Jia Y.1, G. Ni2, T. Kinouchi3, J. Yoshitani4, Y. Kawahara5 and T. Suetsugi6
1JST Research Fellow, Public Works Research Institute, Japan*
2Visiting Researcher, Public Works Research Institute, Japan
3Chief Researcher, Urban River Division, Public Works Research Institute, Japan
4Head of Urban River Division, Public Works Research Institute, Japan
5Prof., Dept. of Safety Systems Construction Engr., Kagawa Univ., Japan
6Head of River Division, Public Works Research Institute, Japan
*Correspondence: Yanwen JIA, JST Research Fellow, Urban River Division,
Public Works Research Institute, Ministry of Construction
Asahi 1, Tsukuba, Ibaraki 305-0804, Japan
Tel. 81-298-642211 ext.4233, Fax 81-298-641168, E-mail jia55@pwri.go.jp
Abstract: A distributed hydrological model - WEP (Water and Energy transfer Process) model is improved at first by adding components of overland flow routing and storm-water detention pond simulation. The modified model is then utilized to assess the flood-reduction effects of storm-water detention ponds and infiltration trenches in the Ebi river watershed (27 km2). Adopting a design storm hyetograph of a central-concentrated shape, 4 cases, namely, with ponds, with trenches, with ponds and trenches, and without ponds and trenches are studied at 6 sites over the watershed. Through comparing the discharge hydrographs at the 6 sites in the 4 cases, it is found that the storm-water detention ponds play much bigger roles than infiltration trenches in reducing flood peaks in upstream reaches, whereas infiltration trenches distributed over the watershed play dominant roles in downstream reaches. The joint installations of storm-water detention ponds and infiltration trenches are thought to be effective for flood damage mitigation in the whole watershed.
Keywords: hydrological cycle, urbanization, Ebi river, distributed model, storm-water detention pond, infiltration trench
Hydrological cycles in urban watersheds are greatly changed with urban developments, leading to larger discharges and shorter concentration times during flood periods. On the other hand, discharges become much less and water quality deteriorates during dry seasons. Installations of infiltration facilities, storm-water detention ponds and sewers are believed to have mitigation effects against these issues.
Jia et al. (2000) developed a distributed hydrological model, WEP (Water and Energy transfer Process) model, and applied it to study the impact of urban development and the installation effect of infiltration trenches on water budgets in the Ebi river watershed. On the other hand, the rational formula is usually adopted for designing inflow peaks into storm-water detention ponds while correct determination of runoff coefficients and concentration times are not so easy.
In this study, effects of storm-water detention ponds on reducing flood peaks are assessed by using the distributed WEP model that is improved by adding components of overland flow routing and storm-water detention pond simulation. Adopting a central-concentrated rainfall event with a duration time of 24 hours and a maximum intensity of 50mm/hour, 4 cases, namely, with ponds, with trenches, with ponds and trenches and without ponds and trenches are studied at 6 sites over the watershed.
The WEP model was developed to simulate spatially variable water and energy processes in watersheds with complex land covers. The model structure within a grid cell is shown in Fig.1 (a). Land use is divided into 3 groups, namely a water body group, a soil-vegetation group and an impervious area group. The soil-vegetation group is further classified into bare soil, tall vegetation (forest or urban trees) and short vegetation (grass or crops). The impervious area group consists of impervious urban cover and urban canopy. The areal average of water and heat fluxes from all land uses in a grid cell produces the averaged fluxes in the grid cell. For the soil-vegetation group, 9 vertical layers, namely an interception layer, a depression layer, 3 upper soil layers, a transition layer, an unconfined aquifer and 2 confined aquifers, are included in the model structure.
Concerning the horizontal structure, the WEP model so far conducted river flow routing for every tributary and a main river by using the kinematic wave method and simplified overland flow as lateral inflow to rivers. To accurately predict river hydrographs during flood periods the WEP model is improved by adding a component of overland flow routing with a time interval of 10 minutes in this study. A schematic illustration of the horizontal structure after the improvement is shown in Fig.1 (b). After deriving the flow direction of every grid cell according to elevation data and limits of urban infrastructures, overland flow is routed from the grid cells in upstream area to those in downstream area or adjacent rivers using the kinematic wave method in the scheme of 1-D sheet flow. With the lateral inflows from adjacent grid cells, river flow routing is conducted for every tributary and a main river using the dynamic wave method in 1-D scheme. In addition, a two-dimensional simulation of multi-layered aquifers, i.e., a quasi-3D simulation of groundwater flow is performed using the Boussinesq equation.
Calculations of infiltration trenches and storm-water detention ponds are carried out as follows.
Calculation of an infiltration trench (Herath 1994):
(1)
(2)
(3)
(4)
where St is the storage in an infiltration trench, Qsr the surface runoff in a grid cell, Qinf the infiltration to the soil below the trench, Qovf the overflow from the trench, n the porosity of filled material in the trench, L the trench length, W the trench width, H the trench depth, Hm the maximum design depth, K0 the saturated hydraulic conductivity of the soil below the trench, a, b and c the constants.
Calculation of a storm-water detention pond:
(5)
where V is the storage in a pond, Qin the inflow that comes from rivers or overland flows and Qovf the outflow from an outlet, a pump and/or a spillway.
