Methodology & Sources

The Monitor surfaces county-level uninsured drought exposure across U.S. row crops: where current drought severity is high and federal crop-insurance coverage is thin — i.e., where loss exposure stays with the producer rather than shifting to the federal indemnity backstop. Crops tracked: corn, soybeans, wheat (winter + spring), upland cotton, grain sorghum.

Headline math

exposure               = drought_severity × (1 − coverage)
coverage               = clip(RMA insured acres ÷ NASS planted acres, 0, 1)
drought_severity_score = (D0·1 + D1·2 + D2·3 + D3·4 + D4·5) / 100   # 0–5
loss_ratio             = indemnity ÷ premium                         # KPI + table only

A county lands on the watch list when it has data from all sources and coverage < 70%. The list ranks by Risk = at-risk acres = planted acres × uninsured share × (drought ÷ 5) — the drought- and coverage-weighted slice of planted acres "at risk." It blends how exposed a county is with how much acreage is at stake, so a large-but-not-dry county doesn't outrank a smaller, severely-stressed one (every column is still click-sortable). The map's Crop Acreage Risk layer colors by the same figure.

The timeline

The play button animates USDM drought severity over time — monthly back to 2000 — so you can watch where drought has persisted across multiple years. It is the one variable with deep history. Coverage, the CPC forecast, and the exposure score always reflect the present: coverage is annual and slow-moving, and the CPC outlook is a single forward-looking issuance, so animating them would not be meaningful. The timeline answers "where has it been dry, and for how long"; the static views answer "given today's coverage, where is the uninsured risk."

Forecast blend

The CPC Seasonal Drought Outlook is directional (defined relative to current conditions), so forecast severity blends the current USDM value with the outlook category rather than using the category as a standalone score:

Counties without a CPC polygon assignment render gray ("no forecast available").

Coverage smoothing

RMA publishes the Summary of Business monthly, so the open crop year under-reports per-county insured acres until late season. The Coverage View control (Advanced) switches between:

• Adjusted (default) — observed, unless a stable county's open-year value is missing or drops > 10pp below its 5-year mean (RMA lag), then the 5-year mean.
• Observed — raw current year as published. Useful to see the lag.
• Smoothed — always the 5-year mean for stable counties (closed-year stdev < 5pp).

Sources

All sources join on 5-digit county FIPS.

Known limitations

• Prevented-plant acres are not netted out of the denominator, so coverage in flood/PP-heavy years is biased downward.
• Wheat is all-classes (winter + spring excl. durum). Durum (~3% of US wheat, ND/MT) is excluded, so durum-heavy counties' coverage is slightly overstated.
• Cotton: upland only. Pima/extra-long-staple (~2%, AZ/CA/NM) is excluded.
• Sorghum: grain only. Silage and hybrid-seed sorghum are excluded.
• RMA quantity-unit filter. Only acre-denominated rows sum into insured acres (yield-protection bushels excluded).
• FIPS join edge cases. VA independent cities, AK/HI county-equivalents, and consolidations can mismatch across sources; counties missing any source are excluded from the composite.
• Crop rotation. A single crop's planted acres swing year to year (corn↔soy). Coverage is plotted as a rate (rotation-robust); the combined All view (across the five tracked crops) is the most stable multi-year lens.
• Estimated acres before NASS publishes. Early in a crop year, county planted acres are estimated (state acres × the county's 3-year share, or prior-year carry) until NASS publishes county-level data, then they auto-promote. Coverage ratios resting on an estimate can shift when the actuals land. The hover/panel flag the acres source.
• Graze-out wheat. In the Southern Plains (e.g., the Texas Panhandle) much wheat is dual-purpose — planted for cattle forage and never harvested for grain, so it isn't insured as grain. NASS "area planted" counts it; RMA grain insurance does not. That legitimately depresses the insurance ratio in graze-heavy counties relative to grain-dominant neighbors.
• Loss ratio is volatile. One weather event in a single closed year can push a county above 5.0 — treat it as a flag, not a verdict.