1 INTRODUCTION

Variation in fine root traits among species impacts the acquisition of belowground resources and rates of decomposition and nutrient cycling (Iversen et al., 2017; Klimešová & Herben, 2023; Petit-Aldana et al., 2019). Decomposition is a physical and chemical process that converts dead plant tissue into its component elements, driving nutrient cycling rates and the soil organic matter pool (Aerts, 1997; Guo et al., 2021; Wambsganss et al., 2022), yet decomposition is still a poorly understood flux of nutrients and carbon within terrestrial ecosystems (See et al., 2019; Sun et al., 2013). Fine root decomposition at the early stage can be influenced by root morphological (Wu et al., 2022) and chemical traits (Bonanomi et al., 2021; Jiang et al., 2021; Silver & Miya, 2001). Morphological root traits are phylogenetically conserved, but chemical traits are less constrained by evolutionary history (Liese et al., 2017; Liu et al., 2019; Sun et al., 2018). Studies exploring the relationship between root traits and decomposition in the context of phylogeny are scarce (e.g. LeRoy et al., 2020). Determining the relative importance of root traits as drivers of decomposition in a phylogenetic context remains a research priority to improve our understanding of decomposition dynamics in terrestrial ecosystems.

Four fine root traits that could drive decomposition rates define two orthogonal belowground trade-offs known as the root economics space (Bergmann et al., 2020). The conservation axis is delineated by a fast versus slow trade-off, while the collaboration axis reflects a do-it-yourself versus outsourcing trade-off that has likely evolved in interaction with mycorrhizal symbiosis (Bergmann et al., 2020). The conservation and collaboration axes may impact decomposition rates differently (Guo et al., 2021; Klimešová & Herben, 2023). The conservation axis is quantified by variation in root nitrogen (RN) and root tissue density (RTD), which are well-known drivers of decomposition rates. High root nitrogen (RN) in acquisitive species enhances the quality of substrate for accelerated microbial decomposition, while the less porous structure of dense root tissue in conservative species makes them more resistant to decomposition (Valverde-Barrantes et al., 2016; Weigelt et al., 2021). The collaboration axis is quantified by variation in root diameter (RD) and specific root length (SRL), which could also influence decomposition rates, but it is still unclear whether thin or thick fine roots decompose faster. Species with high SRL and higher surface area-to-volume ratio may facilitate faster decomposition due to increased substrate accessibility by microbes (Chengfang et al., 2011; Prieto et al., 2016; Wu et al., 2022) and high nutrient release and diffusion from the decomposing material into the soil (Liu et al., 2017), especially in the early stage of decomposition. Variation in root diameter could also influence decomposition (Goebel et al., 2011; He et al., 2019; Van Do et al., 2016). Hobbie et al. (2010) showed that thin roots decompose more slowly than thick roots during the early stage but more quickly than thick roots in the long run. In contrast, thick roots with higher fractions of high-quality cortical tissues that are preferred by microbes may decompose faster.

Root tissue chemistry traits are also known to be dominant drivers of decomposition (Prieto et al., 2016; See et al., 2019; Sun et al., 2013; Tu et al., 2015). In long-term decomposition studies, RN is the best predictor of decomposition in the early stages (Liao et al., 2022), while root lignin is the best predictor of decomposition in the later stages (Bachega et al., 2016). Lignin is widely acknowledged as an important driver of decomposition (Guo et al., 2021; Solly et al., 2014; Sun et al., 2018) because it can inhibit cellulose degradation by protecting it from microbes (Yue et al., 2016), and root litter with a high lignin-to-N ratio decomposes slowly (Prieto et al., 2016; Solly et al., 2014). The lignocellulose index (LCI), computed as LCI = lignin/(lignin + cellulose) (Melillo et al., 1989), has been proposed as an indicator of a structural barrier to decomposition by preventing access to labile carbon compounds by microbes (Austin & Ballaré, 2010). Bachega et al. (2016) reported a low decomposition for Acacia fine roots with high LCI values (ranging from 0.49 to 0.64). In a 10-year decomposition study by Adair et al. (2008), LCI decreased decomposition at the intermediate stage due to cellulose protection by lignin. Phosphorus (P) limitation may also hinder microbial activities during early-stage decomposition (Liao et al., 2022), and global meta-analyses have reported a consistent positive effect of P on decomposition rates (See et al., 2019; Zhao, Tian, et al., 2023) and higher lignin-to-P ratio makes litter more recalcitrant to decomposition. Lastly, root dry matter content (RDMC) has been shown by Roumet et al. (2016) to exhibit a negative relationship with decomposition.

