CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Farm management · Growing

Overbearing and die-back

Why a coffee plant that sets more fruit than its leaves can feed starves its own shoots and roots, so that branches die from the tip down and the next crop fails – and how shade, nutrition and pruning limit it.

Across the chain: Science

Source-checkedby Coffearium verification pass· Last source-checked 23 Oct 2026· Prepared with AI assistance· 11 sources (10 tier A)· 3 contested· Updated 23 Oct 2026· Literature last searched 23 Oct 2026
Reading depth
Contents
  • Die-back is the death of branches from the tip downwards; in coffee it typically follows heavy crops on unshaded trees.1,2
  • Fruit is the priority sink. In a Costa Rican trial, berries received four times the dry matter that went into branch growth.3
  • In a Brazilian field study, about 78% of branches died back below 6 cm² of leaf per fruit, against about 20% above 14 cm².2
  • Die-back deepens alternate bearing; a 2007 review notes that plants may need two or three years to recover.1
  • A 2007 review links shade and nitrogen to less die-back and treats fungi such as *Colletotrichum* and *Phoma* on dying twigs as mostly secondary.1

What die-back is#

Die-back is the death of twigs and branches that starts at the tip and moves downwards.1 The 2007 review discusses it as the descending death of the side (plagiotropic) branches.1 Overbearing means that a plant carries more fruit than its leaves can feed.

The two are linked. Unshaded coffee often bears very heavily. A heavy crop limits how many leaves the plant forms and keeps, which leads to branch die-back.2 As trees enter the reproductive phase with heavy fruit loads, they become more prone to die-back, according to a 2007 review.1

Fungi and bacteria can also kill branch tips. That pathogen die-back is a different problem and is covered further down.1,9

Fruit comes first#

A coffee plant makes its sugars in the leaves. Fruit, new shoots and roots all draw on them. The fruit wins: coffee berries act as priority sinks, as the 2007 review puts it.1 A sink is any part of the plant that uses more sugar than it makes.

In a Costa Rican field trial, berries received four times as much dry matter as branch growth over the season.3 The more fruit a branch carried, the less it grew, and the more often it died back.3 The plant does not seem to control the pull of its fruit finely, the review notes.1

45681237
  1. Mature leaves: the source of sugars for the whole plant.1
  2. Fruit clusters: the priority sink; berries draw sugars and nutrients first.1,3
  3. Shoot tip: new growth, next year’s fruiting wood; it grows less as the crop grows.3,4
  4. Died-back branch: dead from the tip downwards, leaves lost.1,2
  5. Stem reserves: starch stored in the stem and roots, drawn down under a heavy crop.1,4
  6. Roots: in one conilon study cited by the 2007 review, root mass fell by more than half during a heavy crop.1
  7. Flow to the fruit: most of the dry matter goes here.3
  8. Flow to shoots and roots: what is left after the fruit.1,3
Where the sugars of a heavily cropping coffee plant go. Point at a part, its number or its name to highlight it. Schematic, not to scale.

Leaves per fruit#

What matters is how much leaf feeds each fruit. In a Brazilian field study, branches were grouped by leaf area per fruit.2 With less than 6 cm² of leaf per fruit, about 78% of the branches died back. With more than 14 cm², about 20% did.2

The same study saw no die-back the following year, when the crop was small.2 A 1993 Brazilian study found that trees in full sun can carry a very heavy crop without die-back if they keep enough leaf area.1

020406080100% of branches with die-backBelow 6 cm² per fruit(mean 3.5)≈78%2 6.1–14 cm² per fruit(mean 9.4)≈50%2 Above 14 cm² per fruit(mean 23.0)≈20%2
Data table
% of branches with die-backSource
Below 6 cm² per fruit (mean 3.5)≈78%2
6.1–14 cm² per fruit (mean 9.4)≈50%2
Above 14 cm² per fruit (mean 23.0)≈20%2
Share of branches with die-back in three classes of leaf area per fruit, Catuaí Vermelho IAC 99 in full sun at Coimbra, Brazil, 2007. Data: Chaves et al. 2012.2

One heavy year, one light year#

A coffee branch grows new wood and carries fruit in the same year. The two kinds of growth compete.6 After a high crop the trees grow little, so the next crop is smaller.6 This swing between heavy and light crops is called alternate or biennial bearing. Branching and fruiting wood explains why the crop sits on last year’s wood.

