Monday, November 29, 2010

Group Geology Projects

It is the last week of lecture at UC Davis.  It is understandably difficult to get everyone's schedule to align in order to finish up group projects... as a group.

For now, I'm relying on Google Docs to allow group members to collaborate on finalizing the presentation and written report.  I have no idea how well this will work.

In the end, I'll take all the changes from the written report and put it into the master LaTeX document for printing.

While LaTeX is very portable, the world of Microsoft forbids it.  So I export my LaTeX documents to OpenDocument format.  Generally, that works good enough.  Then I save as a Word Doc in Open Office and import it to Google Docs or forward the .doc for collaboration.  Why should I use LaTeX when I have to go through these steps for collaboration?  Mainly, it outputs beautiful pages without any work, cross referencing figures can't be easier, and... BibTeX.  Bibliography can be a joy!

Finals are next week.

Saturday, November 27, 2010

QGIS 1.6 Released

Time to map more geology in a GIS:  QGIS 1.6 is out of the oven!  Mac users can get it here  and everyone else, here.

This release is already awesome, because it has a remote sensing and geology-friendly splash screen.

Friday, November 19, 2010

Georeferencing with GDAL 1.7 and QGIS 1.6 Trunk

Yesterday I set out to georeference the field map I made during the trip to Carrizo Plain National Monument.  I did not expect it to go quickly.  It did.

There is a plugin available for QGIS 1.6 Trunk that provides a very robust way to easily georeference images using GDAL.

You simply load the raster, plot points at your control points on the map (such as graticule crossings), choose a transformation (helmert, linear, polynomial 1-3, thin plate spline), an output raster name, and your target projection.  Then tell it to have at it.

If you've georeferenced using ArcMap 9.3, you'll find the interface in this QGIS plugin to be a bit different.  Instead of having to pan back and forth, zoom in and out, as in ArcMap, you have two views to work with in QGIS.  The plugin view provides only the image you are georeferencing and then there's the map view in the main QGIS window.  As you georeference your point by some coordinate (e.g., decimal degrees or UTM), those points show up in red within the main QGIS window (providing you're in the area you are trying to georeference the raster to).  Thus you can verify your georeferencing as you plot without having to pan and zoom!  I hope I'm being clear on this, because it is a fantastic way to georeference.

Also, as you generate points, they show up in a table below the raster.  You can modify the x and y coordinate for each of those points as necessary.  That is a true time saver.  Finally, you can save your points to a file for future use.

Another great feature with the GDAL georeference plugin is that it can generate a script for you.  I can see this being incredibly useful if you need to batch georeference maps from a class.  Of course, the problem of reliable scans with minimal pixel deviations may ruin such plans.

Sunday, November 14, 2010

Searching for Faults

This is a geology blog, get your head out of the drama gutter.

The trip to Carrizo Plain, as I shall discuss in depth at a later time, consisted of surface mapping (quaternary  stuff) and finding faults alongs the San Andreas.

It was fun.  Especially as it was the first time the professor had ever been out there with students.  So we all got to learn things like: just because there's a road does not mean there isn't a fence blocking the way :-)

And of course we learned about geological stuff and puzzled out offsets of ridges and drainages to help build a case for the presence of faults.

More later...

Friday, November 12, 2010

Carrizo Plain

Tomorrow, November 13, I will be heading out with my structural geology I class to Carrizo Plain National Monument.

I have no idea what the agenda is, but you can bet the San Andreas 1857 scarp will be involved.  There examples of pressure ridges, sag ponds, beheaded and displaced channels, anticlines, synclines, marine sediments, fluvial sediments, and so much more.  

I'll have my Canon G11 and a GPS receiver so that I can spatially correlate images by time.  

Since I have no idea what the agenda is, I made two simple maps: north and south Carrizo Plain.  They use a mosaic of the Taft and Cayuma 30x60 minute USGS quads and have a graticule in NAD83 UTM 11N (thanks to QGIS 1.6 Trunk).  I reprojected the DRGs into NAD 83 UTM 11N so that the graticule will be compatible with my GPS.  The great news about Zone 11N is that the west standard line is so very close to our field location.  I can take comfort in *very* minimal scale distortion :-)  The declination, by the way, on November 13 at the field area, will be 13.13 degrees.

