I tried two more test locations tonight: Saint Louis, Missouri (KSTL), and Memphis, Tennessee (KMEM). The nighttime microphysics satellite image shows a broad region of stratus over Saint Louis and the animated image showed stratus near Memphis also (but the edge of the stratus deck was nearby).
Beginning with Saint Louis, the recent METARs show a persistent low ceiling:
KSTL 300344Z VRB04KT 2 1/2SM BR OVC005 03/01 A2990 RMK AO2 T00330011
KSTL 300321Z 35005KT 3SM BR OVC005 03/01 A2991 RMK AO2 T00330011
KSTL 300251Z 36005KT 4SM BR OVC004 03/01 A2989 RMK AO2 SLP130 60001 T00330011 51005
KSTL 300249Z 35006KT 4SM BR OVC004 03/01 A2989 RMK AO2
KSTL 300232Z 01004KT 4SM BR OVC005 03/01 A2989 RMK AO2 T00330011
And a RAP model sounding for 03Z shows a very stable saturated layer from near the surface to almost 7,000 ft. MSL:
Beginning at ground level in the sim at KSTL, the overcast was very well depicted:
As I climbed into the cloud layer, the ground was still visible until approximately 2,400 ft. MSL. Here is the view at 1000 ft MSL (approximately where the cloud base should be):
And here is the view at 2,400 ft MSL, just as the ground faded from view:
At this height, the tops of the clouds were already visible through the translucent cloudy air. I continued climbing and reached the top of the main cloud layer at around 4,000 ft. MSL:
It looks like the sim drew a nearly overcast layer between approximately 1,000 and 4,000 ft. MSL, and a separate layer with far less coverage around 7,400 ft. MSL:
It’s possible that the Meteoblue NEMS model indicated a second layer, unlike the RAP model sounding above. It’s impossible to know for sure without looking at the data. But overall, this cloud layer was a reasonable representation. Climbing higher to get a broader view, the layer looks stratiform. This is a good representation.
I suspect the layer looks so good because coverage was near 100%. In my tests with the sim’s cloud parameters, the sim will draw a stratiform layer if coverage=100. This is true even if scatter=100. If the coverage were less than 100% in this case, would the sim have drawn building cumulus clouds?
Regardless of that, the density parameter is far too low here. It is doubtful that the ground would still be visible after climbing 1,400 ft. above the cloud base in real life. By the time I reached that altitude, I could already see the cloud tops 1,600 ft above me.
The images I showed in the original post indicated the same conclusion: density needs to be very high for shallow stratus clouds to look realistic. But high density also leads to volcanic ash-like appearance at the cloud top if the cloud is thick. It’s a difficult balance to strike with the current cloud parameters.
Moving on to Memphis, Tennessee:
The last few METARs indicate a persistent low ceiling, but broken clouds an hour ago:
KMEM 300254Z 34004KT 10SM OVC007 13/12 A2983 RMK AO2 SLP100 T01330117 51015 $
KMEM 300154Z 32006KT 10SM BKN007 BKN120 13/11 A2982 RMK AO2 SLP098 T01280111 $
KMEM 300152Z 34006KT 10SM BKN007 BKN120 13/11 A2982 RMK AO2 $
The RAP model sounding for 03Z shows a shallow stable saturated layer near the surface (with a top near 1700 ft. MSL) and another saturated layer near 11,000 ft (similar to the METAR reports of clouds at 12,000 ft).
At ground level in the sim, blue sky and sunshine are easily visible through the clouds, but if you look closely, you can see that the clouds are covering most of the sky. The density parameter is low, resulting in translucent clouds.
Climbing through the clouds was almost like climbing through a hole between clouds. I could see the ground and the sky throughout the climb. Here is the view near the cloud base:
The cloud top was near 3,200 ft. MSL (different than the RAP model, but the sim isn’t using the RAP model). The clouds above this layer were at approximately 13,000 ft. MSL:
The accuracy of cloud heights here is reasonable. From above, the cloud coverage definitely does not look overcast, but as I descended back toward the airport, I could see very thin clouds covering the area:
These clouds are so thin that they are practically invisible when viewed from above.
In this case, I suspect coverage was not 100% in the sim, but it might have been near 90%. And that seems reasonable based on the previous hour’s METARs and the satellite image showing the edge of stratus in the area. The clouds are also passable as stratocumulus. They do not look like building cumulus clouds. Here again, though, the density is far too low to be convincing. The cloud layer is so translucent that climbing through the layer looks like climbing through a hole in the clouds.
And maybe not reducing the coverage below 100% when it should be 100%