The heat transfer in the region of the Mauna Kea (Hawaii)—constraints from borehole temperature measurements and coupled thermo-hydraulic modeling

The objectives of this paper are an understanding of the thermal and hydraulic field because of a negative temperature gradient and cold temperatures in the 1-km-deep borehole of the Hawaiian Scientific Drilling Project (HSDP), located near the coast line. The temperature pattern is attributed to a superposition of thermal and hydraulic processes. In the deeper borehole (HSDP-2, depth 3.1 km) detailed temperature monitoring was performed. Temperature measurements reveal two different thermal regimes. The upper part is characterised by cold temperatures and a negative temperature gradient similar to those observed in the shallow pilot borehole. Below 1100 m, increasing temperatures are observed. Different processes, such as topographically driven groundwater flow, ingress of salt water and conductive heat flow are investigated by numerical modeling. A pure conductive scenario fails to match the temperature measurements, implying that both borehole sections are overprinted by advective conditions. Coupled fluid and heat flow modeling with solute transport yield results that agree with observed temperatures. The results of these simulations suggest that meteoric water flow from the mountain range controls the thermal conditions in the upper part of the borehole. Below this level, the thermal regime is additionally affected by circulation of salt water from the nearby ocean. Each of these flow systems has been observed at other locations: topographically driven fresh water at locations with pronounced topography and ingress of salt water is typical for islands or coastal areas. At Hawaii, they coincide and influence each other, resulting in a salt water interface occurring at greater depth than expected.     

Permeability of the rocks from the Kola superdeep borehole at high temperature and pressure: implication to fluid dynamics in the continental crust

Permeability of the samples collected from the surface and from the depths of 8–11 km in the Kola SG-3 and from the depth of 3.8 in the KTB boreholes was studied at temperatures up to 600 °C and pressures up to 150 MPa. These PT correspond to in situ conditions of the deep parts of the superdeep boreholes and to the conditions of progressive and regressive metamorphism of the Kola series rocks. The experiments were carried out with fluid filtration parallel and normal to rock foliation and parallel to core axis. The temperature–permeability trend behavior depends on effective pressure and depth of sample collection. At low effective pressure, a temperature increase leads first to a permeability decrease and then to its increase. At higher effective pressure, inversions appear on all the temperature trends of the samples collected from great depths. In contrast, permeability of the samples selected at shallow depth (3.8 km) and on the surface decreases within the entire temperature range. As a rule, with flow parallel to foliation, the values of permeability are higher than with flow normal to foliation. The results of microstructure studies allow to conclude that microcrack initiation and closure, due to a competitive influence of temperature and pressure cause such permeability behavior. In the samples, there are two families of microcracks: with low aspect ratio and those with high aspect ratio. Their effect on rock permeability changes with temperature. On sample heating, the low aspect ratio microcracks close and, on the contrary, high aspect ratio ones open. The total effect is expressed by minima in the temperature–permeability trends. Permeability anisotropy increases with temperature, reaches a maximum at 200 °C and then decreases. Sample permeability decreases with different gradients at simultaneous increase of temperature and pressure, simulating in situ depth increase. Hence, the deep seat rocks can vary greatly in permeability and against the common background of permeability decrease with depth, local deep aquifers may occur. At PT of progressive metamorphism the permeability values were high enough to permit the fluid flow to penetrate the whole volume of rock massif. At PT of regressive metamorphism, the permeability values were a few decimal orders lower, so fluid flow could be concentrated in large disjunctive zones only.     

