Prospects Of Long-Lead Seasonal Wildland Fire Predictions For Oahu, Hawaii

We examined statistical relationships between the seasonal Southern Oscillation Index

(SOI) and annual total acreages burned and the number of fire in Hawaii. The results show

that summer is the most favorable season for fire activities. A composite of total acres burned

during four ENSO events reveals that a large total acres burned is likely to occur from spring

to summer in the year following an ENSO event. The correlation is most significant between

the total acres burned in summer and the SOI of the antecedent winter. This relationship

provides a potential for long-lead (i.e. 2 seasons in advance) prediction of wildfire activity in

Hawaii.

Simple regression and logistic regression models are developed to test fire predictability.

Large year-to-year variations of fire data impair the efficiency of the simple regression

prediction. To overcome this problem, logistic regression is applied to predict events of large

acreages burned by wildfires. The goodness of predictions is measured by specificity,

sensitivity, and correctness using a cross validation method. A comparison between the

simple regression and logistic regression shows that a logistic regression model is better in

seasonal fire prediction than the simple regression model.