Simulations of low impact development designs using the storm water management model

Jiyi Jang; Soobin Kim; Sung Min Cha; JongCheol Pyo; Kwang-Sik Yoon; Hankyu Lee; Yongeun Park; In-Su Jang; Kwang-Ju Kim; Jung-Won Yu; Mi-Seon Kang; Hyan-Su Bae; Sang-Soo Baek; Kyung Hwa Cho

doi:10.4491/eer.2023.712

Environ Eng Res > Volume 29(6); 2024 > Article

Jang, Kim, Cha, Pyo, Yoon, Lee, Park, Jang, Kim, Yu, Kang, Bae, Baek, and Cho: Simulations of low impact development designs using the storm water management model

Research

Environmental Engineering Research 2024; 29(6): 230712.

Published online: March 31, 2024

DOI: https://doi.org/10.4491/eer.2023.712

Simulations of low impact development designs using the storm water management model

Jiyi Jang^1,^*, Soobin Kim^2,^*, Sung Min Cha³, JongCheol Pyo⁴, Kwang-Sik Yoon⁵, Hankyu Lee⁶, Yongeun Park⁶, In-Su Jang⁷, Kwang-Ju Kim⁷, Jung-Won Yu⁷, Mi-Seon Kang⁷, Hyan-Su Bae⁷, Sang-Soo Baek^8,^†

, Kyung Hwa Cho^9,^†

¹Division of Ocean and Atmospheric Sciences, Korea Polar Research Institute (KOPRI), Incheon 21990, Republic of Korea

²Disposal Safety Evaluation R&D Division, Korea Atomic Energy Research Institute (KAERI), 111, Daedeok-daero 989 beon-gil, Yuseong-gu, Daejeon 34057, Republic of Korea

³Jeollanam-do Environmental Industries Promotion Institute (JEIPI), Jeollanam-do 59205, Republic of Korea

⁴Department of Environmental Engineering, Pusan National University, Busan 46241, Republic of Korea

⁵Department of Rural and Biosystems Engineering & Education and Research Unit for Climate-Smart Reclaimed Tideland Agriculture, Chonnam National University, Gwangju 61186, Korea

⁶School of Civil and Environmental Engineering, Konkuk University, Gwangjin-gu, Seoul 05029, Republic of Korea

⁷Daegu-Gyeongbuk Research Center, Electronics and Telecommunications Research Institute (ETRI), Daegu, South Korea8Yeungnam University, Gyeongbuk, South Korea

⁸Department of Environmental Engineering, Yeungnam University, Gyeongsan 38541, Republic of Korea

⁹School of Civil, Environmental and Architectural Engineering, Korea University, Seoul 02841, South Korea

^†Corresponding author: E-mail: khcho80@korea.ac.kr (K.H.C), ssbaek@yu.ac.kr (S.-S.B.), Tel: +82-2-3290-3313 (K.H.C), +82-53-810-2542 (S.-S.B.)

* These authors contributed equally to this work.

Received November 20, 2023 Revised March 14, 2024 Accepted March 30, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This study assessed the U.S. Environmental Protection Agency’s Storm Water Management Model (SWMM) for urban water management challenges. This study conducted a sensitivity analysis to identify the most influential factors in the SWMM. Moreover, the performance of SWMM was evaluated with the HYDRUS-1D module in simulating infiltration rates. The sensitivity results showed field capacity as the most significant factor, highlighting the need for advanced modeling techniques to consider factors like field capacity. The SWMM was evaluated by the HYDRUS-1D that SWMM consistently underestimated peak infiltration rates and commenced infiltration calculations only when soil moisture exceeded field capacity. It reveals its limitations in handling unsaturated soil conditions and highlights the consideration of the matric head of the soil during the infiltration calculation in soil media. Moreover, the evaluation of bioretention areas showed larger areas resulting in more substantial flow reductions but with significant variability under different rainfall conditions. Accordingly, this result emphasizes the importance of careful consideration for environmental factors in bioretention design. This study contributes by enhancing understanding of SWMM’s limitations in simulating urban water management challenges. Thus, this research will offer technical assistance to stakeholders focused on challenges such as runoff, hydrologic cycle, and urban flooding in urban areas.

Keywords: Low-impact development, Sensitivity analysis, Storm water management model, Urban modeling

Graphical Abstract

Keywords: Low-impact development, Sensitivity analysis, Storm water management model, Urban modeling

1. Introduction

Urban management needs to address water-associated challenges, such as accelerating runoff and changes in the hydrologic cycle [1, 2]. Urban flooding, when sewer water overflows, not only disrupts the city but also poses health risks due to the release of pollutants [3, 4]. Temperate regions, such as Korea, affected by the East Asian monsoon during summer, experience approximately 60% of their annual precipitation during this season [5, 6]. This has increased public interest in urban water sustainability and planning [7]. In response, the implementation of low-impact development (LID) has been discussed for several decades as a means to transition from conventional pipe-based urban drainage systems to designs focused on storage and infiltration [8–10]. This strategy includes maintaining the hydrological functionality of pre-development and promotes societal benefits such as biodiversity, climate resilience, and cultural amenities [10, 11]. Governments and researchers have made efforts to provide guidelines associated with LID using monitoring and modeling approaches [12, 13]. Modeling approaches can be used for the effective implementation of LID [14, 15].

