# Development of Water Level Prediction Models Using Machine Learning in Wetlands: A Case Study of Upo Wetland in South Korea

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Study Area

^{2}and a river extension of 30 km. In the middle and lower stream of Topyeongcheon where the Upo wetland is located, the watershed becomes geographically unclear, bordering the Hyeonchangcheon, a local stream to the north, and Changnyeongcheon to the south. Moreover, it is connected to the Nakdong River, which crosses the wetland in a south to west direction.

#### 2.2. Data Used

#### 2.2.1. Dependent Variables

#### 2.2.2. Independent Variables

#### 2.3. Machine Learning Techniques

#### 2.3.1. Overview

#### 2.3.2. Artificial Neural Network

#### 2.3.3. Decision Tree

#### 2.3.4. Random Forest

#### 2.3.5. Support Vector Machine

#### 2.4. Metrics for Evaluation

_{obs}is the observed water level and Y

_{sim}is the modeled value from the model. $\overline{{\mathrm{Y}}_{\mathrm{obs}}},\overline{{\mathrm{Y}}_{\mathrm{sim}}}$ are the average values of Y

_{obs}and Y

_{sim}.

## 3. Results

#### 3.1. ANN

#### 3.2. Decision Tree

#### 3.3. Random Forest

#### 3.4. Support Vector Machine

^{(0–7)}and the range of ϵ at 0.1 intervals from 0 to 1. The parameter condition to minimize the error was found to be cost = 8 and ϵ = 0.2.