Spatial characteristics of groundwater temperature in the Ishikari Lowland, Hokkaido, northern Japan: analytical and numerical applications

In porous sediments of the Ishikari Lowland, there is a gradual increase in the background geothermal gradient from the Ishikari River (3–4 °C 100 m^–1) to the southwest highland area (10 °C 100 m^–1). However, the geothermal gradient at shallow depths differs in detail from the background distribution. In spite of convective heat-flow loss generally associated with groundwater flow, heat flow remains high (100 mW m^–2) in the recharge area in the southwestern part of the Ishikari basin, which is part of an active geothermal field. In the northeastern part of the lowland, heat flow locally reaches 140 mW m^–2, probably due to upward water flow from the deep geothermal field. Between the two areas the heat flow is much lower. To examine the role of hydraulic flow in the distortion of the isotherms in this area, thermal gradient vs. temperature analyses were made, and they helped to define the major components of the groundwater-flow system of the region. Two-dimensional simulation modeling aided in understanding not only the cause of horizontal heat-flow variations in this field but also the contrast between thermal properties of shallow and deep groundwater reservoirs. Electronic Publication     

Thermal conductivity of soft Bangkok clay from laboratory and field measurements

This paper presents the results of a study on the thermal conductivity of a soft saturated clay (Bangkok clay) carried out in relation to an investigation into using thermal treatment to enhance the consolidation process of soft soils. The thermal conductivity of clay specimens was measured in the laboratory using a steady state method (divided bar test) and a transient state method (needle probe test). In general, the laboratory test results show that the thermal conductivity increased with the increase in soil density. However, the needle probe test was found to yield greater thermal conductivities than those derived from the divided bar test. Furthermore, to assess the validity of the laboratory test results, the heat transfer results obtained from a full-scale embankment test that employed prefabricated vertical thermo-drains (PVTD) were simulated numerically using the laboratory determined thermal conductivity values. The numerical analysis indicates that the field thermal conductivity was close to the value obtained from the needle probe test. However, it was also found that the changes in thermal conductivity values obtained from the two laboratory methods did not impact significantly on heat flow behaviour, suggesting that the two methods are acceptable for characterizing the thermal conductivity of soils.     

Heat flow in the Altai-Sayan Area: new data

Eleven new estimates of heat flow (q) from the southern Altai-Sayan Folded Area (ASFA) have provided update to the heat flow map of Gorny Altai. Measured heat flow in the area varies from 33 to 90 mW/m², with abnormal values of >70 mW/m^q at four sites. The anomalies may have a deep source only at the Aryskan site in the East Sayan (q = 77 mW/m²) while high heat flows of 75–90 mW/m² obtained for the Mesozoic Belokurikha and Kalguty plutons appear rather to result from high radiogenic heat production in granite, which adds a 25–30 W/m² radiogenic component to a deep component of 50–60 mW/m². The latter value is consistent with heat flow estimates derived from helium isotope ratios (54 mW/m² in both plutons). Heat flow variations at other sites are in the range from 33 to 60 mW/m². The new data support the earlier inferences of a generally low heat flow over most of ASFA (average of 45–50 mW/m²) and of a “cold” Cenozoic orogeny in the area (except for southeastern ASFA), possibly driven by shear stresses associated with India indentation into Eurasia.     

Numerical simulation of groundwater flow for a coastal plain in Japan: data collection and model calibration

Using a three-dimensional finite element model, this study characterizes groundwater flow in a costal plain of the Seto Inland Sea, Japan. The model characterization involved taking field data describing the aquifer system and translating this information into input variables that the model code uses to solve governing equations of flow. Geological geometry and the number of aquifers have been analyzed based on a large amount of geological, hydrogeological and topographical data. Results of study demonstrate a high correlation between the ground surface elevation and the groundwater level in the shallow coastal aquifer. For calibrating the numerical groundwater model, the groundwater flow was simulated in steady state. In addition, the groundwater level and trend in the transient state has also been elucidated. The numerical result provides excellent visual representations of groundwater flow, presenting resource managers and decision makers with a clear understanding of the nature of the types of groundwater flow pathways. Results build a base for further analysis under different future scenarios.     

