Transport and corresponding timescales of waters in the unsaturated zone are yet poorly understood. We use a continuous 3-year record of stable water isotopes from the “WaldLab Forest Experimental Site” in precipitation, streamflow, soil waters from various depths and xylem waters to trace subsurface transport and water uptake.

Most of annual streamflow was older than 2–3 months (82%) reiterating the importance of old waters stored in the subsurface for streamflow generation. Only 3% of streamflow was younger than 1 day, indicating that even at high flows the majority of streamflow does not origin from recent precipitation. The fractions of recent precipitation decreased with depth in the subsurface, both in mobile and bulk soil waters. Fractions of recent precipitation found in soils and streamflow were larger with wet antecedent conditions, showing that wet landscapes can transmit precipitation quicker than dry landscapes. However, mobile soil waters below 2 m depth contained much less recent precipitation than streamflow did, indicating that streamflow must be generated from a mixture of deep subsurface waters, with very little isotopic seasonality and short-term variability, and shallow soil waters, with more pronounced isotopic seasonality and short-term variability.

Overall, most of subsurface water originates from precipitation in the winter half of the year, even though our site receives more precipitation in the summer half of the year; most summer precipitation is intercepted, evaporated or transpired by forest-floor vegetation, thus not dominantly contributing to soil water storage. As a result, forest-tree transpiration is also dominated by winter precipitation. Our results illustrate how flow, storage, and mixing processes linking precipitation to streamflow and evapotranspiration can be constrained by measuring isotopic variability across different hillslope positions, subsurface depths, and time scales.

Monitoring programs that are designed for groundwater protection are mostly based on information obtained from observation wells penetrating the saturated part of the subsurface. However, pollution identification in well water is clear evidence that contamination has already crossed the entire unsaturated zone and accumulated in the aquifer water. Along with that, remediation of contaminated sites depends mainly on the ability to control the hydrological and chemical conditions that are required for the development of pollutants degrading bacteria across the contaminated profile. Accordingly, both protection of groundwater pollution from pollution hazards and optimization of remediation strategies require real-time information on the dynamics of water percolation and contaminant transport in the unsaturated zone.
Vadose zone monitoring systems (VMS) enable real-time measurement of water flow and contaminant transport across the entire unsaturated zone. Continuous measurement of variation in sediment water content, along with the variations in the chemical and isotopic composition of water and gas phases at multiple depths across the unsaturated zone may be used for direct assessment of water and contaminant flux crossing from the land surface to the water table. Moreover, it provides data that enables optimization of remediation activity through real-time adjustment of the chemical and hydrological conditions of the unsaturated zone. The VMS has been successfully implemented worldwide in a variety of projects on land use impact on the quality and quantity of groundwater recharge, as well as optimization of remediation strategies of contaminated sites. These include assessment of potential pollution of various agricultural setups, landfills, PFAS-contaminated sites, and remediation of hydrocarbon, perchlorate, and explosives-contaminated sites.

Surface contamination by per-and poly-fluoroalkyl substances (PFAS) have occurred due to the on-ground application of firefighting foams (AFFF) in Australia and worldwide for more than four decades. Over the last decade, environmental monitoring and increasing research efforts, including those performed at affected Australian Commonwealth sites, have delivered significant progress towards understanding the sheer scale of the problem and have indicated notable retention of PFAS in the vadose zone below discrete AFFF source zones. This has led to the identification of the unsaturated zone as a long-term source for sustained PFAS contamination of groundwater and a target for remedial management.

In this study, we combined a detailed physical and geochemical characterization of a selected AFFF-impacted vadose zone with a model-based interpretation of the collected data. This framework allowed us to unravel the historic fate of PFASs across the vadose zone and, importantly, identify the processes that have, under varying climatic conditions, exerted the key controls on the observed PFAS mobility within the vadose zone. This formed the basis for estimates of PFAS fluxes towards underlying groundwater at the selected study site, which in turn allowed predictive simulations to determine future risks, the factors controlling them, and the efficiency of a range of potential remediation measures.

The Cretaceous Chalk aquifer is a major groundwater resource in Europe and in the Paris basin. Understanding the transfer of water and contaminant in the unsaturated zone (UZ) of Chalk is a crucial issue.
The underground quarry of Saint-Martin-le-Noeud is an exceptional observatory because it opens a window into the chalk aquifer. It is about 1200m long and 300m large. Dug at the limit between the UZ and the saturated zone, it provides direct access to the water table: underground lakes in the deepest parts of the quarry and the output of the UZ (percolation water at the ceiling).
The properties of the UZ vary along the quarry. UZ thickness is between 17 and 30m. The layer of clay-with-flints (CWF) overlying the chalk, characterized using surface geophysics , is from 0 to about 10m thick.
From 2012, geochemical and hydrodynamic data are collected on groundwater for 16 “sites” (lake + percolation) with different UZ characteristics. Time series of water level, elec. cond., temp. and percolation flow are recorded. Samplings are collected every 2 to 3 months to analyze major and trace ions, pesticides, isotopes, tritium, …
Along the quarry, Chalk groundwater has highly spatially variable hydrodynamic behaviour and geochemical properties.
This long-term multidisciplinary dataset provides new insights about
- the local variability in groundwater behaviors upon reaching the water table: little or very high contamination
- the balance of fracture versus matrix flow: the Chalk groundwater is mostly old (a few decades here) and water transfer processes occur mainly in the matrix with velocities range 0.37 to over 0.72 m a−1
- the transport of contaminants: the agricultural contaminants are stored in the UZ (here from 15 and 30m deep) and percolates slowly to the water table
- the role of the CWF cover: When the CWF layer is thick, a perched water table is present, in which degradation processes of pesticides occur, and the recharge is continuous

