Bridging Uncertainty in SWMM Model Calibration: A Bayesian Analysis of Optimal Rainfall Selection

The present study investigated the storm wise runoff collected in farm pond with the runoff estimated by Storm Water Management Model (SWMM) and Soil Conservation Service (SCS-CN) models. The SWMM and SCS-CN models estimated runoff depth storm wise. The runoff depths correspond to the catchment area given the runoff volume from the catchment. The runoff depth estimated from the Storm Water Management Model and Soil Conservation Service model was compared against the depth of runoff estimated from the Water balance model. For small rainfall depths, the runoff estimated from the Storm Water Management Model was at par with the actual runoff volume stored at the pond. It is necessary to know the watershed runoff contribution to the river or streams due to rainfall in order to determine environmental risk or flood potential. In larger rainfall depth, the runoff volume estimated from the SWMM model was less than the stored runoff volume at Farm Pond. The Soil Conservation Service Model gave better results for larger rainfall depth compared to Storm Water Management Model. SWMM was able to simulate runoff depth for small rainfall depths of 2mm. The peak runoff depths were produced by rainfall depths of 35.5mm. Initial abstractions of the study area for antecedent moisture content i.e. AMC I, AMCII and AMCIII are 53.2, 23.91 and 10.43mm, respectively. The comparison showed that both SWMM and SCS-CN models gave better runoff quantification results.

A comprehensive hydrological model, like the storm water management model (SWMM), has been widely used for rainfall-runoff simulation. In recent years, simple and effective modern modeling techniques have also brought great attention to the prediction of runoff with rainfall input. A comparative case study between SWMM and a presently developed fuzzy logic model for the predictions of total runoff within the watershed of Cascina Scala, Pavia in Italy is presented. A data set of 23 events from 2000 to 2003 including with the total rainfall and total runoff are adopted to train fuzzy logic parameters. Other data (1990–1995) with detailed time variations of rainfall and runoff are available for the setup and calibration of SWMM for runoff modeling. Among the 1990–1995 data, 35 independent rainfall events are selected to test the prediction performance of the SWMM and fuzzy logic models by comparing the predicted total runoffs with measured data. Comparisons and performance analyses in terms of the root-mean-squared error and coefficient of efficiency are made between the SWMM and the fuzzy logic model. The predicted total runoffs from either the SWMM or the fuzzy logic model are found to agree reasonably well with the measured data. For large rainfall events, the fuzzy logic model generally outperforms the SWMM unless the modification of the impervious ratio is applied to improve the SWMM results. However, the SWMM can produce the time varying hydrograph whereas fuzzy logic is subject to limitation of the methodology and is unable to generate such an output.

This study investigates the flood mitigation potential of paddy field dams (PFD) in a flood-prone agricultural catchment in central Taiwan using a coupled hydrological-hydrodynamic modeling approach with the Storm Water Management Model (SWMM). Simulations incorporating bi-directional weir structures in paddy field boundaries demonstrate substantial flood peak attenuation, with discharge reductions exceeding 60% in minor drainage channels, while reductions in primary conduits reached 11.5% during a historical storm event and ranged from 10.4% to 17.6% across 10- and 50-year design storm scenarios. Spatial analysis reveals that PFD performance is highly heterogeneous and governed by localized hydrodynamic-topographic coupling. Besides, the retention capacity of PFD is significantly enhanced in low-lying fields near drainage confluences where backwater effects prolong hydraulic interaction. Moreover, a dynamic interaction was identified between the temporal evolution of piezometric pressure within the PFDs and the hydraulic head fluctuations associated with the flood peak in adjacent drainage channels. This hydraulic coupling governs the directionality of transverse weir flow and modulates the effective floodwater storage capacity of the PFDs, thereby exerting a first-order control on their overall flood mitigation performance. These findings demonstrate that the effectiveness of PFDs can be substantially improved when their design and placement account for field elevation gradients and local channel hydraulics, rather than following uniform or spatially arbitrary installation strategies. This study establishes a hydrodynamically informed framework for targeted implementation of decentralized flood mitigation measures, offering a practical alternative to conventional uniform deployment approaches, and highlight the potential of multifunctional paddy field infrastructure to enhance flood resilience. • Evaluates the spatial performance of Paddy Field Dam (PFD) through a sophisticated hydrodynamic model - SWMM. • Assesses hydrodynamics governing bidirectional exchanges between PFD units and drainage channels. • Introduces a strategic framework for the spatial optimization of PFD placement to identify mitigation hotspots. • Demonstrates that PFD performance is sensitive to localized spatial configurations and weir crest elevations.

