The Tatacoa Desert (TD) is a Cenozoic basin between the Central and Eastern Cordilleras in the southern Colombian Andes. Its sparse vegetation and Badlands morphology resemble a semi-arid environment, despite receiving ∼1,300 mm of mean annual precipitation. Based on optically stimulated luminescence dating in two fluvial terraces and drainage morphometric analysis, we propose that such “aridification” was imposed by a drainage reorganization during the Late Pleistocene. No river coming from the neighboring mountains flows into the TD. This makes the Tatacoa rivers unable to balance the evapotranspiration effect, causing a surficial water deficit. A fluvial terrace (TAT-2) works as a divide between the Tatacoa catchments and the Cabrera River, the latter coming from the Eastern Cordillera and bounding the TD towards the north. Paleocurrent measurements on its sedimentary lithic clasts point to a W-NW-directed flow into the TD. Thus, OSL ages around 74.1 ± 7.0 ka in this terrace suggest the TD was drained by an Eastern Cordillera catchment during the Late Pleistocene. At the western end of the TD, a dissected fluvial terrace (TAT-1) lying 80 m above the Magdalena River ages between 6.3 ± 1.5 and 7.7 ± 1.5 ka, representing the Holocene incision rates of 10.8 ± 0.3 mm/y. Coupled OSL data and river longitudinal profile analysis show a high susceptibility to erosion of the Tatacoa rocks and estimate incision rates ranging from 2–5 mm/y to 10–20 mm/y within the TD. The higher rates are concentrated along knickpoints related to anticlines and thrust faults. Thus, a northward shift of the Cabrera River and a westward migration of the Magdalena Valley have drastically reduced the surficial water availability in the Tatacoa and the development of vegetation, increasing erodibility. Potential climatic and tectonic drivers include 1) terrace accumulation during the wet periods of Marine Isotope Stage 5A and 1; 2) fault-induced incision of the Magdalena River in the NW, which would have favored northward migration of tributaries such as the Cabrera River; and 3) activity on the Baraya thrust fault, which would have prevented rivers from the Eastern Cordillera to enter the desert.
Read full abstract