Many studies of the interaction of groundwater and surface water are initiated because of problems related to water quality; therefore, many investigators have used chemical characteristics of both groundwater and surface water to determine the interaction. In addition to using major ions in these studies, interest in organic constituents and biochemical transformations at the sediment-water interface and in the hyporheic zone have increased in recent years. Isotopes, especially environmental isotopes, are also being used with greater frequency for studies of the interaction of groundwater and surface water [ Payne, 1988].
The effects of agricultural practices on organic contamination of water resources is of concern in intensely farmed regions. Thurman et al. [1992] discussed the effects of herbicides on the quality of groundwater and surface water in the midwestern United States. Squillace et al. [1993] found that groundwater in the alluvium adjacent to the main stem of the Cedar River in Iowa was the principal nonpoint source of atrazine and deethylatrazine to the river after the river had been in baseflow conditions for 5 days.
Much of the research related to organic chemistry with respect to the interaction of groundwater and surface water is concerned with processes at or below the streambed. Although this is the hyporheic zone, in large rivers attention seems to be focused on lateral movement through the bank rather than longitudinal movement as is the case with mountain streams. For example, Bourg et al. [1989] found that retention and biodegradation processes at the bank of the Deule River, France, prevented the lateral migration of heavy metals into the contiguous aquifer. In another study of the filtering effects of streambanks, Darmendrail [1988] indicated that the biogeochemical conditions in the banks of alluvial aquifers affect the way pollutants are transported. Bourg and Bertin [1994] found that wells located within 15 m of the Lot River, France, showed significant seasonal variation in dissolved manganese. They hypothesized that part of the reason for this is a local control of the redox conditions by organic matter in aquifer sediments.
Vervier et al. [1993] found that bacteria in gravel bars are important in the processing of dissolved organic carbon in river water. They recommend that restoration of rivers should include development of gravel bars to enhance reduction of dissolved organic carbon. Vervier et al. [1992] further maintain that the groundwater and surface water interface can be considered an ecotone through which matter and energy are exchanged. The ecotone acts as a combination of three filters; photic, mechanical, and biogeochemical. The relative effects of the filters is determined by permeability of the porous medium; for example, if permeability is high, the filter primarily is mechanical, if permeability is low, the dynamics of the filter are controlled by aerobic and anaerobic processes.
In another study of the groundwater and surface water ecotone, Creuze des Chatelliers and Poinsart [1991] found that geomorphologic processes affect the spatial distribution of hyporheic fauna. Where ground-water inflow to the Rhone River is weak and where current velocities are low, biodiversity is low. In contrast, where groundwater inflow is high, interstitial fauna are abundant and diverse. In a third reach, which is in karst terrane, benthic communities are dominated by micro-arthropods such as Ostracoda and Cladocera.
Danielopol [1976; 1989] has studied the ecology of the groundwater-surface water interface of large rivers in Europe, such as the Danube, for many years. His studies indicated the importance of biological entities as indicators of the interaction of groundwater and surface water. Lafont et al. [1992] indicated that oligochaete worms at the groundwater and surface water interface may provide evidence of pollution of one to the other. They found that several rare taxa are found in coarse sediments of rivers upstream and/or downstream of dams or gravel operations, and that they invade surface waters from phreatic or interstitial water when flow is amplified between the two water bodies. Laszlo et al. [1990] found that activities of man, such as dredging and instream flow-control structures, adversely affect the hyporheic ecotone, causing changes in communities and in seepage rates through the bed of the Danube River in Hungary.
Most studies of the interaction of groundwater and surface water in which isotopes were considered used the isotopes to determine relative age of water or the proportion of water that had been exposed to evaporation. Stuyfzand [1989] used tritium to determine that the bulk of the contaminants, such as halogenated hydrocarbons, in water that seeped from the Rhine River in The Netherlands due to pumping of groundwater has done so within the past 25 years. McCarthy et al. [1992] used a time series of data on deuterium and oxygen-18 to determine the amount of Columbia River water that contributed to groundwater pumping in an alluvial aquifer near Portland, Oregon. Using a simple mixing model, they found that the river contributed 40 to 50 percent of the pumped water after 5 to 6 days of pumping. In a study of the Danube and Sava Rivers near Belgrade, Hadzisehovic et al. [1990] used tritium to determine that the aquifer of interest had two strata of water. The upper unit contained older water and had a weak connection to the river, and the lower unit, which is the principal aquifer, had younger water and was well connected to the river.
Because the concentration of radon-222 generally is much higher
in groundwater than it is in surface water, it was used in a
number of studies of the interaction of groundwater and surface
water. Yoneda et al. [1991] used radon-222 to determine
the discrete points of groundwater inflow to a river in Japan.
Radon-222 was also used by Ellins et al. [1990] to quantify
groundwater inputs to a stream in Puerto Rico. In the reach of
interest, the study revealed that the stream gained 1.2 m
s
but that it also lost 0.5 m
s
to
groundwater. In a similar study in Tennessee, Lee and
Hollyday [1991] used radon-222 to determine that 36 percent of
Carters Creek at a time of low flow was contributed by
groundwater.