Numerical models related to LID have been developed by several research groups. These models include SWMM [16], L-THIA-LID [17], MUSIC [18], and SUSTAIN [19]. They are capable of simulating the impact of LID applications on the hydrologic cycle and water quality and have been extensively used to assess the effectiveness of LID in urban areas [15, 20, 21]. These models simulate LIDs by incorporating a combination of one to three soil or ground layers, allowing for calculations of detention storage, infiltration, and evaporation [16, 22, 23]. However, the existing numerical models related to LID are oversimplified when it comes to representing the hydrologic processes involved in each LID practice. For example, soil moisture and matric head are representative contributory factors of oversimplification. Soil moisture is a major component of the hydrologic environment, controlling infiltration, runoff, and evaporation [24, 25]. Moreover, the soil matric head is the leading force responsible for water movement in the soil [26]. The simplified approaches without considering soil moisture and matric heads result in errors in simulated water movement in soil media [27]. Accordingly, the LID models require sensitivity analysis and efficiency evaluation. It enables the LID models to focus on refining those factors for better model performance but also to provide valuable insights into their effectiveness in mitigating urban water-related challenges. Indeed, previous studies have predominantly focused their evaluation of LID performance on factors such as drainage outflow and surface runoff [8, 9, 21].

In this study, we aimed to evaluate the U.S. Environmental Protection Agency (EPA) storm water management model (SWMM) in terms of urban water management issues. The objectives of this study were: 1) to identify the most influential factors in the SWMM with the sensitivity analysis and 2) to evaluate the performance of SWMM by the HYDRUS-1D module along with the infiltration rate and underdrain flow under different rainfall-intensity scenarios. This study will provide technical support to policies and guidelines concerned with rapid runoff, hydrologic cycle alterations, and urban flooding in urban landscapes.

2. Methodology

2.1. Storm Water Management Model (SWMM)

2.1.1. Model description

The overall flow conducted in this study is illustrated in a flowchart in Fig. 1. The EPA SWMM was developed to analyze hydrologic and hydraulic processes of storm water from urban areas. The SWMM can be used to simulate sewer systems in detail (including combined sewers, open channels, and irregular natural channels) and runoff quantity and quality from a continuous or single event [16]. This model has versatile applications ranging from research to urban site planning and materials selection [15, 28]. As the awareness of LID has grown, the SWMM has been upgraded to improve the ability to simulate LID hydrologic performance [16].

2.1.2. LID modules

This study utilized the SWMM version 5.1.010 and specifically selected bioretention, one of the LIDs, to evaluate its performance. Bioretention included three soil layers [29, 30] which comprised surface, soil, and storage layers with different hydrologic and hydraulic behaviors [31] [Fig. 2]. The bioretention module in the SWMM is described in Supplementary Materials.

2.2. New Module Evaluation with the HYDRUS-1D

The performance of the SWMM was evaluated using the HYDRUS-1D software (version 4.04). This software was designed to solve the Richards equation using the finite element method. It has been verified globally and applied to verify simulations with simpler soil water flow models [32–34]. It includes tools for estimating the hydraulic parameters of soils using the Rosetta model [34, 35]. In this study, we compared modeling results from the SWMM with those from the HYDRUS-1D software. Descriptions of the HYDRUS-1D model can be found in Supplementary Materials.

The new module evaluation was conducted under 1,872 rainfall scenarios. These scenarios were computed based on total rainfall volume (10, 15, 20 …, 200 mm), duration (60, 120 …, 720 min), and rainfall shape (Huff curves 1–4). Huff curves were used to determine the temporal pattern of rainfall distribution. This approach has been employed in previous studies to generate rainfall distribution over time [20, 36]. These rainfall scenarios were applied to not only sensitivity analysis but also to model evaluations.

2.3. Sensitivity Analysis Under Various Rainfall Scenarios

We conducted a sensitivity analysis for the SWMM using global sensitivity analysis with the Sensitivity Analysis For Everybody (SAFE) toolbox [37]. Sensitivity analysis has served the purpose of elucidating key factors in water management within predictive modeling studies [38–40]. We applied the Morris method to investigate the sensitivity of parameters in this study [41]. The parameter input spaces (X_i) were discretized into the p levels assuming a uniform distribution. The elementary effects (EEs) of the i^th input were calculated using the following equation [42]:

(1)

{EE}_{i} = \frac{[Y (X_{1}, X_{2}, ...., X_{i - 1}, X_{i} + Δ, \dots X_{k}) - Y (X_{1}, X_{2}, \dots, X_{k})]}{Δ}

where Δ= p/[2(p − 1)] and Y was an objective function. The process was repeated for a predefined number of samples (n). In this study, the number of samples was set to 300, based on the literature [43].

