Hydroclimatic vulnerability and electricity adequacy in Albania from 2000 to 2024 using generalized additive models

Albania’s hydropower-dominated electricity system is structurally sensitive to hydro-climatic variability, generating substantial fluctuations in self-sufficiency and import dependence. Using annual electricity balance data for 2000–2024, this study examines how hydro-climatic and structural factors shape three key electricity-balance indicators: self-sufficiency (SS), import ratio (IR), and loss rate (LR). Within an Electricity Balance Performance Index (EBPI) framework, outcome-specific shrinkage-penalized generalized additive models (GAMs) are estimated and evaluated using leave-one-year-out cross-validation. The results identify hydrological availability, proxied by an inflow index, as the dominant driver of adequacy outcomes. The estimated smooths indicate that hydrological effects are approximately monotonic and near-linear over the observed historical range, suggesting that hydro-climatic variability is transmitted into adequacy outcomes in a broadly proportional manner. Higher inflow is associated with higher self-sufficiency and lower import dependence, while industrial activity provides additional explanatory content for import intensity. In contrast, temperature anomalies and GDP per capita contribute limited separable information once hydrological and structural dynamics are accounted for. Loss dynamics follow a different pattern, being driven primarily by a nonlinear time trend associated with gradual structural and institutional change rather than short-term hydro-climatic shocks. Scenario mapping across dry, normal, and wet inflow regimes (P10/P50/P90) further highlights the policy significance of these relationships, with gross import intensity declining by approximately 15 percentage points between dry and wet conditions. Robustness checks using offset-level reformulations confirm that the main findings are not artefacts of ratio construction.