Numerical investigation of salinity in controlling ore-forming fluid transport in sedimentary basins: example of the HYC deposit, Northern Australia

This paper presents the first hydrogeological model that fully couples transient fluid flow, heat and solute transport associated with the formation of the HYC SEDEX deposit in the McArthur Basin, northern Australia. Numerical results reveal that salinity plays an important role in controlling hydrothermal fluid migration. In particular, it appears that it is the distribution of evaporitic units within a given basin, rather than their absolute abundance, that controls the development of free convection. Relatively saline conditions at the seafloor strengthen the thermally-induced buoyancy force and hence promote free convection of basinal solutions; whereas high salinities at the bottom counteract the thermal function of natural geothermal gradient and suppress the development of convective hydrothermal fluid circulation. In the latter case, higher thermal gradients are required to initiate substantial free convective fluid flow.Numerical experiments also suggest the position of an ore body with respect to its vent system may be controlled by the spatial and temporal salinity distributions in the basin. Vent-distal ore formation, a result of exhalation of brines that are denser than seawater and hence can flow away from the vent region, is promoted by moderate salinity at the seafloor and higher salinity in the aquifer. Vent-proximal ore accumulation, a result of pluming upon exhalation of brines less dense than seawater, is favored by the highest salinity conditions occurring near the level of the seafloor.Editorial Handling: G. Beaudoin     

Absence of a regional surface thermal high in the Baikal rift; new insights from detailed contouring of heat flow anomalies

Heat flow in active tectonic zones as the Baikal rift is a crucial parameter for evaluating deep anomalous structures and lithosphere evolution. Based on the interpretation of the existing datasets, the Baikal rift has been characterized in the past by either high heat flow, or moderately elevated heat flow, or even lacking a surface heat flow anomaly. We made an attempt to better constrain the geothermal picture by a detailed offshore contouring survey of known anomalies, and to estimate the importance of observed heat flow anomalies within the regional surface heat output. A total of about 200 new and close-spaced heat flow measurements were obtained in several selected study areas in the North Baikal Basin. With an outrigged and a violin-bow designed thermoprobe of 2–3-m length, both the sediment temperature and thermal conductivity were measured. The new data show at all investigated sites that the large heat flow highs are limited to local heat flow anomalies. The maximum measured heat flow reaches values of 300–35000 mW/m², but the extent of the anomalies is not larger than 2 to 4 km in diameter. Aside of these local anomalies, heat flow variations are restricted to near background values of 50–70 mW/m², except in the uplifted Academician zone. The extent of the local anomalies excludes a conductive source, and therefore heat transport by fluids must be considered. In a conceptual model where all bottom floor heat flow anomalies are the result of upflowing fluids along a conduit, an extra heat output of 20 MW (including advection) is estimated for all known anomalies in the North Baikal Basin. Relative to a basal heat flow of 55–65 mW/m², these estimations suggest an extra heat output in the northern Lake Baikal of only 5%, corresponding to a regional heat flow increase of 3 mW/m². The source of this heat can be fully attributed to a regional heat redistribution by topographically driven ground water flow. Thus, the surface heat flow is not expected to bear a signal of deeper lithospheric thermal anomalies that can be separated from heat flow typical for orogenically altered crust (40–70 mW/m²). The new insights on the geothermal signature in the Baikal rift once more show that continental rifting is not by default characterized by high heat flow.     

Sensitivity analysis of hydrological parameters in modeling flow and transport in the unsaturated zone of Yucca Mountain, Nevada, USA

The unsaturated fractured volcanic deposits at Yucca Mountain in Nevada, USA, have been intensively investigated as a possible repository site for storing high-level radioactive waste. Field studies at the site have revealed that there exist large variabilities in hydrological parameters over the spatial domain of the mountain. Systematic analyses of hydrological parameters using a site-scale three-dimensional unsaturated zone (UZ) flow model have been undertaken. The main objective of the sensitivity analyses was to evaluate the effects of uncertainties in hydrologic parameters on modeled UZ flow and contaminant transport results. Sensitivity analyses were carried out relative to fracture and matrix permeability and capillary strength (van Genuchten α) through variation of these parameter values by one standard deviation from the base-case values. The parameter variation resulted in eight parameter sets. Modeling results for the eight UZ flow sensitivity cases have been compared with field observed data and simulation results from the base-case model. The effects of parameter uncertainties on the flow fields were evaluated through comparison of results for flow and transport. In general, this study shows that uncertainties in matrix parameters cause larger uncertainty in simulated moisture flux than corresponding uncertainties in fracture properties for unsaturated flow through heterogeneous fractured rock.     

