Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework

In this paper the difficult problem of how to legitimisedata-driven hydrological models is addressed using an example of a simple artificial neural network modelling problem. Many data-driven models in hydrology have been criticised for their black-box characteristics, which prohibit adequate unders...

Full description

Bibliographic Details
Main Authors: Mount, Nick J., Dawson, C.W., Abrahart, R.J.
Format: Article
Published: European Geosciences Union (EGU) 2013
Subjects:
Online Access:https://eprints.nottingham.ac.uk/28052/
_version_ 1848793495830003712
author Mount, Nick J.
Dawson, C.W.
Abrahart, R.J.
author_facet Mount, Nick J.
Dawson, C.W.
Abrahart, R.J.
author_sort Mount, Nick J.
building Nottingham Research Data Repository
collection Online Access
description In this paper the difficult problem of how to legitimisedata-driven hydrological models is addressed using an example of a simple artificial neural network modelling problem. Many data-driven models in hydrology have been criticised for their black-box characteristics, which prohibit adequate understanding of their mechanistic behaviour and restrict their wider heuristic value. In response, presented here is a new generic data-driven mechanistic modelling framework. The framework is significant because it incorporates an evaluation of the legitimacy of a data-driven model’s internal modelling mechanism as a core element in the modelling process. The framework’s value is demonstrated by two simple artificial neural network river forecasting scenarios. We develop a novel adaptation of first-order partial derivative, relative sensitivity analysis to enable each model’s mechanistic legitimacy to be evaluated within the framework. The results demonstrate the limitations of standard, goodness-of-fit validation procedures by highlighting how the internal mechanisms of complex models that produce the best fit scores can have lower mechanistic legitimacy than simpler counterparts whose scores are only slightly inferior. Thus, our study directly tackles one of the key debates in data-driven, hydrological modelling: is it acceptable for our ends (i.e. model fit) to justify our means (i.e. the numerical basis by which that fit is achieved)?
first_indexed 2025-11-14T19:01:13Z
format Article
id nottingham-28052
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:01:13Z
publishDate 2013
publisher European Geosciences Union (EGU)
recordtype eprints
repository_type Digital Repository
spelling nottingham-280522020-05-04T16:35:34Z https://eprints.nottingham.ac.uk/28052/ Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework Mount, Nick J. Dawson, C.W. Abrahart, R.J. In this paper the difficult problem of how to legitimisedata-driven hydrological models is addressed using an example of a simple artificial neural network modelling problem. Many data-driven models in hydrology have been criticised for their black-box characteristics, which prohibit adequate understanding of their mechanistic behaviour and restrict their wider heuristic value. In response, presented here is a new generic data-driven mechanistic modelling framework. The framework is significant because it incorporates an evaluation of the legitimacy of a data-driven model’s internal modelling mechanism as a core element in the modelling process. The framework’s value is demonstrated by two simple artificial neural network river forecasting scenarios. We develop a novel adaptation of first-order partial derivative, relative sensitivity analysis to enable each model’s mechanistic legitimacy to be evaluated within the framework. The results demonstrate the limitations of standard, goodness-of-fit validation procedures by highlighting how the internal mechanisms of complex models that produce the best fit scores can have lower mechanistic legitimacy than simpler counterparts whose scores are only slightly inferior. Thus, our study directly tackles one of the key debates in data-driven, hydrological modelling: is it acceptable for our ends (i.e. model fit) to justify our means (i.e. the numerical basis by which that fit is achieved)? European Geosciences Union (EGU) 2013-01-09 Article PeerReviewed Mount, Nick J., Dawson, C.W. and Abrahart, R.J. (2013) Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework. Hydrology and Earth System Sciences, 17 . pp. 2827-2843. ISSN 1027-5606 data-driven models http://www.hydrol-earth-syst-sci.net/17/2827/2013/hess-17-2827-2013.html doi:10.5194/hess-17-2827-2013 doi:10.5194/hess-17-2827-2013
spellingShingle data-driven models
Mount, Nick J.
Dawson, C.W.
Abrahart, R.J.
Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title_full Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title_fullStr Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title_full_unstemmed Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title_short Legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
title_sort legitimising data-driven models: exemplification of a new data-driven mechanistic modelling framework
topic data-driven models
url https://eprints.nottingham.ac.uk/28052/
https://eprints.nottingham.ac.uk/28052/
https://eprints.nottingham.ac.uk/28052/