Missing the forest for a tree

22 minute read

“The nation that destroys its soil destroys itself”


I’d been a coffee buyer for barely two years and had just gotten my Q when, in 2014, I was invited to a cupping series in Chicago at Intelligentsia’s Roasting Works organized by Michael Sheridan, then of the Coffeelands Project through Catholic Relief Services. I was new to the industry, felt essential and indispensable to the company I worked for and, imagining that the world would fall apart if I left for a week, I declined. I didn’t know it at the time but I’d think about the fact that I wasn’t there for years. 

And, in fact—I’m thinking about it today. So I decided to write about it—to no one in particular. A year later, I still can’t safely go to happy hour at my favorite neighborhood haunt let alone fly to Colombia. 

What else is there to do, anyway?

I’m ambivalent about genetic modification (or “GMO” as we’re used to seeing it in media, on reddit, and in advertising on packaging and restaurant doors targeting the guileless and gullible).

On the one hand, there is no scientific evidence that GMO crops negatively impact human health or longevity [hilariously, as a sidebar: the same people who are concerned about GMOs are often concerned—obsessed—about mycotoxins. And yet, at least in corn, GMO crops had up to 36.5% less presence of mycotoxins. Shit!]. On the other hand, accelerating crop selection through gene splicing rather than traditional breeding is risky and likely to produce unintended outcomes—whether through consolidation of power to one corporation, or by outcompeting heirloom and native species, or simply because humanity still can’t predict the outcomes of random genetic mutations or competitive pressures. And there’s the issue that plant breeding selects for singular attributes—disease resistance, yields, whatever—specifically for application in commercial monocultures.

And yet:

Papaya is, I’m told, a pretty divisive fruit among white people—but I like papaya, and I’m glad it still exists. Without GMO, it probably wouldn’t.

A disease—papaya ringspot virus—threatened global production of the fruit. Because fruit production typically results in the selection of trees from a limited genetic pool with favorable traits for commerce (yield, flavor, ripening, etc.), it leaves them vulnerable to diseases that exploit that genetic makeup. This agro-economic monoculturing process left papaya trees susceptible to PRSV globally—kind of like roya and coffee. It could have caused the species—devoid of immunity—to simply cease to exist.

But scientists—specifically, one working for Monsanto—engineered immunity, licensing the (patentable) gene sequence at no cost to farmers and making seeds available at cost, and thus: papaya still exist.

I remember learning about cloud seeding in one of the geochemistry courses I took in undergrad, and how China used it to will the world to rain and empty the skies ahead of the 2008 Olympic Games. It’s the 21st century garden of Louis XIV*, I think—attempting to force nature to submit to the will of humanity—and it’s proof that Mary Shelley was ahead of her time. 

* The first greenhouse in Europe was built to house a tree presented by the Dutch to King Louis XIV of France in 1715. It was a coffee tree, a cutting from which was smuggled to the new world and eventually planted in plantations on the island of Bourbon—after which the cultivar is named. So much for cloud seeding.

Apparently the strongest tie that binds the past to the present is the constancy with which humanity overestimates its cleverness and fails to predict the consequences.

I love papaya, but: fuck papaya.

In the now-infamous Castillo/Caturra comparison study, the team at CRS and their partners in the coffee industry set out to examine the apparent choice faced by coffee producers of whether to plant Caturra—Colombia’s traditional, high-quality, high-yielding cultivar which is susceptible to rust—or Castillo, a cultivar introduced by Cenicafé to address the threat of coffee leaf rust and CBD.

Castillo, to its proponents, exemplifies the potential of what plant breeding technology can achieve. It’s an F5 progeny developed to be a supertree—as high quality in the cup as the traditional cultivars but with lower nutrient requirements, equal or higher yields, and resistance not only to roya but also coffee berry disease.

Science engineered a better tree than nature could.

Cenicafé’s genetic improvement program began in 1968 in response to the threat of coffee least rust and, over the subsequent years, turned out cultivars still grown in Colombia today—variety Colombia in 1982 (a year before roya arrived in Colombia),  Tabi in 2002, and then, in 2005, Castillo.

Just three years after the release of Castillo—as if on cue—Coffee leaf rust decimated production across Latin America.

Anticipating the threat of roya to the coffee industry—Colombia being the third largest exporter of coffee in the world, roya could have massive economic and political consequences for the country—Cenicafé engaged in a program to push adoption of their new cultivar across Colombia. Cenicafé, through FNC, incentivized renovation of farms with Castillo by essentially offering free money to farmers who replaced their more traditional, rust-susceptible trees with the new rust-resistant cultivar. (financing in coffee warrants its own post, eventually, but suffice to say this sort of deal for producers is highly unusual).

Since FNC already recommends replanting trees every 7 years to promote higher yields—why not trade in your old model for the new one, on the house?

When it was introduced in 2005, Castillo was heralded as a great achievement in the fight against roya and is still regarded by proponents of plant breeding—including long-time and influential members of the Specialty Coffee Association and World Coffee Research—as an unmitigated success and demonstration of the promise of breeding programs. And, as Michael Sheridan reported in his Re:co talk on the subject (very worth a watch), within just 10 years of its introduction, over 40% of the coffee grown in Colombia was Castillo.

Castillo was bred from selections of Timor and Caturra

Broadly, Castillo is a cross between Caturra—a traditional cultivar in Colombia, a natural mutation of Bourbon producing high yields and a high-quality cup on a compact tree with high nutrient requirements and susceptibility to roya—and Hibrido de Timor—a naturally-occurring and rust-resistant hybrid of C. Canephora and C. Arabica (Tipica, specifically) that was discovered in East Timor in 1927.

