except Iceland) under RCP8.5 scenario.
808
The studies of soil moisture droughts showed drought projections range between strong drying and
809
wetting conditions in Central Europe (Orlowsky and Seneviratne, 2013).
810
In hydrological regime an increase of minimum flows in northern parts of Europe, Scandinavia and the
811
Baltic countries will experience a general increase in 20 yr minimum flows of up to 20% – in some
812
inland tributaries up to 40% – by the end of the 21st century (Forzieri et. al., 2014). The decrease of
813
drought magnitude and duration is expected for central and northern Europe (except southern Sweden)
814
according to Roudlier et al. (2014). This reduction of low flow duration and magnitude is mainly caused
815
by less snowfall and more precipitation for areas with low flows in winter and by a general increase of
816
rainfall for areas with low flows in summer (Vautard et al. 2014). Prudhomme et al. (2014), using several
817
climate and hydrological models, find a general increase of hydrological droughts over Europe, but they
818
focus on less extreme droughts, and use RCP 8.5, at the end of the century. The runoff in late spring and
819
summer is likely to decrease in most of the basin, due to the earlier snowmelt, increased
820
evapotranspiration, and, possibly, particularly in the southern parts, reduced summer precipitation
821
(Räisänen (2017). Increasingly severe river flow droughts are projected for most European regions,
822
except central-eastern and north-eastern Europe (Cammalleri et al., 2020). Climate change scenarios
823
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
20
project on average a small decrease in the lowest water levels during droughts in Finland (Veijalainen,
824
2019).
825
2.2.10 Ice seasons
826
Maximum Ice extent of the Baltic Sea (MIB) is one of the essential variables describing climate change
827
and variability in the Baltic Sea. On an average winter, maximum ice extent is 165,000 km2 indicating
828
that the Bay of Bothnia, coastal areas of the Bothnian Sea, the Archipelago Sea, the Eastern Gulf of
829
Finland and the Bay of Riga are ice covered (BACC II, 2015). During extreme cold conditions, the
830
entire Baltic Sea can be ice covered and during the mildest winter only the Bay of Bothnia is ice covered.
831
Based on the MIB time-series which dates back to 1720, Seinä and Palosuo (1996) defined classification
832
on ice winters according to ice extent. Years with MIB less than 81,000 km2 were classified as extremely
833
mild ice winters and MIB larger than 383,000 km2 as extremely severe ice winters. Here we discuss
834
drivers of ice winter extremes and their observed and expected changes.
835
Annual maximum ice extent is a cumulative indicator of the severity of winter. It is largely driven by
836
the large-scale atmospheric circulation and it’s inter-annual variability is well correlated with the NAO
837
index (Omstedt and Chen, 2001; Vihma and Haapala, 2009). They concluded that during the winters
838
with the NAO index > + 0.5, the average MIB is 121,000 km2, with a range from 45,000 to 337,000
839
km2, while during winters with the NAO index < - 0.5, the average MIB is 259,000 km2, with a range
840
from 150,000 to 405,000 km2. Extremely mild ice winters (MIB < 60,000 km2) have occurred in 1930,
841
1961, 1989, 2008, 2015 and in 2020. According to Uotila et al. (2015), winter 2015 was the first winter
842
when the Bay of Bothnia was definitely only partly ice covered. That winter was dominated by strong
843
south-westerlies associated with a record high NAO index. In addition, the enhanced transport of warm
844
Atlantic air masses to the Baltic Sea region, anomalous low ice extent in winter 2015 was partly due to
845
higher than average downward long-wave radiation because of increased cloudiness which decreased
846
heat loss of the ocean surface layer. Also, episodes of warm foehn winds due to cyclones passing over
847
the Scandinavian mountains were observed in that winter. Uotila et al. (2015) concluded that extremely
848
mild winters were more common during the 1985 – 2015 period than in any other 30-year period since
849
1720. After 2015, only one winter has been an average in terms of MIB. The others are classified as
850
mild or extremely mild ice winters. The winter 2020 was all time record low ice winter. In that winter
851
central parts of the Bay of Bothnia were again ice free and the MIB was only 37,000 km2. Extremely
852
severe winters (MIB > 383,000 km2) have not been observed since 1987. During the last 30 years, the
853
most severe winter occurred in 2011 which caused major problems and economical losses for marine
854
traffic (see section 2.1.1.4).
855
Ongoing changes towards a milder climate demands a revision of the Seinä and Palosuo (1996)
856
definition of extremely mild and severe ice winters. Their classification was based on a choice that 11%
857
of the lowest MIB’s were classified as extremely mild winters. Correspondingly, 11 % of the largest
858
MIB’s were counted as an extremely severe winter. If we are utilizing the same thresholds for the last
859
30 years data, limits for the extremely mild and severe winters would be ~ 50,000 km2 and ~ 240,000
860
km2, respectively.
861
According to climate projections Baltic sea ice will experience considerable shrinking and thinning on
862
average in future (BACC I, 2008; BACC II, 2015). This is particularly clear for the Bothnian Sea,
863
Bothnian Bay and Gulf of Finland. However, changes in natural variability and extreme sea ice winters
864
is an open question since the model studies have been focused on changes in mean conditions.
865
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
21
2.2.11 Ice ridging
866
Sea ice extremes depend on temporal and spatial scale in consideration but more importantly on
867
geographical location and climate conditions – five-meter-thick pressure ridges are common off the
868
Hailuoto island in the Bay of Bothnia every winter, but rarely present in the Southern Baltic Sea.
869
Capacity of the society of managing sea ice related hazards depends also on the likelihood of occurrence
870
of sea ice. In some regions, even a thin ice cover can cause large economical losses to society if the sea
871
ice freezing is occurring in a region where marine traffic is operated by non-ice class vessels. On a local
872
scale, the predominant feature of drift ice is its large variation in thickness. Due to the differential ice
873
motion, pack ice experiences opening, closing, rafting and ridging. In the Baltic Sea, the thickest ice,
874
i.e. pressure ridges, can be 30 meters thick but typically they are 2-5 m thick (Leppäranta and Myrberg,
875
2009, Ronkainen et al., 2018). After initial formation of ridges, they remain in the pack ice as obstacles
876
for shipping. Ridges are formed when pack ice experiences convergent motion. In the Baltic Sea, this
877
is common when pack ice is drifting against the fast ice. In those coastal boundary zones (Oikkonen et
878
al, 2016), mean ice thickness can be half a meter thicker than in the pure thermodynamically grown
879
level ice in the fast ice zone (Ronkainen et al, 2018).
880
During the convergent motion, pack ice experiences compression and its internal stress increases.
881
Internal stress, also called ice pressure or ice compression, depends on the strength of wind and currents
882
but also on ice thickness, floe geometry and cumulative area of coherent ice region in motion
883
(Leppäranta, 2011). Ice motion, concentration, thickness and internal stress of pack ice are strongly
884
coupled. Internal stress of pack ice, which reduces ice motion, increases non-linearly with ice
885
concentration and thickness. In an ultimate situation, very thick ice can be stationary even under strong
886
winds.
887
For shipping, ridges are well observed obstacles using radar and visual methods. They mainly impact
888
the duration of sea time but sea ice compression is more difficult to observe and can cause total stoppage
889
or even damage to ships and vessels. Sea ice compression can be directly observed by in-situ sea ice
890
stress measurements but those measurements are rare in the Baltic Sea (Lensu et al, 2013a). Implicitly,
891
ice compression events have been observed by ships navigating in ice.
892
The most severe ice winters during the last ten years occurred in 2010 and 2011 due to the negative
893
NAO (Cattiaux et al, 2010). In winter 2011, 14 ship accidents occurred due to harsh ice conditions
894
(Hänninen, 2018). For a comparison, during the average winters there are only 1-5 accidents. Several
895
compression events were also reported during the same winter 2011. The most hazardous one occurred
896
at the end of February when marine traffic was totally halted for a few days. Below we provide an
897
anatomy of this extreme event.
