Any one who had made flow direction and flow accumulation on raster data for filled evaluation model i Qgis? Which tool and plugins are necessary, i haven't been able to find any guidelines on Google
Your best bet is probably to use the SEXTANTE plugin.
Go: SEXTANTE->Geoalgorithms->Domain Specific->Hydrology
Here you will find a number of tools roughly analogous with some of the hydrology tools in ArcGIS. This link on Hydrological Analysis and this one on Hydrology analysis with TauDEM from the Sextante Cookbook may prove useful.
I forgot to mention 'Fill Sinks' which is in a different part of SEXTANTE. It comes in 3 'flavours' and the 'Fill Sinks (Wang & Lui)' version will also output flow direction and watershed basins.
Go: SEXTANTE->Geoalgorithms->Domain Specific->Terrain Analysis and Geomorphology->Fill Sinks.
I have been able to obtain flow direction & flow accumulation but not with QGIS processing toolbox. I used Grass pluggin - and then since I loaded OSGeo4W - I repeated in GRASS 6.4.4. Process - establish NewMapset - use Open Grass Tools - make sure DEM is in .asc format - that is the best image type that I got to work(I opened QGIS - Raster - conversion- translate) - r.fill.dir - (other formats never finished running this fill step) - r.watershed (this will give you flow direction & flow accumulation...I was told you could use the GRASS conmmands in the QGIS processing toolbox - however I was not able to get them to 'work' they produced outputs - but the output files did not match any other software outputs I had tested (ArcGIS - was one)....
See the image - the one on the left is from GRASS pluggin & the one on the right is from QGIS processing toolbar - to obtain stream order- I was not able to get a good/comparable output from the Processing Toolbox.... I am new and figuring this out - so watch you CELL SIZE - after editing my cell (SET REGION in GRASS) to the correct 30 m (.0027777778) cell size - this process takes a long time on my 1GIG sample area
4 years later - if anyone stumbles across this while using QGIS 3.4, I found the simplest/most-robust solution was simply use the GRASS
r.flow tool. The only real input is your DEM, and it outputs flow accumulation and associated data layers without having to do multiple steps. (for the gui: processing --> toolbox --> GRASS --> Raster --> r.flow).
Make sure you are using a CRS that uses meters, not lat/long. You may need to reproject your DEM.
From there you can recategorize the flow accumulation to retain only zeroes, if you are looking for ridgelines, predict stream locations, etc. Both results (stream and ridge) matched pretty well our empirical data, using a NASA ASTGTM2 DEM.