- Email: [email protected]
Spatial analyses of 52 years of modelled sea state data for the Western Baltic Sea and their potential applicability for offshore and nearshore construction purposes
Ocean Engineering 96 (2015) 284–294
Contents lists available at ScienceDirect
Ocean Engineering journal homepage: www.elsevier.com/locate/oceaneng
Spatial analyses of 52 years of modelled sea state data for the Western Baltic Sea and their potential applicability for offshore and nearshore construction purposes Marcus Siewert n, Christian Schlamkow, Fokke Saathoff Universität Rostock, Chair of Geotechnics and Coastal Engineering, Justus-von-Liebig Weg 6, LAG II, 18059 Rostock, Germany
art ic l e i nf o
a b s t r a c t
Article history: Received 19 February 2014 Accepted 20 December 2014 Available online 30 January 2015
The present paper contains statistical analysis of modelled sea state data for the Western Baltic Sea for a time period of 52 years. Twenty charts were created, showing mean wave heights and frequencies of occurrence for different seasons. Taking a closer look at three potential areas for offshore wind energy in the Western Baltic the following mean significant wave heights were calculated (from west to east): Fehmarnbelt 0.6 m, Kadet Furrow 0.7 m and Arkona Basin 0.9 m. A comparison with waverider buoy measurements at five locations for different time series proves the good quality of the modelled data. These charts impart detailed information on the sea state from a spatial and temporal perspective which can be utilized by a wide range of users from different backgrounds. An exemplary monthly analysis of one location shows the possible application of the data set. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Wave statistics Sea state analysis Spatial distribution Baltic Sea
1. Introduction Sea state data is a governing parameter for the dimensioning of coastal protection structures (dikes and dunes) or harbor facilities. Therefore, the sea state is usually calculated for a small specific area, a line of points, or one single point. The growing market of offshore activities leads to a growing demand for spatial sea state data for the open sea. Wave spectra have been investigated for years and are a common tool for describing the sea state (Pierson and Moskowitz, 1964; Hasselmann et al., 1973). Still, they give no information about the spatial variability of the sea state. There are several publications that deal with sea state and wave heights in the Baltic Sea with different regional focus and different analysis methods (Schmager, 1979; Blomgren et al., 2001; Augustin, 2005; Cieślikiewicz and Paplińska-Swerpel, 2008; Feistel et al., 2008; Soomere et al., 2012). Numerical simulations and measurement campaigns have been carried out for different regions of the Baltic Sea and for different time scales. The way of presenting the analyses differs from single point analyses to spatial analyses for specific regions at different scales. No detailed spatial analyses have been made for the Western Baltic Sea until now. The Western Baltic Sea is highly frequented by ships (sailing boats, commercial fishing vessels, container ships, ferries, cruise ships and oil n
Corresponding author. Tel.: þ 49 381 4983689. E-mail addresses: [email protected] (M. Siewert), [email protected] (C. Schlamkow), [email protected] (F. Saathoff). http://dx.doi.org/10.1016/j.oceaneng.2014.12.029 0029-8018/& 2015 Elsevier Ltd. All rights reserved.
tankers) with no pilot obligation, but narrow and shallow shipping lanes and a growing number of offshore wind farms. Detailed spatial information on sea state conditions for different seasons in this region can be used by many users and gives an essential upgrade to the nautical security in this area. Subsequently, spatial sea state data of the Western Baltic Sea is statistically analyzed and plotted to charts. For the statistical analysis a 52-year data set covering the period 1948 to 1999 was used. Different statistical parameters have been chosen to show characteristic values of the sea state data (mean significant wave height and the frequencies of occurrence for waves with minimum height of 0.5 m, 1.0 m and 1.5 m). The calculations have been made for the whole time series and for different seasons (winter, spring, summer and autumn). This data analysis is part of the research project “BioBind” which is funded by the German Federal Ministry of Economics and Technology. The aim of the project BioBind is to develop an effective and fast airborne oil spill recovery system for nearshore shallow water areas and adverse weather conditions by using biogenic binders. Oil spill response equipment is highly sea state dependent, therefore it is necessary to have a detailed knowledge on the spatial distribution of sea state data to guarantee an effective oil spill response by choosing the appropriate equipment. In this context it was decided to analyze sea state data to evaluate the performance and application limits of spill response equipment. The Baltic Sea is a landlocked intercontinental sea in Western Europe which is dominated by wind waves which are fetch-limited.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
The spatial distribution of wave heights in the Western Baltic Sea is a result of two special features: 1. Winds from westerly directions (southwest is the dominating direction) are the main energy input for the sea state in this region. The longer the fetch length is, the bigger the waves are. Thereby the wave heights increase from west to east. 2. Winds from easterly directions (less frequent) can have a very long fetch length (x 4600 km) and generate high wave heights. These waves travel to the southwestern end of the Baltic Sea and bring even more energy to the areas with high waves heights due to the west wind input (Arkona Basin (e)). For increasing the mean wave heights in the Bay of Lübeck (c) these events are too seldom.
In Fig. 1 the frequencies for wind from different directions at the location Rostock–Warnemünde (situated at the southern coast of Western Baltic Sea) between 1996 and 2005 are shown. Nearly 45% of the wind is coming from directions between 2101 and 3001.
285
This strongly influences the creation and propagation of waves in the Western Baltic.
2. Methodology 2.1. Data basis The data basis for the statistical analysis is a result of different consecutive research projects conducted recently. The wind data was taken from the EU-funded research project HIPOCAS—Hindcast of Dynamic Processes of the Ocean and Coastal Areas of Europe (Feser et al., 2001; Soares et al., 2002). Within this research project the regional atmospheric model REMO was used, driven by the global reanalysis of the National Center for Environmental Prediction, to determine a hindcast of wind velocities for a wide area of northern Europe (Fig. 2). The wind data covered the area from 9.5–221E to 53.5–581N with a spatial resolution of ΔxE 0.51 and ΔyE0.51. The wave heights were taken from the project MUSTOK. In the project MUSTOK—Modeling of Extreme Storm Surges on the German Baltic Sea Coastline (Jensen, 2009), funded by the German Federal Ministry of Education and Research, different models have been used to determine extreme values for sea state as well as water levels for specific locations along the German Baltic Sea coastline. Within the sub-project SEBOK B wind data from the HIPOCAS project was used to calculate hourly significant wave heights for the German Baltic Sea area (Schlamkow and Fröhle, 2009). The sea state simulation was performed using the sea state model SWAN (Booij et al., 2001; The SWAN Team, 2011). For the SWAN simulations a bathymetry of the Baltic Sea with a spatial resolution of Δx E ΔyE1 nm was used (Fig. 2) (Seifert et al., 2001). As a combination of the modelled area (Southern Baltic) and the used input data (HIPOCAS) the model was called SOHIP. 2.2. Accuracy of modelled data
Fig. 1. Frequency of occurrence of wind for the western Baltic at Rostock– Warnemünde (1996–2005).
