Decoding Climate Teleconnections: An Asymmetric Time-Lag Analysis of the Oceanic Niño Index and Chikungunya Dynamics in Indonesia, 2013–2023

Raw macro-climatic and epidemiological time series are inherently prone to non-stationary biases and linear seasonal trends, causing spurious correlations that mask genuine environmental teleconnections. This study aimed to evaluate and measure the long-distance relationship between the global Oceanic Niño Index (ONI) and local Chikungunya virus (CHIKV) transmission dynamics in Indonesia from January 2012 to December 2023 (n = 144 months). Methodologically, a quantitative associative time-series design was employed within RStudio. Both datasets exhibited non-normal distributions (p=0.000) and raw non-stationarity via Augmented Dickey-Fuller testing (p>0.05). To eliminate mathematical bias, a first-order differencing transformation was applied, successfully achieving stationarity (p=0.0100). The Cross-Correlation Function (CCF) was then calculated across an interaction window from Lag -6 to Lag +6 months. The results revealed a highly dynamic double-peak epidemiological pattern, peaking in 2013 (15,324 cases) and 2023 (6,049 cases). While no instantaneous correlation occurred at Lag 0 (r=-0.0180), a significant, optimal positive cross-correlation emerged at Lag -6 (r=0.1246). This confirms an asymmetric teleconnection where an increase in the ONI—shifting toward an El Niño phase—preceding the current period by six months directly drives increased Chikungunya outbreaks in Indonesia, providing a vital baseline for climate-based Early Warning Systems.