Journal cover Journal topic
Earth Surface Dynamics An interactive open-access journal of the European Geosciences Union

Journal metrics

  • IF value: 2.000 IF 2.000
  • IF 5-year<br/> value: 2.000 IF 5-year
  • CiteScore<br/> value: 1.84 CiteScore
  • SNIP value: 0.628 SNIP 0.628
  • SJR value: indexed SJR
  • IPP value: 1.689 IPP 1.689
  • h5-index value: 6 h5-index 6
Earth Surf. Dynam., 2, 67-82, 2014
© Author(s) 2014. This work is distributed
under the Creative Commons Attribution 3.0 License.
Research article
28 Jan 2014
Data-driven components in a model of inner-shelf sorted bedforms: a new hybrid model
E. B. Goldstein1, G. Coco2, A. B. Murray1, and M. O. Green3 1Division of Earth and Ocean Sciences, Nicholas School of the Environment, Center for Nonlinear and Complex Systems, Duke University, P.O. Box 90227, Durham, NC 27708, USA
2Environmental Hydraulics Institute, "IH Cantabria", c/Isabel Torres no. 15, Universidad de Cantabria, 39011 Santander, Spain
3National Institute of Water and Atmospheric Research (NIWA), P.O. Box 11-115, Hamilton, New Zealand
Abstract. Numerical models rely on the parameterization of processes that often lack a deterministic description. In this contribution we demonstrate the applicability of using machine learning, a class of optimization tools from the discipline of computer science, to develop parameterizations when extensive data sets exist. We develop a new predictor for near-bed suspended sediment reference concentration under unbroken waves using genetic programming, a machine learning technique. We demonstrate that this newly developed parameterization performs as well or better than existing empirical predictors, depending on the chosen error metric. We add this new predictor into an established model for inner-shelf sorted bedforms. Additionally we incorporate a previously reported machine-learning-derived predictor for oscillatory flow ripples into the sorted bedform model. This new "hybrid" sorted bedform model, whereby machine learning components are integrated into a numerical model, demonstrates a method of incorporating observational data (filtered through a machine learning algorithm) directly into a numerical model. Results suggest that the new hybrid model is able to capture dynamics previously absent from the model – specifically, two observed pattern modes of sorted bedforms. Lastly we discuss the challenge of integrating data-driven components into morphodynamic models and the future of hybrid modeling.

Citation: Goldstein, E. B., Coco, G., Murray, A. B., and Green, M. O.: Data-driven components in a model of inner-shelf sorted bedforms: a new hybrid model, Earth Surf. Dynam., 2, 67-82, doi:10.5194/esurf-2-67-2014, 2014.
Publications Copernicus