Objectives and scientific hypotheses

Water resources are under increasing pressure because of the growing worldwide urban population and its related pollution and land-use impacts [37]. Furthermore, climate change may impact hydrological processes, leading to water scarcity, at the same time as creation of urban areas leads to the sealing of urban soils, leading to both less groundwater recharge and more discharge of contaminated runoff and storm-waters to the environment. Mitigating such impacts and achieving a sustainable management of water resources are therefore key challenges for all cities in Europe and beyond. The strategy for Smart and Sustainable Cities is based on the development of solutions aimed at reducing the sealing of urban soils and promoting groundwater recharge with water quality improvement [37]. Nature-based solutions, such as water retention ponds, green and permeable surfaces in cities are inspired by nature and provide the functions of infiltration of water in the soil as well as filtration of pollutants. These different types of urban soils provide regulation of the water cycle, flood risk protection and climate change adaptation. Infiltration basins and trenches are one of the most widely-used nature-based ecotechnologies to mitigate the impacts of stormwater on urban communities and waterways [69]. These techniques reduce stormwater at its source by infiltrating runoff water directly into the ground.

However, stormwater is contaminated by a variety of biological, chemical and physical pollutants, originating from anthropogenic activities commonly practiced in urban areas [89]. Common pollutants include heavy metals (e.g. copper, chromium, cadmium, lead, zinc, etc.), metalloids (e.g. arsenic), pesticides (e.g. 2,4-D, Diuron, glyphosate, AMPA) and other organic compounds (COVs, PAHs) [70], most of these being part of the list of priority pollutants proposed by European Community [32, 89]. Emerging pollutants also raise important questions in terms of their impacts on the environment and public health. For instance, there is an increasing utilization of rare earth elements (electronics, medicine, agriculture). The most frequently detected rare earth element is gadolinium (Gd), routinely used as a contrasting agent in magnetic resonance imaging and released into the environment through hospital effluents, under the form of soluble complexes or nanoparticles [11, 73]. Nanomaterials and nanoparticles have recently joined the long list of emerging pollutants commonly found in stormwater [24, 78]. Stormwater is also a carrier of microbial pathogens including Enteroviruses, Pseudomonas aeruginosa, Escherichia coli, and Nocardia cyriacigeorgica [70, 77]. Given the prevalence of this wide range of pollutants, the infiltration of stormwater and the fate of pollutants and bacteria in urban soils must be properly investigated to afford stormwater infiltration while ensuring pollutant filtration for the protection of the quality of groundwater.

The management of water in cities corresponds to a substantial economic sector. According to Eurostat data, the European sewerage and cleantech sectors encompass 58,000 enterprises, 1,269,000 jobs, and a turnover amount of 190,249 M€. In France, the French cities of Lyon and Nantes have invested significant money and energy into the development of techniques dedicated to water infiltration in urban areas and supported nature-based development initiatives, enhancement of urban social, cultural and economic resilience, and climate change adaptation. The Field Observatory for Urban Water Management (Observatoire de Terrain en Hydrologie Urbaine – OTHU) unifies 12 research laboratories from Lyon (France) to develop a long term field-observatory with the support of the Greater Lyon city council and the Rhone-Mediterranean Corsica water agency [6]. The OTHU also forms part of the French urban hydrology observatory network URBIS labelled as SOERE from 2013 to 2015 by Allenvi, along with ONEVU (Nantes) and OPUR (Paris). This observatory network gathers a unique multidisciplinary team with expertise in climatology, hydrology, fluid mechanics, geography, soil sciences, chemistry, biology, microbiology, and social sciences [6] to investigate a wide range of phenomena associated with urban drainage.

To date, no experimental tools exist for the concomitant evaluation of infiltration and filtration functions of soils. This project aims at developing INFILTRON, a commercially-applicable package, which combines an experimental device and numerical model based on field infiltration tests at the metre scale and makes use of eco-friendly nano-tracers (engineered nanoparticles) to trace bacteria and emerging pollutants (anthropic nanoparticles). The INFILTRON package aims at assessing the efficiency of water infiltration and pollutant filtration functions of urban soils, within the framework of the optimization of infiltration system management, but also applicable to other contexts such as the characterization of polluted sites and related environmental and health issues. The INFILTRON project addresses the design and validation of INFILTRON and, at the same time, promotes synergy between the partners on the topic of mass transfer and preferential flow in natural, urban and anthropic soils.

