ARCHITECTURE OF MIOCENE OVERBANK DEPOSITS IN NORTHERN PAKISTAN - DISCUSSION

Willis and Behrensmeyer (1994) have presented a remarkably detailed de scription and analysis of the architecture of ancient overbank deposit s from the Miocene Chinji Formation of northern Pakistan (see also Beh rensmeyer 1987, Willis 1993a). They described meters-thick paleosol-bo unded sequences that are defined by an alternation of relatively matur e paleosols and stratified deposits with a lesser degree of pedogenesi s. They asserted that the mature paleosols represent relatively long p eriods of very low deposition rate, whereas the stratified deposits re present relatively short periods of high deposition rate. They propose d and evaluated five end-member hypotheses for the origin of the paleo sol-bounded sequences. They concluded that the stratified deposits wer e formed by local rapid filling of low areas on the floodplain, indepe ndently of the position of major channels: such filling may have been a more or less continuous process, or may have been a short-lived even t associated with river-channel avulsion. They specifically dismissed the possibility that the nature of the stratified deposits was related to the proximity of major channels and the growth of alluvial ridges. They also dismissed the possibility that paleosol-bounded sequences c ould have been related to regional variations in sediment supply and d eposition rate, or to valley incision and filling. In my view, these f ar-reaching conclusions are flawed, for three main reasons: (1) whethe r or not the overbank sediment bodies appear to hh topographic lows in the paleo-floodplain depends critically on the way the stratigraphic cross sections were constructed, which was not justified adequately; ( 2) the hypotheses for the origin of the stratified deposits are mislea ding; and (3) the sections studied do not extend sufficiently in a dir ection normal to paleoflow direction to test adequately the hypotheses presented. I will expand upon these points in turn.