In order to evaluate the validity of the modified McLaren model for sand transport on a tidal sand bank, 42 grab samples taken from the northern part of the Kwintebank were decalcified, sieved, and analyzed for statistical parameters. The areal distribution of mean grain size, standard deviation, and skewness were defined and used to determine the residual transport directions. The grain-size trends taken into consideration were that the sediment becomes finer, better sorted, and more negatively skewed along the transport pathways (F,B,-) and the sediment becomes coarser, better sorted, and more positively skewed along the transport pathways (C,B,+). These were compared with the transport paths inferred from bedform asymmetries obtained by side-scan sonar records. There were major differences between the sediment transport paths resulting from the two techniques.
When the F,B,- trend and small dune asymmetries are compared, only 5 out of 24 cases showed agreement (±30°), whereas in 9 cases, the trends were 180° ± 30° different. There was even less agreement between the (C,B,+) trend and the asymmetry of the small dunes: only 1 out of 24 cases agreed ±30°, and almost half of the cases were 90° ± 30° different. The sediment transport directions deduced from the morphological analysis of the large dunes all point in the direction of the peak flood current. The large dunes seem to be more suitable indicators of the sediment transport paths than the small ones. However, they conflict with the vectors resulting from the (F,B,-) trend. Only one out of 20 cases agrees ±30°, while in 13 cases the vectors are 180° ± 30° different, and in three cases 90° ± 30°. Neither do the transport directions from the (C,B,+) trend agree with those deduced from the large dunes. Only one case agrees ±30°, while in four out of 20 cases the vectors point in opposite directions. In four cases the vectors are 90° ± 30° different. The asymmetric cross section of the sand bank suggests sediment transport in a northwest direction. It does not seem probable, however, that sediment transport occurs by migration of the whole sand bank. The transport directions inferred from the bank morphology are quite different from the (F,B,-) trend. In 12 out of 42 cases, the vectors agree ±30°, in four cases the vectors point in opposite directions, and in 15 cases they are 90° ± 30° different. There is slightly more agreement between the (C,B,+) trend and the transport directions deduced from the sand bank. 16 out of 42 cases agree ±30°, in nine cases the vectors point in opposite directions, and three cases are 90° ± 30° different.
From this study it seems that the residual sediment transport directions can best be deduced from the morphological characteristics of the large dunes. The departures between the results of the sediment trend analysis according to the modified McLaren model and the morphology of the subaqueous bedforms are partly explained by the location of the study area near the edge of the sand bank. Probably better results of the sediment trend analysis would be obtained if samples over the whole length of the sand bank were analyzed and compared. Perhaps a smaller spacing between the samples would improve the results as well, and maybe the samples should not have been decalcified. Possibly the different transport patterns are attributed to different sediment transport processes, which makes a comparison virtually impossible.