Mycorrhizal association also explains variation in decomposition rate among species (Freschet et al., 2021; Kong et al., 2019). For example, fine roots of arbuscular mycorrhizal (AM) species with high root litter quality (Jacobs et al., 2018; Taylor et al., 2016) decompose faster than those of ectomycorrhizal (EM; Sun et al., 2018; Zhao, Tian, et al., 2023) and ericoid mycorrhizal species (Cornelissen et al., 2001; Ward et al., 2022). However, southern hemisphere temperate rain forests are dominated by ancient conifer lineages (i.e. Podocarpaceae and Araucariaceae) that associate with AM fungi through root nodules (Scheublin et al., 2004) and exhibit dense root tissues with low nutrient concentrations (Kramer-Walter et al., 2016). Therefore, the effect of mycorrhizal association on decomposition rate could depend on the phylogenetic context.

In this paper, we test the relative importance of morphological traits associated with the root economics space (RES; Bergmann et al., 2020) and root chemical traits for explaining variation in fine root decomposition rate. We quantified early-stage fine root decomposition rate as proportion mass loss (pml) after 6 months and fine root functional traits among 63 temperate rain forest tree species in Aotearoa New Zealand. Our objective was to determine which fine root traits best predict early-stage fine root decomposition rates across species while accounting for phylogeny. We hypothesized that (H1) early-stage decomposition (pml) would be more strongly associated with the root conservation axis than with the collaboration axis, and (H2) chemical root traits would be better predictors of early-stage decomposition (pml) than the morphological RES traits (Table 1).

TABLE 1. Description of the trait categories following (McCormack et al., 2017) and their expected relationship with decomposition rate, measured as proportion mass loss (pml). The four core traits used to delineate the RES in Bergmann et al. (2020) are noted as ‘core RES traits’.

