Analysis of the Driving Factors and Decoupling of Carbon Emissions from Energy Consumption in Hainan Province, China

High-energy-consuming and high-emission industries have made enormous contributions to economic development, but their carbon emissions are also substantial. To achieve the “dual carbon” goals as early as possible, this study takes Hainan Province, China, as the study area and employs the LMDI method to decompose the factors influencing carbon emissions. The Tapio model is also used to analyze the decoupling relationship between the driving factors and carbon emissions. The results show the following: (1) Carbon emissions in Hainan Province from 2007 to 2022 exhibited an overall upward trend, with an average annual growth rate of 5.75%. Various oil products accounted for an average share of over 40.11%, but the share of electricity increased, while that of oil decreased. The sectors, ordered from the highest to lowest carbon emissions, are: industry > transportation > residential > agriculture, forestry, animal husbandry, and fishery. (2) The decomposition results indicate that economic output, energy structure, and population size have positive effects on carbon emissions, while energy intensity and industrial structure have negative effects. At the sectoral level, the energy structure factor has a negative effect only on the transportation sector, and positive effects on all other sectors. The energy intensity factor has negative effects on all sectors except “other sectors” and the residential sector, with a cumulative contribution of 2003.65 × 104 tonnes of carbon emissions. The industrial structure factor has negative effects on carbon emissions across all sectors, with a cumulative contribution of 1568.23 × 104 tonnes. The economic output factor promotes emissions in all sectors, with a cumulative increase of 5787.35 × 104 tonnes, of which 2740.88 × 104 tonnes are from the industrial sector. The population factor also promotes emissions across all sectors, with a cumulative contribution of 549.21 × 104 tonnes. (3) The decoupling model analysis shows that from 2007 to 2008, the decoupling state was predominantly an unfavorable negative decoupling. From 2008 to 2010, it shifted to a favorable positive decoupling, but from 2010 to 2011 it returned to an unfavorable negative decoupling. From 2011 to 2022, the decoupling index declined from 1.43 to 0.13, indicating an overall favorable weak decoupling state. (4) The decoupling effects of individual influencing factors reveal that in the 2007–2008 period, the carbon emission decoupling index was mainly composed of the energy intensity effect and the economic output effect. In the 2012–2013 period, the energy structure effect did not change significantly and remained in a weak decoupling state, while the energy intensity effect declined markedly, changing the decoupling state from weak to strong decoupling. The industrial structure effect remained in a strong decoupling state. In 2017–2018, the economic output effect changed from an expansive coupling state to a weak decoupling state, while the other effects all showed relatively favorable positive decoupling states. In 2021–2022, all effects exhibited favorable positive decoupling states, among which the energy structure and energy intensity effects showed strong decoupling. Finally, this study provides a case study for the development of Hainan as a Free Trade Port, a tourism island, a petroleum- and aviation-fuel-intensive province, and a pilot ecological civilization zone.