Topography of the Moho undulations in France from gravity data: their age and origin

The complete gravity data set from France and part of the neighboring countries has been analyzed to compute the topography of the Moho undulations. This work is based on an improved filtering technique and an appropriate assumed density contrast between the crust and the upper mantle. Comparison with deep seismic refraction data reveals that this relief map expresses the continuity and geometry of the Moho undulations better than the sparsely distributed seismic refraction data in France. This gravity Moho map, though may not give absolute depths at places, provides a far better correlation with surface geology than the result from other geophysical techniques. Four domains have been recognized: (a) the Alpine domain where all the Moho undulations are concentric with the Alps; (b) the Armorican domain in which all the undulations are north-west/south-east oriented; (c) the Pyrenean domain, in which the undulations are parallel with the Mountain chain; and (d) the Massif Central Domain which does not show clear structural orientation because of the influence of the strong heat flow located at the lower crust/upper mantle interface. Study of the topography and of the superficial structures associated with these undulations reveals that the undulations delineated in the Alpine Domain result from the Tertiary compression which shaped the Alps. The Armorican Domain was first created during the Lower to Middle Cretaceous opening of the Bay of Biscay. It is now slightly affected by the Tertiary to Quaternary closure of this Bay. The Pyrenean Domain was successively shaped by the Lower Cretaceous oblique opening of the Bay of Biscay and by the Upper Cretaceous to Eocene northward displacement of Spain. Comparison between the Moho undulations map and the stress map of France reveals that most of the undulations are perpendicular to the actual shortening directions. This observation suggests that the Mesozoic, Cenozoic and Quaternary stress directions were roughly the same. Massif Central is characterized by the convergence of these three sets of undulations. Its Post-Oligocene uplift was probably the result of the converging stresses recognized in the three surrounding domains. When the Moho undulations and the topography are compared, two types of periodic crustal instabilities can be recognized. One corresponds to the buckling of the crust developed under compression, the other to boudinage which was associated with extension. Both phenomena show a typical wavelength of 200–250 km which is in agreement with the results of the actual physical and numerical modeling currently available.     

Piperno from Campi Flegrei: a relevant stone in the historical and monumental heritage of Naples (Italy)

This paper reports novel data concerning the volcanic formation of the Piperno, a stone that has been used for a long time in the monumental architecture of Naples and the Campanian region (Southern Italy). A careful field survey of the outcrops or underground sites so far accessible allowed us to draw a lithostratigraphic reconstruction of the formation and enabled the recognition of the three main layers exploited in past centuries and which provided dimension stones for architectural and structural purposes. Petrophysical and mineralogical characterization also enabled us to discriminate the same layers previously identified. Different scoriae/matrix ratio, scoriae dimension and welding degree are the most peculiar features of these horizons. The combination of these three parameters, even though they did not affect the use of Piperno as dimension stone, somehow determined its variable susceptibility to the weathering agents. However, a linear relationship between the exposure time and the decay of the stone itself, at a parity of weathering intensity, was not observed. This assumption was proved by verifying that the specific layers of the volcanic sequence controlled the weathering degree of Piperno from monuments disregarding their age of emplacement.     

Disparity between measured and BSR heat flow in the Xisha Trough of the South China Sea and its implications for the methane hydrate

We calculate the heat flow from the depth of bottom-simulating seismic reflectors (BSRs) on a seismic profile in the Xisha Trough of the South China Sea, and compare them with the probe heat flow measurements. The BSR heat flow turn out to be 32–80 mW/m², significantly lower than the measurements of 83–112 mW/m². Such big disparity cannot be ascribed only to the errors from parameters (parameter errors) that traditionally believed to influence the BSR heat flow. Besides the parameter errors, we discuss emphatically the errors coming from the theoretical assumption for the BSR heat flow determination (theoretical errors), which occur when the BSR depth does not coincide with the base of the methane hydrate stability zone (MHSZ). If BSR stays bellow the base of MHSZ, lying at the top of free gas zone, the derived heat flow would be underestimated. Compared with the parameter errors, the theoretical errors would be relatively larger in some geological settings. The disparity between measured and BSR heat flow in the Xisha Trough might be mainly due to the theoretical error. Based on the theoretical model, assuming that the BSR lying at the top of the free gas zone, the methane flux along the Xisha seismic profile is estimated, and the thickness of the methane hydrate occurrence zone is predicted.     