898
January and February in 2011 were characterized by cold and calm weather in the Northern Baltic Sea.
899
Consequently, the Gulf of Bothnia became totally ice covered already early in February. Because of the
900
weak winds, the Bothnian Sea was mainly covered by 15 – 30 cm thick undeformed ice (Figure 9). This
901
situation created favorable preconditions for an intensive ice compression and ridging event. After a
902
cold and calm period a change in weather pattern occurred on 24th February when a cyclone arrived in
903
the Bothnian Sea region. The wind speed increased up to 18 m/s and strong southwesterly winds
904
prevailed for the following five days. Consequently, pack ice drifted towards the north-eastern sector of
905
Bothnian Sea. Ice field experienced compression, strong deformations and the undeformed level ice
906
field was redistributed to a heavily deformed ice. In the south-west area of the Bothnian Sea a coastal
907
lead was generated due to divergent ice motion (Figure 9). Based on helicopter electromagnetic
908
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
22
measurements (Ronkainen et al., 2018) mean sea ice thickness along ~100 km transects in the heavily
909
deformed areas increased up to 0.9 m and 1.6 m meters. Thickness of individual ridges were 4 - 8 meters
910
(Figure 9). Sea ice compression, or internal stress of ice, has not been regularly measured in the Baltic
911
Sea but the crews of the ice breakers and merchant vessels are reporting observations of ice pressure
912
from the bridge. Indications of the ice pressure include: closing of ship channels, reduction of ships
913
speed, besetting in ice and compression of ice against the hull of the ships. During the period from 24
914
February to 7 March 142 ice compression cases were reported in the Gulf of Bothnia. From these 25
915
reported severe compressions, or 3-4 on a scale of four (FMI ice service; Lensu et al, 2013b).
916
Compression and thick ice caused a total closedown of marine traffic for several days. Even the largest
917
merchant vessels need to be assisted by the ice breakers. In many cases, the ice breakers needed to assist
918
the merchant vessels one at the time i.e. traditional assistance in convoys was not possible.
919
Sea ice extent and thickness are projected to decrease remarkably in the Baltic (Meier et al., 2021). It’s
920
also expected that occurrence of severe ice winters will decrease and consequently heavy ice ridging
921
and compression events will become rare if wind conditions remain the same in future.
922
2.2.12 Phytoplankton blooms
923
One component of the marine ecosystem here considered an extreme event is phytoplankton blooms
924
(for the marine ecosystem in general see Viitasalo et al., 2021). Phytoplankton (algae and cyanobacteria)
925
undergoes typical annual successions, induced by the regular changes of abiotic (solar radiation,
926
temperature, nutrient concentrations) and biotic (feeding, infections, competition, allelopathy) factors.
927
Under favorable conditions, i.e. sufficient nutrient (N, P, Si) concentrations and solar radiation as well
928
as low wind that allows stratification in the upper water layers, massive phytoplankton growth may
929
occur, leading to blooms. Blooms are visible mass-occurrences of phytoplankton after excessive growth.
930
They become visible by increased water turbidity, sometimes even discoloration (red tides) or surface
931
scums. The mass-occurrence of toxic species (harmful algal blooms) may have detrimental impact on
932
the environmental components, lead to toxic incidents, and may also cause economic harm, e.g. by
933
constraints of the touristic use of the coastal waters (Wasmund, 2002). Phytoplankton forms the basis
934
of the pelagic food web and feeds after sedimentation also the benthos. Its blooms are natural
935
phenomena and a vital component of the ecosystem. Only the excessive blooms caused by cultural
936
eutrophication may be considered a nuisance and should be reduced to a natural level (HELCOM, 2007).
937
This natural level is still not achieved in most areas of the Baltic Sea (HELCOM, 2018).
938
Eutrophication was identified as a major problem in the Baltic Sea in the 1960s and 1970s, leading to
939
the foundation of the Helsinki Commission (HELCOM) in 1974 and the induction of complex
940
monitoring in the Baltic Sea since 1979. Meanwhile, the concentrations of growth-limiting
941
macronutrients, dissolved inorganic nitrogen (DIN) and dissolved inorganic phosphorus (DIP), are
942
decreasing (Andersen et al., 2017). Major Baltic Inflows (BMI) are rare events, which lead to re-
943
oxygenation in the deep water and fixation of phosphorus in the sediment. The latest BMI occurred in
944
December 2014 (Mohrholz et al., 2015). Its effect on oxygen concentrations in the deep water was only
945
of short duration and DIP concentrations were increasing again since 2015 both in the deep and surface
946
water of the Gotland Deep (Naumann et al., 2018). It had no clear effect on phytoplankton biomass, and
947
it did not introduce new phytoplankton species into the Baltic Sea. The originally dominating diatoms
948
in the spring blooms have suddenly decreased since the end of the 1980s in the Baltic Proper (Wasmund
949
et al., 2013) and have been replaced by dinoflagellates (Klais et al., 2011). The ratio of diatoms and
950
dinoflagellates may be a sensitive indicator for changes in the ecosystem including the food web. It was
951
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
23
used to develop the Dia/Dino index as an indicator for the implementation of the Marine Strategy
952
Framework Directive (Wasmund et al., 2017).
953
The summer blooms of cyanobacteria are the most impressive ones in the Baltic Proper and the Gulfs
954
of Finland, Riga and Gdańsk. Long-term analyses including historical data revealed that cyanobacterial
955
blooms became a common phenomenon since the 1960s (Finni et al. 2001). Cyanobacteria seem to
956
increase on a world-wide scale due to global warming (Karlberg and Wulff 2013). Cyanobacterial
957
species mostly have higher growth rates at high temperatures than other phytoplankton species and they
958
are favoured in thermally stratified waters (O'Neil et al. 2012). Also increased freshwater inflow, as
959
projected mainly in the north of the Baltic area (BACC II, 2015) will intensify stratification and support
960
cyanobacteria blooms. However, wind-induced upwelling in early summer may induce blooms, which
961
is primarily an effect of phosphorus input into the surface water (Wasmund et al. 2012). If stratification
962
is disrupted by wind, established cyanobacteria blooms may collapse (Wasmund 1997). As the bloom-
963
forming buoyant cyanobacteria occur patchy, representative sampling is difficult and data may be
964
insufficient for a reliable trend analysis. The development of cyanobacteria blooms is annually reported
965
in HELCOM Environment Fact Sheets since 1990 (Öberg, 2017; Kownacka et al. 2020), but general
966
trends could not be identified in these three decades. However, in specific regions, trends may occur,
967
which may be even contradictory (Olofsson et al. 2020). A few recent extreme blooms are selected to
968
be mentioned here.
969
On 20 July 2017, cyanobacteria warnings were issued for eight beaches in the area of the Gulf of Gdańsk
970
and on 22-24 July 2017, three bathing sites were closed due to the decreased water transparency. In
971
2018, all the bathing sites of the Gulf of Gdańsk and Puck Bay were closed for 12 days owing to the
972
formation of toxic scums. In the Gulf of Finland, the exceptionally warm summer 2018 (see also marine
973
heat waves, section 2.2.8) caused the strongest cyanobacterial bloom of the 2010's
974
(https://www.syke.fi/en-
975
US/Current/Algal_reviews/Summary_reviews/Summary_of_algal_bloom_monitoring_2018_S(47752
976
)). Remarkably, the typical cyanobacteria genus of the summer blooms was also abundant in winter
977
under the ice on the western and eastern Finnish coast, as identified for example on 7 January 2019
978
(http://www.syke.fi/fi-FI/Ajankohtaista/Tiedotteet/Viileassakin_vedessa_viihtyvaa_sinilevaa(48957)).