The map of the Ebi River watershed is shown in Fig.2. It is located in the Chiba prefecture of Japan with an area of 27 km2. There are 6 rain gauges within the watershed and its surrounding area, one of which also has the hourly observations of temperature, wind and sunshine, and 2 environmental monitoring stations with the hourly observation of relative humidity. The Thiessen method is used to estimate the meteorological data for each grid. In addition, there are 2 gauges of river water stage and discharge, one of which is the Yasakaebashi station of the Ebi main river. The land use and elevation data with a grid resolution of 10m by 10m are based on the Fine Digital Information System (FDIS), Japan Geographical Survey Institute. There are 4 kinds of soils considered in the study, the dominant ones being Kanto-loam and alluvial soil. The geological boring data indicates that the aquifers in the watershed have a multi-layered structure.
The present model is applied to the Ebi River watershed with a grid cell size of 50m and variable time steps (10 minutes for overland and channel flows, and 1 hour for the other processes). Soil parameters are referred to Herath et al. (1992) and vegetation parameters are referred to Dunn and Mackay (1995) etc. though some parameter adjustments are made during calibrations. The model is at first warmed up and calibrated using the data in 1992 and then verified using the data from 1993 to 1997. Fig.3 shows the comparison of hourly discharges at the Yasakaebashi station. It can be seen that the simulated discharges show overall agreement with the observed ones even though no fitting was made.
Though some infiltration trenches had been installed in the watershed, because the amount was small and the concerned data were not obtained, an assumed scenario is used to study installation effects of infiltration trenches. In the assumed scenario, infiltration trenches are only installed for the runoff from urban canopies (building roofs). The guideline for infiltration trench installations promulgated by the Japan Association for Rainwater Storage and Infiltration Technology recommends that the land slope be lower than 10%, the soil not be clay, the ground water table be 2 m or deeper from the land surface, and the trench density be less than 450m/ha etc. In the Ebi river watershed, the urban canopies account for about 20% of the whole area. Fig.4 shows the computed infiltration trench length in every grid cell in accordance with the guideline.
In the Ebi river watershed, several ponds had been constructed for storm-water detention since the 1980s and there is one now under construction. In this paper, 2 ponds (see Fig.2) are considered. The pond 1 with a command area of 0.33 km2 is located in the upstream of Miyamae tributary for regulating the storm water from a residential area (see Fig.5). And the Pond 2 with a command area of 1.28 km2 is located beside the Nagatsu tributary for regulating its river flow. Design and operation data of the 2 ponds were obtained from Urban River Division of Chiba Prefecture and Urban Infrastructure Development Corporation, Japan.
Assuming a central-concentrated storm hyetograph with a duration time of 24 hours and a maximum intensity of 50mm/hour (return period 8.3 year), discharge hydrographs at 6 sites, i.e., 2 pond outlets, 2 tributary downstream sites, Ebi river middle-stream and Ebi river mouth are compared for 4 cases, namely, without any countermeasures, with the ponds, with infiltration trenches and with both ponds and trenches, and shown in Fig.6. Flood peak reduction effects of ponds and trenches at the 6 sites are compared in Fig.7. It can be seen that the storm-water detention ponds play big roles in reducing the flood peaks of tributaries whereas the infiltration trenches have obvious effects on cutting the flood peaks at the main river. Therefore, the joint installations of infiltration trenches and storm-water detention ponds are thought to be effective to reduce flood damage risks throughout the whole watershed.
In this study, the WEP model was improved for flood simulation by adding overland flow routing and it was then verified through comparing the simulated discharge with the observed one. Using the model, the effects of infiltration trenches and storm-water detention ponds were assessed. It was found that: (1) the storm-water detention ponds play big roles in reducing the flood peaks of tributaries but their effects become smaller in the downstream sites, (2) the infiltration trenches have less effects than ponds on reducing the flood peaks of tributaries but their effects are more obvious than ponds on reducing the flood peak at the main river mouth, (3) the joint installations of infiltration trenches and storm-water detention ponds are suggested for effectively reducing flood damage risks throughout the whole watershed.
Acknowledgements
Thanks are given to Prof. K. Musiake, Tokyo University, and Prof. Y. Takahashi, Chiba Institute of Technology, Japan for providing the observed discharge data of the Ebi river.
References
[1] Dunn, S. M., and Mackay, R. (1995). “Spatial variation in evapotranspiration and the influence of land use on catchment hydrology.”J. Hydro., 171, 49-73.
[2] Herath, S., Musiake, K., and Hironaka, S. (1992). “Field estimation of saturated conductivity using borehole test, influence of unsaturated flow and soil anisotropy.” Ann. J. Hydraul. Eng., JSCE, 36, 435-440.
[3] Herath, S. (1994). “Design of storm water storage and infiltration system.” Rain Water Technology (in Japanese), ARSIT, 12, 131-139.
[4] Japan Association for Rainwater Storage and Infiltration Technology (1998). “The guideline for Rainwater infiltration facilities (draft).” In Japanese.
[5] Jia, Y., Ni, G., Kawahara, Y., and Suetsugi, T. (2000). “Simulation of hydrological cycle in an urban catchment and effect evaluation of infiltration facilities.” Ann. J. Hydraul. Eng., JSCE, 45, 31-36.