Numerical modelling and analysis of the influence of local variation in the thickness of a coal seam on surrounding stresses: Application to a practical case

One of the factors with the most influence in the carrying out of underground coal mining work is the stress state, especially when such work is carried out at great depths. One of the mining methods in which the influence of stresses becomes most patent is that of sublevel caving. For that reason, a variety of authors have studied many of the causes which condition the stress state of mining work of this kind at a general level.However, diverse types of conditioning factors exist at a local level that may substantially vary recorded pressures in the surroundings of the workings. Using numerical models, this paper analyses the specific case of a coal seam mined by the sublevel caving method at a depth of 1000 m which suffers significant increases in stress and major deformations in certain advance headings as a result of variations in the geometry and dimensions of the load itself, in particular, thinning.The results of the research carried out were validated by measuring the stresses in the area under study. Furthermore, on the basis of a retrospective analysis of what was observed in the mine, another series of predictive models were analyzed to establish the existing relation between the increase in stress due to thinning of the seam and the depth at which such thinning is situated.     

Interannual and seasonal variation of flow velocity in Koxkar Baxi Glacier from 2014 to 2020北大核心CSCD

Glacier flow is a key factor in understanding the nature of glaciers, and it is also one of the main research contents of glacier dynamics, which can provide basic support for rational utilization of glacier resources and early warning of glacier disasters. There are many mountain glaciers located in the west of China. The study on the spatiotemporal variation of surface velocity of glaciers also has great significance for the social and economic development of the western China. Koxkar Baxi Glacier, locates on the southern slopes of the Tomur Peak, is a typical dendritic glacier. In order to obtain the conditions of Koxkar Baxi Glacier’flow rates and its variation to further reveal the future of the variation of glacier, the spatiotemporal variability of glacier velocity was surveyed using correlation(COSI-Corr)method on Landsat imagery from 2014 to 2020. The results show that: (1)The average annual flow velocity of the Koxkar Baxi Glacier was 0. 04~0. 05 m·d-1 during 2014 to 2020. (2)The glacier reaches its maximum flow velocity near the center part, and the velocity decreased towards both lateral margins. In a longitudinal profile, ice flow velocity in the accumulation area increased down to the equilibrium line, while decreased towards the glacier terminal. The maximum velocity with 0. 17~ 0. 20 m·d-1 was found near the equilibrium line altitude. (3)The glacier flow velocity in warm seasons were 16. 67% faster than that in cold seasons. (4)The glacier flow velocity from 2014 to 2020 showed a slight decreasing trend, and the average flow velocity decreased 0. 01 m·d-1. (5)Temperature and precipitation had certain influence on the seasonal fluctuation and interannual variation of the flow velocity of the glacier. © 2022 Science Press (China). All rights reserved.     

中国大陆地壳和岩石圈化学成分模型可信吗? ——来自大地热流值的检验

大地热流值是大陆地壳和岩石圈U，Th,K丰度的直接约束；根据地球化学元素丰度值推算出的大地构造单元的区域地壳热流值，必定不能大于区域的平均热流值，根据700余个实测大地热流数据，对目前发表的中国大陆地壳和岩石圈的化学成分模型进行了检验，结果表明，多数模型不能满足大地热充约束，如黎彤等的关于中国大陆及其内部构造单元的地壳和岩石圈成分模型，倪守斌等提出的新疆北部地壳生热率模型，以及高山等提出的扬子地台北缘地壳成分模型，这些模型的U，Th,K丰度值不太可靠，其他强不相容元素的丰度值的可信程度亦值得怀疑，而迟清华，鄢明才提出的华北地台地壳成分模型和高山等建立的中国东部及华北地台和秦岭造山带的地壳成分模型通过了区域大地热流的检验。     