The sensitivity metrics involve the mean (μ) and the standard deviation (σ) of each elementary effect (EE) [44]. The indicates the overall influence of a parameter on the output, while the assesses the potential interaction effect with other parameters [42, 45]. The parameter ranges presented in Table 1 were derived from previous studies [30, 34, 46]. For the underdrain system, the parameter ranges were obtained from NCDENR [46], Cho et al. [47], and Gironás et al. [48].

2.4. Model Evaluation Under Various Rainfall Scenarios

The performance of SWMM was evaluated with the root-mean-square error (RMSE) and the coefficient of determination R². The RMSE and the R² values were obtained using the following equations [8, 32]:

(2)

RMSE = \sqrt{\frac{1}{N} \sum_{i = 1}^{N} {(P_{i} - O_{i})}^{2}}

(3)

R^{2} = {[\frac{\sum_{i = 1}^{N} (O_{i} - \bar{O_{ι}}) (P_{i} - \bar{P_{ι}})}{\sum_{i = 1}^{N} {(O_{i} - \bar{O_{ι}})}^{2} \sum_{i = 1}^{N} {(P_{i} - \bar{P_{ι}})}^{2}}]}^{2}

where P_i is the value predicted based on the SWMM. In general, O_i represents the observed value; however, for this study, the predicted value derived from the HYDRUS-1D was used instead of the observed value. The average of the values predicted by the SWMM is denoted as

\bar{P_{ι}}

, while

\bar{O_{ι}}

represents the average of the values predicted by the HYDRUS-1D.

3. Results and Discussion

3.1. Sensitivity Analysis

The sensitivity results of infiltration have been plotted against ranges of rainfall volume [Fig. 3]. The standard deviation (σ) of EEs determines the spread of the distribution of EE values, indicating that these factors can be strongly influenced by other factors [49]. The total rainfall did not substantially affect the sensitivity of the SWMM. Among all the factors, field capacity (θ_FC) showed the highest sensitivities, with a mean (μ) that was a minimum of 6.0 times greater and standard deviations (σ) of the EEs that were a minimum of 4.3 times higher compared to the other factors [Table S1]. Field capacity is the water content held in the soil after excess rainfall or irrigation [50]. The SWMM could generate infiltration when soil moisture content exceeds field capacity and consistently emphasized that field capacity has a significant impact on the predicted performance in SWMM in previous studies [51–53]. Accordingly, the sensitivity analysis offered essential insights into the behavior of the SWMM and its underdrain module in response to key factors in this study. Specifically, the results underscore the significant impact of field capacity on the model output and factor interactions.

Fig. 4 presents the sensitivity analysis results of the underdrain module in the SWMM under rainfall scenarios. The flow exponent (η) had a minimum of 3.7 times higher mean value of EE than the flow coefficient (C). Whereas, the flow coefficient showed a minimum of 1.2 times greater standard deviation than the flow exponent, in the case of 50–100 mm and 150–200 mm [Table S2]. The results indicate that the flow exponent could substantially affect the underdrain discharge of the SWMM, whereas the flow coefficient showed high interaction effects in this model. This knowledge can guide improvements in modeling and optimize the performance of SWMM in stormwater management applications.

3.2. Evaluation of the Performance on Infiltration Sub-Modules

The results of the comparative analysis between SWMM and the HYDRUS-1D model for simulating infiltration rate are shown in Fig. 5 and Fig. S2. Fig. 5(a) shows the result under the unsaturated soil condition driven by rainfall events (130 mm, duration of 60 min, and Huff’s first quartile). Whereas, Fig. 5(b) shows the simulated infiltration rate driven by rainfall of 200 mm for 60 min and a rainfall shape of Huff’s fourth quartile under the saturated soil condition. In both saturated and unsaturated scenarios, the SWMM consistently underestimated peak infiltration rates and tended to increase infiltration earlier than the HYDRUS-1D model. This discrepancy might be caused by the LID module of the SWMM oversimplifying the infiltration method, which does not adequately account for the matric head of soil in infiltration rate calculations. Specifically, the observed disparities in infiltration rate simulation with 0.38 of R² in Fig. 5(a) and Fig. S2(a), highlight the limitations of the SWMM in handling unsaturated soil conditions. Previous studies have shown that higher infiltration rates increase the moving depth of surface moisture content and moisture content in unsaturated soils [54]. The SWMM commences simulating infiltration only when soil moisture exceeds field capacity, neglecting the role of the matric head and any water above the soil surface in the infiltration process. Therefore, SWMM also needs to be supplemented by considering infiltration rates in unsaturated soils.

In Fig. S3, the simulated infiltration rates of the SWMM were compared with the rates simulated using the HYDRUS-1D model obtained from 1872 rainfall scenarios. The SWMM had relatively low values of R² in the rainfall range from 60–180 mm [Fig. S3(a)], indicating a reduced capability to capture unsaturated infiltration dynamics. This is primarily due to its inability to consider the matric head, leading to inaccuracies in predicting the onset of infiltration. The RMSE of the SWMM increased with the amount of rainfall [Fig. S3(b)]. RMSE is a metric indicating the prediction error of a model, and the higher the rainfall, the more likely the infiltration rate predicted by the model will exceed the observed value. For instance, during abnormally high rainfall such as flood events, the infiltration rates also significantly increase. Such findings have been previously reported, leading to a larger disparity between predicted and observed values and inevitably increasing the RMSE. Although the SWMM provided relatively satisfactory simulation results for the infiltration rate in saturated soil, this model could not account for infiltration processes under unsaturated soil conditions.