To the authors’ knowledge, this study is the first one to

• propose a package including INFILTRON experimental device and INFILTRON model for the assessment of infiltration & filtration functions of urban soils,
• account for preferential flow, including preferential flow in urban heterogeneous soils, and
• deal with the transfer of emerging pollutants (including nanoparticles) and bacteria and
• propose a strategy for the design of eco-friendly nanoparticles for mimicking emerging pollutants and bacteria.

The consortium will have a strong focus on maximising dissemination of the results and their implications for practice through guidelines for practitioners, scientific and technical meetings. In addition to this new knowledge, this project will provide cutting-edge but user-friendly tools for practitioners:

1. INFILTRON-exp, an experimental simple approach, based on water infiltration experiments, to assess the infiltration & filtration functions and to track preferential flows in the field and
2. INFILTRON-mod, a numerical tool for modelling preferential flow in urban soils including infiltration systems but also polluted sites, anthropic soils.

The INFILTRON tool will guide engineers and practitioners to optimize the management of stormwater infiltration basins, but also urban soils and polluted sites, allowing practitioners to consider environmental quality risk criteria. The tools will have excellent potential for future market outcomes.

Originality and relevance in relation to the state of the art

Urban soils are known for their strong heterogeneity. The first source of soil heterogeneity, even natural or urban soils, can be attributed to lithological heterogeneity [26]. The soil can be seen as a patch of several lithofacies with contrasting lithological, geochemical and transfer properties. The scale of the lithological heterogeneity corresponds to the dimensions of lithofacies under the form of inclusions and layers of a few meters or a few tens of meters. In addition to this kind of heterogeneity, each lithofacies may exhibit inherent heterogeneity, resulting in a variation of lithological and transfer properties from one point to another in space, or even at the mm – cm scale (e.g. macropores and cracks) [26, 42, 43]. Macropores, cracks, root channels and holes produced by fauna (e.g. earthworms) and plants constitute an important source of lithofacies inherent heterogeneity, mostly for uppermost horizons [9]. In addition, in urban soils, the embedment of specific materials (like draining materials) constitute an additional source of soil heterogeneity.

Infiltration systems: the perfect model of urban soils

Infiltration basins are the perfect example of urban soils including strongly heterogeneous urban soils. The injection of stormwater brings large amounts of particles and solids that form a sedimentary layer at their surface. Meanwhile, plants colonize the upper soil horizon, and modify the soil texture and structure [7, 33, 38]. Below, the subsoil is often chosen for its high permeability and thus prone to a strong lithological heterogeneity. For example, in the region of Lyon, most infiltration basins are built over a glaciofluvial deposit, well known for its high permeability but also its strong lithological heterogeneity [35]. Such lithological heterogeneity was described for the specific case of Django Reinhardt (DjR) infiltration basin (see Figure 1).

All these parameters lead to the formation of a complex and strongly heterogeneous anthropic soil [86]. At surface, the upper layer concentrate pollution from stormwater, including emerging pollutants [24, 78]. Opportunistic human pathogens such as Pseudomonas aeruginosa and Nocardia cyriacigeorgica were recovered in significant amounts among detention basins and runoff waters. Pathogenic species belonging to Nocardia have also been detected in fecally-contaminated waters [10] and in the surface soil layer of infiltration basins and watershed in the region of Lyon [61] (See Figure 2).

Infiltration basins are simultaneously good representative of urban soils and a potential source of pollution; which make these systems the perfect object to study for the investigation of flow and mass transfer in urban soils.

Soil heterogeneity & preferential flow and mass transfer

The capacity of soils to infiltrate stormwater and filtrate stormwater by removing pollutants is crucial with regards to the proper functioning of infiltration systems and the quality of the groundwater below. Besides, soil heterogeneity must be considered when the question of infiltration and filtration of pollutants is raised. The effect of lithological heterogeneity on flow has been demonstrated for the Django Reinhardt infiltration basin in the Lyon region by combining an intensive lithological characterization of the glaciofluvial deposit below this basin and numerical modelling [8, 20, 36]. The deposit is made of several lithofacies with contrasting properties, including an upper layer of a mixture of gravel and sandy matrix, a uniform mass of a bimodal gravel (mixture of gravel and sandy matrix) and, lastly, inclusions of lenses of sand and matrix-free gravels [20, 35] (see Figure 1a-b). The numerical modelling of flow and solute transfer through sections of soils (~10-15 m long and 2-3 m deep) demonstrated the occurrence of capillary barrier effects at the vicinity of the lenses of gravel and sand [8], resulting in the establishment of preferential flow (see Figure 1c). Preferential flow pathways have characteristic widths in the order of the metre and are more marked under drier states (lower water saturation degrees and infiltration rates imposed at surface). Such flow patterns are proved to impact water infiltration at the surface and thus the infiltration function of the soil over the course of the year [20]. In addition, such a flow pattern may also impact the fate of pollutant in the soil, meaning an impact on the efficiency of the filtration function of soils, as discussed in the following paragraph.

