Thank you for the question. To answer this, you will have to go where no man has gone before.
There are a number of concerns. For example, cdom requires uv (usually down to 200 nm). If you decide to proceed, try the calibration procedure to make sure your diy spectrometer is aligned and is showing good sensitivity at 232 nm. Mercury line from a cfl.
I don't know if you are going to use fluorescent from or absorbance. Whichever, your light source will be a problem. It will need to be moderately high intensity.
And then there is software. Don't know about that.
If you are planning on immersing the spectrometer as a probe, you will have a number of changes to make to the instrument.
Please let us know how the project goes. Good luck. Live long and prosper.