Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fixing an incorrect link #10

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fixing an incorrect link #10

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5e76eaf
Fixing an incorrect link
ryanbanderson Jun 13, 2017
3bcc892
Merge branch 'master' into master
ryanbanderson Jun 15, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 5 additions & 7 deletions _posts/2016-12-18-chemcam.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,12 +14,13 @@ Since ChemCam can be used quickly and from a distance, it is used almost every d

## How it works
Laser-induced breakdown spectroscopy (LIBS) works by concentrating intense pulses of laser light on a target, so that a small amount of the target is “ablated” and turned into a plume of plasma. That plasma plume gives off light that can be passed to a spectrometer and analyzed to identify the emission lines of chemical elements in the target. Under normal Earth-like atmospheric pressure, the LIBS plasma plume can’t expand very much, while under hard vacuum conditions the plasma plume is unconfined and expands really fast. In both cases, it is not especially bright. On the other hand, Mars is an ideal environment for LIBS because its thin atmospheric pressure is “just right,” giving nice bright sparks. As a bonus, the shockwave formed by the expanding plasma has enough force to clear away the pervasive dust that covers everything on Mars.

Since LIBS works by ionizing the target material, it theoretically can detect any element that has emission lines (in other words, all of them). However, it is most sensitive to elements that are easy to ionize and have nice, bright lines. That means ChemCam is particularly good at identifying alkali and alkali earth metals (even very light elements like Li). It is least sensitive to elements like chlorine and sulfur that are harder to ionize, though they are still detectable at high enough abundances. ChemCam uses three spectrometers to span a spectral range from 240 nm to 850 nm where most common elements have emission lines. As shown below, it is possible to say quite a lot about a target just by looking for certain diagnostic lines in the spectra. To help with this sort of analysis, you can use the NIST [spectral database](http://physics.nist.gov/PhysRefData/ASD/lines_form.html), or download [this](http://pds-geosciences.wustl.edu/workshops/chemcam_workshop_2015/4a._cquest_v2.5.0.jar) handy tool developed by Agnes Cousin (also see this [presentation](http://pds-geosciences.wustl.edu/workshops/chemcam_workshop_2015/4._Cousin_CQUEST_2015.pdf) for details on how to use the tool).

![Figure 1](/img/posts/rbanderson/ccam_example_spectra.png)
_Fig. 1: Example ChemCam spectra from Mars of targets with very different compositions._

## A Typical Observation
##A Typical Observation
Before we discuss how the data are processed, it’s worth talking about how a typical ChemCam observation is conducted. For LIBS to work, the laser light must be focused on a small enough spot to generate plasma. The ChemCam spot size varies slightly depending on distance to the target, but is less than 0.6 mm. To get a better idea of the variability of composition in a target, we usually analyze multiple locations or “points” on the target. These points are organized in a grid (e.g. 3x3) or a straight line (e.g. 1x10). At each point, we typically use 30 laser shots which allows any dust to be cleared away (usually within the first 5 shots) with a good number of spectra left over to average together to get a good measurement. The spectra can also be analyzed individually to look for variations with depth as the layer ablates away the upper few microns of the target. ChemCam spectra are grouped together by point, so for a 1x10 observation of a target, there will be 10 files, each of which contains 30 spectra (one per shot). For each point, we also collect a “passive” or “dark” spectrum, where the laser is left off and the spectrometer just collects reflected sunlight.

![Figure 2](/img/posts/rbanderson/ccam_rmi_example.png)
Expand Down Expand Up @@ -73,11 +74,8 @@ The latest ChemCam calibration uses a combination of ICA Regression and sub-mode
![Figure 6](/img/posts/rbanderson/submodels_graphic.png)
_Fig. 6: Example from Anderson et al., 2016 illustrating the concept of sub-model PLS regression._


There’s a lot more involved in optimizing regression methods and estimating their accuracy, but I have already gone on long enough in this post so I will just say that for the PLS models we use multi-fold cross-validation using folds stratified on the composition of interest. If you’d like to know more, please contact me!

## Getting the Data
The main place to look for ChemCam data is on the Planetary Data System (PDS). Raw data is available [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-2-edr-v1/mslccm_0xxx/) and processed data is [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/data/moc/). If you’d rather skip all of the complicated stuff above and just want the derived major element oxide compositions, they are available [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/data/moc/). You may also want to take a look at the “[master list](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/document/msl_ccam_obs.csv)” which contains a lot of useful metadata about all of the ChemCam observations, including indications of spectra that were excluded from the PDS due to saturation, low signal, or major element composition totals greater than 110 wt.% (this indicates a problem with one or more of the major element predictions). For RMI images, they are available on the PDS, but your best bet is to get the annotated RMI mosaics from the ChemCam [website](http://www.msl-chemcam.com/index.php?menu=images_result&rubrique=63&art=587&titre_url=Results%20-%20ChemCam%20-%20Results&titre_url=RMI%20Mosaics%20-%20RMI%20Mosaics#.WFbjDnrAvEY).

## Working with the Data
##Getting the Data
The main place to look for ChemCam data is on the Planetary Data System (PDS). Raw data is available [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-2-edr-v1/mslccm_0xxx/) and processed data is [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/data/). If you’d rather skip all of the complicated stuff above and just want the derived major element oxide compositions, they are available [here](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/data/moc/). You may also want to take a look at the “[master list](http://pds-geosciences.wustl.edu/msl/msl-m-chemcam-libs-4_5-rdr-v1/mslccm_1xxx/document/msl_ccam_obs.csv)” which contains a lot of useful metadata about all of the ChemCam observations, including indications of spectra that were excluded from the PDS due to saturation, low signal, or major element composition totals greater than 110 wt.% (this indicates a problem with one or more of the major element predictions). For RMI images, they are available on the PDS, but your best bet is to get the annotated RMI mosaics from the ChemCam [website](http://www.msl-chemcam.com/index.php?menu=images_result&rubrique=63&art=587&titre_url=Results%20-%20ChemCam%20-%20Results&titre_url=RMI%20Mosaics%20-%20RMI%20Mosaics#.WFbjDnrAvEY).
##Working with the Data
If all of the complicated stuff above makes you want to dig in and get your hands dirty with processing your own data, then you can do that too! If you are a skilled programmer, then you can probably take the raw or CCS-processed data and run with it. However, if you want a head start, I have been working with a student on a NASA PDART-funded effort to develop a collection of code for analyzing ChemCam data (and other spectral data) called PySAT (Python Spectral Analysis Tool). It is still very much in development, but if you’re willing to put up with some bugs it may still be useful for you, and we’re constantly improving it. The “back end” code is available at [this repository](https://github.com/USGS-Astrogeology/PySAT), and the “front end” graphical interface is available [here](https://github.com/USGS-Astrogeology/PySAT_Point_Spectra_GUI). The GUI doesn’t yet implement everything that is possible to do with the back-end, and the back-end in turn doesn’t yet do everything that I eventually hope it will be able to. However, if you’re interested in this sort of thing, please do take a look! If you try the code and run into any issues or would like to request additional features, please get in touch with me so that we can made the code as useful as possible!