To evaluate the SOHIP simulation, the modelled sea state data were compared to wave measurements performed by University of Rostock in recent years (Fröhle, 2000) (except station Darss Sill which is operated by the Helmholz Zentrum Geesthacht Centre for
Fig. 2. Spatial extension of wind data within the HIPOCAS project (left, Weisse and Günther, 2007), Baltic Sea bathymetry and boundaries of the SOHIP model (right, Schlamkow and Fröhle, 2009).
286
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. 3. Western Baltic Sea with investigation sites from west to east Schönhagen, Heiligenhafen, Brodten, Rostock–Warnemünde and Darss Sill. (Picture source: Open Street Map).
Table 1 Acquisition period of wave measurements of the analyzed locations. Location
Schönhagen
Heiligenhafen
Brodten
Rostock Warnemünde
Darss Sill
Acquisition period
1996–1999
1996–1999
1997–1999
1998–1999
1991–1999
Fig. 5. Mean absolute deviation between modeled and measured wave heights at five locations in the Western Baltic Sea. The histogram bar shows the exact value, the digits are rounded on two decimal places.
Fig. 4. Comparison of measured and modelled mean wave heights at five locations in the Western Baltic Sea.
Materials and Coastal Research Deutscher Wetterdienst, 2013). Five locations in the Western Baltic were chosen and compared to their nearest grid point from the SOHIP simulation (Fig. 3). All measurements were performed by waverider buoys situated in water depth of 10 m to 35 m, calculating significant wave heights and wave periods as a mean value of a 30 min measurement. The temporal availability of measured data differs for the five locations (Table 1). For each location, the mean significant wave height for the astronomical seasons spring, summer, autumn and winter had been
calculated. A comparison of measured and modelled wave heights (mean value for one season for all available years) is shown in Fig. 4. The mean absolute deviation of modelled wave heights is 0.05 m for all locations, the maximum mean absolute deviations are 0.13 m for spring in Brodten and 0.11 m for spring in Heiligenhafen and winter in Schönhagen. Fig. 5 represents the mean absolute deviations for modelled data for the full period (histogram bar shows exact value, digits are rounded on two decimal places). For mean wave heights the comparison shows a high accordance between measured and modelled data. Therefore, the modelled wave heights from the SOHIP model can be used for further investigations on the spatial analyses of sea state data for the Western Baltic Sea. The comparison of the modeled data with another sea state model for the Western Baltic Sea (Soomere et al., 2012) shows an improvement of accuracy.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Maximum wave heights were also compared to measured data and showed a clear underestimation of the model. A comparison of measured and modeled wind velocities at five locations showed that lower maximum wind speeds at all investigation sites were modeled and thereby leaded to lower wave heights. The reasons for the lower maximum wind speeds in the model are not assessable from the author’s point of view, insecurities in measuring extreme wind speeds shall also be mentioned. There are possibilities to reduce uncertainty of wind hindcasts (Weisse and Feser, 2003), but the spatial application of wind correction still contains many uncertainties and therefore it was decided not to apply correction measures. As a result of the accuracy analysis, maximum significant wave heights were excluded from further investigations and were not plotted to charts. Due to the use of SWAN, the accuracy of the calculated wave periods is not sufficient for further use and the wave periods are not plotted to charts (Schlamkow and Fröhle, 2008). Nevertheless, wave periods are an important part of the sea state climate. Based on a large number of measurements at different locations, a consistent Table 2 Correlation of wave height and wave period. Wave height [m]
Wave period [s]
1.25 2.0 2.5 3.0 4.0
4.0 5.0 5.5 6.0 7.0
287
correlation for calculating wave periods from the wave heights could be derived for the Western Baltic Sea (Fröhle et al., 2006) (see Table 2). 2.3. Statistical analysis of sea state data The statistical analysis of wave heights was performed in an area ranging from 53.51N to 56.51N and 09.01E to 15.01E. In Fig. 6 the research area is shown, points of interest (a–e) and important islands (1–5) are highlighted. For the statistical analysis the following parameters had been calculated:
mean significant wave height; frequency of occurrence of waves Hs Z 0.5 m; frequency of occurrence of waves Hs Z 1.0 m; frequency of occurrence of waves Hs Z 1.5 m.
The mean significant wave height Hs had been calculated at every point of the model grid for a specific time period by calculating the arithmetical mean x using Eq. (1) (Bosch, 1993). x ¼ 1=nn ðx1 þ x2 þ ⋯ þ xn Þ
ð1Þ
To determine the frequency of occurrence of different wave heights the relative frequency r n had been calculated using Eq. (2): Where hn aj is the total number of events aj (significant wave height Z0.5 m/1.0 m/1.5 m). r n ðaj Þ ¼ 1=nn hn ðaj Þ;
j ¼ 1; 2; …; m
ð2Þ
To illustrate the seasonal changes of the mean sea state, the data was separated into the four meteorological seasons: spring (01.03.– 31.05.), summer (01.06.–31.08.), autumn (01.09.–31.11.) and winter (01.12.–28./29.02.). In addition to the seasonal calculation, the above described parameters had been calculated for the whole year. The seasonal classification is a deliberate data reduction and has been chosen because it contains a large amount of detailed data but still offers a comprehensive view. Due to the modelling process, the sea state data was separated into annual files. The described statistical analysis was performed separately for each year. Afterwards, the mean values for the complete time series were calculated using Eq. (1). All statistical calculations were done with MATLAB (2012) using specially made M-Files. With this set-up of analysis, 20 charts on the spatial distribution of wave heights in the Baltic Sea were created. An overview of the calculated combinations can be found in Table 3.
3. Results Fig. 6. Research area—Western Baltic Sea with points of interest and important islands marked. (a) Bay of Kiel, (b) Fehmarn Belt, (c) Bay of Lübeck, (d) Kadet Furrow, (e) Arkona Basin, (1) Fehmarn, (2) Lolland/Falster, (3) peninsula Fischland – Darss – Zingst, (4) Rügen, (5) Bornholm.
In this section some of the 20 created charts will be shown and discussed. The other charts can be found in the annex (Figs. A1– A16) to this article.
Table 3 Combination and names of calculated values for sea state data in the Baltic Sea. Meteorological seasons
Statistical parameters
Mean significant wave height Frequency of occurrence for waves Hs Z0.5 m Frequency of occurrence for waves Hs Z1.0 m Frequency of occurrence for waves Hs Z1.5 m
Spring
Summer
Autumn
Winter
Whole Year
Hs-Sp F0.5-Sp F1.0-Sp F1.5-Sp
Hs-Su F0.5-Su F1.0-Su F1.5-Su
Hs-Au F0.5-Au F1.0-Au F1.5-Au
Hs-Wi F0.5-Wi F1.0-Wi F1.5-Wi
Hs-Y F0.5-Y F1.0-Y F1.5-Y
288
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
In Fig. 7 the mean significant wave height for the whole year (Hs-Y) is shown. The highest mean significant wave height can be found between the island of Rügen (4) and the island of Bornholm (5) in the Arkona Basin (e) with Hs E0.85 m. In the Kadet Furrow (d), the mean significant wave height is Hs E0.65 m. In the Fehmarn Belt (b) the mean significant wave height is Hs E0.60 m. In the Bay of Kiel (a) the mean significant wave height is Hs E0.45 m. The smallest wave heights can be found in the Bay of Lübeck (c) Hs E0.35 m.
