A groundwater catchment, located in Woodford and Jessamine Counties in the Inner Bluegrass of Kentucky, was instrumented to develop long-term flow and water quality data. The land uses on this 1 620-ha catchment consist of approximately 59% in grasses consisting of beef farms, horse farms, and a golf course; 16% row crops; 6% orchard; 13% forest; and 6% residential. Water samples were analyzed twice a week for, Ca, Mg, Na, Cl-, HCO3-, SO4=, NO3-, total solids, suspended solids, fecal coliforms, fecal streptococci, and triazines. Flow rate and average ambient temperature were also recorded. No strong linear relationship was developed between chemical concentrations and other parameters. The transient nature of the system was emphasized by one event that drastically deviated from others. Pesticide data were summarized and the ''flushing'' phenomena accredited to karst systems was discussed. The total solids content in the spring was consistent at approximately 2.06 mg/L. Fecal bacteria contamination was well above drinking water limits (fecal coliform and fecal streptococci averages were 1 700 and 4 300 colony-forming-units/100 mL, respectively) and the temporal variation in bacterial contamination was not linked to any other variable
The aim of this paper is to appraise the ability of the oxidation of riverine organic matter in the control of limestone dissolution, in a karst network. Biogeochemical processes during infiltration of river water into an alluvial aquifer have already been described for an average flow velocity of 4-5 m d(-1) (Jacobs, L. A., von Gunten, H. R., Keil, R, and Kuslys, M. (1988) Geochemical changes along a river-groundwater infiltration flow path: Glattfelden, Switzerland. Geochim. Cosmochim. Acta 52, 2693-2706; Von Gunten, H. R., Karametaxas, G., Krahenbuhl, U., Kuslys, M., Giovanoli R., Hoehn E. and Keil R. (1991) Seasonal biogeochemical cycles in riverborne groundwater. Geochim. Cosmochim. Acta 55, 3597-3609; Bourg, A. C. M. and Bertin, C. (1993) Quantitative appraisal of biogeochemical chemical processes during the infiltration of river water into an alluvial aquifer. Environ. Sci. Technol. 27, 661-666). Karstic drainage networks, such as in the River Loire-Val d'Orleans hydrologic system (Fig. 1), make possible flow velocities up to 200 m h(-1 a) and provide convenient access to different water samples several tens of km apart, at both extremities of the hydrologic unit (Chery, J.-L. (1983) Etude hydrochimique d'un aquifere karstique alimente par perte de cours d'eau (la Loire): Le systeme des calcaires de Beauce sous le val d'Orleans. These, Universite d'Orleans; Livrozet, E. (1984) Influence des apports de la Loire sur la qualite bacteriologique et chimique de l'aquifere karstique du val d'Orleans. These, Universite d'Orleans). Recharge of the karstic aquifer occurs principally from influent waters from stream sinks, either through coarse alluvial deposits or directly from outcrops of the regional limestone bedrock (Calcaires de Beauce). Recharge by seepage waters From the local catchment basin is small (Zunino, C., Bonnet, M. and Lelong, F. (1980) Le Val d'Orleans: un exemple d'aquifere a alimentation laterale. C. R. somm. Soc. Geol. Fr. 5, 195-199; Gonzalez R. (1992) Etude de l'organisation et evaluation des echanges entre la Loire moyenne et l'aquifere des calcaires de Beauce. These, Universite d'Orleans) and negligible in summer. This karstic hydrologic: system is the largest in France in terms of flow (tens to hundreds of m(3)/s) and provides the main water resource of the city of Orleans. Chemical compositions of influent waters (River Loire) and effluent waters (spring of the river Loiret) were compared, in particular during floods in summer 1992 and 1993 (Figs 2-4). Variation of chloride in the River Loire during the stream rise can be used as an environmental tracer of the underground flow (Fig. 2). Short transit times of about 3 days are detectable (Fig, 2) which are consistent with earlier estimations obtained with chemical tracers (Ref. in Chery, J.-L. (1983) These, Universite d'Orleans). Depending on the hydrological regime of the river, organic carbon discharge ranges between 3-7 and 2-13 mg/l for dissolved and particulate matter respectively (Fig. 3). Eutrophic characteristics and high algal biomasses are found in the River Loire during low water (Lair, N. and Sargos, D. (1993) A 10 year study at four sites of the middle course of the River Loire. I - Patterns of change in hydrological, physical and chemical variables in relation to algal biomass. Hudroecol. Appl. 5, 1-27) together with more organic carbon rich suspended particulate matter than during floods (30-40 C-org % dry weight versus 5-10%). Amounts of total organic carbon and dissolved oxygen (Fig. 3) dramatically decrease during the underground transport, whereas conversely, dissolved calcium, alkalinity and inorganic carbon increase (Fig. 4). Anoxia of outflows map start in April. Dissolution of calcium carbonates along the influent path outweighs closed system calcite equilibrium of inflow river waters (Table 3). The impact of organic matter oxidation on calcite dissolution may be traced by variations of alkalinity and total carbonates in water. Following, Jacobs, L. A., von Gunten, H. R., Keil, R. and Kuslys, M. (1988) Geochemical changes along a river-groundwater infiltration flow path: Glattfelden, Switzerland. Geochim. Cosmochim. Acta 52, 2693-2706), results are shown graphically (Fig. 5). Extent of reactions is controlled by the consumption of dissolved O-2 and nitrate for organic matter oxidation and by the release of Ca2 for calcite dissolution (Table 2). The karstic network is considered to behave like a biological reactor not exchanging with the atmosphere, with steady inhabitant microbial communities (Mariotti A., Landreau A, and Simon B. (1988) N-15 isotope biogeochemisrry and natural denitrification process in groundwater: Application to the chalk aquifer of northern France. Geochim. Cosmochim. Acta 52, 1869-1878; Gounot, A.