A case study of the impact of cyclonic trajectories on sea-level extremes in the Gulf of Finland
Introduction
The coastline of the Gulf of Finland is currently undergoing extensive construction. New port terminals have been designed or built in St. Petersburg, Primorsk, Ust’-Luga, Batareinya Bay, Vistino, Vysotsk and other cities. The second block of the Leningrad Nuclear Power Station (LNPS) is planned on the coast of Koporskay Bay. The St. Petersburg Flood Protection Barrier, which crosses Neva Bay 25 km west of St. Petersburg, is to be completed in the next few years. For the design of these projects predictions of extreme water levels are needed with an annual exceedance probability of up to 0.01% (once in 10,000 years). Extreme water levels in the Baltic Sea are generated by deep cyclones passing over it. To determine extreme levels with rare return periods, a statistical approach is used, which requires a long time series of observations (Gidrometeoizdat, 1973).
The longest time series in the Gulf of Finland is for Kronshtadt for which sea gauge data are available starting in 1805. In 1898 the sea gauge was replaced by a paper tape mareograph. In 1877 a mareograph was installed near the Gornii Institute in St. Petersburg. Floods in St. Petersburg have been recorded since 1703. Data from these two stations were used for the design of the Flood Protection Barrier. It was estimated that in St. Petersburg the extreme water level with annual exceedance probability 0.01% was 540 cm above local mean sea level, and 465 cm in Kronshtadt (Nezhykhovski, 1988). Long series of observations also exist for Hanko (since 1887), Vyborg (since 1889), Tallinn and Narva-Jõesuu (since 1899) and Helsinki (since 1904). For other sites in the Gulf of Finland the time series are shorter. For example, on station Staroe Garkolovo located near the LNPS, water level was measured for 1924–1940 and 1957–1988. For such short records the frequency curves are derived using correlations with other stations. Water levels with 0.01% annual exceedance probability obtained for Staroe Garkolovo using various methods of extrapolation are in the range from 330 to 470 cm and this range of 140 cm is significant.
A different approach to the evaluation of extreme sea-level height is the mathematical modelling of water levels in the Baltic Sea when a deep cyclone propagates along the most dangerous trajectories and with the most dangerous velocities for a given location on the coast. In this paper we try to predict the extreme sea-level heights for specific coastal sites in the Gulf of Finland under their particular “most dangerous” conditions. It is assumed that a cyclone has a round form. An analytical expression is proposed for the determination of the atmospheric pressure in an idealized moving cyclone, which takes into account the existence of a cold front and the time evolution of the cyclone intensity. The empirical parameters in this relation are estimated based on those of the cyclone ‘Erwin’ (known as ‘Gudrun’ in the Nordic countries), which passed over the Baltic Sea region on 8–9 January 2005 when historical maximum water levels were exceeded at almost all stations in the Gulf of Finland and at some stations outside the Gulf. According to Suursaar et al., 2006 this cyclone was among the five most powerful ones in recorded history.
The hydrodynamic model of the Baltic Sea BSM6 is used with these parameters.
Postglacial vertical land movements and the impacts of possible climate change are not considered in this paper.
Section snippets
Observed historical maxima of water level in the Baltic Sea
Observed historical water level maxima from Baltic Sea stations are shown in Fig. 1. Though durations of the measurements vary, this figure shows the general character of water elevations in the Sea. They have maxima in the heads of the gulfs and minima in the western part of the Central Baltic.
