A simple model of leakage induced by deep pumping and its application to pollution beneath a city in Thailand

Many cities in developing countries are dependent upon groundwater for water supply. Frequently this groundwater is pumped from semi-confined aquifers in alluvial deposits. These deeper aquifers are often considered to be protected from polluted shallow water by intervening less-permeable layers. However, where groundwater is pumped from a semi-confined aquifer immediately beneath a city, significant induced leakage of contaminated shallow water can occur. This may lead to a serious deterioration of water quality in deeper aquifers in the longer-term. A simple model has been developed which provides insight into the hydraulic controls on water quality in such semi-confined aquifers. The model provides a tool for the initial assessment and prediction of the impact of urbanization on groundwater quality. Also, the model characterizes the key hydrogeological behaviour through a single parameter, here termed the ‘city leakage factor’, which can be used to assess the vulnerability to contamination by leakage. A case study of a city in Thailand illustrates the use of this model.

---

Résumé Beaucoup de villes des pays en développement dépendent de l’eau souterraine pour leur alimentation en eau. Cette eau souterraine est souvent pompée dans des aquifères alluviaux semi-captifs. Ces aquifères plus profonds sont souvent considérés comme protégés des eaux peu profondes et polluées, grace à des couches intercalaires moins perméables. Cependant, dans le cas où l’eau souterraine est pompée à partir d’un aquifère semi-captif situé directement sous une ville, une drainance importante des eaux peu profondes et polluées peut être induite. Ceci peut entra?ner, à long terme, une détérioration significative de la qualité de l’eau dans les aquifères plus profonds. Un modèle simple a été construit fournissant un aper?u des contr?les hydrauliques agissant sur la qualité de l’eau dans des aquifères semi-captifs. Ce modèle est un outil permettant d’évaluer l’état initial et de prédire l’impact de l’urbanisation sur la qualité de l’eau souterraine. Le modèle caractérise également les comportements hydrogéologiques majeurs à travers un unique paramètre, nommé dans cette étude facteur de drainance de la ville“, et qui peut être utilisé pour évaluer la vulnérabilité de l’aquifère face à une contamination par drainance. L’étude de cas d’une ville en Tha?lande illustre l’utilisation de ce modèle.

---

Resumen Muchas ciudades de paises en desarrollo dependen del agua subterránea para el abastecimiento de agua. Frecuentemente el agua subterránea se bombea de acuíferos semi-confinados en depósitos aluviales. Estos acuíferos más profundos se protegen frecuentemente de agua somera contaminada mediante la intervencción de capas menos permeables. Sin embargo, donde el agua subterránea se bombea de un acuífero semi-confinado inmediatamente debajo de una ciudad, pueden ocurrir fugas significativas inducidas de agua somera contaminada. Esto puede conducir a un serio deterioro de calidad de agua en acuíferos más profundos en el largo plazo. Se ha desarrollado un modelo simple el cual aporta idea acerca de los controles hidráulicos en la calidad del agua en tales acuíferos semi-confinados. El modelo aporta una herramienta para la evaluación inicial y predicción del impacto de urbanización en la calidad del agua subterránea. El modelo también caracteriza el comportamiento hidrogeológico clave a traves de un solo parámetro, que aquí se denomina ′factor de fuga de la ciudad′el cual puede usarse para evaluar la vulnerabilidad a la contaminación por fuga. El uso de este modelo se ilustra con un estudio de caso de una ciudad en Tailandia.

---

     

A Study of Long-Term MTBE Attenuation in the Borden Aquifer, Ontario, Canada

In 1988 and 1989, a natural gradient tracer test was performed in the shallow, aerobic and aquifer at Canadian Forces Base (CFB) Borden. A mixture of ground water containing dissolved oxygenated gasoline was injected below the water table along with chloride (Cl-) as a conservative tracer. The migration of BTEX, MTBE, and Cl was monitored in detail for 16 moths. The mass of BTEX compounds in the plume diminished significantly with time due to intrinsic aerobic biodegradation, while MTBE showed only a small decrease in mass over the 16-month period. In 1995/96, a comprehensive ground water sampling program was undertaken to define the mass of MTBE still present in the aquifer. Since the plume had migrated into an unmonitored section of the Borden Aquifer, numerical modeling and geostatistical methods were applied to define an optimal sampling grid and to improve the level of confidence in the results. A drive point profiling system was used to obtain ground water samples. Numerical modeling with no consideration of degradation pedicted maximum concentrations in excess of 3000 μg/L; field sampling found maximum concentrations of less than 200 μg/L. A mass balance for the remaining MTBE mass in the aquifer eight years after injection showed that only 3% of the original mass remained. Sorption, volatilization, a biotic degradation, and plant uptake are not considered significant attenuation processes for the field conditions. Therefore, we suggest that biodegradation may have played a major role in the attenuation of MTBE within the Borden Aquifer.     

