Cavidades naturales en rocas magmáticaslas cuevas en rocas plutónicas

Abstract

Caves developed in granite terrains are landscapes little know so far. But despite the few and scattered references in the scientific literature to such type of caves, the Atlas on Pseudokarst written by CHABERT & COUBORN (1997) includes a hundred of large cavitites located in plutonic rocks, a number which is increasing every day. This Thesis is in general focused on the caves developed in granites (plutonic rocks) and especially on the large granite cave systems of Iberia Peninsula. In a first phase they will be mapped erosional forms, structural forms and depositional (mainly speleothems and biospeleothems) forms, to develop specific geomorphological maps (plan and elevation views, crossover profiles, 3D models), on which will be intented to define a chronology for these cavities and also the active and fossil dynamics which are present in this type of underground environment. These caves can be of three types. The first type are fissure caves, which are developed taking advantage in those fault or fracture planes where weathering proccesses was well developed and where the alterites were removed after by erosion. The second type, and here belongs to the majority of the samples that will be studied in this Thesis, have been defined taking advantage of existing voids in slope deposits formed by gravitational collapse of blocks. In most cases the size of the gaps among blocks depend entirely on the dimensions of the blocks. Some times when the accumulations of blocks occur at the valey bottoms, the runoff waters are channelized through these gaps, although the water scarcely produces changes in space than pot holes or erosion channels. In other cases boulder caves have been developed by opening convex forms (tor, castle kopje, inselbergs, etc), (that is) in which previous rock structure has been opened by rock(y) creep giving place to the cavity due to the separation of blocks. The deposits inside these cavities preserve sedimentary records of different entity from which we can get information on the evolution of the underground environment since the cave is formed. The third type of caves, of great singularity though more modest dimensions, are tafoni caves. Here the cavity is limited to the scope of the block affected by the process. From the three types of granite caves, this Thesis is specifically focused in the second one. The first part of this Thesis studies the genesis of these caves, and so defining as far as possible, the involved mechanisms that produce block and boulder movements and also the chronology of these movements, both aspects often intractable especially by the lack of an appropriate criteria. The second part of the Thesis studies the microclimatic characteristics that defining the underground environment. Our objective is to build a mathematical model (black-box or gray-box model) which copy the underground microclimatic dynamic inside these type of caves. Many works on karst caves assume that caves are causal dynamic systems, continuous, invariant on time and lineal. However the first data analyzed shows that these type of caves should be processed using no-lineal dynamic system identification strategies. So mathematical models will be developed using artificial neural networks, mainly those specialized on processing time series. These networks are characterized by their learning, generalization and abstraction capacities, and so they are capable to develop and extend continuous models from broken datasets (series). Multiparametric mono or multiobjective optimization will be solved by heuristic approaches. As previous phase to system identification, it must be determined the correlation among physic variables that interact in the system and that in our problem all they are long not-stationary time series. We will use wavelet transforms due they are an specialized tool to study not-stationary process, multiscale, and developed both in finite spatial and time domains. Another part ot this Thesis is focused on the assessment and risk analysis of these systems: On one hand many of these cavities are draining important water courses and so there is a risk of avenues especially dangerous for visitors. Also by the type of lithology and when there isn't drainage through the cave, a potential risk arising from high levels of radon are added, which can become these cavities in traps or natural deposits of this type of gas and their children radioactive subproducts (cesium). On the other hand these caves are accumulations of blocks of large thickness which requires to analyze stability and their potential risk of reactivation.