<font size=2 face="sans-serif">In the GrADS Documentation Index, the description
for the Cressman objective analysis function "oacres" rightfully
cautions against using too few stations in the analysis, relative to the
number of grid points:</font><br><br><font size=3>"The Cressman Analysis scheme can be unstable if
the grid density is substantially higher than the station data density
(ie, far more grid points than station data points). In such cases, the
analysis can produce extrema in the grid values that are not realistic.
It is thus suggested that you examine the results of </font><tt><font size=3>oacres</font></tt><font size=3>and compare them to the station data to insure they meet your needs."</font><br><br><font size=3>I have two questions regarding use of oacres:</font><br><font size=3>1. What is an example of a grid-to-station ratio
that one might consider unstable or unrealistic?</font><br><br><font size=3>I have searched the GrADS User Group archives, but no
luck. Additional searches on the Internet suggest a grid-to-station
ratio of around 81:1 is approaching any useful limits, based on a grid
cell requiring any station data within four grid points of itself (thus,
a 9 x 9 "box" surrounding it). If you've had experience
(good or bad) with using oacres, or were involved in its development, please
let me know what might be considered realistic grid-to-station limits.
Of course, I could go the trial-and-error route to see what makes
sense, but I would rather approach the problem based on others' success.</font><br><br><font size=3>2. What are the function's optional "radii
of influence"? Are they length units and, if so, what? Are
they number of grid cells/points? Or something else?</font><br><br><font size=3>Again, I could not find the answer, either in the Documentation
Index description, or in the User archives.</font><br><br><font size=3>I would be processing mainly surface temperature and precipitation
station data across regional and continental domains.</font><br><br><font size=3>Thanks in advance!</font><br><font size=3>Stephen McMillan</font><br><font size=3>Planalytics, Inc.</font><BR>
<BR>
***************************************************<BR>
The information contained in this e-mail message is intended only for the use of the recipient(s) named above and may contain information that is privileged, confidential, and/or proprietary. If you are not the intended recipient, you may not review, copy or distribute this message. If you have received this communication in error, please notify the sender immediately by e-mail, and delete the original message.<BR>
***************************************************<BR>