1

I have a raster DEM of 90m resolution (srtm90). I have a fishnet(polygon) for the same region of size 25*25 sq. km. For each grid in fish net, I want to store all the pixel values of raster within the same lat-lon range of fishnet grid box. The ultimate aim is to create elevation bands of different range say 0-1000m, 1000-1500m, 1500-2000m and so on with their respective areas inside each fishnet grid box.

The following picture states the problem with a table asking for each elevation pixel with its respective area. Though area can be same for each pixel in a DEM, all pixel values are required in each grid box. These area are later used to determine the area fraction of each elevation bands inside a grid.

Raster to polygon followed by intersection with fishnet was my initial attempt but the file was so big that raster to polygon conversion was never going to stop. I dont think this method is appropriate. Hence I am bound to question here.

Arcgis is preferred.

enter image description here

4

The ultimate aim is to create elevation bands of different range say 0-1000m, 1000-1500m, 1500-2000m and so on with their respective areas inside each fishnet grid box.

When you're doing analysis like this, it's much faster to keep your data in raster format if at all possible. One way to get this done would be to Reclassify your DEM first, creating a new raster with a single value for each of your elevation ranges.

Then, to calculate the areas of each elevation band in each fishnet grid cell, run Tabulate Area, using your fishnet layer as the Zone and the reclassified DEM as the class. Both tools require the Spatial Analyst extension.

  • This worked fine. Thanks to @DanC for great suggestion. But is there a way to find the mean value of each elevation band inside each fishnet grid box? – NirAmar Oct 6 '16 at 5:37
  • @NirAmar Yes, but that's a different question. You'd use Zonal Statistics on your original DEM (not the reclassified one) for that. – Dan C Oct 6 '16 at 12:11
  • Dear @DanC. The Zonal Statistics gave only the mean elevation of a single grid. What I want is the mean elevation of each elevation bands created after reclassification inside a grid. For eg. If a grid has three classes of elevation bands, I want mean values of each of the class inside a grid. Hence I will have three mean values for that grid at the end. A single elevation band of class, say 0-500, might have numerous values in that class in my original DEM. I need to get a mean of all those values that falls within this range for each of the grids. – NirAmar Oct 11 '16 at 4:45
  • @NirAmar You should ask a new question (after searching the existing ones for a possible answer). Zonal statistics is not really related to your original question. – Dan C Oct 11 '16 at 12:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.