Description of the KNMI Operational Wave Forecast Model GONO

Description of the KNMI Operational Wave Forecast Model GONO

Bruinsma, J.
Janssen, P.A.E.M.
Komen, G.J.
Peeck, H.H.
Saraber, M.J.M.
De Voogt, W.J.P.

TU Delft

GONO

1980-05-02

The purpose of this report is to give a description of the GONO computercode, which is operational at KNMI for many years now. The program was developed by J.W. Sanders, and its deep water version is based on a Norwegian wave prediction model. built by C. Haug in the sixties. Shallow water effects are however important in the southern part of the North Sea. giving a limitation of the wave growth and causing important swell dissipation. A discussion of the shallow water effects. as present in GONO, is given by Sanders in Ref. 3. The computer code GONO is written in ALGOL60 and operational on the Burroughs 6700 of KNMI. The program GONO calculates wind speed. wind direction and sea energy at every grid point of the GONO grid (cf. Appendix A) and swell energy at a limited number of points only (we call these points swell points). GONO runs every six hours and it gives a 12 and 24 hours forecast as well as results based on analyzed weather maps. In the winter of '79-'80 the output of GONO was compared with measurements and the operational wave model of Bracknell (United Kingdom). (The latter GONO version differs a little bit from the version here described). The preliminary results of this comparison are given in Ref. 4. A reasonable agreement of significant wave height and low-frequency energy, as given by GONO, with the observations was found. The present version of GONO, with small modifications compared to the previous version, is operational since the end of February 1980. Essentially, the GONO model is based on two steps. First. the sea energy at every grid point is determined. To this end advection of energy is treated by means of a finite difference scheme whereas the growth of the wave energy is calculated by means of an empirical growth curve (assuming that the wave spectrum has a fixed form). The second step is the calculation of the swell. Of course, in principle swell can be treated likewise, but then one has to store swell energy (and its direction) at every time, at every grid point for every frequency band. Also. this finite difference scheme is rather crude for swell propagation. whereas. because of stability reasons. there is an upper bound for the propagation speed (in the present case the upper bound is given by 13.87 m/s). If one is only interested in swell information at particular points (swell points), it is tempting to use a ray technique. The advantage of this technique is that it is very accurate. Swell is determined in this fashion in GONO.

GONO
wave forecast model
code description
computer code
wave model
wave forecasts
swell

TLJ200510
TLJ800400
TLJ300200

http://resolver.tudelft.nl/uuid:e529cb2a-0332-43eb-a8d5-d095b2016852

KNMI

W.R. 80-8

Hydraulic Engineering Reports

report

(c) 1980 KNMI