A NEW DIRECT RADIOCARBON AMS DATE FOR AN UPPER PALAEOLITHIC HUMAN BONE FROM SIBERIA

A new AMS radiocarbon date has been obtained from a Siberian human bone sample taken directly from a cranial fragment. The specimen comes from the north‐east Siberian site of Pokrovka 2 (55⁰19′46.30″N, 92⁰26′48.80″E). The remains consist of the anterior portion of the cranium or the superior part of the face; that is, the forehead and the roof of the eye sockets. Sadly, indifferent features of the frontal bone do not allow a confident diagnosis of the sex of the individual. The evidence suggests, however, that the remains are probably those of a young adult (teenager) and an anatomically modern human, rather than a Neanderthal. Radiocarbon dating was undertaken at the Oxford Radiocarbon Accelerator Unit (ORAU). The radiocarbon age determination of the bone fragment, 27 740 ± 150 bp (OxA‐19850), is one of the earliest direct dates for a modern human from Siberia. The Pokrovka cranial fragment dates to the middle Upper Palaeolithic of Eurasia and is broadly contemporary with the human remains from the famous eastern European site of Sunghir. When the radiocarbon age is compared with the Cariaco Basin ¹⁴C data set and the Greenland oxygen isotope record of NGRIP, it corresponds with Greenland Interstadial 5, a warmer phase of the Last Glacial period, although this is a preliminary conclusion, and is based on climate records that may or may not be wholly synchronous between the two locations.     

Rugged Plateaus and Extensive Floodplains — Modelling Landform Evolution in a Northern Australian Catchment

Medium to large natural catchments are often more spatially heterogeneous than small catchments or single landforms. Attempting to model landform evolution of large areas is consequently more complex. This paper demonstrates that modelling landform evolution in medium to large catchments can be improved by calibrating the model to smaller, more geomorphologically homogenous sub‐catchments. The paper investigates landform evolution in the Ngarradj catchment in the Northern Territory of Australia (a medium scale catchment of approximately 67 km²). The catchment is complex and contains two distinct landform regions; an upland plateau region with highly dissected sandstone and shallow, sandy soils, and a lowlands region with gentle, wooded slopes and floodplains with deep, sandy soils. The SIBERIA landform evolution model is calibrated and applied to the Ngarradj catchment. The complexity of the Ngarradj catchment is incorporated into the modelling by dividing the catchment into three sub‐catchments (Swift Creek (SC), Upper Main (UM) and East Tributary (ET)) which are relatively homogeneous and for which hydrology and sediment transport data are available. A discharge‐area relationship and long‐term, sediment loss rates for the catchment are derived based on an annual series flood frequency analysis of a 20 year runoff record predicted in a previous study. Sediment transport modelling incorporates both suspended and bedload sediment loss. The denudation rates derived using these data are 37, 63 and 77 mm kyr^?1 for the SC, UM and ET sub‐catchments, respectively. Model predictions indicate that the UM sub‐catchment will have the greatest mean erosion. This is balanced by the large amount of deposition that will occur in the upper Ngarradj valley of the UM sub‐catchment. Further deposition occurs on the floodplain of Ngarradj, with the area between the SC and ET/UM (up‐stream) sub‐catchments experiencing a small net accretion of sediment (15 mm kyr^?1).     

Estimation of Soil Erosion Using Field and Modelling Approaches in an Undisturbed Arnhem Land Catchment,Northern Territory,Australia

An assessment of slope erosion at Tin Camp Creek catchment, Arnhem Land, Northern Territory, Australia, was carried out using the fallout environmental radioisotope caesium‐137 (¹³⁷Cs) as an indicator of soil erosion status, two numerical models (SIBERIA and the Revised Universal Soil Loss Equation (RUSLE)) and erosion pins. This undisturbed drainage basin is situated in the seasonally wet‐dry tropics, with high energy storms and a mean annual rainfall of 1389 mm. Tin Camp Creek catchment is unaffected by European agriculture or pastoral activities, but often experiences fire during the dry season. Two transects were sampled for ¹³⁷Cs in 2002 and 2004, and two models were used to convert ¹³⁷Cs measurements into soil loss estimates. Two methods using the theoretical Profile Distribution Model (PDM) gave net soil redistribution rates between +2.72 and –22.19 t ha^?1 yr^?1 and +2.95 and –24.06 t ha^?1 yr^?1, respectively, while an Australian empirical model (AEM) for uncultivated soils produced estimates between +1.84 and –7.00 t ha^?1 yr^?1 (negative values indicate soil erosion, positive, deposition). The RUSLE gave estimated soil losses for the two transects of approximately 10 t ha^?1 yr^?1, while the SIBERIA model produced values between 0.5 and 2 t ha^?1 yr^?1 for the transects and between 3.5 and 11 t ha^?1 yr^?1 for the total catchment. Average net soil losses of 14 and 15 t ha^?1 yr^?1 for the total catchment and slopes, respectively, were measured by erosion pins. The soil losses in the catchment are similar to those for some other transects in the Northern Territory and the Kimberley region of Western Australia (measured by the ¹³⁷Cs AEM), even though these areas are affected by pastoral activities. This may be at least partly explained by erosion in Tin Camp Creek catchment during high intensity rainstorms at the commencement of the wet season, especially if the slopes have been affected by fire during the previous dry season.