SBB高被引论文|有机施肥通过改变土壤团聚性促进作物生产力

土壤团聚体是土壤生态系统的关键功能单元，对生物地球化学循环和植物生长至关重要。然而，施肥管理如何影响土壤团聚体、相关资源和微生物在团聚体中的分布，以及这些变化对其他非生物和生物过程的潜在影响仍不清楚。为此，本研究在中国亚热带地区开展了一个长期玉米种植田间实验，施用了四种施肥处理：不施肥、化学施肥、有机施肥和化学加有机施肥。此外，研究将土壤团聚体分为大团聚体（>2毫米）、小团聚体（0.25–2毫米）和微团聚体（<0.25毫米），并比较了每个团聚体分级和整体土壤中的土壤养分、酶活性、微生物群落，以及作物生产力和植物的碳（C）、氮（N）和磷（P）含量。结果显示，长期有机施肥：（1）增加了宏团聚体中土壤C、N和P的含量，同时提高了所有三个团聚体尺寸分级中的细菌和真菌生物量；（2）增加了氮获取酶活性、C和N酶比率，但无论聚集体分级如何，磷酸酶活性和C酶比率均有所下降；（3）促进了作物生产力，但与化学施肥相比，植物的C、C和N比率有所下降。此外，真菌：细菌比和磷酸酶活性下降，但随着团聚体尺寸减小，革兰阳性菌与革兰阴性菌的比率、C获取酶活性、C和C酶比率增加。偏最小二乘模型证实，宏团聚体对作物表现有强烈影响，而微团聚体则是微生物群落的主要决定因素。总的来说，长期有机施肥通过增加土壤宏团聚体的比例，促进了土壤功能和作物生产力。

土壤团聚体作为土壤生态系统的功能单元，是生物与非生物相互作用的重要决定因素。不同大小的团聚体因其特有的通气性、保水性、有机质含量和生态位可用性而展现出不同的特性。例如，稳定的有机质（如矿物相关有机碳、难降解的植物或微生物副产物）通常不易被生物降解，并且在<0.25毫米的土壤颗粒（如微聚集体、淤泥和粘土）中更为稳定。相反，易降解的有机物，如新鲜的落叶、细根和真菌菌丝，更丰富地存在于大团聚体（>0.25毫米直径）中，为有机物提供物理保护。农业施肥是提高土壤聚集性的最著名驱动因素之一。有机肥料的投入，如来自粪便的堆肥，通过增强大团聚体的形成，增加了有机物的物理保护。

土壤团聚体为土壤生物群体提供了空间异质性。越来越多的研究发现，土壤微生物、胞外酶活性和微型动物（如线虫）在土壤团聚体中的分布模式不同。团聚体大小分布中的微生物活性和群落组成差异预计与养分循环和其他土壤生态过程有关，这些过程提供了多重功能。许多研究表明，土壤团聚体分布对于土壤营养元素的保持和可用性至关重要。到目前为止，大多数相关研究将土壤分为不同的团聚体尺寸，并集中研究每个团聚体单独的功能潜力，而忽略了不同尺寸团聚体对整体土壤功能（例如作物生产力）的具体贡献。

土壤团聚体的大小可以与微生物群落相互作用，从而影响元素循环。因此，不同大小团聚体之间的养分和有机碳（C）周转效率进一步影响植物生长。土壤微生物参与土壤聚集作用，团聚体内的微生物群落组成和功能因此成为探索土壤生态过程内在机制的一个重要途径。此外，土壤微生物衍生的酶对团聚体内部资源变化非常敏感，酶学计量被解释为底物可用性的指标。进一步而言，植物生长通过释放根部分泌物（如增加土壤团聚体稳定性）来反过来影响土壤团聚过程，进而重新排列土壤颗粒。然而，很少有研究集中于施肥对植物、微生物群落和养分循环功能在团聚体层面的复杂相互作用的影响。

本研究因此集中于施肥管理方式（化学肥料与有机肥料）对土壤团聚体规模上微生物群落组成和土壤功能（如养分保持和可用性）的影响，以及其对整体土壤功能（如支持植物生长）的后果，并探讨其与植物表现特性（即元素浓度和生产力）的关系。该研究使用了在中国亚热带地区进行的为期28年的长期田间实验，研究了有机和/或无机施肥管理。为理解上述复杂关系，提出了以下假设：（1）土壤资源属性和微生物群落随团聚体尺寸分级变化，在土壤功能中扮演不同的角色，这些功能由土壤酶活性和计量学特征指示；（2）施肥管理通过重新构建土壤团聚体的分布模式调节土壤功能；（3）有机施肥通过增加大团聚体的比例，从而促进土壤整体碳和养分池，进而提高植物生产力。

