We are wrapping up our analysis of the 2016 Catracha Quality Project (CQP). Royal published a lot on this over the last two months, but if you haven’t been following this series, I’d suggest starting here. I’ll summarize a few of our key takeaways below.

The team has started focusing on the next step, taking the information back to the Catracha producers in Santa Elena. The process is already underway.  Mayra Orellana Powell and Lowell Powell are in Honduras and attended a producer meeting held on October 21 to plan for the next harvest, which will begin in late December. Lowell asked to be included in the programming so that he could present some of our findings from the quality project.  He used the process map to emphasize the strategies used on four different farms to process quality coffee.

This particular meeting was part of the regular calendar of capacity building meetings that Catracha Coffee organizes throughout the year. USAID, a United States development agency that funds training programs in Honduras, provided three training sections; one emphasizing the importance of supporting women as coffee producers; one on properly calibrating a de-pulper and keeping the wet-mill clean; and one on coffee drying strategies.  Each section was held on a different area around Luis Nolasco’s wet-mill (one of the CQP data collection sites).  The day ended with tamales, coffee, and some conversation about the coming harvest.

copy-of-cqpgroup

In all more than 60 Catracha producers–plus another 25 producers interested in joining Catracha–attended the meeting, which is really exciting. It shows the continued commitment to the Catracha model. Everyone was divided into smaller groups and rotated through a few different workshops. Lowell and Mayra had a chance to detail some of what we’ve learned through the data project.

So, what have we learned?

How Do Catracha Producers Process Their Coffee?

The first goal of the Catracha Quality Project was to create a map to visualize how the producers in Santa Elena process coffee from cherry to dried parchment. For two harvest seasons, the CQP team has interviewed producers and observed their practices. With this information, Arvin Juan created a map, a visual tool for identifying best practices for coffee production in Santa Elena.  Lowell translated the map into Spanish for the October producer meeting.

copy-of-mapa-de-processar

We learned that most producers follow the same post-harvest process, but logistics play an outsized role in how decisions are made. Producers process coffee onsite, usually, with their own micro-wet mill rather than collectively at a larger facility. Improvements have been made over time to these micro-mills.

We did learn that there is a variation in picking between producers, who are often faced with a choice. Do they hire many pickers who can execute quickly? Or few pickers that are more selective? We believe that these maps will be the most valuable tool for implementing consistent best practices and promoting collaborative innovation in the years to come.

Lowell and Mayra used the process map to illustrate where selection strategies could lead to improved quality. “I felt like I was really engaging with producers and the map gave me the ability to solidify their own best practices,” Lowell said. “I believe the group of nearly 60 producers will become even more consistent in their processing practices this year and the map will help less experienced newcomers quickly improve,” he said.

What Role does Cherry Selection Play in the Final Cup Score?

Catracha organizers encourage producers in Santa Elena to focus on quality and to select only ripe cherry. As part of the quality project, seven producers agreed to record the weight of both picked and rejected coffee.  This year, after the coffee was processed, forty-three samples were sent to Royal’s lab where they were cupped by coffee professionals from  Blue Bottle Coffee, Flying Goat, Equator, Andytown, Mr. Espresso, and Tico Roasters. Interestingly, cup scores varied between picks. The range was as much as 4 points (83 – 87).

We used this cupping data to try and find correlations with producer practices. We asked is there a correlation between cup score and the percentage of rejected cherry? There does appear to be a “window” where the percentage of rejected cherry correlates to a higher cup score.

overall2

On Luis Nolasco’s farm, the highest cup score correlated to 5 or 6 percent of rejected cherry, with lower cup scores at both higher and lower percentages. We saw a similar window for Santiago Lopez and observed across all seven producers when joined together on a single scatter plot. Does this show the natural progression of cherry ripeness? With early and late picks showing lower cup scores and correlating to low and high percentages of rejected cherry? Lowell and Luis and others are still teasing out the answers to these questions. “I am really looking forward to being in Santa Elena for the entire harvest so I can ask more questions about potential picking strategies that might really impact cup scores,” Lowell said.

How about Fermentation? How does that Process Impact the Final Score of the Coffee?

While coffee was being processed, we recorded the length of fermentation and the coffee’s pH level. Andy Noblet compared that information with cupping scores recorded by our team of professional cuppers. We found that most fermentation time was roughly 46 hours, and final pH was between 3 and 3.4. The coffees, in general, scored between 84 and 85. “Notice how tight distributions are,” Noblet wrote.

fermentation-1024x360

There were some outliers. Atanacio Nolasco’s first experimental pick scored quite high. Interestingly, this sample showed a relatively fast change in pH and it dropped to the lowest recorded final pH value. Fidelina Perez had the next highest scoring coffee. “The fermentation numbers for this run are odd,” Noblet wrote. The fermentation was long, but the pH dropped from 4.5 to 3.7 – starting more acidic but ending less than is average.

These unique samples are worth mentioning—and exploring further with Catracha producers.  Additional equipment will be purchased to measure the temperature of the slurry in the fermentation tank.  

Our more general takeaway is this: Consistency in fermentation seems to lead to consistency in overall cup quality. Noblet again: “If we assume that each producer’s coffee has a baseline potential quality, that level of quality is achieved more regularly when the fermentation process varies less.”

The CQP team will use this year’s harvest season to evaluate potential methods for finding the fermentation sweet spot and strategies to consistently hit the mark.  Lowell thinks that the amount of coffee in the fermentation tank might have an impact on fermentation times and change in pH.  He said, “finding the optimum amounts of coffee to put in the fermentation tank might be achieved with more consistent picking strategies.”    

To GrainPro or Not?

The pH experiments relied on data loggers that measure temperature and relative humidity. They were secured inside of solar dryers and in other places around the mill area.We thought it would be interesting to put two in a bag of coffee while it traveled from Honduras to California on a container ship. Chris Kornman wrote about a similar journey from Colombia.  One logger went inside the GrainPro lining and the other inside the jute bag outside of the liner. How does the journey impact the coffee?  Does the relative humidity and temperature of coffee that’s bagged inside of GrainPro differ from coffee that simply in a coffee sack?

Here’s what we found. Interestingly, the GrainPro lining kept temperature and relative humidity stable, while coffee that was in just a jute bag experienced greater change in both areas.

First, let’s discuss temperature. Until the coffee arrived in Oakland in July, there was no notable variation in temperature. But, while the coffee rested in a storage warehouse, the temperature inside the GrainPro remained relatively stable. But without GrainPro, the temperature fluctuated by as much as 3 degrees Celsius. We found similar results for relative humidity, inside the liner the humidity remained virtually unchanged, while it changed wildly inside the bag alone.

Without reading too much into a single experiment, we can say that temperature storage in GrainPro helped to mitigate swings in temperature and humidity, at least for this journey. This is important information to consider. How does coffee age over time? How do we maintain freshness as best we can? It seems GrainPro might be a good option.

This year’s quality project has created a lot of questions for the data team (Arvin Juan, Andy Noblet, Lowell Powell, John Casazza, and myself).  With a few more years working together with the Santa Elena team (Luis Nolasco, Fidelina Perez, Atanacio Nolasco, Jose Antonio Nolasco, Santiago Lopez, Porfirio Lopez, Antonio Granados, and Keny Amaya) on data collection we might find ways to consistently squeeze a few more points out on the cupping table.  This would mean even better prices for producers and increasing resources to care for their families, not to mention some pretty amazing coffee to boot.  I think we are all in for another exciting harvest and more discovery.