Automated Monitoring of Volcanic Ash Micro and Macro-Physical Properties: A comparison of Future and Current Satellite Ins...
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
<ul><li>The Eyjafjallajökull Eruption: </li></ul><ul><li>Nearly 100,000 canceled flights </li></ul><ul><li>Airlines were l...
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
Volcanic Ash Properties <ul><li>Retrievable ash cloud properties:  effective temperature, emissivity, and a microphysical ...
In order to provide global volcanic ash products at high temporal resolution, the following imaging sensors are needed. X ...
Volcanic Ash Properties <ul><li>Regardless of the channel combination, the same ash cloud properties (ash height, mass loa...
Volcanic Ash Properties <ul><li>Quantitative ash detection (Pavolonis 2010) is expressed as an ash confidence. </li></ul><...
Volcanic Ash Properties <ul><li>Ash mass loading (ton/km 2 ) is needed to determine if jet engine tolerances are exceeded ...
Volcanic Ash Properties <ul><li>The ash cloud top height is critically important for determining if ash is at jetliner cru...
Volcanic Ash Properties <ul><li>The ash cloud effective particle radius is not a required GOES-R product, but it is automa...
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
May 7, 2010 (14:00 UTC) Ash clouds
May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul>Ash clouds
May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul>Ash clouds
May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul><ul><li>11/12 ...
May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul><ul><li>11/12 ...
February 12, 2010 (05:30 UTC) Ash cloud
<ul><li>Single channel IR window </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
<ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
<ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul><ul><li>11/12   m retrieval </li></ul...
<ul><li>Single channel IR window </li></ul><ul><li>11/13.3   m retrieval </li></ul><ul><li>11/12   m retrieval </li></ul...
Ash Cloud Height Validation 11/12   m 11/12/13.3   m 11/13.3   m
11/13.3   m 11/12   m 11/12/13.3   m Ash Mass Loading Validation
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
<ul><li>The ash cloud property products can be used to issue automated ash cloud alerts to VAAC’s. </li></ul><ul><li>Decis...
Ash trajectories initialized using GOES-R retrievals Model trajectories courtesy of Brad Pierce (NOAA/NESDIS)
Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabili...
Summary <ul><li>In order to provide global information on ash cloud properties in a timely manner, a flexible retrieval al...
Eyjafjallajökull (05/08/2010, 15:00 UTC) -  11/13.3   m Algorithm
Eyjafjallajökull (05/08/2010, 15:00 UTC) -  11/12   m Algorithm
Eyjafjallajökull (05/08/2010, 15:00 UTC) -  11/12/13.3   m Algorithm
11/13.3   m 11/12   m 11/12/13.3   m
Soufriere Hills (02/12/2010, 04:00 UTC) -  11/13.3   m Algorithm
Soufriere Hills (02/12/2010, 04:00 UTC) -  11/12   m Algorithm
Soufriere Hills (02/12/2010, 04:00 UTC) -  11/12/13.3   m Algorithm
11/13.3   m 11/12   m 11/12/13.3   m
Upcoming SlideShare
Loading in …5
×

MO4.L10 - AUTOMATED MONITORING OF VOLCANIC ASH MICRO- AND MACRO-PHYSICAL PROPERTIES: A COMPARISON OF CURRENT AND FUTURE SATELLITE INSTRUMENT CAPABILITIES

544 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
544
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
6
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • Aya (day)-fee-ought-lie-urkell The goal of this talk is to illustrate that while we can retrieve macro and micro physical properties of ash clouds using infrared channels available on imaging sensors like GOES and AVHRR, the additional infrared spectral information provided by more advanced sensors, like the next generation GOES imager, allows for a significant increase in product accuracy. We focus on infrared measurements since it is important to be able to monitor volcanic ash clouds consistently day and night. -no algorithm details
  • A visual appreciation for the impacts on aviation in Europe.
  • -Optimal estimation framework provides error estimates -As you might imagine, the three channel combination, which is used by the official GOES-R volcanic ash algorithm, is most desirable -Why does the retrieval need to accommodate three different channel combinations? -Next, introduce each panel in more detail.
