The software uses 100-meter cells and runs on discharge instead of area, as seen in my last model run.

I have been looking at scale dependence. I have done the analysis at

The following table shows the intercept and slope for various basins in the final timestep. (I have omitted the basins that changed rank from one scale to another.) The results seem to fairly consistent for the larger basins, but wander more for the smaller basins. Note also that the path of the main channel can vary with scale or from one timestep to another.

scale intercept slope
basin 1
 500 0.990      -0.338
 100 1.022      -0.341
1000 0.946      -0.332
basin 6
 500 0.783      -0.313
 100 0.766      -0.312
1000 0.555      -0.283
basin 7
 500 0.744      -0.305
 100 0.940      -0.331
1000 0.470      -0.270
basin 8
 500 0.297      -0.236
 100 0.237      -0.230
1000 0.399      -0.248
basin 9
 500 0.704      -0.292
 100 0.543      -0.274
1000 0.599      -0.279
basin 10
 500 0.413      -0.248
 100 0.713      -0.286
1000 0.447      -0.250
basin 11
 500 0.538      -0.263
 100 0.893      -0.311
1000 0.605      -0.270
basin 12
 500 0.880      -0.320
 100 0.931      -0.327
1000 0.512      -0.275
basin 13
 500 0.505      -0.274
 100 0.483      -0.271
1000 0.318      -0.249
basin 14
 500 0.601      -0.269
 100 0.513      -0.258
1000 0.492      -0.252
basin 19
 500 0.378      -0.257
 100 0.460      -0.270
1000 0.116      -0.225
basin 20
 500 0.641      -0.260
 100 0.683      -0.266
1000 0.184      -0.203
basin 21
 500 0.364      -0.222
 100 0.353      -0.222
1000 0.019      -0.179
basin 22
 500 -0.023     -0.191
 100 0.179      -0.217
1000 -0.165     -0.172
basin 31
 500 -1.214     -0.019
 100 -1.579     0.024
1000 -0.986     -0.047

So, do we want to stick with 500-meter cells of go to 100 meter? Should I do a run at 25m or 50m so see if we are settling down or introducing artifacts with fine scales? Do the results for smaller basins seem reliable enough for analysis? Did you have a nice weekend?