# Modeling Salt Redistribution in Fractured Porous Media Caused by Convection Driven Evaporation Within the Fracture Christopher Graham 1 Maria Ines Dragila 1, Clay Cooper 2, Noam Weisbrod 3 (1) Department of Crop & Soil Sciences, Oregon State University, USA; (2) Desert Research Institute; (3) Department of Environmental Hydrology & Microbiology, Institute for Water Sciences & Technologies, BIDR, Ben-Gurion University of the Negev, Israel Objective Evaporation Dynamics Justification ØHow does evaporation from fractures and salt concentration affect total soil evaporation? ØHow does evaporation from fracture affect salt redistribution? Figure 1: Schematic depicting conceptual model. (Left) Convection within soil fracture due to unstable density gradients between atmosphere and fracture air. (Right) Schematic of convection enhanced evaporation. Figure 3: Physical model based on Ritchie and Adams (1972) weighing lysimeter. 5 cm slice represents entire lysimeter due to symmetry. Full depth is modeled. Boundary Conditions Model Details ØSoil lysimeter is modeled using a representative 5 cm slice, 60 wide and 120 cm deep. ØHydraulic properties of Houston Black Clay are modeled using the van Genuchten curves using parameters for clay from Carsel and Parrish. ØLower boundary is constant head, allowing for gravity drainage, right and left boundaries are no flow, for symmetry. Upper boundary and fracture are constant head, representing a well mixed atmosphere. ØLower: Constant Head (Gravity Drainage) ØUpper: Constant Head (Atmospheric Boundary) ØRight: No Flow (Lysimeter edge) ØUpper Left: Constant Head (Fracture Atmospheric Conditions) ØLower Left: No Flow (for Symmetry) MODEL CALIBRATION AND VALIDATION ØTOUGH 2 is a three dimensional, multiphase, multicomponent, finite difference porous media simulator capable of handling severe permeability differences between elements. ØThe EWASG module of TOUGH 2 is designed to simulate water, saline and air fluxes in porous media. EWASG models salt in both the aqueous and solid phases. Evaporation and Salt Diffusion ØEvaporation and Salt Diffusion are modeled as Fickian Diffusion between elements, according to a soil matrix dependent version of Fick’s Law: Moisture and Advective Salt Flux ØSoil pore water flux is governed by Darcy’s Law, and advective salt flux is a function of this flux: Time (days) Figure 4: Simulation and experimental results. First stage evaporation duration and second stage evaporation rates are accurately modeled. f: diffusive flux f: porosity t 0: matrix dependent tortuosity T: saturation dependent tortuosity D: diffusion coefficient DX: change in mass fraction salt between elements. F: advective flux k(q): relative permeability A: cross sectional area X: mass fraction salt in element Dy: change in matric potential Dx: distance between elements N 0 100. 0 Fracture Y 0 120. 5 Low Salt Y 15 116. 6 High Salt Y 100. 1 Table 1: Simulation initial conditions and results. ØPermeability reduction greatly reduces cumulative evaporation. 100 day cumulative evaporation decreases 56% with a reduction of permeability from 1 E-13 to 1 E-15 m 2. ØSalt flux to the fracture is similarly reduced, with a 88% reduction in salt flux over the 100 day simulation. ØMatrix permeability clearly has a large impact on evaporation and salt flux dynamics. Permeability 100 Day Total Evap Average Daily Evap Solid Salt Precipitation Near Fracture (m 2) (kg) (mm/day) (% Pore Space) 1 E-13 1. 96 0. 65 3. 10 1 E-14 1. 13 0. 38 0. 81 1 E-15 0. 85 0. 28 0. 35 Table 1: Permeability simulation results. Time (days) Figure 10: Diffusive and advective salt flux in closest 3 cm of soil matrix to fracture Figure 12: Cumulative evaporation for 100 day simulations, with varying matrix permeability. Figure 13: Total solid phase salt precipitation over time with varying matrix permeability Conclusions Ø 100 day cumulative and daily evaporation increased with presence of fracture. ØFracture evaporation caused increased drying at all depths in soil column. ØIncreased pore salinity reduced daily and cumulative evaporation. ØReduced soil permeability decreased evaporation and salt flux. ØSalt levels increased near the fracture during course of simulations due to transport with water driven by matric potential gradient caused by evaporation. Implications Distance from fracture (m) Figure 5: Scale and hydrometer lab setup Figure 6: Water content iso lines Ritchie and Adams field data (left), and TOUGH 2 numeric simulation (right). Effect of Permeability Modification Time (days) Figure 9: Fraction of initial salt content after 100 day simulation (high initial salt content) Field Data ØBased on soil fracture and lysimeter described in Ritchie and Adams (1974). ØLysimeter is filled with Houston Black Clay ØSoil lysimeter has cross sectional area of 1. 