Hi Amanda,

I have seen this happen when samples get to the lower detection limit of the spectrophotometer method (

Thank you both for these answers! You've given me a few things to check! 

At very low concentrations just the smallest errors can result in negative results; e.g. particles in the cuvette, smudges or scratches on the cuvette etc. Perhaps you should consider using a cuvette with a longer pathlength (e.g. 5 cm), which will result in higher absorbance readings making the spec less sensitive to the low concentrations.

Have you ever try the acetone extraction method? It is easier to run and perhaps you get rid off the negative values on Chl 'a'. Let me make a resume and i´ll send it to you.

Amanda,

I think you already got the best answers for your question. However, it will first be nice to know the system from which your samples came from. Is it eutrophic, oligotrophic (mostly clear waters) or in between? For oligotrophic systems like Lake Tanganyika, we do not do acidification, because we were running into the same problems. Then we started using acetone for extraction and a 5 cm cell, which improved our results. Off course a fluorometer is better. And you may consider using HPLC if the problem continues. Best of luck.

Gavin- that is a good suggestion on the path length, as well as the comment about the impurities, I'll give that a try. 

Leonardo- We have tried an acetone method in the past though found that our results did not differ from an ethanol based-method. Perhaps I will go back to that in the future. 

Ismael- these lakes are actually fairly eutrophic, but that's a good tip about not doing the acidification step for oligotrophic samples, I hadn't heard of that! 

How much water did you filter before doing the extraction?

I use extraction with DMF. Using that method and large filter volumes I get good repeatable results very quickly for all kinds of water samples. The extractions take minutes and the analyses can be completed shortly thereafter. I can complete something like 100 analyses in a normal day with one person doing the work. I have a reference detailing this method available here in RG.

Edit: A couple more things. If your water is eutrophic then you should not need large samples. I'm measuring concentrations near limits of detection in many of my samples. One more thing to keep in mind is that anything that absorbs at that wavelength may 'show' as pheophytin. If there is a lot of non-living seston in that water, I would also advise using HPLC which should be more specific and sensitive.

This only happen to us, even with the acetone method and not so low absorbances in the past. In our case it was the effect of using low quality filters. We switched to a better brand and the problem was resolved.

John, we pretty much quick using the Spec method years ago - except as a cross calibration on about 10% of our samples.  Instead, we use fluorometric detection and a filter volume of 5.0 ml on mini-GFC filters  We extract in DMSO:Acetone directly in a sealed culture tube that serves double duty as a  fluorometer cuvette.  EtOH extraction works well too. At least 100x easier than Spec method and much more sensitive.

Me too but the original post was with regard to using the spectrophotometer approach, hence my response. You're right though, especially for waters that are eutrophic.

 i hope this might be helpful

Youo  could look at

Less is better: Uncorrected versus pheopigment corrected photometric chlorophyll-a estimation. H. B. Stichand A. Brinker, Arch. Hydrobiol. 162 1 111–120

I think this link is useful, 

http://www.instanano.com/characterization/theoretical/concentration-calculation-from-uv-vis-absorbance/