Castillo, Colombia and Caturra growing in Palestina, Colombia

The first time I ever saw a Castillo tree, it looked different. Visually, its leaves were slightly larger and darker green than its neighboring trees (which were Colombia and Caturra, mostly), with nodes densely packed with oblong berries somewhat larger than on other trees. 

And: while much of the farm showed signs of Hemileia vastatrix infection, the Castillo plants did not.

While the role of climate change in the spread of roya is disputed, some strategies a farmer might take to combat roya—such as moving plantings to cooler, higher ground—are increasingly unavailable as the world warms. And as temperatures increase, mutation of the fungus accelerates, further pushing the enemy we seek to control out of our grasp.

Add to that the effects of global economic disruption. 

Coffee farms around the world exist in peril.

What do we do?
Do we seed the clouds and change the weather?
Do we play god and write poetry with genes?

For years, I heard of buyers avoiding Castillo because it “cupped like a Catimor.” They’d describe it as astringent and vegetal, and with less sweetness, structure or complexity than Colombia or Caturra cultivars. George Howell was known to have regarded this as the signature flavor of Timor hybrids, a flavor he called “the tail of the devil.” Damn, George.

And in a famous incident in 2010, a coffee from Finca La Loma that was submitted as 100% Variedad Castillo was awarded a score of 94.9 points, a score that seemed so implausible that the producer, Jose Antonio Gualguan, was accused of lying and falsely labeling a Caturra as Castillo. Trade publications like Sprudge sowed further suspicion, writing in their reporting on the subject:

If every farmer in Colombia believes that they, too, can win the CoE with Castillo trees, a very slippery slope indeed could be underway in Colombia. If the FNC gets to 15 million bags a year by convincing farmers to do away with Caturra and other heirloom varietals under false pretenses, many farmers will lose the opportunity to produce spectacular products and garner higher and higher prices at auction.  

Was Castillo—a coffee heralded as a solution to the threat of roya—really so inherently inferior that a 90+ score warranted skepticism?

Sorting Castillo by hand for defects and ripeness

Many producers I worked with believed so. They talked of needing to pick Castillo at a different color ripeness than other cultivars owing to how the acidity modulated as the fruit developed and how the sugar content of the mucilage seemed different from other trees. Indeed, as I pulled the flesh off of one cherry, I noticed that the mucilage of coffee from this tree was thicker; on the tongue, it was more astringent and less sweet than neighboring trees. This producer also noticed that heavy rains caused the Castillo to drop fruit, a problem other cultivars didn’t seem to have (a problem I would later learn can often be caused by other agronomic stresses, such as calcium and boron deficiency in the soil—but I digress).

One of the producers was in the process of replanting all of the Castillo on his farm with other trees, such as Tabi, which he regarded as offering superior quality with respectable rust resistance. While the average score of coffees from his farm was 86+ and while he would go on to place in Colombia’s Cup of Excellence in coming years, the Castillo on his farm rarely scored higher than 83 or 84. With lower scores came lower prices. How much was rust resistance worth when the C-market was low and neither the FNC nor private exporters were offering a liveable price for that coffee?

We worked together and processed a 10-bag lot of Castillo using a protocol we developed together, and I paid a premium, hoping that this technological fix would find a way around the quality issue the cultivar seemed to present, offer some extra money to help offset the costs of renovating that sector of the farm and preserve his volumes and rust resistance along the way.

We were able to raise the cup score against a reference lot from 84.5 to 86.5, so when he renovated that part of his farm and replaced all of the Castillo, we decided to replicate the process at his neighbor’s farm, which sat at a slightly lower altitude. The neighbor’s Castillo cupped about the same, and while the agronomists were happy we again wondered if there was something more we could do to improve this devilish cultivar.

[George Howell tasted this lot of the neighbor’s coffee in 2019 at a seminar at SCA in Boston and gave it an 88]

While I waited for those samples to arrive at my lab in Cleveland, another producer from a different part of Colombia sent me coffee from their farm, indicating that it was a Castillo—their standard wet process of a dry fermentation for 16-18 hours then washed and slowly dried in partial shade—and that they hoped I’d enjoy it.

As I opened the small parcel and emptied the green coffee from its sleeve onto a sample tray, I noticed something: it looked different. The seeds were larger than the Castillo I’d processed the previous years, and larger than the seeds from the Castillo trees on the farm I’d just visited. What the hell?

And then I cupped it.

It scored an 86—which is all nice and well, but—having come up in the industry of the early 2010s, a thought crossed my mind: Is this really a Castillo? It tasted more like what I expected from Variedad Colombia or Caturra. It was sweet, balanced, and clean with notes of panela and citrus notes and a long sweet-savory finish. 

What the hell.

I remembered the CoE auction from 2010 and the CRS study and pulled the cupping notes I’d been forwarded from the cupping at Intelly that I’d missed. In their Castillo vs. Caturra trials, the panel of cuppers found statistically no difference in quality score between the two cultivars. (They did, however, find qualitative differences).

What’s going on here? How could something be a Castillo but not taste like a Castillo?

A couple of years ago, in researching for a follow-up on his book on Ethiopian cultivars, Tim Hill came across some research that seemed to point me in the right direction to explaining this phenomenon. Among coffee buyers working in Kenya, SL-28 and SL-34 are the favored cultivars, while Batian and Ruiru 11 are widely regarded as inferior. 

But what if not every Ruiru 11 or Batían are created equal?

Take Ruiru 11, which was bred by researchers at the Coffee Research Foundation (CRF) and introduced in 1985 . While regarded as a singular cultivar by buyers, it “is made up of at least 66 different F1 siblings with slightly different parentage.” It’s not just one thing. (African Journal of Food Science Vol. 6(18) pp. 456-464, 26 September, 2012)

Further, as an F1 hybrid, it is the first generation from distinct parents and thus its genes are unstable. Because of this, growers aren’t able to use the seeds from their Ruiru 11 trees to plant nurseries as the genes of the offspring would be distinct from the tree they’re coming from. However, this is not known by some growers in Kenya, and has since been an issue.

