Carbon stock dynamics across land use/land cover in the urbanizing Akaki river catchment, Ethiopia

Abstract

Terrestrial ecosystems play a pivotal role in climate regulation, yet rapid rural–urban land-use and land-cover (LULC) change is profoundly reshaping carbon dynamics across developing landscape. Conventional static assessments often fail to capture these spatiotemporal dynamics, leading to underestimation of carbon fluxes and emissions. This study quantifies spatiotemporal carbon storage and sequestration dynamics across the rapidly urbanizing Akaki River Catchment over the 1984–2024 period using multi-temporal benchmark years (1984, 1994, 2004, 2014, and 2024), and provides the first spatially explicit long-term rural–urban landscape assessment through the novel integration of medium- and high-resolution remote sensing-derived LULC maps, ARIES vegetation carbon datasets, and InVEST modeling. Aboveground biomass (AGB), belowground biomass (BGB), and soil organic carbon (SOC) were derived from the integrated datasets, and carbon stocks were subsequently quantified at both pixel and land-use category levels using spatial zonal statistics in ArcGIS Pro. The results revealed pronounced spatiotemporal carbon stock dynamics across the Akaki River Catchment, driven primarily by land-use conversion and rapid urbanization. Woodland and grassland total carbon stocks declined markedly from 91,738 to 13,770 tC and from 100,164 to 21,691 tC, respectively, between 1984 and 2024, whereas plantation forests increased from 15,058 to 39,537 tC. Although croplands exhibited relatively low per-hectare carbon storage, their extensive spatial coverage resulted in the largest total carbon stocks, increasing from 169,577 to 253,378 tC. Over the study period, plantation forests recorded the highest net carbon gain (+ 24,479 tC; +162.7%), while woodlands (− 85.0%) and grasslands (− 78.3%) experienced substantial carbon depletion. Cumulative model simulations further indicated that urban expansion contributed approximately 9.4 million tC of carbon loss, with the most intense emissions occurring during 2004–2014, coinciding with accelerated rural–urban landscape transformation. Carbon storage in this study refers to total ecosystem carbon stocks within LULC categories, whereas carbon sequestration represents net gains or losses driven by land-use transitions over time. Across the Akaki River Catchment, gross carbon losses reached approximately 9.62 million tC, partially offset by gains from cropland and plantation forests, resulting in a net carbon loss of about 4.82 million tC over the 1984–2024 period. These estimates, derived from integrated remote sensing data and ecosystem service models (InVEST and ARIES), inherently reflect uncertainties associated with input data quality, model structure, and parameterization. The results demonstrate the dominant influence of unplanned land-use change, particularly rapid urban expansion, underscoring the urgent need for spatially explicit land-use zoning, restoration of high-carbon ecosystems, and integration of carbon-sensitive rural–urban planning to enhance carbon sequestration and maintain long-term ecosystem carbon stability in rapidly urbanizing highland landscapes.