Hydrodynamic modeling of the impact of residential rainwater harvesting systems on stormwater runoff and drainage networks

Transcription

1 Hydrodynamic modeling of the impact of residential rainwater harvesting systems on stormwater runoff and drainage networks Agnieszka Stec 1,* 1 Department of Infrastructure and Water Management, Rzeszow University of Technology, al. Powstańców Warszawy 6, Rzeszów Abstract. This paper presents the results of hydrodynamic modeling of urbanized catchment, where rainwater harvesting systems (RWH) was applied. The catchment model was developed in the Storm Water Managment Model program and RWH was simulated as one of the LID practices available in this program - rain barrels (RB). The research was carried out for various rain barrels implementation scenarios (50% -100%) in a single-family housing estate. However, the results of the research showed that the use of rain barrels (RB) to capture rainwater discharged from the roof of buildings was not effective in significantly reducing the outflow of water from the catchment, and thus reducing the occurrence of pressure flows in the analyzed sewage system. 1 Introduction The transformation of natural areas into residential, service and industrial ones causes large environmental and hydrological changes [1-2]. These changes have a negative impact primarily on the quantity and quality of rainwater causing an increase in the speed and volume of runoff, a reduction in infiltration, an increased risk of flooding [3], and the hydraulic overload of sewer systems [4]. Water changes and the functioning of drainage networks are also adversely affected by climate change, especially by an increase in rainfall intensity [5]. In order to limit these negative changes, some actions are taken that consist of applying the principle of sustainable development in the aspect of spatial planning and rainwater management. These include, primarily objects and devices that reduce and delay the time of outflow of water from the basin through their infiltration and retention [6-9]. In recent years, there have been many strategies for rainwater management in urban areas, which, unlike traditional drainage systems, are more environmentally friendly. One of them is the low impact development (LID) strategy, which, through the use of appropriate techniques, aims at restoring the hydrological state of the catchment before its management [10]. Low impact development practices mainly include decentralized devices and objects whose operation is to imitate the natural hydrological processes taking place in the catchment such as infiltration, evaporation and the retention of rainwater. The benefits * Corresponding author: stec_aga@prz.edu.pl The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (

2 resulting from the implementation of LID practices (porous pavement, green roof, rain garden, harvesting system) have been presented in numerous publications [11-21]. One of the LID practices is the collection and use of rainwater as both potable and nonpotable water. The method of using rainwater is influenced by their quality and quantity as well as the methods of their purification [22]. However, in the vast majority of cases rainwater is a source of water for toilet flushing [23-26], cleaning, washing [27], car washing [28], irrigation of green areas and crop irrigation [29-30]. The popularity of RWH systems is increasing all over the world. This is caused not only by water shortages in many countries, but also by an increase in environmental awareness of the society and the introduction of appropriate legal provisions. Especially the latter affects the intensive development of construction with a reduced demand for water and energy [31-35]. The use of RWH systems not only protects natural water resources but also brings financial benefits resulting from the lower consumption of tap water [36]. By reducing the outflow of water from the catchment, it also has a positive effect on the functioning of sewage systems and objects interacting with them. Despite these advantages, rainwater harvesting systems are quite rarely used in Poland. Taking the above into account, research was conducted to determine the impact of RWH application on the volume of rainwater outflow from the catchment and on the functioning of the sewage system. An existing urban catchment was selected for the study, for which a hydrodynamic model was developed in the Storm Water Management Model (SWMM). 2 Materials and methods 2.1. Study area The city of Przemysl is located in the south-eastern part of Poland, on both sides of the San River, which belongs to the European Ecological Natura 2000 Network and is the main receiver of sewage from the city (Figure 1). Its area covers 46 km 2 and is inhabited by over 65,000 inhabitants. Due to the high degree of urbanization, the left-bank part of the city covering the district of Zasanie was selected for the research. There is a combined sewerage system in its area, which in rainy weather conditions significantly affects the operation of wastewater treatment plants and the quality of water in the San River. The main problem in the operation of this sewage system is the occurrence of pressure flows in the main sewage collectors and an excessive number of discharges of storm sewage into the river. Based on the maps and design of the existing sewerage system, a hydrodynamic model of the analyzed catchment was developed, and which is described in detail in [4]. The district of Zasanie model developed in the Storm Water Management Model (SWMM) is shown in fig. 2. The total area of this part of the city is 632,88 ha. It mainly consists of residential and service areas. When analyzing the area of the district of Zasanie using an orthophoto map, an area of over 9 ha was identified in one of the sub-divisions, where 146 single-family residential lots were identified (Figure 2). Then, on these plots, the surface of rooftops, other sealed areas and green areas were measured. The area of the plots ranged from 372 to 1659 m 2, 614 m 2 on average, and the rooftops in area from 64 to 240 m 2, 132 m 2 on average. The detailed data characterizing the analyzed part of the basin, which were input into the SWMM program, are presented in table 1. 2