Batian, too, is regarded as one cultivar offering inferior cup quality. In reality, it’s at least three distinct cultivars of different parentage featuring the best male offspring of the Ruiru 11 trees and an SL28. According to Tim they are, specifically:

And though unlike Ruiru 11, Batian is considered genetically stable and offspring can be produced from seed, each individual type presents differently in the cup.

In other words: when we’re dealing with a hybrid, maybe there’s more to it than a name. Maybe it’s not just that Castillo is equal to Caturra, as CRS data shows, but that it depends on which subtype we’re evaluating because, like Batian and Ruiru 11, Castillo isn’t just one cultivar: it’s a series of cultivars with a divergent set of quality attributes.

[Aside: In my super humble opinion the decline in cup quality from Kenya seen over the last few years—often attributed to the presence of Batian or Ruiru 11 (Roast Magazine Mar/Apr 2021 page 89)—is actually a result of changes to processing protocols in recent years, such as abandoning a post-wash soak often colloquially referred to as “secondary fermentation” as well as changing drying protocols. Many of the “traditional” methods of processing in Kenya were a result of a need to handle more coffee than factories were built for. The second soak before drying was a way to hold coffee post-fermentation for longer in order to gain enough room on the patios or beds for drying. By holding it under water, spoilage was decreased—but also in a way that changed the flavor profile by favoring acetobacter activity in the highly-acidic, alcoholic interstitial liquid around each seed but diluting the resulting metabolites and distributing it uniformly into the seeds. Similarly, the “drying pens” where coffee was moved from patios after reaching ~16% moisture to make room for freshly-soaked, very-wet coffee are no longer necessary as there is sufficient capacity—but those pens gave coffee a chance to stabilize before further drying. Both of these practices, from what I’ve observed across Kenya, have been largely abandoned or abridged—to the detriment of quality—as exploding development around Nairobi and the continued demand for real estate and tea have reduced coffee production nationally]

The cultivar we call “Castillo” could be one of seven different distinct cultivars which have been bred from Caturra and Timor but with different variations to give each adaptability to particular climates around the country. 

If parsed further, would CRS data show differences in cupping scores between Castillo la Trinidad (adapted to Tolima), or El Rosario (Antioquia), or Paraguaicito (Quindio)? Or Castillo Naranjal, Pueblo Bello, Santa Barbara or El Tambo?

What if not all Castillo are created equal?

Unfortunately, the cupping sheet I have is still blinded—but perhaps Michael or someone else is willing to give me a key to extract a bit more granularity from this data.  [The summary sheet from that cupping in 2014 is available here.]

Part of what makes Castillo a resilient, rust-resistant cultivar is its genetic diversity. In the words of coffee breeding scientist Hernando Cortina, “We need to keep different lines in anticipation and preparation for the evolution of the disease.”

It’s no papaya.

Some 16 years after its introduction across Colombia, Castillo still resistant to roya. Plant breeders and organizations who spend double digit-percentages from their budgets on breeding programs, view Castillo as a success for this reason and argue that breeding programs can be one of the most effective tools in the fight against roya.

But if not all Castillo are created equal, what good does that do if a producer is doomed to either succumb to rust or succumb to the fate of a lower-cupping regional variant?

Like papaya, though, commercially cultivated coffee has relatively little genetic diversity outside of Ethiopia, where Caffea arabica evolved and grew in mixed forestry settings. Plant breeders and programs such as those conducted by World Coffee Research argue that breeding can create greater genetic diversity in coffee, thereby securing the future of coffee cultivation. The history of plant breeding, however, suggests that this effort is a bit more complicated than merely producing a variety with desirable traits using selection of parent trees—and may, in fact, result in unintended genetic bottlenecks through lack of adoption by farmers, market failure, eradication of local landraces, or other unintended consequences such as the emergence or mutation of disease.

Only in the long term does breeding seem to generate additional genetic space. The decentralized nature of plant breeding in the coffee industry—an approach baked into WCR’s strategymay in fact lead to further genetic bottlenecks, creating unintended vulnerabilities until these bottlenecks are resolved through further breeding:

Both the decentralization and the disproportionate inbred use reduce effective population size and constrain the accessible genetic space. Under these conditions, long term response to selection is not expected to be optimal under the classical infinitesimal model of quantitative genetics.

Further decentralizing this work, many producing countries operate their own research centers and government programs to support coffee research, like Colombia’s Cenicafé: Brazil (like Instituto Biológico de São Paulo and Agronomic Institute of Campinas), El Salvador (Coffee Research Institute of El Salvador) Guatemala (Anacafé), Honduras (IHCAFE), and India (The Central Coffee Research Institute), to name a few. There is a lot of money and lots of advanced degrees behind agronomic research for the coffee industry, with much of the focus on breeding for climate change hardiness and disease resistance.

The limited genetic diversity present in coffee today is, of course, a result of selective breeding and monoculturing in the first place—this cycle of needing further technological interventions to address solutions of the past is an inherent problem of technological fixes.

Breeding programs can be an essential tool for protecting the future of coffee, just as they have increased the productivity, disease resistance and genetic diversity of other crops (as well as their nutrition), but they’re far from a silver bullet for the problems facing coffee farms today—problems up to and including threat of extinction. 

Even as these programs command headlines and the attention of much of the coffee world, their resource requirements and longitudinal approach are often lost in the misdirection and excitement of novel technology.

