Seasonal dynamics of soil CO2 efflux across land use systems and implications for mitigation under a changing climate

作者：Famoussa Dembélé, Stephen Adu‐Bredu, Reginald Tang Guuroh, Eunice Okyere Agyapong, Padmore Boateng Ansah, Nat Owusu-Prempeh, Bismark Owusu, Dzigbordi Solomon-Ayeh, Suleiman Usman Yunusa, Aboubakar Bengaly, Larissa Raatz, Roman Hinz, Atinuke Adebanji, Samuel Oppong, Rüdiger Schaldach, Amanuel W. Gebremichael, Anja Linstädter · 发表于：Scientific Reports · 年份：2026 · DOI：10.1038/s41598-026-50708-7 · 研究领域：Soil Carbon and Nitrogen Dynamics、Plant Water Relations and Carbon Dynamics、Agroforestry and silvopastoral systems

Abstract Carbon dioxide (CO 2 ) is a major greenhouse gas driving climate change. In Ghana, the Agriculture, Forestry, and Other Land Use (AFOLU) sector remains a significant source of CO 2 emissions, largely due to land use change and degradation. This study assessed seasonal dynamics of soil respiration rates (SRR) across four land-use types, forest, fallow, maize, and rice fields, within the semi-deciduous forest zone of Ghana. The aim was to provide baseline data and identify key soil and environmental factors influencing SRR across these systems. SRR was measured twice monthly over 13 months using a closed-chamber system, with concurrent measurements of soil moisture and temperature, while baseline soil properties (organic matter, pH, and texture) were determined from initial soil sampling. Correlation and stepwise regression analyses were performed to determine the variables most strongly associated with SRR. Results revealed clear temporal and land-use differences, although seasonal patterns were not uniform across sites. Fallow land and croplands (maize and rice fields) recorded the highest SRR values within the study area, whereas forest plots consistently showed the lowest efflux, largely due to persistent moisture limitation rather than temperature or substrate availability. Soil OM, pH, moisture, and silt content were the most influential predictors of SRR, with the final regression model explaining 58% of the observed variability. These findings highlight the imp...