Results generated from Calc is 185% higher the actual estimates

Dear Dorji,

can you please check your Sampling Design in Calc, whether you have selected Cluster Sampling. If yes, then in this case you should also select "Apply cluster sampling only for variance" too.

Regards, Lauri

Hi,

In Saiku, I assume that you have clicked "Refresh Cubes" button before showing the results. Otherwise Saiku will show results from the old processing run and not the current one.

Are you familiar with RStudio? If yes then export your scripts int RStudio, run then up to source("user/005-base-unit-weight.R") (this script also!), and check in R Console the sum of plot weights (you may have different entity names than here).

sum(plot$weight)

In your case this should be 250 if all plots get an equal weight. If this is ok, run tree area script and check plot_area for trees

unique(tree$plot_area)

if there are NA's then something goes wrong. You can also check these two details after running all your scripts because it is possible that some module and your script is doing something unexpected...

Regards, Lauri

Dear Dorji:

Apparently the reason is that the plot area is NOT 0.05 ha but pi*(0.1262^2)/4= 0.0125086 ha.

If you do a quick calculation, taking into account the result you get in Excel (123) and the ratio between the area you claimed (0.05) and the actual area in your inventory (0.0125086), then: 123*.05/0.0125086 = 491.7 trees/ha...which is pretty close to your initial calculation.

Also, sum(plot$weight) = 350 (not 250) accessible plots

.. and nrow(plot)= 368...the total number of plots you had.

Hence, your Calc workspace is working correctly. It was something in your Excel calculation that was not correct, unless your plot size is indeed 0.05 and the Calc workspace is not correct.

Javier