Genotypes shape nutrient export in canephora

Nutrient export is a key cost and sustainability factor in Coffea canephora production. With every harvest, nutrients are permanently removed from the soil because they are allocated to fruits and leave the system with the crop. The magnitude of this export depends not only on site conditions, yield, and management. The evidence increasingly suggests that genetics also plays a role. An open-access study from Brazil addresses exactly this point: 42 Conilon clones (a Brazilian C. canephora group) were compared at the same site across two harvest years to quantify how strongly genotypes differ in nutrient export and whether these patterns are stable enough to be treated as a physiological trait (Silva et al., 2025).

The authors planted 42 clonally propagated Conilon genotypes in the same field and followed them across two consecutive harvests. They harvested ripe cherries. The fruit material was dried, and macro- and micronutrients in the dried cherries were then determined analytically. Using dry mass and nutrient concentration, they calculated how much of each element is removed from the system with the harvest.

What matters for interpretation is that the results were standardized to a defined production amount, meaning the nutrient export per tonne of coffee produced. In that sense, the study primarily describes a genotype’s export intensity. It does not automatically indicate whether a genotype is agronomically “better” overall.

Across all genotypes, the data show a clear ranking of the elements most strongly exported with harvest. Among macronutrients, nitrogen and potassium dominate, followed by calcium and magnesium. Sulfur and phosphorus are lower. Among micronutrients, manganese and iron typically contribute more than boron, copper, and zinc. In the simplified order reflected by the study’s data, this gives: N > K > Ca > Mg > S > P > Mn > Fe > B > Cu > Zn (Silva et al., 2025).

For practice, this ranking is not a minor detail. It helps explain why harvest-driven nutrient management in Coffea canephora is often discussed primarily in terms of N and K, while micronutrients tend to become visible in the cost picture only once deficiencies or imbalances emerge.

The central finding of the study is not the ranking itself, but the spread across genotypes. In the cluster analysis, many clones behave similarly, but some deviate systematically. These deviations are not isolated. They affect entire nutrient export patterns. Some genotypes export clearly more than the population mean, while others export clearly less. At the extremes, the differences are roughly on the order of about 30–40% higher export and about 30–50% lower export relative to the mean of the studied population (Silva et al., 2025). For a field experiment under the same site conditions, this is a practically relevant signal because it suggests that nutrient flows can be organized in a clone-specific way.

For orientation, three examples are useful because they make different “strategies” visible.

Clementino represents an export-intensive profile. In the data, this genotype stands out through above-average values for key macronutrients, especially nitrogen and potassium. Magnesium is also clearly elevated compared with low-export genotypes. The result is a consistently “high” export pattern across multiple elements (Silva et al., 2025).

LB1 is the counterpart as an export-efficient profile. Here, the exported amounts per tonne of coffee are clearly lower, both for macronutrients and for several micronutrients. In practical terms, under the same conditions less nutrient is removed from the field per defined production unit. This is not a guarantee of higher yield or better quality, but it is a clear signal of a different nutrient economy in this genotype (Silva et al., 2025).

Verdim R is particularly interesting because it shows what “outliers” can look like without immediately standing out on macronutrients. In this profile, macronutrients sit closer to the population mean, while individual micronutrients, especially zinc and also iron, can be clearly elevated. Such patterns matter for monitoring and fertilization strategies because they highlight micronutrient flows that can be easily underestimated when looking only at population averages (Silva et al., 2025).

Many nutrient accumulations are positively correlated. Selection decisions targeting one trait can therefore unintentionally shift other export patterns at the same time. At the same time, Mn and Cu stood out due to a lack of significant correlations, which may indicate a more independent regulation of these elements.

The study shows consistent, clone-specific differences in how nutrients are allocated to fruits and exported with harvest. Because these patterns are structured and repeatable, the findings support treating nutrient export as a physiological trait that is at least partly shaped by genetics. At the same time, the boundary is clear: the work does not directly address how these export patterns relate to yield, resilience, or cup quality. An export-intensive genotype may have high yields or not, and an export-efficient genotype may deliver strong quality or not. The available data alone do not allow those conclusions.

For interpretation, it also matters that standardizing “per tonne” deliberately adopts a specific perspective. It helps compare genotypes by export intensity independent of yield level. For farm decisions, this later needs to be combined with the yield perspective (per hectare), because in practice both dimensions determine total nutrient removal from the field.

If genotypes consistently export different amounts of N, K, Ca, or selected micronutrients per defined production unit, fertilization becomes less of a generic “canephora” question and more of a system question: which clones are planted where, under what baseline soil fertility, with which target profile, and with which nutrient return strategy?

In systems with limited input budgets or weaker soils, it can be attractive, both from a breeding and a farm-management perspective, to treat nutrient efficiency as an explicit selection target. This shifts clone choice away from “yield only” toward “yield in the context of nutrient cycling and long-term soil durability” (Silva et al., 2025; Partelli et al., 2024).

To build a robust bridge toward quality and roasting practice, follow-up data are needed. A sensible next step would be multi-site trials that jointly capture export intensity, yield, leaf and soil analytics, and sensory evaluation. Only then can we test whether specific export profiles are associated with more stable ripening curves, lower stress, or indirectly with quality parameters. Until then, the main takeaway is this: the results provide a strong argument for taking genetics in Coffea canephora seriously as a lever for nutrient management and production system design.

Under the conditions of this study, Conilon genotypes in Coffea canephora differ substantially in how strongly they export nutrients via harvest. These differences are large enough to make operational consequences plausible, while also being structured enough to discuss nutrient export as a clone-specific, genetically shaped trait. Clone choice is therefore not only a question of agronomy and sensory performance, but also of nutrient economy. Practical application starts where export data are integrated with yield, site response, and nutrient cycling strategies.