POST
|
Hi, I have a population dataset raster I created from a point to raster conversion. I need to run some analysis on it and need a raster attribute table. I would also like to keep the accuracy of my data, as it contains decimals. I understand that I cannot create a raster attribute table with a 32 bit floating point raster. I tried a work around by multiplying by 100 to get integers (while maintaining my true numbers), however, I keep on getting errors in my raster calculator: File "<string>", line 8, in rcexec AttributeError: 'str' object has no attribute 'save' Failed to execute (RasterCalculator). This has been a wall. Any help with building a raster attribute table with decimals and/or fixing this error in raster calculator? Note: I cannot simply use the integer tool as I lose too much information. Thanks, Phil (Not a primary GIS person)
... View more
06-12-2015
02:25 PM
|
0
|
1
|
3908
|
POST
|
Philip, What size are your pixels with respect to your zonal data? Are you trying to compare for example a 1Km grid with a polygon that are a few hundred meters wide? I'm guessing some sort of re-sampling error is occurring. Ideally you want to ensure that your zonal data has the same cell size as your raster. May be you should create a raster version of this first and make sure its correct. You'll also want to ensure the snap to raster environment setting has been set. Duncan Thanks Duncan. The raster Land Cover dataset has a 250 m grid. The polygon zones (census blocks) vary wildly, from less than 1 m2 to 30,000,000,000 m2. I'll try creating a raster and report back. Thanks again for any help, PC
... View more
02-25-2014
10:04 AM
|
0
|
0
|
574
|
POST
|
This appears to be a seemingly easy calculation. But I am having problems. Question: What is the percentage of a certain land cover class in each census block? Data: 1) Census polygon 2) Land cover raster Method: Tabulate Area tool - with census polygon as my zone and land cover as the class of interest. Problem: My output gives me areas much BIGGER than my census polygon (i.e. Percent land cover class is often greater than 100%...sometimes in the 1000s%). These numbers are obviously incorrect, as area of vegetation can't be greater than the area of the census block (zone). Both have linear units of meters and are projected. What am I doing incorrectly? Are there better ways to calculate area of raster in polygons? Thanks in advance, PC
... View more
02-24-2014
01:27 PM
|
0
|
2
|
3584
|
Online Status |
Offline
|
Date Last Visited |
11-11-2020
02:23 AM
|