All Issue

2025 Vol.35, Issue 2 Preview Page

Research Article

Regression Analysis for Predicting Nitrate-N Concentrations in Groundwater in Western Jeju

제주 서부지역 지하수의 질산성질소 농도 예측을 위한 회귀분석

Gyu-Han Kim1
,
Hyeong-Sin Kim2
,
Yong-Seok Seo3*
김 규한1
,
김 형신2
,
서 용석3*
1Deputy Director, Korea Rural Community Corporation
2Lecturer, Department of Earth and Environmental Sciences, Chungbuk National University
3Professor, Department of Earth and Environmental Sciences, Chungbuk National University
1한국농어촌공사 차장
2충북대학교 지구환경과학과 강사
3충북대학교 지구환경과학과 교수
*Corresponding Author
30 June 2025. pp. 231-248
PDF XML Citation
Abstract

This study aims to identify the key controls on nitrate nitrogen (NO3-N) concentrations in groundwater and proposes regression-based models for prediction of the concentrations, focusing on 32 wells in western Jeju Island. Conventional groundwater vulnerability indices such as DRASTIC, DRASTIC_jeju, and WELCUP had limited explanatory power for NO3-N concentrations (adjusted R2 < 0.1). In contrast, multiple regression models combining the size of agricultural area and the distance to pollution source (e.g., 1/d, 1/d2) showed significantly higher predictive performance with Adj. R2 values of ~0.49 and statistically significant p-values (p < 0.05). Error analysis using mean absolute error (MAE), root mean square error (RMSE), and relative excess error (REE) showed that some models maintained stable prediction performance even in the presence of outliers (i.e., high-concentrations of NO3-N). These findings suggest that spatial variables play a critical role in explaining the distribution of NO3-N and that the proposed models could be practically used for groundwater pollution management and identifying nitrate-vulnerable zones in groundwater on Jeju Island.

Keywords
nitrate nitrogen (NO3-N)
groundwater vulnerability
DRASTIC
WELCUP
multiple regression model
error analysis

본 연구는 제주도 서부 지역의 지하수 관정 32공을 대상으로 질산성질소(NO3-N) 농도에 영향을 미치는 주요 요인을 분석하고, 예측 가능한 회귀모델을 제시하고자 하였다. 지하수 오염취약성 평가를 위해 DRASTIC, DRASTIC_jeju, WELCUP 지수를 산정하였으나, 이들 지수만으로는 NO3-N 농도의 예측력이 제한적이었으며(adjusted R2 < 0.1), 통계적 유의성도 확보되지 않았다. 반면, 농경지 면적과 점오염원(개인하수처리시설, 축산폐수배출시설)과의 거리 기반 지표(1/d, 1/d2)를 조합한 회귀모델에서는 Adj. R2 ≈ 0.49 수준의 설명력과 p < 0.05의 유의성을 확보하였다. 또한 평균절대오차(mean absolute error, MAE), 평균제곱근오차(root mean square error, RMSE), 상대초과 오차율(relative excess error, REE)을 활용한 예측오차 분석 결과, 일부 모델은 고농도 이상치가 존재하는 경우에도 예측의 안정성을 유지함을 확인하였다. 이러한 결과는 NO3-N 분포를 설명하는 데 있어 공간적 변수가 핵심적인 역할을 하며, 제안된 회귀모델이 제주도 지하수 오염관리 및 오염취약지역 선별을 위한 실용적인 분석 도구로 활용 가능함을 시사한다.

키워드
질산성질소(NO3-N)
지하수 오염취약성 평가
DRASTIC
WELCUP
회귀모델
예측오차 분석
References
1

Aller, L., Bennet, T., Lehr, J.H., Petty, R.J., and Hacket, G., 1987, DRASTIC: A standardized system for evaluating ground water pollution potential using hydrogeologic settings, US EPA 600/2-87-035.

2

Baek, W.K., Park, S.H., Yu, J.W., Yoon, Y.W., Jung, H.S., 2021, Mapping of groundwater pollution vulnerability in entire South Korea using DRATIC model, Geo Data, 3(2), 32-38 (in Korean with English abstract).

10.22761/DJ2021.3.2.005
3

Choi, S.Y., Han, K.Y., Kim, B.H., 2012, Comparison of different multiple linear regression models for real-time flood stage forecasting, Journal of the Korean Society of Civil Engineers, 32(1B), 9-20 (in Korean with English abstract).