By separating the analysis into the meteorological seasons a more differentiated result was achieved. Fig. 8 shows the mean significant wave heights for all seasons (a) spring, (b) summer, (c) autumn and (d) winter). The comparison of these four images shows that there is a decrease of wave heights from spring to summer and an increase of wave heights from summer to winter. The mean significant wave heights for the four seasons at the locations a, b, c, d and e can be found in Table 4. The smallest wave heights can be found in summer and the highest wave heights can be found in winter. Again, an increase of wave heights from west to east is obvious. By setting up a threshold and calculating the frequency of occurrence for waves equal or above this threshold a second analysis was done. The thresholds were set to 0.5 m, 1.0 m and 1.5 m. Fig. 9 shows the frequency of occurrence of waves Hs Z0.5 m for the four seasons. In Fig. 10 the frequencies of occurrence for Hs Z0.5 m, Hs Z1.0 m and Hs Z1.5 m for winter are shown. Table 4 Mean significant wave heights for the four seasons at five specific locations [m]. Chart name Location a (Bay of Kiel) b (Fehmarn Belt) c (Bay of Lübeck) d (Kadet Furrow) e (Arkona Basin)
Hs-Sp (m) Hs-Su (m) Hs-Au (m) Hs-Wi (m) 0.45 0.55 0.35 0.60 0.75
Fig. 7. Hs-Y—mean significant wave height Hs (1948–1999).
Fig. 8. Mean significant waves heights for (a) spring, (b) summer, (c) autumn and (d) winter.
0.35 0.50 0.30 0.50 0.60
0.40 0.65 0.35 0.70 0.90
0.50 0.75 0.45 0.80 1.05
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
289
Fig. 9. Frequency of occurrence for waves Hs Z0.5 m for (a) spring, (b) summer, (c) autumn and (d) winter.
Fig. 10. Frequency of occurrence for waves for (a) Hs Z 0.5 m, (b) Hs Z 1.0 m and (c) Hs Z 1.5 m for winter.
Table 5 Frequencies of occurrence of waves for Hs Z 0.5 m (F0.5), Hs Z 1.0 m (F1.0) and Hs Z 1.5 m (F1.5) for winter (Wi) and summer (Su) at five specific locations [%].
Location
Chart name
F0.5-Wi (%)
F1.0-Wi (%)
F1.5-Wi (%)
F0.5-Su (%)
F1.0-Su (%)
F1.5-Su (%)
a (Bay of Kiel) b (Fehmarn Belt) c (Bay of Lübeck) d (Kadet Furrow) e (Arkona Basin)
35 63 32 65 77
10 25 5 30 47
2 7 0.5 10 25
15 38 13 40 50
1 9 0.4 10 17
0.07 2 0.03 2 5
290
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. 11. Monthly analysis of sea state at the offshore wind farm Baltic I with utilization frequency of activities limited to sea state higher than Hs Z 0.5 m, Hs Z 1.0 m and Hs Z1.5 m and oil skimmer efficiency.
Table 5 shows that the occurrence of higher waves, especially in the western part of the German Baltic Sea, is relatively low. For example, in the Bay of Lübeck (c) the occurrence of waves of Hs Z1.0 m is about 5% in winter and 0.4% in summer. Most of the Western Baltic Sea has a very low occurrence for waves higher than 1.5 m in winter and in summer. The maximum frequency of waves Hs Z1.5 m can be found in the Arkona Basin (d) with 25% in winter and 5% in summer. In fact, there are several points in the investigated area where the frequency of waves Hs Z1.5 m is close or equal to zero. Even though the mean wave heights seem to be relatively small, there are a lot of coastal and offshore activities which got limited applicability at wave heights higher than one meter. The efficiency of oil spill response equipment (booms and skimmers) decreases to 50% at wave heights between 1.0 m and 1.5 m (Fingas, 2010). Significant wave heights of Hs ¼1.0 m lead to a dredging overdepth of 0.1 m to 0.6 m and a dredging overwidth of 1.0 m to 4.0 m for trailing suction hopper dredgers (Wang and Lu, 2011). For the construction of offshore wind turbines and coastal protection structures (e.g. breakwaters or groynes), the relevant sea state conditions depend mainly on the type and size of the construction vessel, a universal limit is thereby hard to specify. Coastal construction companies often stop their work completely at wave heights around one meter, even though there is no literature on the construction limits. To calculate the window of opportunity for marine operations in general, vessel specific working limits are needed, correlated with the frequency of occurrence of specific wave heights (Germanischer Lloyd, 2012). A monthly analysis of the sea state at the location of the German offshore wind farm Baltic I (541360 50″N, 121400 00″) was conducted. Fig. 11 shows the utilization possibility of sea state dependent activities at this location. Depending on the chosen wave height limit, the operation possibility can be very low. In January operations with a wave height limit of 0.5 m can be conducted on 25% of the time, for a wave height limit of 1.0 m possible operation time raises to 65%, operations feasible to wave heights up to 1.5 m can be carried out 85% of the time. As an example the throughput efficiency (Fingas, 2011) of an average oil skimmer (percentage of oil presented to a skimmer versus that recovered in percent) was included in Fig. 11. Mean significant wave heights of Hs 40.75 m lead to a decrease of throughput efficiency of more than 50%. It needs to be mentioned that every skimmer is different, but still the impact of sea state on the efficiency stays significant. Thereby it can be stated, that
the effective use of average oil skimmers (ƞ450%) is only possible about half of the year. Containment booms are in the same range of efficiency. This influences the contingency plans of oil spill response activities significantly and calls for sea state independent alternatives and new developments.
4. Conclusion A spatial analysis of modelled sea state data for the Western Baltic Sea has been carried out using existing wind and sea state data from recent research projects. A range of 20 charts showing different statistical parameters (mean annual wave height, mean seasonal wave height, frequency of occurrence for waves higher than 0.5 m, 1.0 m and 1.5 m) have been created using data from more than 50 years. As an example for the application of this data, oil spill response equipment and its sea state dependence was described more closely. Next to the charts a NetCDF-File (Rew et al., 2011) with all computational results has been written. This makes it possible to extract the exact value of a specific location if needed for the increase of accuracy compared to the chart illustration. These further analyses are possible for the seasonal breakdown as well as for the monthly breakdown. With this data it is possible to plan and evaluate sea state dependent work at the open sea such as the construction of offshore wind turbines, the operational capability of spill response equipment or the execution of marine field investigations including sampling.
