· Analytic
Z3138 Corn Belt Tmax Silking Heat
The count of July days where daily maximum temperature exceeded 32°C across Iowa, Illinois, Indiana, and Nebraska averaged by state, z-scored and July-gated. The signal only fires during the critical July silking window , delivering a precision read on pollination-window heat-stress risk.
Current reading
+0.000
NORMAL
DORMANT · WAITING FOR JUL
DAILY · SIGNAL
As of Oct 30, 2026. Updated nightly. · NEXT WINDOW JULY 2027
Current composite
0.0000
On the plus-or-minus one scale; signal is dormant outside July
SEASONAL STATE
DORMANT · WAITING FOR JUL
The signal is July-gated by construction. From August 1 through June 30 each year the composite reads exactly 0.000. The next live window opens July 1, 2027.
Signal state
Normal
59 consecutive days in episode
All-time percentile
91.6th
Upper-decile , most of history is NORMAL (dormant outside July) Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
EXTREME_POS corn fwd 60d
+7.96%
88 days (1.0%), HAC t +3.66 sig at 1% , precision silking-heat signal Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
MODERATE_POS corn fwd 60d
-2.72%
673 days (7.5%), HAC t -2.23 sig at 5% , pre-priced moderate July heat Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
Signal is July-gated by construction and fires only during a ~22-trading-day window each year; historical sample size is correspondingly small.
Current contribution to ML Auto Corn Composite V9
+0.000
1.7 percent weight × -1 polarity × 0 value
OBJECTIVE
The silking phenology stage is when corn plants extrude silks for pollination. During silking, kernel count is set , the single most consequential 7-10-day window for corn yield each year. Heat stress above approximately 32°C during silking directly reduces kernel count via pollen damage and desiccation. In most years, July 15-25 is the peak-silking window across the Iowa-Illinois-Indiana-Nebraska belt (collectively ~55 percent of US corn production). This signal queries OpenMeteo daily maximum temperature data for each of the four states, counts days per state where Tmax exceeded 32°C in the trailing 7-day window, averages across the four states, z-scores on a 504-day rolling window, applies a JULY-GATE (multiplies by 1 during July, 0 otherwise), and tanh-bounds. The July gate means the signal is DORMANT 11 months of the year and only carries information during July. Across 36 years of history, 91.5 percent of sample is in NORMAL tier (zero signal, outside July). Within the active July window, 7.5 percent of sample is MODERATE_POS and 1.0 percent EXTREME_POS. EXTREME_POS state (88 state-days, July with widespread above-32C temperatures) delivers mean corn fwd 60d of plus 7.96 percent (HAC Newey-West t plus 3.66, significant at the 1 percent level) , a strong precision signal of July silking-window heat stress. MODERATE_POS state (July with moderate above-32C days) delivers minus 2.72 percent (HAC t minus 2.23, significant at 5 percent, 673 state-days). The MODERATE_POS state’s negative mean reflects the pre-priced market dynamic: by the time moderate July heat is observed, corn has typically already rallied substantially; the forward window is often where the heat-stress narrative resolves. EXTREME_POS is the extreme precision firing where heat is so severe the market was NOT yet pricing it. The signal reads 0.0000 as of October 23 2026 in NORMAL territory for 59 consecutive trading days , the signal is dormant outside July. The signal contributes 1.7 percent weight to ML Auto Corn Composite V9 ML Auto Corn Composite (second-lowest weight) with polarity sign minus one; current contribution zero.
LIVE CHART
Thirty-six years of July days-above-32°C count across Iowa + Illinois + Indiana + Nebraska, July-gated.
COMPOSITE
0.0 / 100
STATE
Normal
PHASE
0
FWD 60d
-0.37% (baseline +2.31%)
SAMPLE N
9,039 days
TODAY’S PLAYBOOK READING
STATE
NORMAL
WHAT THE CHART SHOWS
The composite reads exactly 0.000, by construction. The signal is July-gated: it only fires between July 1 and July 31 each year, reading the count of days with daily max temperature above 32 degrees Celsius across the four production-dominant belt states. Outside July the engine zeroes the composite by design. July 2026 produced a mild heat-day count (composite mean +0.30 through the live window). CBOT front-month corn (orange, right axis) sits near $4.48/bu.
TOP INPUTS FIRING
- Primary composite 0.000 (July-gated) Signal is OUT OF SEASON. The engine returns exactly 0.000 by construction from August 1 through June 30 each year. Next live window opens July 1, 2027.
- July 2026 silking-heat window reading The 2026 July composite averaged +0.30, reading a mild count of belt heat days above 32 degrees Celsius during the pollination window. Not an extreme-heat year, not a cool-July year.
- IC vs CBOT corn fwd 60d: tail +0.138 (p=0.056) Rolling-IC tail is marginally significant; the July-gated design produces a thin firing sample (approximately 22 trading days per year), so power is limited to in-season observations only.
