My goal was to show how the river’s sinuosity and path changed over time. I could see that the river eked out meanders and abandoned them when they became too twisty, and varied it’s course quite a lot over the years. To understand how the river changed, I traced the paths of the river while viewing one year’s imagery at a time (in one shapefile, to keep things simple). Then I added the year info for each line to the file in the Attribute Table. I made them different colors too. Then to calculate sinuosity I downloaded a toolbox file from ArcGIS’s website that ran a python script. Unfortunately, someone at ArcGIS messed up and flipped the equation for finding sinuosity, which yielded sinuosity values under 1 (they should be 1 at the very lowest. A sinuosity value of 1 would indicate a perfectly straight river). To fix this, I opened the script in a Python viewer program, and simply changed around the part that described which thing to divide by. I saved the newly edited script and opened it in ArcGIS. I applied it to my river channel shapefile, which calculated the sinuosity for each year’s path. Turns out, the river was most sinuous in 1940, and least sinuous in 2002. I would have expected it to be most sinuous in 1867 and least in 2009, but I’m guess that a few factors are at play here:
- I didn’t trace the pathways accurately. a. Low quality of imagery in some cases (trees blocking for example) could have prevented me from capturing the real path of the river. b. I did not trace the parts of the river where it branched off, so my sinuosity is probably an underestimate.
- I don’t know what month the imagery is from, so it could be the dry season in one year and the wet season in another, which would make comparison less viable.
- Human intervention (damming) could have altered the river’s behavior from what would be the expected decreasing sinuosity over time as the river seeks the most efficient course to Puget Sound.
- A flood in intervening years could have changed things dramatically