1. Title, Journal and Authors
Title : Development of a Nowcasting System Using Machine Learning Approaches to Predict Fecal Contamination Levels at Recreational Beaches in Korea
Journal : Journal lf Environmental Quality
Authors : Yongeun Parka, Minjeong Kimb, Yakov Pachepskyc, Seoung-Hwa Choid, Jeong-Goo Chod, Junho Jeone, and Kyung Hwa Chob*
a Civil and Environmental Engineering, Konkuk Univ., Seoul 05029, Republic of Korea;
b School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology, Ulsan 44919, Republic of Korea;
c USDA–ARS, Environmental Microbial and Food Safety Lab., 10300 Baltimore Ave. Building 173, BARC-EAST, Beltsville, MD 20705;
d Busan Metropolitan City Public Health and Environment Research Institute, Busan, 46616, Republic of Korea;
e Environmental Engineering and Graduate School of FEED of Eco-Friendly Offshore Structure, Changwon National Univ., Changwon, Gyeongsangnam-do 51140, Republic of Korea.
2. Summary
Currently, determining the level of fecal contamination takes 24 h. The objective of this study is to predict the current level of fecal contamination (enterococcus [ENT] and Escherichia coli) using readily available environmental variables. Artificial neural network (ANN) and support vector regression (SVR) models were constructed using data from the Haeundae and Gwangalli Beaches in Busan City. The input variables included the tidal level, air and water temperature, solar radiation, wind direction and velocity, precipitation, discharge from the wastewater treatment plant, and suspended solid concentration in beach water. The dependence of fecal contamination on the input variables was statistically evaluated; precipitation, discharge from the wastewater treatment plant, and wind direction at the two beaches were positively correlated to the changes in the two bacterial concentrations (p < 0.01), whereas solar radiation was negatively correlated (p < 0.01). The performance of the ANN model for predicting ENT and E. coli at Gwangalli Beach was significantly higher than that of the SVR model with the training dataset (p < 0.05). Based on the comparison of residual values between the predicted and observed fecal indicator bacteria concentrations in two models, the ANN demonstrated better performance than SVR. This study suggests an effective prediction method to determine whether a beach is safe for recreational use.
3. Originality & Creativity
This study demonstrates that useful tools can be developed for the reliable and rapid simulation of FIB concentrations using available explanatory variables in beach waters. It is expected that a machine learning model can be readily constructed and queried by decision makers working with recreational coastal waters.
4. Application to research
The ANN demonstrated better performance than SVR based on the comparison of prediction accuracy and residual values between two models.
5. Contact
Jeongwoo Moon / Intern student
Environmental Systems Engineering Lab.
School of Earth Sciences & Environmental Engineering
Gwangju Institute of Science and Technology
1 Oryong-dong Buk-gu Gwangju, 500-712, Korea
Phone : +82-10-9384-8271
E-mail : mjwabc@naver.com