spurious connections

[aestrad-intera]

sfrmaker.Lines.from_shapefile() followed by lines.to_sfr() results in many spurious connections (as in Figure 5 of Leaf et al., 2021 in Groundwater).

After programmatically removing those spurious cells from sfrdata.reach_data and running sfrdata._reset_routing(), the diagnostics indicate segments with non-adjacent reaches and nodes with overlapping conductances. What is, then, the correct way to avoid the generation of spurious connections whne using sfrmaker.Lines.from_shapefile()?

[aleaf]

Hi @aestrad-intera, sorry for the slow reply. The diagnostics can be overly picky. Have you run the SFRData.write_shapefiles() command and looked at the layers generated? (especially *_sfr_routing.shp, *_sfr_lines.shp, and *_sfr_cells.shp) If the connections shown in the *_sfr_routing.shp file look good, then you're probably OK. *_sfr_lines.shp shows a representation of the SFR package mapped on the fragments of the original flowlines that are the basis for each reach; *_sfr_cells.shp shows the reaches represented as model cells. To avoid having more than one reach in a cell (common when streams come together within a model cell), you can lines.to_sfr() with the one_reach_per_cell=True argument.