Research Projects


Hop Variety Profiling

We've summarized the analytical data for the 2014 hop harvest by creating variety profiles as featured below. The height of the bar above zero represents the average oil content in 2014, broken down by the dominant oil components. The height of the bar below zero represents acid concentration, broken down to alpha and beta components. These variety profiles are useful for characterizing hop varieties, tracking variation, and for quality analysis.

Characterizing Varieties: The single most common question during hop harvest is "Can you ID the hop that's growing in my back yard?" Well... not exactly. In the absence of a comprehensive genetic database, the best we can do is try to match an unknown variety to our database of variety profiles. For example, in the attached figure, we could easily distinguish between Ahtanum™ YCR 1 cv. and Amarillo® VGXP01 based on their alpha acid concentrations. Similarly, we could distinguish between Amarillo® and Brewer’s Gold based on Farnesene levels. However, the differences between Mt. Hood and Mt. Rainier are fairly subtle, as are Northern Brewer and Perle, etc. The moral of the story is that that there are well over one hundred varieties, some of which are closely related, so at best, profile matching usually turns up several candidate varieties while ruling out many more.

Quantifying Variation: We'll be digging a lot deeper into the variety profiles over the next few weeks. In some cases, we've got dozens of samples per variety, so we're able to determine if a variety grown in Oregon has a different average profile than one grown in Washington, Colorado, or elsewhere. We can also explore the effect of harvest date, kilning process, or any of a dozen other parameters. Moving forward, we'll be able to track annual variation, potentially making inferences about the factors that influence that variation.

Quality Analysis: From a quality perspective, hops that lie outside of the average profile are of particular interest. These outliers may represent a particularly special hop (the brewing equivalent of an award winning vineyard-designated wine grape) or they may be just the opposite. In either case, we can look at the factors that differentiate the outlier from the average in an attempt to better understand the factors that influence hop development.


Hop Aroma Correlations

In preparation for the upcoming Craft Brewer's Conference, we'd like to publish the following hop aroma correlations. The image below is a correlation matrix with a heat map showing aromatic similarity between several hop varieties. Cells that are bright green indicate two varieties that were judged to have similar aromas while the darkest orange-brown pairs were least similar. For example, in the top-left corner, Amarillo® VGXPO1 and Ahtanum™ YCR 1 cv. are neither similar nor dissimilar in aroma as indicated by the intermediate color, while in the bottom-right corner, the strong green indicates that Willamette and Warrior® YCR 5 cv. have fairly similar aroma profiles.

All varieties were ranked from 0-9 for the intensity of aroma in a dozen categories: Menthol, Tea, Green Fruits, Citrus, Green, Vegetal, Cream Caramel, Woody Aromatic, Spicy / Herbal, Red Berries, Sweet Fruits, and Floral. The evaluation was put on by Joh. Barth & Sohn and the panelists included beer sommeliers and a perfumist. The correlations presented here represent the overall similarity across all 12 aroma categories.

Hop Aroma Correlations

Now for the caveats. First, while this shows a mathematical similarity between varieties, that similarity doesn’t take into account the specific aromatic properties of the hops. For example, for some palates, spicy/herbal aromas are related to vegetal aromas and are very distinct from citrus aromas.

  Spicy/Herbal Vegetal Citrus
Variety 1 7 1 7
Variety 2 1 7 7
Variety 3 7 7 1


In the table above, varieties 1 and 2 would be similar to those palates because they are both predominately citrusy and have some spicy/herbal/vegetal characteristics. In contrast, variety 3 would be different because it features virtually no citrus, but is still spicy/herbal/vegetal. However, as far as mathematical correlations go, these three varieties are equally similar.

In addition, this chart does not take bitterness or oil content into account. If two hops smell similar but one has 3% alpha while another has 18% alpha, they will make poor substitutes.

Data Sources:
Schönberger, Christina, ed.The Hop Aroma Compendium: A Flavour Guide. Vol. 1. Nuremberg: Joh. Barth & Sohn, 2011. Print.
Schönberger, Christina, ed.The Hop Aroma Compendium: A Flavour Guide. Vol. 2. Nuremberg: Joh. Barth & Sohn, 2012. Print.