04 — Do heat waves need a soil-moisture threshold?
Supplied working report; original verdicts not validated.
Review status: Needs review. This is an archival source note, not a passed proof.
README.md
# 04 — Do heat waves need a soil-moisture threshold? **Result: yes, there is a threshold, and it is low. Summer days with the driest 20% of antecedent soil moisture run 0.99 °C hotter than the wettest 20% (95% CI 0.65–1.37). The relation is not a smooth ramp: it bends at about the 23rd percentile of soil moisture, above which temperature is nearly insensitive to further drying. 25 of 29 stations show the effect.** ## The proposal Soil moisture is supposed to be a switch as well as a dial: evaporative cooling can absorb part of a heat input while water is available, and once the reservoir is empty the same atmospheric forcing produces a larger temperature response. That predicts a kink — a threshold — rather than a straight line from wet to dry. ## What was tested NOAA's U.S. Climate Reference Network: 29 stations with co-located air temperature and soil moisture (5, 10 and 20 cm) sensors, June–August 2011–2025, 37,487 station-days. For each station-day: - **soil moisture** — mean of the three shallow sensors, averaged over the 30 days *before* the day in question and expressed as a within-station percentile. Using antecedent moisture means the drying caused by the hot day itself is not counted as its cause. - **temperature anomaly** — daily maximum temperature minus a station-specific harmonic seasonal cycle fitted on the station's whole record. The contrast is within-station, so latitude, elevation, aridity and sensor calibration cannot explain it. Intervals resample stations, not days. ## What happened | | value | |---|---| | station-days / stations | 37,487 / 29 | | driest fifth minus wettest fifth | **+0.99 °C** (95% CI 0.65 to 1.37) | | breakpoint of the two-slope fit | **23rd percentile** | | slope below breakpoint | −0.032 °C per percentile | | slope above breakpoint | −0.008 °C per percentile | | F-test, two slopes against one | 32.2 (p << 0.001) | | stations with a positive dry-minus-wet contrast | **25 of 29** (sign test p = 0.0001) | Decile means run monotonically from +0.77 °C in the driest tenth to −0.38 °C in the wettest. **P1 and the threshold prediction are both supported.** Sensitivities: the same-day contrast is much larger (+2.59 °C), as expected if the day's own drying is included — that number is contaminated by reverse causation and is reported only to show the size of the contamination. Restricting to 2018–2025 gives +0.83 °C, slightly smaller but the same sign. ## What it means This is the cleanest of the five tests: a large, consistent, within-station effect that survives a station-level sign test and shows the predicted kink. The kink sits at a low moisture percentile, which is itself informative — most of the time, in most of the network, further wetting does not cool the daily maximum, because the atmosphere is not demanding the extra evaporation. Amplification is a property of the dry tail. The mechanism cannot be identified from this design. Dry soils and hot days are mutually reinforcing, and the 30-day antecedent window only reduces the simultaneity; it does not remove it. Land–atmosphere coupling studies use lagged and conditional designs for exactly this reason. ## What this test cannot see The network is 29 stations in the contiguous United States, not the world, and its siting avoids the urban and irrigated landscapes where the coupling is most consequential. Soil moisture at 5–20 cm is not the whole root-zone reservoir. And the seasonal-cycle subtraction is station-specific but not year-specific: a station with a strong multi-year drought trend will contribute a persistent anomaly that is not a heat wave. Preprint draft · not peer reviewed · designed and executed with AI assistance.
hypothesis.md
# The proposal, quoted Source: Incoprea notebook, *Earth & Climate* #30, "Heat Waves Need Dry Soil". > Heat waves require a soil moisture deficit threshold before they can amplify, so the same atmospheric setup produces a heat wave over dry ground and a mild spell over wet ground. Stated prediction: - Surface-air temperature anomalies should amplify only once soil moisture falls below a threshold, rather than rising smoothly as soils dry. (No separate refutation is recorded in the notebook entry; the implied refutation is a smooth, threshold-free relation.)