Rwanda + Burundi Coffee Fundamentals

With a focus on East Africa, we discuss how quality coffee happens, and the hard work behind it…

Updated 9/10/2019

A lot has happened since this article was published 6 years ago. The quality of coffee coming out of these two regions has improved, and our commitments to Rwanda and Burundi coops and coffee washing stations have grown exponentially. The background information has only been slightly modified to reflect on the years that have passed between then and now. Chris’s roasting comments hold up well, and we still tend to follow his original recommendations outlined below. 

Originally written by Thompson Owen and Christopher Schooley published  6/4/13

What Goes Into Producing Top Rwandan Coffees 

Rwanda has one of the most interesting East African coffee histories. It is a place where the production of high-quality coffee is inextricably linked to the rising spirit of a population after the tragic genocidal civil war of the 1990s. Known as the “Land of a Thousand Hills,” many of them are cultivated in high-grown coffee areas between 1700 and 2000 meters. Rwanda coffee can be world class, with clean bright flavors that go beyond the best Central America coffees, more balance than Kenyas, attractive fruited sweetness, often spiced and sometimes floral, and uniquely tea-like in the finish.

It is believed that coffee was introduced in Rwanda in 1904 by German missionaries. Around 1930, a considerable interest in coffee developed as it was the sole revenue-generating commodity for rural families. Coffee is grown in many sectors, but most coffee comes from the South and Western districts. It is grown all along Lake Kivu, from the northern area of Gisenyi through to the central areas of Kibuye and Nyamasheke, down to Cyangugu. In the South there is a lot of production in the vicinity of Butare. The North has more limited coffee growing, with much in Rulindo district north of the capital Kigali. And the East produces a decent amount of volume, but much of it is lower-grown at altitudes of 1300 meters or so, whereas many farms in other areas range from 1700 to 2000 meters. There is some good high-grown coffee in the East of course.

The government encouraged (actually, they mandated) high-volume, crudely-processed coffee production. Even with this low grade coffee production, coffee played a considerable role in the economic development of the country because it was one of the few cash crops. But with the collapse of world coffee prices at the international market level, the push to export low-grade arabica made less and less sense. Historically, Rwanda had been the 9th largest producer of arabica in Africa, with 500,000 small farms averaging less than 1 hectare each. Farms have usually not been measured in land area: being so small, they were measured in number of trees. The average is 165 trees per farmer, minuscule compared to other nations! The season in Rwanda harvest is generally March – July putting coffee in the US have closer to October – February.

This coffee cherry at Mushonyi station will undergo two grading stages before it’s even depulped

Rwanda coffee was traditionally processed by each small producer, and there was only one true wet-process coffee mill (called a washing station here) in the country. This home-processed coffee still accounts for the majority of production in the country, and is called “ordinary” or “semi-washed” coffee since most of it doesn’t have complete fermentation to remove all the fruit mucilage from the parchment layer of coffee. It is estimated that 60 to 70% of current production goes into semi-washed blends. This type of processing isn’t inherently bad, but when you mix it all together, well-processed batches with poorly-picked and processed ones result in the lowest common denominator for cup quality. As I ask producers, what happens when you mix a glass of clean spring water with a glass of muddy river water? Two bad glasses of water!

This type of low-grade production never returned much to the farmers, but there was so little export production of any kind from Rwanda, it had an outsized significance to the country and to the individual coffee producer. Then the genocide occurred, and how any society returns to a “normal” life after the tragedy of monumental scale is difficult to imagine. But the recovery in Rwanda has occurred with an unflinching openness to the genocide. After the genocide, as the floodgates opened for assistance to the Rwandese population, revitalizing coffee production was seen as an important goal. To do this, organizations like the PEARL project and SPREAD standardized and trained farmers and new cooperative washing stations in traditional techniques of coffee production based on other East African countries. Burundi in particular offered a good model for production.

From the farm through to the washing station, coffee production in Rwanda is quite ideal for a small-holder farmer type system. The original ways of planting and pruning the tree are well done, with the already-mentioned varietal selection (Bourbons-types, generally), plant spacing, mulching for water-retention, organic material input and weed control, light shading of coffee with trees, and Kenya-type pruning techniques. The government via their coffee board NAEB distributes fertilizer to farmers via the wet mills. Bourbon types grown in Rwanda include these attractively named plants: POP3303/21; Jackson 2/1257; BM 139.

