Data-driven modelling: some past experiences and new approaches

Solomatine, Dimitri; Ostfeld, Avi

doi:10.2166/hydro.2008.015

Cited by 558 publications

(266 citation statements)

References 88 publications

Supporting

Mentioning

259

Contrasting

Unclassified

Order By: Relevance

“…In effect, model structure itself then becomes a target for assimilation approaches. The learning systems component of such a framework has the potential to advance rapidly given developments in cyberinfrastructure and computational techniques for data analysis and hydro-informatics (Beck, 2010;Hanson, 2007;Porter et al, 2009;Solomatine and Ostfeld, 2008).…”

Section: Model-data Learningmentioning

confidence: 99%

Developing predictive insight into changing water systems: use-inspired hydrologic science for the Anthropocene

Thompson

Sivapalan

Harman

et al. 2013

Hydrol. Earth Syst. Sci.

130

View full text Add to dashboard Cite

Abstract.Globally, many different kinds of water resources management issues call for policy-and infrastructure-based responses. Yet responsible decision-making about water resources management raises a fundamental challenge for hydrologists: making predictions about water resources on decadal-to century-long timescales. Obtaining insight into hydrologic futures over 100 yr timescales forces researchers to address internal and exogenous changes in the properties of hydrologic systems. To do this, new hydrologic research must identify, describe and model feedbacks between water and other changing, coupled environmental subsystems. These models must be constrained to yield useful insights, despite the many likely sources of uncertainty in their predictions. Chief among these uncertainties are the impacts of the increasing role of human intervention in the global water cycle -a defining challenge for hydrology in the Anthropocene. Here we present a research agenda that proposes a suite of strategies to address these challenges from the perspectives of hydrologic science research. The research agenda focuses on the development of co-evolutionary hydrologic modeling to explore coupling across systems, and to address the implications of this coupling on the long-time behavior of the coupled systems. Three research directions support the development of these models: hydrologic reconstruction, comparative hydrology and model-data learning. These strategies focus on understanding hydrologic processes and feedbacks over long timescales, across many locations, and through strategic coupling of observational and model data in specific systems. We highlight the value of use-inspired and team-based science that is motivated by real-world hydrologic problems but targets improvements in fundamental understanding to support decision-making and management. Fully realizing the potential of this approach will ultimately require detailed integration of social science and physical science understanding of water systems, and is a priority for the developing field of sociohydrology.

show abstract

Section: Model-data Learningmentioning

confidence: 99%

Developing predictive insight into changing water systems: use-inspired hydrologic science for the Anthropocene

Thompson

Sivapalan

Harman

et al. 2013

Hydrol. Earth Syst. Sci.

130

View full text Add to dashboard Cite

show abstract

“…Machine learning and computational intelligence provides a wide variety of approaches for data-driven 12 model discovery [Solomatine & Ostfeld, 2008]. We performed initial experiments using decision trees 13 and random forests, but focused on genetic programming (GP) [Koza, 1989[Koza, , 1992[Koza, , 1994], which has been 14 successfully applied to a wide variety of fields (e.g., to generate new integrated circuits, antennas and 15 controllers in circuit design [Koza et.al, 2003…”

Section: Data Fusion With Genetic Programming 10 11mentioning

confidence: 99%

Testing the accuracy of low-cost data streams for determining single-person office occupancy and their use for energy reduction of building services

Newsham

Xue

Arsenault

et al. 2017

Energy and Buildings

View full text Add to dashboard Cite

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n'arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. Questions? Contact the NRC Publications Archive team atPublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information. NRC Publications Archive Archives des publications du CNRCThis publication could be one of several versions: author's original, accepted manuscript or the publisher's version. / La version de cette publication peut être l'une des suivantes : la version prépublication de l'auteur, la version acceptée du manuscrit ou la version de l'éditeur. NRC Publications Record / Notice d'Archives des publications de CNRC:http://nparc.cisti-icist.nrc-cnrc.gc.ca/eng/view/object/?id=8dd905f5-6b28-4f14-9119-6435285da6a7 http://nparc.cisti-icist.nrc-cnrc.gc.ca/fra/voir/objet/?id=8dd905f5-6b28-4f14-9119-6435285da6a7 We explored methods of detecting occupancy in single-person offices using data already collected by the 7 occupant's PC, or data from relatively cheap sensors added to the PC. We collected data at 15-second 8 intervals for up to 31 days in each of 28 offices. A combination of low/no cost sensors (webcam-based 9 motion detection, and keyboard and mouse activity) was much more accurate at detecting occupancy than 10 a commercial ceiling-based passive infrared (PIR) sensor, and provided overall daytime accuracy >90%, 11 with very low false negative rates. This enhanced detection performance would enable a reduction in the 12 timeout periods for building service curtailment on space vacancy. For example, lighting switch-off 13 timeout could be reduced from the current energy code standard of 20 minutes to less than 5 minutes, 14 increasing energy savings potential by 25-45%. We then deployed this system in a proof-of-concept 15 demonstration, using it to control lighting, heating, ventilation, and air conditioning (HVAC), and plug 16

show abstract

“…Shortridge et al [14] compared a machine learning algorithm with a statistical approach-multiple regression to simulate streamflow over five rivers of Ethiopia. On the machine learning side, support-vector machines [15,16], regression tree based approach [5,17], and ANN [18,19] were used for rainfall-runoff modeling.…”

Section: Introductionmentioning

confidence: 99%

Streamflow Simulation Using Bayesian Regression with Multivariate Linear Spline to Estimate Future Changes

Bhowmik

Seo

Sahoo³

2018

Water

View full text Add to dashboard Cite

Statistical models for hydrologic simulation are a common choice among researchers particularly when catchment information is limited. In this study, we adopt a new statistical approach, namely Bayesian regression with multivariate linear spline (BMLS) for long-term simulation of streamflow on a Hydroclimate Data Network (HCDN) site in the United States. The study aims to: (i) evaluate the performance of the BMLS model; (ii) compare the performance of climate model outputs as predictors in hydrologic simulation; and (iii) estimate the changes in streamflow caused by anthropogenic climate change which is defined as the projected change in precipitation and temperature under different greenhouse gas emission scenarios. Performance of the BMLS model is compared with climatology for the validation period. Results suggest that the BMLS model forced with observed monthly precipitation and average temperature exhibits information that is not presented in the climatology of the validation period. Later, we consider Coupled Model Intercomparison Project Phase 5 (CMIP5) historical and hindcast runs to simulate streamflow at the HCDN site. The study found that sea-surface temperature-initialized decadal hindcast runs are performing no better than 20th century historical runs regarding hydrologic simulation. Finally, the changes in mean and variability in streamflow at the HCDN site are estimated by forcing the model with CMIP5 future projections for the period 2000-2049.

show abstract

Data-driven modelling: some past experiences and new approaches

Cited by 558 publications

References 88 publications

Developing predictive insight into changing water systems: use-inspired hydrologic science for the Anthropocene

Developing predictive insight into changing water systems: use-inspired hydrologic science for the Anthropocene

Testing the accuracy of low-cost data streams for determining single-person office occupancy and their use for energy reduction of building services

Streamflow Simulation Using Bayesian Regression with Multivariate Linear Spline to Estimate Future Changes

Contact Info

Product

Resources

About