3 Fig. 1. Location of case study city in Poland. Fig. 2. SWMM model of the district of Zasanie located in Przemysl city with marked analyzed catchment. 3

4 Table 1. Input data for the hydrodynamic model of the analyzed part of catchment. Parameter Value Land surface slope 0.5%-5.0% Manning s coefficient for impervious surfaces Manning s coefficient for pervious surfaces 0.25 (e.g. dense grass) Impervious depression storage 1.5 mm Pervious depression storage 7.0 mm Percent imperviousness: Rooftop Green areas Other impervious surfaces 100% 10%-20% 40%-70% 2.2. RWH as the LID practice in SWMM In the research, the SWMM program in version was applied, which allows the use of low impact development (LID) practices in the catchment which limit the outflow of rainwater and increase their infiltration [37]. These solutions also include rainwater harvesting (RWH), which in the program is modeled as rain barrels (RB) allowing for the temporary retention of rainwater and their use, e.g. for watering. Rain barrels can be modeled as a storage layer with an exhaust valve placed above an impermeable bottom. The outflow of rainwater from the roof is directly into the rain barrels through the gutter and their functioning in the program is described by the equation (1). If the inflow to the tank exceeds its capacity, the excess of water is discharged through the upper overflow. Q out = Q 1 Q 2 Q 3 (1) where Q 1 is the amount of surface inflow captured by the barrel, Q 2 is the barrel overflow and Q 3 is the underdrain outflow from the barrel [10]. SWMM allows the drain valve to be closed prior to a rainfall and then opened at some required number of hours after rainfall finishes. If the valve is closed, then Q 3 would be 0. The volume flow through the underdrain results from the equation (2). The duration of this outflow is controlled by the value of the C coefficient which can be calculated from the equation (3). Q 3 = C(h-H)n (2) where C is the drain coefficient, n is the drain exponent, h is the barrel height and H is the drain offset. C = 0,6(A 3/A 1) 2g 1/2 (3) where A 1 is the surface area of the barrel, A 3 is the area of the drain valve opening and g is the acceleration of gravity. In the modeling of rain barrels in SWMM, an important parameter is drain delay time, which means the time after the precipitation stops and after which the emptying occurs. If a smaller outflow is required, for example when watering the lawn, this can be achieved by leaving the valve partially closed. This situation in SWMM can be simulated using a reduced valve surface at the time of determination and a drain flow coefficient. In SWMM both outflows, top and bottom ones, can be directed to the sewage system, on the pervious surface or to another LID object. In these studies, it was assumed that the underdrain outflow from rain barrels will be directed at the pervious area to simulate the 4

