← Back to scans

How bad will CO2 atmospheric concentration get before 2030? — At least 450

KXCO2LEVEL-30-450 · World · 2026-07-05
3%
Agent
22%
Market Price
-19.0%
Edge
80%
Confidence
Volume: 3,853
Spread: 9.0c
Days to resolution: 1275
Markets in event: 5
Final Rationale
Physical evidence is decisive: current seasonal peak ~430 ppm, ~8-10 ppm short of 450, and all models (Cambridge MLG central 2034, earliest 95% bound summer 2032) place crossing after the Jan 2030 deadline. Reaching 450 requires unprecedented sustained 4+ ppm/yr, well above the ~2.6 ppm trend or even the record 2024 jump. The 22% Kalshi price appears overpriced and likely reflects thin-market noise. Resolution ambiguity doesn't close the gap. I nudge slightly below the forecasters' 4% given the critique's valid point about over-anchoring on the market.
Price vs. prediction
How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.
$n = 10$ follow-ups
Re-scan Context
This market has been scanned before. Previous predictions:
DatePredictedMarket PriceConfidence
2026-06-28 2% 22% 80%
2026-06-21 3% 13% 80%
2026-06-14 4% 12% 80%
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_direct claude_news code_execution kalshi_related polymarket_related
Sub-questions (Fermi decomposition)
  1. What is the current atmospheric CO2 concentration (Mauna Loa/NOAA global mean) as of the latest reading?
  2. What is the average annual growth rate of atmospheric CO2 over the past 5-10 years?
  3. What is the seasonal peak (typically May) CO2 concentration vs the annual mean, and which does the market measure?
  4. Projecting the current trend, what year would CO2 first reach 450 ppm?
  5. How does the Kalshi market currently price this outcome?
Planner reasoning
CO2 atmospheric concentration follows a well-established, predictable upward trend (Keeling Curve) with strong seasonal cycles peaking in May. The key question is whether the monthly/annual mean will hit 450 ppm before 2030, which is a near-deterministic extrapolation of the current ~425 ppm level growing ~2.5 ppm/year. The math strongly suggests 450 is unlikely to be reached by 2030 unless a shorter measurement window (seasonal peak) is used.
2. Research Parallel tool calls
ToolStatusTimeItemsSummary
kalshi_direct OK 0.2s 1 ## This Market's Kalshi Data **At least 450** (KXCO2LEVEL-30-450) - Current price (probability): 22.00% - 7-day price change: +10.00% - 30-day price change: +0.00% - Average daily volume: 71 contracts - Price range: 12.00% - 26.00% - Data points: 33 days
claude_news OK 27.9s 11 Here are the key findings for the forecasting question: --- **Current CO₂ Levels (Mauna Loa / Global)** - For the first time in history, CO₂ concentrations surpassed **430 ppm** during their seasonal peak; Mauna Loa showed a May 2025 monthly average of **430.2 ppm** (Scripps) / **430.5 ppm** (NO
code_execution OK 38.9s 1 Here are the key findings from the CO2 projection analysis: --- ## 📊 CO2 Concentration: Path to 450 ppm — Key Findings - **Current baseline (2024):** Annual mean CO2 ≈ **425.0 ppm**; May seasonal peak (Mauna Loa) ≈ **426.9 ppm**, with the seasonal cycle adding ~3–4 ppm above the annual mean each
kalshi_related OK 0.6s 1 1 related markets / summaries. series KXCO2LEVEL: 0 markets (skipped 5 no-signal) | keyword 'CO2 concentration': ok
polymarket_related OK 0.6s 0 Scanned 100 active Polymarket markets, kept 0 matches. keyword 'CO2 concentration': 0 markets | keyword 'atmospheric CO2': 0 markets
3. Evidence Brief Sonnet · 4925 chars
# Current state The Kalshi market asks whether atmospheric CO2 will reach ≥450 ppm before Jan 2030. CO2 is currently ~427–430 ppm (seasonal peak); the market resolves YES if any official reading hits 450 before the deadline — the resolution criterion is unclear on annual mean vs. instantaneous/seasonal peak, which matters at the margin. # Timeline of key events - **2024 full year**: Global annual mean CO2 = 422.7–423.9 ppm (NOAA/WMO) — confirmed - **2024**: Year-over-year increase of 3.58–3.75 ppm — largest single-year jump on record (NOAA/WMO) — confirmed - **May 2025**: Mauna Loa monthly average = 430.2–430.5 ppm, first time seasonal peak exceeded 430 ppm (Scripps/NOAA) — confirmed - **2026 forecast**: Met Office projects 2026 annual mean at Mauna Loa = 429.4 ± 0.6 ppm; May 2026 peak = 432.2 ± 0.6 ppm — reported - **2033–2035**: Central projection for 450 ppm crossing (Cambridge MLG model: April 2034 ± 17 months) — modeled --- # Event Will atmospheric CO2 concentration reach at least 450 ppm before January 2030? # Outcomes to forecast - **Yes**: CO2 reaches ≥450 ppm before Jan 1, 2030 - **No**: CO2 does not reach 450 ppm before Jan 1, 2030 # Kalshi market anchor **Current YES price: 22.00%** — up +10 pp over 7 days, flat over 30 days. Volume: ~71 contracts/day. Range over 33 days: 12%–26%. The recent 7-day spike is notable but 30-day flat suggests the spike may be noise/thin market rather than new information. # Sub-question answers 1. **Current CO2 concentration?