한국수자원학회논문집 (Journal of Korea Water Resources Association)

  • 제54권2호
  • /
  • Pages.121-133
  • /
  • 2021
  • /
  • 2799-8746(pISSN)
  • /
  • 2799-8754(eISSN)
한국수자원학회 (Korea Water Resources Association)
권호 표지
DOI QR코드

DOI QR Code

기계학습모형을 이용한 다분광 위성 영상 기반 낙동강 부유 물질 농도 계측 기법 개발

Development of suspended solid concentration measurement technique based on multi-spectral satellite imagery in Nakdong River using machine learning model

  • 권시윤 (서울대학교 건설환경공학부) ;
  • 서일원 (서울대학교 건설환경공학부) ;
  • 백동해 (서울대학교 건설환경공학부)
  • Kwon, Siyoon (Department of Civil and Environmental Engineering, Seoul National University) ;
  • Seo, Il Won (Department of Civil and Environmental Engineering, Seoul National University) ;
  • Beak, Donghae (Department of Civil and Environmental Engineering, Seoul National University)
  • 투고 : 2020.12.18
  • 심사 : 2021.01.15
  • 발행 : 2021.02.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

하천에서 발생하는 부유 물질은 주로 유역으로부터 유입되거나 하천 내에서 자생으로 발생하기도 하며, 퇴적되어 중장기적인 수질 오염을 초래할 수도 있는 중요한 수질 인자이다. 하지만, 부유물질의 재래식 계측방식은 점 단위 계측이기 때문에 노동 집약적이며 방대한 양의 자료를 취득하기는 어렵다. 따라서, 본 연구에서는 고해상도 다분광 위성영상을 제공하는 Sentinel-2 위성 자료를 이용하여 낙동강 전역에 대한 원격탐사 기반 부유 물질 농도 계측 기법을 개발하였다. 개발된 기법은 기존 원격탐사 기반 회귀식들의 한계점을 개선하고 낙동강 전체 영역의 지역적 특성을 반영하기 위해 기계학습 모형인 서포트 벡터 회귀(Support Vector Regression, SVR) 모형을 이용하여 다양한 파장대의 분광 밴드들과 밴드비(band ratios)를 고려하였으며, 이를 입력 변수들의 최적 조합으로 재귀적 특징 제거법(Recursive Feature Elimination, RFE)과 SVR의 각 변수별 가중계수를 활용하여 도출하였다. 가장 중요도가 높은 분광 밴드로는 Red-edge 파장대 영역에 속하는 705 nm 밴드가 산출되었으며, 최종적으로 구축된 SVR 모형을 선행 연구들에서 제시한 회귀식들과 비교한 결과, 가장 정확한 계측 결과를 제공하는 것으로 밝혀졌다. 본 연구에서 개발된 SVR 모형은 RFE를 통해 산출된 최적 분광 밴드 조합을 바탕으로 하기 때문에 기존 단일 분광 밴드 혹은 밴드비를 기반으로 구축된 회귀식들이 가지는 변수 의존도를 낮추는 동시에 더욱 정확한 부유물질 농도 공간분포를 제공할 수 있을 것으로 판단된다.

Suspended Solids (SS) generated in rivers are mainly introduced from non-point pollutants or appear naturally in the water body, and are an important water quality factor that may cause long-term water pollution by being deposited. However, the conventional method of measuring the concentration of suspended solids is labor-intensive, and it is difficult to obtain a vast amount of data via point measurement. Therefore, in this study, a model for measuring the concentration of suspended solids based on remote sensing in the Nakdong River was developed using Sentinel-2 data that provides high-resolution multi-spectral satellite images. The proposed model considers the spectral bands and band ratios of various wavelength bands using a machine learning model, Support Vector Regression (SVR), to overcome the limitation of the existing remote sensing-based regression equations. The optimal combination of variables was derived using the Recursive Feature Elimination (RFE) and weight coefficients for each variable of SVR. The results show that the 705nm band belonging to the red-edge wavelength band was estimated as the most important spectral band, and the proposed SVR model produced the most accurate measurement compared with the previous regression equations. By using the RFE, the SVR model developed in this study reduces the variable dependence compared to the existing regression equations based on the single spectral band or band ratio and provides more accurate prediction of spatial distribution of suspended solids concentration.

