All Issue

2025 Vol.58, Issue 4 Preview Page

Research Article

Prediction of total organic carbon using multi-model ensemble based on satellite imagery in estuarine reservoir

담수호의 위성영상 기반 Multi-model Ensemble을 통한 TOC 예측

Jinuk Kima
,
Wonjin Jangb
,
Jin hwi Kimc
,
Yong-Gu Leed
,
Jaeki Shine
,
Yongeun Parkf
,
Seongjoon Kimg*
김 진욱a
,
장 원진b
,
김 진휘c
,
이 용구d
,
신 재기e
,
박 용은f
,
김 성준g*
aPh.D. Student, Department of Civil, Environmental and Plant Engineering, Konkuk University, Seoul, Korea
bPh.D. Student, Department of Civil, Environmental and Plant Engineering, Konkuk University, Seoul, Korea
cResearcher, Future and Fusion Lab of Architectural, Civil and Environmental Engineering, Korea University, Seoul, Korea
dResearcher Professor, Environmental Engineering, Kangwon National University, Chuncheon, Korea
eCSO President, Limnoecological Science Research Institute Korea, THE HANGANG, Miryang, Korea
fProfessor, School of Civil and Environmental Engineering, College of Engineering, Konkuk University, Seoul, Korea
gProfessor, School of Civil and Environmental Engineering, College of Engineering, Konkuk University, Seoul, Korea
a건국대학교 일반대학원 사회환경플랜트공학과 석박사통합과정
b건국대학교 일반대학원 사회환경플랜트공학과 석박사통합과정
c고려대학교 미래건설환경융합연구소 박사후연구원
d강원대학교 환경공학과 연구교수
e수생태원 한강 원장
f건국대학교 공과대학 사회환경공학부 교수
g건국대학교 공과대학 사회환경공학부 교수
*Corresponding Author
30 April 2025. pp. 345-358
PDF XML Citation
Abstract

This study aimed to indirectly estimate Total Organic Carbon (TOC) in the Namyang Reservoir using a combination of satellite remote sensing and a machine learning-based Multi-Ensemble model. Sentinel-2A/B satellite data were used to construct reflectance datasets for the study area, and the HSIC-Lasso model identified 10 significant input variables (B4/B3, B4/B5, B2/B3, B8/B7, B4/B2, B1/B3, B1/B5, B8/B6, B5/B2, B2/B5) highly correlated with TOC. To predict TOC, a Stacking Ensemble model was developed using Support Vector Regression (SVR), Random Forest Regression (RFR), eXtreme Gradient Boosting (XGB), and Multi-Layer Perceptron (MLP) as base models, combined with six metamodels, including Partial Least Squares (PLS) and Ridge Regression (RID). The Stacking model achieved the highest predictive performance with R2 values of 0.963(train) and 0.886(test), and MAE and RMSE values of 0.697 mg/L and 1.556 mg/L, respectively, outperforming individual machine learning models. This study integrates satellite data and machine learning models to establish a cost-effective and sustainable framework for monitoring TOC. The accumulation of long-term TOC data is expected to enhance its applicability as a practical tool for water quality management and policy development.

Keywords
Total Organic Carbon
Remote sensing
Machine learning
Stacking model
HSIC-Lasso

본 연구는 위성 원격탐사와 머신러닝 기반 Multi-Ensemble 모델의 일종인 Stacking 모형을 결합하여 남양호에서 총유기탄소(Total Organic Carbon, TOC)를 간접적으로 추정하는 것을 목표로 하였다. Sentinel-2A/B 위성 데이터를 활용하여 연구 지역의 반사도 데이터를 구축하였으며, HSIC-Lasso 모델을 통해 TOC와 높은 상관성을 가지는 10개의 주요 입력변수(B4/B3, B4/B5, B2/B3, B8/B7, B4/B2, B1/B3, B1/B5, B8/B6, B5/B2, B2/B5)를 도출하였다. TOC를 예측하기 위해 Support Vector Regression (SVR), Random Forest Regression (RFR), eXtreme Gradient Boosting (XGB), Multi-Layer Perceptron (MLP)을 베이스모델로 사용하고, Partial Least Squares (PLS) 및 Ridge Regression (RID)를 포함한 6개의 메타모델을 결합한 Stacking Ensemble 모델을 개발하였다. Stacking 모델은 훈련 데이터와 테스트 데이터에서 각각 R² 값 0.963과 0.886, MAE 값 0.697 mg/L, RMSE 값 1.556 mg/L의 값을 보였으며, 단일 머신러닝 모델의 예측 성능보다 개선됨을 보여주었다. 본 연구 결과는 위성 데이터와 머신러닝 모델을 통합하여 TOC를 비용 효율적이고 지속 가능한 방식으로 모니터링할 수 있는 기반을 제시하며, 추후 장기간의 TOC 데이터 축적을 통해 수질 관리 및 정책 개발을 위한 실용적인 도구로 활용될 수 있을 것으로 기대된다.