WCR, the preeminent non-governmental organization for coffee industry research in the Western Hemisphere, operates across Latin America conducting field work with its breeding program supported by researchers at coffee origin as well as from CIRAD in France, Texas A&M, and at their research center in Santa Ana. WCR’s research is funded through a mix of grants and industry donations, both through direct giving as well as a one-to-ten penny per pound fee that roasters and importers can pay to earmark money for WCR’s research (you may be paying this without even knowing it, depending where you buy spot coffee). Through this fairly brilliant mechanism, WCR has effectively consolidated the attention of the consuming countries to itself and its own research efforts.

Famously, WCR received funding through USAID during the Obama era, announcing a $7 million initiative to create “a public/private model for a pipeline of pest-resistant, higher-yielding and higher-quality varieties.” You read that right: the future of coffee is still a future away.

Among WCR’s largest donors and board members are the most recognizable names in coffee: Dunkin’, Smucker’s, Keurig Dr. Pepper, Peet’s, and S&D. Clearly: Big Coffee cares about the future viability of coffee cultivation, to the tune of millions of dollars per year (in their 2019 annual report, WCR disclosed $3,058,433 of coffee industry donations, out of $3.6m in revenues). [Notably, these are all predominantly United States corporations run by white men, which colors WCR’s claims and appeals of “science to save coffee” with a tinge of white saviorism]

Because their reported expense categorizations changed in 2019, it’s difficult to parse exactly what went where, but we can infer from the written report that most of the expenses were in support of the breeding research—roughly $3.1 million dollars, going toward research that will ostensibly benefit all coffee producers, but aggressively pursues expensive, complicated, longitudinal solutions rather than tactics that are immediately actionable, scalable, and affordable and putting that $3 million per year in the hands of the people who produce coffee—leaving lower-hanging fruit untouched, so to speak. 

Even in a modern context, where the cost of genetic sequencing has decreased as speed has increased and international collaboration is more possible than ever, it’s difficult to speed up breeding programs. The much-heralded F1 hybrids previewed by WCR are just that: first generation releases, unstable and, like Ruiru 11, not all able to be replicated by seed (which is, as discussed, problematic). The speed of breeding is only as fast as the next generation can bear fruit—in the case of coffee trees, anywhere from 2-4 years. In other words, if a selection is successful, to successfully breed to a stable, 5th generation cultivar, it is likely to take a minimum of 15 years.

Development of Castillo, for example, took 23 years, funded by a government organization subsisting on a tax on coffee exports and overseen by government scientists building on the work Cenicafé did in developing variedad Colombia

Well: a lot can happen in 23 years—or fewer.

In 2017, an outbreak of Hemileia vastatrix race II, the most-prevalent and aggressive race of roya seen in Central America, hit Honduras—a country that just 5 years prior experienced devastation as a result of roya. In response to that outbreak in the early 2010s, many farmers planted varieties that WCR advised were highly resistant to roya, such as Lempira, Parainema, and IHCAFE 90. Lempira, first released in Honduras in 1998, like Castillo was bred from selections of Timor (Timor Hybrid 832/1) and Caturra. In 2017, Lempira accounted for 42% of all coffee trees in Honduras. 

After overcoming Lempira’s resistance to rust, the new infection spread to over 50% of Honduras’ coffee growing regions within months, owing to the shorter cycle of this strain of rust.

Hemileia vastatrix isn’t a genetic monolith. There are currently known to be 40 different races of coffee leaf rust, and no cultivar of coffee is known to be resistant to every strain of roya. Over time, even resistant cultivars can lose that resistance as roya mutates, just as it appears likely that it may have in Honduras in 2017.

What happens if, during the 20-year development cycle of a new super-cultivar, rust mutates again—or a new strain migrates from another growing region—effectively rendering that future F5 progeny obsolete?

Climate modeling suggests that in that time, average monthly temperatures in Colombia will rise by anywhere from 0.6-2ºC, with precipitation in the coffee lands increasing during the rainy season (which will itself extend longer) and increasing the likelihood of drought during the dry season. As climate change leaves previously cooler high-altitude plantings unprotected from roya—and as climate change stressors and plant breeding selection practices may directly lead to the evolution of stronger fungal pathogens—how can coffee producers respond to this threat?

How could these resources have been better distributed and efforts diversified? If breeding is a 20-year moonshot requiring massive resources, what strategies exist for the present?

“The foundation to plant health starts at the ground level.  Farms must have healthy, fertile soils to produce health coffee plant” 

At the cupping session at SCA 2019 in Boston, alongside my bespoke-processed Castillo, we presented a series of three coffees from a farm in Guatemala provided by Sam Knowlton of soilsymbiotics. On their website, soilsymbiotics writes:  “Our work translates cutting edge science into an actionable, principle based program that produces higher quality crops, increased yields, and economic vitality,” a claim that sounds indistinguishable from breeding approaches such as those of WCR, up and until its final clause: “all the while regenerating soil.”

According to Sam, the central goals sought by technology companies such as WCR—disease resistance, quality improvements, productivity enhancements—can be achieved on a near-short term basis by improving the overall health and microbiology of the entire soil/plant system.

WCR, to its credit, published a roya prevention guide (in Spanish) for coffee producers, and includes a section on the connection between plant nutrition and disease resistance (though it is rife with issues and lacks important context). The guide also recommends conducting soil analysis at least every two years to determine the level of macro- and micronutrients in the soil as well as their uptake in the plant, particularly in its leaves. Once the ratios and quantities of nutrients are known, a fertilizer regimen can be implemented to remediate any issues.

