Flagstaff Seasonal Outlook

Data: Monthly precipitation at Flagstaff Pulliam Airport (KFLG, station USW00003103) from January 1950 through . Ocean index data (PDO, AMO) from NOAA NCEI ERSSTv5, Niño 3.4 from CPC ERSSTv6 Relative (detrended — tropical-mean SST trend removed, replacing NOAA's discontinued plain-anomaly ERSSTv5 series as of June 2026).

Phase classification (leading-indicator design): Every analog year is classified using its own PDO, AMO, and Niño 3.4 state as of the current data month — the same calendar point "now" occupies — never using that year's own eventual outcome season. PDO and AMO use a 12-month trailing average ending at the current month (loosely inspired by the annual-regime framing in McCabe et al. 2004, though this tool's per-year, per-current-month analog matching is a different method from their 20-year moving-window drought-frequency analysis); ENSO uses a 3-month trailing Niño 3.4 mean ending at the current month, classified at a custom ±0.4°C threshold on the detrended Relative anomaly — narrower than NOAA's own operational ONI/RONI convention (±0.5°C), chosen to keep analog-year sample sizes usable. NOAA's own operational ENSO determination additionally requires persistence across seasons and supporting atmospheric evidence (e.g. the Southern Oscillation Index, trade winds, convection patterns), not an SST threshold alone — this tool's classification is a simplified, SST-only proxy. The outcome shown for each analog year is that year's next occurrence of the selected season or month — which lands in the following calendar year whenever the target hasn't happened yet this year (e.g. a July reference predicts the upcoming Dec–Mar winter, not the one just completed).

Composites: For each phase combination, the tool identifies all historical years with matching conditions (analog years) and displays the distribution of precipitation in those years — the full range, the middle half of outcomes, and the median. The long-term median is shown for comparison.

Probability bars: Show what fraction of analog years had precipitation below, near, or above the long-term median for that period. "Near normal" means within 15% of the long-term median.

How to use it forward-looking: The tool assumes the current AMO phase will persist — reasonable given that the AMO shifts primarily on decadal timescales. The PDO is less reliably persistent: it shows meaningful seasonal and interannual variability and can shift phase well within a single year, so treat the current PDO reading as a snapshot rather than a stable regime the tool can safely assume holds for the full outlook window. ENSO classification and the outcome-year mapping are both derived automatically from the current data month, for every season and month view alike — there's no separate winter-specific or monsoon-specific ENSO field anymore. If you want to explore what happens under a different ENSO state (e.g., a CPC forecast for developing La Niña), note which analog years in the list correspond to that state and assess them directly.

Important limitations: This tool reflects historical tendencies, not a numerical weather forecast. Small sample sizes (fewer than 10 analog years) should be interpreted cautiously — the full list of analog years is shown so you can judge the evidence yourself. Climate is always variable; analog years span a wide range of outcomes.

New Zealand Index (NZI): The Current Conditions panel shows the most recent Jul–Sep (JAS) sea surface temperature for the New Zealand Index region (25°S–40°S, 170°E–200°E) from NOAA ERSSTv6 — displayed there only October–March, the window during which it's specifically a predictor for the upcoming winter; outside that window the panel doesn't show it as "current." Mamalakis et al. (2018) identified this southwestern Pacific region as a stronger predictor of Southwest U.S. winter (Nov–Mar) precipitation than ENSO. The Jul–Sep NZI correlates about r ≈ −0.55 with Nov–Mar winter precipitation in the recent period. A stronger value, about −0.7, appears with the Sep–Nov window, which is not available yet. These are the paper's regional Southwest U.S. averages, not Flagstaff values. Warm JAS SST in this region (above the 1950–present median of 16.907°C) is associated with drier Southwest winters; cool SST with wetter winters. This signal was weaker before ~1983. The paper reports correlations did not exceed 0.45 in 1950–1983. When Winter is selected, an annotation appears below the outlook showing the current NZI direction. NZI does not filter the analog years — sample sizes are too small for a reliable three-way split.

Niño 1+2 (monsoon context): The Current Conditions panel shows the most recent monthly Niño 1+2 anomaly (eastern equatorial Pacific, 0–10°S, 80–90°W) and, once available, the most recent complete pre-monsoon spring 3-month mean (AMJ, MAM, or FMA — whichever window has most recently closed) from NOAA CPC's OISSTv2.1 Relative (detrended) index, available from 1981 onward. This region is the most volatile ENSO index — it warms first during developing El Niño events. Whether warm spring Niño 1+2 suppresses monsoon moisture flux into the interior Southwest is treated here as a hypothesis under investigation, not an established relationship: Higgins et al. (1999) documented meaningful interannual variability in the North American warm-season precipitation regime, but Adams & Comrie (1997) concluded that monsoon interannual variability overall is not strongly linked to ENSO or other common circulation indices — a caution worth weighing against this annotation. The signal is weaker than ENSO's relationship with winter precipitation and does not filter the analog years.

Southern Oscillation Index (SOI) & teleconnection confidence: The Current Conditions panel shows the most recent monthly SOI (Australian Bureau of Meteorology, Tahiti-Darwin pressure difference) alongside the most recent complete June–November seasonal average — the lag window Redmond & Koch (1991) identified as predictive of the following October–March winter precipitation — displayed there only October–March, the window during which it's specifically that predictor; outside that window the panel doesn't show it as "current." SOI and Niño 3.4 normally move in opposite directions (BOM's convention: positive SOI = La Niña-like, negative = El Niño-like). McCabe & Dettinger (1999) found that when SOI and Niño 3.4 diverge from this expected relationship, ENSO's teleconnection to western US precipitation weakens or strengthens depending on the concurrent PDO phase. When Winter is selected, an annotation below the outlook states the current teleconnection confidence state (computed from SOI, Niño 3.4, and PDO — see the workbook's Climate_Indices sheet for the full classification logic). SOI does not filter the analog years or change the probability statement itself — it is a contextual qualifier, like NZI. A parallel note using the July–September SOI average appears for monsoon-season views.

References

1. McCabe, G.J., Palecki, M.A., and Betancourt, J.L. (2004). Pacific and Atlantic Ocean influences on multidecadal drought frequency in the United States. Proceedings of the National Academy of Sciences, 101(12), 4136–4141.
2. Mamalakis, A., Yu, J.-Y., Randerson, J.T., AghaKouchak, A., and Foufoula-Georgiou, E. (2018). A new interhemispheric teleconnection increases predictability of winter precipitation in southwestern US. Nature Communications, 9, 2332.
3. Adams, D.K. and Comrie, A.C. (1997). The North American Monsoon. Bulletin of the American Meteorological Society, 78(10), 2197–2214.
4. Higgins, R.W., Yao, Y., and Wang, X.L. (1997). Influence of the North American Monsoon System on the United States summer precipitation regime. Journal of Climate, 10(10), 2600–2622.
5. Higgins, R.W., et al. (1999). Interannual variability of the North American warm season precipitation regime. Journal of Climate, 12(3), 653–680.
6. Mitchell, D.L., Ivanova, D., Rabin, R., Brown, T.J., and Redmond, K. (2002). Gulf of California sea surface temperatures and the North American monsoon: Implications for forecasting. Journal of Climate, 15(17), 2261–2281.
7. Redmond, K.T. and Koch, R.W. (1991). Surface climate and streamflow variability in the western United States and their relationship to large-scale circulation indices. Water Resources Research, 27(9), 2381–2399.
8. McCabe, G.J. and Dettinger, M.D. (1999). Decadal variations in the strength of ENSO teleconnections with precipitation in the western United States. International Journal of Climatology, 19(13), 1399–1410.