The translucent algae might be diffusing the light and making it appear brighter at certain angles. If the light source is already diffuse, this effect of the algae might not be noticeable. Maybe something milky between the light and the dish would provide a more even light so the exact angle of the spectrometer (path from slit to grating) was not as critical. To get a better measurement of which wavelengths are absorbed by the algae, a continuous spectrum light source like a halogen bulb might be better.

Just a few thoughts, Yes, there may well be some "glowing" (refraction) effect of the substance as the light shines through it, but I think that's not the whole concept to consider here.

Consider doing a test with food coloring (no physical structures to refract). With Green coloring, what your eye sees is "mostly" green as the transmission of light is attenuated for all other wavelengths -- i.e. relatively transparent for green. Yes, there is a little attenuation in the Green band but the major impact you can measure is attenuation of other spectra.

Yes, as noted, for such measurements, you do not want the sharp-band spectr of the CFL -- instead you want a very broadband light source -- like a Halogen (Solux is pretty good). That way you are watching a much smoother curve attenuate outside of the green.

That said, yes, I'd agree that there should be some relationship between the attenuation in the green band and the density of the substance. (Concept: If you only look at the green light where the chlorophyll is blocking the light you might get attenuation of green -- however, there might be a lot of refraction obscuring the measurement). I'm not sure which is more measurable -- attenuation in the Green or attenuation outside the green band I suspect the later is more sensitive. If you want to measure in-band attenuation, you could experiment with using a green food coloring reference (instead of clear water) (attenuate all else so as to see just the green) and then add the substance.

The alternate concept is to look at reflected spectr instead of transmitted spectra (albeit that is harder). With just water over a non-reflective black, adding chlorophyll would then add a green spectral band -- in which case you'd be viewing reflection, not absorption.

Go back to the simple, use a broadband source and try green food coloring, or add coloring to a drop of oil floating on the water reference.

Hope something here is useful. Cheers, Dave

Sorry, I don't have anything to add on your methods but you've got a nice peak around 700 nm which near the chlorophyll fluorescence peak! Quote from Tufillaro 2011 (see link below): "Chlorophyll has an absorption minimum near 660 nm, and a flourescence peak near 680. However, at very high algal levels this peak tends to shift to the read and move as far to the right as 710 nm."

http://www.aquahue.net/aquahue/..files/Tuf_Chapt.pdf