Predicting and measuring waves, long waves and winds

Twitter
Facebook
LinkedIn
Email

Authorship

Peter J. McComb and David L. Johnson, MetOcean Solutions Ltd, New Zealand

Publication

A high-resolution weather forecasting tool for marine operations management in ports and harbors

Introduction

Many ports and harbors experience occasions when energetic weather conditions lead to operational issues. These may relate to underkeel clearance for the entrance or exit transit, agitation at the berth due to swell penetration or longwave surge and high winds influencing vessel management for example. Other activities, such as dredging, survey or construction typically require low energy conditions for safe execution. Effective planning for these activities requires reliable quantification of site specific weather parameters over short and medium (five to seven days) forecast range.

To meet this need, a set of web-based tools (MOVs – MetOceanView) has been developed to allow access and management of harbor-scale weather.

The core forecast system architecture is founded on modern high-resolution ocean and atmospheric models, customized at appropriate scales for the actual port or harbor location as well as the shipping approaches. The numerical models allow the detailed transformation of waves from offshore and into the entrance regions, as well as the coastal currents and water levels. The wind fields influenced by the regional topography can also be resolved. Ingestion of real-time measured wave and wind data into MOV allows the forecast accuracy and trends to be co-plotted and therefore assessed on a daily basis. The core forecast configurations also assimilate observed data where possible to improve short range forecasts. Examples of the forecast system are presented here for four operational situations. Port Taranaki and Port Geraldton lie within complex reef and near-shore bathymetry, and experience high energy swell conditions and occasions with problematic longwaves. Das Island in the Persian Gulf provides a lower energy example for the management of dredging operations, while the Port of Los Angeles application shows the benefit of spatial overlay maps of the local marine conditions.

Forecast system overview

The operational forecast system is configured at global, regional and local scales; allowing a truly customized solution to be established for any location. The high-resolution domains can be rapidly deployed, and a typical port system can be established within two to three days of notification. Full access to the forecast data is provided by an innovative set of web-based tools, and through delivery by email, as discussed in subsequent sections.

Global scale model

An in-house global WaveWatch3 (WW3) domain is run four times per day at 0.5o resolution using wind fields from NCEP’s Global Forecast System (GFS). Remote-sensed wave data are assimilated into each cycle. Forecasts produced with a sevenday horizon. As part of the WW3 implementation, full spectral boundaries are also generated for the nested regional domains.

Regional and local scale models

A series of multi-nested SWAN [2] wave model domains provide detailed resolution at the necessary regional and local scales. For a port location, a three-stage nest from the global domain is typically prescribed, although an additional level may be required in regions with highly complex bathymetric settings. An example of the nested wave domains used for the operational Port Taranaki implementation is shown in Figure 1.

Wind fields within these nested domains are sourced from either the NOAA GFS or from customized WRF atmospheric model domains. Local validation is initially performed to determine if increased resolution over the GFS scale is warranted. A 6–8km WRF domain is often beneficial in coastal regions with significant topographic influence, such as New Zealand waters or the Bass Strait (see Figure 2).

Effective marine forecasting requires data sources that suit the local topographic complexity and wave generation characteristics. Regional current models are run in some domains, with resolution of between 1km and 5km. A 2D implementation of POM [5] is used for current modeling, with wind velocity components and atmospheric pressure interpolated in space and time onto the model grid. The TPXO7.2 global inverse tidal solution [1] is used to prescribe the tidal elevation and current velocity boundaries.

Any number of reporting sites can be set within the domain, producing hourly site-specific data over the seven-day forecast horizon.

 

To read the full article download PDF

Cookie Policy. This website uses cookies to ensure you get the best experience on our website.