5 watering of greenery around the buildings. The excess water from the barrels will also be discharged via the emergency overflow at the pervious area. A simplified diagram showing the assumed assumption in SWMM is shown in figure 3. Fig. 3. Simplified diagram of rain barrels functioning in SWMM. Due to the fact that in the SWMM program the rain barrels are treated as tanks located on the surface, and taking into account the availability of such solutions in the Polish market, and the fact that the analyzed area was an estate of single-family houses (utility and aesthetic reasons), the barrels included in the tests were of a capacity of 1000 liters. In order to take into account the uncertainty associated with residents' acceptance for installing rainwater barrels at their homes, various RWH implementation scenarios were adopted in the analyzed catchment (Table 2). The rain barrels and drain delay time were also variable parameters in the tests. Table 2. Implementation scenarios of rain barrels in households. Scenarios Implementation Households % Scenario Scenario Scenario Result and discussion The simulation tests conducted for real rainfall data from showed that the use of rain barrels in the analyzed catchment as one of the LID methods would result in slight changes both in the catchment itself and in the sewage system. When analyzing the change in hydrological conditions in the basin, depending on the number of rain barrels used (50% -100%), it was noted that the peak runoff was reduced in the range from 9,7% to 19,8% for scenarios 1 and 3 respectively, in relation to scenario 0 (catchment in the existing state without RB). Shaping the outflow from the catchment, depending on the duration of rainfall for selected rainfall, is shown in figure 4. The assumption that underdrain flow and overflow from the rain barrels will be recirculated back to the pervious surface caused an increase in the volume of rainwater that infiltrated the ground. If RB will be implemented in 50% households in the catchment (Scenario 1), then infiltration will be intensified at 7,3% compared to Scenario 0. This percentage is 10,8 and 14,1 for Scenario 2 and Scenario 3, respectively. 5

6 Fig. 4. Peak runoff analyzed for precipitation from September 5-6, Considering that in the analyzed sewage system there are periodically pressure sewage flows, one of the research objectives was to determine the impact of rainwater harvesting systems on the hydraulic conditions of this network. However, the tests carried out on the hydrodynamic model showed that for the adopted RB parameters their influence on the reduction-outflow hydraulic overload of the sewers was insignificant. The selected research results in this respect are shown in figure 5. The implementation of RB even in all households located in the catchment area under consideration will not eliminate the phenomenon of pressure flows in the main sewers, and will only shorten the time of their occurrence. Fig. 5. Filling in the main sewer discharging sewage from the analyzed basin for precipitation from September 5-6, The studies also analyzed the impact of rain barrels drainage time and drain delay on how they function in the catchment. Extending the RWH emptying time limited the unit's ability to capture subsequent flows and this caused an increase in overflow and an increase in the amount of rainwater discharged as overflow. A similar situation was observed when increasing the delay time of the outflow of water from the barrel. With reference to the base 6

7 size of 24h, the ratio of overflow to drain outflow was 2,82 and the extension of drain delay to 48h resulted in an increase of this ratio to 6,64. However, the shortening of the drain delay to 6h reduced this relation to 1,04. 4 Conclusions The results of the research showed that the use of rain barrels (RB) to capture rainwater discharged from the roof of the building was not effective in a significant reduction of water outflow from the catchment, and thus reducing the occurrence of pressure flows in the sewage system analyzed. Perhaps the use of other LID practice in the modeled catchment, also in combination with RB, would bring the expected results, as shown by research results for other catchments. Similar conclusions, but also in terms of limiting flood phenomena in urban catchments, were drawn by other researchers [38]. An unquestionable advantage of implementing rainwater harvesting systems is the financial benefits resulting from the reduced demand for tap water and the protection of water resources. However, in Polish conditions, there are still no guidelines for the design of these systems and a clearly formulated strategy for rainwater management encouraging residents to use alternative water sources contributing to limiting the use of natural water resources. References 1. R. Gunn, A. Martin, B. Engel, L. Ahiablame. Urban Water J. 9 (4), (2012) 2. M. O'Driscoll, S. Clinton, A. Jefferson, A. Manda, S. McMillan. Water. 2 (3), (2010) 3. C.P. Konrad. Geological Survey Fact Sheet (2014) 4. A. Stec, D. Słyś. Ecol. Chem. Eng. S. 20, (2013) 5. B. Kaźmierczak, A. Kotowski. Theor. Appl. Climatol. 118, (2014) 6. K. Pochwat, D. Słyś, S. Kordana. J. Hydrol. 549, (2017) 7. K. Pochwat, E3S Web of Conferences 17, (2017) DOI: /e3sconf/ M. Starzec, J. Dziopak, D. Słyś, Underground Infrastructure of Urban Areas 4, (2018) 9. M. Starzec, J. Dziopak, Underground Infrastructure of Urban Areas 4, (2018) 10. USEPA (US Environmental Protection Agency) Low Impact Development (LID). A Literature Review Office of Water, Washington, D.C (2000) EPA-841-B L.M. Ahiablame, B.A. Engel, I. Chaubey. Water Air Soil Pollut. 223, (2012) 12. M.E. Dietz. Water Air Soil Pollut. 186, (2007) 13. E.S. Bedan, J.C. Clausen. Journal of the American Water Resources Association, 45 (4), (2009) 14. H. Qin, Z. Li, G. Fu. J. Environ. Manag., 129, (2013) 15. F. Kong, Y. Ban, H. Yin, P. James, I. Dronova. Environmental Modelling & Software 95, (2017) 16. Y. Yang, Ting Fong May Chui. J Environ Manage. 206, (2018) 17. T.F. May Chui, X. Liu, W. Zhan. J Hydrol. 533, (2016) 7