키워드
총유기탄소
위성영상
머신러닝
Stacking 모형
HSIC-Lasso
References
1

Ahmed, K., Sachindra, D.A., Shahid, S., Iqbal, Z., Nawaz, N., and Khan, N. (2020). "Multimodel ensemble predictions of precipitation and temperature using machine learning algorithms." Atmospheric Research, Vol. 236, 104806.

10.1016/j.atmosres.2019.104806
2

Amri, M.R.E., and Marrel, A. (2021). "Optimized HSIC-based tests for sensitivity analysis: Application to thermalhydraulic simulation of accidental scenario on nuclear reactor." Quality and Reliability Engineering International, Vol. 38, No. 3, pp. 1386-1403.

10.1002/qre.2954
3

Bonelli, A.G., Loisel, H., Jorge, D.S.F., Mangin, A., d'Andon, O.F., and Vantrepotte, V. (2022). "A new method to estimate the dissolved organic carbon concentration from remote sensing in the global open ocean." Remote Sensing of Environment, Vol. 281, 113227.

10.1016/j.rse.2022.113227
4

Breiman, L. (2001). "Random forests." Machine Learning, Vol. 45, pp. 5-32.

10.1023/A:1010933404324
5

Chang, N.B., Vannah, B.W., Yang, Y.J., and Elovitz, M. (2014). "Integrated data fusion and mining techniques for monitoring total organic carbon concentrations in a lake." International Journal of Remote Sensing, Vol. 35, No. 3, pp. 1064-1093.

10.1080/01431161.2013.875632
6

Chen, T., and Guestrin, C. (2016). "XGBoost: A scalable tree boosting system." KDD '16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, U.S., pp. 785-794.

10.1145/2939672.2939785
7

Chen, Y., Song, L., Liu, Y., Yang, L., and Li, D. (2020). "A review of the artificial neural network models for water quality prediction." Applied Sciences, Vol. 10, No. 17, 5776.

10.3390/app10175776
8

Cho, D.J., Yoo, C.H., Im, J.H., Lee, Y.S., and Lee, J.S. (2020). "Improvement of spatial interpolation accuracy of daily maximum air temperature in urban areas using a stacking ensemble technique." GIScience & Remote Sensing, Vol. 57, No. 5, pp. 633-649.

10.1080/15481603.2020.1766768
9

Chung, J.H., Lee, Y.G., Kim, J.U., Jang, W.J., and Kim, S.J. (2023). "Soil moisture estimation using the water cloud model and Sentinel-1 & -2 satellite image-based vegetation indices." Journal of Korea Water Resources Association, Vol. 56, No. 3, pp. 211-224.

10.3741/JKWRA.2023.56.3.211
10

Gascon, F., Bouzinac, C., Thepaut, O., Jung, M., Francesconi, B., Louis, J., Longjou, V., Lafrance, B., Massera, S., and Gaudel- Vacaresse, A. et al. (2017). "Copernicus Sentinel-2A calibration and products validation status." Remote Sensing, Vol. 9, No. 6, 584.

10.3390/rs9060584
11

International Ocean Colour Coordinating Group (IOCCG) (2018). Earth observations in support of global water quality monitoring. Reports and Monographs of the International Ocean Colour Coordinating Group No. 17, Dartmouth, Canada.

12

Jang, W.J., Kim, J.U., Kim, J.H., Shin, J.K., Chon, K.M., Kang, E.T., Park, Y.E., and Kim, S.J. (2024). "Evaluation of Sentinel- 2 based Chlorophyll-a estimation in a small-scale reservoir: Assessing accuracy and availability." Remote Sensing, Vol. 16, 315.

10.3390/rs16020315
13

Jeong, D.H., Chung, H.M., Cho, Y.S., Kim, E.S., Kim, C.S., Park, J.W., Lee, W.S. (2018). "A study on operation and management for TOC removal of public sewage treatment works." Journal of Korean Society of Water and Wastewater, Vol. 32, No. 6, pp. 535-550.

