Nutrient dynamics in Canephora cherry, bean, and parchment
– What the new study means for practice and quality
A new open-access study in Scientific Reports about nutrient dynamics shows when and where which nutrients accumulate in the coffee cherry of Coffea canephora – separated into bean, parchment/fruit husk (husk/pod) and whole cherry. The results provide a scientifically sound basis for controlling fertilization in a more targeted, variety-specific, and environmentally friendly manner. It is particularly important to note that potassium (K) dominates export throughout the entire cherry, while nitrogen (N) plays the main role in the bean, with a critical demand phase of 35–45 weeks after flowering (WAF).
Why this study is important
Coffee is of great economic importance worldwide, and C. canephora contributes significantly to production. In many places, high nutrient requirements are met with nutrient-poor soils, which, due to over- or undersupply, reduce yield and quality and pollute the environment (soil/water). Varieties differ in their ripening process and nutrient uptake, which means that general recommendations are insufficient.
Aim of the study: To precisely quantify dry matter accumulation and macro/micronutrient dynamics (N, P, K, Ca, Mg, S; Cu, Fe, Mn, Zn, B), separated by bean, husk, and whole cherry.
Practical benefits: Knowing when the plant needs which nutrients and where allows fertilizer applications to be split, timed, and dosed for stable yields, better cup quality, and lower losses.
Study design clearly explained
Location & conditions
Tropical climate (Aw, dry winters) on a nutrient-poor Yellow Dystrophic Argisol.
Plant material
Six C. canephora genotypes (Pirata, Bamburral, A1, Clementino, Beira Rio 8, and P1) representing different maturity cycles.
Planting & management
≈5,000 plants per hectare and ~10,000 main stems (orthotropic axes) per hectare; controlled irrigation; N, P₂O₅, and K₂O applied in six split doses over the season.
Sampling
Nine time points from 33 to 49 weeks after flowering (WAF), covering the period from the end of fruit expansion to full ripeness. Manual harvest from pre-marked branches; drying and weighing of whole cherries; separation and individual weighing of beans and husk; calculation of husk mass.
Analytical methods & statistics
- Nutrient determination: Kjeldahl for N; optical emission spectroscopy (OES) for P, S, and B; atomic absorption spectroscopy (AAS) for K, Ca, Mg, Cu, Fe, Mn, and Zn.
- Standardization: Nutrient amounts were standardized to a production of 1,000 kg of beans at 49 weeks after flowering (WAF).
- Statistical analysis: Two-way ANOVA (genotype × time point) to test for significant differences.
Fig. 2. The visual ripening evolution of berries and corresponding dry beans was tracked for six Coffea
canephora genotypes (Pirata, Bamburral, A1, Clementino, Beira Rio 8, and P1) from the end of the berry
expansion phase to full ripening. This process was monitored at 33, 35, 37, 39, 41, 43, 45, 47, and 49 weeks
after flowering (WAF). For each genotype, the upper line represents the visual appearance of freshly collected
berries at each sampling point, showing the gradual change in colour, size, and ripeness from early stages to
full maturity, whereas the lower line displays the corresponding dry beans for each collection date, illustrating
the transformation in bean size, shape, and weight as the ripening process advances. Selvador et al., 2025, pp.6
Key findings
1) Dry matter (DM) — where does the biomass go?
Bean: Rapid DM increase up to ~41 WAF, followed by a plateau/slight decline.
Husk: Delayed DM accumulation, stable until ~41 WAF, then sharp increase with a peak at 45–49 WAF.
Genotypic differences: Beira Rio 8 has the highest DM values; P1 accumulates the latest.
Interpretation: In the late phase, the plant strongly reallocates photoassimilates to the bean; towards the end, respiration increases; the bean’s DM gain levels off, while the pod gains DM in the final stage.
Concentrations: how do nutrient levels shift over ripening?
Bean (trends across ripening):
- Nitrogen and potassium generally decline.
- Magnesium decreases until about 45 WAF.
- Zinc rises sharply and more than doubles between 35 and 37 WAF.
- Boron gradually decreases.
Husk:
- Potassium is very high and increases until about 43 WAF.
- Zinc shows the opposite pattern compared with the bean. It rises in the bean and falls in the husk.
- Iron is consistently higher than the other micronutrients in both the bean and the husk.
3) Accumulated nutrient amounts standardized to 1,000 kg of beans
Bean:
Nitrogen is the most accumulated nutrient and increases across ripening.
Husk:
Potassium is the most accumulated nutrient, peaking around 41 to 45 WAF depending on genotype.
