Wave-induced responses are normally one of the main criteria for planning of marine operations. Estimates of response are usually based on forecasted values of significant wave height Hs and mean period Tz. Standard spectral distributions such as Pierson-Moskowitch or Jonswap types are applied. Weather forecasts of height and period of total sea and swell are therefore critical parameters for marine operations. In this work we discuss an alternative approach: evaluating the response using directional wave spectra given from a numerical wave forecast model. All information in the sea state will then be fully included. Responses estimated by including full wave spectral information from the wave forecast model WAM run at DNMI (Norwegian Meteorological Institute) are used as "base case". Alternative response estimates using the standard text based weather forecast, denoted "rig forecast'', are compared to the base case. The heave, pitch and roll responses are estimated for a semi-submersible, mono-hull and twin-hull. The uncertainties in the transfer functions are neglected. A case study is used to evaluate improvements achieved by such a response forecast. The selected wave situation clearly demonstrates the limitations of on-board response estimations based on the "rig forecast". Different methods give a wide spread on the results. Compared to wave model spectra, text based forecasts often simplify a sea state description and introduce the "human factor" in the evaluation of response.
Severe weather- and sea state conditions put major constraints on offshore operations, especially during the winter season. "Severe conditions" are both site and operation specific. Limitations are mostly related to wind and wave criteria. For floating systems, criteria are often related to long swell which induces response (heave for example) on the construction. In most cases these environmental conditions may be forecasted days ahead.