A quantitative approach to understanding dated dune stratigraphies

Abstract

Attempts to reconstruct past changes in climate-related forcing of dryland landscapes are hampered by the lack of an adequate quantitative framework for understanding the production and interpretation of dated sedimentary records. In drylands, as in other environments, information on past forcing conditions is progressively modified, degraded and removed from the available stratigraphic record by a series of 'filters' involving changes in the primary forcing factors themselves, geomorphological processes and the sampling/dating procedures. In this paper we describe a quantitative model that includes these effects, and use the model to examine the nature of preserved dryland sedimentary records and their relationships to primary forcing conditions: thicker preserved sedimentary records reflect periods of more intense aeolian activity; localized switching between erosion and deposition results in discontinuous and highly variable stratigraphic sequences; a preservation bias towards younger deposits is observed, potentially leading to a continuum of accumulation that decays approximately in proportion to 1= ffiffiffiffiffiffiffi ffi age p . Time periods not represented by deposition can in some cases be interpreted as periods of higher precipitation and/or lower wind energy. An asymmetry exists between the efficiency with which past 'drier' and 'wetter' episodes can be identified, which relates to the timeseparation of depositional periods and the correct distinction between hiatuses due to forcing conditions and those due to undersampling. Relevant to this is the effect of random dating errors (statistical uncertainty), which (increasingly with age) filter-out higher frequency events from the record. A new data treatment method (termed Accumulation Rate Variability) provides an efficient proxy for accumulation rates, and therefore the intensity of aeolian activity, with significant improvements over existing date-frequency methods. The filtering problem discussed applies to all attempts at understanding the timing and nature of past events, independent of the proxies and dating methods employed. Further explicit analysis of these issues would be beneficial.

Fluvio-lacustrine deposits reveal precipitation pattern in SE Arabia during early MIS 3

“…Aeolian sediments have often been interpreted as indicators of phases of aridity in Arabia (Preusser et al, 2002Bray and Stokes, 2004;Radies et al, 2004;Stokes and Bray, 2005;Atkinson et al, 2011), but aeolian deposition also reflects other factors such as sediment availability and preservation (e.g. Preusser, 2009;Bailey and Thomas, 2014;Leighton et al, 2014). Past humid periods in Arabia have been identified by analysing speleothems (Burns et al, 1998Fleitmann et al, 2003) as well as relict lake deposits Rosenberg et al, 2011Rosenberg et al, , 2012; these pluvials concur with maxima in fluvial activity and alluvial fan generation (Maizels, 1990;Rodgers and Gunatilaka, 2003;Blechschmidt et al, 2009;Parker, 2009).…”

Interpreting and modelling late Quaternary dune accumulation in the southern Arabian Peninsula

“…Conversely, when there are periods of high net accumulation, large uncertainties on the ages may lead to cautiousness in interpretation which, whilst necessary, lessens the environmental conclusions that may be drawn from a sequence of ages. Rapid fluctuations are therefore unlikely to be resolved by OSL dating, as shown by Bailey and Thomas (2014), and under-sampling may result in the misinterpretation of apparent hiatuses as periods of erosion or low levels of aeolian activity. False assumptions of the meaningfulness of apparent peaks of accumulation are then a likely consequence, as noted by Stone and Thomas (2008).…”

“…Goudie et al, 2000;Forman et al, 2005;Chase andThomas, 2006, 2007;Preusser, 2009;Roskin et al, 2011;Yang et al, 2012). Bailey and Thomas (2014) describe the problem as one of successive 'filtering', whereby the information of initial forcing is transformed through geomorphological response and depositional and erosional processes to produce the sedimentary sequence available for dating and interpretation. The interaction of these factors to 'produce' dune structures is not straightforward.…”

“…2. The windspeed background level is 7 ± 1.5 m/s, (the threshold of sediment entrainment is 6.4 m/s for the model described by Bailey and Thomas, 2014). Peaks in windspeed at 31, 24, 16.9 and 12.9 ka represent H3-1 events and the Younger Dryas respectively.…”

“…Following the qualitative description given in the preceding sections, a more rigorous approach to assessing aeolian system sediment state is introduced here. Sediment supply, sediment availability and transport capacity have been modelled to show change in dunefield height over the last 30 ka using a numerical model described by Bailey and Thomas (2014). The model calculates whether erosion or deposition takes place on a surface at each time step, and hence the change in surface height over time.…”