For the final week of our Maya Pyramids project, we focused on tools that allow us to share our ArcGIS maps on Google Earth. We created .kmz files that captured our potential pyramids sites, some of our map views of the sites, and our final map of potential sites (above).
The graduate students completed a "bonus" section creating similar composite views (NDVI in my case) for Angkor Wat which were then used to train and create a supervised classification. My final classification is above.
The Angkor Wat site seemed fairly "trainable" though I did run into some initial problems. Apparently it is quite cloudy at the site. My first download (very large, as we know) of Landsat images ended up covering the site completely. The second set of images from an earlier date, which is what I used, still had a fair amount of cloud cover, but I was able to work around it.
The site seems to contain a lot of canals and ponds. These are highlighted quite well by a False Color and especially by the NDVI view.The Supervised Classification, however, seems to not be able to recognize the water (and it is strangely classified as "Urban"). However, the linear nature of the canals and various man-made structures is still obvious.
The linear indicators helps highlight the extensiveness of the site beyond the monumental core. The site seems to extend at least as far east as it does west (far left of the large rectangular pond). Similarly, the site seems to extend north and south about the same distance as the height of the central square region. I'm not sure the Supervised Classification provided a better view of the nature of the site as the plain False Color and NDVI images similarly expose the extent of the site.
No comments:
Post a Comment