The filtration function of soils results from several mechanisms that ensure pollutant sorption to soil particles, pollutant degradation, etc. In the calcareous deposit that covers most of the plain of Lyon, heavy metals are retained by precipitation and sorption to calcite particles [51]. In the same deposit, nanoparticles undergo a large panel of retention mechanisms [72],  including attachment to solid-water and air-water interfaces, aggregation and straining, or film straining [67]. Undoubtedly, in all cases, sorption and retention mechanisms require good access of pollutants to the soil reactive particles [44, 51, 52]. However, this access is ruled by the transport of pollutants by water and thus by flow pattern. In the case of the Django Reinhardt basin (Figure 1c), preferential flow limits pollutant access to reactive particles located along flow pathways, and prevents retention or sorption onto particles far from these pathways. Such restriction may prevent pollutant from reaching more than half of the soils, thus precluding the contribution of these soil particles in pollutant retention. This can greatly reduce the efficiency of pollutant retention . In summary, preferential flows are expected to lower significantly the function of filtration of pollutants of urban soils, and thus must be accounted for in the design and evaluation of stormwater infiltration and filtration systems.

Lack of infiltration techniques

To date, no tool has been developed for the assessment of infiltration and filtration functions of urban soils in a way that also considers the effect of preferential flow. Soil and agricultural science communities have pioneered a large range of techniques based on water infiltration to quantify and characterize water infiltration in soils. These techniques were developed within the framework of the optimization of irrigation and the management of the risks of soil erosion [1]. However, these techniques were mostly developed for the characterization of infiltration processes at small scales (~cm scale) and do not address hydrological processes at larger scales (e.g. [47]). Secondly, very few of these techniques are appropriate for addressing the processes of preferential flows since their related scales are much larger than the usual dimensions of infiltration devices (in the order of few cm) [42, 43, 49]. Thirdly, most of these techniques focus on water and do not pay much attention to the fate of pollutants [17], whereas infiltration techniques based on tracer injection are needed to address the fate of pollutants and to evaluate the filtration function of soils. Given this context, there is an urgent need to develop an infiltration assessment tool and a physically-based method that allows for concomitant assessment of the water infiltration and pollutant filtration functions of urban soils.

The INFILTRON project aims at :

• the development of an infiltration device (infiltrometer) that addresses the meter scale to properly account for preferential flows,
• the injection of nano-tracers to track and quantify the transfer of emerging pollutants (including nanoparticles and bacteria), and
• the use of non-intrusive geophysical methods techniques to catch the pathways of water and the nano-tracer.

The monitoring of water infiltration at surface will allow the quantification of the function ‘infiltration’. The monitoring of the nano-tracers below the infiltrometer and the quantification of the mass of nano-tracers reaching a certain depth will allow the quantification of the function ‘filtration’. Finally, the picture of water and nano-tracer pathways provided by geophysical methods will give insight on the physical processes related to flow and mass transfer. INFILTRON will use synthetic nanoparticles as physical and biological nano-tracers to represent several families of pollutants and bacteria; which strengthens the novelty of project. Silica nanoparticles will be tested as candidates, since nanomaterials are ubiquitous in stormwater and silica is cost-effective and eco-friendly (dissolution in water leading to innocuous silicic acid). Silica nanoparticle size can be adjusted over a large range of values (e.g. from 10 nm to a maximum of 5 µm), and their surface charge can be finely adjusted by grafting silanes with various hydrophobicity degrees [18, 41]. Gadolinium (Gd) is also a good candidate, in particular for lab investigation, given its property of contrasting agent for Magnetic Resonace Imaging (MRI). AguIX (Activation and Guiding of Irradiation by X-ray) particles, currently used as multimodal theragnostic probes for cancer therapy, can be easily engineered and controlled in terms of surface properties [78]. This use of synthetic nanoparticles as nano-tracers is novel and quite promising.

Next: Methodology & Risk management