Whole berry:
Export order K > N > Ca > P ≈ Mg > S for macronutrients, and Fe > B > Mn > Cu > Zn for micronutrients.
Critical window:
Nitrogen demand is especially high between 35 and 45 WAF.
What does this mean for fertilization, harvest, and quality?
1) Time fertilization to the crop’s phenology
- Apply N during the bean filling phase (35 to 45 WAF). This is when bean filling and the protein and enzyme balance are decided.
- Provide K over a longer period (to beyond 43 WAF). K supports osmoregulation, sugar transport, enzyme activation, nitrogen and carbon metabolism, and photosynthesis, and it contributes to bean filling and mass.
- Tune Zn management. Rising Zn in the bean while Zn in the husk falls points to remobilization. Avoid both deficiency and excess.
- Keep an eye on Fe. As a dominant micronutrient it can become limiting.
2) Think by genotype
- Beira Rio 8 shows high dry matter, while P1 accumulates late. Adjust harvest windows and split applications by genotype. A single plan for all plants leaves performance on the table.
3) Environment and cost
- Split, demand driven applications reduce losses from leaching and emissions and lower costs, especially on nutrient poor Argisol soils.
Practical guidelines
Think of these guidelines as a clear workflow:
Start from the flowering date. Use it to calculate weeks after flowering (WAF) and plan actions around the key windows. Before the season begins, run soil and leaf analyses so fertilization and corrections are not done blind.
From 35 to 45 WAF, focus on nitrogen. Use several small split applications that support bean filling rather than a single large dose that is easily lost. Keep potassium going longer (beyond 43 WAF) because it supports water balance, sugar transport, and therefore yield and cup structure; here too prefer moderate, repeated applications over a single heavy one.
Track the micronutrients in parallel. Zinc rises in the bean and falls in the husk, which signals remobilization; iron remains especially important. Check leaf values in the 35 to 45 WAF window for N and Zn, and in 41 to 45 WAF for K, and link applications to water availability (derived from the findings: no applications before heavy rain, incorporate into the soil where possible or apply close to the soil surface).
Pay attention to genotype. Late maturing lines shift the windows. Dry matter and nutrient curves are not identical, so time harvest and splits by genotype. Record everything (fertilization, weather, ripeness, yield, screen size and density, and cup quality) and build a feedback loop: review quarterly, correct deviations, and refine rates and timing. This reduces losses and environmental load, saves costs, and turns the study’s physiological insights into robust, quality focused field practice.
Limitations of the study
The results are strong, but they are context dependent. The work was carried out at a single site with an Aw climate (dry winters) on a nutrient poor Yellow Dystrophic Argisol, conditions that do not transfer one to one to other soils, elevations, or rainfall patterns. The genetic scope is also limited: six C. canephora genotypes with different maturity cycles offer valuable insights, but they are not a substitute for a broad varietal evaluation. The time window from 33 to 49 weeks after flowering (WAF) captures the period from the end of fruit expansion to full ripeness, but it leaves earlier developmental stages outside the analysis.
Nutrient management was done under controlled irrigation with predefined split applications of N, P₂O₅, and K₂O; other fertilizer strategies, organic sources, or strictly rainfed systems were not tested. Standardizing nutrient amounts to a production of 1,000 kg of beans at 49 WAF helps comparison, but it is a modeling step that only approximates farm level yields and harvest dates. Finally, the study does not directly link nutrient dynamics to cup quality, yield efficiency, or environmental outcomes. This remains work for follow up, practice oriented and multisite field studies.
Conclusion
Despite these limitations, the study sets a clear marker for practice oriented nutrient management in canephora. Potassium emerges as the lead nutrient across the whole cherry and should be supplied consistently and in line with demand through late ripening. In the bean, nitrogen drives accumulation, with a critical demand window between 35 and 45 WAF, when targeted N splits support bean filling and the potential for yield and cup quality. Measurable differences among genotypes in dry matter and nutrient trajectories argue against blanket recipes. Harvest windows and fertilization should be planned by phenological stage and by genotype. Teams that translate these findings into field calendars, leaf and soil monitoring, and split strategies reduce losses, save resources, and bring the quality potential of canephora closer to the cup. The next step is multi site trials that link the timing shown here directly to cup scores, profitability, and environmental outcomes, so that precise diagnosis turns into consistently better decisions.
Sources
Salvador, H. P., Vieira, H. D., Gontijo, I., Marques, I., Ramalho, J. C., & Partelli, F. L. (2025). Nutrient dynamics in the berry, bean, and husk of six Coffea canephora genotypes throughout fruit maturation. Scientific Reports, 15(1). https://doi.org/10.1038/s41598-025-13171-4
