Training data and simulations; Modeling uncertainty with engression: a deep generative time-series approach

This collection contains training data and simulations for the paper: Basil Kraft, Steven Stalder, William Hugo Aeberhard, Nicolás Herrington Ruiz, Nicolai Meinshausen, Xinwei Shen, and Lukas Gudmundsson: Modeling uncertainty with engression: a deep generative time-series approach. Authorea. May 27, 2025. data.zip (harmonized_basins.zarr): training data engression_runs.zip: engression runs quantiles_runs.zip: quantile regression runs We recommend using this training data only to reproduce our work. In this case, please cite all original datasets (see below). Otherwise, it is highly recommended to use the CAMELS-CH dataset directly. The training data contains data from the following sources, aggregated to catchment scale: CAMELS-CH data https://zenodo.org/records/15025258 Static catchment properties: The New Swiss Glacier Inventory SGI2016: From a Topographical to a Glaciological Dataset, Linsbauer et al (2021): 10.3389/feart.2021.704189 Hochauflösende Bodenkarten für den Schweizer Wald, Baltensweiler et al. (2022): 10.3188/szf.2022.0288 The Habitat Map of Switzerland: A Remote Sensing, Composite Approach for a High Spatial and Thematic Resolution Product: Price et al. (2023), 10.3390/rs15030643 swissALTI3D, swisstopo (2018): https://www.swisstopo.admin.ch/en/height-model-swissalti3d#Additional-information, last access: 1 April 2024 SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty, Poggio et al. (2021): 10.5194/soil-7-217-2021 Meteorological features were retrieved from MeteoSwiss. https://www.meteoschweiz.admin.ch/dam/jcr:818a4d17-cb0c-4e8b-92c6-1a1bdf5348b7/ProdDoc_TabsD.pdf, last access: 1 April 2024 https://www.meteoschweiz.admin.ch/dam/jcr:4f51f0f1-0fe3-48b5-9de0-15666327e63c/ProdDoc_RhiresD.pdf, last access: 1 April 2024