Model interaction with surface boundary forcing requires realistic simulation of surface processes. How did the differences in SST between the operational Eta and EtaX get translated into a different forecast for East Coast cyclogenesis? To illustrate, we show 2-m temperature, 2-m specific humidity, and the convective precipitation predicted by the identical surface physics and convective schemes being forced by different SSTs in the graphics below.
2-m temperatures
|
Model |
Eta |
EtaX |
|
|
0 hour |
|
|
|
|
12 hour fcst |
|
|
The 2-m temperatures are different even initially, because the EDAS has been cycling with the different SST fields. The effects of the SSTs on boundary-layer temperatures over the oceans has developed from sensible heat fluxes in the model and been transmitted through the first guess into the initial conditions, producing initial boundary layer differences consistent with cold continental air flowing over the underlying SST fields. Note that the coastal baroclinic zone is more tightly packed and right up against the coast in operational Eta while the EtaX has an offshore north-south oriented pattern with a warm pocket east of northern North Carolina. This results from the absence of cold shelf waters in the operational SST field. After 12 hours, the 2-m temperature gradient relaxes further near shore in the EtaX, while the warm pocket shifts slightly northward and increases in size, as the sensible heat flux enters the model atmosphere over the warm Gulf stream eddy.
2-m specific humidity comparison, Eta vs. EtaX
|
Model |
Eta |
EtaX |
|
0 hour |
|
|
|
12 hour fcst. (00 UTC 12/30/00) |
|
|
The 2-m specific humidity similarly reflects the underlying SSTs, again even at the initial time. The near-shore is drier and the gradient is weaker in the EtaX than in the operational Eta, because of lower SSTs and the resulting smaller latent heat flux from the water surface. However, the EtaX initially contains a bulge of relatively moist air over the 24°C warm pool in the 2-D Var SST field, and this feature becomes even more pronounced by 12 hours into the forecast. In the operational Eta, the coastal gradient strengthens, with 10 g/Kg air very close to shore from southern NC southward.
Now, consider the effect of these temperature and moisture fields on model convection, the next link in how the SST field influences the cyclone development.
Convective precipitation
|
Model |
Eta |
EtaX |
inches |
|
0-12 hour accumulation |
|
|
|
|
12-24 hour accumulation |
|
|
Convection transports these boundary layer effects through a deep layer, coupling
dynamics to the boundary layer fluxes through the distribution of convective
heating. The horizontal heating distribution depends on where the boundary layer
has destabilized sufficiently for convection. In the BMJ scheme (used in the
Eta model), convection occurs where there is CAPE along with sufficient moisture
in the cloud layer. This favors model convection in regions of some vertical
motion over a cool wedge, where boundary layer moisture is lifted to above the
base of the cloud layer.
Convection in the EtaX was coherent
over the warm pool well offshore of North Carolina during the first 12
hours, intensifying and shifting to the NNE-SSW oriented offshore surface
baroclinic zone by 24 hours. In contrast, the operational Eta had no coherent
pattern of convection during the first 12 hours and a NNW-SSE pattern of
maximum amounts along its surface baroclinic zone right on the coast and
into the lower Chesapeake Bay by 24 hours.
While it's unclear without further investigation whether the model convection is primarily responding to interaction between the mid-level wave and the surface baroclinic zone or is substantially affecting the location of the surface development, in either case it can serve as a flag to the forecaster. The precipitation axis in the model cyclones followed the convection orientations and proceeded to pivot cyclonically. The orientation in the operational Eta was already leaning back further toward the west than in the EtaX, contributing to the prediction of heavy precipitation over the Washington-Baltimore area.