In addition, comparing the soil moisture simulated with the SWMM and the HYDRUS-1D model is shown in Fig. S4. The R² values between the SWMM and the HYDRUS-1D increased with higher rainfall amounts [Fig. S4(a)]. However, the RMSE decreased after 150 mm of rainfall [Fig. S4(b)], highlighting a complex relationship between rainfall volume and model performance. This difference is probably because the SWMM initiated the simulation of soil media infiltration only when soil moisture content surpassed field capacity. Thus, these results suggest that the model should consider the matric head of the soil during the infiltration calculation in soil media.

These results underscore the need for enhanced modeling approaches and the incorporation of matric head considerations within the SWMM to improve its accuracy, particularly in simulating infiltration processes under unsaturated soil conditions. Furthermore, they emphasize the importance of robust model evaluation metrics when assessing stormwater management models like the SWMM. In the context of climate change, there is a notable shift in global atmospheric characteristics [55, 56]. Particularly, a study projecting annual rainfall in Korea, expects an increase of up to 13.6% by 2090, compared to the period from 2007 to 2030 [57]. Therefore, these insights contribute to the ongoing efforts to advance urban hydrological modeling and more effectively manage stormwater in various environmental conditions and scenarios, including climate change.

3.3. Evaluation of the Performance on Underdrain Sub-modules

Fig. 6 presents the total and peak underdrain flows in the SWMM. The total underdrain flow of both the SWMM tended to increase with increasing volume of rainfall [Fig. 6(a)]. Moreover, the peak flow of the SWMM gradually increased until the rainfall reached 180 mm, stabilizing at approximately 0.8 in/h [Fig. 6(b)]. Although the volume of rainfall was relatively small, the underdrain flow of the SWMM was easily generated, because this flow was affected by the infiltration rate that started earlier as a result of determining the soil flow grounded in the field capacity. The SWMM used a relatively steady infiltration rate in simulating the peak flow under saturated soil conditions without considering ponding depth. The peak flow of the SWMM was also easily generated because of the oversimplified infiltration process determined by the field capacity. Contrary to our results, a prior study simulating SWMM with 8 years of rainfall intensity data in the Middle East reported variations in flooding rate with rising rainfall intensity [4]. This variability was attributed to factors like residues in the sewage system. Thus, our study underscores the need to simulate actual peak flow rates, considering intricate factors such as tidal depth and recognizing the complexity inherent in our specific environmental context.

3.4. Evaluation of the Effectiveness of Bioretenion

For evaluating the hydrologic effect of the bioretenion, the SWMM was applied to the study area where the flow calibration was previously implemented [58]. The bioretention size means the area of an engineered ecosystem designed to maximize rainfall and water storage for plant growth [29]. Fig. 7 presents the reduction rates of the total flow as influenced by bioretention size and various rainfall scenarios. The SWMM showed that as the size of bio-retention areas increases, there is a corresponding increase in the reduction of both total flow and peak flows. However, it is worth noting that the observed standard deviations in the peak flow and total flow within the SWMM outputs were relatively high, especially in scenarios with smaller bioretention sizes and lower rainfall volumes [Fig. 7(a)]. These results reveal significant variability in bioretention effectiveness under different conditions. The rainfall pattern over Asia is strongly influenced by the monsoon [59], making it crucial to address safety concerns in the design of biological conservation systems for managing outflows from extreme storm events [60]. This consideration remains pertinent when the flows exhibit significant variations, even under conditions of lower rainfall, as observed in this study.

These findings emphasize the critical influence of rainfall duration and pattern on the performance of bioretention systems under conditions of low rainfall volumes. The distinctive differences in the reduction effects of bioretention, as revealed by the SWMM, underscore the need for careful consideration of various environmental factors when designing and implementing bioretention facilities. This knowledge is particularly pertinent in the context of LID and urban drainage system design [26]. Moreover, given the wide utilization of the SWMM for the assessment and design of bioretention facilities and alternative urban drainage systems, the study underscores the importance of refining SWMM-based modeling approaches and incorporating environmental variability considerations. For instance, recent studies in water quality prediction have demonstrated improved model performance by integrating learning-based models for autonomous parameter calibration [61]. These contribute to the ongoing efforts aimed at enhancing urban stormwater management strategies, ensuring the effectiveness of bioretention systems, and promoting sustainable urban development practices.

