I made Avalanche() work in parallel, which I missed before. I parallelized this routine in a manner consistent with surface and subsurface routing. Because of what I consider logic errors in the original, my parallel version may not produce spatial Snow[][].Swq identical to the original, but the aggregate Swq seems identical in the tests I ran.
I am concerned that Snow[][].PackWater and Snow[][].SurfWater could be left with values that are inconsistent with redistributed Snow[][].Swq. Further testing may be needed.
I made
Avalanche()
work in parallel, which I missed before. I parallelized this routine in a manner consistent with surface and subsurface routing. Because of what I consider logic errors in the original, my parallel version may not produce spatialSnow[][].Swq
identical to the original, but the aggregate Swq seems identical in the tests I ran. I am concerned thatSnow[][].PackWater
andSnow[][].SurfWater
could be left with values that are inconsistent with redistributedSnow[][].Swq
. Further testing may be needed.