Preferential flows and the legacy of Peter F. Germann (1944-2020)

<p>Peter Germann died on December 6<sup>th</sup> 2020 in Bern, Switzerland. Known for a wide range of contributions to the physics of soil-water interactions and flow, his name (along with Keith Beven, his career-long collaborator and fiend) is recognized by an entire generation of soil physicists and hydrologists who studies macropore and preferential flows. They both co-authored the classic, and highly cited 1982 review paper in Water Resources Research on Macropores and Water Flow in Soils. Peter’s PhD work between 1976-1980 was a study of soil-water relations based on maintaining a network of 35 nests of tensiometers at 10 different depths down to 3m. At that time, these were still manual tensiometers coupled to mercury manometers that were read every 2 to 3 days for 3 years. One of the features that this remarkable data set revealed was that during infiltration, wetting in some cases occurred at depths, apparently by-passing the tensiometers above. This is what we all now know as preferential flow. Another was the large heterogeneity in responses between sites and between wetting events. For the major part of his research career, Peter was a strong advocate for a reconsideration of the physics of water flow through soils and, in particular, for the limitations of the Darcy-Buckingham-Richards flow theory. Peter later developed the kinematic wave approach into a theory of viscosity (rather than capillarity) dominated film flows subject to Stokes’ law during infiltration. He summarised his research work in his 2013 book on the subject published by the University of Bern. Peter held academic positions at the University of Virginia in Charlottesville, at Rutgers University, and at the University of Bern back in Switzerland where he stayed until he retired in 2009, and held an Emeritus position until 2015.   He continued to publish papers until shortly before his death which followed 2 major strokes. In this talk, we will go over Peter’s main contribution and research highlights in the area of macropores and preferential flows. Peter was no stranger to EGU, and many know him and have met him in this session or others. For those who knew Peter, they will miss his enthusiasm, his critical mind, his genuine care for the state of soil physics, his thoughtful responses, and his humour. He was a great source of inspiration to us and many others. Peter will be missed by many in soil science.</p>

May macropores increase contaminant retention?

<p>Preferential flow is quite usual in natural environments. Non-uniform and preferential flows co-exist or alternate, impacting water transport and contaminant transfer through the vadose zone. In this study, we investigated how macropore-induced flow affects manufactured nanoparticles, as emerging contaminants reactive transfer. Previous studies showed that the presence of a macropore into water-saturated soil columns can foster preferential water flow within the macropore. One could expect that this preferential flow may increase contaminant transfer and reduce retention by the matrix in the case of contaminant, as previously reported. In this study, we injected pulses of silver nanoparticles to assess their transfer through sand columns with and without a macropore. Both systems (with and without macropore) were studied under similar conditions. An unexpected result was obtained: more nanoparticles were retained in the system with a macropore, i.e., with a preferential flow. This result is quite counter-intuitive. It appears that the relation between flow homogeneity and contaminant retention is not straightforward. Some possible explanations, related to chemical and physical kinetics, are put forward to explain the experimental results.</p>

Groundwater recharge estimated by land surface models: an evaluation in the conterminous U.S.

Estimating diffuse recharge of precipitation is fundamental to assessing groundwater sustainability. Diffuse recharge is also the process through which climate and climate change directly affect groundwater. In this study, we evaluated diffuse recharge over the conterminous U.S. simulated by a suite of land surface models (LSMs) that were forced using a common set of meteorological input data. Simulated annual recharge exhibited spatial patterns that were similar among the LSMs, with the highest values in the eastern U.S. and Pacific Northwest. However, the magnitudes of annual recharge varied significantly among the models and were associated with differences in simulated ET, runoff and snow. Evaluation against two independent datasets did not answer the question of whether the ensemble mean performs the best, due to inconsistency between those datasets. The amplitude and timing of seasonal maximum recharge differed among the models, influenced strongly by model physics governing deep soil moisture drainage rates and, in cold regions, snowmelt. Evaluation using in situ soil moisture observations suggested that true recharge peaks 1-3 months later than simulated recharge, indicating systematic biases in simulating deep soil moisture. However, recharge from lateral flows and through preferential flows cannot be inferred from soil moisture data, and the seasonal cycle of simulated groundwater storage actually compared well with in situ groundwater observations. Long-term trends in recharge were not consistently correlated with either precipitation trends or temperature trends. This study highlights the need to employ dynamic flow models in LSMs, among other improvements, to enable more accurate simulation of recharge.