979
In the past decade, blooms of toxic dinoflagellates have increasingly been observed in shallow coastal
980
waters of the Baltic Sea. Neurotoxic A. ostenfeldii now regularly forms dense bioluminescent summer
981
blooms in the Åland archipelago and the Gulf of Gdańsk (Hakanen et al. 2012). Highest cell
982
concentrations so far recorded for this species were measured in the Åland area in August 2015 (Savela
983
et al. 2016). In July 2015, a dense bloom of Karlodinium veneficum, killing fish in a shallow bay at the
984
SW coast of Finland raised the attention of regional authorities (https://www.syke.fi/en-
985
US/Current/Press_releases/Last_summers_fish_kill_was_caused_by_a_t(38306)). Global warming is
986
generally becoming a threat that may influence the phytoplankton stronger (Cloern et al., 2016; Reusch
987
et al., 2018). Future changes in eutrophication and as well as a changing climate will influence the
988
occurrence of harmful algal bloom. If the Baltic Sea Action Plan is implemented successfully, it is
989
suggested that record-breaking cyanobacteria blooms will not occur in the Baltic Sea in the future (Meier
990
et al., 2019).
991
A phenomenon worth mentioning is the extension of the growing season of phytoplankton in the oceans
992
(Gobler et al., 2017), but also in the Baltic Sea (Groetsch et al., 2016). The period with satellite-
993
estimated chlorophyll a (chl a) concentrations of at least 3 mg m-3 has doubled from approximately 110
994
days in 1998 to 220 days in 2013 the central Baltic Sea (Kahru et al., 2016). Based on weekly
995
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
24
measurements of phytoplankton biomass and chl a concentrations at a coastal station in the Bay of
996
Mecklenburg from 1988 to 2017, Wasmund et al. (2019) found an earlier start of the spring bloom with
997
a rate of 1.4 days/year and a later end of the autumn bloom with 3.1 days/year and a corresponding
998
extension of the growing season (Figure 10). The earlier start of the growing season was correlated with
999
a slight increase in sunshine duration during spring whereas the later end of the growing season was
1000
correlated with a strong increase in water temperature in autumn. As the growing season extends
1001
recently from February to December at the investigated site, a further extension is practically not
1002
possible. However, this process may be still ongoing in other regions of the Baltic Sea.
1003
2.3 Possible implications for society
1004
Extreme events and projected changes caused by e.g. global warming or changes in the atmospheric
1005
circulation could have large and potentially disastrous consequences to Baltic societies. This section
1006
examines the potential implications of extremes and changes of extremes on forest fires, coastal
1007
flooding, offshore wind activities and shipping in the Baltic Sea area, all of which are linked to key
1008
economic sectors. These are also linked to the Multiple drivers of the Baltic Sea systems (Reckerman
1009
et a., 2021).
1010
2.3.1 Forest fires
1011
Fires play a key role in the natural succession and maintains biological diversity in boreal forests, they
1012
also pose a threat to property, infrastructure and people’s lives (e.g., Rowe and Scotter, 1973;
1013
Zackrisson, 1977; Esseen et al., 1997; Virkkala and Toivonen, 1999; Ruokolainen and Salo, 2006).
1014
Large forest fires are often associated with long-lasting drought and heat waves. During the
1015
exceptionally warm and dry summer of 2018, numerous large fires burned a total of almost 25,000
1016
hectares of forest in Sweden (Statens offentliga utredningar, 2019; Sjöström and Granström, 2020).
1017
Also, during the heat wave of 2014, a single conflagration in Västmanland burned nearly 15,000
1018
hectares. In Russia, the persistent heat wave of 2010 resulted in devastating forest fires (Bondur, 2011;
1019
Witte et al., 2011; Vinogradova et al., 2016). Fires have a deteriorating impact on air quality (Konovalov
1020
et al., 2011; R’Honi et al., 2013; Popovicheva et al., 2014), in extreme cases even in regions hundreds
1021
of kilometers away from the actual fire (Mei et al., 2011; Mielonen et al., 2012; Vinogradova et al.,
1022
2016). The emissions of gases and aerosols through fires as well as changes in surface albedo also have
1023
impacts on climate. Due to increasing fire activity, boreal forests may even shift from carbon sink to a
1024
net source of carbon to the atmosphere, resulting in a positive climate feedback (Oris et al., 2014; Walker
1025
et al., 2019). The impact of aerosols is more complex, yet generally short-lived. However, heat-trapping
1026
soot from large conflagrations can enter into the stratosphere and persist there for months (Ditas et al.,
1027
2018; Yu et al., 2019). Changes in surface albedo due to fires tend to decrease radiative forcing in the
1028
long term (e.g., Randerson et al., 2006; Lyons et al., 2008).
1029
Compared to other boreal regions, forest fires in Northern Europe are small. This is mainly due to
1030
effective fire suppression. In addition, heterogeneity of Fennoscandian forests with lakes and swamps
1031
creates natural obstacles for fires. Large fires are more common in Russia, Canada and Alaska (e.g.,
1032
Stocks et al., 2002; Vivchar, 2011; Smirnov et al., 2015). However, also in Fennoscandia large fires
1033
were not uncommon before the cultural transition to modern agriculture and forestry led to a steep
1034
decline in annual burned area by the end of the 19th century (Wallenius, 2011).
1035
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
25
The natural source of fire in boreal forests is lightning. Nowadays lightning strikes ignite about 10% of
1036
fires in Sweden and Finland (Granström, 1993; Larjavaara et al., 2005a). In northern Europe, the
1037
distribution of lightning-ignited fires follows approximately the thunderstorm climatology with less
1038
ignitions in the north (Granström, 1993; Larjavaara et al., 2005b). In recent years, many of the largest
1039
fires have been caused by forest machinery operations (Sjöström et al., 2019).
1040
Irrespective of the ignition source, weather influences the conditions for the spreading. In northern
1041
European boreal forests, climate and particularly precipitation variability has been an important decadal-
1042
scale driver of fires even during the recent centuries with strong human influence on fire occurrence
1043
(Aakala et al., 2018). In boreal forests in general, interannual variability in burned area can be by a large
1044
part be explained by fluctuations in lightning activity (Veraverbeke et al., 2017) and also by variations
1045
in large-scale atmospheric circulation patterns (Milenković et al., 2019). Usually, only a few years with
1046
large forest fires account for the majority of burned area at decadal to centennial time scales (Stocks et
1047
al., 2002). Although during the last century these large fire years have tended to occur in northern
1048
Scandinavia in association with warm and dry summers, historically years with large forest fires have
1049
occurred more frequently during cooler than warmer periods (Drobyshev et al., 2016). Drobyshev et al.
1050
(2016) related this to coupled ocean–atmosphere dynamics favouring high pressure systems over
1051
Scandinavia in association with low sea surface temperatures in the North Atlantic. Moreover, fire
1052
regimes in northern and mid-boreal forests have appeared to be more sensitive to climate variations
1053
compared to fire regimes in southern boreal forests (Drobyshev et al., 2014). Drobyshev et al. (2014)
1054
hence concluded that fire regimes across Scandinavia might show even an asynchronous response to
1055
future climate changes.
1056
In response to global warming, the forest-fire danger is generally projected to increase across the
1057
circumboreal region (e.g., Flannigan et al., 2009; Wotton et al., 2010; Shvidenko and Schepaschenko,
1058
2013; Sherstyukov and Sherstyukov, 2014). This is particularly due to enhanced evaporation in a
1059
warmer climate. Already within the recent decades, long-lasting drought events have become more
1060
intense throughout Europe (see section 2.2.9), increasing temperatures having been the main driver of
1061
the change (Manning et al., 2019). According to the most extreme warming scenarios, summer months
1062
with anomalously low soil moisture, occurred in northern Europe recently once in a decade, may occur
1063
more often than twice in a decade in the late 21st century (Ruosteenoja et al., 2018).
