Abstract: Soil organic carbon (SOC) stock can play an important role in the soil carbon (C) sequestration in sustainable agriculture. It is of great significance for the global carbon cycle and food security. Farm practices are often required to realize the C sequestration of the “4 per 1000” initiative in recent years. Meanwhile, some challenges also remain in estimating SOC under climate change. However, it is still lacking in the interaction effects of farm recommendation and climate change on SOC storage. This study aims to predict black soil carbon sequestration potential for optimal farm recommendation using meta-analysis and the denitrification-decomposition (DNDC) model. A case study was taken of Lishu County in Jilin Province, China. Firstly, 2275 groups of field observations were collected from 187 articles in the black soil region of Northeast China. Agricultural practices were investigated in SOC stocks using meta-analysis, including tillage (no-tillage (NT), reduced tillage (RT), and deep tillage (DT)), straw return (SR), manure fertilization (MF), and chemical fertilization (CF). Secondly, the temporal dynamics of SOC stock were verified and then predicted in 1980-2024 and 2025-2100 under different farm practices using the DNDC model. Lastly, the multiple scenarios were designed to consider climate change and farm strategies. Soil carbon sequestration potential was predicted for the farmland in 2025-2100. The results showed that: 1) The MF shared the greatest effect on the increase of SOC, with an increase of 19.44%, when the application amounts <20 t/hm², 23.97% in the 20-30 t/hm², and 31.13% in the >30 t/hm² groups of application amounts. The MF with NT and RT increased the SOC, with the effect size increasing from 7.41% (NT) to 13.21% (NT+MF) and from 10.78% (RT) to 20.09% (RT+MF), respectively. Among them, the greatest effect size (44.45%) was achieved under the MF+CF combination, with the MF application amount >30 t/hm2. A long application period (>10 a) shared a greater effect (40.12%) on SOC sequestration than short application periods (<3 a (17.16%) and 3-10 a (20.34%)). The CF+SR+MF combination made a larger effect than that from the separate practice, and all effect sizes increased over different duration periods, compared with the separate CF practice. DNDC model verified that there was a significant linear relationship (R²=0.93, RMSE=0.60 g/kg) for all the SOC simulations, compared with the measured values. All MF+CF, RT+MF, and MF+CF+SR practices achieved a considerably high carbon sequestration on the SOC in the short term (<3a) under the MF application amounts of 450 t/hm². Moreover, the MF+CF+SR achieved a higher carbon sequestration on the SOC in the medium term (3-10a) and long term (>10a) under the SR ratios of 1/3 or 1/2 conditions. There were significant differences in the effects of various farm modes on soil carbon sequestration. MF practice shared the most outstanding increase in the SOC storage. The carbon sequestration potential was predicted in farmland under future climate scenarios. The DNDC model can be expected to simulate historical changes in the carbon stocks with high accuracy. This finding can provide a strong reference for the farmland quality and high yield under climate change in the black soil region of Northeast China.