Overlapped paleomagnetic vectors and fold geometry: A case study in the Balzes anticline (Southern Pyrenees)

Abstract

When two or more magnetic components share the same temperature and/or coercivity spectrum in a rock, the paleomagnetic components are said to be overlapping. Failure to separately identify these components in the laboratory may lead to substantial errors in the declination and inclination of the primary component (including apparent changes in polarity and with a marked impact in stability tests). Recently, numerical models have been proposed to understand this problem in relation to fold geometry; these allow the expected declination and inclination errors at different locations across the fold surface to be identified and estimated. In this paper, we present a real case study of overlapped paleomagnetic directions from the northern part of the Balzes anticline (South Central Pyrenees). A new dataset is established based on 206 paleomagnetic samples taken in a wide variety of structural locations. Several facies including limestones, siltstones and marls from Eocene marine and deltaic sediments were studied. This dataset shows the occurrence of paleomagnetic overlapping controlled by lithology. That is, most paleomagnetic directions derived from marls show hardly any significant declination and inclination errors and they are consistent in all structural positions. In contrast, in most of the limestones there is strong overlapping and significant errors related to the fold geometry. The dataset also enables the evaluation of errors as a function of structural position. The numerical model for Balzes anticline overlapped directions shows qualification and quantification of errors in all possible structural positions and the effect of overlaps on the fold test is also analyzed. Through the use of numerical modeling we propose a new tool to deal with data with a strong overlapping of components as in this case study. The identification and quantification of the overlapping ratio potentially enables data to be filtered to eliminate such errors and also allows suitable structural positions where error is minimized to be identified.