Trait category	Traits	Definition	Expected relationship with decomposition (pml)	References
Morphology	Root tissue density (RTD; core RES trait)	Ratio of dry fine root mass to fresh fine root volume	Species with high RTD possess thicker cell walls that limit microbial access and are less attractive to decomposers (−)	de la Riva et al. (2021), Kong et al. (2019) and Ryser (1996)
	Root dry matter content (RDMC)	Ratio of dry to fresh fine root mass	Species with high RDMC possess thicker cell walls that limit microbial access and are less attractive to decomposers (−)	Bachega et al. (2016) and Shipley and Vu (2002)
	Root diameter (RD; core RES trait)	Average diameter of fine roots	Species with thin roots with a higher surface area-to-volume ratio may facilitate faster decomposition due to increased substrate accessibility by microbes. In contrast, thick roots with higher fractions of high-quality cortical tissues that are preferred by microbes could also decompose faster (− or +)	Chengfang et al. (2011), Hobbie et al. (2010) and McClaugherty et al. (1985)
	Specific root length (SRL) (core RES trait)	Length of fine roots per unit dry mass	Species with high SRL with a higher surface area-to-volume ratio may facilitate faster decomposition due to increased substrate accessibility by microbes. In contrast, low SRL roots with higher fractions of high-quality cortical tissues that are preferred by microbes could also decompose faster (+ or −)	Goebel et al. (2011), Prieto et al. (2016) and Wu et al. (2022)
Chemistry	Root nitrogen (RN; core RES trait)	Concentration (%) of nitrogen in fine roots	Roots with high nitrogen have high-quality tissue that accelerates microbial decomposition (+)	Chen et al. (2002), Prieto et al. (2016), Roumet et al. (2016) and Sun et al. (2013)
	Root phosphorus (RP)	Concentration (%) of phosphorus in fine roots	Roots with high phosphorus have high-quality tissue that accelerates microbial decomposition (+)	Birouste et al. (2012), Liu (2021), Xu et al. (2013) and Zhang et al. (2008)
	Root cellulose	Concentration (%) of cellulose, a complex carbohydrate in fine roots	High root cellulose contributes to recalcitrant cell walls that slows decomposition (−)	Birouste et al. (2012), Genet et al. (2005), Jiang et al. (2021), Taylor (2008) and Valverde-Barrantes et al. (2016)
	Root tannin	Concentration of tannin (%), a polyphenol and secondary metabolite in fine roots	Tannins can reduce microbial activity due to their toxicity and the formation of complexes with proteins and enzymes, making them unpalatable to decomposer organisms (−)	Coulis et al. (2009), Dong et al. (2016) and Makkonen et al. (2012)
	Root phenol	Concentration of phenol (%), an aromatic organic compound and secondary metabolite, in fine roots	Phenols can reduce microbial activity and phenol compounds decompose slowly (−)	Boerjan et al. (2003), Wang et al. (2015) and Zwetsloot et al. (2020)
	Root lignin	Concentration of lignin (%), a complex organic polymer in fine roots	Lignin limits microbial access to labile components in root tissues (−)	Bonanomi et al. (2021), Poirier et al. (2018), Prieto et al. (2016) and Zhang and Wang (2015)
	Lignin-to-N ratio	Ratio of fine root lignin to fine root nitrogen	Species with high root lignin-to-N ratios are more recalcitrant and are lower quality substrates for microbes (−)	Bonanomi et al. (2021), Prieto et al. (2016) and Walela et al. (2014)
	Lignin-to-P ratio	Ratio of fine root lignin to fine root phosphorus	Species with high root lignin-to-P ratios are more recalcitrant and are lower quality substrates for microbes (−)	Patil et al. (2020)
	Root lignocellulose index (LCI)	Ratio of lignin to the addition of lignin and cellulose in fine roots [LCI = lignin/(lignin + cellulose)]	High LCI leads to high resistance to microbial decomposition due to elevated lignin in root tissues (−)	Assefa et al. (2018) and Melillo et al. (1989)
Architecture	Root branching Index (RBI)	Ratio of the number of root tips per fine root length	Species with high RBI have greater surface area for microbial attachment (+)	Comas and Eissenstat (2004), Poirier et al. (2018) and Stokes et al. (2009)

中文翻译：

---

与守恒梯度相关的性状是早期细根分解速率的最强预测因子

 1 引言

物种之间细根性状的差异会影响地下资源的获取以及分解和养分循环的速度（Iversen et al.， 2017;Klimešová & Herben， 2023;Petit-Aldana等人，2019 年）。分解是一个物理和化学过程，将死亡的植物组织转化为其组成元素，推动养分循环速率和土壤有机质库（Aerts， 1997;Guo et al.， 2021;Wambsganss 等人，2022 年），但分解仍然是陆地生态系统中营养物质和碳的通量知之甚少（见等人，2019 年;Sun等人，2013 年）。早期的细根分解会受到根形态 （Wu et al.， 2022） 和化学性状 （Bonanomi et al.， 2021;江等人，2021 年;Silver & Miya，2001 年）。形态学根性状在系统发育上是保守的，但化学性状受进化历史的限制较小（Liese 等人，2017 年;Liu et al.， 2019;Sun等人，2018 年）。在系统发育的背景下探索根性状与分解之间关系的研究很少（例如 LeRoy 等人，2020 年）。确定根性状在系统发育背景下作为分解驱动因素的相对重要性仍然是提高我们对陆地生态系统分解动力学的理解的研究重点。

