Ground Water Circulation between the
Devil’s Ear Cave System and the Santa Fe River
The fact that the cave system experienced more river intrusion during a period of lower recorded river stage indicates that ground water – surface water exchange is not a simple direct function of the stage of the Santa Fe river. Rainfall and ground water level data collected from the field area and the Northern Highlands by the Suwannee River and St. Johns River Water Management Districts during this investigation provide the clues necessary to explain this complication. Those data show that there was significantly more rainfall in the highlands than in the lowlands during the February 1992 sampling period when the greatest amount of intruded river water was measured in the cave system. On the contrary, there was more rainfall in the lowlands than in the highlands during the June 1993 sampling period when the least amount of intruded river water was recorded. Ground water levels were measured in a Department of Transportation (DOT) well located approximately 2 km west of the field area. Those data show that the water level in the Floridan aquifer was higher during the June 1993 sampling period when there was more rainfall over the lowlands and less river water intrusion to the cave system.
The hydraulic data and the results of this investigation reveal that the ground water – surface water exchange process is dictated by a net head difference between the aquifer and the river. Figure 10 is a conceptual model of the mechanisms controlling ground water circulation between the Devil’s Ear cave system and the Santa Fe river. Though the entire geologic section beneath the river including the cave system is part of the Floridan aquifer, for the purpose of this discussion, the river-aquifer system is divided into three parts: the Santa Fe river, the conduits in the aquifer, and the saturated Ocala Limestone in between.
The 30-m thick section of saturated Ocala Limestone above the cave system is a leaky confining layer separating the extremely permeable conduits from the overlying river. The confining layer allows a hydraulic gradient to develop between the river and the cave and contains water in storage that is displaced when either a downward or upward hydraulic gradient develops. The rate at which water moves through this layer is dependent on the magnitude and duration of the hydraulic gradient.
The river receives water from both runoff in the highlands and spring discharge in the lowlands. Therefore, head in the river is dependent on both the quantity of aquifer discharge to the river and the quantity of surface runoff received by the river and it’s tributaries. In contrast, head in the cave is primarily dependent only on the quantity of recharge. Thus, surface runoff, which is greater in regions where the aquifer is confined, is the independent variable that causes the head difference. The magnitude and direction of the hydraulic gradient between the river and the cave is, therefore, a direct function of the distribution of major regional precipitation events.
When precipitation is concentrated on the Northern Highlands where the Santa Fe river is not in hydraulic connection with the Floridan aquifer, the water accumulates in the river as overland flow and the flood pulse moves downstream onto the unconfined part of the aquifer. The subsequent increase in river stage produces a downward hydraulic gradient causing large amounts of river water to invade the cave through the leaky confining layer (Figure 9a). Observations of water clarity reductions in the cave, by the author as well as other cave-divers, after large flood events originating in the highlands of the upper Santa Fe river reveal that river water intrusion to the aquifer takes place in as little as one day.
Figure 9. Ground water – surface water exchange between the Santa Fe river and the extremely permeable caves in the Floridan aquifer across a 30 m thick section of Ocala Limestone which is a leaky confining layer. (A) Rainfall in the highlands creates a flood pulse in the river that raises the stage of the river above the potentiometric surface of the aquifer creating a downward gradient. (B) Rainfall in the lowlands infiltrates directly to the aquifer raising the potentiometric surface above the stage of the river creating an upward gradient. Copyright Todd R. Kincaid.
Conversely, when precipitation is concentrated on the lowland regions where the Floridan aquifer is unconfined, recharge to the aquifer results from direct infiltration with no resulting flood wave in the river. The hydraulic head in the cave rises above that of the river where a rising river stage is caused only by increased spring discharge. The resulting upward hydraulic gradient results in flow from the cave to the river (Figure 9b). The water in the cave will clear as the tannin surface water from the Santa Fe river is flushed up and out through the confining layer.