1064
In Finland, the climate change impact on forest-fire risk has been evaluated in several studies
1065
(Kilpeläinen et al., 2010; Mäkelä et al., 2014; Lehtonen et al., 2014b, 2016). The projected decrease in
1066
soil moisture content has been reflected as a projected increase in fire risk. Assuming the current
1067
relationship between weather and the occurrence of forest fires, Lehtonen et al. (2016) estimated that in
1068
Finland, the number of fires larger than 10 ha in size may double or even triple during the present
1069
century. Nevertheless, there is considerable uncertainty in the rate of the change, largely due to the
1070
uncertainty of precipitation projections. Yang et al. (2015) predicted that in northern Sweden, the fire
1071
risk could even decrease in the future.
1072
In addition to meteorological conditions, fire potential is largely determined by the availability of
1073
flammable fuels in forests. In southern Europe, the biomass availability may become a limiting factor
1074
for increasing fire activity (Migliavacca et al., 2013). However, in northern Europe this is unlikely, as
1075
forest productivity and biomass stock are projected to increase under a warming climate (Kellomäki et
1076
al., 2008; Dury et al., 2011).
1077
2.3.2 Coastal flooding
1078
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
26
The projected regional sea level rise (e.g. Grindsted et al. 2015) coupled with the expected
1079
intensification of sea level extremes (e.g. Vousdoukas et al. 2018) discussed in Section 2.2.3 will widely
1080
affect both natural and human systems along the Baltic Sea.
1081
In the past, several major floods have occurred on the Baltic Sea coast. While there are few surviving
1082
sea level measurements or other historical records dated before the 19th century traces of extreme floods
1083
are found from sand layers. Studies of coastal sediments, compared with historical records, imply that
1084
the flood in 1497, which damaged cities on the southern Baltic coast, was the largest storm surge on the
1085
Polish coast in 2000 years (Piotrowski et al., 2017). St. Petersburg has also proved vulnerable to coastal
1086
and fluvial flooding, and the highest documented surge occurred in 1824, when the water level rose to
1087
367 cm at Kronstadt, and possibly even to 410 cm at St. Petersburg (Bogdanov and Malova, 2009) over
1088
local mean sea level. In the era of tide gauges, the most severe flood along the southern Baltic coast
1089
happened in 1872. This storm caused large damages at the German and Danish coast and 271 lives were
1090
reported lost (Rosenhagen and Bork, 2008). At Travemünde, Germany, the sea level rose to 340 cm
1091
(Jensen and Müller-Navarra, 2008); at Skanör, along the southern Swedish coast, the sea level reached
1092
approx. 240 cm (Fredriksson et al., 2016). For the Gulf of Finland and the Gulf of Riga, the most severe
1093
flooding on record was caused by the Gudrun wind storm in 2005, when the observed sea level reached
1094
197 cm in Hamina (Finland), 230 cm in St. Petersburg (Russia), 207 cm in Ristna (Estonia), and 275
1095
cm in Pärnu (Estonia) (Suursaar et al., 2006).
1096
In a European perspective, the uncertain influence of climate change on the frequency and intensity of
1097
waves and wind as a predictor of future damage costs due to coastal flooding is decreasing relative to
1098
the observed and projected influence of sea level rise on storm surge heights. Hence, Vousdoukas et al.
1099
(2018) finds that the indirect effect of mean sea level rise, uplifting high sea levels under extreme
1100
weather conditions, serves as the main driver of the increased coastal flood damages in the future and
1101
accounts for 88–98% of the total damages. Interestingly, the highest relative contribution from changes
1102
in cyclones is here projected along the Baltic Sea coast. This stems from a combination of low relative
1103
sea level rise along the Baltic Sea catchment that is due to the land uplift and intensifying waves and
1104
storm surges due to climate change based on the projections used by Vousdoukas et al. (2017). In
1105
general, there is no consensus whether the wind storms are expected to become more frequent (Sec
1106
2.2.1). In particular, for Finland and Sweden - due to land uplift - the physical footprint of sea level rise
1107
in future damage estimates is weakened. Conversely, socioeconomic development along the coast is
1108
likely to be a main driver and modulate the intensification of coastal hazards amongst Baltic Sea
1109
countries.
1110
In the absence of improved coastal management practices and coastal adaptation, the expected
1111
population exposed to coastal flooding along the Baltic Sea coastline annually as well as the expected
1112
annual damages (EAD) due to coastal flooding are both likely to increase by orders of magnitude (e.g.
1113
Forzieri et al. 2016, Vousdoukas et al. 2018, Mokrech et al. 2014, Brown et al. 2018). While the impacts
1114
on managed as well as natural coastal and near-coastal terrestrial ecosystems may be significant, Baltic
1115
coastal cities are likely to be mainly responsible for future coastal flood losses due to their high
1116
concentration of people, infrastructure and valuable assets. To keep future coastal flood losses low,
1117
climate change adaptation measures urgently need to be installed or reinforced (Vousdoukas et al. 2020,
1118
Abadie et al. 2019) to withstand extreme sea levels, which could exceed 4-5 metres in some locations
1119
(see Section 2.2.3).
1120
Apart from recent work by Paprotny and Terefenko (2017) for Poland, environmental and economic
1121
impact assessments at the regional to national level generally belong to the grey literature. Similarly,
1122
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
27
impact assessments at the local (city) level have so far mainly been carried out by, e.g., engineering
1123
consultancies, to facilitate the development of local adaptation strategies (Thorarinsdottir et al. 2017).
1124
Due to local constraints and a lack of best practices, the methodologies behind such detailed assessments
1125
often vary greatly and are not comparable.
1126
Figure 11 shows different damage estimates related to coastal flooding, including for some of the most
1127
exposed cities along the Baltic Sea. Here Copenhagen stands out, Prahl et al. (2017) has calculated a set
1128
of macroscale damage cost curves (Figure 11, main part), i.e., damage cost as a function of flood height,
1129
for the largest 600 cities in Europe, including all of the major cities along the Baltic Sea. Land-use
1130
information is being used and not population coupled with GDP per capita as the basis for approximating
1131
the location of assets; i.e., this ensures that flooded assets are inherently co-located with the city. For
1132
the hydrological modelling, a high-resolution digital elevation model for Europe is used together with
1133
a simple static-inundation model that only accounts for hydraulic connectivity. While this approach
1134
readily allows for estimation of the damage costs associated with flooding for any European coastal
1135
city, the “coarseness” of the methodology (including the underlying empirical and categorical
1136
information on land-use and flood defenses, which goes into the calculations), can lead to
1137
overestimation of the damage cost curves, especially for low-lying urban and high-value areas. This is
1138
particularly found to be the case for (but not restricted to) Copenhagen (Figure 11, main part).
1139
For comparison, Abadie et al. (2016) have carried out a variant set of economic impact assessments for
1140
Copenhagen, Helsinki and Stockholm in 2050 based on an improved version of the same large-scale
1141
modelling framework, cf, the insert of Figure 11 (lower rows). Using the same input as Prahl et al.
1142
(2017), Abadie et al. (2016) have developed a European scale assessment framework, where a
1143
continuous stochastic diffusion model is used to describe local sea level rise, and Monte Carlo
1144
simulations yield estimates of the (risk) damage caused by the modelled sea level rise. This is paired
1145
with an economic damage function developed for each city and point in time. The results found by
1146
Abadie et al. for a RCP8.5 scenario is shown in Figure 11. For Copenhagen and Stockholm the damage
1147
cost estimates of Prahl et al. are largely consistent with those of Abadie et al. (2016).