Four fine root traits that could drive decomposition rates define two orthogonal belowground trade-offs known as the root economics space (Bergmann et al., 2020). The conservation axis is delineated by a fast versus slow trade-off, while the collaboration axis reflects a do-it-yourself versus outsourcing trade-off that has likely evolved in interaction with mycorrhizal symbiosis (Bergmann et al., 2020). The conservation and collaboration axes may impact decomposition rates differently (Guo et al., 2021; Klimešová & Herben, 2023). The conservation axis is quantified by variation in root nitrogen (RN) and root tissue density (RTD), which are well-known drivers of decomposition rates. High root nitrogen (RN) in acquisitive species enhances the quality of substrate for accelerated microbial decomposition, while the less porous structure of dense root tissue in conservative species makes them more resistant to decomposition (Valverde-Barrantes et al., 2016; Weigelt et al., 2021). The collaboration axis is quantified by variation in root diameter (RD) and specific root length (SRL), which could also influence decomposition rates, but it is still unclear whether thin or thick fine roots decompose faster. Species with high SRL and higher surface area-to-volume ratio may facilitate faster decomposition due to increased substrate accessibility by microbes (Chengfang et al., 2011; Prieto et al., 2016; Wu et al., 2022) and high nutrient release and diffusion from the decomposing material into the soil (Liu et al., 2017), especially in the early stage of decomposition. Variation in root diameter could also influence decomposition (Goebel et al., 2011; He et al., 2019; Van Do et al., 2016). Hobbie et al. (2010) showed that thin roots decompose more slowly than thick roots during the early stage but more quickly than thick roots in the long run. In contrast, thick roots with higher fractions of high-quality cortical tissues that are preferred by microbes may decompose faster. 

根组织化学性状也是分解的主要驱动因素（Prieto等人，2016 年;参见 et al.， 2019;Sun等人，2013 年;Tu et al.， 2015）。在长期分解研究中，RN 是早期分解的最佳预测因子（Liao et al.， 2022），而根木质素是后期分解的最佳预测因子（Bachega et al.， 2016）。木质素被广泛认为是分解的重要驱动力（Guo等人，2021 年;Solly等人，2014 年;Sun等人，2018 年），因为它可以通过保护纤维素免受微生物侵害来抑制纤维素降解（Yue等人，2016 年），并且具有高木质素与 N 比率的凋落物分解缓慢（Prieto等人，2016 年;Solly et al.， 2014）。木质素纤维素指数（LCI），计算为LCI = 木质素/（木质素+纤维素）（Melillo等人，1989年），已被提议作为通过防止微生物接触不稳定碳化合物来阻止分解的结构障碍的指标（奥斯汀和巴拉雷，2010年）。Bachega 等人（2016 年）报告了金合欢细根的低分解率，LCI 值高（范围从 0.49 到 0.64）。在 Adair 等人（2008 年）的一项为期 10 年的分解研究中，由于木质素对纤维素的保护，LCI 在中间阶段的分解减少了。磷 （P） 的限制也可能阻碍早期分解过程中的微生物活动（Liao et al.， 2022），全球荟萃分析报告了 P 对分解速率的一致积极影响（参见 et al.， 2019;Zhao， Tian， et al.，2023 年）和较高的木质素与 P 比率使凋落物更难以分解。最后，Roumet 等人（2016 年）表明根干物质含量 （RDMC） 与分解呈负相关。

菌根关联还解释了物种之间分解速率的变化（Freschet等人，2021 年;Kong等人，2019 年）。例如，具有高根凋落物质量的丛枝菌根 （AM） 物种的细根（Jacobs等人，2018 年;Taylor等人，2016 年）的分解速度比外生菌根 （EM;Sun等人，2018 年;Zhao， Tian， et al.， 2023） 和 ericoid 菌根物种 （Cornelissenet al.， 2001;Ward等人，2022 年）。然而，南半球温带雨林以古老的针叶树谱系（即罗汉松科和南洋杉科）为主，它们通过根瘤与 AM 真菌结合（Scheublin 等人，2004 年），并表现出营养浓度低的致密根组织（Kramer-Walter 等人，2016 年）。因此，菌根关联对分解速率的影响可能取决于系统发育环境。

在本文中，我们测试了与根经济学空间 （RES;Bergmann et al.， 2020） 和根化学性状来解释细根分解速率的变化。我们量化了新西兰 Aotearoa 63 种温带雨林树种的早期细根分解率为 6 个月后的质量损失比例 （pml） 和细根功能性状。我们的目标是确定哪些细根性状最能预测跨物种的早期细根分解速率，同时考虑系统发育。我们假设 （H1） 早期分解 （pml） 与根守恒轴的相关性比与协作轴的相关性更强，并且 （H2） 化学根性状比形态学 RES 性状更能预测早期分解 （pml） （表 1）。

TABLE 1. Description of the trait categories following (McCormack et al., 2017) and their expected relationship with decomposition rate, measured as proportion mass loss (pml). The four core traits used to delineate the RES in Bergmann et al. (2020) are noted as ‘core RES traits’.

