DEP_IMP estimate depth of impervious layer

[Timo Brussee]

Hey All,

I was wondering what a usual value is for DEP_IMP ( the depth to the impervious layer)?

It is set as a default to the maximum of 6000mm in ArcSWAT 2012, when I change it to zero as recommended in the ArcSWAT manual in situation where there are no perched water tables in the HRU, my discharge is halved. (from 21% of precip to 11% of precipitation).

My watershed is located at the boarder of Kenya/Tanzania (its called Mara river basin) and I have no idea about the existence of perched water tables there.

In the input/out documentation of ArcSWAT 2012 DEP_IMP is described as the Depth to impervious layer in soil profile (mm):

" Perched water tables are created when water percolating through the soil profile reaches a layer of low hydraulic conductivity that causes water to pond at the upper boundary of the impervous layer. This variable defines the depth to the impervious layer in the soil profile and is required if perched water tables, depressional

storage areas/potholes, or tile drainage is being modeled in the HRU (or subbasin for depressional storage areas).

If perched water tables do not occur in the HRU leave this variable set to 0.
If a generic depth is defined using DEPIMP_BSN (.bsn), set DEP_IMP = 0 to use the basin- level value."

According to the ArcSWAT theoretical documentation version 2009 the Dep_Imp affects the height of the water table by:

" hwtbl = depth_imp * (SW - FC) / ((POR - FC)*(1-airfilledporosity)) "