Acknowledgements The “BioBind” project was performed within the framework research program: Shipping and Offshore Technologies for the 21st Century 2011–2015, funded by the German Federal Ministry of Economics and Technology under grant number 03SX308. Additional support was provided by Dr. Martin Powilleit and Steffi Dimke of the University of Rostock. Appendix A Figs. A1–A16.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A1. Mean wave height in spring.
291
Fig. A4. Mean wave height in winter.
Fig. A2. Mean wave height in summer. Fig. A5. Mean wave height whole year.
Fig. A3. Mean wave height in autumn.
Fig. A6. Frequencies of occurrence for waves higher than 0.5 m whole year.
292
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A7. Frequencies of occurrence for waves higher than 1.0 m in spring.
Fig. A10. Frequencies of occurrence for waves higher than 1.0 m in winter.
Fig. A11. Frequencies of occurrence for waves higher than 1.0 m whole year. Fig. A8. Frequencies of occurrence for waves higher than 1.0 m in summer.
Fig. A9. Frequencies of occurrence for waves higher than 1.0 m in autumn.
Fig. A12. Frequencies of occurrence for waves higher than 1.5 m in spring.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A13. Frequencies of occurrence for waves higher than 1.5 m in summer.
Fig. A14. Frequencies of occurrence for waves higher than 1.5 m in autumn.
Fig. A15. Frequencies of occurrence for waves higher than 1.5 m in winter.
References Augustin, J., 2005. Das Seegangsklima der Ostsee zwischen 1958 und 2002 auf Grundlage numerischer Daten (German), PhD University of Hamburg. GKSS Research Center.
293
Fig. A16. Frequencies of occurrence for waves higher than 1.5 m whole year.
Blomgren, S., Larson, M., Hanson, H., 2001. Numerical modeling of the wave climate in the Southern Baltic Sea. J. Coastal Res. 17 (2), 342–352. Booij, N., Holthuijsen, L., Battjes, J., 2001. Ocean to near-shore wave modeling with SWAN. Coastal Dyn. 01, 335–344. Bosch, K., 1993. Statistik-Taschenbuch (German), second ed. R. Oldenbourg Verlag, Munich. Cieślikiewicz, W., Paplińska-Swerpel, B., 2008. A 44-year hindcast of wind wave fields over the Baltic Sea. Coastal Eng. 55, 894–905. Deutscher Wetterdienst , 2013. Die Deutschen Klimabeobachtungssysteme, Inventarbericht zum Global Climate Observing System (GCOS) (German), Selbstverlag DWD, Offenbach a.M. Feistel, R., Nausch, G., Wasmund, N., 2008. State and Evolution of the Baltic Sea, 1952–2005: A Detailed 50-year Survey of Meteorology and Climate, Physics, Chemistry, Biology and Marine Environment. Wiley, New Jersey. Feser, F., Weisse, R., von Storch, H., 2001. Multi-decadal atmospheric modeling for Europe yields multi-purpose data. Eos, Trans. Am. Geophys. Union 82, 305–310. Fingas, M., 2010. Oil Spill Science and Technology, Elesevier, Amsterdam. Fröhle, P., 2000. Messung und statistische Analyse von Seegang als Eingangsgröße für den Entwurf und die Bemessung von Bauwerken des Küstenwasserbaus (German), Rostocker Berichte aus dem Fachbereich Bauingenieurwesen Heft 2, Rostock. Fröhle, P.; Schlamkow, C.; Horlacher, H.-B.; Carstensen, D.; Drägerdt, S., 2006. Bemessungsseegang Außenküste Mecklenburg-Vorpommern (German), Final Report University of Rostock, Institute for Environmental Engineering, Chair of Coastal Engineering, Rostock. Germansicher Lloyd, General Guidelines for marine Projects, Technical policy board. 2012. Hasselmann, K., Barnett, T.P., Bouws, E., Carlson, H., Cartwright, D.E., Enke, K., Ewing, J.A., Gienapp, H., Hasselmann, D.E., Kruseman, P., Meerburg, A., Müller, P., Olbers, D.J., Richter, K., Sell, W., Walden, H., 1973. Measurements of wind wave growth and swell decay during the joint North Sea wave project (JONSWAP). Ger. Hydrogr. Z. 12 (Suppl. A8). Jensen, J., 2009. MUSTOK—Modellgestützte Untersuchungen zu extremen Sturmflutereignissen an der deutschen Ostseeküste: Eine Einführung (German). Die Küste 75, 1–9. MATLAB, 2012. MATLAB—The Language of Technical Computing Documentation R2011b. Mathworks, WWW-Page 〈http://www.mathworks.de/de/help/docu mentation-center.html〉. Pierson, W.J., Moskowitz, L., 1964. A proposed spectral form for fully developed wind seas based on similarity theory of S. A. Kitaigorodskii. J. Geophys. Res. 69 (24), 5181–5190. Rew, R., Davis, G., Emmerson, S., Davies, H., Hartnett, E., Heimbigner, D., 2011. The NetCDF Users Guide, Unidata Programm Center, Boulder. Schlamkow, C.; Fröhle, P., 2008. Wave period forecasting and hindcasting. In: Proc. COPEDEC VII—International Conference on Coastal and Port Engineering in Developing Countries. Schlamkow, C., Fröhle, P., 2009. Entwicklung von Methoden zur Bestimmung maßgebender hydrodynamischer Bemessungsparameter für Küstenschutzanlagen an der Ostsee (German). Die Küste 75, 207–230. Schmager, G., 1979. Atlas zur Ermittlung der Wellenhöhen in der südlichen Ostsee (German), Seehydrographischer Dienst der DDR. Seifert, T., Tauber, F., Kayser, B., 2001. A high resolution spherical grid topography of the Baltic Sea—revised edition, In: Proc. Baltic Sea Sci. Congr.. pp. 25-29. Soares, C., Weisse, R., Carretero, J., Alvarez, E., 2002. A 40 years hindcast of Wind, Sea Level and Waves in European Waters. In: Proc. 21st Int. Conf. Offshore Mech. Arct. Eng.. pp. 669-675. Soomere, T., Weisse, R., Behrens, A., 2012. Wave climate in the Arkona Basin, the Baltic Sea. Ocean Sci. 8, 287–300. The SWAN Team, 2011. USERMANUAL:SWAN Cycle III Version 40.85. Available: 〈http://swanmodel.sourceforge.net/online_doc/swanuse/node3.html〉.
294
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Wang, Y., Lu, K., Motion mode and dredging accuracy of trailing suction hopper dredger, In Proc. Mechanical Automation and Control Engineering (MACE), Second International Conference on, 2011. Weisse, R., Feser, F., 2003. Evaluation of a method to reduce uncertainty in wind hindcasts performed with regional atmosphere models. Coastal Eng. 48, 211–225.
Weisse, R., Günther, H., 2007. Wave climate and long-term changes for the Southern North Sea obtained from a high-resolution hindcast 1958–2002. Ocean Dyn. 57 (3), 162–171.