Stormwater runoff is a major concern in urban areas which is mostly the result of vast urbanization. To reduce urban stormwater runoff and improve water quality, low impact development (LID) is used in urban areas. Therefore, it is vital to find the optimal combination of LID controls to achieve maximum reduction in both stormwater runoff and pollutants with optimal cost. In this study, a simulation-optimization model was developed by linking the EPA Storm Water Management Model (SWMM) to the Multi-Objective Particle Swarm Optimization (MOPSO) using MATLAB. The coupled model could carry out multi-objective optimization (MOO) and find potential solutions to the optimization objectives using the SWMM simulation model outputs. The SWMM model was developed using data from the BUNUS catchment in Kuala Lumpur, Malaysia. The total suspended solids (TSS) and total nitrogen (TN) were selected as pollutants to be used in the simulation model. Vegetated swale and rain garden were selected as LID controls for the study area. The LID controls were assigned to the model using the catchment characteristics. The target objectives were to minimize peak stormwater runoff, TSS, and TN with the minimum number of LID controls applications. The LID combination scenarios were also tested in SWMM to identify the best LID types and combination to achieve maximum reduction in both peak runoff and pollutants. This study found that the peak runoff, TSS, and TN were reduced by 13%, 38%, and 24%, respectively. The optimal number of LID controls that could be used at the BUNUS catchment area was also found to be 25.

The reliability of the low-impact development (LID) module of the widely used Storm Water Management Model (SWMM) for modeling the runoff reduction performance of permeable pavement systems was evaluated through example applications with rainfall data from Atlanta. The method of calculating infiltration through the pavement layers of permeable pavement systems of the LID module was found to be inadequate when the depths of the pavement layer are less than 120 mm and the computational time steps are longer than 30 min. An alternative method of representing permeable pavement systems as equivalent regular subcatchments is proposed. Using this method, the hydrologic operation of permeable pavement systems can be modeled by SWMM or other hydrologic models.

Rainfall records with time steps less than 15 min, which are required for designing storm water sewerage in urban areas, are rarely available throughout the Moroccan territory. The study was conducted with the objective to tackle this problem through the use of design storm structures with time steps that vary from 5 to 15 min. For this purpose, we used two structural families of design storms: (i) design storms derived from Intensity-Duration-Frequency (IDF) curves of Chicago, Composite, Desbordes, Watt, Weibull, Uniform, and Triangular design storm types; (ii) design storms from the Soil Conservation Service (SCS) synthetic rainfall. The selected design storm structures are those adapted to the rainfall distribution of Tangiers experimental basin and their resulted peak flows are comparable to the flow generated using the synthetic rainfall events. We also did sensitivity analysis for the Storm Water Management Model (SWMM) to changes of different design storm structural parameters, which includes storm duration, time steps, and intensity peak position. We evaluated how changes on these parameters affect peak flows and runoff volumes. In addition, Principal Component Analysis (PCA) method was used to validate the sensitivity analysis by identifying key storm parameters that significantly affect peak flows and Runoff volumes. SWMM model calibration results showed that the Desbordres and Chicago structures best fitted observed flow. Moreover, comparison of different SCS synthetic rainfall forms to observed rainfall events retains that of type 1. Furthermore, the Weibull and watt storms have both a shape parameter that has been calibrated by using the observed events. Overall, simulation results confirmed that peak flow was greatly affected by design storm structures, but was not closely correlated with rainfall intensity.

Multiobjective optimization of low impact development stormwater controls

Simulation of the interaction between an urban drainage system (UDS) and surface flow is a challenging issue. Conventional rainfall–runoff models are only able to simulate the flow in UDSs without considering the interaction between surface runoff and UDS. In this study, SWMM (storm water management model), as a well‐known urban drainage simulation model, and a two‐dimensional overland flow model developed based on the cellular automata approach, have been coupled for UDS simulation. In the proposed model, sewer flows are one‐dimensionally simulated using the SWMM model and at the flooded manholes, the surcharge hydrographs are determined. Then the two‐dimensional overland flow model is used for routing the surcharged flows. The coupled model could consider the complex interaction between the drainage system and the overland flow. The proposed model has been used in a study area located in Tehran, capital of Iran. The results show that almost 40% of surcharged flow returns back to the UDS, which caused 1.5% increase in the UDS's outflow volume. This can greatly affect the UDS performance downstream of the study area. The implementation results show the model's capability in simulation of flood behaviour as well as the urban inundated area with acceptable accuracy and much less computational effort. Copyright © 2017 John Wiley & Sons, Ltd.

Scientific simulation calculation and capacity evaluation of urban drainage system can provide theoretical guidance for the adjustment and optimization of urban drainage system planning scheme. Taking the drainage system of the old urban area of Tongshan County as an example, based on the analysis of the present drainage and waterlogging prevention facilities, this paper evaluates the existing drainage network of the city by using the rainstorm runoff management model SWMM (Storm Water Management Model), analyzes the actual drainage capacity, and puts forward some suggestions for the optimization and transformation of the pipe network. The research ideas and methods in this paper are universal and can be used for reference for the evaluation of drainage capacity in other areas.

