# COSPAR space weather catalogue

The information showed here is collected in the framework of COSPAR through a web interface at http://www.spaceweathercatalogue.org/. In the spirit of COSPAR, the information is available without restriction to the whole community.

If your asset is not listed, please register it at http://www.spaceweathercatalogue.org/index.php?page=add&action=add (You need to register first.)

You can download the whole catalogue in PDF format at http://www.spaceweathercatalogue.org/COSPARCatalogue.pdf

Name | Description | |
---|---|---|

BAS global dynamic radiation belt model | The BAS global dynamic radiation belt model calculates the energetic electron flux in the Earth's radiation belts. It can also be used to forecast changes in the radiation belts using a time series of Kp and data from the GOES satellites. It covers a region from L=2-7 and energies of 0.1-5 MeV. It is being developed as part of the FP7 SPACECAST project. | website |

Celeste3D | Kinetic plasma simulation tool. | |

CLUSTERING BASED MODEL – A pre-process to obtain a forecast curve for the ionospheric critical frequency foF2 | Cluster analysis is used to classify monthly medians of the ionospherical critical frequency foF2 data for 13 stations in Europe during the period 1958-1998. The algorithm used agglomerates the data consisting of 4801 samples of daily variations into 6 sets of sizes ranging from 1334 to 431 samples, characterized mainly by R12 and seasons. Any dependency on the geomagnetic coordinates was investigated and the dependency on the calendar year by a dependency on R12 was identified, hence over looking any possible dependency on atmospherical conditions [Mizrahi et al., 2002]. The work aims to be a pre-process to obtain a forecast curve for the ionospheric critical frequency foF2. | website |

COOLFLUID | Component based scientific computing platform for space weather simulation based on Ideal MHD equations. Allows to compute the interaction of solar wind with planetary magnetosphere (including bow shock), simulate trajectory of Coronal Mass Ejection and its effect on the solar wind. | |

Democritus | Modelling the interaction of the space environment with spacecrafts | |

Disturbed D-region Electron Density Model | Physical model for computing time and height electron density profiles during solar flares. The model rests on the electron continuity equation, relating the space measured GOES X-ray flux (0.1-0.8 nm) and the ground measured response of VLF signals in terms of amplitude and phase extrema. | |

Drag Temperature Model (DTM) | Semi-empirical thermosphere model. Temperature, density and partial densities are modeled. The predictions are given for a single position, as is usual and convenient in orbit computation. | website |

eHeroes Dose Simulator | dose for astronauts | |

Exospheric Solar Wind Model | This is an exospheric model of the solar wind with only protons and electrons (we defer the inclusion of heavy ions to upcoming versions of the code), with a non-monotonic total potential for the protons and with a Lorentzian (kappa) velocity distribution function (VDF) for the electrons. This code is developed for the coronal holes. | website |

FLIP3D-MHD | Global Fluid Code based on the Lagrangian-Eulerian particle-based approach FLIP. | |

Geodetic and Geophysical Research Institute,Hungarian Academy of Sciences | Geophysical Observatory Nagycenk of the Hungarian Academy of Sciences recording elements of the geomagnetic field (as member of INTERMAGNET ionospheric sounding observation of whistlers recording Schumann resonance frequencies measurement of atmospheric electric potential gradient registration of point discharge currents | website |

GETY foF2 MAP MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches are shown to be promising for such cases. The only basic requirement is the availability of some representative data. Modelling capabilities with the use of Genetic Programming approach were introduced [Tulunay Y., et al., 2007]. In these approaches, for particular cases, some of the Near Earth Space parameters and their effects on Geospace have been investigated. One of the parameters is the Ionospheric F layer critical frequency, foF2, which is a parameter designating the maximum usable frequency. In addition to diurnal, seasonal and solar variability of foF2, during disturbed conditions induced by Solar Storms, the physics of Ionosphere become more complex and non-linear. Despite the fact that foF2 is a crucial parameter of telecommunication there are limited number of ionosondes over the world. Moreover, during disturbed conditions, for some of the ionosondes, the quality of measurements decrease and missing number of data increases. Different than the previous works, genetic programming approach is used and an algebraic mapping function is constructed to generate foF2 maps [Tulunay Y., et al., 2007]. The Genetic Programming by Tolga Yapıcı (GETY) foF2 Map Model is employed to map the instantaneous foF2 or forecast foF2 values [Tulunay Y., et al., 2007] [Altuntaş E., et al., 2007]. | website |

iPIC | 3D electromagnetic implicit Particle-in-Cell code for the simulation of magnetic reconnection in planetary magnetospheres. | |

