'Trajectory analysis for the OTAG domain...' is one of 10 general tasks identified on the OTAG Air Quality Analysis (AQA) 'wish list'. While no extramural funding costs were identified to support this task, the States of VT and TX had indicated some ability to contribute in-kind services. During a 1/24/96 AQA Workgroup conference call, several participants expressed capability/interest in assisting with this task. Subsequently, several other individuals have expressed interest in assisting with OTAG 'Trajectory Analysis' Tasks, and it has been suggested that we form an 'Ad Hoc Air Trajectory group' (AHAT) within the OTAG Air Quality Analysis Workgroup. This memo is intended to initiate discussions in the following areas:

- Identify individuals interested in participating on an OTAG AHAT team (Who's Interested?)

- Identify most effective means for AHAT communications and data exchange (What's the Line?)

- Seek consensus definition(s) of "Trajectory Analysis" in OTAG context (What's a Trajectory?)

- Identify the potential uses of Air Trajectory Analysis in the OTAG process (What's the Use?)

- Seek recommendations on Trajectory Analysis Approaches for OTAG(What's the Method?)

- Seek recommendations on specific Trajectory analysis tasks for OTAG (What are the Jobs?)

- Identify individuals or groups who will conduct the specific tasks (Who are the Workers?)

- Provide information on VT DEC Trajectory Analysis progress to date (What's up in VT?)

Please consider this a first draft of a dynamic document. I will state my opinions, and ask each reader to add suggestions, comments, revisions, etc. Your feedback and participation are essential!

Who's Interested?

The following individuals may have expertise and/or interest in Trajectory Analysis - as applied to OTAG goals. Sorry if I've misspelled, forgotten some or added some with no interest. Please advise.

What's the Line?

What are the most effective means of communication and collaboration among participants of the OTAG AHAT group? (Yet another series of bi-weekly) conference calls, (another 100 pages/week of "urgent") faxes, and (another series of expensive, time consuming) face to face meetings - are not conducive to "doing work" (although occasional conference calls, faxes and meetings are desirable). The interactive structure, function and content of the OTAG Air Quality Analysis Internet Server (http://capita.wustl.edu/OTAG) provide an effective, efficient vehicle for exchange of technical ideas, data, results and reports. Group or individual Internet E-mail, and Internet FTP file transfer could also facilitate exchange of "short" communications and "large" data files. For those who lack Internet connections, you almost certainly have a local Internet access provider who offers a modem-based connection for all of the above at a very modest cost (less than $1/hour of use). Commercial services - such as America On Line and CompuServe - offer full Internet access at a nominal cost of about $10/ month. I think the advantages of organized electronic communications outweigh the limitations, and will gladly provide technical assistance to any AHAT participant who lacks connections - via telephone at (802) 241-3840.

What's a Trajectory?

Trajectory analysis is not a precisely-defined term, and may need some clarification for OTAG purposes. In the simplest sense, a trajectory is an estimation of the path of air movement over space and time. While meteorologists often calculate trajectories "by hand" from atmospheric meteorological data, calculations are commonly facilitated by several currently available computerized trajectory models, such as ARL-ATAD (Heffter, 1981), HY-SPLIT (Draxler, 1992) and CAPITA Monte Carlo (Schichtel, 1995). Trajectory models require input of gridded, 4 dimensional meteorological data - which itself must be derived from a meteorological model, such as NGM (Rolph, 1992), RAMS (Pielke, 1992), etc. Trajectories are typically run in a backward mode (path of air movement arriving at a receptor location) or forward mode (path of air movement leaving from a source location). They can also be calculated with alternative assumptions of critical atmospheric physics, for example, at fixed (or mixed) elevations, pressures or isentropes in the atmosphere.

