A global land aerosol fine-mode fraction dataset (2001–2020) retrieved from MODIS using hybrid physical and deep learning approaches

[摘要] The aerosol fine-mode fraction (FMF) is valuable fordiscriminating natural aerosols from anthropogenic ones. However, mostcurrent satellite-based FMF products are highly unreliable over land. Here,we developed a new satellite-based global land daily FMF dataset (Phy-DLFMF) by synergizing the advantages of physical and deep learning methods ata 1 ∘ spatial resolution covering the period from 2001 to 2020. ThePhy-DL FMF dataset is comparable to Aerosol Robotic Network (AERONET)measurements, based on the analysis of 361 089 data samples from 1170AERONET sites around the world. Overall, Phy-DL FMF showed aroot-mean-square error (RMSE) of 0.136 and correlation coefficient of 0.68,and the proportion of results that fell within the ± 20 % expectederror (EE) envelopes was 79.15 %. Moreover, the out-of-site validationfrom the Surface Radiation Budget (SURFRAD) observations revealed that theRMSE of Phy-DL FMF is 0.144 (72.50 % of the results fell within the ± 20 % EE). Phy-DL FMF showed superior performance over alternative deeplearning or physical approaches (such as the spectral deconvolutionalgorithm presented in our previous studies), particularly for forests,grasslands, croplands, and urban and barren land types. As a long-termdataset, Phy-DL FMF is able to show an overall significant decreasing trend(at a 95 % significance level) over global land areas. Based on the trendanalysis of Phy-DL FMF for different countries, the upward trend in the FMFswas particularly strong over India and the western USA. Overall, this studyprovides a new FMF dataset for global land areas that can help improve ourunderstanding of spatiotemporal fine-mode and coarse-mode aerosol changes. Thedatasets can be downloaded from https://doi.org/10.5281/zenodo.5105617 (Yan, 2021).

[学科分类] 眼科学