1148
Vousdoukas et al. (2018, 2019, 2020) has estimated the EAD from coastal flooding for all countries in
1149
Europe (excluding adaptation) by combining future climate model projections with a set of gridded
1150
projections of gross domestic production, population dynamics and exposed assets based on select
1151
shared socioeconomic pathways. Flood defenses are considered as recorded in the FLOPROS database
1152
(Scussolini et al. 2015). As seen in the Table in Figure 11 (upper rows), at the end of the century
1153
Denmark is expected to suffer the largest damages from increased coastal flooding due to climate change
1154
due to its long coast line, followed by Germany, Poland and Sweden.
1155
The large observed variation in cost estimates related to future coastal flooding in the Baltic Sea may
1156
easily be ascribed to different approaches, data and scales used for the impact modelling, including key
1157
assumptions, in particular relating to the economics. To improve confidence in impact assessments, a
1158
comparable assessment of methods, models, and assumptions are needed in order to establish more solid
1159
evidence within the area. Likewise, impacts due to compound events where for example extreme coastal
1160
water levels are (locally) exacerbated by associated high water levels in nearby rivers or high intensity
1161
rainfall (Bevacqua et al. 2019) are largely unaccounted for in most damage cost assessments.
1162
2.3.3 Offshore wind energy activities
1163
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
28
Offshore wind farms are growing rapidly in the Baltic Sea. Figure 12 shows the expansion of wind farm
1164
clusters in southern parts of the Baltic Sea and in the North Sea. According to recent reports, offshore
1165
wind power in the Baltic Sea is far from fully exploited and could reach 83 GW (Cecchinato 2019;
1166
Freeman et al. 2019).
1167
Compared to onshore situations, offshore wind energy benefits from richer wind resources. It is also
1168
greatly challenged by the harsher offshore environmental conditions, which makes the so-called
1169
Levelized Cost Of Energy (LCOE) significantly higher. LCOE accounts for, among others, the
1170
transportation of energy from sea to land, the trips to the farms for maintenance, and water depth where
1171
the turbines will be installed. The maintenance and construction become more challenging when storms
1172
are present as storms cause rougher conditions for the turbines and farms at sea than over land. There
1173
are no land obstacles to effectively consume the atmospheric momentum, instead, waves are generated;
1174
swells develop and propagate, and waves break. This can put tremendous load on construction of fixed
1175
as well as floating turbines. At the same time, breaking waves release water drops and sea salt into the
1176
air. This, together with severe precipitation at sea during storms, has a significant impact on the erosion
1177
process of the turbine blades and affects the turbine performance (e.g. Mishnaevsky 2019). At sea, the
1178
role of icing on blades was considered generally small (e.g. Bredesen et al. 2017), while over the Baltic
1179
Sea ice cannot be ignored (Heinonen et al. 2019). The storm winds at sea reach the cutoff speed of 25
1180
ms-1 at hub height more frequently, causing more fluctuation in power production and accordingly
1181
significant challenges in the power integration system (e.g. Sørensen et al. 2008; Cutululis et al. 2013).
1182
At the same time, strong winds and large waves directly affect the activities such as installation,
1183
operation and maintenance (O&M). See e.g. Diamond et al. (2012), Leiding et al. (2014), Dangendorf
1184
et al. (2016) and Kettle (2018,2019).
1185
Several sections in this report summarized studies on the climatological changes of a number of relevant
1186
parameters including storms, waves, temperature, icing, precipitation and water levels. Effort is needed
1187
in coordinating the analysis and implementing these changes of the environmental parameters in
1188
offshore wind energy planning. Design parameters need to be calculated to avoid placing turbines in a
1189
dangerous wind environment and to identify the suitable turbine design class. Turbulence and the 10-
1190
min value of the 50-year wind at hub height are two key design parameters (IEC 61400-1) requiring
1191
improved estimation.
1192
In the presence of storms over the sea, special organized atmospheric features develop, contributing to
1193
turbulence over broader frequency/wave number range than the typical stationary surface layer
1194
conditions. These features include gravity waves, low level jets, open cells and boundary layer rolls.
1195
Over the Baltic Sea, gravity waves and boundary rolls are present (e.g. Larsén et al. 2012; Svensson et
1196
al. 2017, Smedman 1991). Over the North Sea, it was found that open cells can add an extra of 20 - 50%
1197
to the turbulence intensity (Larsén et al. 2019b).
1198
For the studies of the extreme winds over Scandinavia for wind energy applications, groups in Sweden
1199
and Denmark pioneered by using long-term wind measurements (e.g. Abild 1991; Bergström 1992;
1200
Kristensen et al. 2000). Later, long-term global reanalysis products are used, including the Baltic Sea
1201
area (e.g. Frank 2001; Larsén and Mann 2009). At early stages of the wind energy development, the
1202
reference height of 10 m was most relevant for engineering applications. Today, the turbines are much
1203
bigger and the largest (offshore) turbine has a 220-m rotor and 107-m blade. At the same time, wind
1204
energy is developing to larger global coverage over various land/sea conditions. These make the use of
1205
the mesoscale models an attractive option. The three-dimensional mesoscale numerical model, the
1206
MIUU-model for the 50-year wind speed was used to calculate both 10-min mean and 3-s gust values,
1207
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
29
with a grid space of 1 km (Bergström and Söderberg 2008). In addition, a variety of mesoscale models
1208
have been used for wind resource assessment as well as extreme wind calculations, such as the
1209
HIRHAM model, the (e.g. Clausen et al. 2012; Pryor et al. 2012), the KAMM model (e.g. Hofherr and
1210
Kunz 2010; Larsén and Badger 2012), the REMO, the CCLM models (Kunz et al. 2010) and the WRF
1211
model (Bastine et al. 2018). For long-term data the models are run for decades. In compensation with
1212
the computational cost, most of these models have been run at a spatial resolution of tens of kilometers.
1213
The effect of spatial and temporal resolution of these mesoscale modeled winds was investigated in
1214
Larsén et al. (2012) using modeled data from WRF, REMO and HIRHAM. Larsén et al. (2012)
1215
developed a so-called spectral correction method to fill in the missing variability in the modeled time
1216
series, thus reducing the underestimation of the extreme wind. To calculate the extreme wind, Larsén et
1217
al. (2012, 2019) also developed a selective dynamical downscaling method to efficiently allocate
1218
modeling resources to storms at high resolution, i.e. 2 km. The southern part of the Baltic Sea was
1219
included in these calculations.
1220
The development of approaches for calculating design parameters over the Baltic Sea has provided
1221
different estimations through time. The difference in these estimations (more than 10%) is bigger than
1222
the effect from climate change calculated from different climate scenarios (a few percent). Climate
1223
modeling describes future scenarios and provides a coherent calculation of the whole set of
1224
environmental parameters, including wind, temperature, icing and precipitation. One of such outputs is
1225
from the research project Climate and Energy Systems (CES) supported by the Nordic Research Council
1226
(Thorsteinsson 2011). This study features both opportunities and risks within the energy sector
1227
associated with climate change up to the mid-21st century. Fifteen combinations of Regional and Global
1228
Climate Models were used. The results however did not portrait a consensus on the change in storms
1229
and extreme winds in the future over the Scandinavian seas (see also section 2.2.1 and Belusic et al.,
1230
2019).
1231
2.3.4 Shipping
1232
There are several aspects where changes in extreme events and natural disasters have the potential of
1233
influencing shipping, one relates to ice conditions. As stated above (Section 2.2.10 and 2.2.11) winters
1234
on the Baltic Sea can be different with highly varying ice conditions. This has been observed, when the
1235
ice loads encountered by ships have been measured in full scale by instrumenting ship hulls for ice load
1236
measurements, see example in Figure 13 (Kujala, 2017). Typically, the highest loads occur when ships
1237
are moving through heavily ridged areas or are stuck in moving, compressive ice. The highest measured
1238
loads occurred in severe ice winters, such as 1985 and 1987. Extreme events can also cause remarkable
1239
damages on the ship shell structures as shown in Fig 14 (Kujala, 1991). Typically, the ice-induced
1240
damages are local dents on the shell structures, with the depth about 50-100 mm and width as well
1241
height about 0.5m*0.5 m. The figure shows an example of the extensive damage outside Luleå (upper
1242
figure), when the ship left the harbour independently without icebreaker assistance and got stuck in
1243
compressive ice, and then the whole shell structures got permanent damage with the depth of about 0.5
1244
m and the length and height several meters. The ice-strengthened ships are not designed for this type of
1245
situation as the design principle is that icebreakers will prevent ships from getting stuck in ice.