Die-back makes the swing stronger. Plants may need two or three years to recover, according to the 2007 review.1 Cenicafé, Colombia’s national coffee research centre, notes that the swing is stronger in ageing plantations.6

Sun, nutrition and water#

According to the 2007 review, shade reduces profuse flowering, overbearing and biennial production, and so branch die-back.1 Branch die-back does not usually arise under natural conditions where coffee grows under shade.1 In a Costa Rican trial, 45% shade cut yield by 18% but reduced alternate bearing.7 Shade and agroforestry covers the trade-offs.

Fruit also takes most of the nutrients. In late fruit growth, berries may take over 95% of the plant’s current uptake of potassium, phosphorus and nitrogen.1 In a Colombian trial, the leaves of fully loaded trees fell short of nitrogen and potassium.5 Nutrition and fertilisation covers rates and timing.

Stress adds to the load. The 2007 review links die-back to dry soil and air, high temperatures and strong sun.1 A coffee textbook chapter states that full-sun coffee without fertilisers, pesticides and irrigation suffers strong biennial production and die-back.8

Source–sink physiology#

Allocation. In the 2005 trial on five-year-old Costa Rica 95 in their third production cycle, fruit load was set at 0, 25, 50 and 100% of the initial load.3 Increasing fruit load significantly reduced branch growth and berry dry mass at harvest; ring-barking the branches reduced growth further.3 Dry matter allocation to berries was four times that to branch growth. Branch die-back and berry drop rose significantly with fruit load.3 The authors read this as competition for carbohydrates between berries and shoots, and among berries.3

Fruit photosynthesis. Green berries photosynthesise a little. At the bean-filling stage and field light levels of 200–500 µmol m⁻² s⁻¹, their photosynthesis covered about 30% of their daily respiration and 12% of their total carbon needs.3

Leaf photosynthesis. Whether leaves raise their output when the fruit pulls harder is contested. In the 2005 trial, leaf net photosynthesis rose with fruit load. Leaves on loaded, non-isolated branches reached three times the rate of leaves on ring-barked branches without berries.3 A 2008 study cited by the 2007 review found more than 50% higher photosynthesis and stomatal conductance in trees with a full crop and half the leaf area than in defruited trees.1 The 2012 Brazilian study found no compensatory increase as leaf area per fruit fell.2 A 2024 Colombian study found no significant change in leaf gas exchange across fruit loads.5 See Contested.

Shoot growth. At Viçosa, Brazil (20°45′S, 650 m), vegetative and reproductive growth of field-grown arabica run at the same time.4 Removing the fruit raised branch growth and leaf area gain markedly. The authors attributed much of the restriction in fruiting trees to starch exhaustion, alongside supra-optimal temperatures.4 Nitrogen supply in amino form did not differ enough to explain the growth difference.4

Reserves, roots and nutrients#

Starch. Early studies cited by the 2007 review linked die-back to strong falls in starch in the root and trunk.1 In a 2006 study cited by the review, heavily bearing arabica (3,765 kg of dry beans per hectare, as reported) had 30% less starch than defruited trees, with no fall in total soluble sugars.1 A 1993 Brazilian study found that branches of bearing trees died back regardless of their starch content.1 The review concludes that carbohydrate stores are probably not the only factor that sets off die-back.1

Roots. A 2005 study cited by the review followed irrigated, five-year-old conilon at Linhares, Brazil, through a year of 12,000 kg of dry beans per hectare.1 Root dry matter fell from 3.30 kg per plant in October (pinhead fruit) to 1.50 kg in January (bean filling) and 1.60 kg in April (early ripening).1 It recovered to 2.18 kg in July and 3.50 kg in October.1 Root dry matter did not change significantly in earlier years with smaller, regular crops. The author concluded that heavy crops may badly impair root function and so aggravate die-back and decline.1

