Fire Probability

 

     This was probably one of the most straightforward and  simple labs we have had this quarter, if everything had gone smoothly.  I was able to get the fuel layer from http://frap.cdf.ca.gov/data/frapgisdata/download.asp?spatialdist=2&rec=fmod, the elevation layer from  http://seamless.usgs.gov/website/seamless/viewer.htm, and the fire zone coverage layer from the the UCLA website for geography 7.  That was the easy the part. What normally would have only taken me a few hours to do the tutorial and then my own analysis ended up taking multiple hours because of all of the problems I encountered.  I first had difficulty converting the elevation map into percent.  I got ridiculously high values that I might have been able to work with but were not in fact correct and may have tripped me up at later points in the project.  To fix this I had to convert the elevation layer from the projection it was in to UTM zone 11 which proved to be the most difficult part of the entire venture. The file may have been to large, but every time I attempted to convert the layer, ArcGIS simply shut down.  This happened multiple times on multiple computer.  I eventually had to reload the file and attempt the entire process over from scratch.  

     Once I was finally able to convert the file and actually start the project, I had difficulty finding the metadata to reclassify the fuel values.  Because the values were not simply given to me as in the tutorial, I also had to more or less guess at what the values should be and hope that the values were correct.  Fortunately i was able to find a separate layer online that had already classified the values, and my own interpretation was fairly similar, giving me confidence that I had adequately valued the fuels.  Once I had the values all accumulated I ran into even more problems in attempting to use the Raster Calculator and determine the ultimate risk of fire.  The first couple of times I attempted to find the values, they did not come out as classified values but were rather discrete, which was not correct and did not give me anything to work with.  After manipulating a few aspects I was finally able to get everything together and finish the lab.  All in all this should not have taken me more than 3 hours to do everything, but because of the multiple obstacles it took more than twice as long as that.

     Ultimately all of the issues probably taught me more about ArcGIS than the actual lab did.  I had to deal with so many issues on my own that I learned two things.  First, how to deal with those specific issues in the future should they arise.  The second, and more important revelation, is that I can in fact figure out some of these problems on my own.  The fact that I do in fact have a bit of a GIS toolbox is reassuring, even if it is still rather basic.  I hope that in the future similar projects will not take quite as long, but if they do at least I can be assured that I will probably learn valuable skills in the process.