The four suites in the advanced physics test differ in multiple aspects beyond physics, including dynamic options, physics-dynamics interface, and age of code base. A limited test was conducted to evaluate the impact of these differences. A configuration was created using the same physics parameterizations as Suite 1, but using the dynamics options, physics-dynamics interface, and code base of Suite 4. Therefore, in this test, Suite 1 was run using the CCPP interface. The test was run for a single case -- a Hurricane Irma case initialized on 20170907 at 00 UTC. The results can be intercompared between the original runs conducted for the physics test and the new run conducted for purposes of this sensitivity test. Results indicate that the weak TCs in the Suite 4 runs are caused by the physics suite and not by the dynamics options or by the use of the CCPP. This can be seen in the figure below, which shows that the TC intensity of Suite 1 does not differ significantly between the original run (green) and the rerun using the settings employed by Suite 4 (red). For reference, Suite 4 (blue) and observations (CARQ; gray) are also shown.

More results can be found at:
GMTB's Final Report