Feature Tracking Process Analysis
-
Upload
amethyst-day -
Category
Documents
-
view
39 -
download
0
description
Transcript of Feature Tracking Process Analysis
![Page 1: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/1.jpg)
Feature Tracking Process Analysis
5th Annual CMAS Conference
Friday Center
October 16-18, 2006
http://ftpozone.sph.unc.edu
Barron Henderson, William Vizuete, and Harvey Jeffries
![Page 2: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/2.jpg)
Outline
• History and Functionality• Recent Enhancements• Potential Applications
![Page 3: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/3.jpg)
Photochemical Grid Models: Explaining the Unexplainable
• As Oreskes (1994) and later Beck (2002) have demonstrated, atmospheric models are “open systems” that have “essentially unknowable” inputs
• Can have a wide variety of inputs– Generated by different groups– Minimum level of detail– Come from models with their own uncertainty
• Easily suffer from compensating errors– Getting the ‘right answer’ for the ‘wrong reasons’
• Model Performance Evaluations• Process Analysis
![Page 4: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/4.jpg)
Process Analysis Quantifies Model Processes
• 1994 - Jeffries, H. E., and Shawn Tonnesen. A Comparison of two Photochemical Reaction Mechanisms Using Mass Balance and Process Analysis. Atmospheric Environment 28 (18):2991-3003.
• In model algorithm:– Process Rates– Reaction Rates– Time-Step Averaged
• Post Processor– Extraction– Aggregation
y
x
z
x
y
![Page 5: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/5.jpg)
UT/UNC Collaboration Adds Variable Mixing Height
• 2005 - Vizuete, William. Implementation of Process Analysis in a Three-Dimensional Air Quality Model, Chemical Engineering, University of Texas - Austin, Austin.– Time Variable Mixing Height– Convex Shapes
z
x
z
x
y
![Page 6: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/6.jpg)
Outline
• History and Functionality
• Recent Enhancements• Potential Applications
![Page 7: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/7.jpg)
Enhancement 1: Converted Process Analysis to Python
• Python Based Process Analysis– Urban Airshed Model (UAM) File Interfaces – Spatial Aggregation– Entrain and Detrainment
• Increased Volume Shape Flexibility• Automated Mixing Height Identification
![Page 8: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/8.jpg)
Enhancement 2: Allow for Spatially Variable Mixing Height
cells
z
z
x
y
![Page 9: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/9.jpg)
Enhancement 3: Enable Focus Volume to Follow Ozone Peak
11 12
13 14
![Page 10: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/10.jpg)
Enhancements Require New Algorithms for En(De)trainment
y
x
x
z
• Vertical– Simultaneous
Entrainment and Detrainment
• Horizontal– Simultaneous
Entrainment and Detrainment
z
x
y
![Page 11: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/11.jpg)
Outline
• History and Functionality• Recent Enhancements
• Potential Applications
![Page 12: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/12.jpg)
Potential Applications
• Features of Interest– Concentration Peaks– Chemical Plumes– Impacts of Mega-cities on surroundings– Transcontinental Chemical Transport– Wildfires– Airplane Observations– Any Moving Feature!
• Moving Process Analysis allows us to quantify transported and local processes and their interactions
![Page 13: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/13.jpg)
Acknowledgments
• Funding From:– 8 Hour Ozone Coalition– HARC H60 “Regional Transport Modeling for East Texas” -
Jay Olaguer, Project Officer
• Thanks to:– Jim Smith and TCEQ for providing CAMx ready files– Dr. Kimura at UT for his work on the previous versions of
Process Analysis– Dr. Byeong-Uk Kim at Georgia Dept. of Natural Resources– The rest of the UNC MAQ Lab Group
![Page 14: Feature Tracking Process Analysis](https://reader035.fdocuments.us/reader035/viewer/2022062422/56813919550346895da0c413/html5/thumbnails/14.jpg)
Questions?11
12
13
14
http://ftpozone.sph.unc.edu/