Check out:
http://www.epa.gov/ednnrmrl/reports/SR00016/cover.pdf
Boyd,Bufill,&Knee (1993). "Pervious and impervious runoff in urban catchments" Hydrological Sciences Journal 38(6) 463-478.
Boyd,Bufill,&Knee (1994) "Predicting pervious and impervious storm runoff from urban drainage basins" HSJ 39(4) 321-332.
Something else to consider is that a modeler will typically setup the SWMM catchment as X percent impervious, and 100-X percent pervious, with the same length (subcatchment width) and slope, yet differing values for depression storage and roughness. If the 'typical' overland flow length over pervious land is much different than that for impervious land, then calibrated infiltration parameters can be quite different than the measured ones. That's of course assuming the infiltration measurements are good, which they often aren't. The above comments apply to determination/measurement vs. calibration of 'effective' or 'connected' impervious area as well.