In practice, though, many coffee farms implement a practice of fertilizing based on nutrient removal rate, predicated on the belief that if a tree is removing nutrients from the soil, it’s important to replenish those nutrients, which is often done in a 1:1 linear manner. However, in practice, not all of the nutrients added to soil are immediately available to the plant, nor does this method take into account the baseline level of nutrients already in the soil, or mineral synergies and antagonisms. Further, plants rely on the microorganisms and fungi in the soil to process, refine and synthesize compounds in order to be able to absorb them through a process of biological nutrient cycling or immobiliztion/mineralization. A lack of concencentration of NP or K in the leaves might indicate a need for supplemental NP or K in the soil—it might, however, also indicate an imbalance in the soil that needs to be addressed, regardless of the concentration of those three minerals in the soil. In other words, while it has been established that higher concentrations of Nitrogen and Potassium in the leaves is correlated with lower frequency of rust, it may require more work than simply adding Nitrogen or Potassium to the soil to correct the imbalance.

These relationships are complicated but important to manage, particularly with regard to disease pressures. In response to my questions about WCR’s roya manual, Sam wrote:

It’s well known that excess N, especially in the form of nitrate, is the cause of increased disease pressure. Excess K antagonizes the uptake of Ca which is responsible for building cell wall integrity, among other disease and pest resistant qualities. There are plenty of studies and literature showing the influence of specific nutrients in suppressing fungal diseases. Zinc, Manganese, Silica, Calcium, etc.

We have done some trials with foliar applications of these nutrients versus bordeaux mix and triazole fungicides and had better and at least comparable results with the nutritional approach, plus benefits to overall plant health.

The coffees Sam presented challenged cuppers to question their notions of the importance of cultivar and altitude as well as the role of soil microbiology and farm management in coffee flavor. On the table was a Sarchimor—one of the more common rust-resistant cultivars planted in Guatemala and one renowned for its vegetal, astringent flavor, devil-tail and all—grown at a mere 850 meters, far below the expected altitude for quality specialty coffees. It cupped an 84—higher than expected, and as high as the reference sister lot of the Castillo that George Howell liked just a few cups away, which was grown at twice the altitude, 1700 meters.

The farm where this Sarchmor was grown had, a few years earlier, reached out to soilsymbiotics to address some of the challenges it faced, challenges common to coffee farmers in Latin America: roya and other diseases, declining productivity and declining quality. On his initial visit, Sam took stock of what he saw. The soil was depleted and years of synthetic fertilizers and pesticides had thrown the soil’s chemistry, microbiology, and mycelium out of balance. Unsurprisingly, Roya and ojo de gallo flourished on the farm, which had both traditional as well as rust-resistant cultivars planted anywhere from 800 meters above sea level to over 1700 at its highest. By his estimate, 60% of the trees were impacted by roya, and the farm next door had roya on 70% of its coffee trees. 

Laborers walked through the fields, spraying fungicides from backpack sprayers, wearing only plastic vests for protection that left them doused in chemicals at the end of each day through the entire harvest—the unaccounted for human and environmental toll concealed by an expensive and ostensibly necessary line item. WCR had come to the farm a few years earlier, provided some protocols and limited agronomic training, but left and never returned. No follow-ups, no inquiries.

Things looked bleak.

Decades of traditional farming practices led to a situation common to coffee farms—the highly leachable tropical soils had imbalances that were addressed by the use of narrowly-targeted synthetic fertilizers while weeds, other species of plants, and fungi were knocked back by the application of pesticides as coffee grew densely in a commercial monoculture with inadequate, inappropriate, or insufficient wind breaks, shade, or diversity. The landscape was packed with row after row of coffee trees—a single, brittle source of revenue for the farm owner, and a far cry from the forests from which coffee evolved and traveled from.

This is the farm that breeders have in mind when they make their selections.

These conditions, consequently, left the soil prone to erosion, reduced its ability to hold water, and, of course, weakened the plants’ ability to combat disease and produce higher yields and higher quality fruit. Without intervention, the trees would not produce as expected—directly as a result of previous interventions. So, intervention occurred, further depleting the soils and throwing the system out of balance. 

It’s a cycle of a cycle: the trap of a technological fix. It makes money for everyone but the farmer.

The benefit of mixed-forestry cultivation on both coffee berry borer populations and roya dispersal has been shown in the literature—and yet, this practice is often not implemented on larger estates, which may face devastating collapse as a result of borer beetles or roya. Often, these estates will secure their Rainforest Alliance certification by preserving the requisite hectares of forest at the fringes of their farms, or in areas not hospitable to coffee cultivation owing to grade or altitude. This hardly serves as an integrated solution for combating disease or pests. 

So soilsymbiotics set to work with the farm management to devised a plan for the farm, building an integrated system to address imbalances in the soil, increase the diversity of the coffee plants, co-plant coffee with other cash crops, and develop forests on the land—including the planting of hardwood trees whose harvests every 20 or 30 years would exceed the totality of revenues generated by coffee in the commensurate period. They immediately halted application of fungicides—applied to keep roya in check, but which would make it impossible to verify the results of any of the new measures, and which contributed to a cycle of soil depletion and over-fertilization. Replacing it, they used foliar sprays and organic soil inputs highly customized to match specific soil and plant needs as well as the physiological development of coffee trees and restore the balance of micro- and macro-nutrients and microorganisms in the soil. Some of these products were created from the farm’s byproducts.

Between the rows of trees, life erupted where pesticides previously kept undergrowth in check, holding soil and water in place and helping to control populations of harmful pests and, within one year, while the neighboring farm saw the number of trees impacted by roya soar to 90%, the number of trees on this farm impacted by roya plummeted to merely 20%.

All without fungicides or pesticides—and without the aid of brand-new resistant cultivars. And, the new intercropping introduced created additional economic stability while contributing to the soil—in years that coffee was damaged by storms, or that market volatility suppressed the price, diversification in revenues ensured prosperity. Even if the c-market drove coffee prices below profitability, the farm could continue to grow coffee and remain profitable. 

