Articles | Volume 13, issue 4
Model experiment description paper
28 Apr 2020
Model experiment description paper |  | 28 Apr 2020

Configuration and intercomparison of deep learning neural models for statistical downscaling

Jorge Baño-Medina, Rodrigo Manzanas, and José Manuel Gutiérrez

Related authors

Refining Remote Sensing precipitation Datasets in the South Pacific: An Adaptive Multi-Method Approach for Calibrating the TRMM Product
Óscar Mirones, Joaquín Bedia, Sixto Herrera, Maialen Iturbide, and Jorge Baño Medina
EGUsphere,,, 2023
Short summary
Downscaling multi-model climate projection ensembles with deep learning (DeepESD): contribution to CORDEX EUR-44
Jorge Baño-Medina, Rodrigo Manzanas, Ezequiel Cimadevilla, Jesús Fernández, Jose González-Abad, Antonio S. Cofiño, and José Manuel Gutiérrez
Geosci. Model Dev., 15, 6747–6758,,, 2022
Short summary
Statistical downscaling with the downscaleR package (v3.1.0): contribution to the VALUE intercomparison experiment
Joaquín Bedia, Jorge Baño-Medina, Mikel N. Legasa, Maialen Iturbide, Rodrigo Manzanas, Sixto Herrera, Ana Casanueva, Daniel San-Martín, Antonio S. Cofiño, and José Manuel Gutiérrez
Geosci. Model Dev., 13, 1711–1735,,, 2020
Short summary

Related subject area

Atmospheric sciences
Implementation of a satellite-based tool for the quantification of CH4 emissions over Europe (AUMIA v1.0) – Part 1: forward modelling evaluation against near-surface and satellite data
Angel Liduvino Vara-Vela, Christoffer Karoff, Noelia Rojas Benavente, and Janaina P. Nascimento
Geosci. Model Dev., 16, 6413–6431,,, 2023
Short summary
The capabilities of the adjoint of GEOS-Chem model to support HEMCO emission inventories and MERRA-2 meteorological data
Zhaojun Tang, Zhe Jiang, Jiaqi Chen, Panpan Yang, and Yanan Shen
Geosci. Model Dev., 16, 6377–6392,,, 2023
Short summary
Rapid O3 assimilations – Part 1: Background and local contributions to tropospheric O3 changes in China in 2015–2020
Rui Zhu, Zhaojun Tang, Xiaokang Chen, Xiong Liu, and Zhe Jiang
Geosci. Model Dev., 16, 6337–6354,,, 2023
Short summary
Description and evaluation of the new UM–UKCA (vn11.0) Double Extended Stratospheric–Tropospheric (DEST vn1.0) scheme for comprehensive modelling of halogen chemistry in the stratosphere
Ewa M. Bednarz, Ryan Hossaini, N. Luke Abraham, and Martyn P. Chipperfield
Geosci. Model Dev., 16, 6187–6209,,, 2023
Short summary
A robust error correction method for numerical weather prediction wind speed based on Bayesian optimization, variational mode decomposition, principal component analysis, and random forest: VMD-PCA-RF (version 1.0.0)
Shaohui Zhou, Chloe Yuchao Gao, Zexia Duan, Xingya Xi, and Yubin Li
Geosci. Model Dev., 16, 6247–6266,,, 2023
Short summary

Cited articles

Ba, W., Du, P., Liu, T., Bao, A., Luo, M., Hassan, M., and Qin, C.: Simulating hydrological responses to climate change using dynamic and statistical downscaling methods: a case study in the Kaidu River Basin, Xinjiang, China, J. Arid Land, 10, 905–920,, 2018. a
Baño Medina, J., Manzanas, R., and Gutiérrez, J. M.: SantanderMetGroup/DeepDownscaling: GMD paper accepted for publication (Version v1.2), Zenodo,, 2020. a, b
Bedia, J., Baño-Medina, J., Legasa, M. N., Iturbide, M., Manzanas, R., Herrera, S., Casanueva, A., San-Martín, D., Cofiño, A. S., and Gutiérrez, J. M.: Statistical downscaling with the downscaleR package (v3.1.0): contribution to the VALUE intercomparison experiment, Geosci. Model Dev., 13, 1711–1735,, 2020. a, b, c, d
Cannon, A. J.: Probabilistic Multisite Precipitation Downscaling by an Expanded Bernoulli-Gamma Density Network, J. Hydrometeorol., 9, 1284–1300,, 2008. a
Chapman, W. E., Subramanian, A. C., Monache, L. D., Xie, S. P., and Ralph, F. M.: Improving Atmospheric River Forecasts With Machine Learning, Geophys. Res. Lett., 46, 10627–10635,, 2019. a
Short summary
In this study we intercompare different deep learning topologies for statistical downscaling purposes. As compared to the top-ranked methods in the largest-to-date downscaling intercomparison study, our results better predict the local climate variability. Moreover, deep learning approaches can be suitably applied to large regions (e.g., continents), which can therefore foster the use of statistical downscaling in flagship initiatives such as CORDEX.