KC SPECIALIST · Softs / Brazil Drought
Z6524 – [KC] Brazil Water Deficit Signal
Brazil Cerrado and Sul-Minas water deficit (PET minus AET from TERRACLIMATE), as a true-water-stress drought lens for Arabica.
At a glance
DORMANT · WAITING FOR MAY
Current reading
-0.742
SEASONAL STATE
DORMANT · WAITING FOR MAY
Brazil dry-season window closed
Signal state
Water Surplus
234 days in Stress phase
Updated
Nov 04, 2026
Monthly TERRACLIMATE release
Sample
24-year window
Monthly cadence; 2002-12 onward
As of Nov 04, 2026. Updated nightly. · NEXT WINDOW MAY 2027
SIGNAL OVERVIEW
Brazil ships roughly 35 percent of the world Arabica crop and the Cerrado plus Sul-Minas growing region can run through a full crop year with normal rainfall yet still experience severe water stress when potential evapotranspiration outpaces actual evapotranspiration. The signal reads TERRACLIMATE monthly water deficit (PET minus AET) across Cerrado and Sul-Minas, z-scored on a 10-year rolling window and tanh-bounded. Positive composite equals severe water deficit equals yield stress equals bullish Arabica setup. Daily Spearman IC against Arabica forward 60-day returns is plus 0.050 at p-value 1e-4 across 5,951 observations; Phase +1 water-deficit firings deliver a plus 1.18 pp edge above baseline; Phase -1 water-surplus firings deliver minus 0.79 pp. 1.97 pp directional spread with a 24-year TERRACLIMATE window. Full regime stats and historical case episodes below.
LIVE CHART
KC Brazil Water Deficit Signal
What the chart is showing right now. The dark line is the TERRACLIMATE water-deficit composite, tanh-bounded to plus or minus one on the left axis. Positive composite means PET significantly exceeds AET across Cerrado and Sul-Minas (severe water stress, bullish Arabica via yield damage). The bronze line in the middle panel is ICE US Coffee C first-nearby in cents per pound. The bottom panel tracks the trailing 252-day Spearman information coefficient between the composite and Arabica forward 60-day returns. Green band fills mark Water Deficit extremes (historically bullish Arabica); red band fills mark Water Surplus extremes (historically bearish Arabica).
COMPOSITE
-0.742 / tanh
STATE
Water Surplus
PHASE
-0.95
DAILY IC
+0.050 (p = 1e-4)
PHASE SPREAD
1.97 pp
TODAY’S PLAYBOOK READING
STATE
Moderate Neg
STATE AND HISTORY
Water Surplus firing at composite -0.742, 234 trading days running in the STRESS phase. Phase +1 (Water Deficit) delivers plus 3.90 percent fwd60d / plus 1.18 pp above plus 2.72 pct baseline across 2,031 days. Phase -1 (Water Surplus) delivers plus 1.93 percent / minus 0.79 pp below baseline across 3,139 days. Daily IC plus 0.050 at p = 1e-4. Non-overlap rho plus 0.079 at p = 0.43 across 100 samples; sign-consistent with daily IC but not independently significant.
THE WATER-DEFICIT MECHANISM
Brazil produces about 35 percent of world Arabica, concentrated in Cerrado and Sul-Minas regions. TERRACLIMATE PET-AET captures the HOT-DRY interaction distinct from precipitation alone; rainfall can be normal but high evapotranspiration still stresses yield. The 10-year rolling z-score isolates the drought-of-record anomaly from longer-term climate drift.
HOW TO POSITION
Short-to-medium-lead supply read on Arabica (1 to 3 months). Water-deficit firings (composite above plus 0.5) historically coincide with Brazilian drought-of-record episodes (2014 drought, 2020-2021 compound drought). Water-surplus firings coincide with wet-Brazil regimes where yield is favorable. Pair with the Cecafe realized-price signal (transaction confirmation), the ENSO signal (long-lead ocean forcing), ICE Europe COT positioning.
WHAT THE CHART SHOWS
The composite sits at -0.742 in the Water Surplus band. TERRACLIMATE PET-AET across Cerrado and Sul-Minas has run significantly below its 10-year trend. The current STRESS episode began December 2025; current episode has run 234 trading days which meets the BTD danger threshold.
TOP INPUTS FIRING
- Primary composite -0.742 State Water Surplus, 234 days running in STRESS phase.
- Cerrado plus Sul-Minas water balance abundant TERRACLIMATE PET-AET has run below its 10-year rolling trend through 2026; actual evapotranspiration is keeping up with potential demand.
