Wave Height Estimation Using Disaggregation Models

Abstract

Wave height data are sometimes available at different space locations in the form of average values over larger time intervals. Difficulties may arise when such data are to be used for applications such as small-term operational forecasting for which information at closer intervals of time would be necessary. This paper presents an effort made to deduce wave heights over smaller time periods from their values sampled over longer durations, with the help of the disaggregation models traditionally used in hydrology. Use has been made of basic, extended as well as condensed disaggregation models. Considering the reliability and accuracy of wave rider buoy data, the same has also been employed along with satellite data for comparison purposes and for conducting some additional investigations. It has been found that statistically advantageous disaggregation mod­ els produce more satisfactory estimates of small-interval waves and that the smaller the level difference between high-level, large-interval series and low-level, small-interval ones, the larger the accuracy of resulting estimates. ABSTRACT: Wave height data are sometimes available at different space locations in the form of average values over larger time intervals. Difficulties may arise when such data are to be used for applications such as small-term operational forecasting for which information at closer intervals of time would be necessary. This paper presents an effort made to deduce wave heights over smaller time periods from their values sampled over longer durations, with the help of the disaggregation models traditionally used in hydrology. Use has been made of basic, extended as well as condensed disaggregation models. Considering the reliability and accuracy of wave rider buoy data, the same has also been employed along with satellite data for comparison purposes and for conducting some additional investigations. It has been found that statistically advantageous disaggregation mod­ els produce more satisfactory estimates of small-interval waves and that the smaller the level difference between high-level, large-interval series and low-level, small-interval ones, the larger the accuracy of resulting estimates. Information regarding heights of ocean waves at a site can be collected by deploying a variety of instruments kept under submerged, floating, or air-borne conditions. Each instrument, such as echo sounder, wave rider buoy, and radar altimeter would involve a different method of data retrieval and synthe­ sis. Typically, sensing of wave heights by a radar altimeter mounted overboard a satellite is such that wave information can be more conveniently presented in the form of space and time averages over a wide number of space and time intervals. Such information may appear attractive since it is usually available at nominal cost and, further, when more accurate instruments like a wave rider buoy cannot be deployed in view of large water depths, longer distances offshore, high costs, and time requirements. However, difficulties may arise when such data are to be used for applications like offshore opera­ tional forecasting wherein wave information over closer time intervals would become necessary. Examples of the offshore operations are towing and installation of offshore jackets, pipe laying, and routine navigation where the works may span over say one or more weeks.