RECORDING PALAEO-EVENTS IN THE CENTRAL ASIAN DRYLANDS- THE PERIMONTANE AND PERIDESERT LOESS REGIONS INTERACT

The thick loess in Uzbekistan could one day rival the deposits in north China as an accessible store of environmental proxy archives. Thick loess deposits contain many palaeosols and other indicators of Quaternary climate change, but it is useful if the sedimentological setting is both simple and fairly well understood. The nature of a loess deposit does have some influence on its use as a 'climate register'. While Central Asia was part of the USSR there were many opinions in favour of the eluvial/ in-situ/ soil formation theory of loess development(the famous Berg theory) and this top-down approach is impossible to reconcile with 100m thick deposits and multiple palaeosols (and observed hydrocollapse). Many Central Asian investigators, particularly in Uzbekistan, retained the loess ideas of A.P.Pavlov(c.1900) which essentially divided loess into deluvial (slope) loess and proluvial (plains) loess. Direct deposition from water was required, and there are deposits, e.g. associated with the Chirchik river, that appear to be water-laid. Obruchev, by 1952, had reconciled aeolian deposition with the Pavlov scheme- but more progress in this direction is required. The PTD scheme allows loess particles to originate in the Tien Shan and eventually form airfall deposits. It is possible that the D1 deposits in the current scheme are what Pavlov would call 'deluvial' loess, and that the D2 deposits are 'proluvial' loess. A good case can be made for the D3 deposits being aeolian, and the D4 deposits the result of wind erosion. Once the true nature of the thick deposits has been determined, valuable palaeoclimatic data can be obtained. The 'desert' loess problem is very visible here because the desert edge is very close to the Tien Shan; there are important Quaternary processes still to be investigated in the overlap region.