重力约束反演构建郯庐断裂带泗洪浅覆盖区隐伏基岩地质模型

利用重力异常数据建立覆盖区的基岩地质模型往往需要大量的地质与岩石物性资料,同时受地质先验信息、人为因素等干扰严重。运用重力异常钻孔约束反演的方法尝试构建浅覆盖区基岩地质模型,探讨在人为介入较少的情况下快速高效构建包含基岩面起伏特征、断裂展布、密度分布等要素的地质模型的可行性。在郯庐断裂泗洪地区的地质调查实际应用中,对区内布格重力异常采用小波多尺度分解的方法进行基岩面异常场提取,随后联合运用小波断裂分析、线性回归反演、视密度填图等方法构建了该区域的隐伏基岩地质模型。结果显示,在不需要大量地质先验信息的情况下,该系列方法能够快速高效地构建覆盖区地质模型。结果模型能够较好地反映基岩面构造及物性特征,并且细化郯庐断裂带泗洪地区的区域构造特征,对于郯庐断裂地区的构造演化研究具有一定的参考意义,同时为探索较少人为干预的地球物理数据反演覆盖区隐伏基岩的方法研究提供一定的实践价值。     

剪切指向转换的韧性剪切带——中国大陆科学钻探工程(CCSD)主孔中韧性剪切带(深度2010～2145m)的EBSD特征及运动学研究

位于苏鲁超高压变质地体南部的中国大陆科学钻探工程(CCSD)主孔深度1596～2038m的榴辉岩段和2038～2500m的片麻岩段之间存在一条厚一百余米的韧性剪切带(深度2010～2145m).韧性剪切带由糜棱岩化退变榴辉岩、花岗质糜棱岩等强应变岩石组成,韧性剪切带的面理倾向SEE,倾角由上部平均52°向中、下部平均32°转变,拉伸线理产状与面理倾向近一致,是220～200Ma期间折返应变的产物.糜棱岩化退变榴辉岩和花岗质糜棱岩的显微构造与石英晶格优选方位显示了折返阶段早期自SEE向NWW逆冲剪切指向以及后期自NWW向SEE正滑剪切指向转化的应变行为.CCSD主孔2010～2145m韧性剪切带的形成与它位于以榴辉岩为主的岩性.构造单元与其下以片麻岩为主的岩性.构造单元的界线附近有密切联系.第一期超高压阶段南北向韧性剪切作用形成的叶理、拉伸线理产状及相关的岩石类型与第二期、第三期阶段SEE-NWW向折返变形明显不同.     

Effect of advective mass transfer on field scale fluid and solute movement: Field and modeling studies at a waste disposal site in fractured rock at Oak Ridge National Laboratory, Tennessee, USA

Advective mass transfer is a pore scale mass-transfer process that affects fluid and solute movement between pore domains such as fracture and matrix in a structured porous medium. Mechanistically similar to advection in the advection-dispersion of solutes in non-structured porous medium, it redistributes solutes by moving solute and solvent simultaneously between pore domains. While there is much research on diffusive mass transfer that is often referred to as matrix diffusion, there is a lack of information and study for advective mass transfer in the literature. The objective of this research is to study the effects of advective mass transfer on fluid and solute movement between pore domains. First, field hydraulic measurements at a waste disposal site in fractured rock at Oak Ridge National Laboratory (ORNL), Tennessee, USA, are used to calibrate a fracture-matrix, two-pore-domain groundwater flow model. Latin-hypercube sensitivity analysis suggests that the uncertainty of the calibrated model parameters is small and the calibrated flow model is nearly the optimal. Fracture spacing thus obtained is used to calculate diffusive mass transfer coefficients. The individual effects of advective and diffusive mass transfer on solute movement are then quantitatively evaluated. The calculations indicate that pore structure conceptual models may significantly affect the role of advective mass transfer in field and pore-scale mass transfer. In the particular ORNL field site and with a fracture-matrix pore structure model, contribution of advective mass transfer to solute mass movement is about three to eight orders of magnitude smaller than that of diffusive mass transfer.

