This study examines the complex relationship between renewable energy adoption, fossil fuel dependency, and CO₂ emissions in Fiji's electricity generation sector, utilizing multivariate regression modeling from 1980 to 2009. Building on the literature examining Pacific Island energy transitions, we develop an integrated model that incorporates total CO₂ emissions from liquid and solid fuels, emissions per unit of GDP, population dynamics, renewable energy generation, and installed capacity. Our findings reveal a robust negative correlation (r = −0.656) between solid fuel consumption and CO₂ emissions, indicating the successful enforcement of green energy initiatives. In contrast, a strong positive relationship exists between liquid fuel consumption and emissions (r = 0.969). The regression model achieves an R² of 0.9990 with a statistically significant F-value (p < 0.001), indicating that the associated variables adequately explain the variance in emissions. Critically, our analysis reveals a trade-off between renewable energy expansion and total emissions reduction, suggesting that technological transitions alone are insufficient without complementary demand-side management and fuel-switching strategies. The model projects CO₂ emissions of 1,192,303 metric tons by 2016, representing a 10.5% increase from 2009 levels—a trajectory inconsistent with regional climate commitments. We recommend integrated policy reforms, including the accelerated adoption of biodiesel, the conversion of heavy fuel oil (HFO) with enhanced environmental controls, grid demand management, and strategic investment in hydroelectric capacity. This study offers novel insights into the Pacific Island energy literature by demonstrating that emission control necessitates the simultaneous consideration of electricity supply diversity, population growth dynamics, and economic development trajectories.

Keywords

CO₂ emissions, renewable energy transition, Pacific Island economies, multivariate regression, energy policy, Fiji, climate change mitigation, electricity generation forecasting