Contents lists available at ScienceDirect
Ocean Engineering journal homepage: www.elsevier.com/locate/oceaneng
Spatial analyses of 52 years of modelled sea state data for the Western Baltic Sea and their potential applicability for offshore and nearshore construction purposes Marcus Siewert n, Christian Schlamkow, Fokke Saathoff Universität Rostock, Chair of Geotechnics and Coastal Engineering, Justus-von-Liebig Weg 6, LAG II, 18059 Rostock, Germany
art ic l e i nf o
a b s t r a c t
Article history: Received 19 February 2014 Accepted 20 December 2014 Available online 30 January 2015
The present paper contains statistical analysis of modelled sea state data for the Western Baltic Sea for a time period of 52 years. Twenty charts were created, showing mean wave heights and frequencies of occurrence for different seasons. Taking a closer look at three potential areas for offshore wind energy in the Western Baltic the following mean significant wave heights were calculated (from west to east): Fehmarnbelt 0.6 m, Kadet Furrow 0.7 m and Arkona Basin 0.9 m. A comparison with waverider buoy measurements at five locations for different time series proves the good quality of the modelled data. These charts impart detailed information on the sea state from a spatial and temporal perspective which can be utilized by a wide range of users from different backgrounds. An exemplary monthly analysis of one location shows the possible application of the data set. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Wave statistics Sea state analysis Spatial distribution Baltic Sea
1. Introduction Sea state data is a governing parameter for the dimensioning of coastal protection structures (dikes and dunes) or harbor facilities. Therefore, the sea state is usually calculated for a small specific area, a line of points, or one single point. The growing market of offshore activities leads to a growing demand for spatial sea state data for the open sea. Wave spectra have been investigated for years and are a common tool for describing the sea state (Pierson and Moskowitz, 1964; Hasselmann et al., 1973). Still, they give no information about the spatial variability of the sea state. There are several publications that deal with sea state and wave heights in the Baltic Sea with different regional focus and different analysis methods (Schmager, 1979; Blomgren et al., 2001; Augustin, 2005; Cieślikiewicz and Paplińska-Swerpel, 2008; Feistel et al., 2008; Soomere et al., 2012). Numerical simulations and measurement campaigns have been carried out for different regions of the Baltic Sea and for different time scales. The way of presenting the analyses differs from single point analyses to spatial analyses for specific regions at different scales. No detailed spatial analyses have been made for the Western Baltic Sea until now. The Western Baltic Sea is highly frequented by ships (sailing boats, commercial fishing vessels, container ships, ferries, cruise ships and oil n
Corresponding author. Tel.: þ 49 381 4983689. E-mail addresses: [email protected] (M. Siewert), [email protected] (C. Schlamkow), [email protected] (F. Saathoff). http://dx.doi.org/10.1016/j.oceaneng.2014.12.029 0029-8018/& 2015 Elsevier Ltd. All rights reserved.
tankers) with no pilot obligation, but narrow and shallow shipping lanes and a growing number of offshore wind farms. Detailed spatial information on sea state conditions for different seasons in this region can be used by many users and gives an essential upgrade to the nautical security in this area. Subsequently, spatial sea state data of the Western Baltic Sea is statistically analyzed and plotted to charts. For the statistical analysis a 52-year data set covering the period 1948 to 1999 was used. Different statistical parameters have been chosen to show characteristic values of the sea state data (mean significant wave height and the frequencies of occurrence for waves with minimum height of 0.5 m, 1.0 m and 1.5 m). The calculations have been made for the whole time series and for different seasons (winter, spring, summer and autumn). This data analysis is part of the research project “BioBind” which is funded by the German Federal Ministry of Economics and Technology. The aim of the project BioBind is to develop an effective and fast airborne oil spill recovery system for nearshore shallow water areas and adverse weather conditions by using biogenic binders. Oil spill response equipment is highly sea state dependent, therefore it is necessary to have a detailed knowledge on the spatial distribution of sea state data to guarantee an effective oil spill response by choosing the appropriate equipment. In this context it was decided to analyze sea state data to evaluate the performance and application limits of spill response equipment. The Baltic Sea is a landlocked intercontinental sea in Western Europe which is dominated by wind waves which are fetch-limited.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
The spatial distribution of wave heights in the Western Baltic Sea is a result of two special features: 1. Winds from westerly directions (southwest is the dominating direction) are the main energy input for the sea state in this region. The longer the fetch length is, the bigger the waves are. Thereby the wave heights increase from west to east. 2. Winds from easterly directions (less frequent) can have a very long fetch length (x 4600 km) and generate high wave heights. These waves travel to the southwestern end of the Baltic Sea and bring even more energy to the areas with high waves heights due to the west wind input (Arkona Basin (e)). For increasing the mean wave heights in the Bay of Lübeck (c) these events are too seldom.
In Fig. 1 the frequencies for wind from different directions at the location Rostock–Warnemünde (situated at the southern coast of Western Baltic Sea) between 1996 and 2005 are shown. Nearly 45% of the wind is coming from directions between 2101 and 3001.
285
This strongly influences the creation and propagation of waves in the Western Baltic.
2. Methodology 2.1. Data basis The data basis for the statistical analysis is a result of different consecutive research projects conducted recently. The wind data was taken from the EU-funded research project HIPOCAS—Hindcast of Dynamic Processes of the Ocean and Coastal Areas of Europe (Feser et al., 2001; Soares et al., 2002). Within this research project the regional atmospheric model REMO was used, driven by the global reanalysis of the National Center for Environmental Prediction, to determine a hindcast of wind velocities for a wide area of northern Europe (Fig. 2). The wind data covered the area from 9.5–221E to 53.5–581N with a spatial resolution of ΔxE 0.51 and ΔyE0.51. The wave heights were taken from the project MUSTOK. In the project MUSTOK—Modeling of Extreme Storm Surges on the German Baltic Sea Coastline (Jensen, 2009), funded by the German Federal Ministry of Education and Research, different models have been used to determine extreme values for sea state as well as water levels for specific locations along the German Baltic Sea coastline. Within the sub-project SEBOK B wind data from the HIPOCAS project was used to calculate hourly significant wave heights for the German Baltic Sea area (Schlamkow and Fröhle, 2009). The sea state simulation was performed using the sea state model SWAN (Booij et al., 2001; The SWAN Team, 2011). For the SWAN simulations a bathymetry of the Baltic Sea with a spatial resolution of Δx E ΔyE1 nm was used (Fig. 2) (Seifert et al., 2001). As a combination of the modelled area (Southern Baltic) and the used input data (HIPOCAS) the model was called SOHIP. 2.2. Accuracy of modelled data
Fig. 1. Frequency of occurrence of wind for the western Baltic at Rostock– Warnemünde (1996–2005).