키워드

참고문헌

  1. Arisanty, D., and Nur Saputra, A. (2017). "Remote sensing studies of suspended sediment concentration variation in barito delta." IOP Conference Series: Earth and Environmental Science, Yogyakarta, Indonesia, Vol. 98, pp. 0-6, doi: 10.1088/1755-1315/98/1/012058.
  2. Beschta, R.L., Bilby, R.E., Brown, G.W., Holtby, L.B., and Hofstra, T.D. (1987). "Stream temperature and aquatic habitat; fisheries and forestry interactions." Streamside management forestry and fishery interactions, Edited by Salo, E.O., Cundy, T.W., University of Washington, Institute of Forest Resources, Contribution No 57: Seattle, WA, pp. 191-232.
  3. Bhargava, D.S., and Mariam, D.W. (1991). "Light penetration depth, turbidity and reflectance related relationships and models." ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 46, No. 4, pp. 217-230, doi: 10.1016/0924-2716(91)90055-Z.
  4. Caballero, I., Steinmetz, F., and Navarro, G. (2018). "Evaluation of the first year of operational Sentinel-2A data for retrieval of suspended solids in medium- to high-turbidity waters." Remote Sensing, Vol. 10, No. 7, p. 982, doi: 10.3390/rs10070982.
  5. Chen, Z.M., Hanson, J.D., and Curran, P.J. (1991). "The form of the relationship between suspended sediment concentration and spectral reflectance‒Its implications for the use of Daedalus 1268 data." International Journal of Remote Sensing, Vol. 12, No. 1, pp. 215-222, doi: 10.1080/01431169108929647.
  6. Chi, M., Feng, R., and Bruzzone, L. (2008). "Classification of hyper-spectral remote-sensing data with primal SVM for small-sized training dataset problem." Advances in Space Research, Vol. 41, pp. 1793-1799, doi: 10.1016/j.asr.2008.02.012.
  7. Chu, V.W., Smith, L.C., Rennermalm, A.K., Forster, R.R., Box, J.E., and Rech, N. (2009). "Sediment plume response to surface melting and supraglacial lake drainages on the Greenland ice sheet." Journal of Glaciology, Vol. 55, No. 194, 1072e1082. https://doi.org/10.3189/002214309790794904
  8. Dekkera, A.G., Vosb, R.J., and Petersb, S.W.M. (2001). "Comparison of remote sensing data, model results and in-situ data for to- tal suspended matter zTSM/in the southern Frisian lakes." Science of the Total Environment, Vol. 268, pp. 197-214, doi: 10.1016/S0048-9697(00)00679-3.
  9. Dethier, E.N., Renshaw, C.E., and Magilligan, F.J. (2020). "Toward Improved accuracy of remote sensing approaches for quantifying suspended sediment: Implications for suspended sediment monitoring." Journal of Geophysical Research. Earth Surface, Vol. 125, No. 7, e2019JF005033, doi: 10.1029/2019JF005033.
  10. Doxaran, D., Froidifond, J.M., and Castaing, P. (2003). "Remote-sensing reflectance of turbid sediment-dominated waters, reduction of sediment type variations and changing illumination conditions effects by use of reflectance ratios." Applied Optics, Vol. 42, No. 15, 2623e2634. https://doi.org/10.1364/AO.42.002623
  11. Du, Y., Zhang, Y., Ling, F., Wang, Q., Li, W., and Li, X. (2016). "Water bodies' mapping from Sentinel-2 imagery with modified normalized difference water index at 10-m spatial resolution produced by sharpening the swir band." Remote Sensing, Vol. 8, No. 4, p. 354. doi: 10.3390/rs8040354.
