Abstract Predawn leaf water potential (Ψpd) is widely used to assess plant water status. Also, pigments concentration work as proxy of canopy’s water status. Spectral data methods have been applied to monitor and assess crop’s biophysical variables. This work developed two models to estimate Ψpd using a hand-held spectroradiometer (400−1010 nm) to obtain canopy and foliar reflectance in four dates of 2018 and a pressure chamber to measure Ψpd. Two modelling approaches, combining spectral data and several machine learning algorithms (MLA), were used to estimate Ψpd in a commercial vineyard in the Douro Wine Region. The first approach estimated Ψpd through vine’s canopy reflectance; several vegetation indices (VIs) were computed and selected, namely the SPVIopt1_950;596;521; SPVIopt2_896;880;901; PRI_CI2opt_539;560,573;716 and NPCIopt_983;972, as well as a time-dynamic variable based on Ψpd (Ψpd_0). The second modelling approach is based on pigments’ concentrations; several VIs were optimized for non-correlated pigments of vine’s leaves, assessed by its hyperspectral reflectance. The following variables for Ψpd estimation were selected through stepwise forward method: Ψpd_0; NRIgreen_LUT520;532; NRIgreen_LWC540;551. The B-MARS algorithm performed the best results for both modelling approaches, presenting a RRMSE in both validation modelling approaches between 13–14%.

中文翻译：

---

基于高光谱数据和杜罗葡萄酒产区 Vitis vinifera L. 色素浓度评估黎明前叶水势

摘要 黎明前叶水势（Ψpd）被广泛用于评估植物水分状况。此外，颜料浓度可作为冠层水分状况的代表。光谱数据方法已被应用于监测和评估作物的生物物理变量。这项工作开发了两种模型来估计 Ψpd，使用手持光谱仪（400-1010 nm）获得 2018 年四个日期的冠层和叶面反射率，并使用压力室测量 Ψpd。结合光谱数据和几种机器学习算法 (MLA) 的两种建模方法用于估计杜罗葡萄酒产区商业葡萄园的 Ψpd。第一种方法通过藤蔓的冠层反射率估计Ψpd；计算并选择了几个植被指数 (VI)，即 SPVIopt1_950;596;521; SPVIopt2_896;880;901; PRI_CI2opt_539;560,573;716 和 NPCIopt_983;972, 以及基于 Ψpd (Ψpd_0) 的时间动态变量。第二种建模方法基于颜料的浓度；通过高光谱反射率评估葡萄叶的非相关色素，优化了几个 VI。Ψpd 估计的以下变量是通过逐步向前方法选择的： Ψpd_0；NRIgreen_LUT520;532; NRIgreen_LWC540;551。B-MARS 算法在两种建模方法中都取得了最好的结果，在两种验证建模方法中的 RRMSE 都在 13-14% 之间。Ψpd_0; NRIgreen_LUT520;532; NRIgreen_LWC540;551。B-MARS 算法在两种建模方法中都取得了最好的结果，在两种验证建模方法中的 RRMSE 都在 13-14% 之间。Ψpd_0; NRIgreen_LUT520;532; NRIgreen_LWC540;551。B-MARS 算法在两种建模方法中都取得了最好的结果，在两种验证建模方法中的 RRMSE 都在 13-14% 之间。