Details of ENSO Risk Calculations
To calculate the relative risk of an extreme of a variable given the state of ENSO, the SOI index was first sorted for each season for the 100 years available. Then, the top and bottom 20 years (i.e. the La Niña's and El Niño's) were examined with respect to the variable in question. This variable was sorted as well for each climate division. Then, a contingency table was constructed for the number of years that were both El Niño and warm, El Niño and cold, La Niña and warm and La Niña and cold.
ENSO Phase | Extreme Cold | Neutral | Extreme Warm |
---|---|---|---|
El Niño | 4 | 12 | 4 |
Neutral | 12 | 36 | 12 |
La Niña | 4 | 12 | 4 |
For the top/bottom extreme ENSO years, we would expect
(20/100)*20=4.0or the 4 corners of the table above as the number of extreme warm/cold years that would match by chance.
For each climate division, we plotted
[(actual number of matches-4)/(matches expected by chance)]*100for the warm and cold cases. Regions that represent increases that are significant at the 82.7%,93.6%, and 98.2% level, respectively, are shaded.
Since there are many climate divisions and 4 seasons, one would expect a certain number of chance occurrences where the number of matches is significantly higher or lower than the 4 expected. Field significance tests addressed this question and results (not shown) indicate that our results are significant.