Modelling algal abnormal proliferation in a reservoir using support vector regression: a case study
- 30 September 2014
- journal article
- research article
- Published by Wiley in Ecohydrology
- Vol. 8 (6) , 1109-1118
Abstract
An important issue in lake and reservoir management is to prevent algal proliferation and its problematic symptom: the algal blooms that deteriorate water quality and ruin its use. The aim of this study was to obtain a predictive model able to perform an early detection of algal abnormal proliferation using as a predictor the chlorophyll a (Chl a) concentration of previous days. For this purpose, some biological parameters (phytoplankton species expressed in biovolume) in addition to the most important chemical and physical–chemical parameters were considered. The implementation of the statistical learning method known as support vector machines (SVMs) allowed us to predict the Chl a concentration, one of the most frequently used estimators of algal biomass, and to study its evolution in the Trasona reservoir (Principality of Asturias, Northern Spain). In the first place, the main goal was to obtain the dependence relationship of chlorophyll concentration as a function of not only several chemical and physical–chemical variables but also the biological ones. This dependence relationship was useful in setting the significance order of the variables involved in the Chl a prediction. Secondly, a model for forecasting the algal abnormal proliferation is obtained. The agreement between experimental data and the model confirmed the good performance of the latter. Finally, conclusions of this innovative research work are exposed. Copyright © 2014 John Wiley & Sons, Ltd.Keywords
This publication has 22 references indexed in Scilit:
- Application of an SVM-based regression model to the air quality study at local scale in the Avilés urban area (Spain)Mathematical and Computer Modelling, 2011
- Eutrophication of lakes cannot be controlled by reducing nitrogen input: Results of a 37-year whole-ecosystem experimentProceedings of the National Academy of Sciences, 2008
- Predicting motor vehicle crashes using Support Vector Machine modelsAccident Analysis & Prevention, 2008
- Machine learning techniques applied to the determination of road suitability for the transportation of dangerous substancesJournal of Hazardous Materials, 2007
- Integration of data mining techniques and heuristic knowledge in fuzzy logic modelling of eutrophication in Taihu LakeEcological Modelling, 2003
- Neural network modelling of coastal algal bloomsEcological Modelling, 2003
- Case studies on the use of neural networks in eutrophication modelingEcological Modelling, 2000
- Predictive Model of the Effects on Lake Metabolism of Decreased Airborne Litterfall through Riparian DeforestationConservation Biology, 1995
- Support-vector networksMachine Learning, 1995
- A trophic state index for lakes1Limnology and Oceanography, 1977