The Effect of Land Use Change Imposed on Carbon storage and
Multi-Scenario Prediction in Hainan Island Using InVEST and CA-Markov
Models
Abstract
As a fundamental element of global carbon storage, the storage carbon in
terrestrial ecosystem is significant for climate change mitigation. Land
use/cover change (LUCC) is a main impact element of ecosystems’ carbon
storage. Evaluating the relation between land use change and carbon
storage is vital for lowering global carbon emissions. Taking Hainan
Island as an example, this paper employs the InVEST as well as the
CA-Markov models to assess and predict how different land use affects
carbon storage in various situations from 2000 to 2020 and from 2030 to
2050 on Hainan Island. The influence factors, together with driving
mechanisms of carbon storage spatial distribution are quantitatively
analyzed as well in this paper. The results demonstrate that, from 2000
to 2020, Hainan Island’s net increase in built land was 605.49 km2,
representing a growth rate of 77.05%. Over the last 20 years, Hainan
Island’s carbon storage and density have decreased by 5.90 Tg and 1.75
Mg/hm2, respectively. The sharp rise in built land mainly makes the
carbon storage decline. From 2030 to 2050, land use changes on Hainan
Island are expected to result in differing degrees of carbon storage
loss in various scenarios. In 2050, Hainan Island’s carbon storage will
decline by 17.36 Tg in the Natural Development Scenario (NDS), 13.61 Tg
in the Farmland Protection Scenario (FPS), and 8.06 Tg in the Ecological
Protection Scenario (EPS) compared to 2020. The EPS can efficiently
maintain carbon sequestration capability, but it cannot effectively
prevent cropland area loss. Regarding the carbon storage’s spatial
distribution, Hainan Island generally exhibits a pattern of high carbon
storages in the low and middle carbon storages in the surrounding areas.
Geographic detection presented the spatial differentiation of carbon
storage in Hainan Island is mainly influenced by factors like slope,
land use intensity, and DEM, as well as its interaction with other
factors is significantly strengthened (p<0.05)