Land use change and carbon storage in the Qinling-Loess Plateau transition zone: processes, drivers and multi-scenario projections for the Tianshui city-region, Northwest China

Transition zones between contrasting biomes hold a large share of regional carbon, and land use inside them shifts quickly, yet the link between the two is still weakly resolved at landscape scale. This study works on the Tianshui city-region (14,267.26 km2), which lies on the belt where the northern flank of the Qinling Mountains meets the southern margin of the Loess Plateau in Northwest China. Precipitation falls from about 800 mm in the south-east to about 400 mm in the north-west, so the area sits on the dry sub-humid to semi-arid margin rather than inside the drylands. Land use for 2000, 2010 and 2023 was combined with carbon densities measured at 45 field sites in the InVEST carbon model, land use for 2030 was projected with PLUS under three scenarios, and the spatial pattern of carbon storage was analyzed with the Geodetector. Two expectations were tested: that the carbon released by shrinking cultivated land would outweigh the carbon gained from expanding forest and grassland; and that precipitation and soil type would set the ceiling of carbon storage while socioeconomic pressure would decide how closely that ceiling is approached. Carbon storage fell by 0.69 Tg C (6.89 × 105 Mg C) between 2000 and 2023, although forest gained 68.95 km2 and grassland 135.51 km2 over the same period. Most of that decline is attributable to the 321.26 km2 of cultivated land that was lost. The 2030 scenarios separate less than we had assumed: ecological protection loses 0.57 Tg C, natural development 0.66 Tg C and urban development 0.84 Tg C. Soil type and annual precipitation returned the highest single-factor q-values, and pairing either of them with nighttime light intensity or population density lifted the explained variance above the sum of the two factors acting alone. Two points follow for dryland management. Carbon already held in cultivated soil deserves as much attention as new planting, and a seven-year horizon works against restoration, since young stands carry little biomass by 2030.