03691201234Months after October (pinhead fruit stage)Root dry matter (kg per plant)Root dry matter1
Data table
SeriesMonths after October (pinhead fruit stage)Root dry matter (kg per plant)Source
Root dry matter03.31
Root dry matter31.51
Root dry matter61.61
Root dry matter92.181
Root dry matter123.51
Root dry matter of five-year-old irrigated conilon plants through a high-yield year (12,000 kg of dry beans per hectare), Linhares, Brazil. Data: Bragança 2005, as reported in the 2007 review.1

Nutrients. In the 2024 Colombian trial, three-year-old Cenicafé 1 trees in full sun at the Naranjal station (Chinchiná, Caldas, 1,383 m) were thinned to nine fruit loads from 20 to 100%.5 At 20–40% load, leaf nitrogen averaged 2.14%; at 100% load it was 1.77%, with potassium at 1.02%.5 The authors describe deficiencies of nitrogen, potassium, sulphur and copper at high fruit load. They attribute them to the pull of the fruit during endosperm filling.5 Leaf total soluble sugars fell linearly as fruit load rose.5 That study did not record die-back.

Nitrogen trials. A 1961 Costa Rican study cited by the 2007 review found 44% less die-back with 100 g of nitrogen per plant per year and 66% less with 200 g, compared with no nitrogen.1 The 1993 Brazilian study found no relationship between leaf nitrogen or potassium and die-back.1 In the 2012 study, die-back showed only weak, though significant, relationships with leaf nitrogen and starch.2

The 2012 leaf-to-fruit study#

The study ran on seven-year-old Catuaí Vermelho IAC 99 at Coimbra, Brazil (720 m), in full sun and without irrigation, planted at 2.0 × 1.0 m.2 Branches were sorted into three classes of leaf area per fruit: below 6 cm² (mean 3.5), 6.1–14 cm² (mean 9.4) and above 14 cm² (mean 23.0).2 Die-back was counted where a branch had at least 10 cm of dead tissue at its tip.2

In 2007 about 78%, 50% and 20% of branches died back in the three classes, regardless of canopy position.2 Branch growth rose with leaf area per fruit in every canopy position.2 In the low-yield season of 2007–2008 there was no die-back.2 The authors found die-back probably not closely related to carbohydrate, amino acid and mineral concentrations.2

Older figures set a higher bar for leaf area. A 1983 study in Zimbabwe, cited by the 2007 review, put the leaf area needed for one fruit at about 2,000, 3,200 and 7,200 mm² at the top, middle and bottom of the crown.1 That is 20, 32 and 72 cm² (our conversion). A 2002 study found no further gain in bean size above 2,000 mm² per fruit.1 Even the best-supplied class of the 2012 study averaged 23.0 cm² per fruit and still lost about 20% of its branches.2

Varieties and decline#

The 2007 review states that Bourbon and Caturra are more prone to die-back than Mundo Novo.1 In Brazil, Catimor is highly productive but rapidly enters a decline marked by sudden die-back after a few years of high yields.1 Plants that recover over two or three years while still carrying an appreciable crop may finally degenerate, according to the review.1

Die-back caused by pathogens#

The descending death of side branches has often been blamed on fungi such as Colletotrichum and Phoma.1 A growing body of evidence suggests that these infections follow the weakening of the tissue rather than cause it, according to the 2007 review.1 Some pathogens do cause die-back as a disease of their own. In Brazil, diseases of the Colletotrichum complex include anthracnose of branches, fruit and leaves, and die-back; they are attributed to C. gloeosporioides.10 The same study notes that coffee berry disease is absent from Brazil and restricted to Africa.10 Phoma leaf spot damages coffee by drying of stems, flower necrosis and mummified pinhead fruit. In a 2025 Brazilian study the disease was higher at 15–20 °C and in windier periods; higher plantations are more exposed to cold winds.11 Bacterial halo blight (Pseudomonas syringae pv. garcae) causes lesions on leaves, flowers and pinhead berries and die-back of twigs and branches. In Brazil most strains collected since 1997 came from regions near or above 800 m.9 These diseases show lesions on leaves, flowers or young fruit.9,10,11 No source consulted for this entry gives a tested field key for telling them apart from physiological die-back.

