Development of a Local Analysis System over Highly Variable Terrain
John Horel, Steven Lazarus, and Carol Ciliberti
NOAA Cooperative Institute for Regional Prediction
University of Utah
The Utah Arps Data Assimilation System (ADAS)
Based on the Oklahoma ARPS (Advanced Regional Prediction System) Data Assimilation
System
- Developed by the Center for Analysis and Prediction of Storms (CAPS)
Project Goals:
- Obtain real time, high resolution analyses over complex terrain
- Assist in local weather forecasting
Description of the Utah ADAS
3-Dimensional mesoscale analysis system
220x220 km domain
2 km horizontal resolution
Stretched vertical coordinate
Run at one-hour intervals
Analyses run in two parts
- temperature, relative humidity, pressure and wind analysis
- cloud analysis
Data Used in the Analyses
Main Analysis:
- RUC2 analysis used for initial background field
- Regression data replaces T, Td at first model level
- Surface obs from Utah mesonet
- NWS rawinsonde at 0 and 12 UTC
- NWS WSR-88D velocity and reflectivity data
Cloud Analysis
- 3-D background field from ADAS RH analysis
- Option to use METAR data
- Visible and Infrared satellite imagery
- Radar reflectivities
ADAS Features
Bratseth Interpolation Scheme
- Successive correction technique
- Accounts for background to observation error
- Generally insensitive to observation density
Automated Quality Control
New Vertical Weighting Added to Analysis
- Limit influence of high elevation observations on free atmosphere
- Spread high elevation observation information to data sparse high elevation regions
Future Goals
Incorporate Additional Data Sources
- Satellite sounder data
- Dugway Proving Grounds wind profiler data
- ACARS data
- FAA radar data
Implement Linear Regression Q-C Flags
Experiment with domain size and resolution
Use ADAS to initialize ARPS
- short term, high resolution forecasts
Products Available