A long-standing method for determining tannin levels in wine developed at UC Davis is being challenged by an outside research group.
Larry Brooks of LM Brooks Consulting published a study that claims to invalidate the Adams-Harbertson tannin assay, developed at UC Davis by Dr. Doug Adams and Dr. Jim Harbertson. The assay is a scientific method for tracking tannin concentration in wine.
Tannins, which are astringent, bitter plant polyphenols that bind or precipitate proteins, cause the dry and puckery sensation in the mouth after drinking red wine or unripe fruit.
The 1999 Adams-Harbertson tannin assay was adapted for wine from the original 1978 Hagerman-Butler method, a protein precipitation-based tannin method applied in ecology and grain tannin.
“We didn’t really publish the paper to develop a method,” said Adams, professor at the UC Davis department of viticulture and enology. “We used a method that has been used since 1978. [The assay] has been used in other fields and has been used for around 30 years. It’s a well-accepted method. All we did was adapt it for wine.”
According to the paper by Brooks refuting the assay, “an extensive invalidation of the assay results with luxury wine data shows that the assay cannot distinguish bottled wine with reasonable accuracy.”
Adams, however, said that there is a distinction to be made – the paper does not document a validity test but a competency test of the labs.
“[Brooks] didn’t really conduct a test on the assay itself,” Adams said. “He sent wines to different labs and got different numbers. What you will see is there is no data that talks about validity of the assay but different labs’ ability to test.”
Brooks works as an independent consultant in the wine business in winegrowing and winemaking. When one of his winemaking clients wanted to start using this assay, Brooks had doubts about its accuracy and precision.
“Before I could recommend the client start using this assay as a decision making tool I wanted to validate it using standard protocols,” said Brooks in an e-mail interview. “In these protocols different labs are sent blind coded samples of wines to analyze. If the results are in agreement within certain parameters the analysis is considered valid.”
Brooks had four winery labs and one commercial lab in California use the Adams-Harbertson to assay nine replicates of three bottled wines available in California supermarkets.
“Our study found that this assay had unacceptable variation in results both within and between laboratories. We were not able to validate [the assay],” Brooks said.
In a written statement by UC Davis, Andrew Waterhouse, chair of the viticulture & enology department, said that like any analytical procedure, the Adams-Harbertson assay requires time and effort to ensure that a lab or analyst can deliver reproducible results.
“Regular comparisons of routine assays between winery labs show large variations – up to 20 percent in some cases,” he said. “So Brooks’ study would appear to re-substantiate this common occurrence rather than invalidate the tannin assay.”
Adams said UC Davis does not have plans to review its research.
“The reason we don’t have any inclination to go back and take a look at it is because we have to write research grants,” he said. “It would be difficult to get research money [when] the method is well accepted and used for lots for other things, like in ecology.”
Brooks does not believe the Harbertson-Adams assay should be utilized for basic research in winegrowing and winemaking.
“I think that the desire to have an inexpensive winery-operated analysis for tannin is a valid goal,” he said. “I would love to see it achieved. I don’t think we are there yet, and don’t believe that this assay works for this.”
Brooks co-authored the paper with Leo McCloskey and Doug McKesson from Enologix, a Sonoma based consulting company, and Marshall Sylvan from UC Santa Cruz. It appeared in the September 2008 issue of the Journal of the Association of Official Analytical Chemists.
POOJA KUMAR can be reached at firstname.lastname@example.org.