Data from: Intraspecific trait variation across multiple scales: the leaf economics spectrum in coffee

Understanding species differences in plant functional traits has been critical in developing a mechanistic understanding of terrestrial ecological processes. Greater attention is now being placed on understanding the extent, causes and consequences of intraspecific trait variation (ITV).
ITV is especially important in governing ecological processes in cropping systems, where only a small number of species or genotypes exist in high abundances. However, it remains unclear if key principles of trait-based ecology – namely the leaf economics spectrum (LES) – also describe intraspecific variation in crop functional biology. There also remains a need to understand whether ITV within crops is random, or structured across environmental, management-related or biological levels of organization in agroecosystems.
We employed a nested design field survey to evaluate ITV in leaf traits in coffee (Coffea arabica), one of the world's most widespread tropical crops. We evaluated ITV in eight physiological, morphological and chemical leaf traits, across five nested categorical levels (sites, management systems, spatial location, plant identity, branch identity). We compared patterns of LES trait covariation in coffee, to interspecific patterns observed across over 700 wild plant species.
Patterns of bivariate and multivariate ITV in coffee were broadly consistent with, but considerably weaker than, interspecific patterns associated with the LES, indicating that crops may systematically diverge from global patterns of trait trade-offs observed in wild plants.
Physiological traits varied most widely (coefficient of variation (cv) 42–107%), followed by morphological traits (cv = 15–38%) and chemical traits (cv = 3–11%). Physiological ITV was best explained by the site in which a coffee plant was growing (17–55% explained), while ITV for chemical traits was best explained by management treatments within sites (25–36%); morphological ITV was higher even at the individual tree level or branch level and remained largely unexplained.
Our results support the hypothesis that artificial selection and high-resource agricultural environments lead crops to systematically deviate from patterns of leaf trait covariation observed across wild plants species. Coupled with an understanding of how different traits vary systematically across multiple levels of biological organization, these findings help integrate ITV into future analyses of agroecosystem structure and function.

解析植物功能性状的物种间差异，对于深入理解陆地生态过程的机制具有核心意义。当前，学界对种内性状变异（intraspecific trait variation, ITV）的发生范围、成因及其生态后果的关注度持续提升。
种内性状变异在调控农田种植系统的生态过程中尤为关键——这类系统中仅少数物种或基因型以极高丰度存在。然而，基于性状的生态学核心原则——即叶片经济谱（leaf economics spectrum, LES）——是否同样适用于阐释作物功能生物学的种内变异，目前仍未有定论。此外，我们仍需明确作物内部的种内性状变异究竟是随机分布，还是会沿农业生态系统的环境、管理相关或生物组织层级呈现结构化特征。
本研究采用嵌套设计野外调查方案，针对全球分布最广的热带作物之一——小粒咖啡（Coffea arabica）的叶片性状种内变异开展评估。我们针对8项生理、形态及化学叶片性状，在5个嵌套分类层级（样地、管理系统、空间位置、植株个体、枝条个体）上分析其种内变异，并将咖啡叶片性状的协变模式，与已报道的700余种野生植物物种的种间性状协变模式进行对比。
咖啡的双变量及多变量种内性状变异模式，整体上与叶片经济谱相关的种间模式相符，但强度显著弱于后者，这表明作物可能系统性偏离野生植物所观测到的全球性状权衡规律。
生理性状的变异幅度最大（变异系数（coefficient of variation, cv）为42%~107%），其次为形态性状（cv=15%~38%）与化学性状（cv=3%~11%）。生理性状的种内变异最主要由咖啡植株生长的样地解释（解释度达17%~55%），而化学性状的种内变异则主要由样地内的管理措施所解释（解释度为25%~36%）；形态性状的种内变异即便在单株乃至单枝条水平仍较高，且在很大程度上仍无法被现有因素所阐释。
本研究结果支持下述假说：人工选择与高资源投入的农业环境，会导致作物系统性偏离野生植物物种所观测到的叶片性状协变模式。结合不同性状沿多级生物组织层级呈现系统性变异的规律，本研究结果有助于将种内性状变异纳入未来农业生态系统结构与功能的分析框架中。