Publications

SCI Publications

Waqas M, Kim SM. Seasonal Groundwater Trends and Predictions in Greenhouse Agriculture of Gyeongsangnam-Do Using Statistical and Deep Learning Models. Water. 2026; 18(4):444. https://doi.org/10.3390/w18040444
Waqas, M., & Kim, S. M. (2025). Hybrid Deep Learning Versus Empirical Methods for Daily Potential Evapotranspiration Estimation in the Nakdong River Basin, South Korea. Water, 18(1), 32. https://doi.org/10.3390/w18010032 (Q2, SJR 2024-0.752)
Kim, S. M., Kang, M. S., & Jang, M. W. (2018). Assessment of agricultural drought vulnerability to climate change at a municipal level in South Korea. Paddy and Water Environment, 16(4), 699-714. https://doi.org/10.1007/s10333-018-0661-z (Q2)

KCI Publications

Ha, J. Y., & Kim, S. M. (2025). Spatio-temporal Analysis of Potential Evapotranspiration and Precipitation for Nakdong River Basin. Journal of The Korean Society of Agricultural Engineers, 67(2), 45-54. https://doi.org/10.5389/KSAE.2025.67.2.045
Estimation of Discharge Flow and Pollutant Load After Using Water Curtain Water in Greenhouse Complexes During Winter Season https://scholarworks.gnu.ac.kr/handle/sw.gnu/81451
Lee, S. H., & Kim, S. M. (2024). Spatio-temporal Distribution Characteristics of Chlorophyll-a and Correlation Analysis with Water Quality Factors in the Namgang Dam Region. Journal of The Korean Society of Agricultural Engineers, 66(6), 71-81. https://doi.org/10.5389/KSAE.2024.66.6.071
Baek, M. K., & Kim, S. M. (2023). Time-series analysis and prediction of future trends of groundwater level in water curtain cultivation areas using the ARIMA model. Journal of The Korean Society of Agricultural Engineers, 65(2), 1-11. https://doi.org/10.5389/KSAE.2023.65.2.001
Baek, M. K., & Kim, S. M. (2023). Analysis of groundwater conductivity and water temperature changes in greenhouse complex by water curtain cultivation. Journal of the Korean Society of Agricultural Engineers, 65(6), 93-103. https://doi.org/10.5389/KSAE.2023.65.6.093
Baek, M. K., & Kim, S. M. (2022). Analysis of groundwater level changes near the greenhouse complex area using groundwater monitoring network. Journal of the Korean Society of Agricultural Engineers, 64(6), 13-23. https://doi.org/10.5389/KSAE.2022.64.6.013
Cho, H. K., & Kim, S. M. (2021). Comparison of Generated Loads by Hydroponics of Strawberry, Tomato, and Paprika in Gyeongsangnam-do. Journal of The Korean Society of Agricultural Engineers, 63(5), 73-81. https://doi.org/10.5389/KSAE.2021.63.5.073
Kwak, E. T., & Kim, S. M. (2020). A study on the spatial variation of target water quality and excess rate at 41 stations in nakdong river basin after the total maximum daily loads. Journal of the Korean Society of Agricultural Engineers, 62(6), 97-109. https://doi.org/10.5389/KSAE.2020.62.6.097
Kim, S. R., & Kim, S. M. (2020). Analysis of livestock nonpoint source pollutant load ratio for each sub-watershed in Sancheong watershed using HSPF model. Journal of the Korean Society of Agricultural Engineers, 62(1), 39-50. https://doi.org/10.5389/KSAE.2020.62.1.039
Sun, D., Kim, J., Kim, S., & Jang, M. W. (2020). Analyzing the spatio-temporal trend in TMDL water quality for Gyeongnam using emerging hot spot analysis. Journal of Korean Society of Rural Planning, 26(4), 53-65. https://doi.org/10.7851/ksrp.2020.26.4.053
Jang, M. W., Cho, H. K., & Kim, S. M. (2019). Analysis of a Spatial Distribution and Nutritional Status of Chlorophyll-a Concentration in the Jinyang Lake Using Landsat 8 Satellite Image. Journal of Korean Society on Water Environment, 35(1), 1-8. https://doi.org/10.15681/KSWE.2019.35.1.1
Cho, H. K., & Kim, S. M. (2019). A study on the estimation of irrigation water for sewage treated water reuse for agriculture. Journal of the Korean Society of Agricultural Engineers, 61(2), 97-104. https://doi.org/10.5389/KSAE.2019.61.2.097
Cho, H. K., & Kim, S. M. (2019). Estimation of the Hapcheon dam inflow using HSPF model. Journal of the Korean Society of Agricultural Engineers, 61(5), 69-77. https://doi.org/10.5389/KSAE.2019.61.5.069
Cho, H. K., Lim, H. J., & Kim, S. M. (2018). Comparison of water quality before and after four major river project for water monitoring stations located near 8 weirs in Nakdong River. Journal of agriculture & life science, 52(6), 89-101. https://doi.org/10.14397/jals.2018.52.6.89
Cho, H. K., & Kim, S. M. (2018). Water quality correlation analysis between sewage treated water and the adjacent downstream water in Nakdong River basin. Journal of Korean Society on Water Environment, 34(2), 202-209. https://doi.org/10.15681/KSWE.2018.34.2.202

