What it does
A sieve analysis gives the mass of sediment left on each sieve. This page turns those masses into the numbers sedimentologists report:
- the cumulative curve (weight % coarser against grain size in φ, where φ = −log₂ of the size in mm) and the percentiles φ5 to φ95, read by straight-line interpolation between sieves;
- the Folk and Ward (1957) graphic measures: mean Mz = (φ16 + φ50 + φ84)/3, sorting σI, skewness SkI and kurtosis KG, each with its published verbal class (for example “moderately well sorted”, “fine skewed”, “mesokurtic”);
- the method of moments in three forms: logarithmic (φ), geometric (sizes in mm, worked on their logarithms) and arithmetic (sizes in mm), using the mid-point of each class;
- D10, D50, D90 (the sizes that 10, 50 and 90 % of the sample is finer than) and the uniformity D60/D10;
- the weight % in each Udden-Wentworth class, from boulders to clay, and the gravel, sand and mud totals.
Several samples can go in side by side: one column of masses per sample.
How to use it
- Paste your table: the sieve opening in the first column and the mass retained on it in the next ones. Write pan for what passed the finest sieve. A first row of names labels the samples. You can also open a CSV or Excel file.
- Choose whether the sieve sizes are in mm, µm or φ. Rows can be in any order.
- Set how far the two open classes reach. Grains on the top sieve could be any size bigger than it, and the pan holds everything finer than the last sieve; the page spreads that mass evenly over the width you set (1 φ by default). If more than 5 % sits in either open class the page warns you, because the outer percentiles then depend on that choice.
- Read the summary table, look at the curves, pick a sample to see its percentiles and classes, and download the statistics, the curves or the plots.
Worked example
The example has three made-up samples on half-φ sieves from 4 mm to 63 µm. The Beach sample (111.8 g) gives φ16 = 1.104, φ50 = 1.686 and φ84 = 2.261, so the Folk and Ward mean is 1.684 φ (311 µm), medium sand. Sorting is 0.591, moderately well sorted; skewness −0.014 (symmetrical) and kurtosis 1.059 (mesokurtic). The logarithmic moments give a mean of 1.677 φ and a sorting of 0.586 φ, close to the graphic values, as expected for a tidy sample.
The River bar is gravelly: 20.40 % gravel, 77.77 % sand and 1.83 % mud, a mean of 0.342 φ (coarse sand) and a sorting of 1.554, poorly sorted. The Estuary sample has 32 % in the pan, so the page warns that φ84 and φ95 lie inside the pan class and move if you change how far it reaches.
Tips
- Weigh the pan. Leaving it out makes the sample look coarser and better sorted than it is.
- For muddy samples, sieve the sand and measure the fines with a laser or pipette, then join the two into one table in φ.
- Folk and Ward kurtosis is 1.00 for a normal curve; kurtosis by moments is 3. The two are on different scales.
- In φ, a positive skewness means a tail of fine grains. In mm (the geometric moments) the sign is the other way round.
FAQ
Why are my percentiles different from another program? Programs differ in how they interpolate (straight lines in φ here, as is usual) and in how they treat the open classes. Check the pan and top-class widths first.
Can I use laser data? Yes: give each size bin’s upper limit as the “sieve” and its volume % as the mass.
Is anything uploaded? No. All the sums run in your browser and nothing leaves your device.
Where the numbers come from
- Methods (formulas and class limits only, no data copied): Folk and Ward (1957) graphic measures and verbal classes; Krumbein φ scale; Udden (1914) and Wentworth (1922) size classes; moment formulas as written out by Blott and Pye (2001) for GRADISTAT.
- Example sieve data: synthetic, made up by fibub.
Standard scientific values, such as constants, atomic weights and melting points, are facts found in every textbook; we name where ours were checked. If you are an author or publisher and want a value, a credit or a table changed or removed, tell us and we will act quickly.
Last updated 2026-11-09.