首页|Report Summarizes Machine Learning Study Findings from Nanjing Tech University ( Assessing a machine learning-based downscaling framework for obtaining 1km daily precipitation from GPM data)
Report Summarizes Machine Learning Study Findings from Nanjing Tech University ( Assessing a machine learning-based downscaling framework for obtaining 1km daily precipitation from GPM data)
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 – Research findings on artificial intell igence are discussed in a new report. According to news reporting from Nanjing, People’s Republic of China, by NewsRx journalists, research stated, “Hydrometeo rological monitoring through satellites in arid and semi-arid regions is constra ined by the coarse spatial resolution of precipitation data, which impedes detai led analyses. The objective of this study is to evaluate various machine learnin g techniques for developing a downscaling framework that generates high spatio-t emporal resolution precipitation products.” Our news reporters obtained a quote from the research from Nanjing Tech Universi ty: “Focusing on the Hai River Basin, we evaluated three machine learning approa ches-Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Back Propagati on (BP) neural networks. These methods integrate environmental variables includi ng land surface temperature (LST), Normalized Difference Vegetation Index (NDVI) , Digital Elevation Model (DEM), Precipitable Water Vapor (PWV), and albedo, to downscale the 0.1° spatial resolution Global Precipitation Measurement (GPM) pro duct to a 1 km resolution. We further refined the results with residual correcti on and calibration using terrestrial rain gauge data. Subsequently, utilizing th e 1 km annual precipitation, we employed the moving average window method to der ive monthly and daily precipitation. The results demonstrated that the XGBoost m ethod, calibrated with Geographical Difference Analysis (GDA) and Kriging spatia l interpolation, proved to be the most accurate, achieving a Mean Absolute Error (MAE) of 58.40 mm for the annual product, representing a 14 % imp rovement over the original data. The monthly and daily products achieved MAE val ues of 11.61 mm and 1.79 mm, respectively, thus enhancing spatial resolution whi le maintaining accuracy comparable to the original product.”
Nanjing Tech UniversityNanjingPeople ’s Republic of ChinaAsiaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
