Millions of hectares of agricultural land have been abandoned globally in recent decades, presenting opportunities for secondary vegetation growth and restoration. While abandoned fields have the potential to return to ecological communities with similar species diversity to their pre-agricultural state, they alternatively may transition to novel ecosystems or persist in degraded states that may have alternative functions that impact ecological and human communities. Yet we lack an understanding of how vegetation naturally recovers on disturbed lands. Using remote sensing and land survey data, we characterized the structure, composition, and function of secondary vegetation canopies on former sugarcane fields in Hawaiʻi that were abandoned between 4 and 117 years ago. We used a species distribution model to identify patches of uncultivated land with similar environmental conditions to abandoned sugarcane fields to serve as reference ecosystems. Using these reference ecosystems, we evaluated how secondary ecosystems at different ages since abandonment compare in terms of canopy structure, composition, and function. Grasses were prevalent in the years immediately following abandonment, but shrubs and trees dominated canopy structure on fields that had been abandoned more than 20 years. Non-native species constituted most of the secondary vegetation, but native vegetation cover increased on sugarcane fields that had been abandoned longer than 25 years. Secondary vegetation recovered canopy functional traits in ≤ 53 years since abandonment. Completely recovering the structural properties of reference ecosystems would require over a century. Abandoned sugarcane fields are unlikely to recover the native composition of reference ecosystems through unassisted vegetation recovery. Our findings contribute to a growing body of literature that characterizes whether and when the globally increasing area of abandoned agricultural land may passively recover, which can direct restoration efforts on abandoned lands to enhance ecosystem services or guide alternative management to achieve socio-cultural objectives.

近几十年来，全球有数百万公顷的农业用地被废弃，为次生植被生长和恢复提供了机会。虽然废弃的土地有可能恢复到具有与农业前状态相似的物种多样性的生态群落，但它们也可能过渡到新的生态系统或持续退化状态，这些状态可能具有影响生态和人类群落的替代功能。然而，我们对植被如何在受干扰的土地上自然恢复缺乏了解。我们使用遥感和土地调查数据，描述了 4 至 117 年前废弃的夏威夷前甘蔗田次生植被冠层的结构、组成和功能。我们使用物种分布模型来识别环境条件与废弃甘蔗田相似的未开垦土地块，作为参考生态系统。使用这些参考生态系统，我们评估了自废弃以来不同年龄的次级生态系统在冠层结构、组成和功能方面的比较。草在被遗弃后的几年里很普遍，但灌木和树木在已经被遗弃 20 多年的田地中占据了冠层结构的主导地位。非本地物种构成了次生植被的大部分，但在废弃时间超过 25 年的甘蔗田上，本地植被覆盖率有所增加。次生植被在废弃后 ≤ 53 年内恢复了冠层功能特征。完全恢复参考生态系统的结构特性需要一个多世纪的时间。废弃的甘蔗田不太可能通过独立的植被恢复来恢复参考生态系统的原生组成。我们的研究结果有助于越来越多的文献描述全球范围内不断增加的废弃农业用地是否以及何时可以被动恢复，这可以指导废弃土地的恢复工作以增强生态系统服务或指导替代管理以实现社会文化目标。