# Re: 2D shaded time series plot

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
(970)491-1972```

```--
_________________________
Steve Knox
NREL