4. Conclusions

In this study, we found the most influential factor for SWMM by conducting a sensitivity analysis. Moreover, the performance of the SWMM was evaluated with the HYDRUS-1D module, considering soil moisture conditions and specific schemes of underdrain pipes under 1,872 rainfall scenarios. The following conclusions were drawn from this study:

Field capacity emerged as the most influential factor in the SWMM modeling for enhanced accuracy.
Both peak infiltration and matric head need consideration in the SWMM for precise simulation, especially in unsaturated soil conditions.
The SWMM requires meticulous consideration of environmental factors such as rainfall when designing bioretention facilities for urban stormwater management.

Supplementary Information

eer-2023-712-Supplementary.pdf

Acknowledgement

This work was supported by the DGIST R&D Program of the Ministry of Science and ICT (2023010400), and partially supported by Konkuk University Researcher Fund in 2021.

Notes

Conflict-of-Interest Statement

The authors declare that they have no conflict of interest.

Author Contributions

J.J. (Ph.D.) and S.K. (Ph.D.) developed the conceptualization and methodology, and wrote the manuscript. S.M.C. (Ph.D.), J.P. (Professor), K.-S.Y. (Professor), H.L. (Ph.D. Student), and Y.P. (Professor) provided valuable research insights into the study and helped to review the manuscript. I.-S.J. (Ph.D.), K.-J.K. (Ph.D.), J.-.Y. (Ph.D.), M.-S.K. (Ph.D.), and H.-S.B. (Ph.D.) provided valuable research insights into the study, reviewed the manuscript, and helped with publishing. S.-S.B. (professor) and K.H.C. (Professor) conceptualized, reviewed, and edited the manuscript.

References

1. Eshtawi T, Evers M, Tischbein B, Diekkrüger B. Integrated hydrologic modeling as a key for sustainable urban water resources planning. Water Res. 2016;101:411–428. https://doi.org/10.1016/j.watres.2016.05.061

2. Weng Q. Modeling Urban Growth Effects on Surface Runoff with the Integration of Remote Sensing and GIS. Environ. Manage. 2001;28(6)737–748. https://doi.org/10.1007/s002670010258

3. Chen Y, Zhou H, Zhang H, Du G, Zhou J. Urban flood risk warning under rapid urbanization. Environ. Res. 2015;139:3–10. https://doi.org/10.1016/j.envres.2015.02.028

4. Hassan WH, Nile BK, Al-Masody BA. Climate change effect on storm drainage networks by storm water management model. Environ. Eng. Res. 2017;22(4)393–400. https://doi.org/10.4491/eer.2017.036

5. Kim HG, Cho KH, Recknagel F. Time-series modelling of harmful cyanobacteria blooms by convolutional neural networks and wavelet generated time-frequency images of environmental driving variables. Water Res. 2023;246:120662. https://doi.org/10.1016/j.watres.2023.120662

6. Abbas A, Park M, Baek S-S, Cho KH. Deep learning-based algorithms for long-term prediction of chlorophyll-a in catchment streams. J. Hydraul. 2023;626:130240. https://doi.org/10.1016/j.jhydrol.2023.130240

7. Childers DL, Pickett STA, Grove JM, Ogden L, Whitmer A. Advancing urban sustainability theory and action: Challenges and opportunities. Landsc. Urban Plan. 2014;125:320–328. https://doi.org/10.1016/j.landurbplan.2014.01.022

8. Li F, Yan X-F, Duan H-F. Sustainable Design of Urban Stormwater Drainage Systems by Implementing Detention Tank and LID Measures for Flooding Risk Control and Water Quality Management. Water Resour. Manage. 2019;33(9)3271–3288. https://doi.org/10.1007/s11269-019-02300-0

9. Tansar H, Duan H-F, Mark O. Catchment-Scale and Local-Scale Based Evaluation of LID Effectiveness on Urban Drainage System Performance. Water Resour. Manage. 2022;36(2)507–526. https://doi.org/10.1007/s11269-021-03036-6

10. Vogel JR, Moore TL, Coffman RR, Rodie SN, Hutchinson SL, McDonough KR, McLemore AJ, McMaine JT. Critical Review of Technical Questions Facing Low Impact Development and Green Infrastructure: A Perspective from the Great Plains. Water Environ. Res. 2015;87(9)849–862. https://doi.org/10.2175/106143015X14362865226392

11. EPA US. Managing Wet Weather with Green Infrastructure: Action Strategy [Internet] Retrieved 5/23, 2015. U.S. Environmental Protection Agency; Cincinnati, OH: 2008. [cited 20 November 2023]. Available from: https://nepis.epa.gov/Exe/ZyPDF.cgi/P1008SI8.PDF?Dockey=P1008SI8.PDF

12. US DoD. Unified Facility Criteria for Low Impact Development [Internet]. Department of Defense (DoD); 2004. [cited 20 November 2023]. Available from: https://www.wbdg.org/ffc/dod/unified-facilities-criteria-ufc/ufc-3-210-10

13. EPA US. Community solution for Stormwater management [Internet]. U.S. Environmental Protection Agency; Cincinnati, OH: 2016. [cited 20 November 2023]. Available from: https://www.epa.gov/sites/default/files/2016-10/documents/draftlongtermstormwaterguide_508.pdf