The farm worked as a system—symbiotically, together, and giving back to the earth in a productive way rather than merely extracting value from it with diminishing returns over time. That’s the thing about WCR’s research—it doesn’t change the way that farming is conducted, or that farms interact with the world, markets, capitalism, currency exchanges, or agricultural supply companies. It merely attempts to supplant one tree for another so that the true beneficiaries of that work—those titans of industry and capitalist enterprises from consuming countries—can continue to operate as they were, can continue to consume. Issues of colonialism, imperialism or equity aren’t addressed; the reasons why producers lack the resources to address roya and climate change aren’t discussed, nor is that condition raised. 

Trees are merely replaced with new trees.

It—quite literally—misses the forest for a tree.

Food stability shouldn’t rely on luck or good fortune; it shouldn’t rely on roya’s failure to, thus far, encroach on a given set of coordinates. All over the coffee belt, monocultures of trees selected for the taste of roasters abroad, engineered to produce as much coffee as possible, lie in wait—trees dosed to perform, their weakness and vulnerability masked by designer drugs of nitrogen and fertilizer.

While we wait for WCR and its gilded silver bullets to come to fruition, there’s a lot of work that can be done to ensure coffee’s resiliency and protect its viability—if we’d only pay attention, and if we’d only give as much support and resources to soil work as we do to breeding or studying how bad coffee tastes when it’s brewed to equal strength but with cooler water.

Who cares about supertrees if we destroy the world before these supertrees can even be planted? And if coffee were produced in integrated, diversified farming systems globally, would we even need supertrees?

If Typica and Tabi can’t exist in the new climate-changed world of the 2020s, I’ll be thinking about the Sarchimor from that cupping table in Boston, grown at 850 meters in Guatemala in an integrated farming system planned around soil regeneration.

Tail of the devil, be damned.

[UPDATE: On April 29, 2021, after I first published this piece, Reuters reported that “Colombia’s coffee federation has discovered nine new, more aggressive variants of the fungus that causes coffee rust.” It’s as of yet unknown whether any of these new variants—or which—can affect the resistant cultivars planted in Colombia, but the government-sponsored race to replant is, once again, on. As Reuters reports: ‘”I call on coffee producers to build or renew their crops with resistant varieties, using certified seeds,” the federation’s director general, Roberto Velez, said in a statement.’ Resistant varieties account for over 84% of all coffee plants in Colombia.]

[UPDATE 2 – May 10, 2021: This is a problem and a conversation in the banana industry, too, as Tropical Race 4, the fungal strain that causes Fusarium wilt, has spread to Latin American and threatened the industry. Banana, like coffee, is grown in traditional monocultures. Slate published a solid piece, addressing the problem and overviewing proposed technological fixes and their pitfalls.]

[UPDATE 3 – May 30, 2021: At some point, these updates warrant their own post, but for now: peer-reviewed evidence now exists showing that one effect of shade-grown coffee—in addition to creating favorable microclimates and healthier soil via polycultureis a reduction in coffee leaf rust]

My dear friend and immeasurably talented colleague Kathy Altamirano sent me an article some months ago about coffee’s future and lack of resiliency in the face of climate change. That piece inspired much of this entry.


That's just, like, your opinion, man

  1. En mi humilde opinión el mal se origina al eliminar la biodiversidad de los microorganismos del suelo con la aplicación de compuestos químicos. Mientras más nos acerquemos a cultivar el café en un ambiente de diversidad de especies con madera les aromáticos como membrillo, cedro, los ficus llamados laurel en Barahona los frutales como cítricos naranja amarga o agria, China o naranja dulce, toronja y muchas especies más que influyen en el suelo aportando buenas condiciones para el desarrollo de la diversidad microbiological con hongos antagónicos que atacan a los invasores, el desarrollo de micorrizas etc. En Colombia un agricultor ideó el sistema de cultivo al que llamó sancocho que consiste en una diversidad de especies plantadas como maiz, guineo, platano, habichuela, cítricos etc con resultados alagado res en cada especie plantada comparada con los rendimientos obtenidos en monocultivo de esas especies. Demostrando que el suelo se enriquece y aporta a las plantas una mejor combinación, calidad, y cantidad de elementos nutritivos para su desarrollo además de un ambiente sano porque las es ias susceptibles al ataque de una plaga era protegida por la capacidad de repeler o dañar a la plaga que le brindaba otra especie plantada. El aporte dde mantillo hojarasca y suelo superficial de bosque brinda a los suelos y cultivos gran capacidad de protección contra agentes dañinos aportada por microorganismos existentes en ese mantillo.
    Entiendo que debemos tomar estas lecciones en serio y tratar de hacer una Agricultura más integral y menos agredida por químicos y lograremos recuperar esa riqueza biológica que mantiene al suelo vivo porque el suelo es medio vivo de una riqueza inmensa en diversidad biológica microbiana

    1. Gracias para leer y por tu comentario. Claro, restaurar el equilibrio del bosque y establecer la biodiversidad es muy importante para el futuro del café y la prosperidad de los caficultores en todo el mundo. Las plantas, los hongos y los animales trabajan juntos como un sistema. Me gustaría mucho aprender más sobre este sistema de sancocho!

  2. Thanks for a great, thought provoking piece Chris.

    I’m wondering whether you intended to cast the plant-breeding/WCR work and the soil regeneration/economic restructuring work in the somewhat zero-sum light that is presented here?

    The power and resources of WCR feels somewhat overstated in this piece. Consider the calamitous future coffee faces in the wake of climate changes, roughly 3 million dollars of coffee-industry contributions per annum seems rather paltry. And to argue that this money should be directed towards short-term, easily actionable initiatives seems misguided–somebody has to be doing the hard, longitudinal science work, right? That’s not to say that there doesn’t need to be more money injected into initiatives that make tangible improvements today, but I’m not sure slicing up WCR’s pie is the right approach. 

