Data from: Remote sensing of plant trait responses to field-based plant–soil feedback using UAV-based optical sensors

Plant responses to biotic and abiotic legacies left in soil by preceding plants is known as plant–soil feedback (PSF). PSF is an important mechanism to explain plant community dynamics and plant performance in natural and agricultural systems. However, most PSF studies are short-term and small-scale due to practical constraints for field-scale quantification of PSF effects, yet field experiments are warranted to assess actual PSF effects under less controlled conditions. Here we used unmanned aerial vehicle (UAV)-based optical sensors to test whether PSF effects on plant traits can be quantified remotely. We established a randomized agro-ecological field experiment in which six different cover crop species and species combinations from three different plant families (Poaceae, Fabaceae, Brassicaceae) were grown. The feedback effects on plant traits were tested in oat (Avena sativa) by quantifying the cover crop legacy effects on key plant traits: height, fresh biomass, nitrogen content, and leaf chlorophyll content. Prior to destructive sampling, hyperspectral data were acquired and used for calibration and independent validation of regression models to retrieve plant traits from optical data. Subsequently, for each trait the model with highest precision and accuracy was selected. We used the hyperspectral analyses to predict the directly measured plant height (RMSE = 5.12 cm, R2 = 0.79), chlorophyll content (RMSE = 0.11 g m−2, R2 = 0.80), N-content (RMSE = 1.94 g m−2, R2 = 0.68), and fresh biomass (RMSE = 0.72 kg m−2, R2 = 0.56). Overall the PSF effects of the different cover crop treatments based on the remote sensing data matched the results based on in situ measurements. The average oat canopy was tallest and its leaf chlorophyll content highest in response to legacy of Vicia sativa monocultures (100 cm, 0.95 g m−2, respectively) and in mixture with Raphanus sativus (100 cm, 1.09 g m−2, respectively), while the lowest values (76 cm, 0.41 g m−2, respectively) were found in response to legacy of Lolium perenne monoculture, and intermediate responses to the legacy of the other treatments. We show that PSF effects in the field occur and alter several important plant traits that can be sensed remotely and quantified in a non-destructive way using UAV-based optical sensors; these can be repeated over the growing season to increase temporal resolution. Remote sensing thereby offers great potential for studying PSF effects at field scale and relevant spatial-temporal resolutions which will facilitate the elucidation of the underlying mechanisms.植物对前茬植物遗留于土壤中的生物与非生物遗留物产生的响应，被称为植物-土壤反馈（plant–soil feedback, PSF）。植物-土壤反馈是解释自然与农业系统中植物群落动态及植株表现的重要机制。然而，由于田间尺度量化植物-土壤反馈效应存在实际限制，绝大多数相关研究均为短期小尺度研究，而在可控性较低的条件下开展田间实验以评估真实的植物-土壤反馈效应，实属必要。 本研究借助搭载于无人机（unmanned aerial vehicle, UAV）的光学传感器，探究能否通过遥感手段量化植物-土壤反馈对植物性状的影响。我们设置了农业生态野外随机实验，种植了来自禾本科（Poaceae）、豆科（Fabaceae）、十字花科（Brassicaceae）3个植物科的6种覆盖作物及其组合。以燕麦（Avena sativa）为受试材料，我们通过量化覆盖作物遗留效应对植株关键性状的影响，检验植物-土壤反馈的作用，受试关键性状包括株高、鲜生物量、氮含量及叶片叶绿素含量。 在破坏性采样前，我们获取了高光谱数据，并将其用于回归模型的校准与独立验证，以通过光学数据反演植物性状。随后，针对每一项植物性状，我们选取了精度与准确度最高的反演模型。本研究通过高光谱分析实现了对实测植株性状的预测：株高的均方根误差（RMSE）为5.12 cm，决定系数（R²）为0.79；叶绿素含量的RMSE为0.11 g·m⁻²，R²为0.80；氮含量的RMSE为1.94 g·m⁻²，R²为0.68；鲜生物量的RMSE为0.72 kg·m⁻²，R²为0.56。 总体而言，基于遥感数据得到的不同覆盖作物处理的植物-土壤反馈效应，与原位测量得到的结果一致。平均而言，响应于箭舌豌豆（Vicia sativa）单作遗留效应及其与萝卜（Raphanus sativus）混作遗留效应的燕麦冠层最高，叶片叶绿素含量也最高，分别为100 cm与0.95 g·m⁻²，以及100 cm与1.09 g·m⁻²；而响应于黑麦草（Lolium perenne）单作遗留效应的燕麦冠层与叶绿素含量最低，分别为76 cm与0.41 g·m⁻²；其余处理的遗留效应则带来中等程度的响应。 本研究证实，田间确实存在植物-土壤反馈效应，且该效应会改变多项关键植物性状；借助搭载于无人机的光学传感器，可通过非破坏性方式实现这些性状的遥感感知与量化，且可在整个生长季重复开展观测以提升时间分辨率。由此可见，遥感技术为在田间尺度与相关时空分辨率下开展植物-土壤反馈效应研究提供了巨大潜力，将有助于阐明其背后的潜在机制。