#### 3.5. Evaluation Results

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## References

- Mitsch, W.J.; Gosselink, J.G. Wetlands, 4th ed.; Van Nostrand Reinhold: New York, NY, USA, 2007. [Google Scholar]
- Kwak, J.W.; Kim, G.H.; Kim, J.W.; Singh, V.P.; Kim, H.S. Assessment of hydrological regimes for vegetation on riparian wetlands in Han River Basin, Korea. Terr. Atmos. Ocean. Sci.
**2017**, 28, 1055–1067. [Google Scholar] [CrossRef] [Green Version] - Kim, J.W.; Lee, B.E.; Kim, J.G.; Oh, S.H.; Jung, J.W.; Lee, M.J.; Kim, H.S. Functional Assessment of Gangcheon Replacement Wetland Using Modified HGM. J. Wetl. Res.
**2017**, 19, 318–326. [Google Scholar] - Environment Canada. Where Land Meets Water: Understanding Wetlands of the Great Lakes; Canadian Wildlife Service: Downsview, ON, Canada, 2002.
- Keddy, P.A. Wetland Ecology: Principles and Conservation; Cambridge University: Cambridge, UK, 2000. [Google Scholar]
- Bittmann, E. Grundlagen und Methoden des biologischen Wasserbaus. In Der Biologische Wasserbau an den Bundesstrassen; Bundesanstalt f. Gewaesserkunde: Koblenz, Germany, 1965; pp. 1–56. [Google Scholar]
- Kim, J.W. Prediction and Evaluation of Hydro-Ecology, Functions, and Sustainability of a Wetland under Climate Change. Ph.D. Thesis, Inha University, Incheon, Korea, 2019. [Google Scholar]
- Environment Agency. A Guide to Monitoring Water Levels and Flows at Wetland Sites; Environment Agency: Bristol, UK, 1996.
- Ministry of Environment. The 3rd Wetland Conservation Master Plan; Ministry of Environment: Sejong, Korea, 2018.
- Kumar, A.P.S.; Sudheer, K.P.; Jain, S.K.; Agarwal, P.K. Rainfall runoff modeling using artificial neural networks: Comparison of network types. Hydrol. Process.
**2005**, 19, 1277–1291. [Google Scholar] [CrossRef] - Yu, P.S.; Chen, S.T.; Chang, I.F. Support vector regression for real-time flood stage forecasting. J. Hydrol.
**2006**, 328, 704–716. [Google Scholar] [CrossRef] - Chau, K.W. Particle swarm optimization training algorithm for ANNs in stage prediction of shing mun river. J. Hydrol.
**2006**, 329, 363–367. [Google Scholar] [CrossRef] [Green Version] - Wilson, G.; Khondker, M.H. Data selection for a flood forecasting neural network. In Proceedings of the 4th International Conference on Hydroinformatics, Cedar Rapids, IA, USA, 23–27 August 2000. [Google Scholar]
- Bazartseren, B.; Hildebrandt, G.; Holz, K.P. Short-term water level prediction using neural networks and neuro-fuzzy approach. Neurocomputing
**2003**, 55, 439–450. [Google Scholar] [CrossRef] - Tiwari, M.K.; Chatterjee, C. Development of an accurate and reliable hourly flood forecasting model using wavelet-bootstrap-ANN (WBANN) hybrid approach. J. Hydrol.
**2010**, 394, 458–470. [Google Scholar] [CrossRef] - Jun, H.D.; Lee, J.H. A Methodology for Flood Forecasting and Warning Based on the Characteristic of Observed Water Levels Between Upstream and Downstream. J. Korean Soc. Hazard Mitig.
**2013**, 13, 367–374. [Google Scholar] [CrossRef] [Green Version] - Byeon, S.J.; Lee, S.H.; Choi, G.W.; Jung, J.G. Use of Gauged Water Level and Precipitation Data to Predict Short Term Water Level Changes. Korean Rev. Crisis Emerg. Manag.
**2014**, 10, 247–264. [Google Scholar] - Castillo, J.M.M.; Cspedes, J.M.S.; Cuchango, H.E.E. Water Level Prediction Using Artificial Neural Network Model. Int. J. Appl. Eng. Res.
**2018**, 13, 14378–14381. [Google Scholar] - Shamseldin, A.Y.; O’Connor, K.M. A real-time combination method for the outputs of different rainfall-runoff models. Hydrol. Sci. J.
**1999**, 44, 895–912. [Google Scholar] [CrossRef] - Georgakakos, K.P.; Seo, D.J.; Gupta, H.; Schaake, J.; Butts, M.B. Towards the characterization of stream-flow simulation uncertainty through multi-model ensembles. J. Hydrol.