-M. (1991) Ecologie microbienne des eaux ei des sediments souterrains. Hydrogeologie, 239-248). Thus, energy requirements only are considered, not carbon assimilation. Moreover, there is no necessity to invoke any delay for nitrification enhancement, as observed elsewhere, after waste water discharge into the river (Chesterikoff, A., Garban, B., Billen, G. and Poulin, M. (1992) Inorganic nitrogen dynamics in the River Seine downstream from Paris (France). Biogeochem. 17, 147-164). Main microbial processes are assumed to be aerobic respiration, nitrification and denitrification. Reactions with iron and manganese, real but not quantitatively important, were neglected. Sulphate reduction and methane formation, certainly not active, were not considered. Denitrification, which is suggested by low nitrate and ammonium concentrations and anoxia in the outflow, is known to be rapid enough to be achieved in a short time (Dupain, S. (1992) Denitrification biologique heterotrophe appliquee au traitement des eaux d'alimentation: Conditions de fonclionnement et mise au point d'un procede. These, Universite Claude Bernard, Lyon). Reaction are somewhat arbitrary but conform to general acceptance (Morel, M. M. and Hering, J. G. (1993) Principles and Applications of Aquatic Chemistry. Wiley, New York). Anaerobic ammonium oxidation (Mulder A., van de Graaf, A. A., Robertson, L: A. and Kuenen, J. G. (1995) Anaerobic ammonium oxidation discovered in a denitrifying fluidized bed reactor. FEMS Microbiol. Ecol. 16, 177-184). although possible, was not considered. In fact, C/N ratio of the reactive organic matter has only mild repercussions on the results; i.e. in the same range as the analytical errors for alkalinity and total carbonates. The objective was simply to roughly confront characteristics of outflowing waters and the calculation. Respective roles of aerobes and denitrifiers, for instance, are not certain. Several periods during low water or floods were selected with various ranges for calcium dissolution or nitrate and oxygen concentrations. The result is that in most cases simulation and data are in reasonable accordance (Fig. 5). Amounts of organic matter in River Loire are generally sufficient to sustain the process (Table 3. Particulate organic matter is probably the most reactive. The balance of oxidation of organic matter indicates that about 65 mu g C-org/l.h are oxidized during the transport without much variation with the river regime or organic discharge. It is concluded that limestone dissolution is directly dependent on organic matter oxidation, but variation occurs (7-29 mg CuCO3/l) with the level of bases that can be neutralized in the River Loire water. (C) 1998 Elsevier Science Ltd. All rights reserved
Geochemical data indicate that the Springfield Plateau aquifer, a carbonate aquifer of the Ozark Plateaus Province in central USA, has two distinct hydrochemical zones. Within each hydrochemical zone, water from springs is geochemically and isotopically different than water from wells. Geochemical data indicate that spring water generally interacts less with the surrounding rock and has a shorter residence time, probably as a result of flowing along discrete fractures and solution openings, than water from wells. Water type throughout most of the aquifer was calcium bicarbonate, indicating that carbonate-rock dissolution is the primary geochemical process occurring in the aquifer. Concentrations of calcium, bicarbonate, dissolved oxygen and tritium indicate that most ground water in the aquifer recharged rapidly and is relatively young (less than 40 years). In general, field-measured properties, concentrations of many chemical constituents, and calcite saturation indices were greater in samples from the northern part of the aquifer (hydrochemical zone A) than in samples from the southern part of the aquifer (hydrochemical zone B). Factors affecting differences in the geochemical composition of ground water between the two zones are difficult to identify, but could be related to differences in chert content and possibly primary porosity, solubility of the limestone, and amount and type of cementation between zone A than in zone B. In addition, specific conductance, pH, alkalinity, concentrations of many chemical constituents and calcite saturation indices were greater in samples from wells than in samples from springs in each hydrochemical zone. In contrast, concentrations of dissolved oxygen, nitrite plus nitrate, and chloride generally were greater in samples from springs than in samples from wells. Water from springs generally flows rapidly through large conduits with minimum water-rock interactions. Water from wells flow through small fractures, which restrict how and increase water-rock interactions. As a result, springs tend to be more susceptible to surface contamination than wells. The results of this study have important implications for the geochemical and hydrogeological processes of similar carbonate aquifers in other geographical locations. Copyright (C) 2000 John Wiley & Sons, Ltd