特征类别	性状	定义	与分解 （pml） 的预期关系	引用
形态学	根组织密度 （RTD;核心 RES 性状）	干燥细根质量与新鲜细根体积的比率	具有高 RTD 的物种具有较厚的细胞壁，这限制了微生物的进入，并且对分解者的吸引力较低 （−）	de la Riva 等人（2021 年）、Kong 等人（2019 年）和 Ryser （1996 年）
	根干物质含量 （RDMC）	干根与新鲜细根质量的比率	具有高 RDMC 的物种具有较厚的细胞壁，这限制了微生物的进入，并且对分解者的吸引力较低 （−）	Bachega 等人 （2016） 和 Shipley 和 Vu （2002）
	根直径 （RD;核心 RES 性状）	细根的平均直径	由于微生物对底物的可及性增加，具有较高表面积与体积比的细根物种可能促进更快的分解。相比之下，微生物喜欢的具有较高比例高质量皮质组织的粗根也可能分解得更快（− 或 +）	Chengfang et al. (2011), Hobbie et al. (2010) and McClaugherty et al. (1985)
	比根长度 （SRL）（核心 RES 性状）	每单位干质量的细根长度	由于微生物对底物的可及性增加，具有高 SRL 和较高表面积与体积比的物质可能促进更快的分解。相比之下，微生物首选的具有较高比例的高质量皮质组织含量高的低 SRL 根也可能分解得更快（+或 −）	Goebel et al. (2011), Prieto et al. (2016) and Wu et al. (2022)
化学	根系氮 （RN;核心 RES 性状）	细根中氮的浓度 （%）	高氮的根系具有高质量的组织，可加速微生物分解 （+）	Chen et al. (2002), Prieto et al. (2016), Roumet et al. (2016) and Sun et al. (2013)
	根磷 （RP）	细根中磷的浓度 （%）	高磷的根具有加速微生物分解的高质量组织 （+）	Birouste et al. (2012), Liu (2021), Xu et al. (2013) and Zhang et al. (2008)
	根纤维素	纤维素的浓度 （%） （一种存在于细根中的复合碳水化合物）	高根纤维素有助于形成顽固的细胞壁，从而减缓分解 （−）	Birouste et al. (2012), Genet et al. (2005), Jiang et al. (2021), Taylor (2008) and Valverde-Barrantes et al. (2016)
	根单宁	单宁浓度 （%），细根中的多酚和次生代谢物	单宁由于其毒性以及与蛋白质和酶形成复合物而降低微生物活性，使它们难以分解生物 （−）	Coulis et al. (2009), Dong et al. (2016) and Makkonen et al. (2012)
	根酚	细根中苯酚 （%） 的浓度 （%），一种芳香族有机化合物和次生代谢物	酚类物质会降低微生物活性，酚类化合物分解缓慢 （−）	Boerjan et al. (2003), Wang et al. (2015) and Zwetsloot et al. (2020)
	根木质素	木质素浓度 （%），一种在细根中的复合有机聚合物	木质素限制了微生物对根组织中不稳定成分的获取 （−）	Bonanomi 等人（2021 年）、Poirier 等人（2018 年）、Prieto 等人（2016 年）以及 Zhang 和 Wang （2015 年）
	木质素与 N 的比率	细根木质素与细根氮的比例	具有高根木质素与 N 比率的物种更顽固，并且是微生物质量较低的底物 （−）	Bonanomi 等人（2021 年）、Prieto 等人（2016 年）和 Walela 等人（2014 年）
	木质素与 P 的比率	细根木质素与细根磷的比例	具有高根木质素与 P 比率的物种更顽固，并且是微生物质量较低的底物 （−）	Patil 等人 （2020）
	根木质素纤维素指数 （LCI）	木质素与细根中木质素和纤维素添加量的比率 [LCI = 木质素/（木质素 + 纤维素）]	由于根组织中木质素升高，高 LCI 导致对微生物分解的高抗性 （−）	Assefa 等人 （2018） 和 Melillo 等人 （1989）
建筑	根分支索引 （RBI）	每细根长的根尖数的比率	具有高 RBI 的物种具有更大的微生物附着表面积 （+）	Comas 和 Eissenstat （2004）、Poirier 等人 （2018） 和 Stokes 等人 （2009）