| Proposal | Pilot Study | Geology and Hydrogeology of the Fault-Partitioned Hickory Aquifer |
|---|---|---|
| Executive Summary | Pilot Study Objectives and Results | Abstract, Introduction |
| Stratigraphy | ||
| Proposal | Structure | |
| Hydraulic Compartmentalization | ||
| Hydraulic Conductance of Faults |
The primary groundwater data are water-level elevation measurements in wells distributed across the study area. Measurements were made on a regular basis from 4/15/88 to 5/7/89, which encompasses one complete irrigation and recovery cycle. From 1990 to present, water levels have been measured in the late Spring of most years. Well groups straddling faults have been measured more frequently. Since the original study in 1988-90, five additional wells have been drilled or became available for measurement in the area and provide additional constraints on earlier interpretations. Results of five pump tests also provide insight into aquifer response characteristics, the nature of hydraulic communication over a short time interval for wells in the region of influence of the pump test, and effects of specific faults on well-drawdown characteristics.
The presentation below is a summary of an in-depth presentation (Randolph, 1991; MS thesis, TAMU) of the groundwater data and analysis. Emphasis here primarily is on interpretations derived from analysis of the groundwater data.
The study entails data from 28 (32 in 1995) uncased wells distributed across the study area . Most wells penetrate to the granite basement, except in the northern part of the area (Region D and the NE Region). Fourteen wells are used primarily for irrigation. Irrigation wells typically pump at rates of 300-500 gpm, and during a typical summer irrigation period each well produces in excess of 100 acre-ft (0.6 million barrels).
The pumping-induced hydraulic response of the Hickory Sandstone aquifer depends upon the stratigraphic position of the water table/piezometric surface. If the Hickory Sandstone is fully-saturated and overlain by the low permeability Cap Mountain Limestone, the aquifer responds as a confined aquifer. Pump tests reveal that when the potentiometric surface is positioned within the Hickory Sandstone, the hydraulic response is that of a transversely anisotropic, unconfined aquifer. The higher the stratigraphic position of the potentiometric surface, the greater the degree of transverse anisotropy, which reflects the increase of laterally extensive, clay-rich interbeds in the middle Hickory. Results of three interpretable pump tests indicate average horizontal permeabilities of 0.5 to 1.2 Darcy, with ratios of vertical to horizontal permeabilities ranging from 0.2 to 0.001. Permeability measurements on core samples with clay and silt laminae show that the permeability parallel to bedding is often several orders of magnitude larger than the permeability perpendicular to bedding.
The elevation contour map of the water-level data for 4/17/89 (Figure, 49K) shows an interpretation of the configuration of the water table that incorporates the effects of faults on the groundwater system as deduced from the extensive analysis of groundwater data. The water-level data are contoured to account for effects of known or inferred low conductance faults and no-flow boundaries identified in the analysis stage. No-flow boundaries are assumed where the saturated zone of the aquifer is faulted against low permeability granite; hence, contour lines terminate against and are perpendicular to a no-flow boundary. Contour lines obliquely intersect known or inferred low conductance faults where leakage across the fault is inferred. The contour map is highly speculative in areas with little or no well data; these specific areas are: 1) the central corridor north of Region A and between Regions B and C; 2) Region E and the southern portion of Region D; and 3) the Eastern Region. Spatial variations of the contour spacing clearly illustrate that the hydraulic gradient changes significantly across the area. These differences reflect either singly or in combination: 1) differing average specific discharges or 2) differing aquifer permeabilities. In some areas, such as Region A, the variation of hydraulic gradient along the flow direction may reflect a systematic change of the wetted-cross sectional area of the saturated zone in the aquifer.
The observed spatial and temporal variations of water levels in wells in the study area clearly show that faults impede lateral flow of groundwater and influence both the short- and long-term hydraulic responses of fault-defined regions. The impedance of groundwater flow across a fault is indicated by: 1) an anomalously large head change across the fault, 2) a significant variation of the hydraulic gradient in proximity to and across the fault, and 3) poor or indiscernible hydraulic communication between wells on either side of the fault. One or more of these diagnostic features is observed associated with several of the faults that separate fault-defined regions. Most notable are: 1) fault F2, which separates the Western Region from Region C, 2) the fault separating Region A from Region C, and 3) the faults separating Region B from Regions C and E (Figure). Fault F1 between Region A and B and the inferred fault separating Regions C and D also affect groundwater flow, but their effects are smaller compared to the other faults.