To evaluate the SOHIP simulation, the modelled sea state data were compared to wave measurements performed by University of Rostock in recent years (Fröhle, 2000) (except station Darss Sill which is operated by the Helmholz Zentrum Geesthacht Centre for
Fig. 2. Spatial extension of wind data within the HIPOCAS project (left, Weisse and Günther, 2007), Baltic Sea bathymetry and boundaries of the SOHIP model (right, Schlamkow and Fröhle, 2009).
286
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. 3. Western Baltic Sea with investigation sites from west to east Schönhagen, Heiligenhafen, Brodten, Rostock–Warnemünde and Darss Sill. (Picture source: Open Street Map).
Table 1 Acquisition period of wave measurements of the analyzed locations. Location
Schönhagen
Heiligenhafen
Brodten
Rostock Warnemünde
Darss Sill
Acquisition period
1996–1999
1996–1999
1997–1999
1998–1999
1991–1999
Fig. 5. Mean absolute deviation between modeled and measured wave heights at five locations in the Western Baltic Sea. The histogram bar shows the exact value, the digits are rounded on two decimal places.
Fig. 4. Comparison of measured and modelled mean wave heights at five locations in the Western Baltic Sea.
Materials and Coastal Research Deutscher Wetterdienst, 2013). Five locations in the Western Baltic were chosen and compared to their nearest grid point from the SOHIP simulation (Fig. 3). All measurements were performed by waverider buoys situated in water depth of 10 m to 35 m, calculating significant wave heights and wave periods as a mean value of a 30 min measurement. The temporal availability of measured data differs for the five locations (Table 1). For each location, the mean significant wave height for the astronomical seasons spring, summer, autumn and winter had been
calculated. A comparison of measured and modelled wave heights (mean value for one season for all available years) is shown in Fig. 4. The mean absolute deviation of modelled wave heights is 0.05 m for all locations, the maximum mean absolute deviations are 0.13 m for spring in Brodten and 0.11 m for spring in Heiligenhafen and winter in Schönhagen. Fig. 5 represents the mean absolute deviations for modelled data for the full period (histogram bar shows exact value, digits are rounded on two decimal places). For mean wave heights the comparison shows a high accordance between measured and modelled data. Therefore, the modelled wave heights from the SOHIP model can be used for further investigations on the spatial analyses of sea state data for the Western Baltic Sea. The comparison of the modeled data with another sea state model for the Western Baltic Sea (Soomere et al., 2012) shows an improvement of accuracy.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Maximum wave heights were also compared to measured data and showed a clear underestimation of the model. A comparison of measured and modeled wind velocities at five locations showed that lower maximum wind speeds at all investigation sites were modeled and thereby leaded to lower wave heights. The reasons for the lower maximum wind speeds in the model are not assessable from the author’s point of view, insecurities in measuring extreme wind speeds shall also be mentioned. There are possibilities to reduce uncertainty of wind hindcasts (Weisse and Feser, 2003), but the spatial application of wind correction still contains many uncertainties and therefore it was decided not to apply correction measures. As a result of the accuracy analysis, maximum significant wave heights were excluded from further investigations and were not plotted to charts. Due to the use of SWAN, the accuracy of the calculated wave periods is not sufficient for further use and the wave periods are not plotted to charts (Schlamkow and Fröhle, 2008). Nevertheless, wave periods are an important part of the sea state climate. Based on a large number of measurements at different locations, a consistent Table 2 Correlation of wave height and wave period. Wave height [m]
Wave period [s]
1.25 2.0 2.5 3.0 4.0
4.0 5.0 5.5 6.0 7.0
287
correlation for calculating wave periods from the wave heights could be derived for the Western Baltic Sea (Fröhle et al., 2006) (see Table 2). 2.3. Statistical analysis of sea state data The statistical analysis of wave heights was performed in an area ranging from 53.51N to 56.51N and 09.01E to 15.01E. In Fig. 6 the research area is shown, points of interest (a–e) and important islands (1–5) are highlighted. For the statistical analysis the following parameters had been calculated:
mean significant wave height; frequency of occurrence of waves Hs Z 0.5 m; frequency of occurrence of waves Hs Z 1.0 m; frequency of occurrence of waves Hs Z 1.5 m.
The mean significant wave height Hs had been calculated at every point of the model grid for a specific time period by calculating the arithmetical mean x using Eq. (1) (Bosch, 1993). x ¼ 1=nn ðx1 þ x2 þ ⋯ þ xn Þ
ð1Þ
To determine the frequency of occurrence of different wave heights the relative frequency r n had been calculated using Eq. (2): Where hn aj is the total number of events aj (significant wave height Z0.5 m/1.0 m/1.5 m). r n ðaj Þ ¼ 1=nn hn ðaj Þ;
j ¼ 1; 2; …; m
ð2Þ
To illustrate the seasonal changes of the mean sea state, the data was separated into the four meteorological seasons: spring (01.03.– 31.05.), summer (01.06.–31.08.), autumn (01.09.–31.11.) and winter (01.12.–28./29.02.). In addition to the seasonal calculation, the above described parameters had been calculated for the whole year. The seasonal classification is a deliberate data reduction and has been chosen because it contains a large amount of detailed data but still offers a comprehensive view. Due to the modelling process, the sea state data was separated into annual files. The described statistical analysis was performed separately for each year. Afterwards, the mean values for the complete time series were calculated using Eq. (1). All statistical calculations were done with MATLAB (2012) using specially made M-Files. With this set-up of analysis, 20 charts on the spatial distribution of wave heights in the Baltic Sea were created. An overview of the calculated combinations can be found in Table 3.
3. Results Fig. 6. Research area—Western Baltic Sea with points of interest and important islands marked. (a) Bay of Kiel, (b) Fehmarn Belt, (c) Bay of Lübeck, (d) Kadet Furrow, (e) Arkona Basin, (1) Fehmarn, (2) Lolland/Falster, (3) peninsula Fischland – Darss – Zingst, (4) Rügen, (5) Bornholm.
In this section some of the 20 created charts will be shown and discussed. The other charts can be found in the annex (Figs. A1– A16) to this article.
Table 3 Combination and names of calculated values for sea state data in the Baltic Sea. Meteorological seasons
Statistical parameters
Mean significant wave height Frequency of occurrence for waves Hs Z0.5 m Frequency of occurrence for waves Hs Z1.0 m Frequency of occurrence for waves Hs Z1.5 m
Spring
Summer
Autumn
Winter
Whole Year
Hs-Sp F0.5-Sp F1.0-Sp F1.5-Sp
Hs-Su F0.5-Su F1.0-Su F1.5-Su
Hs-Au F0.5-Au F1.0-Au F1.5-Au
Hs-Wi F0.5-Wi F1.0-Wi F1.5-Wi
Hs-Y F0.5-Y F1.0-Y F1.5-Y
288
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
In Fig. 7 the mean significant wave height for the whole year (Hs-Y) is shown. The highest mean significant wave height can be found between the island of Rügen (4) and the island of Bornholm (5) in the Arkona Basin (e) with Hs E0.85 m. In the Kadet Furrow (d), the mean significant wave height is Hs E0.65 m. In the Fehmarn Belt (b) the mean significant wave height is Hs E0.60 m. In the Bay of Kiel (a) the mean significant wave height is Hs E0.45 m. The smallest wave heights can be found in the Bay of Lübeck (c) Hs E0.35 m.
