Data set related to the article titled 'Determination of Root Species Composition in Grass-Legume Mixtures Using Fourier Transform Infrared (FTIR) Spectroscopy

This dataset accompanies the article titled 'Determination of Root Species Composition in Grass-Legume Mixtures Using Fourier Transform Infrared (FTIR) Spectroscopy". Dataset Description This dataset contains calibration and validation files, structured to support the building of an identification and quantification FTIR model for species-specific root composition in grass-clover mixtures. The data files are organized systematically, with a standardized naming convention and clear metadata to ensure reproducibility and transparency. File naming conventionThe dataset titles are coded as following: Species_DatasetType_MixtureTypeExample: PR_calib_artmix.csv Metadata fields Species: The plant species included in the dataset are coded as: PR = Perennial ryegrass RC = Red clover WC = White clover Dataset type: Indicates the type of dataset used for modeling and validation: calib = Calibration dataset blindval = Blind validation dataset loo = Leave-one-out validation dataset Mixture type: Specifies the type of sample composition: artmix = Artificial mixture pure = Pure roots File format: All data are provided as CSV files to ensure compatibility with common statistical and data analysis tools. Columns: Each dataset contains treatments and variables, including: Measured contents (%) Predicted contents (%) – the predicted values were obtained from spectra of the samples using OPUS software (Version 8.7.31, Bruker 2021). (The file tittled"FTIR_model_building_experiment_metadata" also explains the file naming convention and abbreviations)本数据集配套于题为《利用傅里叶变换红外（Fourier Transform Infrared, FTIR）光谱法测定禾本科-豆科混播植物的根系物种组成》的学术论文。 数据集说明 本数据集包含校准文件与验证文件，旨在支撑禾本科-三叶草混播根系物种组成的鉴定与定量FTIR模型构建。数据集文件采用系统化组织方式，遵循标准化命名规则并配备清晰的元数据，以确保实验可复现性与结果透明度。 文件命名规则 数据集文件命名遵循以下编码格式：物种_数据集类型_混样类型，示例：PR_calib_artmix.csv。 元数据字段 1. 物种：本数据集涉及的植物物种编码规则如下： PR = 多年生黑麦草（Perennial ryegrass） RC = 红三叶（Red clover） WC = 白三叶（White clover） 2. 数据集类型：用于标识建模与验证所用的数据集类别： calib = 校准数据集（Calibration dataset） blindval = 盲测验证数据集（Blind validation dataset） loo = 留一法验证数据集（Leave-one-out validation dataset） 3. 混样类型：用于指定样品组成类型： artmix = 人工混配样品 pure = 纯根系样品 文件格式：所有数据均以CSV格式存储，以兼容主流统计与数据分析工具。 字段说明：每份数据集包含处理项与变量项，具体包括： - 实测含量（%） - 预测含量（%）—— 预测值通过OPUS软件（版本8.7.31，布鲁克2021）对样品光谱进行分析得到。 （名为"FTIR_model_building_experiment_metadata"的文件亦对文件命名规则与缩写进行了说明）