Re: 2D shaded time series plot

[Steve Knox <stevek@xxxxxxxxxxxxxxxxxx>]

Thanks for pointing me to the reshape command.
I seem to be doing something wrong, however, when I try to follow the example.
 Why do I get an error from UNITS=MON. (If I omit "UNITS=MON" I get no warning
message but the results are the same).
The reshaped grid seems ok but when I list my_reshaped I get no data.

The time variable in my file tmp.cdf is :        
                float time(time) ;
                time:units = "months since 1900-01-15 00:00:00" ;
                time:title = "Time" ;
                time:long_name = "Time axis" ;
                time:time_origin = " 1900-01-15 00:00:00" ;
The variable time has values:
 time = 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28,
...
...
When I follow the example this is what I get:

use tmp.cdf
DEFINE AXIS/t=1-JAN-1900:31-DEC-1903/NPOINTS=48/UNITS=MON tcal
 *** NOTE: /UNIT=MONTHS is ambiguous ... using 1/12 of 365.2425 days
Replacing definition of axis TCAL
LET my_time_series = tmp[gt=tcal]  ! tmp is my average temp variable
DEFINE AXIS/t=1900:1903:1 tyear
DEFINE AXIS/Z=1:12:1 zmonth
LET out_grid = Z[GZ=zmonth]+T[GT=tyear]
LET my_reshaped = RESHAPE(my_time_series, out_grid)
SHOW GRID my_reshaped
    GRID (G002)
 name       axis              # pts   start                end
 normal    X
 normal    Y
 ZMONTH    Z                   12 r   1                    12
 TYEAR     T                    4 r   1900                 1903
list my_reshaped
             VARIABLE : RESHAPE(MY_TIME_SERIES, OUT_GRID)
             FILENAME : tmp.cdf
             SUBSET   : 12 by 4 points (Z-T)
              1   2   3   4   5   6   7   8   9   10  11  12  
               1   2   3   4   5   6   7   8   9  10  11  12
 1900   / 1:................................................
 1901   / 2:................................................
 1902   / 3:................................................
 1903   / 4:................................................

Ansley Manke wrote:

Hi Steve,
The question of how to represent data as year vs month-of-year happens
to be explained as an example of the RESHAPE function.  Take a look in
the Users Guide under reshape.

Ansley Manke
 
 

Steve Knox wrote:

 

  I have a netcdf file with 1 variable, tmp  for latitudes -90 to 90 (by 1/2 deg) and longitudes -180 to 180 (by 1/2 degree)
The time unit is "months since 1900-01-15 00:00:00". There are 1200 months (100 years).
I would like to plot year along the x-axis and month (1-12) on the y-axis for a paricular spatial aggregation.
The "value" plotted should be a shaded area, representing the value at each month. I want to look for monthly
or seasonal variations over time.

12  B     A


 
 
 
 
 
 ...
 3   A
 2   B    A
 1   B    B

    1901     1902    ...

Where the letters" images would actually be shaded areas.

Is this possible with Ferret?

yes? use tmp
yes? show data
     currently SET data sets:
    1> ./tmp.cdf  (default)
 name     title                             I         J         K         L
 TMP      Surface air temperature          1:720     1:360     ...       1:1200
 

-- 
_________________________
Steve Knox
NREL
Colorado State University
(970)491-1972

-- 
_________________________
Steve Knox
NREL
Colorado State University
(970)491-1972