After the coffee is picked, the cherries are often floated in water to remove light beans. Pulping is often done with Kenya-type disc pulpers than include a grader for light and heavy beans. The light beans are taken out and go to a secondary huller and a separate low-grade fermentation tank, for B and C grade coffees. The heavy dense beans go to the fermentation tank for A grade beans, and these are eventually designated as A1, A2 and A3 qualities.

One of the best things in Rwanda processing is the fact that all the coffee, after it is fermented and washed down the channels to remove the fruity mucilage layer, goes to the “skin-drying tables.” These are raised shaded beds where the wet parchment coffee is picked over to remove defects that are especially apparent in the still-wet parchment. In particular, the workers remove Antestia affected coffee, under-ripe beans, pulper-nicked coffee, fruit skins, or beans where the parchment was mistakenly removed, having been affected by the ferment water in the tanks.

You can tell the drying phase was gentle as this parchment is nearly crack-free, Gashonga coop

This skin-drying phase not only allows an extra chance to remove defects, it slows down the initial drying of the parchment, which I feel increases cup quality. The coffee then goes out the the raised drying tables, where it takes 15-20 days (ideally) to reach a moisture level of 11% or so. In the heat of the day, the workers cover the coffee to prevent too-rapid drying under direct sun. The result with the best Rwanda coffee is a totally white-colored parchment coffee with little to no cracks from rapid drying. Why is this good for quality? Because each little bean has it’s own little drying environment, buffering it from the outside, a shell that allows a slow and even loss of moisture, resulting in less loss of organic compounds that are good for cup flavor, and result in a green coffee that can be stored longer without losing taste quality.

One of the challenges in Rwanda are the poor organic material content in the soil. Every small patch of land is cultivated in this country, and has been for many decades. The soil is depleted, and there isn’t enough sources for new organic material to add back. Some level of chemical fertilization is needed, as well as returning every possible type of compost to the ground. The NGOs brought the California Red Worm here to introduce vermiculture as a source for improved compost, especially to break down the coffee pulp (the skin and outer layer of the fruit) created during processing.

Another issue that is often overlooked is the business of the washing station. In years where the coffee market is high, there is a rush of investors to build washing stations and buy coffee cherry, often without forethought about how they will finance their coffee cherry purchasing, without knowledge of how they will sell the coffee (so they often overpay for cherry thinking there is a big payout on the other side), and general ignorance about the relation between quality and price. Additionally, cooperatives are often poorly-managed, and in fact some are not coops as we imagine them to be. They might be a group of some larger local farmers and a few people from Kigali, the capital, who have never farmed coffee in their lives. It might be 10 “coop members”, but the farmers they buy cherry from are not actually members of the coop, nor allowed to be. With both these less-than-ideal scenarios, the problem for the farmer is that the market for their coffee cherries is not reliable. One year the station might operate, the next year it doesn’t. They always have the option to process themselves and sell it as semi-washed, but the price is much lower.

For our part we have always sought out private and cooperative run stations that are serious about the coffee business, that understand it well, are from the area, and want to return something beneficial to the farmers they buy from. Some of our original sources were supported by  programs like the Technoserve NGO in Rwanda, and now receive support from the service providers born out of that effort.

Transportation can be a problem with Rwanda coffee too. The coffee has historically been transported across Uganda to Mombasa, Kenya for shipment to Europe, a trip that can damage the coffee, and one that relies on economic and political stability in the region. The result is that the coffee cannot reach market, so the price and the incentive to produce top-grade coffee had diminished greatly for the village coffee farmer. The fact that rural people can tend their crops and get higher export prices for them is a good sign for Rwanda, and for us, because we benefit from the high quality of the resulting coffees. And in fact we have had less problems with shipping and receiving our Rwanda coffees each year.

What Goes Into Producing Top Burundi Coffees

Burundi coffee bears a striking resemblance to that of neighboring Rwanda, in both cup character, and in the culture surrounding coffee. Bourbon-type varietals flourish in both countries and Rwanda has imitated Burundi’s traditional practice of wet-processing coffee cherry. Their cup profiles can be dynamic and bright, with red fruits, berry or citrus, complex baking spices and with a great sweetness lingering through the finish.

