Hi Hussen,
One trick (old, before the introduction of patterns in Ferret!) is to make a "transparent" mask using your significance values.
Say we have variables "corr" and "sig_corr" which are on the same grid and we want to highlight values above "sigval"
First method:
Drop out values above the significance in a checkerboard fashion values below will be filled solidly.
let masked_corr = if sig_corr lt `sigval` or mod(i+j,2) eq 1 then corr
shade masked_corr
Second method:
Overlay with a checkerboard mask in monochrome. This works better if you need to use FILL for you underlying plot.
! Only consider locations above the significance level in checkerboard style.
let mask = if sig_corr gt `sigval` and mod(i+j,2) eq 1 then 1
shade corr
shade/ov/nolab/pal=black mask
You can play around with the mod function to get a satisfactory effect. Maybe even regrid to a finer grid to alter the density.
Cheers,
Russ
On 23/04/14 19:31, Hussen Seid wrote:
HussenThanks,Dear ferret/pyferret users,
I installed pyferret to take advantage of its enhancements(improved graphics and new statistical functions), unfortunately fill/shade+patterns are not yet supported by pyferret. Instead I was trying to use cross-hatching, but not succeeded.
So I'm wondering how to produce filled/shade colour figures with cross-hatching?
To make it clear:
I have two files; one with correlation and the other with significant correlation values. what I want to do is to overlay the significant correlations using cross-hatching.
--
Hussen Seid Endris
Climate System Analysis Group
Department of Environmental and Geographical Science
University of Cape Town
Private Bag Rondebosch, 7701
Tel:(+27) (0)74 222 4357