10.11001/jksww.2018.32.6.535
14

Jiang, D., Matsushita, B., Pahlevan, N., Gurlin, D., Fichot, C.G., Harringmeyer, J., Sent, G., Brito, A.C., Brotas, V., and Werther, M. (2023). "Estimating the concentration of total suspended solids in inland and coastal waters from Sentinel-2 MSI: A semi-analytical approach." ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 204, pp. 362-377.

10.1016/j.isprsjprs.2023.09.020
15

Jung, K.Y., Lee, Y.K., Yoo, H.Y, Nam, G.S., Hur, J. (2021). "Changes in spectroscopic and molecular weight characteristics of dissolved organic matter in an agriculture reservoir during a summer monsoon." Journal of Korean Society on Water Environment, Vol. 37, No. 6, pp. 458-468.

16

Kahru, M., and Mitchell, B.G. (1998). "Spectral reflectance and absorption of a massive red tide off southern California." Journal of Geophysical Research, Vol. 103, No. C10, pp. 21601-21609.

10.1029/98JC01945
17

Kim, H.C., Kim, D.Y., and Kim. B.S. (2024). "Performance evaluation of stacking models based on random forest, XGBoost, and LGBM for wind power forecasting." Journal of Wind Energy, Vol. 15, No. 3, pp. 21-29.

18

Kim, J.K, Shin, M.S., Jang, C.W., Jung, S.M., Kim, B.C. (2007). "Comparison of TOC and DOC distribution and the oxidation efficiency of BOD and COD in several reservoirs and rivers in the Han River system." Journal of Korean Society of Water Quality, Vol. 23, No. 1, pp. 72-80.

19

Kim, J.U., Jang, W.J., Shin, J.K., Kang, E.T., Kim, J.H., Park, Y.E., and Kim, S.J. (2021). "Estimating TOC concentrations using an optically-active water quality factors in estuarine reservoirs." Journal of Korean Society on Water Environment, Vol, 37, No. 6, pp. 531-538.

20

Kim, S.J., Kim, S.J., and Kim, P.S. (2002). "Development and application of Freshwater Lake Water Quality Management System (FLAQUM) through the linkage of watershed and freshwater lake." Journal of the Korean Society of Agriucltural Engineers, Vol. 44, No. 6, pp. 124-136.

21

Kim, Y.J., Kang, E.J., Cho, D.J., Lee, S.W., and Im, J.H. (2022). "Improved estimation of hourly surface ozone concentrations using stacking ensemble-based spatial interpolation." Journal of the Korean Association of Geographic Information Studies, Vol. 25, No. 3, pp. 74-99.

22

Kim, Y.S., and Lee, E.J. (2019). "Establishment of target water quality for TOC of total water load management system." Journal of Korean Society on Water Environment, Vol. 35, No. 6, pp. 520-538.

23

Kutser, T., Verpoorter, C., Paavel, B., and Tranvik, L.J. (2015). "Estimating lake carbon fractions from remote sensing data." Remote Sensing of Environment, Vol. 157, pp. 138-146.

10.1016/j.rse.2014.05.020
24

Kwak, Y.J., Go, C.Y., Kwang, S.Y., and Eem, S.H. (2023). "Development of a high-performance concrete compressive-strength prediction model using an ensemble machine-learning method based on bagging and stacking." Journal of the Computational Structural Engineering Institute of Korea, Vol. 36, No. 1, pp. 9-18.

10.7734/COSEIK.2023.36.1.9
25

Latwal, A., Rehana, S., and Rajan, K.S. (2023). "Detection and mapping of water and chlorophyll-a spread using Sentinel-2 satellite imagery for water quality assessment of inland water bodies." Environmental Monitoring and Assessment, Vol. 195, 1304.

10.1007/s10661-023-11874-737828127
26

Lee, K.H., and Kam, J.H. (2024). "Spatiotemporal patterns of water volume and total organic carbon concentration of agricultural reservoirs over South Korea." Water Research, Vol. 256, 121610.

10.1016/j.watres.2024.12161038663212
27

Lee, Y.J., Hur, J., and Shin K.H. (2014). "Characterization and source identification of organic matter in view of land uses and heavy rainfall in the Lake Shihwa, Korea." Marine Pollution Bulletin, Vol. 84, No. 1-2, pp. 322-329.

10.1016/j.marpolbul.2014.04.03324841714
28

Panahi, J., Mastouri, R., and Shabanlou, S. (2022). "Insights into enhanced machine learning techniques for surface water quantity and quality prediction based on data pre-processing algorithms." Journal of Hydroinformatics, Vol. 24, No. 4, pp. 875-897.

