Closed penguian closed 6 years ago
@jxs599@nci.org.au changed status from new
to closed
@jxs599@nci.org.au changed resolution from ` to
fixed`
@jxs599@nci.org.au changed milestone from 6. Report
to 1. Closed
@ccc561@nci.org.au set keywords to nogit
keyword_nogit
resolution_fixed
type_defect
| by inh599The reference levels %zref_uv and %zref_tq are the locations where the meteorological forcing for CABLE is applied. These are set within the roughness module and take the value as heights above displacement - subject to a minimum value.
In some circumstances, e.g. tall but low LAI canopies, the resultant reference heights can be below the height of the canopy and/or smaller than the roughness length (e.g. %ref_uv < %hruffs-%disp or < %z0m). In such circumstances the assumed wind speed and turbulence profiles used to determine the resistances (%rt1usc) and transfer coefficients (%cd_uv, %cd_tq) are no longer valid. It is possible to generate -ve resistances and transfer coefficients in this way.
The roughness module and associated code in the canopy module would benefit from a full review and rewrite. However as a temporary measure it is suggested that alternate limits are applied to %zref_uv and %zref_tq to avoid the problem. Specifically (in both locations where these are determined)
rough%zref_uv = MAX(3.5+rough%z0m, rough%za_uv) rough%zref_uv = MAX(3.5+rough%hruffs-rough%disp, rough%zref_uv)
and similarly for the rough%zref_tq.
Note that in the vegetated surface case these calculations would need to be placed after the calculation of rough%z0m (currently they occur before)
Issue migrated from trac:148 at 2023-11-27 11:20:52 +1100