[Thread Prev][Thread Next][Index]

Re: anomaly



Title: Re: anomaly
Shen
   I  think your problem may be that the time axis in your longterm monthly mean file (TIME1) is not "modulo".  Try setting it to modulo and let me know if it still doesn't work.  Lynn

At 10:05 AM -1000 12/13/01, shen yingshuo wrote:
hi lynn,
 
  it works fine if the two data have same time dim.  say if i use
  http://apdrc.soest.hawaii.edu/cgi-bin/nph-nc/data/soda7/tmp/soda_tmp_1950.cdf  (this file has 12 time steps from jan. 15, 1950 to dec. 15, 1950)
 http://apdrc.soest.hawaii.edu/cgi-bin/nph-nc/data/soda7/tmp/soda_tmp_mean.cdf  (this file has 12 time steps from jan. 15 to dec. 15 this is a longterm monthly mean)
this way, i can only get a Time line for one year 1950.
 
but i have a des file which links all the monthly data from 1950 to 1999 (which has 600 time steps from jan. 15, 1950 to dec. 15, 1999).  i am trying to make an anomaly for all these years..   it seems to me that i need some way to expand the long term mean (12 time steps) to match all these years (600 time steps).  i wonder you and anyone on the list can help me.. this is more a ferret problem.
 
----- Original Message -----
From: Lynn deWitt
To: shen yingshuo ; las_users@ferret.wrc.noaa.gov
Sent: Wednesday, December 12, 2001 10:26 AM
Subject: Re: anomaly

Hi Shen
   I have anomalies implemented on our LAS for GTSPP and WOD98 subsurface temperature data.

http://las.pfeg.noaa.gov/las

See the section called "Subsurface Temperature, Isotherm Depth and Anomalies".  The first two GTSPP items contain a monthly anomaly option, and the third computes seasonal anomalies.  Both of these are done with simple custom .jnl files.  See the LAS documentation on custom initialization scripts.

Here is the part of one of my .xml files that points to a custom .jnl file:

 <!-- GTSPP Subsurface Temperature -->
    <subsurf1deg_nc name="1. GTSPP Real-Time, Monthly, Aug 1996 - present" url="file:/las/las_data/subsurf1deg.nc" doc="doc/subsurf1deg.nc.html">
 <institution name="Pacific Fisheries Environmental Lab" url="http://www.pfeg.noaa.gov"/>
     <variables>
...
...
...
      <anomaly name="3. Subsurface Temperature Anomaly">
       <link match="/lasdata/grids/subsurf1deg_nc_lon_lat_depth_time_series_grid"/>
       <properties>
        <ferret>
         <init_script>subsurf_anom</init_script>      (add a line like this to your xml file)
         <fill_levels>"(-4,4,1)(-10,-4,2)(4,10,2)"          
         </fill_levels>
         <contour_levels>(-11000,11000,22000)           (don't want contours)
         </contour_levels>
         <palette>light_centered.spk          
         </palette>
        </ferret>
       </properties>
      </anomaly>

The file subsurf_anom.jnl is located in  ../lasxml/server/custom
and has the following in it:

set data "/las/las_data/woa98_clim.nc"
set data "$1"
define axis/x=2.5:357.5:5/unit=degrees/modulo xaxis1
define axis/y=-87.5:87.5:5/unit=degrees yaxis1
define grid/x=xaxis1/y=yaxis1 subsurf_grid
let anomaly=t_mean[gxy=subsurf_grid@AVE]-temp[gxy=subsurf_grid@AVE,d=1]
set var/title="GTSPP Temperature Anomaly (Deg C)" anomaly

The first line is the climatology file
The second line gets the name of the monthly mean file (subsurf1deg.nc defined in xml file) from the LAS, but you could insert a file name here too if you want.
Ignore the define axis and grid lines - I compute 5-degree averages from the 1-degree averages in the
subsurface data files.  Also ignore the gxy=...@AVE, but remember to put include the "d=1" for the climatology dataset number.

That's all there is to it (I think - I could be forgetting something else I had to do).

I also have an option to compute seasonal (3-month) averaged anomalies, but that initialization script is a little more complicated.   Let me know if you have any further questions.  Lynn

 

At 8:56 AM -1000 12/12/01, shen yingshuo wrote:
hi there, i wonder if anyone can help me out.  i want my las to show anomaly (monthly - long term mean)...
of course las can show the difference of two file (say  the diff between jan 1979 and longterm mean of jan).  but that does not satify some researchers who want to get an anomaly  t line plot.  say from jan 1950 to dec. 1999  which means.. ( jan.1950-Ltmean jan, feb.1950-Ltmean feb, and ....  dec. 1999 - Ltmean dec).  i have already created all the monthly mean data (which can be linked all together by a des file) and also i have created a long term mean data (averaged all jan, feb, ... dec. from 1950 to 1999 respectively)..
 
for reference, please see http://apdrc.soest.hawaii.edu/las    check SODA data
 
Shen Yingshuo
IPRC-SOEST
Univ. of Hawaii

--------------------------------------------
Lynn deWitt

ldewitt@pfeg.noaa.gov

Pacific Fisheries Environmental Laboratories
1352 Lighthouse Avenue 
Pacific Grove, CA 93950-2097
(831)-648-9036

--------------------------------------------
Lynn deWitt

ldewitt@pfeg.noaa.gov

Pacific Fisheries Environmental Laboratories
1352 Lighthouse Avenue 
Pacific Grove, CA 93950-2097
(831)-648-9036

[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP
Contact Us | Privacy Policy | Disclaimer | Accessibility Statement