Spatiotemporal evolution of ecosystem carbon storage under land use/land cover dynamics in the coastal region of Central Vietnam

Abstract

Ecosystem carbon storage is a cost-effective strategy for global climate change mitigation, and its fluctuation is markedly shaped by land use/land cover (LULC) dynamics. Taking Danang city as an example of Central Coastal Vietnam, this study aims to assess LULC changes and analyze the spatiotemporal evolution of carbon storage from 2023 to 2050 under four LULC change scenarios, including natural trend scenario (NTS), ecological protection scenario (EPS), economic development scenario (EDS), and cropland protection scenario (CPS), by integrating the support vector machine-cellular automata-Markov (SVM-CA-Markov) model and the InVEST model. The Optimal Parameters-based Geographical Detector (OPGD) model was subsequently employed to elucidate the impacts of driving factors on the spatial distribution of carbon storage. The results showed that, from 2007 to 2023, Danang city experienced a dramatic back-and-forth transformation between LULC types, with the predominant transitions being from natural forest to acacia tree-dominated plantation forest (6492.31 ha), and from cropland to settlements, acacia tree-dominated plantation forest, and other land (5483.05 ha, 3763.66 ha, 2762.35 ha, respectively). Between 2023 and 2050, LULC transformations in Danang city are projected to yield varying degrees of carbon storage levels across different scenarios. Specifically, carbon storage is anticipated to dwindle by 0.221 Mt, 0.223 Mt, and 0.298 Mt under NTS, EDS, and CPS, respectively, while enhancing by 0.141 Mt under EPS. Regarding the spatial distribution of carbon storage, high values will be chiefly found in the western high-elevation mountainous region, while low values will be concentrated mostly in the eastern lower-lying areas of the city. Additionally, elevation and temperature acted as the two most significant driving factors influencing the spatial distribution of carbon storage, with Q values of 0.88 and 0.86 (p-value < 0.05), respectively. For interaction detection, the combination of elevation and soil exhibited a synergistic reinforcement effect on the spatial partitioning of carbon storage, with a high Q value of 0.9566 (p-value < 0.05). Our study highlights the necessity of ecological conservation measures in Danang city in the on-track pursuit of national net-zero carbon emissions by 2050.