Composite volume versus grid surface volume calculations

[rschaff]

I am working on a drawing that a contractor producer. They created a volume calculation on a 11 Sq. mile site. They calculated a composite and grid cut and fill added them together and got the average. My question is that I have always used a composite volume calculation. Is there any documentation that describes the best method to use?

[mjfarrell]

In theory the Composite method is more accurate that the grid method.

However they may both vary widely from each other due to the level of detail used to create either, the Existing Ground model, or the Finished model, and the grid spacing.
Changing the grid spacing alone can greatly impact the reported volumes. I used to run preliminary numbers with a fair large grid space (LDT) to save time in processing. Then for final numbers I would run the volume(s) with increasingly smaller grid values until the percent change from the previous was very small.
Although I would still do what the contractor did. I would run a Grid Volume, AND a Composite volume; not only to check one against the other, to sum them up and average them out.
As with any analysis, the more data points you have to compare the higher the accuracy of the results derived from your sample set. (In Theory)

[sinc]

I view grid volumes as obsolete for most purposes, and pretty much never use them anymore.

If you keep setting your grid spacing to smaller and smaller values, you should get results that converge pretty close on the composite volume. Both methods are limited by the accuracy of your Surface. But if you have a bad Surface, you won't get better volumes by using grid instead of composite, unless you just happen to get lucky (e.g., you have a spike in one of your Surfaces, and your grid points happen to miss the spike). I don't like relying on luck, so I don't view comparing Grid and Composite volumes as a good check.