Although field studies of the interaction of surface water and groundwater are common, several studies are discussed below because of the different field methods that were used for the investigations. Two studies of alluvial aquifers in the prairies of the United States included extensive well installations in the alluvial aquifers. Both studies were initiated to evaluate the propagation of flood waves through the alluvium. The study of the Platte River near Grand Island, Nebraska [ Hurr, 1983], indicated that the propagation of water through the alluvium caused by high stages in the river resulted in the formation of temporary wetlands on islands that were important habitat for migrating birds. The other study, which was of an alluvial aquifer in Kansas, is worthy of note because it indicated that water moves through paleochannels in large systems in a similar manner as the `substreams' mentioned previously for small, high-gradient mountain streams [ Sophocleous, 1991]. Groundwater levels in wells in the highly conductive paleochannels responded well to changes in river stage, whereas groundwater levels in wells between the channels had small fluctuations and showed little correlation with river stage. The study also indicated that the extent and speed of pulse propagation is more sensitive to variations of streambed roughness and stream channel slope than it is to any of the aquifer parameters.
The use of temperature data as a means to understand the interaction of groundwater and surface water was investigated by Lapham [1989] and by Silliman and Booth [1993]. Lapham [1989] used a Fourier series solution to describe the simultaneous, 1-dimensional flow of heat and groundwater beneath streams. This approach provides an indirect method of determining vertical flow rates, and the effective hydraulic connection, between sediments and overlying streams. The method, which consists of varying the Darcy velocity in the solution until the predicted temperatures match those measured in the field, was successfully applied to streams in New England. Use of temperature data in the study by Silliman and Booth [1993] involved measurement over time of both sediment temperature and temperature at the base of the overlying water column. This approach resulted in identification of gaining and losing reaches of Juday Creek, Indiana.
Direct measurement of seepage by seepage meters, a technique commonly used in studies of lakes, has also been applied to rivers. Lee and Hynes [1978] used seepage meters to define zones of groundwater inflow to Hillman Creek, Ontario, Canada. More recently, Belanger and Walker [1990] used seepage meters to determine the rates and distribution of seepage in the Indian River Lagoon, Florida, and Cruickshank et al. [1988] used them in a study of Winter River, Prince Edward Island, Canada.