  12. Fang, G., Chen, S., Wang, H., Qian, J., and Zhang, L. (2010). "Detecting marine intrusion into rivers using EO-1 ALI satellite imagery: Modaomen Waterway, Pearl River Estuary, China." International Journal of Remote Sensing, Vol. 31, No. 15, 4125e4146. https://doi.org/10.1080/01431160903229218
  13. Gin, K.Y.H., Koh, S.T., and Lin, I.I. (2003). "Spectral irradiance profiles of suspended marine clay for the estimation of suspended sediment concentration in tropical waters." International Journal of Remote Sensing, Vol. 24, pp. 3235-3245. doi: 10.1080/01431160110114934.
  14. Guyon, I., Weston, J., Barnhil,l S., and Vapnik, V. (2002). "Gene selection for cancer classification using support vector machines." Machine Learning, Vol. 46, pp. 389-422. https://doi.org/10.1023/a:1012487302797
  15. Islam, A., Gao, J., Ahmad, W., Neil, D., and Bell, P. (2003). "Image calibration to like- values in mapping shallow water quality from multi temporal data." Photo- grammetric Engineering & Remote Sensing, Vol. 69, No. 5, 567e575. https://doi.org/10.14358/PERS.69.5.567
  16. Islam, M.R., Yamaguchi, Y., and Ogawa, K., (2001). "Suspended sediment in the Ganges and Brahmaputra Rivers in Bangladesh: Observation from TM and AVHRR data." Hydrological Processes. Vol. 15, pp. 493-509, doi: 10.1002/hyp.165.
  17. Ismail, K., Boudhar, A., Abdelkrim, A., Mohammed, H., Mouatassime, S., Kamal, A., Driss, E., Idrissi, E., and Nouaim, W. (2019). "Evaluating the potential of Sentinel-2 satellite images for water quality characterization of artificial reservoirs: The Bin El Ouidane Reservoir case study (Morocco)." Meteorology Hydrology and Water Management, Vol. 7, No. 1, pp. 31-39, doi: 10.26491/mhwm/95087.
  18. Joshi, I.D., D'Sa, E.J., Osborn, C.L, and Bianchi, T.S. (2017). "Turbidity in Apalachicola Bay, Florida from Landsat 5 TM and field data: Seasonal patterns and response to extreme events." Remote Sensing, Vol. 9, p. 367. https://doi.org/10.3390/rs9040367
  19. Lim, J., and Choi, M. (2015). "Assessment of water quality based on Landsat 8 operational land imager associated with human activities in Korea." Environmental Monitoring and Assessment, Vol. 187, pp. 1-17. doi: 10.1007/s10661-015-4616-1.
  20. Liu, H., Li, Q., Shi, T., Hu, S., Wu, G., and Zhou, Q., (2017). "Application of sentinel 2 MSI images to retrieve suspended particulate matter concentrations in Poyang Lake." Remote Sensing, Vol. 9, p. 761. doi: 10.3390/rs9070761.
  21. Lodhi, M.A., Rundquist, D.C., Han, L., Kuzila, M.S. (1997). "The potential for remote sensing of loess soils suspended in surface waters." Journal of the American Water Resources Association, Vol. 33, No. 1, pp. 111-117. https://doi.org/10.1111/j.1752-1688.1997.tb04087.x
  22. Ma, R., and Dai, J. (2005). "Investigation of chlorophyll-a and total suspended matter concentrations using landsat ETM and field spectral measurement in Taihu Lake, China." International Journal of Remote Sensing, Vol. 26, pp. 2779-2795, doi: 10.1080/01431160512331326648.
  23. Novo, E.M.M., Hansom, J.D., and Curran, P.J. (1989). "The effect of sediment type on the relationship between reflectance and suspended sediment concentration." International Journal of Remote Sensing, Vol. 10, No. 7, pp. 1283-1289. doi: 10.1080/01431168908903967.