By separating the analysis into the meteorological seasons a more differentiated result was achieved. Fig. 8 shows the mean significant wave heights for all seasons (a) spring, (b) summer, (c) autumn and (d) winter). The comparison of these four images shows that there is a decrease of wave heights from spring to summer and an increase of wave heights from summer to winter. The mean significant wave heights for the four seasons at the locations a, b, c, d and e can be found in Table 4. The smallest wave heights can be found in summer and the highest wave heights can be found in winter. Again, an increase of wave heights from west to east is obvious. By setting up a threshold and calculating the frequency of occurrence for waves equal or above this threshold a second analysis was done. The thresholds were set to 0.5 m, 1.0 m and 1.5 m. Fig. 9 shows the frequency of occurrence of waves Hs Z0.5 m for the four seasons. In Fig. 10 the frequencies of occurrence for Hs Z0.5 m, Hs Z1.0 m and Hs Z1.5 m for winter are shown. Table 4 Mean significant wave heights for the four seasons at five specific locations [m]. Chart name Location a (Bay of Kiel) b (Fehmarn Belt) c (Bay of Lübeck) d (Kadet Furrow) e (Arkona Basin)
Hs-Sp (m) Hs-Su (m) Hs-Au (m) Hs-Wi (m) 0.45 0.55 0.35 0.60 0.75
Fig. 7. Hs-Y—mean significant wave height Hs (1948–1999).
Fig. 8. Mean significant waves heights for (a) spring, (b) summer, (c) autumn and (d) winter.
0.35 0.50 0.30 0.50 0.60
0.40 0.65 0.35 0.70 0.90
0.50 0.75 0.45 0.80 1.05
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
289
Fig. 9. Frequency of occurrence for waves Hs Z0.5 m for (a) spring, (b) summer, (c) autumn and (d) winter.
Fig. 10. Frequency of occurrence for waves for (a) Hs Z 0.5 m, (b) Hs Z 1.0 m and (c) Hs Z 1.5 m for winter.
Table 5 Frequencies of occurrence of waves for Hs Z 0.5 m (F0.5), Hs Z 1.0 m (F1.0) and Hs Z 1.5 m (F1.5) for winter (Wi) and summer (Su) at five specific locations [%].
Location
Chart name
F0.5-Wi (%)
F1.0-Wi (%)
F1.5-Wi (%)
F0.5-Su (%)
F1.0-Su (%)
F1.5-Su (%)
a (Bay of Kiel) b (Fehmarn Belt) c (Bay of Lübeck) d (Kadet Furrow) e (Arkona Basin)
35 63 32 65 77
10 25 5 30 47
2 7 0.5 10 25
15 38 13 40 50
1 9 0.4 10 17
0.07 2 0.03 2 5
290
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. 11. Monthly analysis of sea state at the offshore wind farm Baltic I with utilization frequency of activities limited to sea state higher than Hs Z 0.5 m, Hs Z 1.0 m and Hs Z1.5 m and oil skimmer efficiency.
Table 5 shows that the occurrence of higher waves, especially in the western part of the German Baltic Sea, is relatively low. For example, in the Bay of Lübeck (c) the occurrence of waves of Hs Z1.0 m is about 5% in winter and 0.4% in summer. Most of the Western Baltic Sea has a very low occurrence for waves higher than 1.5 m in winter and in summer. The maximum frequency of waves Hs Z1.5 m can be found in the Arkona Basin (d) with 25% in winter and 5% in summer. In fact, there are several points in the investigated area where the frequency of waves Hs Z1.5 m is close or equal to zero. Even though the mean wave heights seem to be relatively small, there are a lot of coastal and offshore activities which got limited applicability at wave heights higher than one meter. The efficiency of oil spill response equipment (booms and skimmers) decreases to 50% at wave heights between 1.0 m and 1.5 m (Fingas, 2010). Significant wave heights of Hs ¼1.0 m lead to a dredging overdepth of 0.1 m to 0.6 m and a dredging overwidth of 1.0 m to 4.0 m for trailing suction hopper dredgers (Wang and Lu, 2011). For the construction of offshore wind turbines and coastal protection structures (e.g. breakwaters or groynes), the relevant sea state conditions depend mainly on the type and size of the construction vessel, a universal limit is thereby hard to specify. Coastal construction companies often stop their work completely at wave heights around one meter, even though there is no literature on the construction limits. To calculate the window of opportunity for marine operations in general, vessel specific working limits are needed, correlated with the frequency of occurrence of specific wave heights (Germanischer Lloyd, 2012). A monthly analysis of the sea state at the location of the German offshore wind farm Baltic I (541360 50″N, 121400 00″) was conducted. Fig. 11 shows the utilization possibility of sea state dependent activities at this location. Depending on the chosen wave height limit, the operation possibility can be very low. In January operations with a wave height limit of 0.5 m can be conducted on 25% of the time, for a wave height limit of 1.0 m possible operation time raises to 65%, operations feasible to wave heights up to 1.5 m can be carried out 85% of the time. As an example the throughput efficiency (Fingas, 2011) of an average oil skimmer (percentage of oil presented to a skimmer versus that recovered in percent) was included in Fig. 11. Mean significant wave heights of Hs 40.75 m lead to a decrease of throughput efficiency of more than 50%. It needs to be mentioned that every skimmer is different, but still the impact of sea state on the efficiency stays significant. Thereby it can be stated, that
the effective use of average oil skimmers (ƞ450%) is only possible about half of the year. Containment booms are in the same range of efficiency. This influences the contingency plans of oil spill response activities significantly and calls for sea state independent alternatives and new developments.
4. Conclusion A spatial analysis of modelled sea state data for the Western Baltic Sea has been carried out using existing wind and sea state data from recent research projects. A range of 20 charts showing different statistical parameters (mean annual wave height, mean seasonal wave height, frequency of occurrence for waves higher than 0.5 m, 1.0 m and 1.5 m) have been created using data from more than 50 years. As an example for the application of this data, oil spill response equipment and its sea state dependence was described more closely. Next to the charts a NetCDF-File (Rew et al., 2011) with all computational results has been written. This makes it possible to extract the exact value of a specific location if needed for the increase of accuracy compared to the chart illustration. These further analyses are possible for the seasonal breakdown as well as for the monthly breakdown. With this data it is possible to plan and evaluate sea state dependent work at the open sea such as the construction of offshore wind turbines, the operational capability of spill response equipment or the execution of marine field investigations including sampling.
