This study evaluates the trade-offs between solar photovoltaic (PV) deployment, land-use change, and vegetation carbon storage in Taiwan. Satellite imagery and official land-use datasets were used as primary data sources, while Geographic Information Systems (GIS) – based spatial overlay analysis was employed to map nationwide solar panel distributions and quantify land-use conversions. Spectral indices, including the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Built-up Index (NDBI), were applied as supporting tools for solar panel identification and land-cover discrimination. Results indicate that southern Taiwan—particularly Pingtung County—has experienced the most intensive expansion of ground-mounted and floating PV systems, accompanied by substantial conversions of farmland and small woodlands. In Checheng Township, agricultural land declined sharply as electricity facilities and anthropogenic land uses expanded. Carbon accounting and biomass-based modeling reveal that existing vegetation, even in small woodland patches, provides persistent long-term carbon storage that is forfeited when converted to PV sites. Quantitative comparison shows that solar PV deployment yields higher and more stable annual carbon reduction benefits (5. 65–6. 57 kt-CO₂e/yr) than the mean annual carbon sequestration of vegetation (approximately 0. 38 kt-CO₂e/yr), although vegetation contributes important long-term carbon stocks and ecosystem services. These findings highlight a clear trade-off between immediate energy-driven emission reductions and the loss of nature-based carbon sinks. The study recommends prioritizing rooftop PV installations and the reuse of idle or low-ecological-value lands to minimize land-use conflicts. Incorporating vegetation carbon storage into renewable energy siting criteria can better align net-zero targets with sustainable land management.
Chang et al. (Sun,) studied this question.