---

Resumen La transferencia de masa advectiva es un proceso de transferencia de masa en escala intersticial que afecta el movimiento de fluido y soluto entre ámbitos porosos tal como fractura y matriz en un medio estructurado poroso. Este proceso, mecánicamente similar a la advección en la dispersión-advección de solutos de medios porosos no estructurados, redistribuye los solutos mediante el movimiento simultáneo de soluto y solvente entre ámbitos porosos. Mientras que existe bastante investigación en transferencia difusiva de masa que frecuentemente se conoce como difusión en matriz, existe falta de información y estudio de transferencia advectiva de masa en la literatura. El objetivo de esta investigación es estudiar los efectos de la transferencia advectiva de masa en el movimiento de fluido y soluto entre ámbitos porosos. Primero se utilizaron mediciones hidráulicas de campo en un sitio de depósito de residuos en roca fracturada en el Laboratorio Nacional Oak Ridge (ORNL), Tennessee, USA, para calibrar un modelo de flujo de agua subterránea de ámbito de dos poros fractura-matriz. Análisis de sensitividad hipercúbico-latino sugieren que la incertidumbre de los parámetros del modelo calibrado es pequeña y que el modelo de flujo calibrado es aproximadamente el óptimo. El espaciamiento de fracturas así obtenido se utiliza para calcular los coeficientes de transferencia de masa difusiva. Luego se evalúa cuantitativamente los efectos individuales de transferencia de masa advectiva y difusiva en el movimiento de soluto. Los cálculos indican que los modelos conceptuales de estructura porosa pueden afectar significativamente el papel de transferencia de masa advectiva en escalas de campo e intersticial de transferencia de masa. En el sitio de campo específico ORNL y con un modelo de estructura porosa de matriz-fractura, la contribución de transferencia de masa advectiva al movimiento de masa soluto es aproximadamente tres a ocho órdenes de magnitud más pequeño que la contribución por transferencia de masa difusiva.

---

Résumé Résumé Le transfert advectif de masse est un processus de transfert de masse à léchelle du pore qui affecte les mouvements du fluide et du soluté entre les différents domaines de pores, tel les fractures et la matrice dans un milieu poreux structuré. Mécaniquement similaire à ladvection dans le concept advection-dispersion de solutés dans les milieux non structurés, ce transfert redistribue les solutés simultanément avec le fluide entre les différents domaines poreux. Alors que de nombreuses recherches portent sur les transferts de masse par diffusion, se référant généralement à une diffusion par la matrice, il y a un grand manquement dinformations et détudes sur les transferts de masse par advection dans la littérature. Lobjectif de cette recherche est détudier leffet du transfert advectif de masse de fluide et de soluté entre les différents domaines poreux. Premièrement, les mesures hydrauliques de terrain sur la décharge en milieu fracturé du laboratoire national dOak Ridge ORNL, Tennessee, USA, sont utilisées pour calibrer un modèle hydrogéologique à double porosité fracture-matrice. Lanalyse de sensibilité latin-hypercube suggère que lincertitude sur les paramètres du modèle est faible et que le calibrage est pratiquement optimal. Lespace de fracture résultant permet de calculer les coefficients de transfert de masse par diffusion. Les effets individuels de ladvection et de la diffusion sur les mouvements de solutés sont dés lors évalués. Les calculs indiquent que le modèle conceptuel de la structure des pores peuvent significativement affectés le rôle du transfert advectif de masse à léchelle du pore et du terrain. Dans le cas du site de lORNL et avec un modèle structuré fracture-matrice, la contribution de ladvection au transfert de masse est de lordre de trois-huitième du transfert de masse par diffusion.