**2004**, 298, 222–241. [Google Scholar] [CrossRef] - Young, C.C.; Liu, W.C. Prediction and modelling of rainfall–runoff during typhoon events using a physically-based and artificial neural network hybrid model. Hydrol. Sci. J.
**2015**, 60, 2102–2116. [Google Scholar] [CrossRef] - Jothiprakash, V.; Magar, R.B. Multi-time-step ahead daily and hourly intermittent reservoir inflow prediction by artificial intelligent techniques using lumped and distributed data. J. Hydrol.
**2012**, 450, 293–307. [Google Scholar] [CrossRef] - Booker, D.J.; Woods, R.A. Comparing and combining physically-based and empirically-based approaches for estimating the hydrology of ungauged catchments. J. Hydrol.
**2014**, 508, 227–239. [Google Scholar] [CrossRef] [Green Version] - Yang, T.; Gao, X.; Sorooshian, S.; Li, X. Simulating California reservoir operation using the classification and regression-tree algorithm combined with a shuffled crossvalidation scheme. Water Resour. Res.
**2016**, 52, 1626–1651. [Google Scholar] [CrossRef] [Green Version] - Bishop, C.M. Pattern Recognition and Machine Learning; Springer: New York, NY, USA, 2006. [Google Scholar]
- Kim, J.; Han, H.; Johnson, L.E.; Lim, S.; Cifelli, R. Hybrid machine learning framework for hydrological assessment. J. Hydrol.
**2019**, 577, 123913. [Google Scholar] [CrossRef] - Quinlan, J.R. Induction of decision trees. Mach. Learn.
**1986**, 1, 81–106. [Google Scholar] [CrossRef] [Green Version] - Breiman, L. Random forests. Mach. Learn.
**2001**, 45, 5–32. [Google Scholar] [CrossRef] [Green Version] - Vapnik, V. The Nature of Statistical Learning Theory; Springer: New York, NY, USA, 1995. [Google Scholar]
- MacQueen, J. Some methods for classification and analysis of multivariate observations. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, CA, USA, 21 June–18 July 1965; University of California Press: Berkeley, CA, USA, 1967; Volume 1, pp. 281–297. [Google Scholar]
- Kohonen, T. Self-organized formation of topologically correct feature maps. Biol. Cybern.
**1982**, 43, 59–69. [Google Scholar] [CrossRef] - Adamowski, J.; Chan, H.F. A wavelet neural network conjunction model for groundwater level forecasting. J. Hydrol.
**2011**, 407, 28–40. [Google Scholar] [CrossRef] - Partal, T.; Cigizoglu, H.K. Estimation and forecasting of daily suspended sediment data using wavelet-neural networks. J. Hydrol.
**2008**, 358, 317–331. [Google Scholar] [CrossRef] - Rajaee, T.; Nourani, V.; Mohammad, Z.K.; Kisi, O. River suspended sediment load prediction: Application of ANN and wavelet conjunction model. J. Hydrol. Eng.
**2011**, 16, 613–627. [Google Scholar] [CrossRef] - Adnan, R.; Ruslan, F.A.; Samad, A.M.; Zain, Z.M. Flood Water Level Modelling and Prediction Using Artificial Neural Network: Case Study of Sungai Batu Pahat in Johor. In Proceedings of the 2012 IEEE Control and System Graduate Research Colloquium, Shah Alam, Malaysia, 16–17 July 2012; pp. 22–25. [Google Scholar]
- Kisi, O.; Shiri, J.; Nikoofar, B. Forecasting daily lake levels using artificial intelligence approaches. Comput. Geosci.
**2012**, 41, 169–180. [Google Scholar] [CrossRef] - Hipni, A.; El-Shafie, A.; Najah, A.; Karim, O.A.; Hussain, A.; Mukhlisin, M. Daily forecasting of dam water levels: Comparing a support vector machine (SVM) model with adaptive neuro fuzzy inference system (ANFIS). Water Resour. Manag.
**2013**, 27, 3803–3823. [Google Scholar] [CrossRef] - Young, C.C.; Liu, W.C.; Hsieh, W.L. Predicting the Water Level Fluctuation in an Alpine Lake Using Physically Based, Artificial Neural Network, and Time Series Forecasting Models. Math. Probl. Eng.
**2015**. [Google Scholar] [CrossRef] [Green Version] - Rezaeianzadeh, M.; Kalin, L.; Anderson, C. Wetland Water-Level Prediction Using ANN in Conjunction with Base-Flow Recession Analysis. J. Hydrol. Eng.
