Historical analogue confidence
JADE ANALYSISThe question is not which year looks similar. It is how much confidence the historical outcome deserves as an analogue for today.
Current state · 2026
- oni
- -0.54 z -0.58
- soi
- +1.33 z +0.78
- iod
- -0.40 z -0.57
- antecedent_mm
- +3070.80 z +4.08
- Outcome
- 1,566.6 mm
Analogue usefulness
0.3501
1 − (stdev of closest-8 outcomes / climatological stdev). Positive values mean analogues tighten the outcome distribution. Not a probability.
Closest-8 outcome range 607.3–1,491.2 mm · climatology σ 505.7 mm
Not in the state vector
- · MJO
- · IPO/PDO
- · SST fields
- · atmospheric moisture
- · soil moisture
- · synoptic regime
Closest historical states
| Year | Distance | Outcome | Matches | Differs |
|---|---|---|---|---|
| 2001 | 2.1778 | 916 mm | oni, soi, iod | antecedent_mm |
| 1975 | 2.2802 | 681.6 mm | oni, soi, iod | antecedent_mm |
| 1999 | 3.0026 | 932.8 mm | soi, iod | antecedent_mm |
| 1954 | 3.0997 | 1,307 mm | iod | antecedent_mm |
| 1961 | 3.2232 | 607.3 mm | oni, soi | antecedent_mm |
| 2012 | 3.2276 | 1,286.2 mm | oni, soi | iod, antecedent_mm |
| 2010 | 3.4775 | 1,491.2 mm | — | oni, soi, antecedent_mm |
| 1956 | 3.5332 | 1,331 mm | soi, iod | antecedent_mm |
- 2001: matched oni, soi, iod; differed antecedent_mm; wet-season outcome 916.0 mm.
- 1975: matched oni, soi, iod; differed antecedent_mm; wet-season outcome 681.6 mm.
- 1999: matched soi, iod; differed antecedent_mm; wet-season outcome 932.8 mm.
- 1954: matched iod; differed antecedent_mm; wet-season outcome 1307.0 mm.
- 1961: matched oni, soi; differed antecedent_mm; wet-season outcome 607.3 mm.
- 2012: matched oni, soi; differed iod, antecedent_mm; wet-season outcome 1286.2 mm.
- 2010: matched nothing closely; differed oni, soi, antecedent_mm; wet-season outcome 1491.2 mm.
- 1956: matched soi, iod; differed antecedent_mm; wet-season outcome 1331.0 mm.
Euclidean distance on z-scored [ONI SON–DJF, SOI Sep–Feb, IOD ASO–NDJ, previous wet-season rainfall]. Equal weights. No SST fields, MJO or soil moisture.
- · Scalar indices are not a substitute for field similarity.
- · No universal percentage is reported because the distance has not been independently validated.
- · Closest years can still produce widely different rainfall if a cyclone or monsoon burst occurs.