Our understanding of unstable wetting phenomena in soils is limited. Therefore, lysimeter experiments were conducted in the laboratory to validate current wetting front instability theories. Four different grades of sieved and air‐dried perlite and quartz sand were used as the experimental material. Water was applied by a sprinkler system at rates within the range of natural precipitation rates in New Mexico. Experiments were conducted in small lysimeters (diameter 30 cm, height 50 cm) as well as a large one (diameter 100 cm, height 150 cm). The experimental results show that wetting front instability will cause fingering phenomena in a homogeneous soil system. This observation confirms experimental and theoretical results of other workers. The diameter of fingers was observed to be a function of the grain size of the sand. Small fingers (3–4 cm diameter) were found in coarse sand (grain size 1.41–0.84 mm); large diameter fingers (12 cm diameter) were observed in fine sand (grain size 0.42–0.25 mm). Our experimental results in the coarse sand show that, for infiltration rates varying between 0.3 and 12 cm/h, finger diameters remain more or less constant. This observation also agrees with existing theories. However, at infiltration rates lower than 0.12 cm/h, the coarse sand experiments show that the wetting fronts became stable. For rates between 0.3 and 0.12 cm/h, the wetting is semistable; that is, there is incomplete wetting without distinct development of fingers. A similar trend was observed in the experimental results of sands with grain sizes of 0.841 to 0.594 and 0.594 to 0.42 mm. This phenomenon has not been observed in previous experimental studies and is not predicted by current wetting front instability theories. Our experimental data under infiltration rates similar to natural precipitation intensities may explain why unstable wetting has rarely been observed in wettable field soils.

mental safety assessment in common situations where time and resources are severely limited. Comparative simulations of a large-scale field infiltration experi-As the question is quite broad, we narrowed it down ment at the Maricopa Agricultural Center (MAC) near Phoenix, AZ, were conducted using a hierarchy of models based on public, generic, by asking what is the ability of simple models and relaand site data joined with pedotransfer functions and an inverse procetively inexpensive data to reproduce and predict reliadure. Our purpose was to investigate the ability of simple models and bly the time evolution of water content profiles in nine relatively inexpensive data to reproduce and predict reliably the time 10-m-deep neutron monitoring boreholes during largeevolution of water content profiles in nine 10-m-deep neutron moniscale infiltration experiments at the MAC near Phoenix, toring boreholes at the site. By relying solely on public sources of in-AZ? To address this narrower question, we conducted formation one might conclude that soil at the MAC site is uniform to comparative simulations of the experiments, using moda depth of 16 m, with a water table at about 22 m. Upon collecting soil samples at the site, we learned the soils are layered and laterally els of increasing complexity and a hierarchy of supportdiscontinuous, with a perched water table at about 13 m. To identify ing data. The fastest and least expensive way to assess the least level of complexity required to simulate infiltration at the the geologic makeup and hydraulic properties of a site MAC, we compared models that consider one-and two-dimensional flow is to rely on public sources of information coupled with in a uniform soil, a soil consisting of uniform layers, and a stratified generic databases. Such information and data seldom soil with laterally distinct zones. There is a paucity of hydraulic characsupport more than a very simple model of the site. A terization data for the site. To investigate the feasibility of obtaining more accurate but time-consuming and costly approach hydraulic parameter estimates for the models on the basis of soil type, we ascribed uniform properties to each layer or zone using mean values is to describe site geology on the basis of disturbed soil of three generic databases. To improve these estimates, we ascribed samples collected at the site and to assess their hydraulic variable soil hydraulic properties to individual soil samples using reproperties using pedotransfer functions. This may justify gression and neural network pedotransfer functions based on soil type the postulation of a more detailed site model. Even and bulk density; we then used them to obtain Bayesian updates of more accurate but demanding and expensive is to collect mean hydraulic properties in each layer or zone. We used the various relatively undisturbed soil samples for laboratory determodels and mean parameter estimates to simulate water contents mination of their hydraulic properties. Additional alterduring one of several infiltrati...

No abstract

Abstract. This paper provides a theoretical stability analysis of gradual wetting fronts based on perturbation analysis. A traveling wave solution of the one-dimensional vertical flow Richards' equation is used as the basic flow on which three-dimensional perturbations are introduced. By locally linearizing the diffusivity form of the three-dimensional Richards' equation a linear partial differential equation is obtained which governs the perturbation variables. The stability of each point at the wetting front is considered in a local coordinate system. The analysis of this perturbation equation at these points of the wetting front provides not only the relationship between the finger sizes and the nonponding infiltration rates at the soil surface but also the traveling speeds of the fingers rooted from these points. Once a perturbation is introduced at some point on the wetting front, there are three possibilities for the development of the perturbation. (1) The perturbation will monotonically decline with time; in this case, no fingers will form and the system is stable. (2) The perturbation does not decline with time, but its downward velocity is less than that of the stable basic wetting front; thus the distribution layer will gradually cover the fingers and the system will become stable. (3) The perturbation will increase with time and have a downward velocity greater than that of the stable wetting front; in this case, the finger will persistently grow in front of the stable wetting front and the system will become unstable. This analysis can be applied to an unsaturated homogeneous soil profile with uniform initial water content for the prediction of instability and for the estimation of finger characteristics over a wide range of infiltration rates.

Abstract. In this study, we verify the stability analysis of the unsaturated flow equation presented in paper 1. We compare finger sizes measured in laboratory experiments with predictions by the stability model. Using measured unsaturated hydraulic properties which are fitted and extrapolated over the entire water content range, the model enables us to predict finger sizes over a wide range of nonponding infiltration rates. The model predicts an increase of finger diameter at high and low infiltration rates for all soils. Such increases were also observed in our experiments. The model yields reliable predictions of finger diameters at high and intermediate infiltration rates but does not perform as well at infiltration rates <1 cm/h. This is caused by experimental limitations in accurately measuring hydraulic conductivities at low infiltration rates. A sensitivity analysis for initial water content and the shape parameter of the hydraulic conductivity curve showed that a small change of the parameters resulted in a relatively small change of finger size at high and intermediate infiltration rates but at low infiltration rates the finger size change was large, i.e., several orders of magnitude. We conclude that the stability model presented in paper 1 can be used for the assessment of wetting front instabilities in homogeneous soils over a wide range of nonponding infiltration rates.