14. Perez-Pedini C, Limbrunner JF, Vogel RM. Optimal location of infiltration-based best management practices for storm water management. J. Water Resour. Plan. Manag. 2005;131(6)441–448. https://doi.org/10.1061/(ASCE)0733-9496(2005)131:6(441)

15. Elliott AH, Trowsdale SA. A review of models for low impact urban stormwater drainage. Environ. Model Software. 2007;22(3)394–405. https://doi.org/10.1016/j.envsoft.2005.12.005

16. Rossman LA, Huber WC. Storm Water Management Model Reference Manual, Volume I – Hydrology (Revised), EPA/600/R-15/162A [Internet]. U.S. Environmental Protection Agency; Cincinnati, OH: 2016. [cited 20 November 2023]. Available from: https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100NYRA.txt

17. Ahiablame LM, Engel BA, Chaubey I. Representation and evaluation of low impact development practices with L-THIA-LID: An example for site planning. Environ. Pollut. 2012;1(2)1. http://dx.doi.org/ep.v1n2p1

18. Lee JG, Selvakumar A, Alvi K, Riverson J, Zhen JX, Shoemaker L, Lai F-h. A watershed-scale design optimization model for stormwater best management practices. Environ. Model Software. 2012;37:6–18. https://doi.org/10.1016/j.envsoft.2012.04.011

19. Wong TH, Fletcher TD, Duncan HP, Coleman JR, Jenkins GA. A model for urban stormwater improvement: conceptualization. Global Solutions for Urban Drainage. 2002;1–14.

20. Baek S-S, Choi D-H, Jung J-W, Yoon K-S, Cho K-H. Evaluation of a hydrology and run-off BMP model in SUSTAIN on a commercial area and a public park in South Korea. Desalin. Water Treat. 2015;55(2)347–359. https://doi.org/10.1080/19443994.2014.939502

21. Elliott AH, Trowsdale SA, Wadhwa S. Effect of aggregation of on-site storm-water control devices in an urban catchment model. J. Hydrol. Eng. 2009;14(9)975–983. https://doi.org/10.1061/(ASCE)HE.1943-5584.0000064

22. Palla A, Gnecco I, Lanza LG. Compared performance of a conceptual and a mechanistic hydrologic models of a green roof. Hydrol. Process. 2012;26(1)73–84. https://doi.org/10.1002/hyp.8112

23. Versini PA, Ramier D, Berthier E, De Gouvello B. Assessment of the hydrological impacts of green roof: From building scale to basin scale. J. Hydraul. 2015;524:562–575. https://doi.org/10.1016/j.jhydrol.2015.03.020

24. Tan X, Shao D, Gu W, Liu H. Field analysis of water and nitrogen fate in lowland paddy fields under different water managements using HYDRUS-1D. Agric. Water Manage. 2015;150:67–80. https://doi.org/10.1016/j.agwat.2014.12.005

25. Jiménez MN, Pinto JR, Ripoll MA, Sánchez-Miranda A, Navarro FB. Impact of straw and rock-fragment mulches on soil moisture and early growth of holm oaks in a semiarid area. CATENA. 2017;152:198–206. https://doi.org/10.1016/j.catena.2017.01.021

26. Krüger E, Schmidt G, Brückner U. Scheduling strawberry irrigation based upon tensiometer measurement and a climatic water balance model. Sci. Hortic. 1999;81(4)409–424. https://doi.org/10.1016/S0304-4238(99)00030-8

27. Herrada MA, Gutiérrez-Martin A, Montanero JM. Modeling infiltration rates in a saturated/unsaturated soil under the free draining condition. J. Hydraul. 2014;515:10–15. https://doi.org/10.1016/j.jhydrol.2014.04.026

28. Kim S, Abbas A, Pyo J, Kim H, Hong SM, Baek S-S, Cho KH. Developing a data-driven modeling framework for simulating a chemical accident in freshwater. J. Clean. Prod. 2023;425:138842. https://doi.org/10.1016/j.jclepro.2023.138842

29. Roy-Poirier A, Champagne P, Filion Y. Review of bioretention system research and design: past, present, and future. J. Environ. Eng. 2010;136(9)878–889. https://doi.org/10.1061/(ASCE)EE.1943-7870.0000227

30. Rossman LA, Huber WC. Storm Water Management Model Reference Manual, Volume III – Water Quality, EPA/600/R-16/093 [Internet]. U.S. Environmental Protection Agency; Cincinnati, OH: 2016. [cited 20 November 2023]. Available from: https://nepis.epa.gov/Exe/ZyPDF.cgi/P100P2NY.PDF?Dockey=P100P2NY.PDF

31. Qin H-P, Li Z-X, Fu G. The effects of low impact development on urban flooding under different rainfall characteristics. J. Environ. Manag. 2013;129:577–585. https://doi.org/10.1016/j.jenvman.2013.08.026

32. Hilten RN, Lawrence TM, Tollner EW. Modeling stormwater runoff from green roofs with HYDRUS-1D. J. Hydraul. 2008;358(3)288–293. https://doi.org/10.1016/j.jhydrol.2008.06.010

33. Qin H-p, Peng Y-n, Tang Q-l, Yu S-L. A HYDRUS model for irrigation management of green roofs with a water storage layer. Ecol. Eng. 2016;95:399–408. https://doi.org/10.1016/j.ecoleng.2016.06.077

34. Šimůnek J, Van Genuchten MT, Sejna M. The HYDRUS-1D software package for simulating the one-dimensional movement of water, heat, and multiple solutes in variably-saturated media. University of California-Riverside Research Reports. 2005;3:1–240.

