This paper presents a new machine learning-based nowcasting model for hourly summer precipitation over the Eastern Alps. An artificial neural network (ANN) using the multi-layer perceptron algorithm was applied and evaluated against the Integrated Nowcasting through Comprehensive Analysis (INCA) nowcasting system and a multiple linear regression (MLR) model. Results show that the ANN model has a better nowcasting skill than the INCA model and the MLR model. The MLR model performs, too, also better than the INCA model. The improvement of precipitation intensity accuracy is substantial for both the morning to late evening period and for large rainfall thresholds. This study suggested that the machine learning approach is a promising methodology for precipitation forecasting.

中文翻译：

---

东部阿尔卑斯山夏季降水临近预报的机器学习方法

本文针对东阿尔卑斯山夏季每小时降水提出了一种基于机器学习的新的临近预报模型。应用了使用多层感知器算法的人工神经网络（ANN），并通过综合分析（INCA）临近预报系统和多元线性回归（MLR）模型对综合临近预报进行了评估。结果表明，与INCA模型和MLR模型相比，ANN模型具有更好的临近预报技术。MLR模型的性能也优于INCA模型。对于早晨到傍晚以及较大的降雨阈值，降水强度准确度的提高都是可观的。这项研究表明，机器学习方法是一种有前途的降水预测方法。