Ground-penetrating radar surveys for the detection of preferential flow into soils

<p>Preferential flow is more the rule than the exception, in particular during water infiltration experiments. In this study, we demonstrate the potential of GPR monitoring to detect preferential flows during water infiltration. We monitored time-lapse ground penetrating radar (GPR) surveys in the vicinity of single-ring infiltration experiments and created a three-dimensional (3D) representation of infiltrated water below the devices. For that purpose, radargrams were constructed from GPR transects conducted over two grids (1 m × 1 m) before and after the infiltration tests. The obtained signal was represented in 3D and a threshold was chosen to part the domain into wetted and non-wetted zones, allowing the determination of the infiltration bulb. That methodology was used to detect the infiltration below the devices and clearly pointed at nonuniform flows in correspondence with the heterogeneous soil structures. The protocol presented in this study represents a practical and valuable tool for detecting preferential flows at the scale of a single ring infiltration experiment.</p>

Preferential fluid flow and chemical transport in saturated fractured porous media and in heterogeneous vadose zones: Two sides of the same coin

<p>Preferential fluid flow and chemical transport occur on scales ranging from pores to aquifers and catchments, in both fully and partially water-saturated geological formations. Preferential flows can be considered, in a general sense, manifestations of self-organization that hinders perfect mixing within a system, and leads to faster throughput of water and chemicals. However, unified concepts for the onset, spatiotemporal patterning, and magnitude of such preferential flows are generally difficult to define; and this is compounded by the difficulty – or practical impossibility – of obtaining detailed measurements of the structure and hydraulic functioning of vadose zones, catchments, and aquifers. We propose that conceptualizations and quantitative characterizations of preferential fluid flow and chemical transport in all of these systems can be unified in terms of tools that connect them in a dynamic framework. Here, we discuss key, shared features of fluid flow and chemical transport dynamics in each of these two systems, based on both laboratory and field measurements, and numerical simulations. We show how even well-connected fracture networks can display highly non-uniform preferential paths for fluid and chemicals. We then recognize that this behavior is similar to that of rapid infiltration in soils and the vadose zone, which exhibits strongly localized preferential pathways in root channels, cracks, worm burrows or connected inter-aggregate pore networks. Moreover, both types of domains can display “memory effects”, in terms of the location and functioning of preferential paths even during perturbations in the velocity gradient and/or rates of infiltration. We argue that the ubiquity of unresolved (or uncharacterized) heterogeneity at all spatial and temporal scales necessitates the use of effective medium models that enable an accounting of a wide range of flow and transport behaviors. For chemical transport, we focus on a probabilistic modelling framework that can capture the dynamics in heterogeneous vadose zones and fractured (or otherwise heterogeneous) geological formations. We then demonstrate application of this model to interpret field-scale tracer breakthrough curves (concentration vs. time) in a highly fractured karst formation over length scales of up to more than 7 km.</p>