In terms of where WCR focus their energies: I hear what you’re saying in terms of varieties being bred to thrive using traditional farming practices, however that’s not to say that these varieties will perform suboptimally in healthier soils and environments (though I’m definitely not qualified to have an opinion here). But, for pragmatic/funding reasons, I can see why the breeding programs proceed in this fashion.

    I also don’t see how the WCR approach is in any sort of conceptual conflict with soil regeneration, improving biodiversity, or economic restructuring. The fact is, most coffee varieties will not be happy as things heat up (to name one threat to coffee), and that’s a problem that breeding programs are arguably best positioned to fix. I’m not sure that it’s wise to leave our eggs in the sarchimor/catimor baskets?

    Last point: I don’t believe it’s accurate to say that selective breeding has led to coffee’s limited genetic diversity. This is predominantly due to arabica’s youthfulness and their slow reproduction rate. Selective breeding has in fact expanded arabica’s genetic diversity over its 10-20,000 years of existence (see Scalabrin et al 2020). There is limited genetic diversity in varieties planted outside of Ethiopia, sure, but there are a multitude of reasons for this.

    1. Hi Tuli,

      Thank you for taking the time to read as well as for your thoughtful reply.

      One note before I reply: I do prefer “Christopher” (I like all 11 letters. As you might suspect, I don’t understand making things short simply to make them shorter).

      I present the Sarchimor as a device in this story to obliquely address two heuristics in the minds of coffee buyers (1) that Timor hybrids are inherently inferior, and (2) that coffees grown at lower altitudes are inherently inferior. The larger takeaway for me is that when grown in more fertile conditions, we have the potential to unlock greater genetic potentials for quality than we’ve observed, particularly as many of the secondary metabolites produced in coffee plants require nutrition unavailable across many current farm conditions. It’s also important to observe that this is in many ways a test of climate resiliency — the temperature difference between 1600 MASL and 850 MASL in this scenario was >3ºC—in other words, hotter than the expected increase in temperatures over the next decade as a result of climate change. Coupled with disease resiliency, improved erosion and water control, and appropriate shade, this demonstrates that mixed forestry cultivation of coffee (much more similar to where coffee evolved—in that single polyploidization event at the origin of the tetraploid genome of Coffea arabica—in the forests of Ethiopia) is in fact a pursuable answer to climate change.

      I hadn’t seen the journal article you shared—I enjoyed it very much, thank you. It’s fascinating to read that much of what we understand about coffee’s spread is verified by the genetic record—that there is very limited diversity in the landraces and accessions spread through the world for cultivation over the recent centuries and that the native species in Ethiopia possess greater (“untapped”) diversity. It’s distressing to read that “regional human overpopulation appears to be the main cause of accelerated destruction of the biodiversity in the montane forests of southwestern Ethiopia” (though my own observation is more mixed—it seems to have to do with land use, not just population growth) but seems to imply in that very statement that mixed forestry settings—such as those being implemented now in Kaffa by the government to protect native species, forests, and biodiveristy—are one way to protect and grow genetic diversity. There is some recent study that seems to support this notion. It’s also distressing to read the amount of genetic material available is quite limited, and equally distressing to read that “many of the traditionally used approaches in plant breeding, trait mapping and gene isolation less efficient and the development of novel alternative approaches a definitive must.” This, of course, goes directly to the allusions to GM as well as the traditional approaches used by organizations such as WCR.

      WCR, of course, isn’t the only group working on plant breeding—I noted a half dozen agencies who also are working on this. They are, though, the only organization quite visibly and vocally represented in our industry and the only one based in a consuming country. I don’t meant to present it as a zero sum—but rather assert that theirs is a longitudinal approach when we need something with greater immediacy. My point is this: Climate change is already occurring; we can’t wait 10 to 15 years. $3 million in the broader scheme may not be much money, but when that $3 million represents the totality of consuming countries’ efforts to combat and mitigate the impacts of climate pressures, it’s problematic that the discourse is so heavily directed by an organization with many ties to SCA and seeks to solve a problem with a technological fix that will later necessitate a further technological fix.

      Thank you again for your kind and thoughtful reply!

      1. Tuli’s follow-up comment got caught in the Spam filter (now disabled). I’ve reposted it here.


        Hi Christopher,

        First up, apologies for the inadvertent name-shortening. Unlike you, I’m quite lazy and shorten things by impulse.

        Thank you for your very thoughtful and detailed reply. I think that your vignettes about Sarchimor, Castillo, and Ruiru 11 were indeed very effective for discussing the negative biases against timor hybrids and lower altitude coffees. You also make a great point regarding Sarchimor’s temperature resilience (when grown in a healthy soil ecosystem, among biodiversity, etc).

        I also found the Scalabrin article fascinating, so am glad you did also. In fact, I found it so interesting that I recorded a podcast with the senior author, Prof. Michele Morgante, a few weeks ago in which we discussed the paper and arabica’s evolutionary lineage (you can find it if you search “Decoding the Coffee Genome” in a podcast app). And thanks for sharing the Schuit et al. (2021) paper–I wasn’t aware that Aaron Davis’ team have been so prolific. Loads to read.

        I agree wholeheartedly with your summary as presented in the final paragraph of your reply. Plant breeding is not a panacea, but given arabica’s low genetic diversity (among the cultivated landraces in particular) and vulnerability to pests and climate shifts, fortifying the genetic stock is essential work. You are right that WCR have been very effective at holding the spotlight, but I don’t think that’s any sort of indictment on them or their vision. In an ideal world, there will be greater experimentation with and investment in more ecologically harmonious farming practices as well as greater investment in breeding programs.

