AHAT Conference Call Minutes - 3/29/96


TO: OTAG Ad Hoc Air Trajectory (AHAT) Team

FROM: Rich Poirot, Quasi Chair

SUBJECT: Minutes of AHAT Conference Call on March 29, 1996, at 1:00 p.m.

DATE: April 3, 1996

Mr. Poirot opened the discussion by reviewing the proposed agenda beginning with AHAT goals and ground rules.

AHAT Goals and Ground Rules

Mr. Poirot suggested creating a forum for the exchange of technical information among individuals who are interested in applications of air trajectory analysis to ozone and related pollutants. He added that it would be useful to create a forum to help identify, conduct, and review specific air trajectory analysis tasks for submission to the Air Quality Analysis Subgroup.

Mr. Poirot's ground rules included:

He reiterated that the process should remain open with few rules in order to foster the exchange of information. Anyone is permitted to submit any item or review comments; however, a record of any formal actions should be maintained. Mr. Poirot concluded that he has no personal interest in being the chairperson and suggested a revolving leadership.

Bob Collom commented on time frames. He said that the Air Quality Analysis Subgroup needed to report to the Policy Group by June or July.

Mr. Poirot encouraged participants to use the Web server to make the public aware of the group's existence. Cliff Michaelsen reported that John Merrill and Jennifer Moody from the University of Rhode Island would be interested in participating in the mini-workgroup because they have done trajectory analysis for NOAA, and he feels that they would be valuable additions to the group. Mr. Michaelsen will contact them himself. Jerry Keeler at the University of Michigan may also be interested in participating and would like his name added to the list.

Mr. Poirot then commented on communication devices for AHAT group members and encouraged communicating via the Web server or E-mail instead of facsimile. Carol Sue Malinowski of Walcoff & Associates, Inc., urged members to forward all information to her in electronic format. She also requested everyone's mailing address in the event that the size of documents prevents electronic transmission.

Mr. Poirot defined the trajectory calculation as much more than just simply a map view of an estimated path of horizontal air mass movement. Mr. Collom reported that the trajectory explanation should include what input data is used in the models. Consensus opinions should also be considered. Mr. Poirot concurred and suggested re-working the outline into a report.

Mr. Poirot reported that both back and forward trajectories will be useful.

Paul Wishinski stated that because individual groups studying trajectories may employ different initial assumptions and different input data, the results should be carefully explained and qualified before they are brought before the Air Quality Analysis Subgroup. Mr. Poirot agreed that an explanation of statements is very important because of a sensitivity issue.

Bret Schichtel commented that by including multiple trajectory analyses from different models with different assumptions and different input data--similarities rather than merely differences­­could and should be examined. Mr. Poirot concurred.

Mr. Poirot added that the potential uses of trajectories in support of OTAG goals would include-

(1) Confirmation, evaluation, and illustration of some of the gridded model results.
(2) Explanation for poor fits of models in certain locations and times.

One advantage trajectory analysis has over photochemical grid models is that it is cheaper and can be done more rapidly for much longer time periods using fewer resources. The variety of approaches to trajectory analysis is large, and the literature is full of different ways of calculating and interpreting results.

Currently available trajectory models include the HY-Split (Vermont), CAPITA, and the EPA trajectory model. The EPA model was used mainly during the early morning. The HY-Split model is based on archived upper-air data used for weather prediction, which represents good estimations about atmospheric conditions over long periods of time. The HY-Split model is better suited for long-range transport work. Mr. Wishinski stated that Vermont's focus was not to look at individual trajectory but rather to examine backward trajectories in a statistical way over long periods of time. It was mentioned that Dick McNighter with the University Of Alabama at Huntsville is the chief scientist in meteorology and modeling, and he may be a good resource.

Mr. Collom said that if the study were completed by September, results would still be considered by the Policy Group. However, if it could be finished in time to put into the Air Quality report by June or July, it would have a greater impact.

Mr. Collom suggested emphasizing the episode times that OTAG has selected and said that the trajectories may be beneficial 1 week before the episode and throughout the episode. Mr. Poirot agreed and said they would be useful long-term as well.

Update on VT DEC HY-Split Analysis (Mr. Paul Wishinski) Mr. Wishinski reported that large data sets are needed for HY-Split. This data can be made available as long as it doesn't have to be circulated over and over. He noted that the six elevation sites discussed in the report (from northern New York to Tennessee, and one on New York City's World Trade Center) were examined over a period of 7 years, four per day, with the assumption that these high elevation sites would represent a long-range transport situation because the ozone data measured at those sites do not represent a strong diurnal pattern. These were presented in draft form. Since then, 12 other sites in the OTAG domain were selected (five along the East Coast from Maine to Delaware, one in southern Vermont inland, and six upper Mid-West sites) to calculate trajectories for 7 years, four per day. When completed, statistical approaches can be applied.

Mr. Wishinski commented on the criteria regarding interpolation schemes and posed the following questions:

(1) What criteria are there for selecting how many levels should be interpolated?
(2) At what levels should they be interpolated?

He reported that so far interpolations have been performed at 50 meters, 100 meters, and 150 meters below the first level of MGM data. All seems to be working quite well on the coastal sites; however, it hasn't been critically examined from a meteorologist's point of view. Analysis has been performed for 7 years for all 12 new sites assuming the following: (1) the HY-Split is run with a default mode assuming no interpolation in the lower level; and (2) with an assumed three-level interpolation. Within a couple of weeks, draft summary plots should be available and will be distributed to the group.

Update on CAPITA Trajectory Server (Mr. Bret Schichtel)

Mr. Schichtel reported on the CAPITA model. He said it will soon be available on the Internet (http://capita.wustl.edu/otag), and it will calculate the trajectory and return a GIF file showing a map view and path of trajectories. Trajectories can be calculated for multiple locations, and the server can also return ASCII data. In addition to his work on the trajectory server, Mr. Schichtel is performing animations of forward and backward trajectories for 6-month time spans from 1991 to 1995. Movies for 1991 have already been generated and will be made available on the Internet.

Mr. Poirot suggested scheduling another conference call, with Mr. Schichtel taking the lead role, sometine after the CAPITA data server was in place. It was agreed that the next conference call would take place April 25, 1996, at 1:00 p.m.

Submit your comments, feedback, questions, and ideas pertaining this page. Your input will be automatically added to the existing annotations. In order to add a new comment, you must be registered with the OTAG/AQA Peoples Page.