Just a cautionary tale:
If the lowermost SOL_K = 0 in some of your HRUs (as might be presumed to be the case when bedrock is the lowest soil layer), that HRU can produce huge amounts of sediment during wet periods. Setting SOL_K to almost anything >0 (like, 1 or 2...) will greatly reduce overland runoff and soil erosion. I would argue that most shallow bedrock has enough fractures to have at least some finite effective permeability.
I found this out when my current model results were telling me that deciduous forest was the dominant source of sediment, in a landscape dominated by agriculture, which is crazy. I narrowed it down to one soil type, in one steep subbasin, where annual sediment yields from forest were over 100 t/ha (10-year averages). I changed the lowermost SOL_K value to match similar shallow soils over bedrock, and the results settled down to something more reasonable. Sediment yields are still significant -- it is shallow soil over bedrock -- but not crazy-high.
Cheers,
-- Jim
--
Director, St. Croix Watershed Research Station