- Non-overlap test not independently significant (honest caveat) Non-overlap IC at 60-day sampling delivers rho plus 0.079 with p-value 0.43 across 100 samples; sign- consistent with daily IC plus 0.050 but not independently significant.
THE TAKEAWAY
What this means for you today
THE WATER-BALANCE CALL
Composite -0.742, Water Surplus. Brazilian Cerrado and Sul-Minas water deficit is significantly below trend; evapotranspiration is not outpacing actual water availability. Signal firing in the Water Surplus bucket for 234 days; forward 60-day edge in Phase -1 historically runs minus 0.79 pp below baseline.
THE EMPIRICAL CONTEXT
Phase stats show clean monotone pattern. Phase +1 (Water Deficit) delivers plus 1.18 pp above baseline; Phase 0 (Normal Balance) delivers plus 0.06 pp; Phase -1 (Water Surplus) delivers minus 0.79 pp. 1.97 pp directional spread with clean monotone ordering.
THE HONEST CAVEAT
Non-overlap test at 60-day sampling delivers rho plus 0.079 with p-value 0.43 across 100 samples; sign-consistent with daily IC but not independently significant at the 5 pct threshold. The 10-year z-score window means early-sample readings carry less precision than mature-window readings. Monthly TERRACLIMATE release cadence means the signal updates slowly.
THE REGIME MAP
The three phases, forward returns and sample share
Computed from 6,011 post-warmup trading days (2002-12-03 onward). Baseline Arabica fwd60d: mean plus 2.72 percent (median plus 0.59 percent). Phase +1 (Water Deficit) delivers +1.18 pp edge; Phase 0 delivers +0.06 pp; Phase -1 (Water Surplus) delivers -0.79 pp. Non-overlap IC not significant at p=0.43 (every-60d sampling, n = 100), though sign-consistent with daily IC.
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Water Deficit | +1 | 33.8% (2031 d) | +3.90% | +1.18pp | 2024 drought, 2021 drought, 2014 drought | |
| Normal Balance | 0 | 14.0% (841 d) | +2.78% | +0.06pp | 2017, 2013, 2011 | |
| Water Surplus TODAY | -1 | 52.2% (3139 d) | +1.93% | -0.79pp | 2026 (current), 2022, 2019 |
The historical pattern
Signature episodes in the 24-year sample
What this measures
TERRACLIMATE monthly potential-evapotranspiration-minus-actual-evapotranspiration (PET minus AET) aggregated across Cerrado and Sul-Minas Brazilian coffee regions, z-scored on a 10-year rolling window and tanh-bounded. Positive composite equals severe water deficit (drought); negative composite equals surplus moisture (wet regime). Distinct from precipitation-only signals; this captures the HOT-DRY interaction where rainfall can be normal but high evapotranspiration still stresses yield.
Why it matters
Why water deficit reads Arabica yield
Brazil produces roughly 35 percent of world Arabica, concentrated in Cerrado and Sul-Minas. Arabica is a rainfall-dependent crop; yield is determined by soil-water availability during flowering (September through January) and cherry-development (October through March). TERRACLIMATE PET-AET captures the HOT-DRY interaction that pure-precipitation metrics miss; rainfall can be normal but high potential evapotranspiration (temperature plus vapor pressure deficit) still stresses yield by pulling water out of the plant faster than roots can replenish.
How to use this signal today
Short-to-medium-lead supply read on Arabica with a 1 to 3 month lead horizon. Water Deficit firings (composite above plus 0.5) historically coincide with Brazilian drought episodes (2014, 2020-21). Water Surplus firings coincide with wet-Brazil regimes. Pair with the ENSO signal (long-lead ocean forcing), the Cecafe realized-price signal (short-lead transaction confirmation), and USDA FAS Brazil coffee attache reports for stacked conviction.
Risk overlay
Validation rests on daily Spearman IC plus 0.050 (p-value 1e-4, N = 5,951) with sign-consistent non-overlap test at 60-day sampling (p-value 0.43, not independently significant). 1.97 pp directional spread between Phase +1 and Phase -1. Primary risks are (a) monthly TERRACLIMATE cadence means slow updates, (b) the 10-year z-score window is young by climate standards (data starts 2000; usable 2002), and (c) individual Phase -1 triggers show heterogeneity on the forward side.
Signal consumers and siblings
Not investment advice · The analysis on this page is informational and educational. Signal readings, historical case figures, and empirical statistics are derived from publicly available data and are provided without warranty. Hypothetical performance results have inherent limitations, including the benefit of hindsight. Readers who intend to trade should consult a licensed investment professional and see the research brief for the full disclaimer.
Last refreshed Nov 05, 2026 05:30 PM ET