Damon
NOTES
2005_02_04: (2 hours)
Disk Space and Directory of LANDSAT TM data is /home/tm, modis is /home/modis, avhrr is /home/avhrr, aviris is /home/aviris, dems are located at /home/dem. Use the REASoN profiles on yer SSH program.
2005_02_07 (3 hours):
Data regarding snowpillows, stream flow, etc. in California can be found at http://cdec.water.ca.gov/snow/.
2005_02_08 (4hours):
The project goals seem to be
• Snow-covered area and snow albedo at subpixel resolution from MODIS, based on algorithms we have developed and tested with AVIRIS and Landsat.
• Snow-covered area at subpixel resolution from AVHRR, for the area accessible from the UCSB antenna, based on algorithms we have developed and tested with Landsat and AVHRR.
• Snow-water equivalence at coarse, global-scale resolution from passive microwavemeasurements with AMSR-E and AMSR, incorporating co-registered snow-coveredarea measurements from AVHRR or MODIS.
• Historical compilation of consistent satellite measurements of snow-covered area atdrainage-basin scale from CORONA, Landsat MSS, Landsat TM, and EOS instrumentsfor the state of California from 1962-present, along with historical time-series ofsnow-water equivalence measurements at snow courses.
• Historical compilation of consistent satellite measurements of snow-covered area andsnow-water equivalence at global scale, combining optical measurements fromAVHRR and MODIS with passive microwave products from SSM/R, SSM/I, AMSR,and AMSR-E.
A few months ago in november, Dozier asked KArl to : " Pick a scene close to (but not necessarily on) April 1st, and then use the snow course and pillow data to estimate the SWE at the time of that scene’s acquisition."
I figure I should put about 220 hours per quarter into this research. 20 hours per week times 11 weeks per quarter.
2005-02-09:
2005-02-10:
2005-02-11:
2005-02-12:
2005-02-13:
2005-02-14:(1 hour)
2005-02-15:(4 hours) Accumulated the daily SWE values from all the pillow sensors inside our three Landsat TM scenes, off the California Department of Water REsources web site.
2005-02-16:(2 hours) Plotted the data.
2005-02-17: (3 hours) I am now collecting Landsat images for Path 42, Row 35. As close to April 1 as possible. I also should put all the image dates onto the SWE graphs.
2005-02-18:
2005-02-19:
2005-02-20: (5 hours) Today I put the existence & price of images ontop of the SWE time series.
2005-02-21:
2005-02-22:
2005-02-23:
2005-02-24:
2005-02-25:
2005-02-26:
2005-02-27:
2005-02-28:
2005-03-01:(2 hours) Berkeley has some CDs with Landsat ETM images from 1999-2000. http://www.lib.berkeley.edu/EART/cdrom.html
2005-03-02:(5 hours) Searched for landsat images.
2005-03-03:(3 hours) Setup ipw. Walter recommends the following a week ago:
Okay. I talked with JD today, and he's fine with running them now slowly
with a small spectral library, then redoing it later if need with a faster
algorithm and a larger library.
So, step 1 is to get uncorrected images to be processed into IPW format.
Step 2 is to run "tmref" on them to apply a nominal atmospheric correction
for location, elevation, season, atmospheric model etc. Step 3 is the slow
SCA script tmemsca.
2005-03-04:(4 hours) Read up on Walters algorithm
2005-03-05:(5 hours) More on the algorithm
2005-03-06:(5 hours) Input more TM images to our wiki list. NOTE: Landsat images have a resolution of 28.5 meters. I am going to run through Walter's SCA algorithm using the landsat data from 19930513_p042r034. The coordinate system is space oblique mercator. The pixels are 28.5 meters by 28.5 meters, but the orientation of the rows/columns are not parallel with the "space oblique mercator" world-lines. The ORIENTATION number in these files gives the angle formed between the world-lines of the pixels of the landsat image and the world-lines of the SOM coordinate system.
