KIWIQUALI Project 2023 - Physiological performance and models

This dataset regards the project KIWIQUALI (Grant Agreement 101066378) which was funded by Marie Slodovska Curie (Horizon Europe). The data was collected in 2023 in a commercial orchard of Actinidia chinensis var chinensis. The kiwifruit vines were subjected to four irrigation treatments as percentages of crop evapotranspiration (ETc). 100% ETc (T100); 68% ETc (T68); 57% ETc (T57); 40% ETc (T40). Four kiwifruit vines per treatment were monitored through the Tmax method regarding sap flux density (from June to October), and in nine days from July to September leaf gas exchange measurements were performed throughout the day, with at least one measurement every hour per kiwifruit vine. Besides the analysis of the stomatal conductance (gs) daily curves, this data was also used to obtain artificial neural network models with different input variables, from environmental data (atmospheric vapor pressure, photosynthetic photon flux and soil water content) to physiological measurements (sap flux density and leaf temperature). The dataset contains five folders to help organize the files regarding their subjects. The description of each folder is detailed in the following table. Folder name Description EnvironmentalConditions Files with data regarding the irrigation volumes from June to October 2023; the meteorological conditions from July to October 2023; and the soil water content from July to October 2023. FruitGrowthQuality Files with fruit dry matter content measured from August to October 2023; fruit growth monitored through fruit gauges from August to October 2023; the fruit growth measured with an automatic caliper from June to October 2023; and the fruit quality at harvest (fruit size, fruit weight, solid soluble content, firmness and acidity). ModelStomatalConductance Files in the format .h5 containing artificial neural network models obtained with different input variables. Physiology Files containing leaf gas exchange measurements performed in 9 days (from July to September 2023), throughout the day, with at least one measurement every hour; sap flux density measured with the Tmax method from June to October 2023. Scripts_DataAnalysis Files in the .R and .ipynb formats containing the scripts in R and Python languages (respectively). The R file regards the analysis of the data and the optimization of the mechanistic model of stomatal conductance (Buckley 2012). The Python file regards the training and testing the artificial neural network models regarding stomatal conductance from different input variables. Each folder contains its own README file (either in .txt or .ods formats) with the explanation of the data.本数据集关联由玛丽·斯克洛多夫斯卡·居里（Marie Slodovska Curie）资助、项目编号为Grant Agreement 101066378的KIWIQUALI项目，该项目获欧盟地平线欧洲（Horizon Europe）计划支持。数据集采集于2023年，采集地点为中华猕猴桃（Actinidia chinensis var. chinensis）商业种植果园。 供试猕猴桃植株设置4个灌溉处理，以作物蒸散量（crop evapotranspiration, ETc）的百分比作为梯度标准，分别为：100% ETc（T100）、68% ETc（T68）、57% ETc（T57）、40% ETc（T40）。 每个处理选取4株猕猴桃，采用Tmax法监测其树干液流密度（sap flux density），监测周期为2023年6月至10月；2023年7月至9月的9个测定日期内，每日对植株进行全天叶片气体交换参数（leaf gas exchange）测定，每株每小时至少完成1次测定。 除用于分析气孔导度（stomatal conductance, gs）的日变化曲线外，本数据集还可用于构建基于多输入变量的人工神经网络（artificial neural network）模型，输入变量涵盖环境数据（大气水汽压、光合光子通量密度、土壤含水量）与生理测定数据（树干液流密度、叶片温度）。 本数据集包含5个文件夹，用于按研究主题整理相关文件，各文件夹的详细说明如下： 1. 环境条件（EnvironmentalConditions）：包含2023年6月至10月的灌溉量数据、2023年7月至10月的气象数据，以及2023年7月至10月的土壤含水量数据。 2. 果实生长与品质（FruitGrowthQuality）：包含2023年8月至10月测定的果实干物质含量、2023年8月至10月通过果实测径仪监测的果实生长数据、2023年6月至10月使用自动游标卡尺测定的果实生长数据，以及采收期果实品质数据（果实大小、单果重、可溶性固形物含量、硬度与酸度）。 3. 气孔导度模型（ModelStomatalConductance）：包含后缀为.h5的文件，存储基于不同输入变量构建的人工神经网络模型。 4. 生理指标（Physiology）：包含2023年7月至9月的9个测定日期内采集的全天叶片气体交换参数数据（每株每小时至少1次测定），以及2023年6月至10月采用Tmax法测定的树干液流密度数据。 5. 数据分析脚本（Scripts_DataAnalysis）：包含后缀为.R和.ipynb的文件，分别为R语言与Python语言脚本。其中R脚本用于数据解析与气孔导度机理模型（Buckley 2012）的优化；Python脚本用于基于不同输入变量的气孔导度人工神经网络模型的训练与测试。 每个文件夹均配备自身的README文件（格式为.txt或.ods），对其中包含的数据进行详细说明。