A probabilistic approach to sea level rise up to the year 2100 at Kołobrzeg, Poland

Abstract

As an example of the linear forecast method, the application of a linear trend to annual mean sea level data from Kołobrzeg is analysed, and the mean sea level rise for a 110 yr time period is forecast. On the basis of the classical trend theory, the possibility of making a linear forecast on the basis of variable confidence intervals is demonstrated and various computational techniques are compared. The method assumes the application of confidence intervals not only to the stochastic component of the forecast model but also to the estimated trend parameters. As a result, depending on the assumptions and measurement data, the linear forecast method can be adapted to a variety of situations where low-lying areas are under threat from rising sea levels. The possibility of including the mean sea level rise forecast in the probabilistic forecast of storm surges is discussed. Example calculations for the period 1901 to 1990 were performed for a time series of maximum annual sea levels at Kołobrzeg, estimated by Gumbel's distribution.