2005-03-07:
2005-03-08: (1 hour) NOTE: We need to verify that our wiki list of TM images http://wiki.snow.ucsb.edu:8888/SnowREASoN/95 matches the excel file of TM images
2005-03-09: (7 hours) Began a document on the "workflow" necessary for processing SCA with Walter Rosenthal's algorithm; available at U:\MyResearch\REASoN_Grant\LandSatTM_Stuff\DetailsOfWaltersAlgorithm
or
~/MyResearch/REASoN_Grant/LandSatTM_Stuff/DetailsOfWaltersAlgorithm/
And I am working on getting one Landsat image (or rather the 7 bands of one image) through Walters Algorithm once. This is all taking place in
/home/dturney/MyResearch/REASoN_Grant/LandSatTM_Stuff/tm/1993/19930513_p042r034/FilesWaltersAlgorithm
2005-03-10:
2005-03-11: (3 hours) Fixed a bug in tmref2.
2005-03-12:
2005-03-13:
2005-03-14:
2005-03-15: (3 hours) began running tmemsca on my first scene - the 1993 scene.
2005-03-16: (1 hour) Prepared the presentation for Snowpack class
2005-03-17:
2005-03-18:
2005-03-19:(20 hours) Helped out during the week of the Snowpack class.
2005-03-20:
2005-03-21:
2005-03-22:
2005-03-23:
2005-03-24:
2005-03-25:
2005-03-26:
2005-03-27:
2005-03-28: (1 hour)
2005-03-29:(10 hours) Began compiling the info necessary for runnig walters algo on all our TM images
2005-03-30:(5 hours) Now I have all the info/metadata needed to run walters algo on all our TM images
2005-03-31:
2005-04-01: (5 hours: total 104) Converted all the tiff images to binary images, using adobe photoshop (saving the images as "raw")
2005-04-02:(5 hours) Finished converting tiffs to binaries, and began to write the batch file for processing walters algorithm
2005-04-03:(4 hours) Now using PBS to process SCA on the cluster.
2005-04-04: (3 hours) tmemsca is now failing. Error messages are: broken pipes,
2005-04-05;
2005-04-06:
2005-04-07:
2005-04-08:(4 hours) got another scene from the library. Began finding the fix to the error messages from above.
2005-04-09:
2005-04-10:
2005-04-11:(9 hours) Finally pieced together the stupid library scene. Still working out problems with PBS and tmemsca.
2005-04-12: (5 hours) Still trying to get PBS qsub to process tmemsca correctly.
2005-04-13:
2005-04-14:
2005-04-15:
2005-04-16: (5 hours) Got PBS working!
2005-04-17:
2005-04-18:
2005-04-19:
2005-04-20:
2005-04-21:
2005-04-22:
2005-04-23:
2005-04-24: (10 hours : total 150) now processing all the data.
2005-04-25:
2005-04-26:
2005-04-27:
2005-04-28:(5 hours) : still messing around w dis shit
2005-04-29:
2005-04-30:
2005-05-01:
2005-05-02:(2 hrs) crap still won't work.
2005-05-03:
2005-05-04:
2005-05-05:
2005-05-06:
2005-05-07:
2005-05-08: (9 hrs) Now have the images in U:\MyResearch\REASoN_Grant\LandSatTM_Stuff\WaltersAlgorithm\Metadata_4_SCA_Processing.xls corrected with tmref2_rev1 and now will begin making SCA images by rlogining into the cluster nodes and running nohup "tmemsca -i image" by hand.
2005-05-09: (2 hrs) Now runnig the process with all input files coming from /tmp This is the cpu's local disk, and this should make the operation much faster.
2005-05-10:
2005-05-11: (5 hrs) Handling the processing.
2005-05-12:
2005-05-13:
2005-05-14:
2005-05-15:(5 hrs) Handling the processing.
2005-05-16:
2005-05-17:
2005-05-18: (3 hrs) Finished her off! phwew
2005-05-19:
2005-05-20:
2005-05-21:
2005-05-22:
2005-05-23:
2005-05-27: (2 hrs) Transfered all the SCA files to /home/tm/products/sca/ , and cleaned up the mess on /home/tm.
2005-06-19: (6 hrs) Redoing some shiznit. Teaching Karl how to redo some shiz..