FOUM67 KWBC 101200
 OUTPUT FROM NGM 12Z MAR 10 97
 TTPTTR1R2R3 VVVLI PSDDFF HHT1T3T5
 DTW//784932 -1818 182815 34009891
 06000476075 -3117 162612 38049995
 12000377014 04107 092220 39090396
 18000537023 00102 062627 37070397
 24000945808 -1007 103220 27999792
 30000574124 -1514 183213 25009490
 36000493219 -1916 223217 23009388
 42000543316 -1917 253215 24999389
 48000634440 -2519 273116 27989491
      ^^^^^^


 NGM FOU forecast for DTW:

 TT - Hours after model execution (12 UTC, 10 MAR)
 PPT - 6-hourly precipitation (hundreths of an inch)
 R1, R2, R3 - Mean relative humidities in 3 atmospheric layers
 VVV - Upward air velocity (microbars/sec)
 LI - Lifted Index
 PS - Sea level surface pressure (mb, last 2 whole digits included)
 DDFF - Mean boundary layer wind direction (36 compass points) & speed (knots)
 HH - 500 mb height (m - first digit omitted)
 T1, T3, T5 - Mean temperatures in 3 model layers 1, 3, & 5; different from those
              for R1, R2, & R3 (deg C, 99 = -1, 98 = -2, etc.)


 Most relevant to cloudiness are R1, R2, & R3.  Model forecasts mean layer 2 relative
 humidity increase from 49 to 70 % during the 12 UTC to 00 UTC period, which agrees
 with cloudiness observed.  Broken cloudiness was forecast to continue past 00 UTC,
 which satellite imagery does not so clearly indicate.  You may notice that because
 clouds are much thinner than layers 1, 2, 3 (roughly corresponding to low clouds to
 lowest high clouds), a *mean layer* relative humidity does not have to be near 100 %
 with clouds some place in it.