THE TAKEAWAY
What this means for you today
THE JULY SILKING WINDOW
Corn kernel count is set during the silking phenology stage, which occurs across most of the US corn belt in mid-to-late July. The window is approximately 7-10 days per plant but varies by state (Iowa typically peaks July 15-25, Nebraska often slightly earlier). Heat stress above 32°C during silking directly damages pollen and desiccates silks, reducing kernel count. The 2012 drought July produced 15+ consecutive days above 32°C across the central belt and resulted in approximately 25 percent corn yield loss , the modern-era extreme example.
THE JULY-GATED PRECISION DESIGN
The engine is DESIGNED to be dormant outside July. The multiplier `in_window = pd.Series(idx.month.isin([7]).astype(float), index=idx)` zeroes the signal in every month except July. The sample distribution reflects this: 91.5 percent of history is in NORMAL tier (dormant). The ACTIVE July window carries highly concentrated information: MODERATE_POS (673 state-days, 7.5 percent of sample) and EXTREME_POS (88 state-days, 1.0 percent of sample) collectively represent the Julys with meaningful above-32°C exposure. EXTREME_POS at HAC t plus 3.66 sig at 1 percent on 88 state-days is a HIGH-PRECISION high-confidence signal despite its low sample weight.
THE HONEST CAVEAT
The MODERATE_POS state delivers NEGATIVE forward returns (-2.72 percent, HAC t -2.23 sig at 5 percent) , opposite the EXTREME_POS +7.96 percent direction. The explanation: moderate July heat (above-32°C for 2-5 days in trailing 7-day window) is NORMAL for the Corn Belt and is already priced into the market by late July. The forward window is where the market resolves the already-priced heat narrative. Only EXTREME_POS (above-32°C for 7+ days in trailing 7-day window , true heat-wave territory) delivers the uncompensated-stress signature. The 32°C threshold is a FIXED threshold not adjusted for microclimate variations; different states have different climatological 32°C baselines (Nebraska typically hotter than Indiana at same calendar day).
THE REGIME MAP
The three belt silking-heat regime states, forward returns and sample share
Behavior across the belt daily-Tmax silking-heat count signal (CBOT corn benchmark). The signal is JULY-GATED by design: it fires only during the pollination window (July 1 through July 31 each year), reads zero on every other calendar day. Outside July, state is Normal by construction. Three states reflect heat-day count intensity during the July window: Normal (none), Moderate (above-baseline count), Extreme (deep tail).
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Normal Silking Heat TODAY | 0 | 91.6% (8281 d) | +1.00% | +0.71pp | Aug 2026, Aug 2025, Aug 2024 | |
| Moderate Belt Silking Heat | + | 8.4% (763 d) | -1.20% | -1.49pp | Jul 2026, Jul 2025, Jul 2024 | |
| Extreme Belt Silking Heat | ++ | 0.0% (0 d) | n/a | n/a | none |
The historical pattern
Methodology
THE TRACK RECORD
State changes
| State | Start | End | Duration | Fwd 60d SPY |
|---|---|---|---|---|
| Moderate Pos | 10/23/2024 | 06/30/2025 | 171d | +4.92% |
| Normal | 07/01/2025 | 07/31/2025 | 22d | +6.73% |
| Moderate Pos | 08/01/2025 | 06/30/2026 | 229d | +10.22% |
| Normal | 07/01/2026 | 07/31/2026 | 22d | +3.43% |
| Moderate Pos | 08/03/2026 | 10/23/2026 | 59d |
Scroll inside the table for the full history. Forward 60-day SPY is measured from each run’s start date; a reader cannot trade off a start-date they only identify in hindsight, so this is a transparency log, not an alpha claim. Asterisked (*) returns are partial. The run is less than 60 trading days old. The full 60-day comparable fills in as more data arrives.
What this measures
The engine queries OpenMeteo daily maximum temperature for Iowa (`OPENMETEO_IOWA_CORN_BELT_TMAX_DAILY`), Illinois (`OPENMETEO_ILLINOIS_CORN_BELT_TMAX_DAILY`), Indiana (`OPENMETEO_INDIANA_CORN_BELT_TMAX_DAILY`), Nebraska (`OPENMETEO_NEBRASKA_CORN_BELT_TMAX_DAILY`). For each state, counts trailing-7-day days where Tmax exceeded 32°C. Averages across the four states. Z-scores on 504-day rolling window with 126-day minimum. Multiplies by month==7 gate (1 during July, 0 otherwise). Tanh-bounds. Five-state machine.
GO DEEPER
Want the full depth?
The full research brief adds methodology detail on how the belt Tmax count is built across four production-dominant states (Iowa, Illinois, Indiana, Nebraska), how the July gating works mechanically, and why a narrow pollination-window reading can surface compressed-yield-risk even in otherwise comfortable growing seasons. Includes the three historical case studies anchored to in-window composite readings, the three-regime state-conditional table, and the methodology for pairing the signal with Silking Pace, Condition Week-over-Week, and the Corn Cycle DNA playbook.
Taxonomy, polarity, state labels
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 02, 2026 11:37 AM ET