The present study investigates the short-term statistical distributions of wave heights
and crest heights arising in intermediate and shallow water depths. These represent
key input parameters in a wide range of engineering applications. As such, their
accurate prediction is of great importance for the effective design and safety of marine
structures and vessels.
The results presented herein are based upon the analysis of a vast database of
field measurements, recorded by wave radars in the North Sea. These are supplemented
by an extensive laboratory investigation of directionally spread sea-states.
Strict quality control procedures were applied to the field measurements and a thorough
validation process was adopted for the experimental investigation. The latter
involved establishing good agreement between results obtained in facilities with very
different operational characteristics. Most importantly, close agreement was obtained
between laboratory and field data arising in similar sea-state conditions.
Considering the statistical distribution of wave heights, the best performing models
under a wide variety of met-ocean conditions are identified using the field data.
Additionally, the analysis of experimental results provided physical insights regarding
the effects of non-linearity, directionality, effective water depth and spectral bandwidth.
Using this physical understanding, a new wave height model is proposed,
providing an improvement over existing models.
With respect to the distribution of crest heights, the existing statistical models
are assessed using both field and experimental measurements. This analysis has provided
conclusive evidence of significant departures from the widely applied Forristall
(2000) distribution. These arise both as amplifications beyond the model’s predictions
in sufficiently steep sea-states and reductions in those cases with significant
wave breaking. These two competing mechanisms are further investigated with respect
to their dependence on the sea-state steepness, directional spreading, effective
water depth and spectral bandwidth. In incorporating these findings, a new crest
height model is formulated building upon the proposed wave height model. Finally,
the accuracy of both the proposed wave height and crest height models is demonstrated
through comparisons with field measurements under a wide range of incident
wave conditions and effective water depths.