35. Chen M, Willgoose GR, Saco PM. Spatial prediction of temporal soil moisture dynamics using HYDRUS-1D. Hydrol. Process. 2014;28(2)171–185. https://doi.org/10.1002/hyp.9518

36. Azli M, Rao AR. Development of Huff curves for Peninsular Malaysia. J. Hydraul. 2010;388(1)77–84. https://doi.org/10.1016/j.jhydrol.2010.04.030

37. Pianosi F, Sarrazin F, Wagener T. A Matlab toolbox for Global Sensitivity Analysis. Environ. Model Software. 2015;70:80–85. https://doi.org/10.1016/j.envsoft.2015.04.009

38. Kwon DH, Hong SM, Abbas A, Park S, Nam G, Yoo J-H, Kim K, Kim HT, Pyo J, Cho KH. Deep learning-based super-resolution for harmful algal bloom monitoring of inland water. GISci. Remote Sens. 2023;60(1)2249753. https://doi.org/10.1080/15481603.2023.2249753

39. Jang J, Abbas A, Kim H, Rhee C, Shin SG, Chun JA, Baek S, Cho KH. Prediction and interpretation of pathogenic bacteria occurrence at a recreational beach using data-driven algorithms. Ecol. Inform. 2023;78:102370. https://doi.org/10.1016/j.ecoinf.2023.102370

40. Kim HG, Cho KH, Recknagel F. Bibliometric network analysis of scientific research on early warning signals for cyanobacterial blooms in lakes and rivers. Ecol. Inform. 2024;80:102503. https://doi.org/10.1016/j.ecoinf.2024.102503

41. Loizu J, Álvarez-Mozos J, Casalí J, Goñi M. Evaluation of TOPLATS on three Mediterranean catchments. J. Hydraul. 2016;539:141–161. https://doi.org/10.1016/j.jhydrol.2016.05.025

42. Morris DJ, Speirs DC, Cameron AI, Heath MR. Global sensitivity analysis of an end-to-end marine ecosystem model of the North Sea: Factors affecting the biomass of fish and benthos. Ecol. Model. 2014;273:251–263. https://doi.org/10.1016/j.ecolmodel.2013.11.019

43. Ge Q, Menendez M. Extending Morris method for qualitative global sensitivity analysis of models with dependent inputs. Reliab. Eng. Syst. Saf. 2017;162:28–39. https://doi.org/10.1016/j.ress.2017.01.010

44. Morris MD. Factorial Sampling Plans for Preliminary Computational Experiments. Technometrics. 1991;33(2)161–174. https://doi.org/10.1080/00401706.1991.10484804

45. Saltelli A, Ratto M, Andres T, Campolongo F, Cariboni J, Gatelli D, Saisana M, Tarantola S. Global sensitivity analysis: the primer. John Wiley & Sons; 2008.

46. NCDENR Stormwater best management practices manual [Internet]. North Carolina Department of Environment and Natural Resources; Raleigh, N.C.: 2007. [cited 20 November 2023]. Available from: http://www.buildhighpoint.com/DocumentCenter/View/285/NC-Stormwater-BMP-Manual-PDF?bidId=

47. Cho S, Kang M, Kwon H, Lee J, Kim S. Evaluation of the effectiveness of low impact development practices in an urban area: non-point pollutant removal measures using EPA-SWMM. J. Korean Soc. Water Environ. 2013;29(4)466–475.

48. Gironás J, Roesner LA, Davis J, Rossman LA, Supply W. Storm water management model applications manual [Internet]. National Risk Management Research Laboratory, Office of Research and Development, US Environmental Protection Agency; 2009. [cited 20 November 2023]. Available from: https://nepis.epa.gov/Exe/ZyPDF.cgi/P100EA7H.PDF?Dockey=P100EA7H.PDF

49. Campolongo F, Braddock R. The use of graph theory in the sensitivity analysis of the model output: a second order screening method. Reliab. Eng. Syst. Saf. 1999;64(1)1–12. https://doi.org/10.1016/S0951-8320(98)00008-8

50. Veihmeyer F, Hendrickson A. The moisture equivalent as a measure of the field capacity of soils. Soil Sci. 1931;32(3)181–194.

