[ferret_users] Re: [las_users] LAS XML Time Configuration

[Roland Schweitzer - NOAA Affiliate <roland.schweitzer@xxxxxxxx>]

You can safely use the configuration which specifies the range of dates that covers the extent of the data set. If somebody makes a plot of a missing day, they will get an empty plot or an interpolated result depending on the plot options they selected in the UI.  Any calculations or time series plots done over the time range will account for the missing days.

You can force addXML to treat a long time series with a few missing days as if it were regular with no missing days with the -a:t command line switch.

Roland

On Wed, Aug 23, 2017 at 9:16 AM, Oots, Penny C. (LARC-E3)[Science Systems & Applications, Inc.] <penny.c.oots@xxxxxxxx> wrote:

I’m not sure if this is a LAS or a Ferret question.  Using LAS 8.6 and Ferret 6.9.5, I have a dataset that contains a Thredds aggregation of 7536 daily files with a .25 x .25 degree grid.  When I run addXML.sh for this aggregation, I get a large file with individual time entries for each day because there are a few missing daily files.  

Previously, this aggregation was configured with a time axes with the number of days specified equal to the number of files plus the number of missing days.

Is using the xml file with the individual dates slower but more accurate?  If using individual dates in the xml file, how will missing dates be handled if a time plot is requested for a period that includes the missing dates?  Are there any other considerations I should think about when deciding which configuration to use. 

Penny Oots 

SSAI, NASA Langley, Hampton, VA

--

The policy of the DOC and NOAA requires me to inform you that the opinions in this email are mine and do not necessarily represent the opinion or policy of the Department of Commerce or the National Oceanic and Atmospheric Administration.