Sizing of sediment basins

I am trying to find some definative information on the sizing of sediment basins in relationship to contributing watershed/rainfall/sediment load. Having a hard time, anyone have any sources?

My comments are in reference to a sedimentation basin for post- development conditions, not the construction period.

The best method for taking into account local hydrology is to use a continuous model, such as SWMM to derive a series of flows for 5 to 10 years (the longer the better) and then route these flows through a sedimentation routine. I wrote my own sedimentation program to track any number of particle sizes in any weighted distribution. The program accounts for both quiescent and dynamic settling. You can use either literature values for particle size distribution and settling velocities (such as USEPA NURP values) or user specified.

A couple of observations from my experience. Don't treat the results as gold. In my experience the modelling with standard EPA distribution and settling velocities will underestimate the actual performance of a settling basin. The modelling is only sedimentation and other factors can/will impact the results (flow path/dead zones, wind resuspension, flocculation effects (gravity, collision, bacterial), etc.). This reinforces the concept that modelling is no substitution for common sense and good engineering. The modelling is just a rough estimate, but still very useful. IMHO the hydrological

modelling is of greater importance since most sedimentation basins are designed to treat the "first flush". The "first flush" is an imperfect term since it varies based on meteorology (intensity frequency-duration) and location of interest (no first flush seen at the end of a large watershed). A key question to ask yourself as a designer is what percentage of the annual runoff volume am I treating with my sedimentation basin ? You will find a breaking point where the design of a larger basin is just wasting money. Once you satisfy yourself with the amount of water you are treating, then you can turn to the sedimentation aspects of the design.

This may seem an arduous methodology to size a sedimentation basin, and it is. However, IMHO the design of sedimentation basins for post- development conditions using a design event is incorrect. A peak flow design does not tell you what percentage of the annual runoff is treated, nor what kind of long term effectiveness you can expect. One might argue that a basin designed with a 5 year storm peak would be better than one designed for 1/3 year storm. However, the 5 year storm may scour and resuspend sediments out of the basin, be 3 times more expensive to build, be more difficult to maintain, and only treat 5% more annual runoff volume. Not a wise tradeoff for the ease of design.