NDR - Values for crit_len out of grasslands/forests ?
I am using NDR model, which requires input for "crit_len" parameter (length after which the maximum retention efficiency "eff" is reached).
This parameter is relatively easy to assess for grassland and forest (which are commonly used for riparian retention), eg. Zhang et al. 2010. But it is far more complicated to evaluate for fields, and urban areas. Intuitively, one would input higher values for crit_len on these land uses, since they don't retain nutrients as well as grasslands. However, InVEST online documentation suggests to set it at the pixel's size - which, in my model, is smaller (5m) than grassland's crit_len (10-15m).
So, should I keep this odd parameterization ? Or should I rather take a fictive, high crit_len (such as 300m, which is a maximum according to online documentation) ?
Many thanks in advance,