基于DNDC模型的气候变化下长期施肥对紫色土区玉米产量和土壤有机碳影响的分析

Analysis of the effect of long-term fertilization on summer maize yield and soil organic carbon in purple soil areas under climate change based on the DNDC model

  • 摘要: 农田土壤有机碳(soil organic carbon, SOC)储量的动态变化与施肥策略和气候变化密切相关, 为探究气候变化下增产固碳的施肥策略, 以紫色土区玉米农田为研究对象, 利用长期定位监测试验对反硝化-分解模型(Denitrification-Decomposition, DNDC)进行校正与验证, 并利用校正后的DNDC模型与CMIP6模式未来情景气候数据结合, 分析了基准、SSP2-4.5和SSP5-8.5 3种情景下无施肥对照(CK)、有机肥和化肥配施(T1)、常规单施化肥(T2)和单施化肥增量(T3) 4种施肥措施对玉米产量和SOC储量的影响。结果表明, 研究区在两种SSP情景下降水量和温度随时间推移均呈上升趋势, 且SSP5-8.5情景降水量和温度的上升趋势更明显。各处理玉米产量和表层SOC含量实测值和模拟值的平均偏差误差(MBE)表现良好, 标准均方根误差(NRMSE)为0.07~0.19, 决定系数(R2)为0.74~0.98, 校正后的DNDC模型模拟效果较好, 能够用于紫色土区玉米产量和SOC储量的模拟。在T1处理下, 各未来情景下SOC储量在大多数年代下无显著差异, 气候变化对有机肥和化肥配施措施下的玉米产量和SOC储量影响较小; 与基准情景相比, SSP2-4.5和SSP5-8.5情景能够显著提高无施肥处理的玉米产量, 但会显著减少常规单施化肥处理的玉米产量。CK、T2和T3处理下各情景SOC储量随时间推移均呈降低趋势, T1处理下各情景SOC储量随时间推移均呈升高趋势; 2100年各处理SOC储量表现为T1>T3≈T2>CK。不同施肥措施下, 2100年SSP5-8.5情景的SOC损失量均高于SSP2-4.5情景, 高排放情景会导致更多SOC损失。在未来气候情景下, 常规施肥模式虽然能够保证作物稳产, 但不利于农田固碳, 而有机肥和化肥配施是紫色土耕地实现固碳增产较为适宜的施肥管理模式。

     

    Abstract: Dynamic changes in the soil organic carbon (SOC) storage in farmland are closely related to fertilization strategies and climate change. To explore fertilization strategies for increasing the crop yield and sequestering carbon under climate change, this study used summer maize farmland in purple soil areas as the research object. The Denitrification-Decomposition (DNDC) model was calibrated and validated using long-term-positioning monitoring experiments, and the calibrated DNDC model was combined with future climate data from the CMIP6 model to study the effects of different fertilization measures (CK: control with no fertilization; T1: combined application of organic and chemical fertilizers; T2: chemical fertilizer application; T3: increased chemical fertilizer application) on summer maize yield and SOC storage under three future scenarios (baseline, SSP2-4.5, and SSP5-8.5 scenarios). Results show that under both SSP scenarios, the precipitation and temperature in the study area showed an upward trend over time, with the SSP5-8.5 scenario showing more pronounced upward trends in precipitation and temperature. The mean bias error (MBE) between measured and simulated maize yield and surface SOC content across treatments showed satisfactory agreement, with normalized root mean square error (NRMSE) of 0.07–0.19 and coefficient of determination (R²) of 0.74–0.98, respectively, indicating a good simulation performance and be able to simulate summer maize yield and SOC in purple soil areas of the model. Under the T1 treatment, there was no significant difference in SOC storage under various future scenarios in most eras, and climate change had a relatively small impact on the yield and SOC storage under combined application of organic and chemical fertilizers. Under CK treatment, the maize yields under SSP2-4.5 and SSP5-8.5 scenarios increased significantly compared to that under the baseline scenario; however, the maize yield under the two SSP scenarios decreased significantly treated with chemical fertilizer alone. Under the CK, T2, and T3 treatments, the SOC storage under each scenario showed a decreasing trend over time, whereas under the T1 treatment, the SOC storage under each scenario showed an increasing trend over time. By 2100, the SOC storage of each treatment showed a trend of T1 > T3 ≈ T2 > CK. Under different fertilization measures, by 2100, the SOC loss under the SSP5-8.5 scenario is higher than that under the SSP2-4.5 scenario, i.e., the high emission scenario leads to pronounced SOC loss. Research has shown that under future climate change, although conventional fertilization methods can ensure stable crop yields, they are not conducive to carbon sequestration in farmland. The combination of organic and chemical fertilizers is a more suitable fertilization management mode for achieving carbon sequestration and increased yield in purple soil farmland.

     

/

返回文章
返回