REPRODUCIBILITY OF SIEVE AND SETTLING TUBE TEXTURAL DETERMINATIONS FOR SAND-SIZED BEACH SEDIMENT

Using 525 beach sediment samples, the reproducibility of sieve and set tling tube textural determinations is examined, in order to assess the suitability of both methods for determining parameters for use in env ironment of deposition or sediment transport models. Agreement between classifications of settling tube and sieve determinations of textural parameters is poor, with 37%, 44% and 77% of replicates having differ ent mean grain size, sorting and skewness classifications respectively . The mass of the sediment sub-sample used affects textural parameters derived by both settling tube and sieve analyses. Mean grain size det ermined by settling tube increases with increasing sub-sample mass, wh ereas the mean grain size derived by sieving shows no trend. Sorting d eteriorates (more poorly sorted) with increasing sub-sample mass for s ettling tube analyses, but improves slightly (better sorted) for sieve analyses. Skewness tends to become more finely skewed with increasing sub-sample mass for sub-sample masses less than 50 g, for both method s. The variability of settling tube results is least for sub-sample ma sses of 20-30 g. These results indicate that settling tube textural pa rameters must be used with caution in any environmental or transport n umerical model based on sediment textural characteristics, at least fo r beach sands of the type used here. In particular, settling tube dete rmined textural parameters cannot be directly substituted into a model derived from sieve analyses. Models based solely on specific numeric ranges of 1 or 2 textural parameters should be avoided. Models based o n trends are more likely to result in reproducible interpretations.