Ten years, 17 countries, 32 varieties: what the world’s largest coffee variety trial has taught us about climate resilience

Key takeaways

• 17 countries, 30 sites, 32 varieties: A new study in Scientific Reports analyzes yield and rust data from the world’s largest dataset for understanding how coffee varieties respond to their environment.
• Climate shapes performance: Cross-disciplinary research collaboration combined climate modelling with global variety performance data to reveal that temperature patterns, especially the gap between day and night temperatures, are associated with which varieties do best where. Rust severity rises in  warm, humid sites.
• Improved varieties lead: Introgressed varieties outperformed pure arabica varieties on both yield and rust resistance, although with distinct climatic preferences and trade-offs in yield stability.

Researchers from World Coffee Research (WCR), the Alliance of Bioversity International and CIAT, and scientists from 21 countries have published the most comprehensive global analysis to date of how arabica varieties perform across climates. The analysis of yield and rust data from the International Multilocation Variety Trial (IMLVT) included 32 varieties across 30 sites in 17 countries. 

A tree planted today is a 20-year commitment to a future climate. Planting the wrong variety costs farmers time, capital and productivity. Most variety trials are in one country, so there is little global evidence on how a variety will perform outside the place it was bred. The IMLVT was designed to close that gap and to provide coffee breeders and other scientists insight into the extent to which variety performance is influenced by environmental factors like heat and humidity.

“Our members invested in this trial a decade ago because no one company or country could answer these questions alone. That patience is now paying off in new scientific understanding and the advancement of varieties,” says Dr. Jennifer “Vern” Long, CEO, WCR.

A trial built for a changing climate

Beginning in 2015, WCR gathered 32 top-performing arabica varieties from breeding programs worldwide and distributed them to national partners in 17 countries for long-term evaluation using shared, standardized protocols. Trial sites were chosen using climate suitability research we commissioned from the Alliance in 2015, and cover most of the climates where arabica grows today and where warmer conditions are expected by 2055. The first published study from the trial focused on rust resistance and showed that relative performance varies between sites. The new Scientific Reports study expands the environmental modeling on rust resistance and presents data on yields and connects both traits to climate factors. The trial lives on and will continue producing data at a subset of sites through 2030.

"What sets this study apart is that we bring coffee breeding and climate science together at an unprecedented global scale,” said Dr. Romain Gautron, co-lead author and Alliance coffee researcher.  “By linking variety performance with contrasting climate conditions across trial sites worldwide, we can characterize how varieties respond to their environment, revealing not just which ones tend to perform well, but under what conditions, and identify promising options for the future of coffee farming." 

Bringing data to life through CafeClima and Innovea

The analysis described in Scientific Reports is the scientific engine behind CafeClima, launched with the Alliance in February 2026. CafeClima pairs IMLVT performance data with Alliance climate projections. It finds “climate twins” (trial sites whose climate today matches what a user’s region will look like in the future), and shows how varieties performed there. The interactive web tool is free, available in English and Spanish, and designed for agronomists, sustainability managers, governments and investors planning renovation.

Performance data from the trial also informed key decisions in selecting some of the best parent material for the Innovea Arabica Breeding Network, WCR’s global effort to develop the next generation of climate-resilient arabica varieties. 

The environment tips the scales

The new environmental modeling found that higher rust severity was associated with trials in higher temperatures, elevated humidity, and shorter dry intervals. For yield production, there was not a clear climate pattern to overall productivity, but the differential variety response was found to be associated with thermal regimes: between cool and high diurnal temperature range to warm and low diurnal temperature range. Across the IMLVT’s 30 sites, the top four yielding varieties overall  were:  EC16 (“Mundo Maya”), H1 Centroamericano, Parainema and Ruiru 11. But the varieties’ performance differed significantly depending on the climate. EC16, Parainema and Ruiru 11 were most successful in cooler sites with big day–night temperature swings. Centroamericano performed best in warmer, more stable climates. EC16  and Centroamericano are both vegetatively propagated hybrid varieties, which can have tradeoffs for farmers (higher cost, and more difficult to access). 

Other varieties, including IPR103, Colombia 5, and S4808, paired good yields with stable performance across sites, meaning that they performed well across many different climates. Conversely,  other  varieties such as Pacamara, K7, SL28, Catuaí, and Mundo Novo yielded less, were more susceptible to rust, and the rust severity varied strongly with climate.

The results confirm that breeding delivers important performance gains. Notably, the most recent varieties, including F1 hybrids and improved inbred lines such as IPR107, Catigua MG2, and Parainema, displayed the highest levels of productivity and rust resistance. Together, these findings underscore that adapting coffee for a changing climate will depend on sustained breeding.

“The IMLVT trial revealed valuable insights into the adaptation of leading global Arabica varieties under India's unique shade-grown conditions. The project provided a rare platform for the Indian coffee community to observe these varieties alongside our local checks. The learnings from the project  will help planters make more informed choices for future planting.” 

—Jeena Devasia, Head, Plant Breeding and Genetics Division, Central Coffee Research Institute (CCRI), Chikmagalur, India.

From trial plots to farmers’ fields

The findings confirm that promising varieties should always be tested under local environmental and management conditions before they are recommended to farmers. This work is underway in Rwanda, where data from the IMLVT trial has been used to support the introduction of five varieties into formal release trials after some yielded more than twice as much as older varieties.

“This is the payoff of years of rigorous trial work: coffee trees suited to Rwanda's changing climate and to the quality our farmers are already known for,” said Simon Martin Mvuyekure, a Senior Coffee Breeding Research Fellow at RAB. 

A return on investment with long-term vision

The decade-long IMLVT trial is funded by WCR’s 200+ member companies, whose pre-competitive investment has produced a mounting body of peer-reviewed science on variety performance under changing climate conditions, as well as a public decision tool to reduce the risk of renovation investments and varieties moving toward farmers’ fields. This analysis received additional support from Taylors of Harrogate and Coffee Circle.