El Niño: Brazil in Focus — What the Timing Means for the 2026/27 Coffee Crop

The Forecast Shift — March Through May

As recently as March 2026, El Niño was not the consensus. ENSO-neutral had just returned as La Niña wound down, and El Niño probability for the summer was below 30% across most model frameworks. By May 19, the IRI had revised that figure to 97–98% probability from May–July onward, sustained through the first quarter of 2027. The NOAA Climate Prediction Center's own table reached 82% El Niño for May–July (MJJ), rising to 96–98% for the July–September through December–February window.^[1,2]

The speed of the revision is itself meaningful. The table below shows the ENSO index progression through 2026, drawn from NOAA's official series and the SoftSignal ENSO data feed.^[3]

The ocean went from deep La Niña (RONI −0.88 in January) to the threshold of El Niño in five months. The raw Nino3.4 monthly anomaly crossed into positive territory for the first time in over two years in May, reaching +0.45°C; the weekly reading had already hit +0.9°C by May 13.^[1] What drove the acceleration: a subsurface heat reservoir built up across the entire La Niña period is now surfacing. Temperatures at 50–150 meters depth in the equatorial Pacific are running up to +6°C above normal across the 150°W–80°W corridor.

The atmospheric signal confirmed the ocean's shift in April. The Southern Oscillation Index (SOI), which had been positively coupled with La Niña at +2.4 as recently as February, flipped to −1.1 in April — the atmosphere catching up to an ocean that had already moved. When both ocean and atmosphere reinforce the same direction, the signal carries more weight.^[3]

What El Niño Does to Brazil — The Regional Split

Brazil is not a single climate story under El Niño. The country spans enough latitude that the same event produces meaningfully different — sometimes opposite — outcomes across its major growing regions. Understanding which crops sit in which zone is the first step in any ENSO-to-commodity analysis.

The mechanism: SACZ suppression

The primary driver of El Niño's impact on southeastern Brazil is the South Atlantic Convergence Zone (SACZ) — a quasi-stationary band of deep convection that runs diagonally from the Amazon basin toward the southeast Atlantic, supplying the summer monsoon moisture to Minas Gerais, Espírito Santo, and northern São Paulo. During El Niño events, anomalous Walker Circulation driven by equatorial Pacific warming creates subsidence over tropical South America, suppressing SACZ activity. The result is reduced rainfall, delayed wet season onset, and elevated temperatures in the coffee belt during the critical summer months.^[4]

Southern Brazil operates on a different system. Enhanced subtropical jet stream activity steers more frontal systems into Paraná, Santa Catarina, and Rio Grande do Sul, producing above-normal rainfall. Moderate El Niño events have historically been associated with improved soybean yields in this zone — though the 2024 Rio Grande do Sul floods, which displaced 600,000 people during the 2023/24 strong El Niño, illustrate that the benefit becomes a liability when events intensify.^[5,6]

The Phenological Calendar and Why Timing Is Everything

Arabica coffee in Brazil follows a defined seasonal clock. El Niño's supply risk is not diffuse — it concentrates in a specific three-month window, and the 2026 event is arriving on a trajectory that puts it squarely against that window.

The Brazil arabica seasonal calendar

The following rainfall normals are derived from ERA5 reanalysis data for the Brazilian arabica growing region, averaged across the Sul de Minas, Cerrado Mineiro, and Zona da Mata subregions.^[3]

The dry season is not the risk — the onset is

The dry season (June–August) is agronomically necessary. Arabica requires accumulated water stress during dormancy to synchronize flower bud differentiation — the tree uses this period to prime for the following season's production. Normal dry-season conditions across the Minas Gerais belt: June receives roughly 5mm, July about 2mm, August barely 3mm. Trees are not drinking; they are priming.^[8]

What El Niño disrupts is the wet season onset. In a normal year, September brings the first tentative return of moisture (~23mm in ERA5 normals for the arabica belt) — thin but enough to signal the transition. October builds to ~71mm, and November arrives at ~168mm. The arabica florada principal — the main flowering flush — is triggered by this rehydration sequence: cooling nights followed by the first warm rains induce simultaneous bud opening across the canopy, setting the framework for the entire following harvest. Disruption of that sequence — through delayed rains, insufficient rainfall, or premature heat events — produces asynchronous flowering: multiple maturity stages on the same tree, reduced fruit set, uneven cherry development, and lower yields.^[8,9]

SACZ suppression under El Niño delays and reduces this onset. During the 2023/24 El Niño, Mato Grosso's October–November rainfall recorded just 163mm — 53% of the 30-year average of approximately 347mm.^[7] While the Minas Gerais response is less predictable (see the transitional-zone caveat above), the mechanism argues for the same direction.

The heat stress component

Strong El Niño events have driven temperatures in coffee-producing municipalities to 40–43°C. This heat arrives during the bud differentiation phase — the dry-season rest period — and can compromise floral viability before any triggering rain arrives. Trees entering this phase depleted from heavy on-year production, as the 2025/26 crop is expected to be, are particularly exposed: carbohydrate reserves that would otherwise buffer against thermal stress are already drawn down.^[9]

A note on crop year accounting: the formation of a Brazil arabica crop begins roughly two years before its harvest. Disruption to flowering in October–November 2026 means a yield shortfall realized at harvest in June–August 2027.^[8]

Crop Impact Timeline: Onset to Recovery

Starting from El Niño onset around July 2026 — the current central forecast — the following traces the agronomic progression through to post-event recovery.

Market Context

KC Coffee futures peaked at 316¢/lb on April 23 and declined roughly 16% to 266¢/lb by late May. During the same period — while El Niño probability was moving from 16% to 82% — managed money net positioning in Coffee C was cut from +31,035 contracts on May 5 to +17,434 contracts on May 26, with short-side additions roughly doubling over those three weeks.^[14]

The market was selling the near-term supply story: the 2025/26 Brazil harvest is strong, and that is being correctly priced. El Niño is not a 2025/26 story. Its impact falls on the flowering window that determines the 2026/27 crop — the one currently forecast at 71.9M bags on the positive biennial. The market corrected on near-term optimism while forward phenological risk was building.

World coffee stocks-to-use stands near multi-decade lows. There is no supply buffer at the global level that absorbs a disrupted flowering season in Minas Gerais. Whether the October–November 2026 wet season onset proceeds on schedule — or is delayed and suppressed under a coupled El Niño — is the supply question for 2027.

What to Watch

• 1

  RONI crossing +0.5°C for the first time in an official 3-month average. This is the formal El Niño declaration threshold under NOAA's current framework.^[15] The declaration itself is a lagging confirmation — the supply risk timeline is driven by the physical progression, not the official label.

• 2

  El Niño type classification as the event matures through austral winter. Central Pacific (Modoki) events drive stronger SACZ suppression in SE Brazil and are therefore more adverse for the coffee belt than Eastern Pacific events, which concentrate impacts on northeast Brazil.

• 3

  Minas Gerais soil moisture and rainfall anomalies for August–September. The transition from dry season to first moisture signals in the coffee belt. Any meaningful departure from normal in these months will indicate whether the wet season onset is tracking on schedule or being delayed.

• 4

  SACZ activity October–November. Whether the convergence zone re-establishes on its normal schedule is the direct flowering-risk signal. Private crop tour data (Volcafe, StoneX) and CONAB's field assessments during this window will be the first reliable read on flowering quality.

• 5

  CONAB's first 2026/27 crop survey, typically released in late 2026. This will be the first official Brazilian assessment of the new crop and will reflect whether the flowering window proceeded normally or was disrupted.