1246
Increasing maritime traffic in areas where icebreaker (IB) assistance is needed will increase the demand
1247
for icebreaking assistance. The workload of an IB in its operational area, at a specific time, is strongly
1248
dependent on the area specific ice conditions and ship traffic. This leads to large area- and time-specific
1249
variations in the demand for icebreaking assistance. Even under constant ice conditions, it is hard to
1250
estimate local demand for assistance solely from the estimated increase or decrease in local maritime
1251
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
30
traffic. There are a number of studies related to the development of the transit simulation models for
1252
ships navigating in ice, (e.g. Patey and Riska, 1999; Kamesaki et al., 1999; Montewka et al., 2015;
1253
Kuuliala et al., 2017 and Bergström, 2017). Typically, all these models simulate the speed variation of
1254
a single ship when it is sailing in varying ice conditions such as level ice, ridged ice and ice channel. In
1255
addition, the real time Automatic Identification System of vessels (AIS) data has been used to study e.g.
1256
the convoy speed when IBs assist merchant ships, see Goerlandt et al. (2017). Monte Carlo random
1257
simulation can also be used to study the uncertainties and variations on the ice conditions and on the
1258
calculation methods to evaluate ship speed in various ice conditions (Bergström, 2017).
1259
The newest development includes simulation tools built around a deterministic IB-movement model
1260
(Lindeberg et al.,2015, 2018). The new approach is that the simulation model also includes the decision
1261
principles of IBs to determine which ships and when they will be assisted. The model also includes the
1262
possible assistance and towing principles of merchant ships behind an IB. The tool can be used for
1263
predicting local demand for icebreaking assistance under changing ice and traffic conditions. It can also
1264
be used to predict how the traffic flow will react to changes in the IB operational areas of the modelled
1265
system, i.e. by adding/removing IBs from the system and/or by modifying the boundaries of IB
1266
operational areas.
1267
Typically, during a normal winter starting in December and ending in April, there are about 10000 ship
1268
visits to our icebound harbors in the Baltic Sea and the traffic is assisted by 5-9 IBs. The developed
1269
model can be used to study e.g. the effect of winter hardness on the IB activities and waiting time for
1270
merchant vessels (Lindeberg et al., 2018). The new environmental requirements will cause a decrease
1271
in the used engine power of ships, which might mean that the need for IB assistance will increase. As
1272
studied by Lindeberg et al. (2018), the new so called EEDI ships will increase the merchant vessel
1273
waiting time 100 % when 50 % of the new ships will fulfill the EEDI requirements, so this means that
1274
in future we might need more IBs to guarantee smooth marine traffic. EEDI is a new energy efficient
1275
requirement, which will decrease the engine power on typical merchant ships.
1276
The model can also be used to study the effect of winter hardness on the amount of needed IB assistance,
1277
e.g. during the hard winter of 2010-2011, the total number of IBs assisting was nine with the total
1278
amount of assisting miles: 77056 nm and during a mild winter of 2016-2017, it was eight IBs and 29502
1279
nm assisted.
1280
In addition to ice conditions, the maritime shipping in the Baltic Sea is affected by wind and wave
1281
conditions and icing due to sea spray. Although the mean wind and wave conditions area relatively low
1282
in the Baltic Sea, some of the high wind events and especially the severest storms affect the maritime
1283
traffic (cf. Section 2.2.2 for extreme wave events). In the severest storms smaller vessels need to find
1284
shelter or alternative routes and large vessels need to reduce speed or increase engine power. Increasing
1285
the vessels engine power during these events will also increase the ship emissions (Jalkanen et al. 2009).
1286
Also getting safely in and out of harbours is an issue during high wind and wave events.
1287
In the changing climate, the ice winters are estimated to get shorter and the ice extent smaller (Section
1288
2.2.10). The time of the year that in the present climate has ice cover, partly coincides with the windiest
1289
time of the year. This means that the wave climate in the Bay of Bothnia and eastern part of the Gulf of
1290
Finland, where there still is ice every winter in the present climate, is estimated to get more severe and
1291
this can cause increasing dynamics of the ice making navigation in ice more demanding.
1292
However, the occurrence of the extreme wave events is not only dependent on the changes in the ice
1293
conditions but also on the changes in the wind conditions. Moreover, the Baltic Sea sub-basins are
1294
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
31
relatively small and the high wind events are often fetch-limited, thus the wind direction plays a large
1295
role in the generation of the high wave events. As the frequency of strong westerly winds is projected
1296
to increase (see section 2.2.1), this will most likely lead to an increase in the high wave conditions from
1297
this sector.
1298
Icing due to sea spray causes problems for maritime traffic in the Baltic Sea every now and then. In a
1299
future climate, this can happen more often as the ice winters get milder and the sea is open during the
1300
time of the year when sea surface temperatures are close to freezing point, so the probability of getting
1301
freezing water on the ship deck will potentially increase.
1302
3 Knowledge gaps
1303
As extreme events by definition are rare, long time series of data and/or large ensembles with model
1304
simulations with high spatial coverage are a necessity for a full understanding of return periods and for
1305
mapping expected changes in intensities of extreme events. When also adding the impact of climate
1306
change and to some extent an unknown response of the climate system to partly unknown changes in
1307
forcing, the uncertainty increases further especially locally. This is in particular true for compound
1308
events (i.e interaction of multiple hazard drivers) and freak events (i.e. events that have very low
1309
probabilities but which potentially can have disastrous impacts). These kinds of events are largely
1310
unexplored in the scientific literature.
1311
As previously discussed, many extreme events in the Baltic Sea region are related to the large-scale
1312
atmospheric dynamics, including storms originating from the North Atlantic region. Knowledge gaps
1313
concerning the response of large-scale atmospheric circulation in a warming climate include the
1314
dynamic response of reduced Arctic sea ice and changing oceanic conditions as well as the possibility
1315
of changes in the jet stream patterns and/or changing blocking frequencies over Europe.
1316
Besides storms that are related to extratropical cyclones, strong winds can also be induced by extreme
1317
convective weather, including downbursts, tornadoes, detached thunderclouds, derechos and other
1318
mesoscale convective systems (Rauhala et al., 2012; Punkka, 2015). Furthermore, wind gusts driven by
1319
convective downdrafts or turbulent mixing can also occur during larger-scale windstorms (Laurila et
1320
al., 2019). All these phenomena may be harmful to infrastructure, the severity of the impacts depending
1321
on the intensity and location of occurrence of the events. New convection permitting climate models
1322
with grid spacing of a few km (Sec. 2.2.4), as well as increasing observation density owing to the use
1323
of weather radars, satellites and lightning-location sensors, open new possibilities to assess their
1324
probabilities of occurrence of in the recent past and projected future climate.
1325
A local characteristic is the uncertainty in local responses to large-scale variability and global change.
1326
One particular feature is soil water response to heat waves, but also features such as changes in
1327
frequency of major Baltic inflows (Lehman et al., 2021; Meier et al., 2021). In the Baltic Sea region,
1328
the state of the cryosphere has already changed remarkably. Past mean changes in frost, snow, icing,
1329
lake and sea ice conditions have been rather well estimated by the regional models, but their future
1330
variability and change ranging from synoptic to centennial time scales are uncertain. Moreover, the
1331
impact of extreme cryosphere changes on forestry, reindeer herding, spring floods, extreme wave
1332
heights or shipping is largely unknown. Concerning flood assessments, the majority of the studies are
1333
devoted to high flood extremes. The low flow periods are less well described due to the absence of
1334
remarkable changes in flow regime especially in northern Europe because of the large model uncertainty
1335
in precipitation during the summer (or warm period) when low flow usually occurs.
