首页|Research from University of Leipzig Yields New Study Findings on Machine Learnin g (Marine cloud base height retrieval from MODIS cloud properties using machine learning)
Research from University of Leipzig Yields New Study Findings on Machine Learnin g (Marine cloud base height retrieval from MODIS cloud properties using machine learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Investigators discuss new findings in artificial intelligence. According to news reporting from Leipzig, Germany, by N ewsRx journalists, research stated, “Clouds are a crucial regulator in the Earth ’s energy budget through their radiative properties, both at the top of the atmo sphere and at the surface; hence, determining key factors like their vertical ex tent is of essential interest.” Financial supporters for this research include H2020 Marie Sklodowska-curie Acti ons. The news editors obtained a quote from the research from University of Leipzig: “While the cloud top height is commonly retrieved by satellites, the cloud base height is difficult to estimate from satellite remote sensing data. Here, we pre sent a novel method called ORABase (Ordinal Regression Autoencoding of cloud Ba se), leveraging spatially resolved cloud properties from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument to retrieve the cloud base height over marine areas. A machine learning model is built with two components to faci litate the cloud base height retrieval: the first component is an auto-encoder d esigned to learn a representation of the data cubes of cloud properties and to r educe their dimensionality. The second component is developed for predicting the cloud base using ground-based ceilometer observations from the lower-dimensiona l encodings generated by the aforementioned auto-encoder. The method is then eva luated based on a collection of collocated surface ceilometer observations and r etrievals from the CALIOP satellite lidar. The statistical model performs simila rly on both datasets and performs notably well on the test set of ceilometer clo ud bases, where it exhibits accurate predictions, particularly for lower cloud b ases, and a narrow distribution of the absolute error, namely 379 and 328 m for the mean absolute error and the standard deviation of the absolute error, respec tively.”
University of LeipzigLeipzigGermanyEuropeCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
