H21D-0740
Evaporation from Near-Drift Fractured Rock Surfaces
The amount of water entering emplacement drifts from a fractured unsaturated rock is an important variable for performance evaluation of a potential high-level radioactive waste repository at Yucca Mountain, Nevada. Water entering the drifts as liquid or gas may enhance waste package corrosion rates and transport released radionuclides. Liquid water in form of droplets may emerge from fractures, or flow along the drift wall and potentially evaporate and condense at other locations. Driven by pressure and temperature gradients, vapor may be transported along fractures, or liquid water may evaporate directly from the matrix. Within the drift, heat-driven convection may redistribute the moisture leading to condensation at other locations. The geometry of the evaporation front around the drift is not fully understood and this, in turn, influences processes related to reflux, rewetting as the thermal pulse dissipates. Existing models focus on processes in the porous media (e.g., two-phase dual-permeability models for matrix and fractures), or on processes in the drift (e.g., gas-phase computational fluid dynamics models). This study focuses on the boundary between these two domains, and the corresponding models, where evaporation at the solid rock/drift air interface appears to play an important role. Studies have shown that evaporation from porous media is a complex process sensitive to factors such as (i) hydrological properties of the porous media, (ii) pressure gradients in the porous media, (iii) texture of the interface or boundary, (iv) local vapor and temperature gradients, and (v) convective flow rate and boundary layer transfer. Experimental observations based on passive monitoring at Yucca Mountain have shown that the formation surrounding the drift is able to provide and transport large amounts of water vapor over a relatively short period. This study will examine the basic processes that govern evaporation in the unsaturated rock surrounding drifts for ambient and thermally-perturbed conditions, and illustrate the effect of the contrasting hydrologic properties of the matrix and fracture continua. Simple analyses to establish bounds on vapor flux into the drift are proposed. A more prominent role for gravity to evaluate potential seepage is proposed. Available models will be evaluated for their applicability for in situ conditions at Yucca Mountain. This abstract is an independent product of the CNWRA and does not necessarily reflect the view or regulatory position of the NRC. The NRC staff views expressed herein are preliminary and do not constitute a final judgment or determination of the matters addressed or of the acceptability of a license application for a geologic repository at Yucca Mountain.
H21D-0741
Impact of Pore Size and Salt Precipitation on the Evaporation From Porous Media
Evaporation is a major process in the global water cycle. The literature on micro-scale evaporation mechanisms and the influence of the resulted salt precipitation within pores on evaporation processes is limited. Moreover, the role of heterogeneity on the combined evaporation – salt precipitation processes is poorly understood. This research focuses on salt deposition inside the matrix pores during the evaporation process and its correlation to gas permeability of the matrix. Another goal is to explore the combined impact of salt precipitation and heterogeneities on evaporation. Two experimental methods are used: (1) quantifying salt deposition in pores with high-resolution X-ray CT; and (2) monitoring of evaporation and salt deposition in controlled columns. In the first experiment, a Berea Sandstone, saturated with five percent NaI solution was allowed to evaporated naturally for five days while the water, air, and salt distributions were monitored periodically using X-ray CT. Klinkenberg permeabilities were determined prior to and following the evaporation process to explore the impact of salt deposition within the matrix on gas permeability. In the second experiment, eight columns were packed with fine and coarse sand as well as fine and coarse sand combined, with a textural interface between the two grain sizes. Four columns were saturated with DI water and the rest with salty solution. The columns were placed in a temperature controlled room with continuous monitoring of evaporation rates. Fifty percent of the salt precipitated at the top seven mm of the sandstone sample, near the evaporating interface. Large pores (>200 micron) remained open with very small amount of precipitated salt crystals, while the smaller pores were mostly clogged with salt. These observations indicate that the solution migrated to the evaporation surface by capillarity. The gas permeability was reduced by ten percent supporting the observation that most of the large-pores remained open. Preliminary results from the column experiments indicate that both grain size and pore-water salinity play a major role in the evaporation rate.
H21D-0742
Experimental Quantification of Evaporation from Surface-Exposed Fractures (ESEF)
Evaporation from surface-exposed fractures (ESEF) was investigated experimentally. Surface-exposed fractures can act as major conduits of flow from land surface to underlying aquifers. This role is intensified for low- permeability matrix and deep vadose zone. ESEF can cause salt accumulation (SA) at and near the fracture surface and within the fracture aperture, drawing solution from the vadose zone towards the fracture walls. Surface waters which penetrate and flow through these fractures during flood or intensive rain events can dissolve the accumulated salt and carry it downwards towards underlying groundwater. In this laboratory study, artificial fractures were formed by placing two rock blocks, 50x50x20 cm, with 1 cm between them, mimicking a large-aperture fracture. Pore solution in the rocks was held under low tension, and flux was continuously monitored. The system was constructed in a way that loss of pore solution could occur only via evaporation from the two fracture surfaces. All measurements were taken under controlled ambient temperatures ranging between 9 and 25°C. It was found that decreasing the ambient atmospheric temperature, while the bottom of the rock blocks were kept at constant temperature of 23°C, increases the evaporation from the fracture surfaces by up to three folds. It is concluded that moist air convection is the controlling mechanism for ESEF rates. Long-term effects of SA on EFEF rates were also studied, where evaporation rates began decreasing after approximately 100 gram of salt per square meter of fracture surface had accumulated. Furthermore, ESEF decreased by ~50% compared to its initial value after approximately 160 gram per square meter had accumulated.
