Application des techniques physiques, géochimiques et isotopiques pour l'évaluation de l'infiltration sous climat tempéré Physical, Chemical and isotopic techniques to evaluate recharge under temperate climates

Résumé

Depuis plusieurs années, la Zone Non Saturée d'une parcelle expérimentale au nord de l'Italie (plaine du Pô) a fait l'objet de nombreuses études sur les caractéristiques physiques et géologiques du milieu, ainsi que sur les modalités du mouvement des eaux d'infiltration. Le site a été équipé de 6 tensiomètres à mercure, 6 bougies poreuses pour l'échantillonnage de l'eau et d'un tube d'accès pour la sonde à neutrons.

L'application des différentes méthodes d'évaluation a fourni, pour la période d'étude (Septembre 1994-Août 1995), des valeurs de l'infiltration sensiblement différentes. La méthode physique, basée sur le calcul du bilan hydrique annuel, et considérant le sol comme un unique réservoir monocouche, a indiqué que 19% du total des précipitations arrive à s'infiltrer. D'autre part, un modèle de simulation intégrant les caractéristiques hydrodynamiques du sol a permis d'évaluer la recharge à 29% des précipitations annuelles. Le calcul de la répartition de la teneur en eau sur le profil porte sur la description des fluctuations temporelles du plan de flux nul, en relation avec les épisodes pluvieux et l'intensité de l'évaporation.

La méthode géochimique du "profil en ions conservés" entre l'eau de pluie et du sol a permis, quant à elle, d'évaluer la lame d'eau infiltrée à environ 12% des précipitations annuelles.

La composition isotopique des eaux extraites du sol manifeste un enrichissement global en isotopes lourds (²H et ¹⁸O) par rapport aux eaux météoriques, conséquence d'une évaporation intense. L'infiltration semble être de type "piston flow" dispersif. Elle suit, seulement pendant l'hiver, un flux où la composante convective dans la microporosité est prédominante sur la composante diffusive qui a lieu, de préférence, dans la macroporosité.

Mots-clés:

• Zone non saturée,
• climat tempéré,
• infiltration,
• bilan hydrique,
• ions conservatifs,
• composition isotopique,
• microporosité,
• macroporosité

Abstract

In groundwater management, it is necessary to define water movement, evaluate infiltration and evapotranspiration rates, and quantify the physico-chemical evolution of transported solutes throughout the unsaturated zone (UZ). Traditionally, in temperate regions, recharge rate is evaluated by the comparison between physical methods (based on direct measurements of hydrological parameters with lysimeters, tensiometers and neutron gauges) and geochemical approaches (conservative ions, stable isotope profiles and artificial tracers).

In the Po Plain (Northern Italy) intensive agricultural irrigation and overexploitation of groundwater profoundly affect the quality and availability of shallow groundwater resources. The region has been intensively cultivated with corn and rice during the last 5 centuries. Irrigation canals have been constructed in order to distribute water from alpine rivers to areas more favourable to agriculture. In the past, only the water balance method has been applied to obtain recharge rates in these situations.

An experimental field site has been operational since 1987 in the ENEA-EUREX Nuclear Centre of Saluggia. The experimental plot represents an unsaturated zone in fluvio-glacial deposits (Holocene) of the Dora Baltea River, formed by gravel and sand, interlayered with silt levels. The water table is usually at a depth of 200 cm but it varies as a function of the river level. The objectives of the present study were to describe water movement throughout the UZ and to evaluate infiltration and evapotranspiration rates using different physico-chemical methodologies. The Saluggia plot was equipped with six tensiometers, a neutron gauge hole and 6 porous cups, up to a depth of 160 cm.

During 10 years of monitoring, the most favourable period for understanding infiltration processes and water movement through the UZ is September 1994 - August 1995. This period is characterised by an exceptional rain event (300 mm in 48 hours), followed by a river flood. Water samples collected with porous cups at various depths have been analysed for chemistry and stable isotope composition, together with rainfall and Dora Baltea River water.

In October 1994 (beginning of the rainy period) and May 1995 (beginning of the dry period) soil samples were collected at different depths and water was extracted under vacuum conditions. Stable isotope analyses were performed on extracted waters. Soil samples were also analysed for granulometry, chemical and mineralogical composition.

Reducing soil to an unique monolayer aquifer, the water balance method suggests that only 19% of the total precipitation infiltrates, whereas a simulation model taking into account soil hydrodynamic characteristics estimates an infiltration rate of 29% of the rainfall. The position of the Zero Flux Plane (ZFP) fluctuates seasonally between the surface and the maximum studied depth, as a consequence of precipitation events and evaporation fluxes. In some periods, the ZFP is level with the water table and direct discharge may occur.

Measurements of the stable isotopic composition (δ²H and δ¹⁸O) of soil water allows a quantitative estimate of direct groundwater discharge. On a δ²H versusδ¹⁸O plot, pore waters from the UZ have an isotopic composition that differs from that of the majority of groundwater samples, plotting below the local rain water line and indicating some degree of evaporation during the recharge process. The isotopic enrichment is particularly significant at the evaporation front, suggesting that pore water in soil reflects a different recharge regime from those of the regional ground waters. Water isotope and solute composition were substantially modified from their original composition during the infiltration process within the soil, via mechanisms such as anion exchange with soil particles, salt precipitation/dissolution or isotopic fractionation.

Arial recharge was also evaluated using the depth distribution of a conservative solute. Assuming that chloride is derived from precipitation alone, Cl- content in soil and in rain water suggests that 12% of total rainfall infiltrates. Therefore, the recharge rate is estimated to be less than 100 mm/a. These data are in agreement with other results obtained by the chloride concentration profile method, in areas of the Po Valley, but are considerably different from those evaluated by the traditional physical methods.

Keywords:

• Unsaturated zone,
• temperate climate,
• recharge,
• water balance,
• conservative ions,
• isotopic composition,
• microporosity,
• macroporosity