Foreword or backward trajectory model output is most commonly displayed as a "simple" 2 dimensional (horizontal) path of air movement on a map view. However, the model calculations and output include substantial additional information - including vertical motion, and various meteorological characteristics (pressure, humidity, clouds, precipitation, temperature, solar radiation, mixing heights, etc.) along the trajectory path). A more comprehensive term to incorporate this important supplemental information is "Airmass History". The gridded photochemical models are, in effect, calculating forward airmass histories for thousands of individual release locations and start times (along with the complex simultaneous influences of fluctuating emissions and atmospheric chemistry).

It should be recognized that any model-derived "Airmass History" is, by definition, a "Revisionist History". Results are dependent on: which trajectory model, which met driver(s), which physical assumptions on horizontal and vertical movement and/or mixing, which starting elevations, which surface influences on flows, etc. Likewise, pollution at a given receptor location has invariably resulted from emissions from a large number of source locations; while emissions from an individual source location ultimately influence a wide range of receptor sites. Where did today's ozone come from? From the models, as from the atmosphere, we should anticipate more than one answer.

What's the Use?

What are the potential uses of trajectory analysis (or airmass histories) for OTAG data analysis?

For OTAG Model Episodes:

• Illustration/visualization of meso-scale and synoptic-scale atmospheric flow during model episodes
• Sensitivity analysis of alternative meteorological drivers (and/or model default settings)
• Evaluation of model results (confirm good fits; explain poor fits; illustrate effective strategies)
• Help distinguish among effects of local, meso-scale, and synoptic-scale transport (or lack thereof)
  

For Other Time Periods:

• Illustrate Transport (or lack thereof) for other short-term episodes
• Evaluate "representativeness" (or lack thereof) of selected OTAG model episodes
• Develop long-term trajectory climatologies for specified source or receptor locations
• Analysis of ambient ozone, precursor data, or statistical indexes (S.T. Rao's Baseline, etc.)
  

Other potential uses:

• Forward Trajectories with emissions data, ambient air quality data, simple chemistry, depletion¼
• Backward trajectories with ambient data to estimate/ confirm source strengths
• Other?

What's the Method?

Currently Available Trajectory Models:

HY-SPLIT (give address of NOAA ARL ftp site) paulw@qtm.anr.state.vt.us

• CAPITA Monte Carlo (Rudy, Brett - ?) bret@mecf.wustl.edu
• EPA "Trajectory"
• ARL ATAD, BAT
• Other?

Currently Available Meteorological Data Drivers:

• NGM
• RAMS
• MM4
• Other?
  

Critical Model Operating Assumptions:

• Starting Elevation(s)?
• Follow Elevation, Terrain, Pressure, etc. ?
• Use of sub-layers near the surface?
• Time duration (hours-days), time periods (years of available data), Missing data periods?
• Accounting for dispersion (forward), "reflective dispersion" (backward), transformation, deposition, precipitation, day vs. night, vertical motion, etc.
• Other?

Analysis and Display of trajectory results:

• Map view, vertical view, single, multiple, source, receptor, animations, etc.
• Statistical sorting and/or aggregation (residence time analysis, cluster analysis, etc.

What are the Jobs? And Who are the Workers?

If there are no funds to support additional trajectory work, it seems critical to couple the questions of "what to do" and "who will (volunteer to) do it". At the same time, if good ideas are developed which require supplemental funding support, the AHAT group should seek to identify those potential tasks and funding needs (sometimes the system works in mysterious ways -and good ideas can sometimes attract sufficient funding support)

Specific Tasks:

• Residence Time Analysis (and other ensemble / trajectory climatology techniques) - VT DEC will participate, and is eager to collaborate with other interested participants
• What else?
• Who else?

What's up in VT?

VT DEC has initiated some exploratory analysis using the NOAA HY-SPLIT model and 1989-95 NGM met data. A 2/29/95 draft summary report is available via OTAG Web server at (address), and is also available in WordPerfect 6.1 format via anonymous ftp at (address) or as hard copy (B&W-only) from VT DEC.