Show simple item record

dc.contributor.authorSsegane, Herberten_US
dc.contributor.authorTollner, E. W.en_US
dc.contributor.authorMohamoud, Yusufen_US
dc.contributor.authorRasmussen, Todd C.en_US
dc.contributor.authorDowd, John F.en_US
dc.contributor.editorCarroll, G. Deniseen_US
dc.date.accessioned2013-02-22T20:26:26Z
dc.date.available2013-02-22T20:26:26Z
dc.date.issued2011-04
dc.identifier.isbn0-9794100-24
dc.identifier.urihttp://hdl.handle.net/1853/46233
dc.descriptionProceedings of the 2011 Georgia Water Resources Conference, April 11, 12, and 13, 2011, Athens, Georgia.en_US
dc.description.abstractThe study explored use of causal feature selection algorithms to select dominant watershed variables that drive high, medium, and low flows. A two step approach was implemented. The first step minimized variable redundancy by examining variable relevance, variable redundancy, and conditional relevance of variable pairs whose correlation was greater than 0.9. The second step used six algorithms that seek to reconstruct a Bayesian network structure around a target variable for each flow percentile. Nineteen (19) flow percentiles were used to characterize high, medium, and low flow conditions of 26 Piedmont watersheds in the Mid-Atlantic. The algorithms included: (1) Grow-Shrink (GS); (2) interleaved-Incremental Association Markov Boundary (interIAMB) (3) Incremental Association Markov Boundary with Peter-Clark (IAMBnPC); (4) Local Causal Discovery (LCD2); (5) HITON-PC; and (6) HITON-MB. A new method was developed to quantify the reliability of each algorithm and its performance was compared to existing reliability methods. The effect of the initial number of variables on the final variable set selected by each algorithm was tested. Fusion of the algorithms was used to determine the overall dominant features for each flow percentile.en_US
dc.description.sponsorshipSponsored by: Georgia Environmental Protection Division U.S. Geological Survey, Georgia Water Science Center U.S. Department of Agriculture, Natural Resources Conservation Service Georgia Institute of Technology, Georgia Water Resources Institute The University of Georgia, Water Resources Facultyen_US
dc.description.statementofresponsibilityThis book was published by Warnell School of Forestry and Natural Resources, The University of Georgia, Athens, Georgia 30602-2152. The views and statements advanced in this publication are solely those of the authors and do not represent official views or policies of The University of Georgia, the U.S. Geological Survey, the Georgia Water Research Institute as authorized by the Water Research Institutes Authorization Act of 1990 (P.L. 101-307) or the other conference sponsors.en_US
dc.language.isoen_USen_US
dc.publisherGeorgia Institute of Technologyen_US
dc.relation.ispartofseriesGWRI2011. Environmental Protectionen_US
dc.subjectWater resources managementen_US
dc.subjectCausal feature selection algorithmsen_US
dc.subjectWatershed variablesen_US
dc.subjectFlow percentilesen_US
dc.titleBeyond Correlation: the Search for Causal Relationships Between Flow Percentiles and Watershed Variablesen_US
dc.typeProceedingsen_US
dc.contributor.corporatenameUniversity of Georgia. Dept. of Biological and Agricultural Engineeringen_US
dc.contributor.corporatenameUnited States. Environmental Protection Agencyen_US
dc.contributor.corporatenameUniversity of Georgia. Dept. of Geologyen_US
dc.contributor.corporatenameDaniel B. Warnell School of Forestry and Natural Resourcesen_US
dc.contributor.corporatenameNational Exposure Research Laboratory (U.S.)en_US
dc.publisher.originalWarnell School of Forestry and Natural Resources, The University of Georgiaen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record