The study area can be subdivided into at least three major hydraulic compartments that correlate to the structural partitioning of the aquifer (Figure, 37K). A major hydraulic compartment is comprised of one or more fault-defined regions. Identification of a hydraulic compartment is based upon the degree of hydraulic communication between adjacent fault-defined regions and their similarity or dissimilarity of irrigation-induced short- and long-term water-level variations. The graben is subdivided into two major hydraulic compartments: compartment I consists of structurally-defined Regions A and B, and compartment II consists of Regions C, D and E. The Western Region comprises compartment III. The degree of hydraulic communication between the NE Region and hydraulic compartment II is uncertain. Faults that separate the major hydraulic compartments have a displacement in excess of 100 ft (30 m).
Compartments I and II within the graben are subdivided further into hydraulic subcompartments by faults with smaller displacements. The subcompartments correlate with specific fault-defined regions.
Several types of data support the proposed subdivision of the local groundwater system into three major hydraulic compartments. Poor hydraulic communication exists between wells in neighboring hydraulic compartments, such that irrigation pumping in one compartment induces in the neighboring compartment an anomalously small drawdown relative to that expected in a laterally uniform aquifer. Relative to faults that separate major compartments, a greater degree of hydraulic communication exists across faults separating the hydraulic subcompartments; fault F1 and the inferred fault separating Regions C and D exemplify faults that subdivide major hydraulic compartments into subcompartments. Faults that separate major compartments are not impermeable but have small hydraulic conductances. Flow into and out of a compartment probably reflects the relative positions of the compartments and the spatial variation of the conductance of the bounding faults. The observed and inferred hydraulic gradients differ from compartment to compartment. Hydraulic gradients in compartments I and III appear to be at least twice as large as that determined in the southern and northern parts of compartment II. The difference in hydraulic gradients may reflect either a higher average permeability of the aquifer in compartment II or a smaller specific discharge in compartment II relative to the other two compartments. If the latter is the case, this implies that the flow into compartment II from compartment I is significantly inhibited by the faults separating the two compartments.
Other characteristics of the spatial and temporal water-level variations also support the differentiation of the aquifer into hydraulic compartments. The residual drawdowns (drawdowns after 5-12 days of recovery) of wells within a compartment are similar but differ significantly across compartment boundaries (Figure, 33K). Differences within a compartment usually correlate with a subdivision of the compartment by a fault; for example, the residual drawdowns in Region B are approximately twice as large as those in Region A, yet there is a high degree of hydraulic communication across fault F1. The observed residual drawdowns reflect a combination of aquifer response characteristics, volume of water extracted during the irrigation interval and the areal extent of the aquifer readily giving up water from storage. The large residual drawdowns observed in Region B correlate with a large volume of water being extracted from a relatively small aquifer subcompartment.
Annual water-level changes exhibit a spatial pattern similar to that of the residual drawdown and are consistent with the division of the groundwater system into major hydraulic compartments. Annual water-level changes, however, do not readily differentiate the hydraulic subcompartments. The cumulative effect of differing water-level declines over the long term in the hydraulic compartments is reflected in the historical water-level declines. The marked difference of the water-level declines in the past 15 to 20 years for the Western Region (10 ft), the Eastern Region (25 ft), Region A (81 ft) and Region C (40 ft) clearly show the significantly different long-term hydraulic response of the three major hydraulic compartments.
The characteristics of the water-level histories of wells during the irrigation/recovery cycle differ significantly (Figure, 43K). Wells within the same hydraulic compartment have more similar histories compared to wells in other compartments. Differences between wells within the same hydraulic compartment appear to reflect primarily the subcompartmentalization, although certain details reflect a differing aquifer response.