Management#

Lever What the sources report Conditions Source
Shade Reduces profuse flowering, overbearing, biennial production and die-back Review DaMatta et al.1
Shade 45% shade: yield −18%, less alternate bearing Costa Rica 95, central valley of Costa Rica Vaast et al.7
Leaf area Full-sun trees can carry a very heavy crop without die-back if enough leaf area is kept Brazil, 1993 study cited in review DaMatta et al.1
Nitrogen Die-back −44% at 100 g N per plant per year; −66% at 200 g Costa Rica, 1961 study cited in review DaMatta et al.1
Full-sun intensification Without the package of fertilisers, pesticides and irrigation: strong biennial production and die-back Textbook chapter Lambot et al.8
Stem renewal Periodic pruning and renewal of upright stems buffers biennial production Conilon, Brazil DaMatta et al.1
Renovation in blocks Renewing part of the farm each year stabilises farm production Colombia Arcila Pulgarín6

Cenicafé gives renewal shares of 33%, 25%, 20% and 17% of the plantation per year for cycles of three to six years.6 Pruning methods, stumping and renovation cycles are covered in Pruning, stumping and renovation. Fertiliser rates by country are covered in Nutrition and fertilisation.

Reference values#

Parameter Value Conditions Source
Dry matter to berries vs branch growth ≈4 : 1 over the cycle Costa Rica 95, 5 years old, Costa Rica Vaast et al.3
Berry photosynthesis, bean filling ≈30% of daily respiration; 12% of total carbon needs 200–500 µmol m⁻² s⁻¹ Vaast et al.3
Branches with die-back ≈78% / 50% / 20% Below 6 / 6.1–14 / above 14 cm² leaf per fruit; Catuaí Vermelho IAC 99, Brazil, 2007 Chaves et al.2
Die-back definition used ≥10 cm of dead tissue at the branch tip Same study Chaves et al.2
Leaf area for one fruit ≈2,000 / 3,200 / 7,200 mm² (top / middle / bottom) Arabica, Zimbabwe, 1983 study cited in review DaMatta et al.1
Bean size plateau No gain above 2,000 mm² leaf per fruit 2002 study cited in review DaMatta et al.1
Starch, heavy vs no crop −30% (no fall in soluble sugars) Arabica, 3,765 kg dry beans/ha, Brazil DaMatta et al.1
Root dry matter, heavy crop 3.30 → 1.50 kg per plant (Oct → Jan); 3.50 by next Oct Irrigated conilon, 5 years old, 12,000 kg dry beans/ha DaMatta et al.1
Fruit share of K, P, N uptake >95% of current uptake, late fruit growth Review DaMatta et al.1
Leaf N, 20–40% vs 100% fruit load 2.14% vs 1.77% Cenicafé 1, 3 years old, 1,383 m, Colombia Leon-Burgos et al.5
Leaf K, 100% fruit load 1.02% Same study Leon-Burgos et al.5
Recovery after die-back 2–3 years Review DaMatta et al.1

Open questions#

  • The trigger. Starch exhaustion, nutrient depletion and root loss have all been proposed. No study consulted for this entry separates them in one design (see Contested).1,2
  • Water and crop load. The 2007 review links die-back to water deficit and heat.1 No study consulted for this entry measured die-back across crop loads under controlled irrigation.
  • Canephora. Most measurements are on arabica. For conilon, the root data come from one study cited in a review.1
  • Telling causes apart. No source consulted gives a tested field key that separates physiological from pathogen die-back.