H21D-0743
An Improved Force-Restore Method for Soil Temperature Prediction
The force-restore method was originally developed to enable soil temperature predictions. The method is currently used in many hydro-meteorological models. The force-restore method assumes that soil is uniform with depth. The method also assumes that thermal conduction is the only heat transfer mechanism necessary for prediction of soil temperature. These assumptions hamper the applicability of the force-restore method to many natural soil conditions. The main objective of this study is to improve the force-restore method by extending it to include the possibility of soil heterogeneity with depth and to include the possible occurrence of convective heat transfer as well as conduction heat transfer in soil. The improved force-restore method presented in this paper is identical to the current force-restore method for the specific condition of vertically uniform, dry soil. The average soil temperatures calculated by the current and the revised force-restore methods for a shallow soil layer were compared with measured soil temperatures at a bare soil site in the China the Loess Plateau during the period from July 22 to 26, 2005. Results showed that the revised force-restore method gave a realistic estimate of soil temperature, especially during daylight periods, and the current force-restore method, on average, overestimated either the diurnal amplitude or the phase shift for the shallow soil layer. The revised force-restore method underestimated nighttime soil temperature because it did not account for the diurnal variation of convection heat transfer and the condensation of water vapor occurring in soil during nighttimes. Overall, the revised force-restore method estimated shallow soil temperature better than the current force-restore method.
H21D-0744
Soil-Water Evaporation Dynamics Determined From Soil Sensible Heat Transfer Measurements
Soil-water evaporation is important in both the hydrologic cycle and the surface energy balance and for processes ranging from microbial ecology to global climate change. Yet, routine measurements are unable to capture rapidly shifting near-surface soil heat and water transfer processes involved in soil-water evaporation. Recent advancements in fine-scale measurement of soil thermal properties provide a new opportunity to observe heat transfer associated with soil-water evaporation in the upper centimeters of the vadose zone. The objective of this study was to determine the depth and location of the evaporation zone within soil using observations of sensible heat transfer. Three-needle heat-pulse sensors were used to monitor soil heat capacity, thermal conductivity, water content, and temperature below a bare soil surface in Central Iowa during natural wetting/drying cycles. Soil heat flux and changes in heat storage were calculated from these data to obtain a balance of sensible heat components. The residual from this balance (i.e., the net heat flux minus the change in heat storage) was attributed to latent heat from water evaporation, and thus, provided estimates of in situ water evaporation. As the soil dried following rainfall, results showed divergence in the soil sensible heat flux with depth. Divergence in the heat flux indicated the location of a heat sink associated with soil-water evaporation. Evaporation estimates from the sensible heat balance provided depth and time patterns consistent with observed soil-water depletion patterns. Evaporation occurred near the soil surface immediately after rainfall and the evaporation zone proceeded below 3 mm in the profile within 2-3 d of rainfall events. As the soil dried, the evaporation zone continued to penetrate deeper into the soil extending below 13 mm within 6 d after rainfall. Peak evaporation rates as high as 0.42 mm/h were observed at the 3-mm depth near midday, with evaporation declining by late afternoon. Evaporation occurred simultaneously at multiple soil depth increments, but with time lag between peak evaporation rates for the deeper depths. Daily heat-balance evaporation estimates compared well with microlysimeter evaporation estimates taken 3 or more d after rainfall providing root mean square error of 0.11 mm/d and r2 = 0.90. Implementation of fine-scale measurement techniques for the soil sensible heat balance provides a new opportunity to improve understanding of soil-water evaporation.
H21D-0745
Measurement and Modeling of Soil Moisture Under Differing Evaporative Conditions
Two experiments that measured bare soil evaporation and soil moisture were performed in a weighing lysimeter located at Five Points, CA. After wetting the soil surface overnight, the evaporation rate was determined for natural conditions by the changes in mass of the lysimeter soil box during a 14-day period in August, 2007. Subsequently, solar radiation was blocked and the experiment was repeated for six days. In both experiments, hourly measurements of soil water content were made at eight depths (0.1-1.0 m) using a capacitance probe. In natural sunlight, variation of soil moisture indicated upward water flow due to capillary forces caused by soil surface evaporation. At night, water flowed downward due to gravity. This diurnal cycle did not occur when solar radiation was blocked. Numerical simulations of one-dimensional water flow in these two experiments provided an estimation of the depth variation of water content reduction due to evaporation.
