Profile AMDAR-Model Statistics Change Details

3-April-2019 (beta)
Corrected a bug in OmF stats for winds. Routine(s) changed: get_amdar_vert_stats.cgi.

24-Oct-2017 (beta)
The "phase = R" option was reinstated, and adjusted to use the HRRR region, not the RUC region. (This allows you to plot model vs. RAOB, along with the normal model vs AMDAR curves, but only for the HRRR region.) Also, a bug was corrected that showed up when there was data at 0 hPa. Routine(s) changed: get_amdar_vert_stats.cgi, Welcome.cgi, DisplayFrame.java.

7-June-2017 (beta)
A java web start (browser-independent) version of this was created, and changes were made to support a java web start version of this: Models now come from getModels.cgi, and an initial time window of 1 month is determined by the java code rather than the calling script. Routine(s) changed: getModels.cgi (new), Welcome.cgi, StatsDisplay.

2-Mar-2016 (production and beta)
Copied to the new web server.

27-April-2016 (production)
All beta changes moved to production

21-April-2016 (beta)
We can now optionally time-match within pairs. That is, one pair can be valid at, say 00Z and the other pair valid at 03Z. Since these pairs have different valid times, they won't all match, but each pair can match. Routine(s) changed: get_amdar_vert_stats.cgi, DisplayFrame, CurvesControl.

20-April-2016 (production)
All beta changes moved to production

14-April-2016 (beta)
Hacked together an option to look at statistics against RAOBs in the RUC region only for models for which those stats are generated. To get these, choose RUC as the region, and 'R' (for RAOB) as the phase. Results are available for bias (forecast minus ob), OmF (ob minus forecast), RMS, N. (Model average and Ob average don't work.) Routine(s) changed: get_amdar_vert_stats.cgi, Welcome.cgi, CurveParameters.

26-Aug-2015 (production and beta)

20-Nov-2014 (beta)
Added FIM_iso_4 (world-wide, 0.5 degree lat/lon) model. Routine(s) changed: on jet.

26-Aug-2014 (production and beta)

24-May-2013 (production and beta)
Corrected a 1h error in the grib file used to compare with the AMDAR obs. Previously, obs were compared with a forecast valid 1h later than the top of the hour nearest the obs times. Checking showed tht 7 day statistics differed very little because of this.

20-May-2013 (production and beta)
First version.


Last modified: Wed Apr 3 17:26:15 GMT 2019