Utilities for working with low-level (raw sample) Incoherent Scatter Radar data, especially from Poker Flat AMISR (PFISR)
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Examples
gfx
isrutils
matlab
precompute
tests
.codeclimate.yml
.gitattributes
.gitignore
.travis.yml
2007-03-18.ini
2007-03-23.ini
2011-03-01.ini
2011-03-02.ini
2012-11-07.ini
2013-04-14.ini
2013-04-14_60-90.ini
2013-05-01.ini
CombineISRhist.py
FakeACF.py
LICENSE
README.md
Wedlund2013.py
setup.cfg
setup.py
singleplot.py

README.md

image image image

AMISR raw data utilities

Utilities for working with Incoherent Scatter Radar data, especially from Poker Flat AMISR.

We work with the complex I+jQ voltage samples, the lowest level data available from the radar, on a single pulse basis. Depending on the beam pattern and pulse modulation, the per-beam pulse cadence is perhaps on the 75 milliscond time scale.

Coming soon, simultaneous plots with high speed multi-camera synchronized video.

Install

python -m pip install -e .

Usage

Several types of "raw" data exist inside the manually-requsted I+jQ voltage files. They can be loaded with several different functions. All of these examples assume first doing:

import isrutils as iu

P is a dict() with parameters such as altitude range, beam number. See the numerous examples for necessary parameters.

fn is the ISR HDF5 .h5 file to process.

  • Raw power hypot(I,Q)
    snrsamp, azel, isrlla = iu.readpower_samples(fn, P)

Plotting

singleplot.py is a main program used to examine raw ISR data. It's configured via .ini files. Some important parameters are:

parameter description
scan CFAR detection of turbulent activity (possible association with Alfven waves)
tlim unless scan=yes, usually you use tlim to only plot over time range of interest (to avoid enormous amount of plots)

Examples

From the Akbari GRL 2012: Anomalous ISR echoes preceding auroral breakup: Evidence for strong Langmuir turbulence doi:10.1029/2011GL050288

Figure 1a Akbari 2012

Figure 3a Akbari 2012

Figure 3b Akbari 2012

Figure 3c Akbari 2012

File Types

Currently, raw ISR data files are not currently contained on Madrigal, you will have to email SRI staff to get them manually.

When requesting raw AMISR data, please request by experiment name as this is more convenient for SRI staff than the date/time.

Here is a limited selection of raw ISR data. It is indexed by date; under each date look for the ISR folder. You will typically want to download all four file types noted in the table below.

  • dt0.h5 Ion Line: Alternating Code
  • dt1.h5 Downshifted Plasma line (negative Doppler shift)
  • dt2.h5 Upshifted Plasma line (positive Doppler shift)
  • dt3.h5 Ion Line: Long Pulse (small Doppler )

Discussion

The "ion line" measurement bandwidth is ~ +/- 100 kHz from the radar center frequency, and contains the data necessary for volume estimates of Electron Density, Ion Temperature, Electron Temperature, and Ion Velocity, under certain assumptions for species composition vs. altitude. Some of the need to make assumptions about atmospheric composition can be mitigated with combined ion/plasma line inversion, among numerous other benefits. The plasma line returns have several MHz of bandwidth, but most of the energy is contained in narrower bands upshifted and downshifted from the center frequency.

No one radar waveform is optimal for all conditions, particularly with regard to the spatio-temporal sampling dilemma. Incoherent scattering from tiny particles gives exceedingly weak returns, and even with many billions of particles in the scattering volume, it takes well over ten thousand radar pulses to build a statistical basis for a usable autocorrelation function (ACF). The shape of the ACF is fitted to estimate certain plasma parameters, given assumptions on the particle population that may be violated, causing in some limited sets of cases either inaccurate fits or a failure to estimate the parameters.