H21D-0746
Solving Steady Evaporation-Driven Flow Through a Shallow Aquifer With Piecewise Continuous Forms of Hydraulic Relationships
Nearly fifty years ago, Wilford Gardner provided analytical solutions to estimate steady evaporation-driven flow from a shallow aquifer. Using analytically tractable functions to describe the relationship between hydraulic conductivity, K, and matric potential, ψ, his solution provides the upward flux and the ψ profile. Though mathematically effective, Gardner's K(ψ) functions typically do not fit field or lab measurements as well as the Mualem-van Genuchten (MvG) or other hydraulic relationships, especially given the large ψ range expected for an evaporation-driven flow system. However, the same large ψ range can cause long run times or non-convergence in the numerical models required to use the MvG relationship. We present a rapid, semi-analytical method based on Gardner's 1958 solution to estimate the evaporation-driven flux through a one-dimensional shallow vadose zone described by the MvG K(ψ) relationship. Like others, piecewise continuous estimates of the MvG K(ψ) relationship are constructed from Gardner's exponential K(ψ) relationship. Unlike other methods, our method provides the unknown evaporation- driven upward flux and the ψ profile across the domain. A system of nonlinear equations is built from the individual Gardner exponential pieces and then solved using Mathematica. The method is rapid, accurate, and can be applied to other K(ψ) relationships. An application to centrifuge-measured hydraulic property data demonstrates the method's utility.
H21D-0747
Characteristic Lengths Affecting Evaporation From Porous Media With Sharp Textural Contrasts
Displacement of evaporation drying front into initially saturated porous medium results from capillary gradients between large pores at the front supplying the evaporation process from smaller pores at the surface. Like other displacement processes, the drying front may exhibit irregular spatial patterns due to pore size variations. In heterogeneous media containing sharp textural contrasts, drying front displacement patterns follow preferential liquid flow from coarse textured regions in support of evaporation from saturated fine textured regions connected to the surface. Hele-Shaw cells with vertical and horizontal sharp textural interfaces between coarse and fine sand domains were used to study water distribution during evaporation using neutron transmission technique and imagery with dyed water. For vertical textural interfaces, evaporation from saturated fine sand was sustained by liquid flow from adjacent coarse sand resulting in preferential advance of the drying front exclusively into coarse sand region. Direct evidence of water flow pathways from coarse to fine sand w obtained with neutron radiography using heavy water as a tracer. A characteristic length defining maximum drying front depth (in the coarse medium) is determined by the difference in air-entry values of the two media. In our experiments, viscous resistance exerted no effect on maximum front depth even when flow cross section (fine sand relative area) was reduced from 75% to 5% for similar external evaporative conditions (viscous limitations would be important for clayey media). For horizontal layers of fine over coarse sand, the drying front initially propagates in fine sand until air first enters the coarse sand resulting in an abrupt and disproportionally large displacement water from coarse to overlaying fine layer driven by capillary pressure difference between air entry values of the sands. Subsequent to rapid pressure relaxation, drying front invades preferentially the coarse layer with no changes in liquid distribution in the overlaying fine layer. Experiments with layers of different thicknesses and positions (depths) relative to evaporative surface revealed the importance of another characteristic length spanned by pore size distribution of the medium. The combination of intrinsic capillary characteristic length and the position of a textural interface below the surface defines the ultimate depth of drying front in layered media (hence magnitude of evaporative losses). Preferential evaporation patterns from texturally-heterogeneous media during capillary driven liquid flows result in an increase in overall evaporative losses relative to porous media represented by homogenous effective properties.
H21D-0748
Neutron Radiography Measurements Of Phase Distributions And Transport Mechanisms During Drying Of Hydrophobic And Hydrophilic Coarse Porous Media
Drying rate and patterns in porous media are strongly influenced by interplay between internal structural and transport properties and external boundary conditions. Liquid flow from the receding drying front to evaporating surface is induced by capillary gradients and sustained by liquid films connecting these domains. In addition to capillary influences, wettability may play a role in maintaining continuity of liquid films required for sustaining high evaporation rates. We used neutron radiography imaging coupled with direct mass-loss measurements of evaporation rates to link phase distributions with dominant transport mechanisms, and to deduce potential impacts on morphology and dynamics of drying fronts in hydrophobic and hydrophilic sands. Results suggest that drying front dynamics may be affected by surface wettability leading to early dominance of vapor diffusion in drying of hydrophobic sand. In contrast, the persistence of a network of liquid films in hydrophilic sand sustains prolonged stage 1 evaporation by liquid flow as the main transport mechanism. Differences in fluid phase distributions in hydrophobic and hydrophilic sands resolved by neutron transmission observations provide new insights into the nonlinear dependency of diffusion coefficient on water content distribution above the drying front, the impact of wettability on dispersion of receding front and liquid phase distribution affected by wettability of porous medium.