51. Fassman-Beck E, Saleh F. Sources and Impacts of Uncertainty in Uncalibrated Bioretention Models Using SWMM 5.1.012. J. Sustain. Water Built Environ. 2021;7(3)04021006. https://doi.org/10.1061/JSWBAY.0000944

52. Limos AG, Mallari KJB, Baek J, Kim H, Hong S, Yoon J. Assessing the significance of evapotranspiration in green roof modeling by SWMM. J. Hydroinformatics. 2018;20(3)588–596. https://doi.org/10.2166/hydro.2018.130

53. Baek S, Ligaray M, Pachepsky Y, Chun JA, Yoon K-S, Park Y, Cho KH. Assessment of a green roof practice using the coupled SWMM and HYDRUS models. J. Environ. Manag. 2020;261:109920. https://doi.org/10.1016/j.jenvman.2019.109920

54. Mohanadhas B, Govindarajan SK. Modeling the sensitivity of hydrogeological parameters associated with leaching of uranium transport in an unsaturated porous medium. Environ. Eng. Res. 2018;23(4)462–473. https://doi.org/10.4491/eer.2017.113

55. Jang J, Park K-T, Yoon YJ, Ha S-Y, Jang E, Cho KH, Lee JY, Park J. Molecular-level chemical composition of aerosol and its potential source tracking at Antarctic Peninsula. Environ. Res. 2023;239:117217. https://doi.org/10.1016/j.envres.2023.117217

56. Jang J, Park J, Park J, Yoon YJ, Dall’Osto M, Park K-T, Jang E, Lee JY, Cho KH, Lee BY. Ocean-atmosphere interactions: Different organic components across Pacific and Southern Oceans. Sci. Total Environ. 2023;878:162969. https://doi.org/10.1016/j.envres.2023.117217

57. Kim BS, Hong SJ, Lee HD. The Potential Effects of Climate Change on Streamflow in Rivers Basin of Korea Using Rainfall Elasticity. Environ. Eng. Res. 2013;18(1)9–20. https://doi.org/10.4491/eer.2013.18.1.009

58. Baek S-S, Choi D-H, Jung J-W, Lee H-J, Lee H, Yoon K-S, Cho KH. Optimizing low impact development (LID) for stormwater runoff treatment in urban area, Korea: Experimental and modeling approach. Water Res. 2015;86:122–131. https://doi.org/10.1016/j.watres.2015.08.038

59. D’ Silva TC, Verma S, Magdaline RM, Chandra R, Khan AA. Environmental resilience and sustainability through green technologies: A case evidence from rural coastal India. Environ. Eng. Res. 2021;27(5)210262–210260. https://doi.org/10.4491/eer.2021.262

60. Greksa A, Grabić J, Blagojević B. Contribution of low impact development practices-bioretention systems towards urban flood resilience: case study of Novi Sad, Serbia. Environ. Eng. Res. 2021;27(4)210125–210120. https://doi.org/10.4491/eer.2021.125

61. Hong SM, Abbas A, Kim S, Kwon DH, Yoon N, Yun D, Lee S, Pachepsky Y, Pyo J, Cho KH. Autonomous calibration of EFDC for predicting chlorophyll-a using reinforcement learning and a real-time monitoring system. Environ. Model Software. 2023;168:105805. https://doi.org/10.1016/j.envsoft.2023.105805

Fig. 1

The overall flow conducted in this study.

Fig. 2

Flow chart of low impact development (LID) simulations in the storm water management model (SWMM): (a) surface layer, (b) soil layer, and (c) storage layer.

Fig. 3

Sensitivity of infiltration-related parameters in the SWMM with different ranges of rainfall volume: (a) 0 50 mm, (b) 50 100 mm, (c) 100 150 mm, and (d) 150 200 mm. The square indicated the mean for the mean and standard deviation of EEs and the error bar showed the standard error for the mean and standard deviation of EEs.

Fig. 4

Sensitivity analysis of underdrain-related parameters in the SWMM based on different rainfall scenarios: (a) 0 50 mm, (b) 50 100 mm, (c) 100 150 mm, and (d) 150 200 mm. The square indicated he mean for mean and standard deviation of EEs. The error bar showed the standard error for mean and standard deviation of EEs.

Fig. 5

Temporal variations in infiltration rate simulated using the SWMM and the HYDRUS-1D models under (a) unsaturated soil conditions and (b) saturated soil conditions.

Fig. 6

Underdrain discharge of the SWMM: (a) total flow and (b) peak flow. The unit of the flows indicate inches per hour.

Fig. 7

The reduction rate of flow depending on LID size (bioretention size) and rainfall scenarios in the SWMM: (a) total flow and (b) peak flow. The color bar and extent of circles indicate the mean and the standard deviation for the reduction rate of the flows, respectively.

Table 1

SWWM LID parameter ranges used in sensitivity analyses.

Model	Module	Parameter	Value	Reference
SWMM	Infiltration module	Porosity	0.43	[16]
		Field capacity	0.26
		Saturated K [cm/min]	0.01733
		K-Slope	35
		Suction head [cm]	8.89
		Soil layer depth [cm]	76

	Underdrain module	Flow Coefficient	1	[16]
		Flow Exponent	0.5	[16]

		Exfiltration rate [cm/min]	8.09*10⁻³	Rural Development Administration, South Korea

		Underdrain layer depth [cm]	25	[16]