If I had more time, I'd find my 8 Gb micro-SD card (probably impossible) and load OpenStreetMap data along with for the area contours.

Anyway, it should be a nice weekend if the weather cooperates.  We'll be learning how to map structures.  I hope that the agenda is such that I won't be sprinting traverses...

A report will follow this overnight trip.

By the way, I got to sit in on a career retrospective brown-bag by Dr. Eldridge Moores today.  

Thursday, November 11, 2010

Lyx 2.0.0 Beta 1 Released

Everyone that knows LaTeX loves it; mostly for it taking care of document formatting (letting you concentrate on writing) and producing beautifully typeset equations.

Once you've learned LaTeX, it's good to go on to the next step.  That would be Lyx.  Lyx is a graphical interface to LaTeX, and frankly, you have to use it to understand how much of a time saver it is.   Writing LaTeX documents / research papers will never be the same :-)

Those with Macs can now fetch Lyx 2.0 Beta 1.   See the feature list.  Windows binaries are sure to follow. 

My favorite feature of Lyx 2.0 Beta 1?  Spell checking on Macs works out of the box!

Wednesday, November 10, 2010

Blender and Geology

Anyone out there ever use Blender to help solve or visualize geologic problems?  I suspect it would be handy in structural geology.  It is scriptable: Python!

I once tried to make a fence diagram using processed seismic imagery from the NRPA (National Petroleum Reserve, Alaska) in Blender, but given the lack of time I had, I failed miserably.  The idea was to see how strata may correlate between shot lines.  This was for a Basin Analysis class.

Eventually I gave up and did it the old fashioned way.  Results from that were apparently pretty good!

Drawing Mohr Circles

This may seem obvious to many Structural Geologists, but I rather enjoyed usng OS X's Grapher to graph a parametric equation of the Mohr Circle.  I also plotted fracture planes and the Mohr-Coloumb Criterion at all of the appropriate angles.  It was my first go of it, so perhaps I missed something important.  But it looks really good.

Monday, November 8, 2010

Open Stratigraphy Log

While I took sedimentology, I created a stratigraphic log using Inkscape.

I forgot about it until now.  It is based on the Nichols (2009) stratigraphy log, but I added a column meant for recording GPS waypoints, photo numbers, and other notes.   This form is field tested.

Download the Stratigraphy Log as a (PDF) (SVG -- for editing the log).

Use and misuse however you like, but please try to keep the attributions somewhere on the log. :-)

It look a lot of time to get this log right and so I really hope it is useful to anyone recording section in the field.  Let me know!

Note: I think the log is different enough from Nichols' that I'm releasing it under Creative Commons Attribute-By. So feel free to modify for your own use! I will revise it on occasion but this is a fairly solid product (though the form does intrude into the right-hand 1" margin, but most printers can handle that!)

Thursday, November 4, 2010

Quick Python Script: Cross Section Topography

I wanted to see if it were possible to take a bunch of map measurements and create an SVG file with topographic cross section.  Turns out it is easy to do, once you find the right Python module.

I installed svg.charts, piped my measurements out to a CSV file (x,y) with the first line as a header, and generated a plot of the topography using my script.  I then loaded the topography into Inkscape, trashed the extraneous vectors (e.g., symbology, axes, legend), and scaled to the on-the-map cross section length and appropriately scaled height.  Remember that cross sections must always have the vertical scale be the same as the horizontal scale.

I could continue on and finish the cross section in Inkscape, except I doubt my TAs would approve.  So I printed it out and traced over the topography.  It matched up with the cross section line perfectly.  Hopefully the next time I'll be more efficient at this.

Although, I was considering that I could have avoided hand measurements by scanning the map and doing the following in a GIS:

  • Plot a point on top of each index contour on the cross section line
  • Give each point the appropriate elevation attribute
  • Generate a distance vs. elevation plot using a QGIS script
  • Use my script and scale appropriately using Inkscape
I won't post the script unless I'm asked.  It's really trivial, though.  But perhaps gnuplot would have been a little faster :-)