2011 IR&D Annual Report

Investigation of the Effect of Epikarst on Recharge and Storage of Groundwater in Karst Aquifers, 20-R8220

Principal Investigators
Ronald Green
F. Paul Bertetti
Ronald McGinnis
Hakan Basagaoglu

Inclusive Dates:  04/01/11 – Current

Background — Increased demands for groundwater have elevated the need to better understand the hydraulics and sustainable yield of aquifers, a major source of water resources. For karst aquifers, a critical need is to better understand how the epikarst layer affects recharge and storage of groundwater. The term epikarst describes the weathered horizon at the top of the vadose zone with high porosity and permeability. Karst aquifers exhibit complex hydraulic reactions in response to recharge events caused by the development of dissolution flow pathways at the surface and at depth. The epikarst layer has a strong spatial and temporal influence on recharge of the aquifer. Epikarst layer effects vary with geographic location, scale of aquifer and the hydrogeologic characteristics of the aquifer of interest. The hydraulics of epikarst at local scales has been characterized with limited success, and formulation of a universal hydraulic conceptual model for the epikarst layer has proven elusive. This project will investigate the hydraulics of epikarst layers in karst aquifers to understand the mechanisms of recharge and groundwater storage in regional-scale karst aquifers and to develop a universal or general model for the hydraulics of epikarst layers.

The hydraulic effect of epikarst on recharge is exhibited by the lag time between the occurrence of recharge events and times when the groundwater elevation and spring discharge respond to precipitation. Not accounting for the lag introduces large errors into groundwater flow models and renders water-resource management decisions inaccurate. Recharge lag time can be determined using hydraulic or water chemistry responses observed in groundwater or spring discharge. Lag times can vary from hours to months depending on the size of the recharge zone, the climatic season of interest (i.e., rainy or dry), the physical nature of the epikarst, and related hydraulic features of the aquifer. In general, larger recharge zones or catchment areas with a more extensive or better developed epikarst layer will have longer lag times. Lag times determined solely by a hydraulic response may be misleading, particularly if the response caused by a pressure pulse or piston flow in which case the response is not reflective of actual groundwater velocity. It is prudent to examine chemical and isotopic tracers, such as trace and major elements, temperature, alkalinity, specific conductance and isotopes, in addition to hydraulic responses to discern actual groundwater flow mechanisms and to parse out the source of the sampled water.

Approach — The technical objective of this project is to quantify the effect of epikarst on recharge of a karst aquifer. To accomplish this, the project will physically characterize features important to the hydraulic capacity of a prominent karst aquifer with a defined epikarst layer; characterize hydraulic inputs (i.e., spatial and temporal distribution of precipitation) and responses exhibited by the regional aquifer system (i.e., spring discharge quantity and chemistry, regional and perched groundwater elevation); and develop a mathematical or quantitative representation of the hydraulic effect of epikarst by relating input to output as a function of system properties. The Edwards-Trinity Aquifer in west-central Texas will serve as the study site. The Edwards-Trinity Aquifer is a large, regional-scale karst aquifer located sufficiently close to San Antonio to facilitate field surveys and data acquisition. The first task will examine the hydrostratigraphy of the Edwards-Trinity Aquifer. This examination will include field surveys, water chemistry testing and analysis, hydraulic response data collection, and assessment of the geological structure. The second task will integrate all relevant data and assemble a coherent conceptual model of the hydraulic mechanisms of the epikarst. The importance of scale dependency to the hydraulic response needs to be recognized and fully incorporated into the conceptual model. The last task will develop a model or algorithm that accounts for these mechanisms in groundwater recharge and storage in karst aquifers.

Accomplishments — To date, research focused on evaluating the internal relationships of recharge, storage and discharge of the targeted study area in the western Edwards Aquifer and the southern Edwards-Trinity Aquifer. Correlations among 39 rain gauges, two index wells (i.e., J-17 and J-27), and two springs (i.e., Comal and San Marcos springs) were analyzed. Results from these analyses indicated low correlation (i.e., no greater than about 0.1) when the entire period of record of the rain gauge data was considered. These results were interpreted that correlations among recharge areas, well locations and spring discharge overlapped both temporally and spatially when taken for long time periods. Subsequent analyses have been undertaken to assess correlations for shorter time periods. Particular emphasis is taken to target a specific time period with a large precipitation event followed by a period of drought or limited precipitation. Such an analysis hopefully will allow assessment of correlations not obscured by multiple overlapping signals. The large rain events of late 2009 and early 2010 followed by prolonged minimal precipitation are targeted for this analysis.

Potential springs and discharge points for water sampling and water chemistry analysis were identified. Protocol for water sampling and target water chemistry parameters have been specified. Two multi-sample in situ samplers were purchased for project use. Each sampler is programmable and capable of collecting up to 24 water samples during specified sampling periods.

Conceptual models for the targeted focused-study areas have been formulated based on available information. Additional data and site-specific hydrogeological information have been compiled to augment and vet these models. Reservoir modeling packages MATLAB, STELLA, and VENSIM have been evaluated for precipitation/groundwater storage/spring discharge assessments. Input data collected during the project are being used in these reservoir modeling exercises. At this time, a simple reservoir model is conceptualized that has a fast and slow phase to recharge, an aquifer storage component and a spring discharge component. The timing of fast recharge, change in storage, and spring discharge is known. The timing of slow recharge and the relative fractions of fast and slow recharge are unknowns. Various candidate lumped-parameter, reservoir models are being assessed to understand the effect of epikarst delaying recharge (i.e., instigating the slow recharge component) into the reservoir system.

Benefiting government, industry and the public through innovative science and technology
Southwest Research Institute® (SwRI®), headquartered in San Antonio, Texas, is a multidisciplinary, independent, nonprofit, applied engineering and physical sciences research and development organization with 10 technical divisions.
04/15/14