1) I set up pre-development hydrographs using the Rational Method, using one large watershed (about 25 acres) to model the pre-development runoff. The max Q came out to 36.31 CFS for the 100 year storm with a 17.44 min Tc.
2) For post-development, I divided the pre-development watershed into routed and bypass sub-areas. I set up my post-development hydrographs and modeled it without the basin routing to quantify how much the runoff would increase as a result of the added impervious surface area.
Here's where I run into a problem. The combined post-development hydrographs (unrouted) resulted in a REDUCTION in the peak runoff, rather than an increase. For the 100-year storm, the max Q was 33.73 CFS with a 17 min. Tc. It seems pretty clear what the problem is. Simply put, the peaks of the sub-watersheds do not line up anymore. The increased runoff from the developed area occurs at Tc=5 min, which is well before the Tc of the much larger bypass watershed, which occurs at 17 min. The developed runoff is already off-site by the time the bypass runoff peaks.
Has anyone dealt with a similar problem before? I don't think it would fly if my SWM report indicated that the development resulted in a decrease in runoff. My instinct is to manually add the peak flows of the individual watersheds. It's over-conservative and it does not accurately model the actual conditions. However, I cannot think of a better solution. Suggestions anyone? Thanks.