Ocena regionalnej i lokalnej zmienności składu piaskowców warstw krośnieńskich metodą analizy wariancyjnej

Abstract

Regional and local variation of composition of the sandstones of the Krosno beds (Flysch Carpathians) estimated by ariance analysis The method of multivariate analysis of mineral composition (Krumbein and Tukey, 1956) was, used in the present study of the sandstones of the Krosno beds. The calculations were made with the use of the ELWRO ODRA 1204 computer, in the Computer Center of the Jagellonian University. STRATIGRAPHY AND PALAEOGEOGRAPHY OF THE KROSNO BEDS OF THE SILESIAN SERIES OF THE CARPATHIAN FLYSCH The present study is concerning the thick-bedded graded sandstones of the Lower Krosno beds of Oligocene age, in the Silesian Series of the Carpathian flysch. The lower boundary of the Krosno beds is diachronous in the sedimentary basin of the Silesian series, being stratigraphically younger in the north. In the southern and western part of the basin, the Lower Krosno beds are ending the flysch succession as the youngest beds present, while in the northern and eastern part of the basin they are covered by the Upper Krosno beds. The thick-bedded sandstones analysed in the present study form the basal member of the low er Krosno beds, and occur below the m arker horizon of the calcareous Jasło shales. Therefore the analysed sandstones are roughly coeval in various parts of the sedimentary basin (Fig. 1). The sandstones of the Lower Krosno beds were deposited by turbidity currents, and the m ajo rity of beds is graded. Studies of palaeocurrent 'directions and facial development (Dżułyński and Ślączka, 1958), provided evidence that the clastic material of the Krosno beds was derived from various source areas. Two of these source areas are well individualized, namely the western source area — i. e. the Silesian cordillera of the Carpathian geosyncline, and the eastern source area, i. e. the Maramures massif. The third, north-western source, the presence of which is inferred from palaeocurrent directions in the Upper Krosno beds, and the presence of exotic pebbles and reworked flysch rocks, is somewhat less distinct than the two other ones. The main problem arising from these previous studies w as: to what extent the provenance of material is reflected in the composition of sandstones? THE SAMPLING SYSTEM The hierarchical sampling system was designed with the aim to separate the variation due to local effects from the regional variation of the mineral composition of the sandstones (Fig. 2). The sedimentary basin of the Lower Krosno beds of the Silesian series was divided into three regions, according to the provenance of the clastic material of the sandstones. Two cross sections, one proximal, the other distal with regard to the inferred position of the source area were selected in each region. Ten beds were sampled in each section. Two samples per bed w ere taken, one from the base, the other from the top of the bed. The regions and the location of the sampled sections are shown in Fig. 3. The sampling system perm its to analyse the variation of mineral composition on the following levels: - between samples in the beds, - between beds in the sections, - between sections in the regions, - between regions in the whole basin. PETROGRAPHY The mineral composition of the samples of sandstones was determined in thin sections; with the use of an ELTINOR point counter 200 points per thin section were counted. The content of the following 8 constituents was recorded: quartz, feldspars, micas, metamorphic rocks fragments, limestone fragments (bioclastic, micritic, sparitic limestones) zoogenic limestone fragments (Stramberg type), intraclasts (fragments of shales and fine-grained sandstones) and cement+matrix. The number of analysed samples was 120, and the total number of observations (determined content of a single component in a sample) amounted to 960. The composition of the analysed sandstones is presented on Table 1 and Table 2. STATISTICAL HYPOTHESES The results of the analysis of variance of the mineral composition of the sandstones of the Lower Krosno beds are presented in Table 3. The interactions of the various levels of hierarchical sampling and the mineral constituents were used to test the following statistical hypotheses (Snedecor’s F test): 1. Ho: the variation of composition between regions is not significantly greater than the variation between sections in the regions. 2. Ho: the variation of composition between sections in the regions is not significantly greater than the variation between the beds in the sections. 3. Ho: the variation of composition between beds in sections is not significantly greater than the variation between samples in beds. 4. Ho: the variation between samples in beds is not greater than the residual variation in a homogenous mixture of the constituents. These hypotheses w ere tested for individual regions, for pairs of regions and for all three regions. The results of tests of the statistical hypotheses are given in Table 3 and summarized in Table 4. CONCLUSIONS The first hypothesis is accepted for all pains of regions and all three regions. It is concluded, that the differentiation of material derived from the various source areas is small. The individual character of the eastern source area (region 3) is better marked by qualitative data on the range of rocks occurring among the fragments of metamorphic rocks and by the lack of zoogenic limestones than by the overall quantitative composition. The second hypothesis is rejected, and it is concluded, that the variation of composition between sections in the regions is greater than between the beds in the sections. This indicates, that the transport of the clastic material on the distance between the proximal and the distal section results in a composition sorting, which is presumably associated with size sorting and selective transportation of micas. The third hypothesis is rejected for the regions 1, 2, for the pair 1 and 2, and for all three regions. In these cases it is concluded that the variation of composition between beds in profiles is greater than between samples in beds. The geological interpretation of this conclusion comprises the following possibilities : - the mineral composition is closely related with grain size, and the variation of grain size distribution parameters between beds is greater than the difference of grain-size distribution parameters at the base and the top of the graded beds. This may imply, that the competence of individual turbidity currents varied considerably; - the variation of mineral composition with in a section is reflecting the erosion of the source area, and consecutive degradation of rocks diferring in pétrographie character; - the structure of the source area was complex and the composition of the clastic material deposited in the littoral zone before redeposition by turbidity currents varied in space and time; thus the individual turbidity currents originating at various points along the shores of the cordillera (e. g. in various submarine canyons) carried different material; - all the above factors operated simultaneously. In the eastern region the variation of composition between beds in profiles is not greater than the variation between samples in beds. It is concluded that the eastern source had a more homogenous composition than the western source or sources. This individualizes the eastern source, while the presumed two western sources corresponding to the regions 1 and 2 do not display differences in their pétrographie character. The more uniform character of the material from the eastern source is reflected also in the pairs of regions: 1 and 3, 2 and 3. The fourth hypothesis is rejected for all regions and all groups of regions. It is concluded therefore, that the mechanism of deposition of graded beds by turbidity currents effectively differentiates the composition of the detrital material. The sandstones of the Lower Krosno beds are characterized by a local variability and a regional homogeneity of their mineral composition. The local variability is related with the processes of transport and deposition of the clastic material by turbidity currents.