Biological Mechanisms May Contribute to Soil Carbon Saturation Patterns: Modeling Archive

This Modeling Archive is in support of a TES-SFA publication “Biological Mechanisms May Contribute to Soil Carbon Saturation Patterns” (Craig et al., 2021). We ran and evaluated a multi-assumption soil organic carbon (SOC) model to investigate whether alternative assumptions regarding constraints on soil microbial biomass could lead to soil carbon saturation patterns. We developed this model in the Multi-Assumption Architecture and Testbed (MAAT, https://github.com/walkeranthonyp/MAAT, tag: v1.2.1_Craig2021; Walker et al. 2018). Using MAAT, we embedded three alternative hypotheses in a microbially explicit three-pool SOC model: 1) the efficiency of mineral-associated SOC formation decreases as mineral-associated SOC approaches a maximum value (“Mineral saturation”), 2) the microbial biomass turnover rate increases with increasing microbial biomass (“Density-dependent turnover”), and 3) community carbon use efficiency decreases as microbial biomass increases toward an upper limit (“Density-dependent growth”). We ran a factorial combination of these hypotheses resulting in eight models for three different classes of model (linear decay, Michaelis-Menten decay, or reverse Michaelis-Menten decay), resulting in 24 models, 12 of which are presented or discussed in the related publication. Models were parameterized using values from previous studies with similar models (Wang et al. 2013, Wieder et al. 2014, Li et al. 2014, Georgiou et al. 2017, Hassink and Whitmore 1997) and ran to an approximate steady state (200 years) at six (6) different C input rates corresponding to 0.5, 1, 2, 4, 7, and 10 times the default input value. Further model details are available in the related publication. This archive contains output from three MAAT simulations, and scripts to run these simulations and process and plot the data. Simulations are labeled “lin”, “MM_highKm”, and “RMM_highKm” reflecting factorial runs for linear, Michealis-Menten, and reverse Michaelis-Menten models, respectively. This archive contains: • 3 R scripts prepended with “init_MAAT_” to initialize runs (1 for each simulation), • 1 csv file containing years over which to run simulations (“met_year.csv”), • 1 bash script (.bs) to run MAAT, • 6 XML files that are output from MAAT describing a run (2 for each simulation), • 3 model output csv files prepended by “out_” (1 for each simulation), and • 1 analysis R script for reproducing figures 3 and 4 in Craig et al. 2021. See included user guide (Craig_2021_modeling_archive_20210315.pdf) for file organization details.本建模存档（Modeling Archive）用于支撑发表于期刊TES-SFA的论文《生物机制或可解释土壤碳饱和格局》（Biological Mechanisms May Contribute to Soil Carbon Saturation Patterns，Craig等，2021）。本研究运行并评估了多假设土壤有机碳（soil organic carbon, SOC）模型，以探究针对土壤微生物生物量约束的不同假设是否会产生土壤碳饱和格局。该模型基于多假设架构与测试平台（Multi-Assumption Architecture and Testbed, MAAT，https://github.com/walkeranthonyp/MAAT，标签：v1.2.1_Craig2021；Walker等，2018）开发。 依托MAAT平台，我们在微生物显式三库SOC模型中嵌入了三种备选假说：1） 当矿物结合态SOC趋近最大值时，矿物结合态SOC形成效率下降（"矿物饱和假说"）；2） 微生物生物量周转速率随生物量增加而升高（"密度依赖周转假说"）；3） 当微生物生物量趋近上限时，群落碳利用效率下降（"密度依赖生长假说"）。 我们对这些假说进行因子组合，针对三类模型（线性衰减、米氏（Michaelis-Menten）衰减、反米氏（reverse Michaelis-Menten）衰减）分别构建8种模型，总计24种模型，其中12种在相关论文中进行了展示或讨论。 模型参数取自过往针对类似模型的研究（Wang等，2013；Wieder等，2014；Li等，2014；Georgiou等，2017；Hassink与Whitmore，1997），并以6种不同的碳输入速率（分别为默认输入值的0.5、1、2、4、7、10倍）运行至近似稳态（时长200年）。更多模型细节可参阅相关论文。 本存档包含3组MAAT模拟的输出结果，以及用于运行模拟、处理并绘制数据的脚本。模拟分别命名为"lin"、"MM_highKm"与"RMM_highKm"，分别对应线性、米氏及反米氏模型的因子模拟实验。 本存档包含： • 3个以"init_MAAT_"为前缀的R脚本，用于初始化模拟（每组模拟对应1个脚本）； • 1个CSV文件，包含模拟运行的年份序列（"met_year.csv"）； • 1个用于运行MAAT的Bash脚本（.bs格式）； • 6个由MAAT输出的描述模拟过程的XML文件（每组模拟对应2个文件）； • 3个以"out_"为前缀的模型输出CSV文件（每组模拟对应1个文件）； • 1个用于复现Craig等2021年论文中图3与图4的分析R脚本。 有关文件组织的详细说明，请参阅附带的用户指南（Craig_2021_modeling_archive_20210315.pdf）。