A stochastic model for the sea level in the Estonian coastal area

Spatial variability in the trends in extreme storm surges and weekly-scale high water levels in the eastern Baltic Sea

“…These data sets have been extensively described and analysed from various viewpoints, e.g., the overall course and periodic components of the water level (Raudsepp et al, 1999), projections of extreme water levels and their return periods (Suursaar and Sooäär, 2007), the feasibility of ensemble approach for projections of water level extremes (Eelsalu et al, 2014) or the separation of the total water level into short-term and weekly-scale components . The gauge at Tallinn Harbour seems to most adequately represent the offshore water level (Eelsalu et al, 2014) and the data from this site is chosen to illustrate our approach.…”

Separation of the Baltic Sea water level into daily and multi-weekly components

“…Although extensive and rapid local changes in the water level are possible, strong spatial correlation of water level recordings extends to 150-200 nautical miles, that is, over the entire Gulf of Finland. The typical de-correlation time is about two weeks (Raudsepp et al, 1999). The autocorrelation function vanishes for much longer lags of 60-90 days, signalling the importance of seasonal-scale processes.…”

“…The external impacts led to the total (annual mean) sea level rise in the range of 7.5-15.3 cm at different locations in 1950-2002. The strongest periodic signal in the water level time series is the annual variation in sea level (Johansson et al, 2001). It ranges between 20 and 25 cm in the Gulf of Finland (Raudsepp et al, 1999) and has increased by about 5 cm in Pärnu and Narva-Jõesuu (Suursaar et al, 2006a(Suursaar et al, , 2015. This variation is usually less than 10% of the total range of water level variations (Suursaar et al, 2006a) and about 50% of the typical (annual) standard deviation of the instantaneous water level recordings (Johansson et al, 2001;Suursaar et al, 2006a).…”

“…The above-mentioned sequences of storms (Post and Kõuts, 2014) may add up to 1 m to the level of the entire sea over a typical time scale of a few weeks (Feistel et al, 2008;Leppäranta and Myrberg, 2009). Therefore, even if various harmonic components have been singled out, as has been done in common models of water level (e.g., Raudsepp et al, 1999), the residual signal is a mixture of reactions of water level to at least two strong drivers (single storms and subtidal variations) with different temporal scales and contains a substantial aperiodic component.…”

An Application of Low-Order Arma and Garch Models for Sea Level Fluctuations

“…They included analyses of the global mean sea level time series (Niedzielski and Kosek, 2005;Iz, 2006) as well as regional site-specific data (Raudsepp et al, 1999;Barbosa et al, 2006Barbosa et al, , 2008Niedzielski and Kosek, 2009). However, to date none of these exercises considers recommendations on a model structure which can be inferred from non-gaussian manner of variability.…”