Abstract
An empirical statistical scheme for predicting the September to December fires in New Caledonia in SW Pacific using a cross-validated Generalized Linear Model has been developed for the 2000-2010 period. The predictor employs July sea surface temperatures (SST) recorded over the Niño 4 box (160°-210°E, 5°S-5°N) which is closely related to austral spring (September-November) rainfall anomalies across New Caledonia. The correlation between the log of observed and simulated total burned area across New Caledonia is 0.87. A decrease in the local-scale skill (median correlation between the log of observed and simulated total burned area in a 20-km radius around rain gauge = 0.46) around the main town (Nouméa) and its suburb in the SW of 'Grande Terre', and also in northern New Caledonia, could be associated either to a weaker climatic forcing from Niño 4 SST index or a small-scale climatic forcing not linearly related to El Niño Southern Oscillation (ENSO) phenomenon. More likely, the decrease is tied to the influence of human-driven factors that blurs the regional-scale climatic signal mostly associated with Central Pacific ENSO events.