（注：以上翻译来着ChatGPT，具体文章内容请以原文内容为准。若解读有误欢迎探讨指正。）

Fig. 1. Soil aggregate distribution (A), mean weight diameter of aggregates (B) under different fertilization treatments. Different lowercase letters indicate significant differences among aggregate size fractions in the same fertilization treatment (P < 0.05), while different uppercase letters indicate significant differences among fertilization treatments for each of the aggregate size classes (P < 0.05). Values are means ± SE (n = 3). Control = no fertilizer; NPK = chemical fertilizer; OM = organic fertilizer; NPK + OM = chemical fertilizer + organic fertilizer.

Fig. 2. Soil organic carbon (A), total nitrogen (B), total phosphorus (C), dissolved organic carbon (D), mineral nitrogen (E), and available phosphorus (F) content within aggregates under different fertilization treatments. The red lines are the means of bulk soils under the different fertilization treatments. Different lowercase letters indicate significant differences among aggregate size fractions in the same fertilization treatment (P < 0.05), while different uppercase letters indicate significant differences among fertilization treatments (P < 0.05). Values are means ± SE (n = 3). Control = no fertilizer; NPK = chemical fertilizer; OM = organic fertilizer; NPK + OM = chemical fertilizer + organic fertilizer. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3. Concentrations of bacterial (A) and fungal (B) PLFA (phospholipid fatty acids) markers, and Gram-positive to Gram-negative (C) and fungi to bacteria (D) PLFA ratios within soil aggregates under different fertilization treatments. The red lines are the means of bulk soils under the different fertilization treatments. Different lowercase letters indicate significant differences and ns indicate no significant differences among aggregate size fractions in the same fertilization treatment (P < 0.05), while different uppercase letters indicate significant differences among different fertilization treatments (P < 0.05). Values are means ± SE (n = 3). Control = no fertilizer; NPK = chemical fertilizer; OM = organic fertilizer; NPK + OM = chemical fertilizer + organic fertilizer. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4. Microbial enzyme activities and enzymatic stoichiometry within aggregates under different fertilization treatments. Shown are activities of normalized C-acquiring enzyme (A), N-acquiring enzyme (B), P-acquiring enzyme (C), C:N enzyme ratio (D), C:P enzyme ratio (E), and N:P enzyme ratio (F). The red lines are the means of bulk soils under different fertilization treatments. Different lowercase letters indicate significant differences and ns indicate no significant differences among aggregate size fractions in the same fertilization treatment (P < 0.05), while different uppercase letters indicate significant differences among different fertilization treatments (P < 0.05). Values are means ± SE (n = 3). Control = no fertilizer; NPK = chemical fertilizer; OM = organic fertilizer; NPK + OM = chemical fertilizer + organic fertilizer. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5. Maize shoot and root biomass (A), plant C:N (B), C:P (C) and N:P (D) ratio of maize shoot and root under different fertilization treatments; values are means ± SE (n = 3). Different lowercase letters indicate significant differences among different fertilization treatments (P < 0.05). Control = no fertilizer; NPK = chemical fertilizer; OM = organic fertilizer; NPK + OM = chemical fertilizer + organic fertilizer.

Fig. 6. The partial least squares path models (PLS-PM) illustrating the direct and indirect effects of chemical and organic fertilizer, large macro-aggregate, small macro-aggregate, and micro-aggregate nutrient pools on maize root and shoot properties (A) and standardized total effects on root and shoot properties from PLS-PM (B) as well as the direct and indirect effects of chemical and organic fertilizer, large macro-aggregate, small macro-aggregate, and micro-aggregate microbial community on bulk soil microbial community (C) and the standardized total effects on bulk soil microbial community from PLS-PM (D). Large macro-aggregate, small macro-aggregate, and micro-aggregate nutrient pools are latent variables, which are indicated by total N, total P, mineral N, and available P content of each aggregate, respectively. Maize root and shoot properties are latent variables measured by biomass, N, and P concentration of root and shoot, respectively. Large macro-aggregate, small macro-aggregate, micro-aggregate, and bulk soil microbial communities are latent variables reflected by PLFA profiles of each aggregate and bulk soil, respectively. The red and blue arrows indicate negative and positive flows of causality, respectively. Numbers on the arrowed lines and thickness of arrows indicate normalized path coefficient. The dotted gray arrows represent non-significant path relationships. R2 beside the latent variables are the coefficients of determination. The GoF index represents the goodness of fit. Asterisks represent significant effects: *, P < 0.05; **, P < 0.01; ***, P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)