PRESENTS: FORECASTING FOR OPERATIONS AND DESIGN February 16 th 2011 – Aberdeen

PRESENTS:

FORECASTING FOR OPERATIONS AND DESIGN

February 16th 2011 – Aberdeen

Marine Weather Forecasts

What tools do we use and how do we prepare them?

The meteorological world has moved on a long way .....

Nowadays marine forecasts are prepared using output from NWP – Numerical Weather Prediction

Global Atmospheric Model

Ensemble Forecasting

Spectral wave model

Global Atmospheric Model

Solves fundamental equations describing the evolution of the atmosphere

Fundamental to the success of a forecast is an accurate knowledge of the initial state of the atmosphere

Short-range forecast - 'first guess’

Observations of actual and derived conditions

The model is run forward in time, typically up to a week ahead

During the run, a range of forecast information is produced at set intervals – say 1 or 3 hours

e.g. temperature, humidity, pressure, winds

Ensemble Forecasting

Model does not fully describe processes in the atmosphere

Model resolution insufficient to capture all atmospheric features

Initial observations are not available at every point in the atmosphere

Observational data cannot be measured precisely

Ensemble Forecasting

One method of quantifying the forecast uncertainty is by running the model a number of times

Each run has slightly different starting conditions

Each run evolves differently with time

Ensemble Forecasting – types of output

Spectral wave model

Solves the spectral energy density equation

Uses input from the Global Atmospheric Model, e.g. winds, sea and air temperature

Spectral wave model

The wave model predicts the energy spectrum at each point at each time step

The energy spectrum is processed to produce significant wave height, wave period, wind wave, swell etc

Preparing a Weather Forecast for Marine Operations

Start by examining raw forecast data of wind, waves etc

Compare starting values with actual observed data

Make allowances for model biases

Prepare the final forecast figures

Disseminate the forecast in required format – e.g. table/graph, operations checklist, 3-D plot etc.

Providers of Marine Weather Forecasts

Most oil companies will not have their own in-house forecasting service but will contract out the provision of weather services to a specialist provider.

Forecast company – either public or private – will prepare forecasts for a number of clients using their available model data

Forecasting companies will use output from one or more global forecast centres e.g. UK Met Office, NCEP, ECMWF

Many companies have their own in-house wave model capability

Extreme Wave Height Gather information on wave climate from observations of wave height over as long a period as possible

Choose a distribution function to fit the ‘upper tail’ (i.e. the highest waves) of the data

Log-normalWeibullFisher Tippett

Extrapolate to a probability of exceedance of 1 in 50 years

Extreme Wave Height (continued)

Calculate the highest likely wave corresponding to the maximum value of 50 year significant wave height

Combine with other sea-level parameters e.g. storm surge, astronomical tide

Other Extreme Metocean Parameters Wind

Current – Tidal, Surge, Residual

Air and Sea Temperature

Snow and Ice

Change of Extremes in the Future Very difficult to predict how the wave climate will respond in the future

Recent changes in North Atlantic wave height have been correlated with the phase of the North Atlantic Oscillation