Impact of El Niño on Panama Canal Hydrology and Interoceanic Port Operations Using Statistical and AI

The El Niño-Southern Oscillation (ENSO) phenomenon alters the hydrology of the Panama Canal basin, resulting in reduced levels of the Gatun and Alajuela reservoirs (National Oceanic and Atmospheric Administration (NOAA, 2025)). This reduction in freshwater availability limits lock operations, forcing the Panama Canal Authority (ACP) to impose operational restrictions on transit capacity (Fuentes & Munim, 2025). Such restrictions reduce the number of effective transits and increase the waiting times (CWT) of non-prioritized cargo ships and generate volatility by reserving slots (Fuentes & Munim, 2025). The ports of the interoceanic area operating in the same space as the Canal may also be affected due to the uncertainty of the latter's restrictions; this scenario has been scarcely studied. The objective of this work is to evaluate the causal relationship of the ENSO warm phase→ hydrology→ Panama Canal restrictions→ transits→ CWT waiting times→ port performance of this interoceanic area, integrating climate, hydrological, operational, and port data through hybrid models based on statistics, and artificial intelligence (Fuentes & Munim, 2025). The study integrates climate data from the ONI Index (NOAA, 2025), hydrological levels of the Gatun and Alajuela reservoirs, operational variables of the Panama Canal and port data of the interoceanic area. A hybrid methodology is implemented, integrating traditional statistical models (ARIMA, SARIMA, SARIMAX) are employed as analytical autoregressive structures, and capturing the linear relationships of the warm phase of ENSO→ hydrology→ Panama Canal constraints. This primary characterization provides the foundation for a subsequent and natural transition toward training and predictive evaluation. Model performance is assessed by comparing projections against observed data using error metrics such RMSE. Furthermore, using machine learning techniques, the residual of the model is analyzed using Random Forest and GRU networks to identify nonlinear patterns and hydrometeorological anomalies (Huang et al., 2022). This hybrid approach allows for an integrated assessment of climate-operational transmission to ports. Preliminary analyses show that warm episodes of ENSO (El Niño) are associated with significant reductions in precipitation and marked declines in the levels of the Gatun and Alajuela reservoirs (NOAA, 2025). These hydrological deficits led the Panama Canal Authority to implement operational restrictions, reducing effective transits and increasing Canal Waiting Time (CWT) for non-prioritized vessels (Fuentes & Munim, 2025). Initial observations also suggest potential indirect effects on interoceanic ports, reflected in increases in Port Waiting Time (PWT), alterations in routing patterns, and variability in monthly cargo flows. Since both systems share the same operational space and anchorage zones, Canal congestion appears to propagate toward port operation, revealing a potential alignment between Canal and port queues during ENSO events. Finally, the hybrid models (SARIMAX + ML) show superior ability to detect hydrometeorological and operational anomalies compared to traditional linear models, reinforcing the hypothesis of climate to operation transmission within the interoceanic logistic system. The study constitutes an integrated approach that quantifies impacts and explores causal pathways of the warm phase of ENSO (El Niño) with hydrological variability, Panama Canal operational restrictions, transit dynamics, Canal Waiting Time, and the performance of Interoceanic ports. It established a methodological pathway from descriptive analysis toward predictive assessment. By combining statistical modeling and machine learning techniques it allows capturing both regular patterns and El Niño-induced disturbances, providing a useful methodological framework to assess the climate vulnerability of the interoceanic logistic system of Panama.