Re: [ferret_users] Counting Data-points

[Russ Fiedler <russell.fiedler@xxxxxxxx>]

Thanks for the glowing reference, Billy, but that reply (nearly 4 years ago!) was in reference to  scattered data and is a bit of overkill for here as the data is gridded regularly. As Ansley points out a simple mask and either @NGD or @SUM does the trick.

As an addition to what Ansley suggested you can mask directly. You won't be able to use the @SUM transform to get the count but you can do things like taking averages of values above the threshold without going through intermediate steps. I've no idea about the efficiency of doing things in steps so that is left as an exercise for the reader...

let var_hi = IF var[x=170E:170W,y=30S:40S,t=10-Jan-2013:10-Dec-2013] GT 11 then var ! instead of 1

Russ

On 05/12/14 14:39, William S. Kessler wrote:

See a typically-elegant solution from Russ Fiedler about 2 years ago. Mick Spillane had another (both in response to a similar question from me). I'm not at a computer so can't look it up, but search "count" and Fiedler and you should find it. 

BK

Sent from my phone .... Please excuse my brevity .... and my typing. 

On Dec 4, 2014, at 6:10 PM, Luke M <coding1227@xxxxxxxxx> wrote:

Dear all

I'm trying to count the number of data points that are above a certain value for a given region and time period, but I am a bit stuck.

Does anyone have any suggestions on how I could find what are, for each month, the total number of data points (pixels) that have a value higher than 11 degrees for a given small area & time period:

x=170E:170W/y=30S:40S/t=10-Jan-2013:10-Dec-2013

Using the sst dataset:
set data "http://oceanwatch.pifsc.noaa.gov/thredds/dodsC/pfgac/monthly"

Thanks!

-