** — May 2025 Mauna Loa monthly mean = 430.2–430.5 ppm (Scripps/NOAA); 2024 global annual mean = 422.7–423.9 ppm (NOAA/WMO). 2. **Average annual growth rate (5–10 yr)?** — ~2.5–2.6 ppm/yr recent estimate (Cambridge MLG); 2024 saw a record 3.58–3.75 ppm single-year jump, but longer-term trend is ~2.5 ppm/yr. 3. **Seasonal peak vs. annual mean?** — May seasonal peak runs ~3–4 ppm above the annual mean. Market rules do not specify; if measuring a seasonal peak reading, May 2030 mid-scenario ≈ 441–442 ppm, still ~8 ppm short. 4. **Projected year to reach 450 ppm?** — Central projection: 2034 (annual mean); May peak crosses 450 in 2033–2034 (Cambridge MLG). Even aggressive growth (2.6 ppm/yr) yields only ~441 ppm annual mean by 2030. 5. **Kalshi market price?** — 22% YES (see anchor above). # Key facts (high-confidence, factual) 1. [NOAA/WMO] 2024 global annual mean CO2 = 422.7–423.9 ppm 2. [Scripps/NOAA] May 2025 Mauna Loa peak = 430.2–430.5 ppm 3. [NOAA] 2023→2024 increase = 3.58–3.75 ppm, largest on record 4. [Cambridge MLG] Current growth rate estimated at 2.60 ± 0.28 ppm/yr 5. [Met Office] 2026 annual mean forecast = 429.4 ± 0.6 ppm 6. [Cambridge MLG] 450 ppm central crossing: April 2034 ± 17 months (95% CI: summer 2032–summer 2035) 7. [Code model] Mid-scenario annual mean in 2030 ≈ 440 ppm; May 2030 peak ≈ 442 ppm — ~8–10 ppm short of 450 # Cross-market signals - **Kalshi related**: No other CO2 concentration markets found in the series with signal - **Polymarket**: No matching markets found - **Sportsbook**: N/A # Analyst opinions and speculation - Cambridge MLG model (mlg.eng.cam.ac.uk): "No evidence of drastic action" in current data; 450 ppm by 2034 central estimate - Met Office (2026 forecast): Rise will be "slightly smaller than last 3 years" due to La Niña strengthening carbon sinks - The 22% Kalshi price appears significantly overpriced vs. physical science; reaching 450 before 2030 requires ~4.1+ ppm/yr sustained — roughly double the long-run trend and 40% above the record 2024 jump # Directional lean per outcome - **Yes (≥450 ppm before 2030)**: - *Supporting*: Record 3.75 ppm jump in 2024; emissions still rising; no credible mitigation pathway on track - *Opposing*: Even at record growth, 2030 annual mean ≈ 440 ppm; seasonal peak ≈ 442 ppm; 450 requires unprecedented ~4+ ppm/yr for 5 consecutive years; no model projects this - **No**: - *Supporting*: All scientific projections (Met Office, Cambridge MLG, code model) place 450 ppm in 2033–2035; 8–10 ppm gap remains in 2030 under any realistic scenario; La Niña moderating growth in near term # Gaps / unknowns - Market resolution criterion: annual mean, instantaneous reading, or seasonal peak? (matters at margin, not enough to change outcome) - Potential for black-swan emissions shock (e.g., runaway permafrost feedback) — not modeled, extremely low probability - Whether recent 7-day Kalshi price spike (+10 pp) reflects new information or thin-market noise # Calibration anchors - **Kalshi current YES price: 22%** — appears 3–5x too high vs. physical evidence - Physical science consensus: ~2–5% probability of 450 ppm before 2030 (only achievable via unprecedented acceleration) - Cambridge MLG 95% CI places earliest possible crossing at **summer 2032** — after the Jan 2030 deadline - Precedent: The record 2024 jump (3.75 ppm) was exceptional; sustaining 4+ ppm/yr for 5 years has no historical analog
4. Forecasters Opus + GPT-5.5 in parallel
claude-opus-4-8
high
Yes 4%
No 96%
Current CO2 ~430 ppm peak; even at record growth, 2030 peak ≈442 ppm, ~8 ppm short. All scientific projections (Cambridge MLG, Met Office) place 450 ppm crossing in 2033-2035, with earliest 95% CI at summer 2032—after the deadline. The Kalshi 22% price appears substantially overpriced versus physical evidence; reaching 450 requires unprecedented sustained 4+ ppm/yr.
