2

two rasters combine to one

So I began with lidar elevation data 1m resolution, and I performed a terrain analysis to extract specific ranges of elevation from the raster (1st picture is 0-1m elevation, 2nd picture is 1-2m elevation). These extractions resulted in 2 different rasters, but for further analyses I would like to combine them back into one raster but now with just these two elevation size classes (I want the third picture but as a new raster I can process further). Mosaic is not working because the two new rasters are the same spatial extent. I have both arcGIS 10 and LIDAR Analyst.

  • 1
    What exactly are you trying to accomplish? Are you trying to assign both raster values to each individual cell? Give us an idea of what you're looking to accomplish in the end so we can provide an idea. – Branco Jun 25 '14 at 19:11
  • I am trying to produce a new raster that contains both elevation ranges, there would be no overlap, the new raster has a single value per cell. the BIGGER situation is that I have about 420 rasters covering the long island seashore, and I tried mosaic-ing all of them together and then reclassifying the mosaic into new elevation classes but that did not work. I need 5 classes of elevations that I am then going to reclassify again for a weighted overlay model. I am trying to find a way to be able to batch process all these rasters – tugboat789 Jun 25 '14 at 20:00
2

There is a tool in the open-source GIS Whitebox Geospatial Analysis Tools called Max Overlay, which will take an arbitrary number of overlapping rasters and output a raster in which each grid cell is assigned the maximum value from the input stack. It sounds like it will do what you're looking for. There is likely a similar tool in other GIS as well (e.g. SAGA, QGIS, etc.), I just don't know their names.

2

This can be done with conditional operators (Con) in the spatial analyst. I am assuming there is no overlap in the cells

Con("Elev1" > 0, "Elev1", "Elev2")
  • I am not sure this will work because I have 5 rasters, each with only 1 elevation class (0-1, 1-2, 2-3, 3-4, 4-5), and these 5 rasters all represent the same exact extent on a map, just different elevation ranges. I need a new single raster containing all these elevation classes, that I can then Reclassify for a weighted overlay. (this same process has to be repeated for 420 rasters covering all of Long Island) – tugboat789 Jun 25 '14 at 20:22
0

You cannot create a single band raster dataset with multiple classes of elevation values contained within each pixel. The closest you could come to that is by creating a multiband raster using the Composite Bands tool. However, this would be a very unusual approach to your analysis. Based on your comments, there is no reason why you cannot add multiple single-band rasters in a weighted overlay analysis.

  • Unfortunately I am getting back some wonky results when I do that where is merges together the two separate elevation layers into one value instead of keeping them two separate values (values being the separate elevation classes 0-1 and 1-2) – tugboat789 Jun 25 '14 at 20:05
  • visually that produces what I would like, but for analysis I cannot reclassify this raster. So What I want is a single raster that has 5 elevation classes. I initially tried this just using reclassify from the original elevation raster BUT in order to batch process 420 rasters, I cannot do this because the Reclassify tool will not iterate through the rasters because each individual raster has a different range of elevation values. I thought mosaic would eliminate this problem, but I cannot reclassify a mosaic dataset – tugboat789 Jun 25 '14 at 20:30

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.