  • Not all sensors are created equal, channels vary from sensor to sensor.
  • -Optimal estimation framework provides error estimates -As you might imagine, the three channel combination, which is used by the official GOES-R volcanic ash algorithm, is most desirable -Why does the retrieval need to accommodate three different channel combinations? -Next, introduce each panel in more detail.
  • While not covered in this talk, we do apply a sophisticated ash detection algorithm to determine which pixels to perform a retrieval on.
  • We are seeking GOES-R RR funding to pursue this. We need to works towards fulfilling the 5 minute warning criteria desired by the aviation industry.
  • MO4.L10 - AUTOMATED MONITORING OF VOLCANIC ASH MICRO- AND MACRO-PHYSICAL PROPERTIES: A COMPARISON OF CURRENT AND FUTURE SATELLITE INSTRUMENT CAPABILITIES

    1. 1. Automated Monitoring of Volcanic Ash Micro and Macro-Physical Properties: A comparison of Future and Current Satellite Instrument Capabilities Michael Pavolonis (NOAA/NESDIS/STAR) Marco Fulle - www.stromboli.net
    2. 2. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    3. 3. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    4. 4. <ul><li>The Eyjafjallajökull Eruption: </li></ul><ul><li>Nearly 100,000 canceled flights </li></ul><ul><li>Airlines were losing $200 million/day </li></ul><ul><li>Total economic impact - $2 billion </li></ul>Before Ash Event During Ash Event Why is Volcanic Ash Monitoring Important?
    5. 5. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    6. 6. Volcanic Ash Properties <ul><li>Retrievable ash cloud properties: effective temperature, emissivity, and a microphysical parameter (mainly related to particle size) </li></ul><ul><li>From these retrieved parameters, the effective ash cloud height, mass loading, and effective particle radius can be calculated. </li></ul><ul><li>The retrieval is performed using optimal estimation (e.g. Heidinger and Pavolonis, 2009). </li></ul><ul><li>Possible channel combinations: </li></ul><ul><li>1). 11/12  m </li></ul><ul><li>2). 11/13.3  m </li></ul><ul><li>3). 11/12/13.3  m </li></ul>
    7. 7. In order to provide global volcanic ash products at high temporal resolution, the following imaging sensors are needed. X LEO VIIRS X GEO SEVIRI X GEO MTSAT X LEO MODIS X GEO FY2 X GEO GOES-R ABI X GEO GOES-12 - GOES-15 Imagers X GEO GOES-11 Imager X LEO AVHRR 11/12/13.3  m channel combination 11/13.3  m channel combination 11/12  m channel combination Orbit Sensor
    8. 8. Volcanic Ash Properties <ul><li>Regardless of the channel combination, the same ash cloud properties (ash height, mass loading, and effective particle radius) are always produced. </li></ul>
    9. 9. Volcanic Ash Properties <ul><li>Quantitative ash detection (Pavolonis 2010) is expressed as an ash confidence. </li></ul><ul><li>Ash detection results can be overlaid on false color imagery to give the user perspective. </li></ul><ul><li>The ash detection can be used to provide automated ash alerts. </li></ul>Quantitative Ash Detection
    10. 10. Volcanic Ash Properties <ul><li>Ash mass loading (ton/km 2 ) is needed to determine if jet engine tolerances are exceeded and to initialize models. </li></ul><ul><li>If a 1 km cloud thickness is assumed, the mass loading is numerically equivalent to ash concentration in mg/m 3 . </li></ul>Ash Mass Loading
    11. 11. Volcanic Ash Properties <ul><li>The ash cloud top height is critically important for determining if ash is at jetliner cruising altitudes (nowcasting component). </li></ul><ul><li>In addition, the ash cloud height is a very important parameter for initializing dispersion models (forecasting component). </li></ul>Ash Cloud Height
    12. 12. Volcanic Ash Properties <ul><li>The ash cloud effective particle radius is not a required GOES-R product, but it is automatically generated as part of the ash retrieval. </li></ul><ul><li>The effective particle radius is well correlated with ash residence time. </li></ul>Ash Effective Radius