10.2166/hydro.2022.022
29

Park, S.W., Lee, S.J., Yun, W.S., Shin, D.Y., Park, S.Y., and Lee, Y.W. (2018). "Estimation of Chlorophyll-a concentration for inland water using red-edge band of Sentinel-2 and RapidEye." Journal of the Korean Association of Professional Geographers, Vol. 52, No. 3, pp. 445-454.

30

Raiyani, K., Goncalves, T., Rato, L., Salgueiro, P., and Silva, R.M.D. (2021). "Sentinel-2 image scene classification: A comparison between Sen2Cor and a machine learning approach." Remote Sensing, Vol. 13, No. 2, 300.

10.3390/rs13020300
31

Ren, W.W., Yang, T., Huang, C.S., Xu, C.Y., and Shao, Q.X. (2018). "Improving monthly streamflow prediction in alpine regions: integrating HBV model with Bayesian neural network." Vol. 32, pp. 3381-3396.

10.1007/s00477-018-1553-x
32

Rumelhart, D.E., Hinton, G.E., and Williams, G.E. (1986). "Learning internal representations by error propagation." Parallel distributed processing: Explorations in the microstructures of cognition, Vol. I, Edited by Rumelhart, D.E., and McClelland, J.L., MIT Press, Cambridge, MA, U.S., pp. 318-362.

33

Satish, N., Anmala, J., Rajitha, K., and Varma, M.R.R. (2024). "A stacking ANN ensemble model of ML models for stream water quality prediction of Godavari River Basin, India." Ecological Informatics, Vol. 80, 102500.

10.1016/j.ecoinf.2024.102500
34

Shang, Y., Liu, G., Wen, Z., Jacinthe, P.A., Song, K., Zhang, B.,, Lyu, L., Li, S., Wang, X., and Yu, X. (2021). "Remote estimates of CDOM using Sentinel-2 remote sensing data in reservoirs with different trophic states across China." Journal of Environmental Management, Vol. 286, 112275.

10.1016/j.jenvman.2021.11227533684799
35

Shin, G.J., Gong, S.H., and Kim, T.G. (2020). "Mapping Chlorophyll-a distribution of drinking water reservoirs using Landsat 8 satellite imagery data." Journal of the Korean Society for Environmental Technology, Vol. 21, No. 1, pp. 40-47.

10.26511/JKSET.21.1.6
36

Sun, W., and Du, Q. (2019). "Hyperspectral band selection: A review." IEEE Geoscience and Remote Sensing Magazine, Vol. 7, No. 2, pp. 118-139.

10.1109/MGRS.2019.2911100
37

Tian, D., Zhao, X., Gao, L., Liang, Z., Yang, Z., Zhang, P., Wu, Q., Ren, K., Li, R., and Yang, C. et al. (2024). "Estimation of water quality variables based on machine learning model and cluster analysis-based empirical model using multi-source remote sensing data in inland reservoirs, South China." Environmental Pollution, Vol. 342, 123104.

10.1016/j.envpol.2023.12310438070645
38

U.S. Geological Survey (USGS) (2023). "Chapter A6." Guidelines for field-measured water-quality properties, U.S. Geological Survey Techniques and Methods, book 9, U.S. Department of the Interior, Reston, VA, U.S.

39

Vapnik, V., Colowich, S., and Smola, A. (1996). "Support vector method for function approximation, regression estimation and signal processing." Advances in Neural Information Processing Systems, Vol. 9, pp. 281-287.

40

Wolpert, D.H. (1992). "Stacked generalization." Neural Networks, Vol. 5, No. 2, pp. 241-259.

10.1016/S0893-6080(05)80023-1
41

Yamada, M., Jitkrittum, W., Sigal, L., Xing, E.P., and Sugiyama, M. (2014). "High-dimensional feature selection by feature-wise kernelized lasso." Neural Computation, Vol. 26, No. 1, pp. 185-207.

10.1162/NECO_a_0053724102126
42

Zhai, B., and Chen, J. (2018). "Development of a stacked ensemble model for forecasting and analyzing daily average PM2.5 concentrations in Beijing, China." Science of The Total Environment, Vol. 635, pp. 644-658.

10.1016/j.scitotenv.2018.04.04029679837
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :4
  • Pages :345-358
  • Received Date : 2024-12-08
  • Revised Date : 2025-03-05
  • Accepted Date : 2025-03-07
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.4.345
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close