Hi folks,
I'm working on a format for storing the output from a particle tracking
model. Once we're happy with it, we'll post here for feedback, but for
the moment a question:
As many particle tracking models create and destroy particles as the run
goes on, we have decided that a "ragged array" representation is the way
to go. But I'm having trouble figure out the "standard" way to represent
ragged arrays. An minimal example:
dimensions:
time = 24 ;
data = UNLIMITED ; // (240 currently)
## The data dimension describes the "ragged array" -- it is stored as a
1-d array
## time is straightforward.
variables:
double time(time) ;
time:units = "seconds since 1970-01-01 0:00:00" ;
time:long_name = "time" ;
time:standard_name = "time" ;
int particle_count(time) ;
particle_count:units = "1" ;
particle_count:long_name = "number of particles in a given timestep" ;
particle_count:CF:ragged_row_count = "data" ;
## here is the question. particle_count is the number of particles at
each time step -- this describes how long each "row" of the ragged array
is. For each time step, it gives the number of particles at that time.
I also found this in the CDM implementation of CF discrete sampling
features (DRAFT):
"ragged_row_count is an alias for sample_dimension standard name"
so should we be using "sample_dimension" instead? ( I like
ragged_row_count better, this isn't a sample )
Is the the way to specify a "ragged array?"
-Chris
--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chris.Barker at noaa.gov
Received on Fri Sep 09 2011 - 10:00:21 BST