        But I see your point: the plant-breeding solution narrative may well be crowding the industry’s mind at the expense of immediately actionable initiatives. It’s great and important that you are banging the drum on this, and my intention with the previous comment was merely to lightly push back in order to reinforce the importance of building up a diverse genetic arsenal.

  3. Christopher, thank you for this thought provoking and lucid piece, which I greatly enjoyed. I would love to engage in a discussion of some of the challenges to such a wholesale conversion to symbiotic interplanting to better contextualice this approach in the toolbox of agronomic choices producers may take. For example, there is a very real opportunity cost and cash outlay in interplanting hardwoods or other crops with very long time horizons. And while it’s great for coffee farmers to diversify cash crops, the reality of investing the upfront time and materials to grow other fruits and vegetables presents an insurmountable barrier to many of the most vulnerable producers.

    Overall I wholeheartedly agree that coffee resiliency should be thought of holistically rather than only in terms of varietals and fertilizers. I just think that actually transitioning a farm to a comprehensive, integrated, symbiotic ecosystem is a lot harder than it sounds (and it sounds very hard!). Perhaps a bit of a discussion of the barriers to implementing soil symbiosis could help to clarify the costs and benefits.


    1. Hey Alex! Thanks for reading and for your contribution here.

      I’m not going to lie—this is difficult work. It takes professionals like soilsymbiotics to guide this process. I’ve seen firsthand many producers attempt this work on their own with mixed or even disastrous results. It’s not enough to simply apply a compost tea—you also need a balance of mineral inputs, different pruning methods, revisit plantings and shade trees, strategies for control of pests, water use, etc. etc. It’s a lot. There are success stores in regenerative ag in other industries—it just seems that coffee is a bit far behind (unsurprisingly based on the fact that it’s still largely a colonialist enterprise—we consuming countries appear, extract value, and if we can’t get what we want at the price we want, move on to somewhere else that will). In Brazil, I know FAF is exploring regenerative agriculture with their farms (you might want to reach out to Felipe Croce about their work) and I know of a few producers in Colombia who have quite beautiful mixed forestry grown coffee.

      It requires training, resources and attention—and I think perhaps because it’s less sexy than using supercomputers to unravel the genomics of coffee, this work tends to get overlooked. I know Sam is working (slowly) on a field manual for coffee producers that he’d make available for free—kind of like WCR’s roya manual, but with metrics and methods for soil regeneration and field renovation. It’s resources like that which would help many, if our industry saw fit to produce and promote them.

      For many producers the cost of *any* change may be high—but we’re approaching a point where growers can’t afford NOT to change. With the c-market hovering below cost of production and domestic prices fluctuating wildly and climate and disease pressures—growing coffee as it has been traditionally done over the last century may no longer be viable.

      Sam’s told me of some of his renovation work on farms—it can be done in stages so as not to disrupt production overall. Let one sector regenerate while selectively treating other sectors. And rotate. It takes time, to be sure, to develop the system fully—but the impacts can be immediate on the areas where these efforts are applied, even with existing plantings.

  4. A comment on papayas (and bananas). I understand the statement could attempt to generalize commercial banana and papaya production, but to say the invention of a GM papaya saved papayas (not just in Hawaii) in general is somewhat outrageous. Like bananas, there might be a handful of flavored varieties for industrial cultivation, but in practice there exist so many other kinds, and these continue to evolve and be unaffected by disease in their diverse settings. In fact, native varieties where I live now were completely unaffected by Fusarium (in fact, only hit intensive plantations). Papayas pop up at many elevations. I doubt that the non-invention of the GM papaya would have led to its total demise.

    1. Hey there, thanks for your thoughtful reply.

      I didn’t really speak on bananas—but as you imply they are grown in commercial mono cultures and suffer from limited genetic diversity.

      The issue of the threat faced by papaya and the papaya industry, though, has been reported far and wide by various outlets. I provide a bit more context in the piece itself, which of course is not actually about papaya but uses it as a vehicle to discuss the nature of technological fixes (and while ambivalent, pretty firmly aligns with your implied sentiment that there’s a better way). While saying papaya wouldn’t exist anywhere in the world and would be extinct would be a bit alarmist, so perhaps more accurate would be to say that the industry faced extinction. But: If the industrial supply of papaya were to collapse and industry to fail, papaya would likely not be available outside of places of origin such as where you live. That would not be sufficient supply to satisfy an international market (which of course is how we got ourselves into this situation in the first place—selection of specific trees for specific attributes of flavor, presentation and shelf stability resulting in a singularly vulnerable population). Shelves where I am in the U.S. would be barren——it would be tantamount to papaya not existing.

      In coffee, if roya were to exterminate Arabica coffee globally outside of its native home of Ethiopia, we’d face a similar predicament.

      1. I understand that the papaya issue has been reported far and wide by news outlets in your country, and that you are just using it as a tool to discuss coffee. And yes it is more accurate to say that papaya would likely not be available outside of places of origin. I’m not sure about others’ sensibilities, but for me that would definitely not be tantamount to papaya not existing, and it is important to state that. I feel this is an important point to make, and it is important not to pin the survival of the plant itself––you may find it pedantic, but in your article you do acknowledge that the tree has an active life outside of industry and export. The “surviving populations” always hold the key to moving the industry forward anyway.

        Just nature’s way of keeping monocultures in check, right? As you touched upon, not all Arabicas are affected if their conditions are biodiverse. On a related note, Liberica had its moment here in the 1800s because of rust, and allowed continued export but also birthed a Liberica-loving zone that (in a way) persists and is culturally important til today.

        PS- my reference to bananas was in response to the note at the bottom of your post

  5. This is also a very good post which I really enjoyed reading. It is not every day that I have the possibility to see something like this..

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.

Back to top