It's true, a trap type contraption is needed to separate

Physical mainly

Chemical and

Gaseous pollutants in rain water.

At 1st, we take filter to separate

fine

medium sized

large sized which do not flow but stay stagnant on stream itself.

It's this property which most hydrologists use to trap the matter .

.. to be continued..

Yeah, it makes sense to me that for something like sediment, a stormwater BMP might have a consistent concentration in the effluent. Say, a large bio-retention area collecting runoff from a farm field. (Think, something like a detention basin or a rain garden). During low-flow conditions, most of the sediment is suspended in the water column well; say, it's largely colloidal. And this passes through the BMP with the water. During high-flow conditions, a lot of the sediment is only temporarily suspended; say, sand; and the overall sediment concentration in the inflow is much higher. Most of this is captured by the BMP, but the colloidal stuff, that would have been there during low-flow conditions anyway, continues to outflow with the water.

a) This effect, of course, depends a lot upon what kind of BMP system we're talking about.

b) This seems less likely for dissolved constituents like nitrate. Again, depending on the BMP being considered.

Dear Greg Williams et al.,

There is no such thing as "irreducible concentration" for stormwater contaminants. Current stormwater BMP are mainly physical (screening, settling, filtratation, ..) and chemical (coagulation-flocculation). These technologies don't remove soluble pollutants. That's why we also integrate advanced biotreatment with short contact time (15 - 20 minutes) to capture and mineralize most soluble pollutants. Typical results are shown on https://www.modelengineering.eu/circulate_water .