/* DATA DESCRIPTION: Citation: Mohn, Helge; Kreyling, Daniel; Wohltmann, Ingo; Lehmann, Ralph; Rex, Markus (2021): Benchmark dataset for 24-hour stratospheric ozone tendencies (SWIFT-AI-DS) [dataset]. PANGAEA, https://doi.org/10.1594/PANGAEA.939121 Abstract: SWIFT-AI-DS is a benchmark dataset that consists of samples that have been derived from two simulation runs (each 2.5 years long) of the chemistry and transport model ATLAS (Wohltmann and Rex, 2009; Wohltmann et al., 2010). This data set of nearly 200 million samples meets the requirements of a labelled data set and is ideally suited for training and testing of a machine learning based surrogate model. Two time periods were considered in the simulation runs: first from November 1998 to March 2001 and the second from November 2004 to March 2007. The dataset covers the entire Earth geographically, but is vertically restricted to the altitudes of the lower to middle stratosphere, for which the SWIFT (Rex et al., 2014; Kreyling et. al, 2017; Wohltmann et al., 2017) approach of 24-hour ozone tendencies can be applied. Applicability was determined in terms of the chemical lifetime of stratospheric ozone, which is a function of solar irradiance and altitude. It can be described by a dynamic upper bound [Kreyling et. Al, 2017]. Within the range where the chemical lifetime is longer than 14 days, ozone is not in quasi-chemical equilibrium. Moreover, this data set focuses on the region of the lower to middle stratosphere because it is the region with the largest contribution to the total ozone column. State-of-the-art physical process models for stratospheric chemistry require enormous computational time. Our research is focused on developing much faster, yet accurate, surrogate models for computing the 24-hour tendencies of stratospheric ozone. Much faster models of stratospheric ozone provide a new application area such as for climate models. These surrogate models benefit greatly from the methodological and hardware improvements of the last decade. Each simulation run uses the full stratospheric chemistry model to solve a system of differential equations involving 47 chemical species and 171 chemical reactions at a very high (<< seconds) and variable temporal resolution. The ATLAS model is driven by ECMWF reanalysis data (either ERA-I or ERA5). The air parcel state has been sampled at a 24-hour time step (00:00 UTC model time). During postprocessing some variables are stored as 24-hour averages, as 24-hour tendencies or as the state at the beginning of the 24-hour time step. The dataset is stored in 12 monthly netCDF-files. Keyword(s): Atmospheric chemistry; Atmospheric physics; climate science; machine learning; ozone; stratospheric chemistry; stratospheric ozone; Surrogate model Related to: Kreyling, Daniel; Wohltmann, Ingo; Lehmann, Ralph; Rex, Markus (2018): The Extrapolar SWIFT model (version 1.0): fast stratospheric ozone chemistry for global climate models. Geoscientific Model Development, 11(2), 753-769, https://doi.org/10.5194/gmd-11-753-2018 Lehmann, Ralph (2004): An Algorithm for the Determination of All Significant Pathways in Chemical Reaction Systems. Journal of Atmospheric Chemistry, 47(1), 45-78, https://doi.org/10.1023/B:JOCH.0000012284.28801.b1 Rex, Markus; Kremser, Stefanie; Huck, P; Bodeker, G E; Wohltmann, Ingo; Santee, M L; Bernath, P (2014): Technical Note: SWIFT - a fast semi-empirical model for polar stratospheric ozone loss. Atmospheric Chemistry and Physics, 14(13), 6545-6555, https://doi.org/10.5194/acp-14-6545-2014 Wohltmann, Ingo; Lehmann, Ralph; Rex, Markus (2010): The Lagrangian chemistry and transport model ATLAS: simulation and validation of stratospheric chemistry and ozone loss in the winter 1999/2000. Geoscientific Model Development, 3(2), 585-601, https://doi.org/10.5194/gmd-3-585-2010 Wohltmann, Ingo; Lehmann, Ralph; Rex, Markus (2017): Update of the Polar SWIFT model for polar stratospheric ozone loss (Polar SWIFT version 2). Geoscientific Model Development, 10(7), 2671-2689, https://doi.org/10.5194/gmd-10-2671-2017 Wohltmann, Ingo; Rex, Markus (2009): The Lagrangian chemistry and transport model ATLAS: validation of