For most stations the historical maxima were registered during 12 storms, shown in the table in Fig. 1. During the flood of 19 November 1824 the absolute maximum was in the shallow head of the Gulf of
The Baltic Sea model BSM6
For the simulation of water level oscillations, the model of the Baltic Sea, BSM6, constructed with the modelling system CARDINAL (Klevanny et al., 1994, Klevanny et al., 2001), was applied. This model (and its earlier versions) has been used successfully at the St. Petersburg Center for Hydrometeorology and Environmental Monitoring for water level forecasts since 23 December 1999 with an advance time of 48 h. The wind and pressure fields are from the HIRLAM model of SMHI. The model is based on
Approximation of wind-pressure fields in an idealized cyclone
Atmospheric pressure in a deep cyclone may be approximated with the relationwhich was used in Miyazaki et al., 1961 for typhoons and is in good agreement with the observations. Here Pa′ is the surface pressure at distance r(t) from the cyclone center xc(t), yc(t), P∞ the pressure outside the cyclone and ΔP the difference between P∞ and pressure in the cyclone center. As it follows from Eq. (2), rТ is the radius, where the pressure gradient and hence the wind velocity are
Estimation of parameters for a deep idealized cyclone
To evaluate parameters rТ, ΔPо, P∞ and ΔT we will focus on cyclone ‘Erwin’. In Averkiev and Klevannyy, 2007 rТ in cyclone ‘Erwin’ was estimated at 200 km. This value will be also used in this paper. For comparison in cyclone ‘Katarina’ (18 March 2003), rТ was in the range of 190–280 km. Additional sensitivity analyses for this parameter should be done in future work. The pressure in cyclone ‘Erwin’ decreased to 960 mbar and we assigned this value as the pressure minimum. A similar value was also
Numerical experiments
With the parameters derived in Section 5 for the deep cyclone ‘Erwin’, numerical experiments were run using the Baltic Sea model BSM6 with different trajectories and cyclone speeds. The aim of these simulations was to determine the extreme water levels possible due to such a cyclone in St. Petersburg, Kronshtadt, LNPS, Vyborg, Kotka, Loviisa (site of the Finnish Nuclear Power Station), Helsinki, Hanko, Tallinn and Pärnu (Fig. 7).
Experiments showed that maximum water level is a monotonic
Results and conclusions
It can be found in the literature that if a cyclone's speed is close to the speed of propagation of gravity waves resonance will occur and wave height will increase. With the average depth in the Gulf of Finland being 36 m, the mean velocity of propagation of long waves is 19 m/s which is higher than the most dangerous ones (11–15 m/s for the Eastern Gulf of Finland, Fig. 4d). However, the depths in the Gulf of Finland are very irregular. The cross-section mean depths decrease from 47 m at the
Acknowledgements
This work is supported by the Science for Peace NATO project ‘Flood Risk Analysis for the Gulf of Finland and Saint Petersburg # 981382. The authors acknowledge I. Dailidiene (Lithuania), A. Kudure (Latvia), T.Kõuts (Estonia), K. Kahma (Finland), B. Broman (Sweden), C. Mudersbach (Germany), V. Huess (Denmark) and M. Kaminska (Poland) for providing historical maximum water levels from their countries.
References (17)
- Apukhtin, A.A., 2007. Leningrad Nuclear Power Station—2. Technical Report. “Atomenergoproject” Design Institute, St....
- et al.
Determining cyclone trajectories and velocities leading to extreme sea level rises in the Gulf of Finland
Russian Meteorology and Hydrology
(2007) Manual for Calculation of Parameters of Hydrological Regime in the Coastal Zone of Seas and in the Estuaries for Engineering Designs
(1973)- Khandozhko, L.A., 1981. Characteristics of wind variability in the coastal zone of the Gulf of Finland. Proceedings of...
Floods in St. Petersburg in the conditions of uncompleted Barrier.
Izvestya Russian Academy of Science Geography
(2006)- et al.
Water level forecasts for the eastern Gulf of Finland
Bulletin of the Maritime Institute in Gdansk
(2001) - et al.
An integrated modelling system for coastal area dynamics
International Journal for Numerical Methods in Fluids
(1994) - Lopatukhin, L.I., Mironov, M.E., Pomeranets, K.S., Trapeznikov, Yu.A., Chernyshova, E.S., 2006. Evaluation of extreme...
Cited by (0)
- 1
Tel.: +7 812 2242155.