An integrated framework that combines machine learning and numerical models to improve wave-condition forecasts

Abstract This study investigates near-shore circulation and wave characteristics applied to a case-study site in Monterey Bay, California. We integrate physics-based models to resolve wave conditions together with a machine-learning algorithm that combines forecasts from multiple, independent models into a single “best-estimate” prediction of the true state. The Simulating WAves Nearshore (SWAN) physics-based model is used to compute wind-augmented waves. Ensembles are developed based on multiple simulations perturbing data input to the model. A learning-aggregation technique uses historical observations and model forecasts to calculate a weight for each ensemble member. We compare the weighted ensemble predictions with measured data to evaluate performance against present state-of-the-art. Finally, we discuss how this framework that integrates data-driven and physics-based approaches can outperform either technique in isolation.

[1]  Paul A. Wittmann,et al.  Evaluations of Global Wave Prediction at the Fleet Numerical Meteorology and Oceanography Center , 2005 .

[2]  Dick K. P. Yue,et al.  Theory and Applications of Ocean Surface Waves: Part 1: Linear AspectsPart 2: Nonlinear Aspects , 2005 .

[3]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[4]  Manfred K. Warmuth,et al.  Exponentiated Gradient Versus Gradient Descent for Linear Predictors , 1997, Inf. Comput..

[5]  R. Betts,et al.  Comparing projections of future changes in runoff from hydrological and biome models in ISI-MIP , 2013 .

[6]  Emanuele Ragnoli,et al.  Characterizing observed circulation patterns within a bay using HF radar and numerical model simulations , 2015 .

[7]  S. James,et al.  A machine learning framework to forecast wave conditions , 2017, Coastal Engineering.

[8]  Richard J. Beckman,et al.  A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code , 2000, Technometrics.

[9]  H. S. Chen,et al.  National Weather Service National Centers for Environmental Prediction Technical Procedures Bulletin Subject : A Variational Wave Height Data Assimilation System for NCEP Operational Wave Models , 2004 .

[10]  M. Donelan,et al.  Dynamics and Modelling of Ocean Waves , 1994 .

[11]  L H Holthuijsen,et al.  WAVE FORECASTING FOR MOVING AND STATIONARY TARGETS , 1988 .

[12]  Vivien Mallet,et al.  Ozone ensemble forecast with machine learning algorithms , 2009 .

[13]  J. Bidlot,et al.  User manual and system documentation of WAVEWATCH III R version 4.18 , 2014 .

[14]  Martin Verlaan,et al.  OpenDA-NEMO framework for ocean data assimilation , 2016, Ocean Dynamics.

[15]  S. Wilson,et al.  A simple method for integrating a complex model into an ensemble data assimilation system using MPI , 2015, Environ. Model. Softw..

[16]  J. Bidlot,et al.  Intercomparison of the Performance of Operational Ocean Wave Forecasting Systems with Buoy Data , 2002 .

[17]  Lars Nerger,et al.  PDAF - The Parallel Data Assimilation Framework: Experiences with Kalman Filtering , 2005 .

[18]  Jesse D. Roberts,et al.  Numerical modeling of the effects of wave energy converter characteristics on nearshore wave conditions , 2016 .

[19]  Jeffrey L. Anderson,et al.  The Data Assimilation Research Testbed: A Community Facility , 2009 .

[20]  L. Rosenfeld,et al.  Addressing ocean and coastal issues at the West Coast scale through regional ocean observing system collaboration , 2012, 2012 Oceans.

[21]  Scott C. James,et al.  An ensemble-based approach that combines machine learning and numerical models to improve forecasts of wave conditions , 2017, OCEANS 2017 – Anchorage.

[22]  Suraje Dessai,et al.  Ensembles and uncertainty in climate change impacts , 2014, Front. Environ. Sci..