METU-FNN1 foF2 FORECAST MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches such as the Neural Network (NN) based modelling are shown to be promising for such cases. The only basic requirement is the availability of some representative data. An artificial NN is a system of inter-connected computational elements, the neurons, operating in parallel, arranged in patterns similar to biological neural nets and modeled after the human brain (Tulunay, E., 1991). Highly nonlinear and complex processes in the Near-Earth Space can be modeled by the METU-NN models, which have been developed by the Group since 1990 (Altinay et al., 1997). The METU-NN has one input one hidden and one output layer. Levenberg-Marquardt Backpropagation algorithms with validation stop are used for training. A Fuzzy-Neural Network (METU-FNN) approach is employed for forecasting ionospheric critical frequency (foF2) during one of the major storms, Halloween 2003 storm [Altuntaş et al., 2007-a] [Altuntaş et al., 2007-b]. The model consists of a fuzzy interference part with two rule spaces and two feedforward neural network models, quiet NN (qNN) and disturbed NN (dNN) is employed to forecast the foF2 values [Altuntaş et al., 2007-a]. The METU-FNN1 foF2 Forecast Model is employed to forecast the foF2 values up to one hour in advance. | website |

METU-FNN2 foF2 FORECAST MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches such as the Neural Network (NN) based modelling are shown to be promising for such cases. The only basic requirement is the availability of some representative data. An artificial NN is a system of inter-connected computational elements, the neurons, operating in parallel, arranged in patterns similar to biological neural nets and modeled after the human brain (Tulunay, E., 1991). Highly nonlinear and complex processes in the Near-Earth Space can be modeled by the METU-NN models, which have been developed by the Group since 1990 (Altinay et al., 1997). The METU-NN has one input one hidden and one output layer. Levenberg-Marquardt Backpropagation algorithms with validation stop are used for training. A Fuzzy Neural Network (METU-FNN) approach is employed for forecasting ionospheric critical frequency (foF2) [Şenalp et al., 2011]. Previous METU-FNN models have been considered [Altuntaş et al., 2007] [Tulunay Y. et al., 2008]. The model consists of a fuzzy interference part providing expert information input to a feedforward neural network module, i.e. METU-NN [Şenalp et al., 2011]. The METU-FNN2 foF2 Forecast Model is employed to forecast the foF2 values up to one hour in advance. | website |

METU-NN SCHUMANN RESONANCE INTENSITY FORECAST MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches such as the Neural Network (NN) based modelling are shown to be promising for such cases. The only basic requirement is the availability of some representative data. An artificial NN is a system of inter-connected computational elements, the neurons, operating in parallel, arranged in patterns similar to biological neural nets and modeled after the human brain (Tulunay, E., 1991). Highly nonlinear and complex processes in the Near-Earth Space can be modeled by the METU-NN models, which have been developed by the Group since 1990 (Altinay et al., 1997). The METU-NN has one input one hidden and one output layer. Levenberg-Marquardt Backpropagation algorithms with validation stop are used for training. A revised METUNN model is developed [Tulunay, Y. et al., 2008]. The model is based on the previous METUNN model described in Tulunay Y. et al (2004). The revision is on initializing the initial layer weights of the NN [Tulunay, Y. et al., 2008]. Schumann Resonances (SR) are the electromagnetic (EM) phenomena which occur in the cavity formed by the conducting Earth and the ionosphere. The METU-NN Schumann Resonance Intensity Forecast Model is employed to forecast the SR values up to 36 hours in advance. | website |

METU-NN-C GPS TEC FORECAST MAP MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches such as the Neural Network (NN) based modelling are shown to be promising for such cases. The only basic requirement is the availability of some representative data. An artificial NN is a system of inter-connected computational elements, the neurons, operating in parallel, arranged in patterns similar to biological neural nets and modeled after the human brain (Tulunay, E., 1991). Highly nonlinear and complex processes in the Near-Earth Space can be modeled by the METU-NN models, which have been developed by the Group since 1990 (Altinay et al., 1997). The METU-NN has one input one hidden and one output layer. Levenberg-Marquardt Backpropagation algorithms with validation stop are used for training. A cascade model based on Hammerstein system modeling has a nonlinear static block cascaded with a linear dynamic block. METU-C is a cascade model. METU-NN estimates the internal state-like variables of the METU-C. Different representations of nonlinearities have been used in the first block of the cascade model. METU-NN-C model consists of one METU-C and one METU-NN (Senalp et al., 2006-a). The METU-NN-C GPS TEC Forecast Map Model with polynomial and Bezier curve nonlinearity representations are employed to forecast the Total Electron Content (TEC) values and maps up to one hour in advance. | website |

