Operational Use of the Rapid Update Cycle

1

• SPRING 2004 RUC CHANGES (13 April 2004)
• 2) RUC analysis
• Improve use of surface data by incorporating
  information on boundary-layer depth.
• Improves surface dew point and CAPE forecasts
  (and temperature to a lesser extent)

Removed 16Z 26 April 2004
2
PBL-based METAR assimilation - Intention Use
METAR data through PBL depth from 1h fcst
RUC oper analysis 18z 3 Apr 02 IAD
Effect of PBL-based METAR assimilation
x
x
x
x
3
Occasional ridiculous features were noted in the
analyses
4
Some of the problems were actually related to
long-standing biases
High CAPE bullseyes in locations between raob
sites.
5
moisture increment added to already moist
profile
1-hr guess
Obs. Dulles sounding
RUC analysis
6
Overview of Changes

• Moisture corrections from METAR obs are applied
  to only 200 mb above the sfc based on 1-hr PBL
  fcst (was more in previous version of code but
  only40 mb in ops)
• Moisture increment values are more limited
• Positive moisture increments are limited if 1-hr
  forecast shows non-zero convective precip
• T and q increments are now further constrained
  going from land to water points

7
New Version often leads to improved CAPE forecasts
8
Biggest Differences at 2-m
Largest diffs in west
Minimal impact on slp
9
Decreasing Impact Aloft
850 mb
500 mb
10
Sample Analysis Sounding Comparisons
Both have same sfc dew point value but very
different PBL structure
11
Downside Initial Fit to Raobs not as good
Ops vs. Para
Obs
12
Verification Scores show worse fit to data at
00-hr but no problems by 03
13
2-m stats
14
Conclusions

• Slightly modified analysis scheme being tested
  for the RUC
• Intent is to have more of a relationship between
  sfc METAR data and the PBL and to limit
  unreasonable extrapolation between stations
• Biggest impacts are on 2-m dew point (especially
  in the west) and low-level moisture structure
  which influences CAPE
• Tradeoff between more realistic low-level
  structure and fit to data