International/National Conference Procedings

Latest SCI-Pulication

Seasonal Groundwater Trends and Predictions in Greenhouse Agriculture of Gyeongsangnam-Do Using Statistical and Deep Learning Models

Muhammad Waqas and Sang Min Kim*

https://doi.org/10.3390/w18040444

Abstract

Seasonal groundwater (GW) pumping and climatic variability significantly impact the dynamics of greenhouse-dominated agricultural systems, yet quantitative evaluations at the local scale remain limited. This study explores non-parametric statistical and deep learning (DL) models for analyzing seasonal GW trends and predicting GW levels near greenhouse agriculture systems in Gyeongsangnam-do, South Korea. The modified Mann–Kendall (MK) test and Sen’s slope estimator were used to estimate long-term seasonal trends for the summer (wet season) and winter (dry season), based on monthly GW-level time series from six monitoring wells. Findings indicate that seasonal asymmetry is strong (winter trends have greater magnitudes and greater variability than summer trends), and that winter trends are negative (ranging from −0.45 to +1.70 m year−1) and summer trends are positive (ranging from −0.02 to +0.31 m year−1). At Jinju1 and Jinju4, statistically significant increases were observed in both seasons (p < 0.05), whereas at the other stations, weak or non-significant trends were observed due to short records or high variance. Long short-term memory (LSTM) and spatio-temporal graph neural network (STGNN) models were deployed and compared to predict at the GW level. The STGNN was found to be superior to LSTM in terms of R2 (0.799–0.994) and reduced RMSE of up to 64.6, especially in winter, when spatially synchronized pumping is dominant in GW behavior. Despite advanced modeling, there is a serious concern about data limitations. Findings show that combining seasonal trend analysis with spatiotemporal modeling of DLs can significantly enhance knowledge and forecasting of GW dynamics in intensive greenhouse farming.

Latest KCI-Pulication

Prediction and Comparison of Potential Evapotranspiration Using RNN, GRU, and LSTM Models in the Nakdong River Basin

Sang Min Kim

https://doi.org/10.5389/KSAE.2026.68.1.043

Abstract
This study aimed to evaluate the applicability of deep learning-based time-series models for predicting reference evapotranspiration (ET₀), benchmarked against the FAO Penman-Monteith (FAO-PM) method. We developed and compared three representative Recurrent Neural Network (RNN) models—basic RNN, Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM)—using daily meteorological data from 21 observation stations across the Nakdong River Basin in South Korea. The results consistently showed that LSTM and GRU models, which incorporate gate mechanisms, significantly outperformed the basic RNN model in prediction accuracy across all stations. The LSTM model demonstrated the best overall performance, achieving the lowest average Root Mean Square Error (RMSE) of 0.871 mm/day and the highest coefficient of determination (R²) of 0.767 during the test period. The GRU model's performance was nearly equivalent to LSTM’s, making it a computationally efficient alternative. While the relative superiority of the models was consistent, the absolute prediction error varied depending on the distinct climatic characteristics of each station. Accuracy was highest at stations with stable wind conditions, whereas errors increased in coastal areas with strong, variable winds. These findings demonstrate that LSTM and GRU are robust and reliable data-driven methodologies for accurately predicting ET₀ across diverse climate environments, highlighting their high potential as effective tools for agricultural water resource management.

Latest SCI-Publication

Hybrid Deep Learning Versus Empirical Methods for Daily Potential Evapotranspiration Estimation in the Nakdong River Basin, South Korea

Muhammad Waqas and Sang Min Kim*

https://doi.org/10.3390/w18010032

Abstract

This study compares the performance of empirical and hybrid deep learning (DL) models in estimating daily potential evapotranspiration (PET) in the Nakdong River Basin (NRB), South Korea, with the FAO-56 Penman–Monteith (PM) method as a reference. Two empirical models, Priestley–Taylor (P-T) and Hargreaves–Samani (H-S), and two DL models, a standalone Long Short-Term Memory (LSTM) network and a hybrid Convolutional Neural Network Bidirectional LSTM with an attention mechanism, were trained on a meteorological dataset (1973–2024) across 13 meteorological stations. Four input combinations (C1, C2, C3, and C4) were tested to assess the model’s robustness under varying data availability conditions. The results indicate that empirical models performed poorly, with a basin-wide RMSE of 5.04–5.79 mm/day and negative NSE (−10.37 to −13.99), and are therefore poorly suited to NRB. In contrast, DL models achieved significant improvements in accuracy. The hybrid CNN-BiLSTM Attention Mechanism (C1) produced the highest performance, with R2 = 0.820, RMSE = 0.672 mm/day, NSE = 0.820, and KGE = 0.880, which was better than the standalone LSTM (R2 = 0.756; RMSE = 0.782 mm/day). The generalization of heterogeneous climates was also verified through spatial analysis, in which the NSE at the station level consistently exceeded 0.70. The hybrid DL model was found to be highly accurate in representing the temporal variability and seasonal patterns of PET and is therefore more suitable for operational hydrological modeling and water-resource planning in the NRB.

