File Download

There are no files associated with this item.

  • Find it @ UNIST can give you direct access to the published full text of this article. (UNISTARs only)
Related Researcher

조경화

Cho, Kyung Hwa
Water-Environmental Informatics Lab.
Read More

Views & Downloads

Detailed Information

Cited time in webofscience Cited time in scopus
Metadata Downloads

Stressor-response modeling using the 2D water quality model and regression trees to predict chlorophyll-a in a reservoir system

Author(s)
Park, YongeunPachepsky, Yakov A.Cho, Kyung HwaJeon, Dong JinKim, Joon Ha
Issued Date
2015-10
DOI
10.1016/j.jhydrol.2015.09.002
URI
https://scholarworks.unist.ac.kr/handle/201301/17934
Fulltext
http://www.sciencedirect.com/science/article/pii/S002216941500685X
Citation
JOURNAL OF HYDROLOGY, v.529, pp.805 - 815
Abstract
To control algal blooms, the stressor-response relationships between water quality metrics, environmental variables, and algal growth need to be better understood and modeled. Machine-learning methods have been suggested as means to express the stressor-response relationships that are found when applying mechanistic water quality models. The objective of this work was to evaluate the efficiency of regression trees in the development of a stressor-response model for chlorophyll-a (Chl-a) concentrations, using the results from site-specific mechanistic water quality modeling. The 2-dimensional hydrodynamic and water quality model (CE-QUAL-W2) model was applied to simulate water quality using four-year observational data and additional scenarios of air temperature increases for the Yeongsan Reservoir in South Korea. Regression tree modeling was applied to the results of these simulations. Given the well-expressed seasonality in the simulated Chl-a dynamics, separate regression trees were developed for months from May to September. The regression trees provided a reasonably accurate representation of the stressor-response dependence generated by the CE-QUAL-W2 model. Different stressors were then selected as split variables for different months, and, in most cases, splits by the same stressor variable yielded the same correlation sign between the variable and the Chl-a concentration. Compared to physical variables, nutrient content appeared to better predict Chl-a responses. The highest Chl-a temperature sensitivities were found for May and June. Regression tree splits based on ammonium concentration resulted in a consistent trend of greater sensitivity in the groups of samples with higher ammonium concentrations. Regression tree models provided a transparent visual representation of the stressor-response relationships for Chl-a and its sensitivity. Overall, the representation of relationships using classification and regression tools can be considered a useful approach to assess the state of aquatic ecosystems and effectively determine significant stressor variables. Published by Elsevier B.V
Publisher
ELSEVIER SCIENCE BV
ISSN
0022-1694
Keyword (Author)
Stressor-responseMachine learningTemperature sensitivityChlorophyll-a
Keyword
FRESH-WATERPHYTOPLANKTON GROWTHNUTRIENT LIMITATIONYEONGSAN RESERVOIRLAKE TAIHUPHOSPHORUSKOREAEUTROPHICATIONCYANOBACTERIANITROGEN

qrcode

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.