BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210916T132457Z
LOCATION:
DTSTART;TZID=Europe/Stockholm:20210706T173000
DTEND;TZID=Europe/Stockholm:20210706T190000
UID:submissions.pasc-conference.org_PASC21_sess182_post185@linklings.com
SUMMARY:P43 - An Ensemble-Based Statistical Methodology to Detect Differen
 ces in Weather and Climate Model Executables
DESCRIPTION:Poster\n\nP43 - An Ensemble-Based Statistical Methodology to D
 etect Differences in Weather and Climate Model Executables\n\nZeman, Schär
 \n\nSince their first operational application in the 1950s, atmospheric nu
 merical models have become essential tools in weather and climate predicti
 on. As such, they are a constant subject to changes, thanks to advances in
  computer systems, numerical methods, more and better observations, and th
 e increasing knowledge about the atmosphere of Earth. Many of these change
 s relate to seemingly unsuspicious modifications like minor code rearrange
 ments, changes in hardware infrastructure, or software updates. Such chang
 es are meant to preserve the model formulation, yet their verification is 
 challenged by the chaotic nature of our atmosphere - any small change, eve
 n rounding errors, can have a big impact on individual simulations. This r
 epresents a serious challenge to a consistent model development and mainte
 nance framework. Here we propose a new methodology for detecting and quant
 ifying differences in weather and climate model executables by using ensem
 ble simulations in combination with statistical hypothesis tests. The meth
 odology can assess effects of model changes on almost any output variable 
 over time. We present first applications of the methodology with the regio
 nal weather and climate model COSMO. The changes considered include a supe
 rcomputer system upgrade, the change from double to single precision float
 ing-point representation, and tiny changes to selected model parameters.
END:VEVENT
END:VCALENDAR