DEPTHS OF DENUDATION IN REMOTE CRATONIC INTERIORS. AN APPRAISAL IN BURKINA FASO (WEST AFRICA) BASED ON FISSION-TRACK ANALYSIS AND THE DATING OF LATERITE

In stable cratonic regions, most tropical weathering mantles record long-term environmental change and develop over long time scales. They may thus also reflect tectonism and its denudational signals as interpreted through apatite fission-track thermochronology, cosmogenic isotope analysis (CIA) and the age-bracketing of laterite by ⁴⁰Ar/³⁹Ar dating of cryptomelanes. Using standard fission-track annealing algorithms, denudation averaged 10-15 m/Myr until the Eocene and subsequently accelerated to 30-35 m/Myr across the Leo uplift of Burkina Faso. Such figures, however, are not immediately consistent with other field and radiometric data. Flighted levels of residual lateritic cap-rock suggest average rates of 3-10 m/Myr for the surrounding topography. The uppermost level is Eocene and bauxitic (50 Ma), while the following generation is late Oligocene. Depths of denudation since 24 Ma have not exceeded 0.08 to 0.17 km. CIA estimates of erosion for laterite cap-rock in Burkina indicate maximum denudation rates of 2 m/Myr, highlighting its long-term stability as a stratigraphic marker.

An understanding of the relationship between tropical soil processes and global tectonics at timescales of 1 to 100 Myr needs to rely on such stratigraphic tools as described above. Reliable radiometric rates, however, must be based on well calibrated methods, whether applicable to field exposures or large regions. Altogether, the "face value" AFT denudation figures of this study overestimate field-based post-Eocene rates by a factor of three. Related discrepancies between stratigraphic or geomorphic and AFT data have been highlighted in other geodynamically stable environments. A possible cause for this may be the inbuilt and debatable assumption, in AFT modelling, that fission tracks do not anneal at temperatures below 60 °C. The hope of discovering compatibilities in denudation rates across time and spatial scales thus also relies on the compatibility between the methods used in measuring them. Careful cross-correlations and a re-examination of methodological underpinnings when field-supported mismatches occur may assist in reconsidering some existing single-proxy estimates of denudation.