Long-term hydraulic and pollution retention performance of infiltration systems
Infiltration techniques are now widely used to manage stormwater in residential and industrial areas. These techniques are used and recognized for their many advantages, such as decreasing stormwater peak flows, reducing pollution of surface waters, recharging groundwater and enhancing urban landscape when infiltration systems are designed as playgrounds or parks for example. However they are still suspected to be unsustainable on the long term. Over time, two major problems are encountered: clogging and eventual pollution of soil and groundwater.
The aim of the paper is to present a framework to evaluate long-term hydraulic and pollution retention performance of infiltration systems. The main goal of the study presented in the paper is more particularly to improve knowledge and modelling of long-term behaviour and especially the clogging speed in relation to the quality of soil and runoff water. The research is based both on experimental data and on numerical modelling designed to optimise the longevity of such systems, the maintenance and operation costs.
On site-monitoring of operational infiltration systems used for that purpose, is part of the On site Observatory for Urban Hydrology (OTHU) launched in France in 1999. An infiltration basin has been selected and is now continuously monitored for various parameters (water flow, water depth in the different parts of the basin, pH, conductivity, turbidity and temperature with a two minute time step). Pollutant loads are also evaluated from samples taken with refrigerated automatic samplers. Following pollutants are currently analysed: suspended solids, organic descriptors (COD, TOC, BOD5, DOC), nutrients (NO3-, TKN, TN, PO43-, PT...), heavy metals (Zn, Pb, Cr, Cu, Cd…), hydrocarbons (total and PAH), pesticides in particular and dissolved fractions. Soil samples are also taken in order to assess the pollution trapped.
The modelling of the evolution of the clogging with time is mainly based on the Bouwer’s model. Bouwer’s model, based on Darcy’s law, evaluates a global hydraulic resistance (R) of the basin. R represents the thickness divided by the hydraulic conductivity of the clogged layer. The calibration of the model gives a value of R for a given storm event. In order to calibrate the model, infiltration flow has to be known; it is computed from the measurement of water depth, inflow in the basin, storage and continuity equation. Measurements have started in January 2004.
By assessing the evolution of R with time, a second step is to analyse its relation with the soil quality and the role of water inflow characteristics on this evolution. For that purpose two aspects have to be studied: first, the pollutant load brought to the infiltration system during a defined period of time, then the load, and accumulation of pollutant in the soil during the same period of time. The first aspect is deduced from continuous and regular event mean concentration measurements. For the second aspect, an innovative and non-destructive method is applied. For three major metals known to be found in stormawater (Zn, Pb, Cu), an X-ray fluorescence (XRF) portable analyser is used to determine the concentration of pollutant on the surface of the basin and in the soil at different depths. More than 100 samples has already been taken and analysed (on a surface of 1 ha) based on a grid of 10 meters. This technique allows to take many soil samples without damaging the basin and to assess the spatial variability. Previous studies have revealed a good correlation between the major stormwater pollutants and their spatial variability, further analyses will be done on other parameters (other metals, hydrocarbons …) but on a reduced number of points.
A first sampling campaign was done at the end of April 2005 and is the starting point of the survey. A map of pollutant concentration can then be drawn and compared to the flooding frequency map in order to determine the correlation between them. At last, the mass of pollutant trapped in the soil can be compared to the mass of pollutant brought to the infiltration system on a given period of time and can be used to determine the pollution retention performance. Such analyses are undertaken on a regular basis (every 6 months) in order to follow the evolution of the pollution with time together with clogging.
The paper will present the global approach that is being carried out to study the long-term hydraulic behaviour of the basin and the relationship between clogging, water inflow and soil quality. It will describe the monitoring system, the Bouwer’s model, its calibration and validation. It will present and discuss the first results on soil pollution investigation and inflow quality and discuss their potential correlation with clogging.