The Road Not Taken Jim Yoe, JCSDA Chief Administrative Officer JCSDA Summer Colloquium Santa Fe, NM 1
Jan 04, 2016
The Road Not Taken
Jim Yoe, JCSDA Chief Administrative Officer
JCSDA Summer Colloquium Santa Fe, NM 1
Outline Motivation (Some) current topics NOT addressed here (Some) comments & extensions about
topics that HAVE been addressed (Some) future directions and
considerations Summary
JCSDA Summer Colloquium Santa Fe, NM 2
Motivation Question: What Road? Answer: Several:
Some Current applications not covered in Colloquium
Some Extensions and connections of topics covered
Some Future Considerations Environments, Applications, Implications
This was to be the closing talk – “roll up” So this might seem out of place
JCSDA Summer Colloquium Santa Fe, NM 3
Topics Not Addressed Land Data Assimilation
See R. Reichle, http://www.jcsda.noaa.gov/meetings_2009SummerColloq.php
GPS Radio Occultation Refer to talks by J. Derber, L.-P. Riishojgaard, See L. Cucurull http://
www.jcsda.noaa.gov/meetings_2009SummerColloq.php
(Note also 2009 GMAO, UKMO DA Summary Talks)
ummer Colloquium Santa Fe, NM 4
Comments and Extensions
JCSDA Summer Colloquium Santa Fe, NM 5
Data Impact Studies• Multiple Methods
• Forecast Sensitivity Diagnostics (FSD; e.g., Adjoints)• Observation System Experiments• Observation System Simulation Experiments
• Multiple Applications• Evaluate DA systems/methods• Evaluate OS components
• Influence decisions regarding future OS components
JCSDA Summer Colloquium Santa Fe, NM 6
500 hPa Anomaly Correlations 15 Aug – 30 Sep 2010 (Riishojgaard and Jung, 2012)
7
No Satellite / No Conventional Data
Northern Hemisphere Southern Hemisphere
5th WMO Observing Systems Impact Workshop
500 hPa Day 3 North America Time Series (LPR/JJ)
JCSDA Summer Colloquium Santa Fe, NM 8
Data Impact Studies• Recommendation (Word of Caution):• Avoid Temptation (resist pressure) to map every
inch of trade space• Recall JCDSA Mission Statement• Analogy to misuse of statistics
• “Statistics are like a street lamp; the wise man uses it for illumination, and the drunk man for support”
JCSDA Summer Colloquium Santa Fe, NM 9
Modeling and DA Constraints
JCSDA Summer Colloquium Santa Fe, NM 10
11
Air Quality
WRF NMM/ARWWorkstation WRF
WRF: ARW, NMMETA, RSM GFS, Canadian Global Model
Regional NAMWRF NMM
North American Ensemble Forecast System
Hurricane GFDLHWRF
GlobalForecastSystem
DispersionARL/HYSPLIT
Forecast
Severe Weather
Rapid Refreshfor Aviation
Climate ForecastSystem
Short-RangeEnsemble Forecast
NOAA’s Model Production Suite
GFS MOM4NOAH Sea Ice
NOAH Land Surface Model
Coupled
Global DataAssimilation
OceansHYCOM
WaveWatch III
NAM/CMAQ
11
Regi
onal
DA
Satellites + Radar99.9%
~2B Obs/Day
NOS – OFSGreat Lakes
Northern Gulf of Mexico
Bays• Chesapeake• Tampa • Delaware
SpaceWeather
ENLIL
Regi
onal
DA
2011
Computing Capability
“reliable, timely and accurate”
Web access to models as they run on the CCS
0
10
20
30
40
50
60
70
80
90
JULJANJULJANJULJANJULJANJULJANJULJANJULJANJULJANJULJANJULJANJULJAN
Popularity of NCEP Models Web Page
2001 2003 2005 2007 2009 2011
Current computers IBM Power6 73.1 trillion
calculations/sec 2 billion observations/day 27.8 million model
fields/day Primary: Gaithersburg, MD Backup: Fairmont, WV Guaranteed switchover in
15 minutes Next generation
computer: by Oct 2013 IBM iDataPlex Intel/Linux 143 trillion calc/sec Primary: Reston, VA Backup: Orlando, FL
Operational NWP Models
JCSDA Summer Colloquium Santa Fe, NM 13
The Global Observing System
• Evolving in which direction? Consider:• Number of US Operational SC and sensors
• Decreasing• Number of Total SC and sensors
• Increasing• Number of observations available
• Increasing• Quality of observations available• Timeliness (latency)
JCSDA Summer Colloquium Santa Fe, NM 14
“It’s not having what you want; it’s wanting what you’ve got”
A word on balance Environmental Modeling as a three-legged stool: Observations
Becoming a commodity? Computational (and communications) capacity Intellectual
Modeling Data Assimilation & QC
JCSDA Summer Colloquium Santa Fe, NM 15
Summary (Satellite) Data Assimilation will grow in
importance More Applications Increasingly coupled applications More varied sensors Part of the most cost-effective component to
invest in for improved modeling.
JCSDA Summer Colloquium Santa Fe, NM 16
Back up slides
JCSDA Summer Colloquium Santa Fe, NM 17