METU-RFNN foF2 FORECAST MODEL | Natural processes such as the near-Earth space are highly complex, nonlinear and time varying. Therefore, mathematical modeling is usually very difficult or impossible. Data-driven approaches such as the Neural Network (NN) based modelling are shown to be promising for such cases. The only basic requirement is the availability of some representative data. An artificial NN is a system of inter-connected computational elements, the neurons, operating in parallel, arranged in patterns similar to biological neural nets and modeled after the human brain (Tulunay, E., 1991). Highly nonlinear and complex processes in the Near-Earth Space can be modeled by the METU-NN models, which have been developed by the Group since 1990 (Altinay et al., 1997). The METU-NN has one input one hidden and one output layer. Levenberg-Marquardt Backpropagation algorithms with validation stop are used for training. A Recurrent Fuzzy-Neural Network (METU-RFNN) approach is employed for forecasting ionospheric critical frequency (foF2) during the major storms of 2003, i.e. the Halloween and the Superstorm of 2003 [Tulunay Y. et al., 2007]. The METU-RFNN foF2 Forecast Model is employed to forecast the foF2 values up to one hour in advance. | website |

MPI-AMRVAC | Numerical code for systems of partial differential equations, with an emphasis on shock-capturing discretizations in a finite volume approach. This open-source version has been initiated at K.U.Leuven's Centre for Plasma Astrophysics (CPA) by Bart van der Holst and Rony Keppens. It offers a large variety of physics modules and discretizations, is parallelized with MPI, and has fully automated grid refinement in any dimensionality. At CPA, it is used also for space weather simulations by exploiting the full (compressional) ideal MHD equations as a base model, with added non-ideal effects. The code is used for coronal reconstruction (using magnetogram data as boundary conditions), 2D and 3D solar wind models (polytropic and MHD), 2.5D and 3D CME initiation and early evolution simulations, interaction of IP CMEs and shocks with the background solar wind and the Earth magnetosphere. It has been applied to Jupiter's magnetosphere as well. It is distributed through a subversion repository. | website |

Plasmapause location | The program retrieves the geomagnetic activity level index Kp observed during the date given as input and 24 hours before. Then, it calculates the position of the plasmapause for the required time period, assuming the corotation and using the convection electric field model E5D (McIlwain, 1986) and the associated magnetic field M2. The mechanism for the formation of the plasmapause is assumed to be the quasi-interchange instability (Lemaire and Gringauz, 1998). | website |

Plasmasphere density | The 3D dynamic model of the plasmasphere provides the number density of the particles and the position of the plasmapause as a function of time (Pierrard and Stegen, 2008). In the simulations presented on this space weather portal, the user gives the date of the event as an input. The program retrieves the observed geomagnetic activity level index Kp and calculates the position of the plasmapause and the number density of the electron predicted by the model in the geomagnetic equatorial plane. The dynamics of the model is directly related to the convection electric field that depends on Kp. Animated simulations show the dynamical plasmasphere every half hour of the simulated day. | website |

Polar wind | This terrestrial polar wind model is based on the exospheric approach and uses the formulas described in Pierrard and Lemaire (1996) for open magnetic field lines considering trapped but no incoming particles. It considers 3 particle species (electrons, protons and O+ ions) for which the user introduces the values at the exobase altitude. The effect of photoelectrons that can increase the O+ flux and the possible minor presence of He+ ions are neglected. The model gives the profiles of number density, electric potential, flux, bulk velocity, temperatures (parallel and perpendicular) and the heat flux of the different particle species (+ crosses for electrons, diamonds for protons, dots for O+ ions) up to the radial distance required by the user. | website |

SOTERIA online solar spectrum nowcast and forecast | online nowcast and forecast of the solar spectral irradiance from 1-400 nm + the total solar irradiance. Forecast is presently 4 days ahead, will soon be > 30 days ahead. Nowcast is updated very hour, with < 15' latency. | website |

SOTERIA Solar Wind Forecasting | Forecasting Tool for solar wind speed, temperature, velocity and density based on data assimilation applied to other pre-existing models such as those based on coronal holes observations (B. Vrsnak, M. Temmer, and A. Veronig, Solar Phys. 240, 315 (2007)) or magnetograms. | |

TSM-assisted-Digisonde (TaD) Profiler | TaD improves the electron density profile in topside F region provided by Digisonde software and extends it in plasmasphere up to 20,000 km. TaD is based on the Topside Sounder Model (TSM), which provides the topside scale height and upper transition height on a global scale, depending on season, solar radio flux F10.7, and Kp. TaD profiler is extensively tested, verified and validated by using all available independent measurement, as GPS-derived TEC, CHAMP reconstruction technique, Millstone Hill, Malvern, and Kharkov ISR topside profiles. | website |

VAC | Numerical code for systems of partial differential equations, with an emphasis on shock-capturing discretizations in a finite volume approach. The open-source version has been adapted at K.U.Leuven's Centre for Plasma Astrophysics (CPA) by Bart van der Holst, and is used for space weather simulations at CPA by exploiting the full (compressional) ideal MHD equations as a base model, with added non-ideal effects. It is parallelized using MPI. The code is used for coronal reconstruction (using magnetogram data as boundary conditions), 2D and 3D solar wind models (polytropic and MHD), 2.5D and 3D CME initiation and early evolution simulations, interaction of IP CMEs and shocks with the background solar wind and the Earth magnetosphere. | website |