Correcting for inclination shallowing of early Carboniferous sedimentary rocks from Kyrgyzstan-indication of stable subtropical position of the North Tianshan zone in the mid-late Palaeozoic

Abstract

High-quality palaeomagnetic data for the early Carboniferous of Central Asia are scarce and the palaeogeographic evolution of this area prior to final amalgamation of the region east of the Ural mountains is still rather obscure. Here, we present palaeomagnetic data for early Carboniferous deposits from two areas in the Kyrgyz North Tianshan (NTS). Detailed rock-magnetic analysis indicates the presence of magnetite and haematite as magnetic carriers in these red sediments. In the Kazakh basin section (KEL), we identify a high-temperature component (HTC) of magnetization during stepwise thermal demagnetization at temperatures of up to ~680 °C yielding a site mean direction of D = 176.2°, I = −36.4°, k = 57.4 and α95 = 8.9° after tilt correction. Two HTCs of magnetization were identified in samples from the Sonkul Basin (DUN) with maximum blocking temperatures of ~600 °C (magnetite) and ~680 °C (haematite). The magnetite component was also identified with alternating field demagnetization. The resulting site mean directions for these two components identified in 16 and 14 sites, respectively, are D = 149.3°, I = −50.3°, k = 73.6 and α95 = 4.3° for the magnetite and D = 139.6°, I = −35.1°, k = 71.6 and α95 = 4.7° for the haematite component. All three mean directions show a significant increase of the precision parameter k after tilt correction indicating acquisition of the high-temperature magnetization prior to the main folding event in the Jurassic. We explain the difference of the two components of DUN by a process of inclination bias due to compaction to which the platy haematite particles are more susceptible. Applying the elongation-inclination (E/I) method to directional data from over 100 individual samples from location DUN results in a negligible correction for the magnetite component (<5°), whereas the inclination of the haematite component corrects from −35.0° to −50.3° (f = 0.6, error interval −41.4° to −57.9°), which is then equal to the uncorrected magnetite inclination.The small number of samples from section KEL does not allow application of the E/I technique and inclination correction based on high field anisotropy of isothermal remanent magnetization was applied, yielding a corrected inclination of −75.2° ± 4°. Assuming comparable degrees of compaction for both study areas and applying the flattening factor obtained in DUN on samples from KEL, however, would result in comparable inclinations. The identification of inclination shallowing at both sections indicates that the age of magnetization is close to the deposition age. Assuming a reversed polarity of the directions from both areas results in palaeolatitudes of ~30°N for section DUN and ~60°N for the anisotropy-based correction of section KEL. The large difference, however, is geologically very unlikely. The inclination of the magnetite component of DUN (unaffected by inclination shallowing) favours a palaeoposition of ~30°N. This is supported by the inclination shallowing corrected haematite component of DUN yielding a comparable inclination. Therefore, our results indicate that the NTS domain was situated at ~30°N in the early Carboniferous. Furthermore, the NTS zone was probably not connected to Baltica or Siberia prior to the late Palaeozoic.