topic Re: Flow Accumulation issues in Water Resources Questions

Flow Accumulation issues

LornaMurison — Fri, 12 Nov 2010 12:37:50 GMT

Hi everyone,
I am trying to create flow accumulation grids from flow direction grids that I downloaded. The flow direction grids can cover quite a large area. For some of them, the flow accumulation grid created has a pixel type of signed integer, and for others it is floating point. Whenever I try to run stream definition on a floating point grid I get an "overflow" error. From what I've read, floating point occurs when values are too great to fit into an integer, so I tried cutting my flow direction grid into smaller pieces and re-computing the flow accumulation. I'm also making sure that all areas of the grid aren't draining into one single point. But I am still getting a floating point grid as a result. I'm going to try doing this on an even smaller area, but I thought I would post about it on here and see if there might be another solution I am missing.
So has anyone else had this issue?
Since every time I compute a new flow accumulation grid it takes at least 3 days, this 'trouble-shooting' is becoming very time consuming.

Thanks!

Re: Flow Accumulation issues

LornaMurison — Fri, 12 Nov 2010 18:57:12 GMT

Update:
I have now created a much smaller flow direction grid and the flow accumulation command still creates a grid with a pixel type of floating point.
The maximum flow accumulation value of the smaller grid is actually much larger than that of the larger grid. There must be something else causing this, other than the actual flow accumulation values.

Re: Flow Accumulation issues

LornaMurison — Tue, 16 Nov 2010 12:15:42 GMT

Apparently it's just as simple as using raster calculator to convert the floating point raster to integer.