Ultimately, I'm trying to generate catchment polygons and stream lines for about 25 small watersheds that are linked to each other in attributes, where the stream segments are linked to each other with a nextDown attribute. I've gone through the basic process in ArcHydro and created this, but I've just realized that I need to calibrate results to stream gauge stations which are not located at the outflow of the watershed or a stream junction. Therefore, I'm trying to figure out how to split a generated catchment at the gaging point (which I did by adding a pour point at the station and calculating the sub watershed and then using Union with the original catchments) but I'm now stuck at trying to regenerate the linking attributes to properly account for the new catchment. I've assigned HydroIDs to the new features (split catchment and split stream segment) with the command from Attribute Tools, but when I try to run the Find Next Downstream Line it runs properly but ignores the inserts (I assume because they haven't been incorporated into the underlying node network). Is there any way to create new nodes to include these new features, or am I approaching it totally wrong and there's some other (earlier in the process) way to do this? Any help would be greatly appreciated, and I can provide more details/screenshots if needed.
... View more
I'm learning ModelBuilder and appreciating how it simplifies life, but I've spent most of today swearing at it because I'm having all kinds of issues trying to deal with FieldMaps. I have a continuous layer of polygons (no gaps) that I'm trying to attribute. Some I can just calculate (based on input data), but in some places, the input data is unavailable, so I want to copy the value from neighbouring polygons of the same type. To do this, I buffer the polygons, intersect the buffers with the source layer, filter out overlaps that are not the right type, summarize the buffer intersections by polygon ID to give MAX area for each source (to find the neighbour with the longest shared side - had issues with the new Polygon Neighbours tool not identifying neighbours) and then selecting out buffers where the area = MAX_Area. In the end, I get a map table with src_ID, nor_ID, src_attrA, nbr_attrA, src_attrB, nor_attrB which i can join to the original dataset based on src_ID and calculate src_attrB = nbr_attrB. To try and simplify things and keep track of which attributes are from Source vs. Neighbour, I've want to rename fields at various points, which I've been able to do using Feature Class to Feature Class and setting up a field map. It worked OK while I was building it and had it linked to a specific source file, but now I want to make the input dataset a parameter (so I can run it on several different files) and it seems like the field maps have to be specified by absolute reference (i.e. C:\Test\db.gdb\Input_FC, field_1) and when I parameterize the input FC (and select C:\Test\db.gdb\FC_1 for example) it says it can't find the expected field - even though the schema is identical. Is there any way to generalize field mappings in model builder, or do I have to resort to python (which I don't know and don't have time to learn at the moment). I can export the model as a python script so - if it's a small code block I could just dump that in and connect it with the rest of the generated code. I've searched all over the net and haven't found anything like this - lots of help about model builder (lacking details about generalizing field maps) and lots of info about details of dealing with field maps in python, but nothing in between. Any thoughts?
... View more