**2015**, 22, 1–11. [Google Scholar] [CrossRef] [Green Version] - Rezaeianzadeh, M.; Kalin, L.; Hantush, M. An Integrated Approach for Modeling Wetland Water Level: Application to a Headwater Wetland in Coastal Alabama, USA. Water
**2018**, 10, 879. [Google Scholar] [CrossRef] [Green Version] - Nakdong River Basin Environmental Office. Conservation Plan of Upo Wetland Reservation Area; Nakdong River Basin Environmental Office: Changwon, Korea, 2016.
- Ministry of Construction & Transportation. A Study on Improving the Ecological-Flood Function in Upo Wetland; Ministry of Construction & Transportation: Sejong, Korea, 2007.
- Cover, T.; Thomas, J. Elements of Information Theory; John Wiley & Sons: New York, NY, USA, 1991. [Google Scholar]
- Ross, B.C. Mutual information between discrete and continuous data sets. PLoS ONE
**2014**, 9, e87357. [Google Scholar] [CrossRef] - Kinney, J.B.; Atwal, G.S. Equitability, mutual information, and the maximal information coefficient. PNAS
**2014**, 111, 3354–3359. [Google Scholar] [CrossRef] [Green Version] - Yaseen, Z.M.; El-Shafie, A.; Jaafar, O.; Afan, H.A.; Sayl, K.N. Artificial inteligence based models for stream-flow forecasting: 2000–2015. J. Hydrol.
**2015**, 530, 829–844. [Google Scholar] [CrossRef] - Rosenblatt, F. The perceptron: A probabilistic model for information storage and organization in the brain. Psychol. Rev.
**1958**, 65, 386. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Choi, C.; Kim, J.; Kim, J.; Kim, H.S. Development of Combined Heavy Rain Damage Prediction Models with Machine Learning. Water
**2019**, 11, 2516. [Google Scholar] [CrossRef] [Green Version] - Choi, C.; Kim, J.; Kim, J.; Kim, D.; Bae, Y.; Kim, H.S. Development of heavy rain damage prediction model using machine learning based on big data. Adv. Meteorol.
**2018**. [Google Scholar] [CrossRef] [Green Version] - Drucker, H.; Burges, C.J.; Kaufman, L.; Smola, A.J.; Vapnik, V. Support vector regression machines. In Advances in Neural Information Processing Systems; MIT Press: Cambridge, MA, USA, 1997; pp. 155–161. [Google Scholar]
- Kim, S.J.; Ryoo, E.C.; Jung, M.K.; Kim, J.K.; Ahn, H.C. Application of support vector regression for improving the performance of the emotion prediction model. J. Intell. Inf. Syst.
**2012**, 18, 185–202. [Google Scholar] - Gunn, S.R. Support vector machines for classification and regression. ISIS Tech. Rep.
**1998**, 14, 5–16. [Google Scholar] - Basak, D.; Pal, S.; Patranabis, D.C. Support vector regression. Neural Inf. Process.-Lett. Rev.
**2007**, 11, 203–224. [Google Scholar] - Moriasi, D.N.; Arnold, J.G.; Van Liew, M.W.; Bingner, R.L.; Harmel, R.D.; Veith, T.L. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Trans. ASABE
**2007**, 50, 885–900. [Google Scholar] [CrossRef] - Hwang, S.H.; Ham, D.H.; Kim, J.H. A new measure for assessing the efficiency of hydrological data-driven forecasting models. Hydrol. Sci. J.
**2012**, 57, 1257–1274. [Google Scholar] [CrossRef] [Green Version] - Lantz, B. Machine Learning with R; Packt Publishing Ltd.: Birmingham, UK, 2015. [Google Scholar]
- Baba, N. A new approach for finding the global minimum of error function of neural networks. Neural Netw.
**1989**, 2, 367–373. [Google Scholar] [CrossRef] - Lewis, N.D.C. Deep Learning Made Easy with R: A Gentle Introduction for Data Science; CreateSpace Independent Publishing Platform: Seattle, WA, USA, 2016. [Google Scholar]
- Falas, T.; Stafylopatis, A.G. The impact of the error function selection in neural network-based classifiers. Int. Jt. Conf. Neural Netw.
**1999**, 3, 1799–1804. [Google Scholar] - Cory, L. Mastering Machine Learning with R; Packt Publishing Ltd.: Birmingham, UK, 2017. [Google Scholar]