1336
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
32
The prolongation of the growing season of phytoplankton is identified, but it may be not only caused
1337
by a simple direct influence of increased radiation and temperature. The temperature may also act via
1338
stronger stratification, shifts in grazing pressure or infections or other factors which still have to be
1339
identified in detail. Earlier phytoplankton spring blooms, a longer summer minimum and a later autumn
1340
bloom may have decisive impact on the food web and need to be investigated. The first major marine
1341
heat wave recorded occurred in the Baltic in 2018. Further research is needed to estimate probabilities
1342
of marine heat waves in the future but also to deepen our understanding how biogeochemical processes
1343
are altered in those conditions.
1344
Simulation of storm tracks and their associated precipitation generally improve with increasing
1345
resolution beyond that used in most current climate models (Michaelis et al., 2017; Barcikowska et al.,
1346
2018), and higher resolution results in more sensitivity to warming (Willison et al., 2015).
1347
Understanding of high-intensity extremes requires improved re-analysis products and carefully
1348
homogenized long time series data as well as higher resolution climate models. Here the better use of
1349
new tools might lead to increased understanding. This includes remote sensing data, new types of sensor
1350
systems in combination with traditional in-situ observational networks. Combining new data with higher
1351
resolution models as well as new methodologies (machine learning, neural networks) has great potential.
1352
Following aspects are the most important to address in relation to future research:
1353
• Coupled high resolution process and Earth System models for detailed understanding of
1354
extremes and feedback mechanisms between different processes (see also Görger et al., 2021).
1355
• Addressing natural variability by assessing long term observational time series and large
1356
samples of simulated states of the climate system.
1357
• Further development of statistical methods (including machine learning) for improved
1358
understanding of risks and return periods of rare events, including compound and freak events.
1359
• Dynamics of the larger scale and regional and local responses. While the local effects of large-
1360
scale circulation changes are reasonably understood, it is not clear which factors control or
1361
change the circulation itself. This is in particular true for changes in velocity and meandering
1362
of the jet stream, effects on blocking frequencies.
1363
• Increase process level understanding of impact of physical extremes on biogeochemical cycles
1364
and fluxes such as an enhanced flux of matter from land to sea during extreme mild and wet
1365
winters or enhanced greenhouse emission from sea bottom to atmosphere during marine heat
1366
wave events.
1367
• Interaction of multiple hazard drivers, since compound events are potentially very damaging
1368
for society.
1369
• Further to quantify economical costs of extreme events as well as impacts on health, ecosystem
1370
and environment.
1371
4. Conclusions and key messages
1372
In this review, we have been focused on extreme events and natural hazards in the Baltic Sea region.
1373
Temporal and spatial scales of the events which are causing these hazards ranges over many orders of
1374
magnitudes. Typical short-term phenomena are dynamical events like storms or heavy precipitation
1375
which are causing severe economical and human losses regionally or locally. Contrastingly, heat waves
1376
and cold spells are gradually developing events which prevail from weeks to months. Their impact on
1377
society and nature can cover the entire Baltic Sea catchment region.
1378
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
33
In Figure 14, we summarize how the hazards are related to the atmospheric, ocean and hydrological
1379
conditions. The weather in the Baltic Sea region is largely determined by the state of the large-scale
1380
atmospheric circulation. In winter, the variability is largely governed by the NAO with dominating
1381
strong westerlies and cyclones in its positive phase while more stable continental weather dominates in
1382
its negative phase. Also, in summer there are large differences between more cyclone-dominated
1383
weather with relatively mild air from the Atlantic and blocking-dominated weather with high pressure
1384
systems and warm continental air. Large scale atmospheric circulation is the main source of inter-annual
1385
variability of seasons and the extreme states are manifested in, for example, the extent of the seasonal
1386
ice cover.
1387
Regional atmospheric events, cyclones and blocking, are causing directly storm damages or triggering
1388
heat waves and forest fires, respectively. Cyclones are also generating storm surges and hazardous
1389
coastal flooding and ocean waves. Summertime blocking situations are frequently causing heat waves
1390
while in winter they are connected to cold spells. For long lasting situations, impacts of blocking are
1391
not restricted to land but also marine heat waves are generated and consequently massive algal blooms
1392
are formed as in 2018.
1393
An important aspect is that the most hazardous events are often combinations of several factors (i.e.
1394
compound events). For example, every cyclone can generate a storm surge, but the level of coastal
1395
flooding depends on the total water volume in the Baltic Sea. Positive water volume, which is caused
1396
by persistent westerlies, can provide an additional 50 cm (Leppäranta and Myrberg, 2009) to the
1397
maximum sea level. Moreover, a single storm is always causing a seiche oscillation and a sequence of
1398
storms can produce combined sea level changes due to the storm surge and seiche oscillation. In cities
1399
located at the river mouth, a sea flood can be further amplified by the river flood.
1400
Trends in circulation patterns are difficult to detect, the long-term temporal behavior of NAO is
1401
essentially irregular. There is, however, weak evidence that stationary wave amplitude has increased
1402
over the North Atlantic region, possibly as a result of weakening and/or northeastward shift of the North
1403
Atlantic storm track. There is an upward trend in the number of shallow and moderate cyclones, whereas
1404
there is no clear change, possibly a small decrease in the number of deep cyclones during the past
1405
decades. Sea level extremes are expected to increase in a changing climate and are directly related to
1406
changes in mean sea level, wind climate, storm tracks and circulation patterns.
1407
European summers have become warmer over the last three decades, partly explained by changes in
1408
blocking patterns (see section 2.1). There is a clear link between warmer summers and an increased risk
1409
of drying (in particular in spring) and heat waves in most of the area. Floods decrease in a large part of
1410
the Baltic Sea in spring but streamflow has increased in winter and autumn during the last decades while
1411
the mean flow shows insignificant changes. Stronger precipitation extremes associated with warmer
1412
climate can have strong impacts on society, in particular in urban regions, and are strongly associated
1413
with flooding and more intense cloud bursts. Results from new, high-resolution convective-permitting
1414
climate models indicate that increases in heavy rainfall associated with cloud bursts may increase even
1415
more than what has previously been found in coarser-scale regional climate models.
1416
Sea-effect snowfall events can be a serious threat to the coastal infrastructure and should be considered
1417
also in the future, although likely with an overall lower risk on an annual basis. More research is still
1418
needed for deepening the understanding of the sea-effect snowfall and for developing a reliable way to
1419
assess the occurrence of such events also in the changing conditions. Another wintertime phenomena of
1420
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
34
potentially hazardous consequences is ice ridging, being one of the sea ice extremes with the greatest
1421
impact potential on coastal infrastructures and shipping.
1422
Phytoplankton blooms are extreme, but natural biological events. However, eutrophication/de-
1423
eutrophication, pollution and changes in irradiation, temperature, salinity, carbon dioxide etc. may
1424
change their magnitude, timing and composition. Examples of extreme and mostly potentially toxic
1425
blooms are given, but reasons can hardly be identified. Their sudden and sporadic appearance
1426
complicates trend analyses and modelling. One trend that seems to be prominent is the prolongation of
1427
the phytoplankton growing season. Climate change is the most probable reason for this.
1428
Table 4 summarizes the changes of some extreme events for the past decades and using scenarios for
1429
the upcoming decades, here a positive trend means increasing probability of occurrence and a negative
1430
trend decreasing probability of occurrence.
1431
Table 4: Selected event and the estimated frequency of occurrence. Scale for changes (major decrease,
1432
minor decrease, no change, minor increase, major increase). Color, confidence scale (Low, medium,
1433
high).
