摘要
一位新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-机器学习的新研究是一篇报告的主题。根据NewsRx编辑在加州戴维斯的新闻报道,研究表明:“杏仁是加州的一种主要作物,世界上80%的杏仁产量来自此。广泛的干旱和地下水管制对种植者构成了重大挑战。”这项研究的财政支持者包括国家食品和农业研究所、加州杏仁委员会赠款项目、美国农业部NIFA奖、加州大学戴维斯分校下一代食品系统AI Contte。我们的新闻记者从加州大学戴维斯分校的研究中引用了一句话：“基于观测到的作物水分状况的灌溉制度可以帮助优化水分利用效率，但一致和准确地测量水分状况可能证明是一个挑战。在杏仁中，作物水分状况最好通过用压力室测量中午茎水势来表征。”本研究旨在利用机器学习(ML)模型,根据林冠光谱反射率、土壤水分等指标,对扁桃园茎水势进行预测。对人工神经网络模型和随机森林模型进行训练,生成覆盖整个果园的茎水盆高分辨率空间图,并对每种ML模型分别训练一个模型预测原茎水势值,另一个模型预测基线调整值。所有模型的平均相关系数为2=0.73，平均均方根误差(RMSE)为2.5 bar，当模型被映射到空间图中时，预测精度显著下降(2=0.33，RMSE=3.31[AVG])，这些结果表明，人工神经网络和随机森林框架都可以用来预测树干水势。但这两种方法都不能完全解释整个或Chard观察到的空间变异性。总的来说，最准确的地图是由随机森林模型绘制的(原始茎水势＝2＝0.47，RMSE＝2.71)。在空间上预测茎水势的能力有助于实施变量灌溉。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting out of Davis, California, by NewsRx editors, research stated, "Almonds are a major crop in California which p roduces 80% of all the world's almonds. Widespread drought and str ict groundwater regulations pose significant challenges to growers." Financial supporters for this research include National Institute of Food and Ag riculture, Almond Board of California Grant Project, USDA NIFA Award, AI Institu te for Next Generation Food Systems at UC Davis. Our news journalists obtained a quote from the research from the University of C alifornia Davis, "Irrigation regimes based on observed crop water status can hel p to optimize water use efficiency, but consistent and accurate measurement of w ater status can prove challenging. In almonds, crop water status is best represe nted by midday stem water potential measured using a pressure chamber, which des pite its accuracy is impractical for growers to measure on a regular basis. This study aimed to use machine learning (ML) models to predict stem water potential in an almond orchard based on canopy spectral reflectance, soil moisture, and d aily evapotranspiration. Both artificial neural network and random forest models were trained and used to produce high-resolution spatial maps of stem water pot ential covering the entire orchard. Also, for each ML model type, one model was trained to predict raw stem water potential values, while another was trained to predict baseline-adjusted values. Together, all models resulted in an average c oefficient of correlation of R2 = 0.73 and an average root mean squared error (R MSE) of 2.5 bars. Prediction accuracy decreased significantly when models were e xpanded to spatial maps (R2 = 0.33, RMSE = 3.31 [avg] ). These results indicate that both artificial neural networks and random forest frameworks can be used to predict stem water potential, but both approaches wer e unable to fully account for the spatial variability observed throughout the or chard. Overall, the most accurate maps were produced by the random forest model (raw stem water potential R2 = 0.47, RMSE = 2.71). The ability to predict stem w ater potential spatially can aid in the implementation of variable rate irrigati on."
NETL
NSTL