This Excel based program from Kenneth Pye is an excellent resource for calculating grain size statistics: http://www.kpal.co.uk/gradistat.html

Plot cumulative frequency on probability paper to illustrate grain size populations / binned fractions as histograms to illustrate skewness etc.

Http://www.cloversgardencenter.com/images/soiltypepyramid.jpg not sure if this what you were going for, but it seems like a start.

Two papers may help you.

1.Veerle vanwesenbeeck,Jean Lanckneus. Residual sediment transport paths on a tidal sand bank:a comparison between the modified Mclaren model and Bedform analysis[J].Journal of sedimentary research,2000,70(3):470~477.

2.Glen S,Visher.Grain size distributions and depositional processes[J].Journal of sedimentary petrology,1969,39(3):1074~1106.

What data are you looking to illustrate with these figures? That will help with figuring out what types of graphs to present. For example, if you're looking for changes with depth, the 100% stacked plots are probably your best bet. If you're trying to show percentage of clay for soil compaction (just an example to illustrate my point) a pie chart is best. If we knew what data you were trying to highlight we'd be able to offer you more help.

Cumulative grain size distribution of the type favoured by soil engineers

Cumulative diagrams which can be characterised by sedimentogical parameters.

Thanks for everyone for.......your suggestions.....i have plotted 100% stacked plot for my grain size vs depth data. I think this would be better because it tells about the variation in grain size according to depth and my age model. I have also plotted D(4,3) values (Volume weighted mean diameter) w.r.t. depth, which gives an idea about how coarse are the grains.

Your samples may or may not work with a sieve analysis. You record the percent

retained on individual sieves with progressively smaller and smaller wire screens.

The screen sizes are standardized.

Praveen,

With sediments, the first question you need to address is really: What do I want to show? Depending on the questions your study wants to address, you may select different representation. Both graphical and statistical methods are available.

Textural analysis can show a number of things and basically it comes down to the descriptive and interpretive methods. An online publication from the USGS gives a great description of what fits the needs best. It talks about marine sediments but you will understand the philosophy behind it.

http://pubs.usgs.gov/of/2000/of00-358/text/contents.htm

Working with lakes myself, I usually use grain-size distributions for the representative part, and do the statistical analysis to obtain as many classifiers as makes sense for a particular study, such as size, roundness, circularity, mineralogy, composition, etc. That describes your sample. Then, to go to the interpretation, you need to go a step further in the analysis, and see how those various classifiers are grouped. For that, you can try multivariate analyses. Those will give you more indications about the families of particles you are dealing with. They can also point to genetic classes, and textural evolution. They are easy to learn (and often times software are doing it these days), and pretty powerful.

Good luck!

OK. I think that all the answers are perfect. I just want to add the statistical analysis. If you have your size analysis data, you can calculate the statistical parameters (such as skewness or kurtosis) from both graphic and moment methods. They are helpful for better understanding of environmental conditions.

This Excel based program from Kenneth Pye is an excellent resource for calculating grain size statistics: http://www.kpal.co.uk/gradistat.html

Plot cumulative frequency on probability paper to illustrate grain size populations / binned fractions as histograms to illustrate skewness etc.

No problem ;)