Burundi is a small landlocked country at the crossroads of East and Central Africa, straddling the crest of the Nile-Congo watershed. Sandwiched between Rwanda, the Democratic Republic of Congo and Tanzania, Burundi has beautiful Lake Tanganyika for much of its western border. This is a country dominated by hills and mountains, with considerable altitude variation, from the lowest point being the lake at 772 meters above sea level to the top of Mount Heha at 2670 meters above sea level.

Coffee has a similar history here as neighboring Rwanda. These are wet-processed coffees, and often employ a two-stage fermentation method, as you might find in Kenya. Their practices in coffee wet-milling are definitely good, provided they are followed. If the coffee that is selected includes unripe cherry, a good washing station will ask the farmer to sort these particular cherries out. The under-ripe coffee can still be submitted separately at some stations and often are purchased for the same price in order to avoid penalizing the farmer (this needs to be considered in terms of quality – stations that pay on different scales based on quality of cherry selection motivates the farmer to pick better).

Sorting through the wet parchment on “skin drying tables” at Kazoza Ni’Kawa

Many washing stations have large concrete basins where the farmers immerse the coffee cherry, skimming off “floaters” that have failed to mature. Floating the coffee cherry is a great step towards a better quality cup. In my experience the first 12-36 hour fermentation is done without water (aerobic fermentation) and the second fermentation is done under water (anaerobic), but this can vary from station to station. The washing station is perched on a slope and the coffee is washed from the first, higher tier of fermentation tanks, and on down a channel where mucilage is agitated off the coffee. It then lands in a second strata of concrete tanks, where it is left submerged in water. Then there is one final wash as the coffee passes down a concrete channel, and is taken to either “skin drying” beds or full sun beds, where the eventual hand-picking removal of defects will take place.

Coffee farming does not have an extraordinarily long history here. The first Arabica coffee tree in Burundi was introduced by the Belgians in the early 1930s and has been growing in the country ever since. Coffee cultivation is an entirely small holder based activity with over 800.000 families directly involved in coffee farming. Their combined total acreage is roughly 60.000 hectares in the whole country and planted with about 25 million coffee trees.

Like Rwanda, Burundi is primarily planted in Bourbon, which is grown at high altitudes ranging from 1250 to 2000 meters. Also similar to Rwanda, smallholder farmers of Burundi tend to about 50 to 250 trees. Historically, coffee from the area was sold as bulked “Ngoma Mild” coffee (Ngoma is a traditional drum). The farmers would bring their coffee to local washing stations, which along with 20-30 other wet mills, made up the Sogestal. All of the coffee collected from the Sogestal members would be blended, and separating qualities was not possible.

Pushing freshly processed coffee through the grading channels at Teka washing station

Several years ago the coffee market was “liberalized”. This meant that individual washing stations could now keep coffees separate, and then market the individual lots to buyers by station, “day lots”, or processing batches. With this comes the new possibility to find gems that were formerly mixed in with the not-so-good lots. So new possibilities emerged in Burundi, and quality has drastically improved through the years.

I’ve made several trips to Burundi over the past decade or so, to visit the cooperatives and washing stations we buy from, and to cup during harvest season. Even still, I’m a relatively late-comer to Burundi coffee, but I still see a mix of potential and great challenges here. When the coffee is good, it can easily top 90 points and pique our interest. But when it’s bad…well, the coffee is no longer considered except maybe in terms of what went wrong along the way (typically bad processing, bad logistics and transport, or by politics of the coffee trade that support unsustainable practices). But thankfully we see less and less of the latter, at least from our Burundi partners, and in our experience Burundi coffee has never been better.

The Potato Defect

The dreaded “potato defect” is another great challenge to producing top coffees in both Rwanda and Burundi. The defect is so named because it smells like an old, sprouted potato in the cup. The problem is specific to the East African lake areas of Kivu, and is also found in Congo coffees. This off-flavor is caused by a bacterial agent that enters the cherry skin and produces a pyrazine chemical toxin that binds to the forming green beans. The bacterial transmission is often brought into the fruit by the Antestia bug, a type of coffee berry borer insect that is attracted to sugars in the coffee fruit. But anything that pierces the cherry wall can allow the bacteria to enter and eventually release the the nasty pyrazine.

