This archive contains the magnetohydrodynamic (MHD) modeling materials associated with the manuscript: "A Near-Real-Time Data-Assimilative Model of the Solar Corona" Cooper Downs, Jon A. Linker, Ronald M. Caplan, Emily I. Mason, Pete Riley, Ryder Davidson, Andres Reyes, Erika Palmerio, Roberto Lionello, James Turtle, Michal Ben-Nun, Miko M. Stulajter, Viacheslav S. Titov, Tibor Török, Lisa A. Upton, Raphael Attie, Bibhuti K. Jha, Charles N. Arge, Carl J. Henney, Gherardo Valori, Hanna Strecker, Daniele Calchetti, Dietmar Germerott, Johann Hirzberger, David Orozco Suárez, Julian Blanco Rodríguez, Sami K. Solanki, Xin Cheng, Sizhe Wu This archive is intended for transparency and reproducibility purposes. It contains the MHD model source code, run inputs, run outputs, synthetic observables, and example python scripts for working with the model data. This includes data from the MHD calculation and the surface flux transport calculation used to create the magnetic boundary conditions for MHD model. See below for more information on each of the subfolders in this archive. Folders A, F, and I together contain all files required for recreating the MHD calculation. # Contents The subfolders are organized as follows: ### A_MAS_source_and_example_run This folder contains the source code for the high-performance MHD code "Magnetohydrodynamic Algorithm outside a Sphere" (MAS) and associated files. MAS is written in Fortran. The dependencies are very straightforward. See `source/README_MAS.txt` and the associated Makefile for compilation instructions. It also contains a working MAS input file and other files that can be used to begin and repeat the time-evolving MHD calculation. See `run_example/README_Run.txt` for more information. Source code for the additional modeling codes used in this work (HipFT and Pot3D) is available on GitHub: https://github.com/predsci/HipFT https://github.com/predsci/POT3D ### B_data_readers This folder contains some example python scripts that illustrate how to read the model data files. This includes an example that will convert the raw 3D data to physical units and place all variables on a common, non-staggered mesh. It also describes to read the other types of data in this archive. See `README_DataReaders.txt` for more information. ### C_run_indexes We provide two ascii datatables that tabulate the snapshots of the time-evolving Live Prediction simulation. These tables can be used to translate between a HipFT, MAS, or FITS sequence number to physical time (e.g. 001820 corresponds to 2024-03-16T19:00:00.000Z). There are two versions of this table: - `C1_run_index_full.ecsv`: All 776 hourly dumps. - `C2_run_index_mhd_subset.ecsv`: The subset of dumps provided in this archive. These files are saved in Astropy's ECSV format (https://docs.astropy.org/en/stable/io/ascii/ecsv.html), so they can easily be read in as rich astropy data tables. ECSV is generally backwards compatible with any machine reader that can handle ascii tables with delimiters (e.g. .csv files or any standard spreadsheet program). ### D_magnetic_data This folder contains all of the mapped magnetic data used for assimilation within the simulation time period. The datafiles are in the standard HDF5 format expected by HipFT. They contain 3 layers: 1) The mapped Br data, 2) the assimilation weight, 3) mu: the cosine of the center-to-limb angle. These datafiles are a produced similarly to how the MagMap package creates HipFT inputs (see https://github.com/predsci/magmap ). HMI Data: - The `2024` folder includes all HMI-NRT data frames organized by month and day. - e.g. `2024/04/08/hmi_map_720s_nrt_20240408T190001_mod.h5` - This data has already had the orbital velocity correction applied (see methods). PHI Data: - The `phi_ec24` folder includes all mapped SolO/PHI FDT data, organized by timestamp. - e.g. `phi_ec24/04/02/phi_custom_assim_vector_0402_03.h5` - There are 12 copies of each frame because they were assimilated over 12 hours. `index_files/all-files.csv`: This is the "index file" used to stage assimilation into HipFT. It has a relative path to all files and is what was used by the HipFT run. ### E_hipft_data This folder contains the HipFT input parameters file and all output files from the run within the simulation time period. The output files are 2D phi,theta maps saved at 1024x512 in HDF5 format. See `INFO_HipFT_Eclipse2024.txt` for some additional details. ### F_boundary_data This folder contains all the boundary condition files read in by MAS during the time-evolving Live Prediction simulation. These files are saved using the same convention as the actual run so the MAS input file provided in A can be used as-is. There are 4 types of files: - `phi`: The energization potential used to compute the energizing electric field. - `br`: The time evolving map of Br at the mesh resolution of MAS. - `vp`: Longitudinal flow file with the large-scale differential rotation profile (relative to Carrington). - `vt`: Co-Latitudinal flow file with the large-scale meridional flow profile. ### G_fits_data This folder contains data files of the synthetic images described in the manuscript. They cover the same sequences as the 3D data files in H_mhd_data. They are saved in the standard FITS format for astronomical images. See `INFO_FITS.txt` inside this folder for more information. ### H_mhd Here we include the largest subset of 3D MHD data that we could fit in this archive. This subset includes: - 1) A 3D dump for every 24 hours of the simulation (19 UT). - 2) One day at 1 hour cadence (2024/04/07 19UT to 2024/04/08 19UT). - 3) The 3D data for all states visualized in figures in the manuscript. For every dump we include 11 files: density, temperature, velocity (r,t,p), magnetic field (r,t,p), and current (r,t,p) Because each dump is large (~2.15 GB) they are separated into individual zip archives to make it easier. ### I_mhd_restart.h5 This is a large "restart" file required for initializing the time-evolving run. It is used by the run input file provided here (see A_MAS_source_and_example_run). ### J_mhd_steady_state_data We also include the final state from 3 of the "time-stationary" a.k.a. "steady-state" MHD calculations that are discussed in the supplementary materials. These correspond to the runs pictured in Fig. S6 panels D-F. The sequence numbers match the times in the live-prediction index file. ## Additional Notes This simulation was created to predict the coronal structure of the April 8th, 2024 total solar eclipse. Please find additional descriptions and materials here: https://www.predsci.com/eclipse2024 MAS is developed and maintained by Predictive Science Inc. (PSI) in San Diego California. If you have questions, concerns, or issues installing or running this code for reproducibility purposes, please contact Cooper Downs .