**Figure 3.**Time series of water level of the Upo wetland over time. Training period is from 1 April 2009 to 31 March 2013 and test period is from 1 April 2013 to 31 March 2015.

**Figure 10.**(

**a**) Relative error by complexity parameter; (

**b**) Level prediction model using decision tree.

**Figure 12.**Correlation coefficient (CC), Nash–Sutcliffe efficiency (NSE), and RMSE values for machine learning models. The dotted lines indicate the average values of each index. The gray rectangles refer to the models showing the best performance.

**Figure 13.**Time series of simulated and observed water level for test period (from 1 April 2013 to 31 March 2015). Four peak values (6 July 2013, 25 August 2013, 11 October 2013 and 22 August 2014) were selected for peak level evaluation.

Water Level | Training Period | Test Period |
---|---|---|

<3 m | 90.91% | 92.54% |

3–4 m | 7.04% | 5.94% |

>4 m | 2.05% | 1.52% |

Variable | Description | Variable | Description | Variable | Description |
---|---|---|---|---|---|

X1.1 | Average temperature (1 day ago) | X1.2 | Average temperature (2 days ago) | X1.3 | Average temperature (3 days ago) |

X2.1 | Minimum temperature (1 day ago) | X2.2 | Minimum temperature (2 days ago) | X2.3 | Minimum temperature (3 days ago) |

X3.1 | Maximum temperature (1 day ago) | X3.2 | Maximum temperature (2 days ago) | X3.3 | Maximum temperature (3 days ago) |

X4.1 | Precipitation (1 day ago) | X4.2 | Precipitation (2 days ago) | X4.3 | Precipitation (3 days ago) |

X5.1 | Maximum instantaneous wind speed (1 day ago) | X5.2 | Maximum instantaneous wind speed (2 days ago) | X5.3 | Maximum instantaneous wind speed (3 days ago) |

X6.1 | Average wind speed (1 day ago) | X6.2 | Average wind speed (2 days ago) | X6.3 | Average wind speed (3 days ago) |

Z1.1 | Water level of Shindang (1 day ago) | Z1.2 | Water level of Shindang (2 days ago) | Z1.3 | Water level of Shindang (3 days ago) |

Z2.1 | Water level of Mokpo (1 day ago) | Z2.2 | Water level of Mokpo (2 days ago) | Z2.3 | Water level of Mokpo (3 days ago) |

Node | Average | Standard Deviation | Node | Average | Standard Deviation |
---|---|---|---|---|---|

1 | 0.19 | 0.08 | 6 | 0.15 | 0.03 |

2 | 0.15 | 0.05 | 7 | 0.16 | 0.03 |

3 | 0.15 | 0.03 | 8 | 0.16 | 0.02 |

4 | 0.15 | 0.03 | 9 | 0.16 | 0.02 |

5 | 0.15 | 0.02 | 10 | 0.16 | 0.03 |

Model | PI | Model | PI |
---|---|---|---|

DT | −0.62 | ANN5 | −0.25 |

RF | 0.19 | ANN6 | −0.18 |

SVM | −0.40 | ANN7 | −1.85 |

ANN1 | −1.21 | ANN8 | −2.45 |

ANN2 | −0.84 | ANN9 | −1.33 |

ANN3 | −0.63 | ANN10 | −1.37 |

ANN4 | −1.32 |

**Table 5.**Differences between peak levels derived from the four machine learning (ML) techniques and observation.

No | Date | ANN | DT | RF | SVM | ||||
---|---|---|---|---|---|---|---|---|---|

Peak (%) | Time (day) | Peak (%) | Time (day) | Peak (%) | Time (day) | Peak (%) | Time (day) | ||

1 | 6 July 2013 | −2.2 | 0 | −9.6 | 0 | −5.8 | 0 | −10.1 | 0 |

2 | 15 August 2013 | 5.4 | 0 | 16.1 | 0 | 10.9 | 0 | 3.6 | 0 |

3 | 11 October 2013 | −3.4 | −2 | −6.5 | 0 | 0.6 | 0 | −2.5 | −2 |

4 | 22 August 2014 | −6.6 | 1 | −18.9 | 3 | −10.9 | 1 | −20.5 | −2 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Choi, C.; Kim, J.; Han, H.; Han, D.; Kim, H.S.
Development of Water Level Prediction Models Using Machine Learning in Wetlands: A Case Study of Upo Wetland in South Korea. *Water* **2020**, *12*, 93.
https://doi.org/10.3390/w12010093

**AMA Style**

Choi C, Kim J, Han H, Han D, Kim HS.
Development of Water Level Prediction Models Using Machine Learning in Wetlands: A Case Study of Upo Wetland in South Korea. *Water*. 2020; 12(1):93.
https://doi.org/10.3390/w12010093

**Chicago/Turabian Style**

Choi, Changhyun, Jungwook Kim, Heechan Han, Daegun Han, and Hung Soo Kim.
2020. "Development of Water Level Prediction Models Using Machine Learning in Wetlands: A Case Study of Upo Wetland in South Korea" *Water* 12, no. 1: 93.
https://doi.org/10.3390/w12010093