Search Results

You are looking at 1 - 2 of 2 items for

  • Author or Editor: Ricardo Martins Campos x
  • Refine by Access: Content accessible to me x
Clear All Modify Search
Ricardo Martins Campos
,
Vladimir Krasnopolsky
,
Jose-Henrique G. M. Alves
, and
Stephen G. Penny

Abstract

Artificial neural networks (ANNs) applied to nonlinear wave ensemble averaging are developed and studied for Gulf of Mexico simulations. It is an approach that expands the conservative arithmetic ensemble mean (EM) from the NCEP Global Wave Ensemble Forecast System (GWES) to a nonlinear mapping that better captures the differences among the ensemble members and reduces the systematic and scatter errors of the forecasts. The ANNs have the 20 members of the GWES as input, and outputs are trained using observations from six buoys. The variables selected for the study are the 10-m wind speed (U10), significant wave height (Hs), and peak period (Tp) for the year of 2016. ANNs were built with one hidden layer using a hyperbolic tangent basis function. Several architectures with 12 different combinations of neurons, eight different filtering windows (time domain), and 100 seeds for the random initialization were studied and constructed for specific forecast days from 0 to 10. The results show that a small number of neurons are sufficient to reduce the bias, while 35–50 neurons produce the greatest reduction in both the scatter and systematic errors. The main advantage of the methodology using ANNs is not on short-range forecasts but at longer forecast ranges beyond 4 days. The nonlinear ensemble averaging using ANNs was able to improve the correlation coefficient on forecast day 10 from 0.39 to 0.61 for U10, from 0.50 to 0.76 for Hs, and from 0.38 to 0.63 for Tp, representing a gain of five forecast days when compared to the EM currently implemented.

Full access
Ricardo Martins Campos
,
Jose-Henrique G. M. Alves
,
Stephen G. Penny
, and
Vladimir Krasnopolsky

Abstract

The error characteristics of surface waves and winds produced by ensemble forecasts issued by the National Centers for Environmental Prediction are analyzed as a function of forecast range and severity. Eight error metrics are compared, separating the scatter component of the error from the systematic bias. Ensemble forecasts of extreme winds and extreme waves are compared to deterministic forecasts for long lead times, up to 10 days. A total of 29 metocean buoys is used to assess 1 year of forecasts (2016). The Global Wave Ensemble Forecast System (GWES) performs 10-day forecasts four times per day, with a spatial resolution of 0.5° and a temporal resolution of 3 h, using a 20-member ensemble plus a control member (deterministic) forecast. The largest errors in GWES, beyond forecast day 3, are found to be associated with winds above 14 m s−1 and waves above 5 m. Extreme percentiles after the day-8 forecast reach 30% of underestimation for both 10-m-height wind (U10) and significant wave height (Hs). The comparison of probabilistic wave forecasts with deterministic runs shows an impressive improvement of predictability on the scatter component of the errors. The error for surface winds drops from 5 m s−1 in the deterministic runs, associated with extreme events at longer forecast ranges, to values around 3 m s−1 using the ensemble approach. As a result, GWES waves are better predicted, with a reduction in error from 2 m to less than 1.5 m for Hs. Nevertheless, under extreme conditions, critical systematic and scatter errors are identified beyond the day-6 and day-3 forecasts, respectively.

Full access