10.12652/Ksce.2012.32.1B.009
4

Han, J.S., 1998, Groundwater environment and contamination, Pakyoungsa, 703-711 (in Korean).

5

Jeong, G.C., 2018, Assessment of groundwater contamination vulnerability by geological characteristics of unsaturated zone, The Journal of Engineering Geology, 28(4), 727-740 (in Korean with English abstract).

10.9720/kseg.2018.4.727
6

JSSGP (Jeju Special Self-Governing Province), 2018, Report for comprehensive management plan for water resources in Jeju Island (in Korean).

7

JSSGP (Jeju Special Self-Governing Province), KRC-J (Jeju Headquarter, Korea Rural Community Corporation), 2020, Report for establishment of groundwater quality measurement network and operation plan (in Korean).

8

Kim, E.Y., Koh, D.C., Ko, K.S., Yeo, I.W., 2008, Prediction of nitrate contamination of groundwater in the northern Nonsan area using multiple regression analysis, Journal of Soil and Groundwater Environment, 13(5), 57-73 (in Korean with English abstract).

9

Kim, Y.I., Lee, K.H., 2023, Accuracy analysis of machine learning methods for predicting PM concentration, Journal of Korean Society for Atmospheric Environment, 39(2), 149-164 (in Korean with English abstract).

10.5572/KOSAE.2023.39.2.149
10

KRC-J (Jeju Headquarter, Korea Rural Community Corporation), 2021, Report on the construction project for groundwater quality monitoring wells (in Korean).

11

Lee, B.H., 2008, Development of a groundwater pollution risk assessment model suitable for the Jeju region, Research Report, 56-84 (in Korean).

12

Lee, H.J., Park, E.G., Kim, K.J., Park, K.H., 2008, A joint application of DRASTIC and numerical groundwater flow model for the assessment of groundwater vulnerability of Buyeo-eup area, Journal of Soil and Groundwater Environment, 13(1), 77-91 (in Korean with English abstract).

13

Lee, S.R., Kim, Y.S., Kim, D.K., 2006, Assessment of regional groundwater pollution hazard using potential pollutant of Pohang area, Journal of the Korean Association of Geographic Information Studies, 9(3), 1-13 (in Korean).

14

Lim, J.J., 2023, Evaluation of compression index of soft clayey ground using regression analysis and machine learning algorithm, Doctoral Dissertation, Chonnam National University, 49-50 (in Korean with English abstract).

15

Piper, A.M., 1944, A graphic procedure in the geochemical interpretation of water-analyses, Eos, Transactions American Geophysical Union, 25(6), 914-928.

10.1029/TR025i006p00914
16

Song, S.H., Hwangbo, D.J., Jang, K.Y., Kim, J.S., Seo, S.K., Yang, W.S., 2021, Evaluation of nitrate nitrogen contamination degree in groundwater wells, Jeju Island, Journal of Soil and Groundwater Environment, 26(4), 8-19 (in Korean with English abstract).

10.7857/JSGE.2021.26.4.008
17

Woo, N.C., 1994, Overview of the groundwater pollution potential assessment method "DRASTIC", Economic and Environmental Geology, 27(6), 611-612 (in Korean).

18

Yoon, J.H., Park, S.H., Choi, H.J., Kim, D.H., Kim, M.S., Yun, S.T., Kim, Y., Kim, H., K., 2020, Analysis of groundwater quality and contamination factors in livestock region, South Korea, Journal of Soil and Groundwater Environment, 25(4), 98-105 (in Korean with English abstract).

10.7857/JSGE.2020.25.4.098
Information
  • Publisher :Korean Society of Engineering Geology
  • Publisher(Ko) :대한지질공학회
  • Journal Title :The Journal of Engineering Geology
  • Journal Title(Ko) :지질공학
  • Volume : 35
  • No :2
  • Pages :231-248
  • Received Date : 2025-05-28
  • Revised Date : 2025-06-23
  • Accepted Date : 2025-06-25
  • DOI :https://doi.org/10.9720/kseg.2025.2.231
Journal Informaiton The Journal of Engineering Geology The Journal of Engineering Geology
Online Submission
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close