Understanding runoff formation in a basin with Peat Bog and Podzol hillslopes

<p>This research deals with the hydrological function of Peat Bog in a catchment where Peat Bog (formed by Histosol or other hydromorphic soils) covers a part of the area (40-60%). In this study, two soil types, creating two main hillslopes of the experimental catchment, form the dominant soil types (Podzol and Histosol) in the Šumava Mountains, Czechia. A modified HBV model was used for the estimation of the contribution of each soil type to common outflow and for the estimation of the water balance. According to previous research and field observations, dominant hydrological processes were described for each hillslope (soil). The model confirmed previous results concerning dominant preferential flows at Peat Bog hillslope and Podzol hillslope; moreover, it quantified a ratio between fast and slow flow in soils. At Peat Bog hillslope, the majority of outflow (67%) was formed from the upper soil layer (Acrotelm). In the mineral soil hillslope, a larger portion of runoff was generated from the lower soil layers or bedrock interface (61%). Peat Bog contributes to a stream mainly during rainfall events; however, the model showed also significant deep percolation at the Peat Bog hillslope and considerable contribution to baseflow during a year. Generally, more precipitation water was turned into runoff at the Peat Bog hillslope by the model, which was indicated by a lower rate of actual evapotranspiration (21% of precipitation), compared to 29% in the case of Podzol hillslope. If we consider land use changes in this locality in terms of expanding or reducing peat areas (draining, drains damming, droughts, etc.), this model could sufficiently estimate hydrological behaviour of local streams and thus, can be potentially used in hydrological planning by local authorities.</p>

Approximate expansions for water infiltration into dual permeability soils

<p>The understanding of hydrological processes requires the investigation of preferential flows. In particular, the infiltration compartment is strongly affected by preferential flows. Recently, Lassabatere et al. (2014) proposed a model for the analytical modelling of the infiltration impacted by preferential flow. These authors extended the model developed by Haverkamp et al. (1994) for single permeability soils to the case of dual permeability soils. However, this model remains implicit, requiring an inversion procedure for the quantification of the bulk cumulative infiltration. Such an implicit feature prevents from direct computation and may annoy any fellow who wants a direct and simple computation procedure. In this paper, we develop two approximate expansions for both transient and steady states. For that, we use the expansions proposed by Haverkamp et al. (1994) for single permeability systems. These expansions are written for each compartment of the dual permeability soils, i.e. the matrix and the fast-flow regions and are combined for the computation of the bulk infiltration. After formulation of these expansions, these are assessed in terms of their capability to accurately reproduce the complete implicit model. Their validity time intervals are also determined and discussed. The main limitation for the use of these expansions results from the fact that the time intervals that define the transient and steady states are contrasted between the matrix and the fast-flow regions. However, some domain of validity can be defined allowing the use of these approximate expansions.</p><p>Haverkamp, R., Ross, P. J., Smettem, K. R. J. and Parlange, J. Y.: 3-Dimensional analysis of infiltration from the disc infiltrometer .2. Physically-based infiltration equation, Water Resour. Res., 30(11), 2931–2935, 1994.</p><p>Lassabatere, L., Angulo-Jaramillo, R., Soria-Ugalde, J. M., Simunek, J. and Haverkamp, R.: Numerical evaluation of a set of analytical infiltration equations, Water Resour. Res., 45, W12415, doi:doi:10.1029/2009WR007941, 2009.</p>

Is breakthrough of solute impacted by the edges of the columns in the case of macropored systems?

<p>In natural systems, preferential flow is the rule rather than the exception. Non-uniform and preferential flows significantly impact mass transport, and by this way most of geochemical processes and pollutant dispersion in the environment. Laboratory columns are experimental devices used for the monitoring of solute transfer through porous. In particular, several studies used such experimental devices for characterizing mass transfer through heterogeneous systems and macropored systems. However, the design of these devices and its impacts on the experimental results has never been investigated in depth so far. In particular, the edge effect is rarely questioned and the transfer is always hypothesized to correspond to a fully developed flow (i.e., flow in an equivalent infinite system). In this study, we question this hypothesis both experimentally and numerically for the case of a macropored system. Tracer elutions, magnetic resonance imaging (MRI), and modeling using multiphysics approaches (Comsol) are conducted to demonstrate that flow is affected by edge effects close to the inlet and the outlet of the column, and that the presence of filters (used to prevent particles from exiting the system and clogging the outlet) do impact the flow and transfer breakthrough. Consequently, these edge effects should be considered when analyzing the results and concluding on the involved processes, in particular for the case of soils and systems with macropores.</p>