Ashik and Novakow: Numerical Modelling of Storm Surges in the Laptev Sea
23
influence of the ice cover on fonning level evaluations was not taken into account.
Let us consider the results for level oscillations in the southern part of the Laptev Sea during
surges in September 1989 and August 1992.
A surge that occurred in September 1989 was caused by a powerful cyclone with core of
gales. The genesis area of the cyclone was in the northern Atlantic from where it moved
eastwards through the northern areas of the Barents, Kara and Laptev seas. The atmospheric
pressure in the centre of the cyclone was estimated to be below 970 Pa. The edge of the drifting
ice was situated near 7SoN, so, the south-eastern part of the Laptev Sea was free of ice. An
elevation of more than 2m above mean sea-level was observed at Nayba.
In Figure 2 we give the time variation of the calculated and observed sea-surface elevations at
the Terpiay-Tumsa Cape, Tiksy Bay and Nayba. The average absolute error of the calculation is
IS-20cm, RMS error is 20-2Scm, the correlation factor is 0,83.
After a maximum level (\ -\ ,S meter) a significant minimum (approximately 2 meters below
mean sea-level) was observed in August 1992 along the southern coast of the Laptev Sea. This
situation was caused by an intense baric evolution in that area. On August 8th a cyclone was
located in the Laptev Sea region, on August 10th it was substituted by an anticyclone. By that
time fast ice in the Laptev Sea was broken, the edge of the drifting ice was situated between the
Peschany Cape and Shirokostan peninsula, so, the southern part of the sea was mainly free of
ice.
The results of the calculation are shown in Figure 3. The average absolute error of this
calculation is 20-30cm, RMS error is 2S-4Scm, the correlation factor - 0,8S.
Discussion
Finite element hydrodynamic models using elements with quadratic basis functions are
considered to be of less numerical efficiency than finite-difference ones. However the use of
modern iteration methods for solution of sparse matrix equations reduce this difference of
efficiency to a minimum.
In spite of the poor spatial resolution of the model in this paper the results are encouraging
and the coastal line approximation is reliable. We believe that such small flexible models
describing coastal zones can become a useful part of compound hydrodynamic models.
Conclusion
A numerical model for solving the shallow water equations, using the finite element method in
the horizontal space domain was presented in this paper. The basis functions of the second
order were used. The model has been applied to the calculation of the surges that occurred in
September 1989 and August 1992 in the Laptev Sea. Computed level oscillations were in
agreement with observations.
References
Ashik, I.M., A.Yu. Proshutinskiy and V.A. Stepanov (1989) Some results and outlooks of the numeric
forecasts of sea level oscillations in the Arctic Seas (in Russian). Meteorology and Hydrology, 8, 74-82
Gudkovich, Z.M. and A.Yu. Proshutinskiy (1988) Modelling of oscillations of the level in the ice covered seas
(in Russian). Proc AARI, v. 413,85-95
Connor, IJ. and C.A. Brebbia (1979) Finite element techniques for fluid flow (in Russian). Sudostroenie,
Leningrad, 264 p.
Peyret, R. and T.O. Taylor (1986) Computational methods for fluid flow (in Russian). Gidrometizdat,
Leningrad, 352 p.
23
influence of the ice cover on fonning level evaluations was not taken into account.
Let us consider the results for level oscillations in the southern part of the Laptev Sea during
surges in September 1989 and August 1992.
A surge that occurred in September 1989 was caused by a powerful cyclone with core of
gales. The genesis area of the cyclone was in the northern Atlantic from where it moved
eastwards through the northern areas of the Barents, Kara and Laptev seas. The atmospheric
pressure in the centre of the cyclone was estimated to be below 970 Pa. The edge of the drifting
ice was situated near 7SoN, so, the south-eastern part of the Laptev Sea was free of ice. An
elevation of more than 2m above mean sea-level was observed at Nayba.
In Figure 2 we give the time variation of the calculated and observed sea-surface elevations at
the Terpiay-Tumsa Cape, Tiksy Bay and Nayba. The average absolute error of the calculation is
IS-20cm, RMS error is 20-2Scm, the correlation factor is 0,83.
After a maximum level (\ -\ ,S meter) a significant minimum (approximately 2 meters below
mean sea-level) was observed in August 1992 along the southern coast of the Laptev Sea. This
situation was caused by an intense baric evolution in that area. On August 8th a cyclone was
located in the Laptev Sea region, on August 10th it was substituted by an anticyclone. By that
time fast ice in the Laptev Sea was broken, the edge of the drifting ice was situated between the
Peschany Cape and Shirokostan peninsula, so, the southern part of the sea was mainly free of
ice.
The results of the calculation are shown in Figure 3. The average absolute error of this
calculation is 20-30cm, RMS error is 2S-4Scm, the correlation factor - 0,8S.
Discussion
Finite element hydrodynamic models using elements with quadratic basis functions are
considered to be of less numerical efficiency than finite-difference ones. However the use of
modern iteration methods for solution of sparse matrix equations reduce this difference of
efficiency to a minimum.
In spite of the poor spatial resolution of the model in this paper the results are encouraging
and the coastal line approximation is reliable. We believe that such small flexible models
describing coastal zones can become a useful part of compound hydrodynamic models.
Conclusion
A numerical model for solving the shallow water equations, using the finite element method in
the horizontal space domain was presented in this paper. The basis functions of the second
order were used. The model has been applied to the calculation of the surges that occurred in
September 1989 and August 1992 in the Laptev Sea. Computed level oscillations were in
agreement with observations.
References
Ashik, I.M., A.Yu. Proshutinskiy and V.A. Stepanov (1989) Some results and outlooks of the numeric
forecasts of sea level oscillations in the Arctic Seas (in Russian). Meteorology and Hydrology, 8, 74-82
Gudkovich, Z.M. and A.Yu. Proshutinskiy (1988) Modelling of oscillations of the level in the ice covered seas
(in Russian). Proc AARI, v. 413,85-95
Connor, IJ. and C.A. Brebbia (1979) Finite element techniques for fluid flow (in Russian). Sudostroenie,
Leningrad, 264 p.
Peyret, R. and T.O. Taylor (1986) Computational methods for fluid flow (in Russian). Gidrometizdat,
Leningrad, 352 p.