Questioning usual monitoring protocols of groundwater contaminated plume: example of a former landfill (Nantes, France)
Abstract
Questioning usual monitoring protocols of groundwater-contaminated plume seems to be necessary in the frame of climate change. Point-source contaminated sites like landfills frequently impact groundwater quality. The evolution and degree of impact is usually observed thanks to point sampling in conditions of low and high groundwater table, when the concentrations of contaminants are supposed extreme. Feedback on existing monitoring appears useful to try and understand the origin and period of peaks and to question the representativeness of groundwater monitoring downgradient of contaminated sites.
Here, we focus more particularly on one pilot site: a former landfill located in Nantes (western France). The underlying alluvial groundwater is contaminated by leaching of waste due to rainwater infiltration. The monitoring data include daily monitoring of groundwater table, electric conductivity, and temperature, and point sampling in early spring and autumn to check the impact of the site on the alluvial groundwater quality. A network of 5 piezometers located along a contamination gradient (upstream, landfill leachates, near and further downstream) is used for sampling. The data, available for several years and including extreme periods, are interpreted in detail: we compare the sampling dates with the evolution of the leachates and groundwater table, looking in particular at lowest and highest levels, the evolution of concentrations and observation of peaks, and the rainy and dry climate events. We assume indeed that important concentration fluctuations are associated with extreme climate periods such as heavy droughts, floods or rainy periods. In a context of climate change, these periods are becoming more frequent in Europe and worldwide.
A clear correlation occurs between contaminant concentrations in the leachates, the landfill groundwater table, perched in the waste deposit, and climate events. Recharge of the landfill groundwater leads to contaminant dilution, whereas higher concentrations and electric conductivity occur in early autumn (usual end of discharge). The same occurs in the alluvial groundwater downstream of the site, with a delay between the two signals. This delay seems to increase with distance from the site. Peaks observed in leachates are more difficult to interpret. Some seem correlated to heavy rain events, leading to a higher than usual landfill groundwater table. This could be linked to the unusual leaching of contaminated waste located usually above the leachate table. For 10 years, the high levels appear at different times of the year (from January to April), depending on very heavy rainfall. This questions the representativeness of the sampling period. The sampling carried out on the pilot site, is not always representative of higher and lower groundwater table as was targeted. The calculated time of transfer of contaminants downstream of the contamination source shows that it is improbable to observe peaks at the same time along the contamination plume. Understanding the origin of peaks anywhere in the system requires a better understanding of the dynamics of the system.
The results obtained on the pilot site underline the need for more effective protocols to improve monitoring strategies. Continuous monitoring using multi-parameter sensors appears useful. Piezometric registration helps knowing when the sampling occurred in the dynamic of the groundwater table evolution. Electric conductivity registration informs on the evolution of the global concentration of contaminants. This is more particularly interesting in case of contamination with major elements, which is the case of landfills. Threshold alert values could also be produced, for groundwater table and electric conductivity, to trigger additional groundwater sampling. Such threshold values would be particularly helpful to build warning systems for sites located upstream of vulnerable receptors such as drinking water catchments.