This forum is shutting down! Please post new discussions at community.naturalcapitalproject.org

RIOS - Pre-processing Tool - Buffer Error

edited September 2014 in RIOS
Hello NatCap fans,

Using RIOS 1.0.0b10 x86,
I'm running into an issue with the Pre-processing Tool.
In the Log file, below at bottom, you can see that the issue is
an unknown error in the Buffer function.

I checked the Stream Shapefiles that the PY program
generates prior to this buffer and that it uses as the
input stream layer for the buffer function.
This is what it looks like at full extent in ArcCatalog:

image

Then zooming in, it looks like this:

image

And even more detailed:

image

So, as you can see, the problem is WAY too much complexity in the stream
pattern, which gets generated in a Raster To Polyline conversion earlier in the Python script.
Artifacts show up, that put a stream around each pixel!

Prior to THAT, a “streams_ras” raster gets created as a derivative from the Flow Accumulation Raster.

THAT one is a direct descendent from the DEM, in our case, the USGS NED 10m pixel dataset.
We had the same error occur with a LiDAR-derived 3m pixel DEM.
Thinking that a coarser DEM would solve the issue, we switched to the NED 10m DEM.

The Buffer function just chokes on the stream pattern that gets created, and hence bombs out.

Anyone have similar problems?
Anyone suggest a work-around or solution??

Many thanks in advance.


============
LOG FILE FOLLOWS:

Executing: RIOS-Pre-Processing true true true true # G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\PreprocessToolCreated_20140923 G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\NASS_NLCD_2012_2006.gdb\lu_coeff G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\Tables\Iowa_LULC_Classification_FromJoshGoldstein_20140911.csv G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\Tables\biophys_table_RIOS_v3_FromJoshGoldstein_20140919.csv G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\NED_10m_01\NED10m.gdb\NED10m_01mtnr G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\LiDEMandDerivatives_20140722.gdb\R_Factor G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\LiDEMandDerivatives_20140722.gdb\Erodibility_01clip G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\LiDEMandDerivatives_20140722.gdb\Soil_Depth_MM_01clip G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\Precipitation\PRISM_ppt_WettestJune_30yr_normal_800m2_02clip.bil G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\IA_Soils\Iowa.gdb\SoilTextureIndex01ftr G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\Precipitation\PRISM_ppt_30yr_normal_800m2_annual_02prj.bil G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\EvapTrans_cm_1971-2000\AnnualEvapTrans_MM_02malg.img 124 60 G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\VectorData.gdb\HUC10_StudyArea20140722_01dslv Test20140923
Start Time: Tue Sep 23 13:51:09 2014
Running script RIOS-Pre-Processing...

Validating arguments...

Erosion Control selected, checking required inputs:
DEM
Erosivity
Erodibility
Soil depth
Land use/land cover
Land use/land cover general class table
RIOS general coefficient table
Threshold flow accumulation
Riparian buffer distance

Phosphorus Retention selected, checking required inputs:
DEM
Erosivity
Erodibility
Soil depth
Land use/land cover
Land use/land cover general class table
RIOS general coefficient table
Threshold flow accumulation
Riparian buffer distance

Nitrogen Retention selected, checking required inputs:
DEM
Soil depth
Land use/land cover
Land use/land cover general class table
RIOS general coefficient table
Threshold flow accumulation
Riparian buffer distance

Flood Mitigation selected, checking required inputs:
DEM
Precipitation depth for wettest month
Soil texture
Land use/land cover
Land use/land cover general class table
RIOS general coefficient table
Threshold flow accumulation
Riparian buffer distance

Checking input raster projections...

Creating hydrology layers...


Processing Erosion Control objective...

Mapping coefficients to landcover...

Creating downslope retention index...

Defining flow direction channels...

Created Erosion downslope retention index: G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\PreprocessToolCreated_20140923\Output\erosion_downslope_retention_index_Test20140923.tif

Creating upslope source...

Created Erosion upslope source: G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\PreprocessToolCreated_20140923\Output\erosion_upslope_source_Test20140923.tif

Creating riparian continuity index...

Created stream layer G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\PreprocessToolCreated_20140923\Output\streams_124_Test20140923.shp

Error creating riparian continuity index: ERROR 999999: Error executing function.
An error occurred during the buffer operation
Failed to execute (Buffer).


Error processing Erosion Control objective: ERROR 999999: Error executing function.
An error occurred during the buffer operation
Failed to execute (Buffer).


Error running script

Traceback (most recent call last):
File "G:\Data\Projects\Iowa_CedarRiver_IA\RIOS\Tools\RIOS_Pre_Processing.py", line 1269, in
raise Exception
Exception

Failed to execute (RIOS-Pre-Processing).
Failed at Tue Sep 23 13:57:01 2014 (Elapsed Time: 5 minutes 52 seconds)

Comments

  • swolnyswolny Member, NatCap Staff
    Howdy -

    In your first zoomed-in stream network, I have seen a similar large-scale grid pattern when I reproject the original DEM using Nearest Neighbor interpolation instead of Bilinear. Does it most every time.

    In the second zoomed-in image, I agree that the generated stream network isn't what you'd expect, but that's how ArcGIS does it, I've seen it many times before. All the preprocessing tool does is run Arc's Flow Accumulation, then selects out the resulting grid cells that are >= your input Threshold Flow Accumulation value. You could try doing a Fill on your DEM, or smoothing it out slightly with a Neighborhood Mean function and see if that helps, but I wouldn't be surprised if these artifacts still appear. Some day, the preprocessing tool will be rewritten using the other InVEST flow algorithms, which should be an improvement. Sorry I can't be more help here.

    Now for the error message about Buffer - I wish I could get better error messages out of the geoprocessor. A few possibilities come to mind: First, do you have ArcInfo? It's the only version that has the particular Buffer options that are required. The script should be checking for this, and letting you know that you have the wrong version, but maybe it didn't for some reason.

    The next idea is that you might be experiencing what seems to be a bug in Arc that has happened to me (with ArcGIS 10.2) - it just bombs on running the Buffer tool, usually just on one side of the stream network, even if I do the Buffer command manually, using the ArcToolbox Buffer tool (which is what the script is using.) To see if this is the case, try manually running Analysis->Proximity->Buffer twice, once with "Side Type" LEFT and the second time with Side Type = RIGHT, using the pre-processing tool output file Output/streams_.shp and see if they both work.

    ~ Stacie

Sign In or Register to comment.