Evaluation of Surrogate Monitoring Parameters for SS and T-P Using Multiple Linear Regression and Random Forest
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Effective nonpoint source (NPS) pollution management requires frequent water quality monitoring, which is, however, often costly to be implementedin practice. Statistical techniques and machine learning methods allow us to identify and focus on fundamental environmental variables that have closerelationships with NPS pollutants of interest. This study developed surrogate models to predict the concentrations of suspended sediment (SS) and totalphosphorus (T-P) from turbidity and runoff discharge rates using multiple linear regression (MLR) and random forest (RF) methods. The RF modelsprovided acceptable performance in predicting SS and T-P, especially when runoff discharge rates were high. The RF models outperformed the MLRmodels in all the cases. Such finding highlights the potential of RF techniques and models as a tool to identify fundamental environmental variablesthat are measured in relatively inexpensive ways or freely available but still able to provide information required to quantify the concentrations o f NPSpollutants. The analysis of relative importance rates showed that the temporal variations of SS and T-P concentrations could be more effectivelyexplained by that of turbidity than runoff discharge rate. This study demonstrated that the advanced statistical techniques such as machine learning couldhelp to improve the efficiency of NPS pollutants monitoring.
Abstract
Effective nonpoint source (NPS) pollution management requires frequent water quality monitoring, which is, however, often costly to be implementedin practice. Statistical techniques and machine learning methods allow us to identify and focus on fundamental environmental variables that have closerelationships with NPS pollutants of interest. This study developed surrogate models to predict the concentrations of suspended sediment (SS) and totalphosphorus (T-P) from turbidity and runoff discharge rates using multiple linear regression (MLR) and random forest (RF) methods. The RF modelsprovided acceptable performance in predicting SS and T-P, especially when runoff discharge rates were high. The RF models outperformed the MLRmodels in all the cases. Such finding highlights the potential of RF techniques and models as a tool to identify fundamental environmental variablesthat are measured in relatively inexpensive ways or freely available but still able to provide information required to quantify the concentrations o f NPSpollutants. The analysis of relative importance rates showed that the temporal variations of SS and T-P concentrations could be more effectivelyexplained by that of turbidity than runoff discharge rate. This study demonstrated that the advanced statistical techniques such as machine learning couldhelp to improve the efficiency of NPS pollutants monitoring.
참고문헌 (36)
- Afendras, G., "Optimality of training", test size and resampling effectiveness in cross-validation, 2019, pp.199, 286~301
- Akaike, H., "Information theory and an extension of the maximum likelihood principle", Second International Symposium on Information Theory, pp.267~281 (1973)
- American Public Health Association(APHA), "Standard Methods for the Examination of Water and Waste Water. 21st ed", ~ (2001)
- Breiman, L., "Random forests", Machine Learning, 45, pp.5~32 (2001)
- Breiman, L., "Classification and Regression Trees" (1984), CRC press
- Handan Çamdevýren, "Use of principal component scores in multiple linear regression models for prediction of Chlorophyll-a in reservoirs", Ecological Modelling, 181(4), pp.581~589 (2005), Elsevier BV[DOI]
- Chattergee, S., "Sensitivity Analysis in Linear Regression" (1988), Wiley
- Chenini, I., "Evaluation of ground water quality using multiple linear regression and structural equation modeling", International Journal of Environmental Science and Technology, 6(3), pp.509~519 (2009)[DOI]
- David K. DeForest, "Multiple linear regression models for predicting chronic aluminum toxicity to freshwater aquatic organisms and developing water quality guidelines", Environmental Toxicology and Chemistry, 37(1), pp.80~90 (2018), Wiley[DOI]
- Ramón Díaz-Uriarte, "Gene selection and classification of microarray data using random forest", BMC Bioinformatics, 7, pp.3~ (2006), Springer Science and Business Media LLC[DOI]
- Francesco Granata, "Machine Learning Algorithms for the Forecasting of Wastewater Quality Indicators", Water, 9(2), pp.105~ (2017), MDPI AG[DOI]
- Mahdi Hasanipanah, "Forecasting blast-induced ground vibration developing a CART model", Engineering with Computers, 33(2), pp.307~316 (2017), Springer Science and Business Media LLC[DOI]
- Jeffery S. Horsburgh, "A sensor network for high frequency estimation of water quality constituent fluxes using surrogates", Environmental Modelling & Software, 25(9), pp.1031~1044 (2010), Elsevier BV[DOI]