83 m x 1. 83 m and depth of 1. 2 m. Øfracture is 50 cm deep, with surface width of 5 cm, running from corner to corner of lysimeter. Control Figure 7: Water content with distance from fracture at Figure 8: Cumulative evaporation for 100 day end of 100 day simulations, with and without fracture and without salt. and salt Salt flux (g/day) Numerical Simulator: TOUGH 2 % of Control Distance from fracture (m) ØSalt content in the 1 cm nearest the fracture increased over 300% over the 100 day simulations. ØA larger increase of 350% was seen in the low initial salt concentration simulation (not shown). ØSalt flux is a combination of positive advective salt flux and negative diffusive flux. Throughout the simulation advective flux overwhelms diffusive flux, though net flux decreases throughout the simulation. ØFlux is greatest nearest the fracture at early time, reversing later in the simulation. Water mass (kg) Figure 2: A) Fracturing and salt precipitation driven by evaporation in clay soil. B) Extreme salt precipitation on sandstone surface. C) Salt precipitation on sandstone and chalk surfaces. g/L Salt Flux Laboratory Data ØWhile the EWASG module of TOUGH 2 has been shown to accurately model saturated and unsaturated water and salt flux, and solid phase salt precipitation, its capabilities regarding evaporation are untested. ØA small scale evaporation experiment was constructed and simulated using the EWASG module of TOUGH 2. ØA hydrometer cylinder was filled with saturated sand allowed to evaporate for 80 days. ØSimulated results closely matched experimental data with minor modifications to the standard modeling procedures. (Y/N) Relative salt content Questions investigated less dense fracture air Total Evaporation Net salt flux (g/day) ØEvaporation from fractures that are open to the atmosphere will lead to salt crusting along the surface which can then either close the fracture or be flushed down to the aquifer during rain events. Fracture [Salt] Cumulative Evaporation (kg) ØEvaporation rate and total evaporation were enhanced by the presence of the fracture. ØDaily evaporation rate and total evaporation were reduced with increasing solute concentration. ØWater content decreased with increased proximity to fracture, with greater decrease seen with reduced salt concentration denser atmospheric air Water Content ØNumerical investigation of solute redistribution within the porous matrix caused by evaporation from a fracture in the matrix. Simulation Time (days) Figure 11: Net salt flux into closest 3 cm of fracture over course of 100 day simulation F. Botts ØWhile the fracture increased evaporation by less than the amount predicted by Ritchie and Adams (1974), this amount can have a significant impact on agricultural water balances, especially in arid ecosystems, which account for one third of the Earth’s landmass. ØEnhanced evaporation due to soil fractures has implications on water balance for agricultural lands as well as natural arid ecosystems such as shallow seasonal lakes in arid environments. ØSalt on the surface of fractures lies in preferential flow paths to the subsurface. In areas such as the Negev Desert of Israel, salt filled rock fractures penetrate meters deep into the subsurface, potentially creating preferential flow paths to the aquifer. This situation creates the possibility of salt contamination of subsurface aquifers. References ØCarsel, R. F. , and R. S. Parrish. 1988. Developing joint probability of soil water retention characteristics. Water Resources Research 24: 755 -769. ØO’Hara, S. L. 1997 Irrigation and land degradation: implications for agriculture in Turkmenistan, central Asia. Journal of Arid Environments 37(1): 165 -179 ØPruess, K. , C. Oldenburg, and G. Moridis. 1999. TOUGH 2 User's Guide, Version 2. 0 Lawrence Berkeley National Laboratory, Berkeley. ØRitchie, J. T. , and J. E. Adams. 1974. Field measurement of evaporation from soil shrinkage fractures. Soil Science Society of America 38: 131 -134. Øvan Genuchten, M. T. 1980. A closed form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Science Society of America 44: 892 -898. ØWeisbrod, N. , M. Dragila, C. Graham, and J. Cassidy. 2005. Evaporation from fractures exposed at the land surface: impact of gas-phase convection on salt accumulation Dynamics of Fluids and Transport in Fractured Rock. American Geophysical Union, submitted August 2004.
Скачать презентацию Modeling Salt Redistribution in Fractured Porous Media
a7ac87d2d1c029ee308dfcb34e78770c.ppt