Besides removing the green beans that have been attacked by the Antestia bug, there is still no mechanical way to identify and remove these beans. The result is that, even with top coffees, an occasional cup will have this potato taste. But farmers that manage their trees well, harvest all the ripe cherry, pick out any scored or pierced cherries, and do not allow cherry to fall to the ground will have much lower incidence of potato defect. And it seems like we see less and less of it each year. It’s a bummer to get a potato cup, but in our opinion, a problem worth managing given the cup quality.

Roasting Coffees from Rwanda and Burundi 

The coffees of Rwanda and Burundi have a lot going for them in terms of roasting. For starters, their bourbon pedigrees grown at high altitudes lend to a crisp yet balanced brightness and long sustained sweetness. But we all know that the best qualities of any coffee are latent and not intrinsic and can take a good deal of effort to ensure that those qualities be front and center. I love to put these coffees next to bourbons from Central America and look at how similar yet varied they are side by side. Roasting wise, they are extremely similar in how they take the heat, but I feel like the Rwanda and Burundis benefit the most from some specific roast profiles.

Burundi has really come a long way in recent years in terms of selection and processing. For years, the Burundi coffees that were available still had a good deal of rustic qualities to them. You could get a sweet raisiny or fig like fruitedness out of them but there was still a mineral-like thread in the coffee. With the honey sweetness this mineral character could be a pleasantly complex element, but as recent offerings have shown us in Burundi, these coffees more and more tend to be crystal clear. The best coffees from the last few harvests have had brilliant acidity with a more spiced thread rather than mineral, and even floral cup character. Instead of tasting rustic, the coffees are crisp and tea-like.

As Tom mentions in talking about the history of coffee in Rwanda, specialty coffee production has greatly benefited from aid programs that focused intently on quality instead of quantity, pushing the value of these coffees. Really special coffees from Rwanda can be juicy and fruited, berry and citrus notes, with even the floral qualities taking on an orange blossom or raw honey characteristics. The sweetness can be refined, raw sugars, honey or bittersweet oriented.

Beautiful old Probat 5-barrel coffee roaster at a Rwanda coffee lab

Rwanda is the coffee that I used for my Stretchin’ Out the Roast articles because of its clean yet dynamic character and because of the ability of the beans to really take the roast. These coffees can take a range of roasts but some of their nicer qualities really benefit most from a quick City roast; they’re high grown, extremely dense bourbon coffee, and they can take the heat. You can push these coffees up into Full City and get sweet balanced flavor profiles. But truly these coffees shine at a lighter roast and the malic, sweet citric, and tartaric qualities really pop.

Look for a vigorous 1st crack and push it through. Take your foot off the gas and let it have a definitive end to 1st crack, but don’t pull back too early and let it sputter through the crack. Giving yourself about 15-30 seconds from the end of the crack will put you squarely in that City, City+ range as long as you’ve pulled back on the roast either through reducing gas or changing the airflow depending on your roaster model.

You can stretch the drying stage in these coffees, and I did get a longer juicier finish in the Burundi which was very pleasant, but didn’t quite have the sweet pop of the white grape. I would say that if you are looking to pull SO espresso with either of these coffees, a stretched drying stage would be a smart move. In the Rwandan coffees the stretching produces a more syrupy than juicy mouthfeel, a little more weighted than the Burundis, and with a candy-like sweetness that lingers long into the finish. The sweetness and mouthfeel are certainly nice, but the sweet citrus/orange note is not as prominent and the floral notes can be more muted.

Honestly, some of my favorite brews and espresso I had were 50/50 blends that I did of my quicker roasts with the roasts that I stretched the drying stage out with, getting the most from the crisp clean acid sparkling in the juicy mouthfeel. This is obviously a rather labor intensive consideration from a production perspective, but it could be a really interesting way to showcase the best of these coffees as well as have a conversation starter about what constitutes a blend and what does Single Origin mean.


Some other articles of interest:

Burundi Coffee in Permanent Crisis
Green Coffee, Defined
Congo Kivu Coffee Photo Tour

Leave a Reply

Your email address will not be published. Required fields are marked *

 

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Reddit
Facebook
Twitter
LinkedIn
Email