Morning Lectures

Topics to cover:

  • ISR theory (Phil/Josh/others? Don Farley?)

  • Signal Extraction/Processing/Fitting at AMISR (Mike Nicolls)

  • Madrigal (Bill Rideout/Anthea Coster)

  • Radar Design/Signal Propagation

  • Experiment design: tradeoffs in fidelity, parameter observability, etc. which impact the final science available with the experiment

  • How to spot an outlier: with UAF staff help (who should be contacted early in the process anyhow), training yourself to see things which don't look quite right:

    1. impossible temperature ratios

    2. negative composition fractions

    3. etc.

Notes:

  1. Don Farley has been invited to help with ISR theory - waiting on his response.

  2. Signal extraction/processing/fitting needs to have the right amount of technical detail. UAF facility people will be taking care of the true gritty details of fitting so this shouldn't be a lecture to produce radar technique experts. We need informed data consumers.

  3. Phil confesses that I don't know what is meant by Signal Propagation. Also, radar and experiment design are other danger areas where we could veer off into transmitter design or other things which would be of interest only to EE students. How did the PARS summer school handle this balance between instrument design theory and science with the final reduced parameters? (NB: the PARS 2006 material is available.)

Afternoon Hands-On Sessions

Homework Assignments (sic): Staged hands-on activities to lead students through

  1. Proposing an AMISR experiment

  2. Scheduling and executing experiments

  3. Scientifically analyzing the resulting data

Potential groupings for student activities - reduces teacher-to-student ratios:

ELIZABETH NOTE: Don't we have 4 afternoons to work with? Friday afternoon will be time for presentations. It seems like we need one more activity. Perhaps we rotate through these activities on M-W and then on Thursday the groups have time to process their data and prepare presentations?

  1. Using Madrigal practically and productively

    1. Easy ways to import Madrigal data into your favorite working language

    2. Contextual parameters (geophysical and geomagnetic indices, IRI model, magnetic coordinates, etc.) available within Madrigal

  2. Millstone ISR tour

    1. Single antenna / transmitter system design

    2. Live radar demonstration

    3. Comparison of single antenna ISR with phased array AMISR

  3. Planning an AMISR experiment, selecting modes, and requesting final data

    1. How to select the right mode

    2. How to schedule

    3. Rules of the road for data access?

    4. Typical data availability times?

Items needed from the AMISR technical team:

  1. Reduced list of three or four experiment choices for student experiments, with a one-page style sheet available for each with the following:

    1. Mode name

    2. Parameters measured

    3. Relative parameter accuracy in standard analysis mode

    4. Altitude resolution

    5. Time resolution

    6. Spatial coverage

  2. Times available for student experiments

  3. Schedule deadlines for student experiments

  4. Data delivery details

    1. will realtime analysis be different than batch (Phil imagines it will), leading to different relative parameter accuracies and other tradeoffs? Accordingly, what will the student need to know about quick look vs. later detailed analysis?

    2. What turnaround for realtime analysis?

    3. What turnaround for batch analysis? Will this happen before the workshop ends?

More afternoon hands-on notes:

  1. What should the followup be with students on their designed experiments? More long-distance help will clearly be needed to get students through to using the data in their science experiments.

  2. Careful grouping of students is the only way Phil can see that we can cycle all students through experiment requests, which means they need to be grouped by common science topic.

ELIZABETH: How many actual experiments did we have proposed? Anthea and I discussed making it really simple and just assigning a science problem to each group. Then they have to think about how to use the radar to solve the problem and get the proper data requested. This week is not really for PhD research - it's to learn the basics of performing an ISR experiment and so it's okay for the science reasons to be "artificial" and hand-picked.

  1. Can we really do this practically?

  2. If not, what is the alternate model we should operate under?

  1. If we run out of time, what will the followup be to ensure all students get a chance to design and execute an experiment?

NOTES FROM CEDAR SUMMER SCHOOL DINNER MEETING

Tentative schedule of events:

  • Monday AM lecture: ISR theory and basic ionospheric physics (and history of ISR by Behnke?)

  • Monday PM: tour facility and introduction to Madrigal (alternating/rotating)

  • Tuesday AM lecture: ISR theory, basic ionospheric physics and some analysis/fitting

  • Tuesday PM: design and submit their PFISR experiments

  • Wednesday AM lecture: analysis and fitting

  • Wednesday PM: retrieve their data and analyze it

  • Thursday AM lecture: science and AMISR

  • Thursday PM: prepare their presentations

  • Friday AM: present their projects and evaluate the workshop

Student organization: 20 students will be divided into 2 or 3 groups for the tour. These will be the same groups that will work together on their PFISR projects.

Lecturers / staff:

  • Phil Erickson

  • Bill Rideout

  • Anthea Coster

  • Anja Strømme

  • Elizabeth Kendall

  • Josh Semeter [exact commitment TBD]

  • Mike Nicolls [provides material to Phil - CEDAR tutorial plus a little supplemental]

  • Rich Behnke: 10-15 min talk on personal ISR history

  • John Holt: helps with afternoon data reduction/interpretation

(It was discussed that keeping the number of lecturers to a minimum may make a more cohesive program.)

PFISR experiments: Each groups of students will design their experiment, submit it, retrieve their data from Madrigal, analyze it and present their results. Longpulse (LP) and alernating code (AC) will be the transmitted pulse scheme. The students will have the option of choosing the beam positions. Some obvious experiment goals are:

  • vector velocity vs altitude

  • vector velocity vs latitude

  • Ne vs latitude

They will submit their experiment via Todd's webpage and it will run during Tuesday night. Their runs will be processed and loaded into Madrigal in time for the Wednesday afternoon session.

Items to incorporate into the lectures:

  • levels of data (raw, fitted, resolved, etc.)

  • scale heights

  • time scales

  • resolution trade-offs

  • pitfalls

  • outliers, out-liers, out-lyers, out-liars, outright liars, big fat liars

  • error bars

  • assumptions

  • No labels