Key studies#

Chaves et al., 2012 – Leaf-to-fruit ratio and branch die-back. Sorted 360 branches of seven-year-old Catuaí in full sun in Brazil by leaf area per fruit. Die-back rose steeply as leaf area per fruit fell, from about 20% to about 78% of branches. Photosynthesis, carbohydrates and minerals changed little, so the authors doubt a simple starvation trigger.

Environmental and Experimental Botany 77, 207–218 · doi:10.1016/j.envexpbot.2011.11.011

Vaast et al., 2005 – Fruit load and carbon allocation. Set fruit loads from none to full on Costa Rica 95 and ring-barked some branches. Berries took four times the dry matter of branch growth; die-back and berry drop rose with load, while leaf photosynthesis also rose.

Tree Physiology 25(6), 753–760 · doi:10.1093/treephys/25.6.753

Amaral et al., 2001 – Fruiting, carbohydrates and nitrogen. Compared fruiting and defruited arabica at Viçosa, Brazil. Defruited trees grew much faster; the authors linked the restriction mainly to starch exhaustion and high temperatures, not to nitrogen supply.

Revista Brasileira de Fisiologia Vegetal 13(1), 66–74 · doi:10.1590/S0103-31312001000100008

Leon-Burgos et al., 2024 – Fruit load and leaf nutrients. Thinned Cenicafé 1 trees in Colombia to nine fruit loads. Full loads left the leaves short of nitrogen, potassium, sulphur and copper and lowered leaf sugars, with no significant change in gas exchange.

Scientia Horticulturae 329, 113005 · doi:10.1016/j.scienta.2024.113005

DaMatta et al., 2007 – Ecophysiology of coffee growth and production. Reviews die-back as a complex, descending branch death tied to heavy crops, full sun and stress. Gathers older field data on starch, roots, nitrogen and varieties, and argues that pathogens are mostly secondary.

Brazilian Journal of Plant Physiology 19(4), 485–510 · doi:10.1590/S1677-04202007000400014

Contested
  • Starch and die-back: a 2001 Brazilian study linked much of the growth restriction in fruiting trees to starch exhaustion,4 and a 2006 study cited by the 2007 review found 30% less starch in heavily bearing than in defruited trees.1 A 1993 Brazilian study cited by the same review found that branches of bearing trees died back regardless of their starch content,1 and a 2012 field study judged die-back probably not closely related to carbohydrate, amino acid or mineral concentrations.2
  • Nitrogen and potassium: a 1961 Costa Rican study cited by the 2007 review found 44% and 66% less die-back with 100 and 200 g of nitrogen per plant per year.1 The 1993 Brazilian study cited in the same review found no relationship between leaf nitrogen or potassium and die-back.1
  • Photosynthesis under a heavy crop: in a 2005 Costa Rican trial leaf photosynthesis rose with fruit load,3 and a 2008 study cited by the 2007 review measured more than 50% higher rates in trees with a full crop and half the leaf area than in defruited trees.1 A 2012 Brazilian study found no compensatory increase as leaf area per fruit fell,2 and a 2024 Colombian study found no significant change in leaf gas exchange with fruit load.5
Share or cite this entry

Change log

  • – The Spearman coefficients for die-back against leaf nitrogen (0.372) and starch (0.289) were removed; the text now says only that the relationships were weak though significant. Reason: The 2012 paper gives the coefficients as positive in its table but calls the relationships negative in its text, so the values cannot be reported without that conflict.
  • – Removed the statement that coffee berry disease is caused by C. kahawae from the pathogen section; it now reports only what the cited study says (absent from Brazil, restricted to Africa). Reason: The cited Brazilian study does not name the agent in that statement; the coffee berry disease entry covers the pathogen.
  • – 'In coffee it mostly hits the side branches' reworded to what the 2007 review says (the descending death of plagiotropic branches); the 2006 starch study is no longer called Brazilian; halo blight strains come from regions 'near or above' 800 m. Reason: The review does not state that die-back mostly hits side branches, nor where the 2006 study was done; the 2018 paper says 'near or above 800 m'.

