| IODP Proceedings Volume contents Search | ||
|
||
| Expedition reports Research results Supplementary material Drilling maps Expedition bibliography | ||
|
doi:10.2204/iodp.proc.311.205.2009 Methods and materialsGrain size analysis was conducted by image analysis of thin sections because the distribution of unlithified sediments from ocean drilling projects can be compared with that of lithified rocks from ancient on-land accretionary complexes. Therefore, the coarser sediments were taken to be analyzed from the core to observe grains using an optical microscope. Fortunately, coarser sediments were relatively abundant in Expedition 311, although the mean grain size is mainly silt size. A total of 23 samples were collected as half-round samples from Sites U1325 to U1329. Samples were carefully moved to plastic half-round liners from archive halves and then covered by shrinkwrap to keep their shape. A subset of 15 samples was analyzed (Table T1) because it was difficult to make thin sections from unlithified sediments. To avoid any disturbance of sample textures, half-round samples were initially impregnated with resin. The center of the sample was cut for thin sectioning (Fig. F1). An advantage of this method is that the original texture of unlithified sediments can be observed. Image analysis was conducted using Image-J software. Maximum and minimum lengths of grains are obtained by the image analysis. Grain size is defined as a root mean of multiplication of maximum and minimum lengths of grains that are obtained from image analysis. The minimum grain size we quantitatively measured is ~1 µm. The number of analyzed grains ranges from 200 to 1000 in each sample. Four statistical parameters were examined to represent characteristics of grain size distributions. Two of them are kurtosis and skewness. The other two are correlation coefficients of fitting by exponential and power functions. Grains <10 µm were excluded from the correlation because the smaller grains introduce measurement error. The two correlation coefficients are named exponential-R and power-R. Figure F2 shows examples of the fitting of these functions. The examples show one of the highest values of exponential-R and power-R among the analyzed samples in this study. |
|