Diffusive and total oxygen uptake of deep-sea sediments in the eastern South Atlantic Ocean:in situ and laboratory measurements

Total O₂ uptake rates were measured by the benthic flux chamber lander ELINOR, and O₂ microprofiles were measured by the profiling lander PROFILUR in the eastern South Atlantic. Diffusive O₂ fluxes through the diffusive boundary layer and the depth distribution of O₂ consumption rates within the sediment were calculated from the obtained microprofiles. The depth integrated O₂ consumption rate agreed closely with the diffusive O₂ uptake at all stations. Total O₂ uptake was 1.2–4.2 times the diffusive O₂ uptake, and the difference correlated with the abundance of macrofauna in the sediment. Diffusive O₂ uptake and O₂-penetration depths correlated with the organic content of the sediments and exhibited an inverse correlation with water depth. Total and diffusive rates of in situ O₂ uptake were higher than previously published data for shelf and abyssal sediments in the Atlantic, but were comparable to rates from upwelling areas in the eastern Pacific. Laboratory measurements on recovered sediment cores showed lower O₂ penetration depths and higher diffusive uptake rates than in situ measurements. The differences increased with increasing water depth. We primarily ascribe this compression of O₂ profiles to a transiently increased temperature during recovery and enhanced microbial activity in decompressed sediment cores. Total O₂ uptake rates measured in the laboratory on macrofauna-rich stations were, in contrast, lower than those measured in situ because of underrepresentation and disturbance of the macrofauna.     

Giant sediment drifts on the continental rise west of the Antarctic Peninsula

Multichannel seismic reflection profiles from the continental rise west of the Antarctic Peninsula between 63° and 69°S show the growth of eight very large mound-shaped sedimentary bodies. MCS profiles and long-range side-scan sonar (GLORIA) images show the sea floor between mounds is traversed by channels originating in a dendritic pattern near the base of the continental slope. The mounds are interpreted as sediment drifts, constructed mainly from the fine-grained components of turbidity currents originating on the continental slope, entrained in a nepheloid layer within the ambient southwesterly bottom currents and redeposited downcurrent.     

The puzzle of high heads beneath the West Cumbrian coast,UK: a possible solution

A region of high heads within the Borrowdale Volcanic Group (BVG; a fractured crystalline rock) beneath the coastal plain of West Cumbria, England (UK), is identified as a possible relic left over by the Late Devensian ice sheet. It was found during investigations in the 1990s. Contemporary modelling work failed to produce a satisfactory explanation of the high heads compatible with the ‘cold recharge’ isotopic signature of the groundwater. This study has reassessed the original hydraulic testing results. By plotting density-adjusted heads versus their depth below the water table in the immediate vicinity of the borehole in which they were measured, a depth profile resembling a ‘wave’ was revealed with a peak value located at 1,100 m depth. The possibility that this wave represents relic heads from the last major ice sheet has been assessed using one-dimensional mathematical analysis based on a poroelastic approach. It is found that a wet-based ice sheet above the West Cumbrian coast was probably thick enough and sufficiently long-lasting to leave such relic heads providing that the hydraulic diffusivity of the BVG is in the order of 10^?6 m s^?1. Initial assessment 20 years ago of the long-interval slug tests suggested that such low values are not likely. More recent interpretation argues for such low values of hydraulic diffusivity. It is concluded that ice sheet recharge is the most likely cause of the raised heads, that the BVG contains significant patches of very low conductivity rock, and that long-interval single-hole tests should be avoided in fractured crystalline rock.     

Contamination of an Unconfined Sand Aquifer by Waste Pulp Liquor: A Case Study

A zone of contaminated ground water has been identified in an unconfined sand aquifer adjacent to a pit into which spent pulp liquor was intermittently discharged from 1970 to 1979. A network of multilevel sampling, bundle-type piezometers was installed. Up to seven depthspecific sampling points were incorporated into each piezometer providing a cost-effective means for three-dimensional mapping of hydraulic head and water quality in the unconfined sand aquifer. Ground-water samples retrieved from this network showed an area of contamination 900 m long, 400 m wide, and more than 25 m deep. This plume is dispersed about the ground-water flow lines passing beneath the waste disposal pit, and it terminates at a vigorous ground-water discharge area located 800 m from the pit. The contaminated ground water is characterized by elevated concentrations of sodium (3,000 mg/1), chloride (590 mg/1), alkalinity (2,700 mg/1), total organic carbon (2,000 mg/1), chemical oxygen demand (10,800 mg/1), biological oxygen demand (2,000 mg/1), tannin and lignin (780 mg/1), and lower sulphate (1 mg/1) compared to background ground waters in the area. The apparent rate of sodium migration is more than 50 m/yr and is close to the average linear ground-water velocity. Removal of some organic matter by biological transformation has produced the increased alkalinity in the contaminated ground water and somewhat reduced pH. Tannin and lignin are relatively inert compared to other organic compounds found in the waste liquor. Extremely low sulphate levels occurring in the highly contaminated ground waters indicate the existence of conditions favorable for microbially-mediated sulphate reduction.     

Distribution of organic matter in a shale clast

The mechanism of petroleum migration is not fully understood. One approach has been to study the distribution of organic matter near contacts between rocks with source character and those with reservoir character. Published studies have been restricted to profiles across the bedding, but the ease of movement of hydrocarbons along and across the bedding may be very different. Gradients for volatile bitumens and kerogen both along and across shale bedding have been established by analyzing a shale clast encased in a sand matrix. In this case, the distribution of volatile bitumens is not controlled by the distribution of kerogen but shows evidence of bitumen movement through the clast. The amount and composition of the volatile bitumens and their distribution suggest that in part they have been introduced into the clast from the surrounding medium, and that the movement of hydrocarbons occurs preferentially along the shale bedding. The composition of the material moving into the clast is not the same as the bulk composition of the external supply but shows a preferential enrichment in normal alkanes, particularly those of shorter chain length.