Improved neural network scatterometer forward models

Dan Cornford, Ian T. Nabney, Guillaume Ramage

Research output: Contribution to journalArticlepeer-review


Current methods for retrieving near-surface winds from scatterometer observations over the ocean surface require a forward sensor model which maps the wind vector to the measured backscatter. This paper develops a hybrid neural network forward model, which retains the physical understanding embodied in CMOD4, but incorporates greater flexibility, allowing a better fit to the observations. By introducing a separate model for the midbeam and using a common model for the fore and aft beams, we show a significant improvement in local wind vector retrieval. The hybrid model also fits the scatterometer observations more closely. The model is trained in a Bayesian framework, accounting for the noise on the wind vector inputs. We show that adding more high wind speed observations in the training set improves wind vector retrieval at high wind speeds without compromising performance at medium or low wind speeds. Copyright 2001 by the American Geophysical Union.

Original languageEnglish
Pages (from-to)22331-22338
Number of pages8
JournalJournal of Geophysical Research
Issue numberC10
Publication statusPublished - 15 Oct 2001

Bibliographical note

Not subject to U.S. copyright. An edited version of this paper was published by AGU. Copyright 2001 American Geophysical Union. Cornford, Dan; Nabney, Ian T. and Ramage, Guillaume, (2001), Improved neural network scatterometer forward models, Journal of Geophysical Research - Oceans, 106, C10, 10.1029/2000JC000417. To view the published open abstract, go to


  • near‐surface winds
  • scatterometer observations
  • ocean surface
  • forward sensor model
  • mapping
  • wind vector
  • measured backscatter
  • hybrid neural network forward model
  • local wind vector retrieval
  • Bayesian framework
  • wind vector inputs


Dive into the research topics of 'Improved neural network scatterometer forward models'. Together they form a unique fingerprint.

Cite this