Acknowledgements The “BioBind” project was performed within the framework research program: Shipping and Offshore Technologies for the 21st Century 2011–2015, funded by the German Federal Ministry of Economics and Technology under grant number 03SX308. Additional support was provided by Dr. Martin Powilleit and Steffi Dimke of the University of Rostock. Appendix A Figs. A1–A16.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A1. Mean wave height in spring.
291
Fig. A4. Mean wave height in winter.
Fig. A2. Mean wave height in summer. Fig. A5. Mean wave height whole year.
Fig. A3. Mean wave height in autumn.
Fig. A6. Frequencies of occurrence for waves higher than 0.5 m whole year.
292
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A7. Frequencies of occurrence for waves higher than 1.0 m in spring.
Fig. A10. Frequencies of occurrence for waves higher than 1.0 m in winter.
Fig. A11. Frequencies of occurrence for waves higher than 1.0 m whole year. Fig. A8. Frequencies of occurrence for waves higher than 1.0 m in summer.
Fig. A9. Frequencies of occurrence for waves higher than 1.0 m in autumn.
Fig. A12. Frequencies of occurrence for waves higher than 1.5 m in spring.
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Fig. A13. Frequencies of occurrence for waves higher than 1.5 m in summer.
Fig. A14. Frequencies of occurrence for waves higher than 1.5 m in autumn.
Fig. A15. Frequencies of occurrence for waves higher than 1.5 m in winter.
References Augustin, J., 2005. Das Seegangsklima der Ostsee zwischen 1958 und 2002 auf Grundlage numerischer Daten (German), PhD University of Hamburg. GKSS Research Center.
293
Fig. A16. Frequencies of occurrence for waves higher than 1.5 m whole year.
Blomgren, S., Larson, M., Hanson, H., 2001. Numerical modeling of the wave climate in the Southern Baltic Sea. J. Coastal Res. 17 (2), 342–352. Booij, N., Holthuijsen, L., Battjes, J., 2001. Ocean to near-shore wave modeling with SWAN. Coastal Dyn. 01, 335–344. Bosch, K., 1993. Statistik-Taschenbuch (German), second ed. R. Oldenbourg Verlag, Munich. Cieślikiewicz, W., Paplińska-Swerpel, B., 2008. A 44-year hindcast of wind wave fields over the Baltic Sea. Coastal Eng. 55, 894–905. Deutscher Wetterdienst , 2013. Die Deutschen Klimabeobachtungssysteme, Inventarbericht zum Global Climate Observing System (GCOS) (German), Selbstverlag DWD, Offenbach a.M. Feistel, R., Nausch, G., Wasmund, N., 2008. State and Evolution of the Baltic Sea, 1952–2005: A Detailed 50-year Survey of Meteorology and Climate, Physics, Chemistry, Biology and Marine Environment. Wiley, New Jersey. Feser, F., Weisse, R., von Storch, H., 2001. Multi-decadal atmospheric modeling for Europe yields multi-purpose data. Eos, Trans. Am. Geophys. Union 82, 305–310. Fingas, M., 2010. Oil Spill Science and Technology, Elesevier, Amsterdam. Fröhle, P., 2000. Messung und statistische Analyse von Seegang als Eingangsgröße für den Entwurf und die Bemessung von Bauwerken des Küstenwasserbaus (German), Rostocker Berichte aus dem Fachbereich Bauingenieurwesen Heft 2, Rostock. Fröhle, P.; Schlamkow, C.; Horlacher, H.-B.; Carstensen, D.; Drägerdt, S., 2006. Bemessungsseegang Außenküste Mecklenburg-Vorpommern (German), Final Report University of Rostock, Institute for Environmental Engineering, Chair of Coastal Engineering, Rostock. Germansicher Lloyd, General Guidelines for marine Projects, Technical policy board. 2012. Hasselmann, K., Barnett, T.P., Bouws, E., Carlson, H., Cartwright, D.E., Enke, K., Ewing, J.A., Gienapp, H., Hasselmann, D.E., Kruseman, P., Meerburg, A., Müller, P., Olbers, D.J., Richter, K., Sell, W., Walden, H., 1973. Measurements of wind wave growth and swell decay during the joint North Sea wave project (JONSWAP). Ger. Hydrogr. Z. 12 (Suppl. A8). Jensen, J., 2009. MUSTOK—Modellgestützte Untersuchungen zu extremen Sturmflutereignissen an der deutschen Ostseeküste: Eine Einführung (German). Die Küste 75, 1–9. MATLAB, 2012. MATLAB—The Language of Technical Computing Documentation R2011b. Mathworks, WWW-Page 〈http://www.mathworks.de/de/help/docu mentation-center.html〉. Pierson, W.J., Moskowitz, L., 1964. A proposed spectral form for fully developed wind seas based on similarity theory of S. A. Kitaigorodskii. J. Geophys. Res. 69 (24), 5181–5190. Rew, R., Davis, G., Emmerson, S., Davies, H., Hartnett, E., Heimbigner, D., 2011. The NetCDF Users Guide, Unidata Programm Center, Boulder. Schlamkow, C.; Fröhle, P., 2008. Wave period forecasting and hindcasting. In: Proc. COPEDEC VII—International Conference on Coastal and Port Engineering in Developing Countries. Schlamkow, C., Fröhle, P., 2009. Entwicklung von Methoden zur Bestimmung maßgebender hydrodynamischer Bemessungsparameter für Küstenschutzanlagen an der Ostsee (German). Die Küste 75, 207–230. Schmager, G., 1979. Atlas zur Ermittlung der Wellenhöhen in der südlichen Ostsee (German), Seehydrographischer Dienst der DDR. Seifert, T., Tauber, F., Kayser, B., 2001. A high resolution spherical grid topography of the Baltic Sea—revised edition, In: Proc. Baltic Sea Sci. Congr.. pp. 25-29. Soares, C., Weisse, R., Carretero, J., Alvarez, E., 2002. A 40 years hindcast of Wind, Sea Level and Waves in European Waters. In: Proc. 21st Int. Conf. Offshore Mech. Arct. Eng.. pp. 669-675. Soomere, T., Weisse, R., Behrens, A., 2012. Wave climate in the Arkona Basin, the Baltic Sea. Ocean Sci. 8, 287–300. The SWAN Team, 2011. USERMANUAL:SWAN Cycle III Version 40.85. Available: 〈http://swanmodel.sourceforge.net/online_doc/swanuse/node3.html〉.
294
M. Siewert et al. / Ocean Engineering 96 (2015) 284–294
Wang, Y., Lu, K., Motion mode and dredging accuracy of trailing suction hopper dredger, In Proc. Mechanical Automation and Control Engineering (MACE), Second International Conference on, 2011. Weisse, R., Feser, F., 2003. Evaluation of a method to reduce uncertainty in wind hindcasts performed with regional atmosphere models. Coastal Eng. 48, 211–225.
Weisse, R., Günther, H., 2007. Wave climate and long-term changes for the Southern North Sea obtained from a high-resolution hindcast 1958–2002. Ocean Dyn. 57 (3), 162–171.