  24. Osadchiev, A. (2015). "A method for quantifying freshwater discharge rates from satellite observations and Lagrangian numerical modeling of river plumes." Environmental Research Letters, Vol. 10, 085009, doi: 10.1088/1748-9326/10/8/085009.
  25. Pal, M., and Foody, G.M. (2010). "Feature selection for classification of hyperspectral data by SVM." IEEE Transactions on Geoscience and Remote Sensing, Vol. 48, pp. 2297-2307. doi: 10.1109/TGRS.2009.2039484.
  26. Peterson, K.T., Sagan, V., Sidike, P., Cox, A.L., and Martinez, M. (2018). "Suspended sediment concentration estimation from landsat imagery along the lower missouri and middle Mississippi Rivers using an extreme learning machine." Remote Sensing, Vol. 10, No. 10, 1503, doi: 10.3390/rs10101503.
  27. Pham, Q.V., Ha, N.T.T., Pahlevan, N., Oanh, L.T., Nguyen, T.B., and Nguyen, N.T. (2018). "Using landsat-8 images for quantifying suspended sediment concentration in red river (Northern Vietnam)." Remote Sensing, Vol. 10, No. 11, 1841. doi: 10.3390/rs10111841.
  28. Shi, H., Cao, Y., Dong, C., Xia, C., and Li, C. (2018). "The spatiotemporal evolution of river island based on Landsat satellite imagery, hydrodynamic numerical simulation and observed data." Remote Sensing, Vol. 10, No. 12, 2046. doi: 10.3390/rs10122046.
  29. Svab, E., Tyler, A.N., Preston, T., Presing, M., and Balogh, K.V. (2005) "Characterizing the spectral reflectance of algae in lake waters with high suspended sediment concentrations." International Journal of Remote Sensing, Vol. 26, pp. 919-928. https://doi.org/10.1080/0143116042000274087
  30. Umar, M., Rhoads, B.L., and Greenberg, J.A. (2018). "Use of multispectral satellite remote sensing to assess mixing of suspended sediment downstream of large river confluences." Journal of Hydrology, Vol. 556, pp. 325-338. doi: 10.1016/j.jhydrol.2017.11.026.
  31. Vanhellemont, Q., and Ruddick, K. (2015). "Advantages of high quality SWIR bands for ocean colour processing: Examples from Landsat-8." Remote Sensing of Environment, Vol. 161, pp. 89-106. doi: 10.1016/j.rse.2015.02.007.
  32. Vanhellemont, Q., and Ruddick, K. (2016) "ACOLITE For Sentinel-2: Aquatic Applications of MSI Imagery." Proceedings of Living Planet Symposium 2016, Prague, Czech Republic, Vol. 740, p. 55.
  33. Vapnik, V. (1995). The nature of statistical learning theory, Springer-Verlag, NY, U.S.
  34. Wang, J.J., and Lu, X.X. (2010). Estimation of suspended sediment concentrations using Terra MODIS: An example from the Lower Yangtze River, China. Science of the Total Environment, Vol. 408, No. 5, 1131e1138. https://doi.org/10.1016/j.scitotenv.2009.11.057
  35. Wang, J.J., Lu, X.X., Liew, S.C., and Zhou, Y. (2010). "Remote sensing of suspended sediment concentrations of large rivers using multi-temporal MODIS images: An example in the middle and lower Yangtze River, China." International Journal of Remote Sensing, Vol. 31, No. 4, pp. 1103-1111. https://doi.org/10.1080/01431160903330339
  36. Wang, J.J., Lu, X.X., Soo, C.L., and Yue, Z. (2009). Retrieval of suspended sediment concentrations in large turbid rivers using Landsat ETM+: An example from the Yangtze River, China. Earth Surface Processes and Landforms, Vol. 34, No. 8, 1082e1092. https://doi.org/10.1002/esp.1795
  37. Wright, D. (2018). "Sentinel-2 as a tool for quantifying suspended particulate matter in the Tamar Estuary." The Plymouth Student Scientist, Vol. 11, pp. 3-33.