This study explored the hydrological impacts of urbanization, rainfall pattern and magnitude in a developing catchment. The Stormwater Management Model was parameterized, calibrated and validated in three development phases, which had the same catchment area (12.3 ha) but different land use intensities. The model calibration and validation by using sub‐hourly hydro‐meteorological data demonstrated a good performance of the model in predicting stormwater runoff in the different development phases. Based on the results, a threshold between minor and major rainfall events was identified and conservatively determined to be about 17.5 mm in depth. Direct runoff for minor storm events has a linear relationship with rainfall; however, events with a rainfall depth greater than the threshold yield a rainfall–runoff regression line with a clearly steeper slope. The difference in urban runoff generation between minor and major rainfall events diminishes with the increase of imperviousness. Urbanization leads to an increase in the production of stormwater runoff, but during infrequent major storms, the runoff contribution from pervious surfaces reduces the runoff changes owing to urbanization. Rainfall pattern exerts an important effect on urban runoff, which is reflected in pervious runoff. With the same magnitude, prolonged rainfall events with unvarying low intensity yield the smallest peak flow and the smallest total runoff, yet rainfall events with high peak intensity produce the largest runoff volume. These results demonstrate the different roles of impervious and pervious surfaces in runoff generation, and how runoff responds to rainstorms in urban catchments depends on hyetograph and event magnitude. Furthermore, the study provides a scientific basis of the design guideline sustainable urban drainage systems, which are still arbitrary in many countries. Copyright © 2015 John Wiley & Sons, Ltd.

LID (Low impact Development) has been widely promoted as a measure against urban flooding which has been recently increasing due to urbanization and climate change. However, studies that consider the combined effects of spatially distributing multiple LIDs simultaneously in an urban area of specific size as urban planning scenarios are few. In this study, two cases of past rainfall data (short duration and long duration heavy rainfall), are used to simulate the effects of 3 different urban planning scenarios of distributed LID (i.e. combined LIDs implementable at the 1) building-level, 2) in open spaces, and 3) combined implementation of both 1&2), in addition to simulating for separate implementation of each LID type using the Storm Water Management Model (SWMM). The results show that in the case of short duration heavy rainfall, all 3 LID scenarios provide a greater decrease in the rate of flood volume with comparison to the case of long duration heavy rainfall. This points to the fact that LIDs have a high potential to mitigate flood volumes resulting from short duration heavy rainfall - the type of rainfall which has become more frequent recently. Furthermore, the paper also organizes the LID types according to their implementability under different urban planning scenarios, and shows that such combined use of LIDs results in higher flood mitigation efficacy.

A new tool for automatic calibration of the Storm Water Management Model (SWMM)

Reliability and financial feasibility assessment of a community rainwater harvesting system considering precipitation variability due to climate change

Urban trees play a significant role in reducing surface runoff through crown interception loss and root-enhanced infiltration. However, the effects of tree crown and roots on surface runoff are often simplified in the Storm Water Management Model (SWMM). This study aims to evaluate stormwater benefits of an urban isolated tree combined with different ground covers through interception-infiltration-runoff processes using SWMM. The crown interception process during 15 rainfall events was measured using a weighing lysimeter and used to obtain the net rainfall (i.e., the rainfall amount reaching the ground surface beneath tree crown) that was input into the rain gauge for simulations. Additionally, the optimization of key hydrological parameters in the bio-retention cell was performed to determine an optimal set of parameters to represent the effect of target tree roots on the infiltration process. This optimal set of parameters was validated using experimental results from two irrigation-drainage events, and the results showed that the coefficient of determination (R2) between simulated and measured results exceeded 0.9. According to the proposed simulation method that considers the effects of tree crown and roots on surface runoff, the interception-infiltration-runoff processes for different combinations of an isolated tree with tree pits, permeable pavement, and water-retaining pavement were analyzed during 15 rainfall events. The results revealed that the total surface runoff reduction from the isolated tree was 11.5%. Meanwhile, the simulation results for different combinations were compared to quantify stormwater benefits, including reductions in total and peak surface runoff. Based on rainfall characteristics (total rainfall amount and rainfall intensity) and surface runoff reduction, this study recommends optimal combination of the tree and ground covers to support urban greening design in stormwater management.

In this study, we performed a statistic analysis of rainfall events (the duration of dry days, the average rainfall intensity, the rainfall duration) using rainfall data for 20 years collected from the Kobe Marine Observatory, and also made a pollutant load run-off analysis using a Storm Water Management Model (SWMM). These data were used to examine stochastic characteristics of rainfall that are closely associated with pollution loads. The results of the analyses revealed that the Kobe area has frequent rainfall events with a combination of a rainfall intensity of 4 mm/hr and a rainfall duration of 25 hours. It was also found that the cumulative amount of pollution load runoff caused by a single round of rainfall was often about 5 to 10 kg in the river basins examined. We examined the effect of a storm water storage tank, in terms of reducing the pollution load, considering the stochastic characteristics of rainfall. It was found that a storage tank would be an effective measure for preventing inundation and also help to reduce the pollution load, through the effect of delaying rainfall run-off.