Malta, a semi-arid island in the centre of the Mediterranean, is among the European countries facing the most significant stress on its natural freshwater resources, a carbonate groundwater system. The island (~315 square kilometres) has an intensive agricultural sector and is one of the most important users of Malta’s groundwater resources with variable groundwater pollution potential. To improve agriculture and groundwater sustainability and management, a nationwide network of sixteen vadose zone monitoring stations (VMS) was established in various sites representing Malta's agricultural land uses. The VMS network provides long-term continuous information on the variation in the rock-water content and chemical composition of the percolating water at multiple depths in the unsaturated zone. Furthermore, this VMS network establishes an extensive hydrological and chemical database on the unsaturated zone from the data derived from sites representing various agricultural, geological, and hydrological conditions. This extensive vadose zone database will be processed through physical-based and machine-learning models, which will be utilized simultaneously on all VMS stations. A comprehensive groundwater recharge model will be established by incorporating factors such as soil type, geological profile, land use, irrigation, fertilizer application, and rainfall. Preliminary results indicate that the spatial variability of precipitation does not necessarily determine the total water storage in the unsaturated zone, despite significant differences in vadose zone water storage across various land uses. Simultaneous analysis of large hydrological and chemical datasets from the unsaturated zone enables, for the first time, a quantitative assessment of agricultural land use impact on the quantity and quality of groundwater recharge.

In the Gilgel Gibe catchment, 90% of the domestic water supply comes from groundwater. As a result of the government's initiative to boost wheat production for food self-sufficiency, the demand for groundwater use in irrigated agriculture is dramatically increasing in the catchment. This may lead to the development of a negative water balance and aquifer depletion, unless appropriate management practice is implemented. Groundwater recharge is pivotal in devising effective management plans. Although, groundwater modelers, watershed planners, and policymakers usually represent a constant value, groundwater recharge remarkably varies spatially and temporally over a region. Currently, various methods are available to estimate the groundwater recharge at remarkably fine spatial and temporal scales. In this study, the Soil Moisture Balance (SMB) method was applied to the Gilgel Gibe catchment to estimate the temporal distribution of groundwater whereas Water and Energy Transfer in Soil, Plants and Atmosphere under a quasi-Steady State (WetSpass) was applied to estimate its spatial distribution. In both cases hydrometeorological 36-year (1985 to 2020) data obtained from 15 ground-based stations and satellite products (CHIRPS and NASA/POWER) were utilized. The input data for WetSpass, including, land slope, land use land cover, soil, and topography were prepared in ArcGIS at 100 m x 100 m spatial resolution. The result shows that the spatial distribution of annual groundwater recharge varies from 0 mm for water bodies to 765 mm for forest-covered areas with an average value of 323 mm. The average annual groundwater recharge estimated by the SMB method is 308 mm of which 4% recharges during spring, 72 % during summer , 24 % during autumn , and nil during winter. The annual value shows significant temporal variation ranging from 101 mm in 2002 to 573 mm in 2019. The study provides useful information for groundwater management and modeling in the catchment.

Groundwater recharge is a key process for understanding and predicting groundwater system responses to changes in hydroclimate. We present a case study from the central Perth Basin, which has experienced a drying climate since the 1970’s. Previous work suggests that recharge on the Gnangara mound is initiated by monthly rainfall amounts in excess of 100-150 mm (Priestley, 2018) and has been assumed to be dominated by piston-flow type dynamics. With the measurement of a 2-year time series of rainfall, multi-depth soil moisture and groundwater level at three sites with different land covers, we aim to improve the recharge process understanding and better inform recharge estimates by including soil heterogeneity.
Despite the differences in soil texture and moisture content, the timing of the onset of recharge is relatively consistent across sites. Event-scale groundwater level rise is not strongly correlated with rainfall amount. Seasonal wetting of the profile is important for supporting recharge in some cases. However, at one of these sites, with the deepest depth to the water table and native vegetation, the soil moisture contents in the middle of the unsaturated zone do not wet up prior to the initiation of recharge identified as the first rise in water table levels. This suggests the presence of different wetting front dynamics such as preferential pathway flow. The presence of different mechanisms has implications for groundwater resource management and the effectiveness of land use change as a lever for ameliorating the impacts of a drying climate.

The impact of climate change as well as shallow groundwater extraction can have a huge influence on groundwater salinity. Especially in the lowland area surrounding the city of Lüneburg where a 5 years project (KliWaMoL) is ongoing to develop a planning criterion to handle salinization with specific management measures in respect to increasing water demands. Therefore, an AEM survey has been conducted. This paper focuses on a somewhat unexpected outcome, namely the ability to map the thickness of the unsaturated zone in the near surface aquifers.

The AEM dataset collected for this project benefits from early time gates within the ramp down. Modelling these gates allows us to obtain a higher near surface resolution in the models. The data were inverted using the smooth and the sharp inversion methodologies. The results were then compared against several geological and hydrological profiles in order to verify and assess the correlation between the resistivity response and the main lithological features. The shallow resistive features seemed to correlate locally with the water table surface. Therefore, we used this surface to evaluate the sensitivity of the inversion model. The near surface high resistivity layers in the area were then mapped using a cutoff value of 500ohmm. The differences between the water table and the 500ohmm cut off surface were then mapped. Through the integration of ancillary data, inversion methodology and high quality AEM data we obtained a map of the discrepancy between the elevation associated with the geophysical 500ohmm cut off surface and the water table. This map describes areas where these surfaces coincide and where they differ.

The method described in this paper is largely applicable in areas characterized by clean granular deposits like sand or gravel. Two of most obvious limitations of this method are the presence of clay that could act as local aquitard and the natural seasonal variations of the water table.