advective transport and mixing. Geoscientific Model Development, 2(2), 153-173, https://doi.org/10.5194/gmd-2-153-2009 Further details: SWIFT-AI-DS: Description of Variables (URI: https://download.pangaea.de/reference/114668/attachments/Description_Variables.pdf) Funding: Helmholtz Association of German Research Centres (HGF) (URI: https://www.helmholtz.de), grant/award no. HIDSS-0005: Helmholtz School for Marine Data Science (MarData) (URI: https://www.mardata.de/) Comment: The benchmark-dataset consists of training- and test-data. Variables are being described in the document Description_variables.pdf. Training-Data: The training data consists of files that include ca. 100 million data samples. Each data sample consists of the input and output features that can be used to train a data-driven model on a regression task Input X: choice of variables (see document Description_variables.pdf) Output y: 24-hour tendency of stratospheric ozone Test-Data: Similar to the training data, but this test-data includes ca. 100 million data samples that have not been used for training. It can be used to assess model performance. Parameter(s): Binary Object (Binary) * PI: Mohn, Helge (helge.mohn@awi.de) Binary Object (Media Type) (Binary (Type)) * PI: Mohn, Helge (helge.mohn@awi.de) Binary Object (File Size) [Bytes] (Binary (Size)) * PI: Mohn, Helge (helge.mohn@awi.de) Binary Object (MD5 Hash) (Binary (Hash)) * PI: Mohn, Helge (helge.mohn@awi.de) Change history: 2022-07-04T10:38:36 - Lehmann, Ralph was added as author. License: Creative Commons Attribution 4.0 International (CC-BY-4.0) (URI: https://creativecommons.org/licenses/by/4.0/) Status: Curation Level: Basic curation (URI: https://wiki.pangaea.de/wiki/Curation_levels) Size: 24 data points Files: All files referred to in data matrix can be downloaded in one go as ZIP or TAR (PANGAEA user account required). To download single files, use the filename from data matrix appended to "https://download.pangaea.de/dataset/939121/files/". */ Binary Binary (Type) Binary (Size) [Bytes] Binary (Hash) SWIFT-AI_test_month_01.nc application/x-hdf 2.5 GBytes f607e12dc39a557a079bf563a5df8f71 SWIFT-AI_test_month_02.nc application/x-hdf 2.2 GBytes 1c5cb7977bb4bd9c1448fe965477652d SWIFT-AI_test_month_03.nc application/x-hdf 2.2 GBytes a3b9b3c7412b99f7d24f38d7b6bd2b77 SWIFT-AI_test_month_04.nc application/x-hdf 1.4 GBytes 08bb77d54c04be0da61eb57359ddd637 SWIFT-AI_test_month_05.nc application/x-hdf 1.3 GBytes 283dedf4a695d06d24f83812698d0f8e SWIFT-AI_test_month_06.nc application/x-hdf 1.4 GBytes dc76bb71cb05e7e04cb60f05e110061f SWIFT-AI_test_month_07.nc application/x-hdf 1.6 GBytes 987cebf363f048a6fa9c03f38f1d3b0c SWIFT-AI_test_month_08.nc application/x-hdf 1.6 GBytes 76109800f6ced1704e9f631baa72f0f3 SWIFT-AI_test_month_09.nc application/x-hdf 1.6 GBytes ec57a0b09920ab28d4e446f6f8b665b8 SWIFT-AI_test_month_10.nc application/x-hdf 1.7 GBytes 5757b40ab8ce2a916772fd5818cba75c SWIFT-AI_test_month_11.nc application/x-hdf 2.3 GBytes ee65d6c447d8425ec5d3136de3aa4598 SWIFT-AI_test_month_12.nc application/x-hdf 2.6 GBytes f010b21fd3130a12caff1ac8441e9d48 SWIFT-AI_train_month_01.nc application/x-hdf 2.5 GBytes 8d496507595be444dc59e95fdd26e166 SWIFT-AI_train_month_02.nc application/x-hdf 2.2 GBytes ddbd866389ef8f279c7858f121a847f8 SWIFT-AI_train_month_03.nc application/x-hdf 2.2 GBytes 92a3e88e1fc9f547899881f49e0f8a99 SWIFT-AI_train_month_04.nc application/x-hdf 1.4 GBytes 4301ef544ef1ce1559d7b2aa5d1ae56d SWIFT-AI_train_month_05.nc application/x-hdf 1.3 GBytes ff9b2d561d88658ae2944bf5655426b1 SWIFT-AI_train_month_06.nc application/x-hdf 1.4 GBytes 42995d9d369b8ebb6fe0ae355112a29e SWIFT-AI_train_month_07.nc application/x-hdf 1.6 GBytes e6373e873853680ee37a725ca8abbf75 SWIFT-AI_train_month_08.nc application/x-hdf 1.6 GBytes e3dd784a9cd7b56b288d996daf52a658 SWIFT-AI_train_month_09.nc application/x-hdf 1.6 GBytes f90cdfc66cf5ba436c95346241be0543 SWIFT-AI_train_month_10.nc application/x-hdf 1.7 GBytes 2bdd198d5e7c3bd28f31463245e69295 SWIFT-AI_train_month_11.nc application/x-hdf 2.3 GBytes 5c39a7562570546cd6f82b7e86f28194 SWIFT-AI_train_month_12.nc application/x-hdf 2.6 GBytes 205077f5cf874f6a041007288ad6723c