For a Better Understanding and Management of Karstic Aquifers: Characterization of the Properties of the Big Spring Karstic System (Missouri)

The results obtained from the study of karstic systems over the last several decades have clearly emphasized their specificities. These specificities are mainly due to the organization of voids in the karst, which, due to the mechanisms of their emplacement, are distributed in an extremely heterogeneous way. The result is an anisotropic permeability, but above all structured according to their position within the karst massif. Moreover, all the voids do not have the same function with respect to the movement of fluids. In saturated media (saturated karst), some voids (“the drains”) have a transmissive function and ensure the propagation of waters. Other systems that are connected to the drainage give an account of their storage and corresponds to the storage function. These properties result in a highly non-linear hydrodynamic behaviour with all its consequences: sensitivity to initial conditions, high unpredictability, and impossibility of using simple deterministic modelling. Due to the permeability contrasts between drains and ancillary systems, depending on the duration of the events (floods), only parts of the aquifer are involved in the hydrodynamics. Therefore, over time the karstic aquifer possesses a variable geometry.

To better study, exploit, and manage karstic aquifers, a new approach using systemic analysis is currently used by Ramboll Environ. The overall behaviour of the aquifer is no longer considered to be the result of the sum of spatially individualized behaviours, which is assumed and imposed by the differential approach, but as the sum of the interactions of these individual behaviours that most of the time are unknown. The karst is thus conceived as a set of dynamic processes (flows) whose study and understanding must lead to the identification of its properties, allow a description, and ultimately facilitate its exploitation and management. While this concept is easy to understand, and seems able to solve the difficulties encountered, its implementation remains subtle.

The Ramboll Environ approach derives from a philosophy different from the conventional hydrogeological approach and requires abundant robust collection of continuous data, using transducers of the various parameters that monitor the functioning of the karstic aquifer (discharges, water levels, rainfall, etc.). This approach involved the development of a set of methods and high-performance software to extract information from the collected data in order to recognize the physical signatures that are responsible for the observed hydrodynamic behaviour, and thus, whatever the difficulty encountered, including non-linearities, characterize the dynamics of systems and predict their behaviours. These methods are called correlative and spectral analyses, continuous or discontinuous wavelet analyses, rescaled range analysis, fractal or multifractal analyses and attractor analyses.

The application of these analyses to Big Spring, Missouri allowed the establishment of an “Identity Card” of the spring, encompassing all information necessary for an optimal management of the spring.