Sources

  1. ADaMatta, F. M.; Ronchi, C. P.; Maestri, M.; Barros, R. S. (2007). Ecophysiology of coffee growth and production. Brazilian Journal of Plant Physiology 19(4), 485–510. doi:10.1590/S1677-04202007000400014 · other copies
  2. AChaves, A. R. M.; Martins, S. C. V.; Batista, K. D.; Celin, E. F.; DaMatta, F. M. (2012). Varying leaf-to-fruit ratios affect branch growth and dieback, with little to no effect on photosynthesis, carbohydrate or mineral pools, in different canopy positions of field-grown coffee trees. Environmental and Experimental Botany 77, 207–218. doi:10.1016/j.envexpbot.2011.11.011 · other copies
  3. AVaast, P.; Angrand, J.; Franck, N.; Dauzat, J.; Génard, M. (2005). Fruit load and branch ring-barking affect carbon allocation and photosynthesis of leaf and fruit of Coffea arabica in the field. Tree Physiology 25(6), 753–760. doi:10.1093/treephys/25.6.753 · other copies
  4. AAmaral, J. A. T.; DaMatta, F. M.; Rena, A. B. (2001). Effects of fruiting on the growth of Arabica coffee trees as related to carbohydrate and nitrogen status and to nitrate reductase activity. Revista Brasileira de Fisiologia Vegetal 13(1), 66–74. doi:10.1590/S0103-31312001000100008 · other copies
  5. ALeon-Burgos, A. F.; Rendón Sáenz, J. R.; Imbachi Quinchua, L. C.; Unigarro, C. A.; Osorio, V.; Sadeghian Khalajabadi, S.; Balaguera-López, H. E. (2024). Varying fruit loads modified leaf nutritional status, photosynthetic performance, and bean biochemical composition of coffee trees. Scientia Horticulturae 329, 113005. doi:10.1016/j.scienta.2024.113005 · other copies
  6. AArcila Pulgarín, J. (2007). Renovación y administración de los cafetales para estabilizar la producción de la finca. In: Arcila Pulgarín, J. et al. (eds), Sistemas de producción de café en Colombia, Cenicafé, ch. 7.
  7. AVaast, P.; Bertrand, B.; Perriot, J.-J.; Guyot, B.; Génard, M. (2006). Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. Journal of the Science of Food and Agriculture 86(2), 197–204.
  8. BLambot, C.; Herrera, J. C.; Bertrand, B.; Sadeghian, S.; Benavides, P.; Gaitán, A. (2017). Cultivating Coffee Quality – Terroir and Agro-Ecosystem. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 2, pp. 17–49. doi:10.1016/B978-0-12-803520-7.00002-5 · other copies
  9. AMaciel, K. W.; Destefano, S. A. L.; Beriam, L. O. S.; Almeida, I. M. G.; Patricio, F. R. A.; Rodrigues, L. M. R.; Guerreiro Filho, O. (2018). Bacterial halo blight of coffee crop: aggressiveness and genetic diversity of strains. Bragantia 77(1), 96–106. doi:10.1590/1678-4499.2016267 · other copies
  10. AOgoshi, C.; Pinto, F. A. M. F.; Santos Neto, H.; Silva, B. M.; Abreu, M. S.; Resende, M. L. V. (2015). Viability and susceptibility of propagation material from coffee plants to Colletotrichum sp.. Semina: Ciências Agrárias 36(6), 3539–3551. doi:10.5433/1679-0359.2015v36n6p3539 · other copies
  11. ASilva, H. R.; Pozza, E. A.; Freitas, A. S.; Freitas, M. L. O.; Barbosa Junior, M. P.; Cirillo, M. A. (2025). Temporal analysis of Phoma leaf spot of coffee plants at different altitudes. Revista Ceres 72, e72009. doi:10.71252/2177-34912025720009 · other copies

Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.

The literature for this entry was last searched on 23 Oct 2026. A statement that no source consulted for this entry shows something refers to the sources above; newer or missing work may exist.

Suggest a change

On Overbearing and die-back. Tell us what you would change and why. A source that supports it helps most.