KCI-Pulication

Spatio-temporal Analysis of Potential Evapotranspiration and Precipitation for Nakdong River Basin

Ji Yeon Ha, and Sang Min Kim

https://doi.org/10.5389/KSAE.2025.67.2.045

Abstract

The purpose of this study was to understand the water resource conditions in the Nakdong River basin through a spatio-temporal analysis of rainfall and potential evapotranspiration. The Penman-Monteith method was used to estimate potential evapotranspiration, and the difference between rainfall and potential evapotranspiration (P-PET) was calculated. To compare and analyze the spatial distribution, meteorological data-such as maximum temperature, minimum temperature, wind speed, humidity, and sunshine duration-from 21 weather observation stations (ASOS) in the Nakdong River basin were used, covering the period from 2011 to 2023. This period was standardized to the shortest observation timeframe for comparison purposes. This analysis revealed that the potential evapotranspiration across 21 stations ranged from 885 mm to 1,112 mm, with an average of 936 mm. The P-PET results showed that the southern region, particularly Busan, had abundant water resources, while northern areas like Bonghwa and Andong experienced water deficits. For analyzing the temporal distribution, 13 stations out of the 21, which had longer data periods, were selected, and data from 1973 to 2023 were used. The Mann-Kendall analysis indicated increasing trends in potential evapotranspiration at five stations.

SCI-Publication

Assessment of agricultural drought vulnerability to climate change at a municipal level in South Korea

Sang-Min Kim, Moon-Seong Kang & Min-Won Jang

https://doi.org/10.1007/s10333-018-0661-z

Abstract

This study aimed to analyze the future vulnerability to agricultural drought of the Korean administrative units of cities (Si) and counties (Gun) following the climate change phenomenon. To assess vulnerability quantitatively and to address various physical and socioeconomic data on the occurrence of agricultural drought, principal component analysis (PCA), a multivariate statistical method, was adopted, and a vulnerability index, the regional vulnerability index to agricultural drought (RVIAD), was proposed. RVIAD, having a range from 0.0 to 1.0, was calculated by rescaling the weighted summation of principal component scores. The analysis was performed with all 21 standardized variables in each administrative unit of Si and Gun: 3 sensitivity variables, 8 exposure variables, and 10 adaptation variables. It resulted in four principal components explaining about 85.7% of the total variance, and the third principal component, comprised of only climate variables, was used for applying future climate data from the RCP 8.5 scenario. The districts of Chungchongnam-Do (M1), Jeollabuk-Do (L1), and Jeollanam-Do (L2) were evaluated as having the highest vulnerability to agricultural drought under the climate change scenario, both in the present and in the future. Despite the limitation inherent in the PCA, the approach in this study could reflect different factors other than climate factors on minimizing subjective interruption, and such is expected to contribute to improving the decision-making for diagnosing the drought adaptation capacity in a region and developing measures to mitigate the drought damage.

KCI-Pulication

Estimation of Discharge Flow and Pollutant Load After Using Water Curtain Water in Greenhouse Complexes During Winter Season

이서영, Kim Sang Min

Abstract

In Korea, large greenhouse complexes are commonly established near rivers to ensure stable access to groundwater, which is intensively used during winter water-curtain cultivation to protect crops from low nighttime temperatures. However, most of the pumped groundwater is discharged directly into adjacent streams, raising concerns regarding both groundwater depletion and water-quality deterioration. This study investigated these issues in a major greenhouse cluster located in Sangnam-myeon, Miryang-si, Gyeongsangnam-do, by quantifying return flows and evaluating associated pollutant loads. From November 2024 to March 2025, four monitoring sites were equipped with water-level loggers, velocity sensors, time-lapse cameras, and automatic sampling instruments to capture hydrological and water-quality dynamics. Stage-discharge relationships were developed for each site and applied to construct continuous return-flow time series. The average return flow per greenhouse unit was 15.72 m3/day (range: 9.33-20.97), with peak discharges observed in December and January when heating demand was highest. Water-quality analysis revealed substantial pollutant loads originating from return flows, including SS (110.36 kg/km2 ⋅day), COD (29.16), TN (18.83), and NO3-N (6.38), whereas TP remained consistently low. These findings demonstrate that water-curtain return flows exert strong influences on both stream hydrology and non-point source pollution, underscoring the need for integrated management strategies that jointly consider groundwater use, winter agricultural practices, and downstream water-quality protection.

Page updated

Google Sites

Report abuse