ENSURF: multi-model sea level forecast – implementation and validation results for the IBIROOS and Western Mediterranean regions

Abstract

Abstract. ENSURF (Ensemble SURge Forecast) is a multimodel application for sea level forecast that makes use of several storm surge or circulation models and near-real time tide gauge data in the region, with the following main goals:1. providing easy access to existing forecasts, as well as to its performance and model validation, by means of an adequate visualization tool;2. generation of better forecasts of sea level, including confidence intervals, by means of the Bayesian Model Average technique (BMA).The Bayesian Model Average technique generates an overall forecast probability density function (PDF) by making a weighted average of the individual forecasts PDF's; the weights represent the Bayesian likelihood that a model will give the correct forecast and are continuously updated based on the performance of the models during a recent training period. This implies the technique needs the availability of sea level data from tide gauges in near-real time. The system was implemented for the European Atlantic facade (IBIROOS region) and Western Mediterranean coast based on the MA-TROOS visualization tool developed by Deltares. Results of validation of the different models and BMA implementation for the main harbours are presented for these regions where this kind of activity is performed for the first time. The system is currently operational at Puertos del Estado and has proved to be useful in the detection of calibration problems in some of the circulation models, in the identification of the systematic differences between baroclinic and barotropic models for sea level forecasts and to demonstrate the feasibility of providing an overall probabilistic forecast, based on the BMA method.

Developing European operational oceanography for Blue Growth, climate change adaptation and mitigation, and ecosystem-based management

“…Increasing use of the ensemble data assimilation method also provides a natural platform for making 3-D ocean ensemble forecasts. For the European seas, multi-model water-level prediction has been developed for European seas in ROOSes and in the ECOOP project, and used for national storm surge forecasts since the early 2000s (Perez et al, 2012). Further development of multi-model ocean forecasting systems has been an active part of MyOcean and CMEMS (Golbeck et al, 2015).…”

Theory and practice of phase‐aware ensemble forecasting

“…For example, one could compute a weighted phase‐aware mean, where the weights would be a function of how well the individual ensemble members are performing during a training period. Such a method would be similar in spirit to the traditionally known Bayesian model averaging (BMA) method in the time domain (Pérez et al, ). Future research is needed to understand how the phase‐aware mean statistics could be improved when combining known techniques.…”

Authors reply to Reviewer 2

“…For example, a stricter coding standard should be applied to ensure the run-to-run reproducibility. A more efficient coding principle such as PSyKAl (Parallel System, Kernel and Algorithm), taken in the GungHo Project which is developing a new dynamical core suitable for the weather and climate simulations, may benefit the UOM development; upgrading the code with the SIMD (single instruction multiple data, Poulsen et al, 2014) feature has proven the benefit for the model by using new vectorization and efficient hybrid threading for multi-core and many-core architectures.…”