- Jeffrey N. Houser, "Upland Disturbance Affects Headwater Stream Nutrients and Suspended Sediments during Baseflow and Stormflow", Journal of Environmental Quality, 35(1), pp.352~365 (2006), Wiley[DOI]
- Johanna, I. F., "A turbidity-based method to continuously monitor sediment, carbon and nitrogen flows in mountainous watersheds", Journal of Hydrology, 513, pp.45~57 (2014)[DOI]
- Amber Spackman Jones, "Surrogate Measures for Providing High Frequency Estimates of Total Suspended Solids and Total Phosphorus Concentrations1", JAWRA Journal of the American Water Resources Association, 47(2), pp.239~253 (2011), Wiley[DOI]
- P. Jordan, "Characterising phosphorus transfers in rural catchments using a continuous bank-side analyser", Hydrology and Earth System Sciences, 11(1), pp.372~381 (2007), Copernicus GmbH[DOI]
- Mareike Ließ, "Uncertainty in the spatial prediction of soil texture", Geoderma, 170, pp.70~79 (2012), Elsevier BV[DOI]
- D. N. Moriasi, "Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations", Transactions of the ASABE, 50(3), pp.885~900 (2007), American Society of Agricultural and Biological Engineers (ASABE)[DOI]
- JAMI L. MONTGOMERY, "The WATERS Network: An Integrated Environmental Observatory Network for Water Research", Environmental Science & Technology, 41(19), pp.6642~6647 (2007), American Chemical Society (ACS)[DOI]
- J.E. Nash, "River flow forecasting through conceptual models part I — A discussion of principles", Journal of Hydrology, 10(3), pp.282~290 (1970), Elsevier BV[DOI]
- Issoufou Ouedraogo, "Aplicação de regressão de floresta aleatória e comparação de seu desempenho com a regressão linear múltipla na modelagem da concentração de nitrato de águas subterrâneas na escala do continente Africano", Hydrogeology Journal, 27(3), pp.1081~1098 (2019), Springer Science and Business Media LLC[DOI]
- Anantha M. Prasad, "Newer Classification and Regression Tree Techniques: Bagging and Random Forests for Ecological Prediction", Ecosystems, 9(2), pp.181~199 (2006), Springer Science and Business Media LLC[DOI]
- M RAZI, "A comparative predictive analysis of neural networks (NNs), nonlinear regression and classification and regression tree (CART) models", Expert Systems with Applications, 29(1), pp.65~74 (2005), Elsevier BV[DOI]
- Victor Rodriguez-Galiano, "Predictive modeling of groundwater nitrate pollution using Random Forest and multisource variables related to intrinsic and specific vulnerability: A case study in an agricultural setting (Southern Spain)", Science of The Total Environment, 476-477, pp.189~206 (2014), Elsevier BV[DOI]
- Hermann Rügner, "Turbidity as a proxy for total suspended solids (TSS) and particle facilitated pollutant transport in catchments", Environmental Earth Sciences, 69(2), pp.373~380 (2013), Springer Science and Business Media LLC[DOI]
- David Scholefield, "Concerted diurnal patterns in riverine nutrient concentrations and physical conditions", Science of The Total Environment, 344(1-3), pp.201~210 (2005), Elsevier BV[DOI]
- Steven Settle, "Determination of Surrogate Indicators for Phosphorus and Solids in Urban Stormwater: Application of Multivariate Data Analysis Techniques", Water, Air, and Soil Pollution, 182(1-4), pp.149~161 (2007), Springer Science and Business Media LLC[DOI]
- Balraj Singh, "Modelling of impact of water quality on infiltration rate of soil by random forest regression", Modeling Earth Systems and Environment, 3(3), pp.999~1004 (2017), Springer Science and Business Media LLC[DOI]
- Michael Snipes, "Model selection and Akaike Information Criteria: An example from wine ratings and prices", Wine Economics and Policy, 3(1), pp.3~9 (2014), Firenze University Press[DOI]
- United States Environmental Protection Agency(USEPA), "“An approach for using load duration curves in the development of TMDLs.” 841-B-07-006", United States Environmentl Protection Agency, 1~68 (2007)
- Verzani, J., "Data sets, etc. for the text Using R for introductory statistics. 2nd ed. Version 2.0-6", ~ (2018)
- Ana Villa, "Determining suspended solids and total phosphorus from turbidity: comparison of high-frequency sampling with conventional monitoring methods", Environmental Monitoring and Assessment, 191(10), pp.605~ (2019), Springer Science and Business Media LLC[DOI]
- Wang, H., "Mapping Robinia pseudoacacia forest health conditions by using combined spectral, spatial, and textural information extracted from IKONOS imagery and random forest classifier", Remote Sensing, 7(7), pp.9020~9044 (2015)[DOI]
- Zambrano-Bigiarini, M., "Goodness-of-fit functions for comparison of simulated and observed hydrological time series. Version 0.3-10", ~ (2017)
- Sediment load monitoring in the Mae Sa catchment in Northern Thailand, "Ziegler, A. D.", ~, pp.IAHS-AISH publication 86-91 (2011)