    13. 13. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    14. 14. May 7, 2010 (14:00 UTC) Ash clouds
    15. 15. May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul>Ash clouds
    16. 16. May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul>Ash clouds
    17. 17. May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul><ul><li>11/12  m retrieval </li></ul>Ash clouds
    18. 18. May 7, 2010 (14:00 UTC) <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul><ul><li>11/12  m retrieval </li></ul><ul><li>11/12/13.3  m retrieval </li></ul>Ash clouds
    19. 19. February 12, 2010 (05:30 UTC) Ash cloud
    20. 20. <ul><li>Single channel IR window </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
    21. 21. <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
    22. 22. <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul><ul><li>11/12  m retrieval </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
    23. 23. <ul><li>Single channel IR window </li></ul><ul><li>11/13.3  m retrieval </li></ul><ul><li>11/12  m retrieval </li></ul><ul><li>11/12/13.3  m retrieval </li></ul>February 12, 2010 (05:30 UTC) Ash cloud
    24. 24. Ash Cloud Height Validation 11/12  m 11/12/13.3  m 11/13.3  m
    25. 25. 11/13.3  m 11/12  m 11/12/13.3  m Ash Mass Loading Validation
    26. 26. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    27. 27. <ul><li>The ash cloud property products can be used to issue automated ash cloud alerts to VAAC’s. </li></ul><ul><li>Decision support systems like this are needed because it is impossible to manually analyze every satellite image in real-time. In addition, the high temporal resolution of future geostationary measurements (like from GOES-R) will not be fully utilized for volcanic cloud monitoring without an automated alert system. </li></ul>Text Warning Quantitative description of ash cloud Product Quick-look
    28. 28. Ash trajectories initialized using GOES-R retrievals Model trajectories courtesy of Brad Pierce (NOAA/NESDIS)
    29. 29. Topics <ul><li>Motivation </li></ul><ul><li>Retrieving volcanic ash properties </li></ul><ul><li>Comparing sensor capabilities </li></ul><ul><li>Applications for decision support </li></ul><ul><li>Summary </li></ul>Marco Fulle - www.stromboli.net
    30. 30. Summary <ul><li>In order to provide global information on ash cloud properties in a timely manner, a flexible retrieval algorithm is needed to accommodate three different infrared channel combinations (11/12, 11/13.3, and 11/12/13.3  m). </li></ul><ul><li>Comparisons to spaceborne lidar indicate that the 11/12/13.3  m channel combination, which was developed for GOES-R, is significantly more accurate than the 2-channel combinations, especially for optically thin high ash clouds. </li></ul><ul><li>Using SEVIRI, the GOES-R products were provided to the London VAAC during the eruption of Eyjafjallajökull. </li></ul><ul><li>The 2-channel combinations still offer valuable information on ash cloud properties, especially if bias corrected. </li></ul><ul><li>Current efforts are focused on developing a global, multi-sensor automated ash alert system and model initialization and assimilation studies. </li></ul>
    31. 31. Eyjafjallajökull (05/08/2010, 15:00 UTC) - 11/13.3  m Algorithm
    32. 32. Eyjafjallajökull (05/08/2010, 15:00 UTC) - 11/12  m Algorithm
    33. 33. Eyjafjallajökull (05/08/2010, 15:00 UTC) - 11/12/13.3  m Algorithm
    34. 34. 11/13.3  m 11/12  m 11/12/13.3  m
    35. 35. Soufriere Hills (02/12/2010, 04:00 UTC) - 11/13.3  m Algorithm
    36. 36. Soufriere Hills (02/12/2010, 04:00 UTC) - 11/12  m Algorithm
    37. 37. Soufriere Hills (02/12/2010, 04:00 UTC) - 11/12/13.3  m Algorithm
    38. 38. 11/13.3  m 11/12  m 11/12/13.3  m

    ×