1434
Event
Past decades
Future scenario
Number of moderate and
shallow extratropical
cyclones
Minor increase
no significant change
Number of deep extratropical
cyclones North Atlantic
Minor increase
Minor increase
Extreme ocean waves
North of 59°N
South of 59°N
no significant change (in
strength and frequency)
no significant change (in
strength and frequency)
minor increase in frequency in
wintertime
no significant change
Extreme sea levels (relative
mean sea level plus storm
surge)
North of 59°N
South of 59°N
minor decrease
minor increase
minor increase
major increase
Ice ridging
unknown
major decrease
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
35
Intense precipitation
minor increase
minor increase
Sea-effect snowfall
Unknown
Unknown
Heat waves
minor increase
major increase
Cold spells
major decrease
major decrease
Marine heat waves
minor increase
increase
Phytoplankton blooms
minor increase
minor increase
Extreme mild ice winters
major increase
major increase
Severe ice winters
major decrease
major decrease
(some uncertainties related to
changes in the large scale
circulation)
Drying
North of 59°N
South of 59°N
decrease
increase
mainly decrease, increase in
the north in the spring
increase in some regions in
spring and summer
River Flooding
increasing in winter/autumn,
decreasing in spring
decrease in spring
increase in winter
(low\high confidence)
For the selected societal elements discussed here, a combination of extremes and their changes are
1435
controlling the development and potential future damage, in addition to numerous other factors. For
1436
forest fires, drought and heat waves might lead to a doubling during the present century in some areas,
1437
however in other areas the risk might decrease due to increased precipitation. The frequency of coastal
1438
flooding responds mainly to sea level, but also wind, wave and precipitation features. The number of
1439
people exposed to coastal flooding in terms of annual damage is expected to increase with orders of
1440
magnitude. Baltic coastal cities are expected to be the main source of future coastal flood losses.
1441
Offshore wind application respond mainly to extreme wind and wave conditions, here loads and
1442
damages are important, but also conditions for operation and management activities imposing
1443
limitations in the potential use. Shipping in the Baltic Sea is affected by wind and wave conditions, icing
1444
due to sea spray and ice conditions, although mean wind and wave conditions are relatively low, the
1445
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
36
most severe storms affect maritime traffic. As ice winters are projected to get shorter, the wave climate
1446
is expected to get more severe (particularly in the eastern part of Bay of Bothnia and Gulf of Finland).
1447
Acknowledgements
1448
Contributions of JH, LT, JS have been supported by the Strategic Research Council at the Academy of
1449
Finland, project SmartSea (grant number 292 985). Contributions of AR and EN have been supported
1450
by FORMAS (grant number 2018-01784). Contribution of XL has been supported by Danish
1451
ForskEL/EUDP project OffshoreWake PSO-12521/64017-0017. The studies of ID were conducted
1452
under the subprogram 1 "The Nature Resources and Ecological Safety" of the State Research Program
1453
during 2016-2020 "The Nature Management and Ecology". Contributions of TO and AL have been
1454
supported by the National Nuclear Waste Management Fund in Finlad, and that of KJ additionally by
1455
the Academy of Finland, project HEATCLIM (grant number 329307) and TO by the Finnish Cultural
1456
Foundation, Satakunta Regional Fund.
1457
1458
5. References.
1459
Aakala, T., Pasanen, L., Helama, S., Vakkari, V., Drobyshev, I., Seppä, H., Kuuluvainen, T., Stivrins,
1460
N., Wallenius, T., Vasander, H., and Holmström, L. (2018) Multiscale variation in drought controlled
1461
historical forest fire activity in the boreal forests of eastern Fennoscandia. Ecological Monographs 88:
1462
74–91.
1463
Aalto, J., P. Pirinen, and K. Jylhä, 2016: New gridded daily climatology of Finland: Permutation-
1464
based uncertainty estimates and temporal trends in climate, J. Geophys. Res. Atmos., 121, 3807-3823.
1465
doi:10.1002/2015JD024651
1466
Aarnes, O. J., Ø. Breivik, and M. Reistad, 2012: Wave extremes in the northeast Atlantic. J. Climate,
1467
25, 1529–1543, https://doi.org/10.1175/jcli-d-11-00132.1
1468
Abild J, Nielsen B. Extreme values of wind speeds in Denmark. Technical Report M-2842, Risø
1469
National Laboratory, Roskilde, Denmark, 1991.
1470
Abadie, LM., Galarraga I., Markandya, A. and Sainz de Murieta, E. (2019). Risk measures and the
1471
distribution of damage curves for 600 European coastal cities. Environ. Res. Lett. 14 064021.
1472
Abadie, LM., Sainz de Murieta, E., Galarraga, I (2016). Climate Risk Assessment under Uncertainty:
1473
An Application to Main European Coastal Cities. Frontiers in Marine Science 3 (16 December 2016),
1474
https://doi.org/10.3389/fmars.2016.00265.
1475
Alfieri, L., Burek, P., Feyen, L., and Forzieri, G. (2015) Global warming increases the frequency of
1476
river floods in Europe, Hydrol. Earth Syst. Sci., 19, 2247-2260.
1477
Andersen, J.H., Carstensen, J., Conley, D.J., Dromph, K., Fleming-Lehtinen, V., Gustafsson, B.G., et
1478
al. (2017). Long-term temporal and spatial trends in eutrophication status of the Baltic Sea. Biological
1479
Reviews 92, 135-149. doi: 10.1111/brv.12221.
1480
Andersson, T., and Nilsson, S., 1990: Topographically induced convective snowbands over the Baltic
1481
Sea and their precipitation distribution, Weather Forecast, 5, 299-312.
1482
https://doi.org/10.5194/esd-2021-13
Preprint. Discussion started: 6 April 2021
c
Author(s) 2021. CC BY 4.0 License.
37
Apsīte E, Bakute A, Elferts D, Kurpniece L, Pallo I (2011) Climate change impacts on river runoff in
1483
Latvia. Clim Res 48:57-71.
1484
Arheimer B. and Lindström G. (2015) Climate impact on floods: changes in high flows in Sweden in
1485
the past and the future (1911–2100). Hydrol. Earth Syst. Sci., 19, 771–784.
1486
Åström D O, Tornevi A, Ebi KL, Rocklöv J, Forsberg B. 2016. Evolution of minimum mortality
1487
temperature in Stockholm, Sweden, 1901–2009. Environ Health Perspect 124:740–744;
1488
http://dx.doi.org/10.1289/ehp.1509692
1489
BACC Author Team (2008). Assessment of Climate Change for the Baltic Sea Basin. Springer-Verlag
1490
Berlin Heidelberg, p. 474
1491
BACC Author Team (2015). Second Assessment of Climate Change for the Baltic Sea Basin.
1492
Springer International Publishing, p. 501
1493
Baker-Austin, C., Trinanes, J. A., Salmenlinna, S., Löfdahl, M., Siitonen, A., Taylor, N. G., &
1494
Martinez-Urtaza, J. (2016). Heat Wave-Associated Vibriosis, Sweden and Finland, 2014. Emerging
1495
infectious diseases, 22(7), 1216–1220. https://doi.org/10.3201/eid2207.151996
1496
Barcikowska, MJ, Weaver SJ, Feser F, Russo S, Schenk F, Stone DA. et al. (2018) Euro-Atlantic
1497
winter storminess and precipitation extremes under 1.5°C vs. 2°C warming scenarios. Earth Syst Dyn
1498
9:679–699, doi:10.5194/esd-9-679-2018
1499
Barnes EA (2013) Revisiting the evidence linking Arctic amplification to extreme weather in
1500
midlatitudes. Geophys Res Lett 40:1–6. doi: 10.1002/grl.50880
1501
Bastine D., Larsén X. G., Witha B., Dörenkämper M,