Automated catchment definition for simulations of city-scale stormwater network : a case study to improve Helsinki combined sewer system model
Ashofteh Beyraki, Moein (2023)
Ashofteh Beyraki, Moein
2023
Master's Programme in Environmental Engineering
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-01-17
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202212299814
https://urn.fi/URN:NBN:fi:tuni-202212299814
Tiivistelmä
The growing impact of climate change on the frequency of extreme weather events and the quality of surface water has necessitated more accurate urban stormwater modeling. Establishing stormwater models can save resources and minimize detrimental environmental impacts when the most critical parts of the network are identified first. The EPA Storm Water Management Model (SWMM) is widely used to study runoff in urban areas. Rainfall-runoff modeling with SWMM requires precise characterization of sub-catchments. However, the delineation and parametrization of accurate sub-catchments for large urban areas and a city-scale model is a time-consuming and complex process, which makes it tedious and prone to designers' errors.
Manual catchment delineation and parametrization challenges indicate the need for automated tools to save modelers a significant amount of time and prevent manual errors. Nonetheless, automated methods can only be used if they are proven to demonstrate their ability to provide realistic results. Furthermore, selecting the spatial resolution of the sub-catchments remains a challenge for simulating the models without a high computational burden.
The main objectives of the thesis were to assess methods for automated delineation and parametrization of SWMM sub-catchments for city-scale modeling applications. The target was to avoid manual work as much as possible while keeping the results consistent using varying Geographic Information System (GIS) approaches and literature values. Testing the automated methods was investigated in two main steps. In the first step, four different GIS-based methods are used, namely: the old HSY method, QGIS, GISTOSWMM, and SCALGO. SWMM sub-catchments were created using these methods in four selected case areas within the Helsinki combined sewer network (CSN) in Finland. The methods were compared with each other, focusing on the fluency of the process, hydraulic results, spatial resolution, and the capability to be used in a city-scale model. In addition, the thesis discusses the impacts of using automation, a new imperviousness layer, and varying levels of detail in catchment definition. In the second step, the best method was used for the whole Helsinki CSN for evaluation in an extensive city-scale model.
The results indicate that the SCALGO method can be used to make hydrological models that range from small to city-scale due to its fast and accurate catchment definition, adjustable spatial resolution, and good model performance. It was found that including stormwater inlets in the SCALGO catchment definition method had a minor effect on the hydraulic results. The use of merged sub-catchments with a minimum adjustable area via the SCALGO toolbox was found practical for finding a suitable subcatchments size. Furthermore, the new data (La-serVesi) obtained from an automated imperviousness surface detection model was useful in estimating the sub-catchments imperviousness parameter. The results of this study make it easier to update sub-catchments for city-scale models. In Helsinki, this is particularly interesting as the network is upgraded annually and more separate sewers are built. While this study focused on automated catchment definition methods for city-scale networks, the findings provide in-depth information about SWMM models' automatic implementation for urban catchments without calibration.
Manual catchment delineation and parametrization challenges indicate the need for automated tools to save modelers a significant amount of time and prevent manual errors. Nonetheless, automated methods can only be used if they are proven to demonstrate their ability to provide realistic results. Furthermore, selecting the spatial resolution of the sub-catchments remains a challenge for simulating the models without a high computational burden.
The main objectives of the thesis were to assess methods for automated delineation and parametrization of SWMM sub-catchments for city-scale modeling applications. The target was to avoid manual work as much as possible while keeping the results consistent using varying Geographic Information System (GIS) approaches and literature values. Testing the automated methods was investigated in two main steps. In the first step, four different GIS-based methods are used, namely: the old HSY method, QGIS, GISTOSWMM, and SCALGO. SWMM sub-catchments were created using these methods in four selected case areas within the Helsinki combined sewer network (CSN) in Finland. The methods were compared with each other, focusing on the fluency of the process, hydraulic results, spatial resolution, and the capability to be used in a city-scale model. In addition, the thesis discusses the impacts of using automation, a new imperviousness layer, and varying levels of detail in catchment definition. In the second step, the best method was used for the whole Helsinki CSN for evaluation in an extensive city-scale model.
The results indicate that the SCALGO method can be used to make hydrological models that range from small to city-scale due to its fast and accurate catchment definition, adjustable spatial resolution, and good model performance. It was found that including stormwater inlets in the SCALGO catchment definition method had a minor effect on the hydraulic results. The use of merged sub-catchments with a minimum adjustable area via the SCALGO toolbox was found practical for finding a suitable subcatchments size. Furthermore, the new data (La-serVesi) obtained from an automated imperviousness surface detection model was useful in estimating the sub-catchments imperviousness parameter. The results of this study make it easier to update sub-catchments for city-scale models. In Helsinki, this is particularly interesting as the network is upgraded annually and more separate sewers are built. While this study focused on automated catchment definition methods for city-scale networks, the findings provide in-depth information about SWMM models' automatic implementation for urban catchments without calibration.