Interpretable tree-based ensemble model for predicting beach water quality

2022 ◽  
pp. 118078
Author(s):  
Lingbo Li ◽  
Jundong Qiao ◽  
Guan Yu ◽  
Leizhi Wang ◽  
Hong-Yi Li ◽  
...  
2021 ◽  
Author(s):  
Ramien Sereshk

It is commonly assumed that the persistence model, using day-old monitoring results, will provide accurate estimates of real-time bacteriological concentrations in beach water. However, the persistence model frequently provides incorrect results. This study: 1. develops a site-specific predictive model, based on factors significantly influencing water quality at Beachway Park; 2. determines the feasibility of the site-specific predictive model for use in accurately predicting near real-time E. coli levels. A site-specific predictive model, developed for Beachway Park, was evaluated and the results were compared to the persistence model. This critical performance evaluation helped to identify the inherent inaccuracy of the persistence model for Beachway Park, which renders it an unacceptable approach for safeguarding public health from recreational water-borne illnesses. The persistence model, supplemented with a site-specific predictive model, is recommended as a feasible method to accurately predict bacterial levels in water on a near real-time basis.


2012 ◽  
Vol 6 (3) ◽  
pp. 164-180 ◽  
Author(s):  
W. Thoe ◽  
S.H.C. Wong ◽  
K.W. Choi ◽  
J.H.W. Lee

2015 ◽  
Vol 14 (1) ◽  
pp. 97-108 ◽  
Author(s):  
Wai Thoe ◽  
King Wah Choi ◽  
Joseph Hun-wei Lee

A beach water quality prediction system has been developed in Hong Kong using multiple linear regression (MLR) models. However, linear models are found to be weak at capturing the infrequent ‘very poor’ water quality occasions when Escherichia coli (E. coli) concentration exceeds 610 counts/100 mL. This study uses a classification tree to increase the accuracy in predicting the ‘very poor’ water quality events at three Hong Kong beaches affected either by non-point source or point source pollution. Binary-output classification trees (to predict whether E. coli concentration exceeds 610 counts/100 mL) are developed over the periods before and after the implementation of the Harbour Area Treatment Scheme, when systematic changes in water quality were observed. Results show that classification trees can capture more ‘very poor’ events in both periods when compared to the corresponding linear models, with an increase in correct positives by an average of 20%. Classification trees are also developed at two beaches to predict the four-category Beach Water Quality Indices. They perform worse than the binary tree and give excessive false alarms of ‘very poor’ events. Finally, a combined modelling approach using both MLR model and classification tree is proposed to enhance the beach water quality prediction system for Hong Kong.


2008 ◽  
Vol 18 (2) ◽  
pp. 149-158 ◽  
Author(s):  
Richard Gersberg ◽  
Jürgen Tiedge ◽  
Dana Gottstein ◽  
Sophie Altmann ◽  
Kayo Watanabe ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document