gpt-5.5
high
Yes 4%
No 96%
Kalshi YES is anchored at 22%, but the physical evidence points much lower: current May 2025 seasonal peak is only about 430.5 ppm, so reaching 450 before 2030 would require roughly 4+ ppm/year sustained growth. The recent 5–10 year trend is about 2.5–2.6 ppm/year, and even the record 2024 jump of about 3.6–3.8 ppm would not be enough if merely repeated. The seasonal-peak ambiguity modestly helps YES, but projections still put May 2030 around 441–442 ppm, well short of 450. Cambridge MLG’s central crossing is 2034, with even its 95% early bound around summer 2032, so this evidence points strongly toward No despite the market price.
5. Devil's Advocate Sonnet · red-team critique
1. Both forecasters correctly identify the physical science gap but neither explicitly addresses the resolution ambiguity's upside tail: if the market resolves on any instantaneous hourly or daily reading rather than a monthly mean, the effective target is slightly lower, yet even this does not close the ~8 ppm gap to 450, so the ambiguity is correctly dismissed as immaterial. 2. Neither forecaster quantifies the black-swan tail from abrupt permafrost/methane feedback or a volcanic/ENSO cascade that could spike CO2 anomalously — the brief flags this as "not modeled, extremely low probability," but neither forecaster assigns even a rough probability to it, leaving the 4% YES figure weakly justified beyond linear extrapolation. 3. Both forecasters anchored their final probability at exactly 4%, which is suspiciously round and identical; given the Cambridge MLG 95% CI places the earliest crossing at summer 2032 (still ~2.5 years after the deadline), the physical evidence arguably supports a lower figure like 1–2%, suggesting both may be slightly over-anchored on the elevated 22% Kalshi price rather than purely on the science. 4. The forecasters do not distinguish between the probability that growth rates accelerate enough versus the probability that the resolution criterion is ambiguous enough to trigger YES — conflating these two independent probability channels could matter if one is trying to decompose the residual YES probability correctly. 5. Neither forecaster critically interrogates why the Kalshi price spiked +10 pp in 7 days despite no new emissions data; the brief notes it may be thin-market noise, but both forecasters simply accept the 22% anchor and argue against it without considering whether the spike itself contains any information about resolution-rule interpretation by sophisticated traders.
6. Aggregator Opus · final calibrated forecast
3% (80% confidence)
Yes: 3% No: 97%
Physical evidence is decisive: current seasonal peak ~430 ppm, ~8-10 ppm short of 450, and all models (Cambridge MLG central 2034, earliest 95% bound summer 2032) place crossing after the Jan 2030 deadline. Reaching 450 requires unprecedented sustained 4+ ppm/yr, well above the ~2.6 ppm trend or even the record 2024 jump. The 22% Kalshi price appears overpriced and likely reflects thin-market noise. Resolution ambiguity doesn't close the gap. I nudge slightly below the forecasters' 4% given the critique's valid point about over-anchoring on the market.
Pipeline Timing
Total pipeline time: 109.0s
Per-tool research timings shown in the Research section above.