Training Data for PRISM-DNN
收藏官方服务：
资源简介：
This dataset supports the training and validation of PRISM-DNN (Pretrained Remote sensing pIxel-based Spatial-teMporal Deep Neural Network), a deep learning framework designed for fine- and coarse-mode aerosol optical depth (AOD) retrieval from satellite observations. The dataset includes satellite-derived variables (e.g., MODIS reflectance, viewing geometry, and meteorological parameters) along with collocated ground-based aerosol measurements. The model implementation and training code are available at DOI: 10.5281/zenodo.15788087.
提供机构：
Zenodo创建时间：
2025-07-02