8 18. N.B. Chang, J.W. Lu, T. F. May Chui, N. Hartshorn. Land Use Policy. 70, (2018) 19. K. Eckart, Z. McPhee, T. Bolisetti. Sci Total Environ. 31, (2017) 20. E. Burszta-Adamiak, M. Mrowiec. Water Sci Technol. 68, (2013) 21. Z. Poorova, Z. Vranayova. IOP Conference Series: Materials Science and Engineering: WMCAUS Bristol: IOP Publishing. 245, 1-7, ISSN (2017) 22. G. Markovič, M. Zeleňáková. ICITSEM 2014: International conference on innovative trends in science, engineering and managment, Dubaj, UAE. [Bangalore]: Mudranik Technologies. ISBN , , (2014) 23. D. Słyś, A. Stec. Ecol. Chem. Eng. S 21, (2014) 24. D. Słyś, A. Stec, M. Zeleňáková. Ecol. Chem. Eng. S 19, (2012) 25. G. Markovič, D. Káposztásová, Z. Vranayová. Transactions on Environment and Development. 10, (2014) 26. M.P. Jones, W.F. Hunt. Resour Conserv Recy. 54, (2010) 27. T. Morales-Pinzón, R. Lurueña, X. Gabarrell, C.M. Gasol, J. Rieradevall. Sci Total Environ , (2014) 28. E. Ghisi, D.F. Tavares, V.L. Rocha. Resour Conserv Recy. 54, (2009) 29. J. Devkota, H. Schlachter, D. Apul. J Clean Prod. 95, (2015) 30. Unami K, Mohawesh O, Sharifi E, Takeuchi J, Fujihara M. J Clean Prod. 88, (2015) 31. A. Stec, S. Kordana. Resour. Conserv. Recy. 105, (2015) 32. S. Kordana, D. Słyś. E3S Web of Conferences. 22, (2017) 33. A. Stec, A. Mazur, D. Słyś. E3S Web of Conferences. 22, (2017) 34. A. Mazur, D. Słyś. E3S Web of Conferences. 17, (2017) 35. M. Zelenakova, G. Markovic, D. Kaposztasova, Z. Vranayová. 16th International Conference on Water Distribution System Analysis (WDSA) Location: Bari, ITALY. URBAN WATER HYDROINFORMATICS AND STRATEGIC PLANNING Book Procedia Engineering 89, (2014) 36. A. Stec. E3S Web of Conferences 17, (2017) 37. USEPA, Storm Water Management Model, version with Low Impact Development (LID) Controls. (accessed January 2018) 38. L. Ahiablame, R. Shakya. J Environ Manage. 171, (2016) 8