     

Rates of salinization by free convection in high-permeability sediments: insights from numerical modeling and application to the Dutch coastal area

Numerical modeling and dimensional analysis is used to study the salinization of thick, high-permeability aquifers by free convection from a salt source at the surface. Current understanding of this process mainly concerns the initial stages of salinization only (boundary-layer development, break-up into fingers and initial phase of finger descent). In the modeling, special attention is paid to the role of two processes in the long-term salinization rate: (1) the progressive loss of salt from fingers by lateral diffusion, and (2) the coalescence of fingers during their descent. From the numerical simulations a relationship is derived that describes the development of the horizontally averaged salinity with depth and time as a function of permeability and initial-density contrast for aquifer Rayleigh numbers up to Ra =6,000. This relationship is consistent with and provides an extension to previous generalized relationships of the rate of finger descent. Its applicability to real-world aquifers (Ra >10⁵) that include complexities due to anisotropy, heterogeneity, and mechanical dispersion is discussed. Application to the Pleistocene coastal aquifer of the Netherlands (thickness 200 m, permeability 10^-11 m²) suggests that salinization of the aquifer during historic episodes of inundation by seawater occurred within decades.

---

Resumen Se utiliza modelos numéricos y un análisis dimensional para estudiar la salinización de acuíferos potentes de alta permeabilidad por convección libre a partir de una fuente salina superficial. El conocimiento acutal de este proceso se limita a las fases iniciales de la salinización (desarrollo de la capa de contorno, creación de digitaciones y fase inicial de la progresión de éstas). En la modelación, se presta atención especial al papel desempeñado por dos procesos de salinización a largo plazo: (1) la pérdida progresiva de sal por difusión lateral desde las digitaciones, y (2) la coalescencia de las digitaciones durante su avance. A partir de las simulaciones numéricas, se obtiene una relación que describe el desarrollo de la salinidad con la profundidad y el tiempo, promediada horizontalmente, el cual depende de la permeabilidad y del contraste inicial de densidad para números de Rayleigh inferiores a 6.000. Esta relación es coherente con índices previos generalizados del avance de las digitaciones, y representa una extensión a estos. Se discute su aplicabilidad a acuíferos reales (con números de Rayleigh superiores a 105), que tienen complejidades asociadas a la anisotropía, la heterogeneidad y la dispersión mecánica. La aplicación al acuífero costero Pleistoceno de los Países Bajos (20 m de potencia y 10-11 m2 de permeabilidad) sugiere que la salinización tuvo lugar en décadas, debido a episodios históricos de inundación por aguas marinas.

---

Résumé Une modélisation numérique et une analyse dimensionnelle ont été mises en oeuvre pour étudier la salinisation d'aquifères épais et à forte perméabilité par convection libre d'une source de sel en surface. La compréhension habituelle de ce processus concerne principalement les étapes initiales de la salinisation seule (développement d'une couche limite, partition en digitations et phase initiale de développement des digitation). Dans la modélisation, une attention particulière a été portée au rôle de deux processus du taux de salinisation à long terme: (1) la perte progressive de sel dans les digitations par diffusion latérale et (2) la coalescence des digitations au cours de leur développement. À partir de simulations numériques, une relation a été obtenue qui permet de décrire l'extension de la salinité horizontalement en profondeur et au cours du temps en fonction de la perméabilité et du contraste initial de densité pour des nombres de Rayleigh de l'aquifère jusqu'à Ra =6,000. Cette relation est compatible avec ces résultats et fournit une extension des relations précédemment généralisées du taux de développement des digitations. On discute son applicabilité à des aquifères réels (Ra >105) incluant des complexités liées à l'anisotropie, l'hétérogénéité et la dispersion mécanique. L'application à l'aquifère côtier du Pléistocène des Pays-Bas (épaisseur environ 200 m, perméabilité environ10–11 m²) laisse penser que la salinisation de cet aquifère au cours d'épisodes historiques d'inondation par la mer s'est produit durant des décennies.

---

     

Construction of three-phase data to model multiphase flow in porous media: Comparing an optimization approach to the finite element approach

Multiphase flow modelling is a major issue in the assessment of groundwater pollution. Three-phase flows are commonly governed by mathematical models that associate a pressure equation with two saturation equations. These equations involve a number of secondary variables that reflect the fluid behaviour in a porous medium. To improve the computational efficiency of multiphase flow simulators, several simplified reformulations of three-phase flow equations have been proposed. However, they require the construction of new secondary variables adapted to the reformulated flow equations. In this article, two different approaches are compared to quantify these variables. A numerical example is given for a typical fine sand.