Olga Khabarova Heliophysical Laboratory Institute of Terrestrial Magnetism

Olga Khabarova Heliophysical Laboratory, Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation RAS (IZMIRAN), Moscow, Russia (olik 3110@aol. com) The short-term forecasts are based on information from spacecraft in the Sun-Earth libration point and different statistical models based on near-Earth plasma data and geomagnetic field disturbance level analysis. Such forecasts are rather exact, up to ~90%, but their alert time (ΔТ~1 hour) is too small for preventing of a storm hazard. The long-term forecasts try to predict general space weather and geomagnetic situation in relatively far future, using solar observations and different statistical models. There is no correct information about the accuracy level of this type of forecasts. The medium-term forecasts are most valuable for practical aims. Methods of their realization are mainly based on the recognition of the approach of CME-like structures to the Earth. ICME interactions with the Earth magnetosphere are considered as a cause of superintense geomagnetic storms. As a result, CME-like structures are considered to be producing moderate and weak magnetic storms as well. Most of the medium-term forecast methods are oriented towards the prediction of probability of the onset of severe storms (with Dst < - 80 n. T) only. So, the medium-term forecasts’ quality remains rather modest: even during a solar maximum (when the CME number is large) the successful forecasting rate is ~ 75%. The rate of successful forecasts falls down catastrophically during solar minimum. ~ 90% Short-term 100% diu Me rm -te mte ng NO CORRECT INFORMATION rm Lo QUALITY OF FORECASTS < 75% In spite of our growing knowledge we can predict only ~75% of geomagnetic storms with Dst<-80 n. T, but number of such storms is merely ~10% of the total (W. D. Gonzales, 1994). A superposed epoch analysis for Kataoka and Miyoshi list of 22 magnetic storms of mixed origin with Dst<-100 n. T, 1997 -2004 The onset day is shown by the vertical line. We can predict only 75% from 10% of the total number of magnetic storms. Kataoka, R. , and Y. Miyoshi (2006), Flux enhancement of radiation belt electrons during geomagnetic storms driven by coronal mass ejections and corotating interaction regions, Space Weather, 4, S 09004, doi: 10. 1029/2005 SW 000211. ) 75% 10% 7. 5% Dst < -80 n. T Dst<-30 n. T Geomagnetic storm prediction remains on the probabilistically casual level ~50%, as scientific community is interested in investigations of severe storms’ origin only. We can predict only 7. 5% of all geomagnetic storms. There is no assurance that weak and moderate storms obey the rules found for severe storms We can predict nothing… The solar wind becomes more dense and turbulent 1 day before severe geomagnetic storms of mixed origin. Long-lasting negative Bz + HIGH density+ stably LOW speed = geomagnetic storm Bz~ -3 n. T Bz~ -5 n. T N < 4 1/cm 3 N < 8 1/cm 3 V> 600 km/s Long-lasting negative Bz + HIGH speed + LOW density geomagnetic storm time delay between sharp growth of N and Bz falling lower than -5 n. T Long-lasting negative Bz + HIGH density+ stably LOW speed = geomagnetic storm Problems of medium-term magnetic storm forecasting seem to be a result of the shift of the interest of scientific community to the prognosis of severe magnetic storms only and to estimation of probability of CMEs’ (and, more rarely, CIRs) occurrence at 1 AU as the main prognostic factor. The most hopeful way of problems’ solving is changing of the dominating paradigm, and investigation of the weak and moderate magnetic storms’ origin from the prognostic point of view. Superposed epoch time series of solar wind and magnetospheric data for the set of 389 magnetic storms with Dst>-80 n. T, 1964 -2003. The onset day is shown by the vertical line. Mc. Pherron’s List of magnetic storms: http: //cdaw. gsfc. nasa. gov/geoma g_cdaw/data/cdaw 4/Mc. Pherron/ Mag. Storms 6403. txt The solar wind becomes more dense and turbulent 1 -2 days before moderate and weak geomagnetic storms. The solar wind density plays a more significant geoeffective role than it was previously assumed. A sharp density increase and consequent negative Bz can produce weak, moderate and even strong magnetic storms without any significant changes of the solar wind velocity. Most geomagnetic storms are associated with “high-density”, but not with high-velocity streams of the solar wind. The well-known rule: “High speed + long-lasting negative Bz + compression = severe geomagnetic storm” must be supplemented with the rule: “Sharp solar wind density increase + consequent negative IMF Bz = weak or moderate geomagnetic storm”. Intensity of the density-driven geomagnetic storms depends on maximum of the density, minimum of Bz values and a time-lag between them. Solar wind turbulence growths ~1 day before magnetic storms. All these facts may be used for prognostic aims. 1. Khabarova O. V. , Current Problems of Magnetic Storm Prediction and Possible Ways of Their Solving. Sun and Geosphere, http: //arxiv. org/ftp/arxiv/papers/0805. 0547. pdf , 2(1), pp. 32 -37, 2007 2. Khabarova O. , Pilipenko V. , Engebretson M. J. , and Rudenchik E. , Solar wind and interplanetary magnetic field features before magnetic storm onset. Proceedings of the Eighth International Conference on Substorms (ICS-8), edited by Syrjäsuo and Donovan, University of Calgary Press, Alberta, Canada, 127 -132, 2007, http: //ics 8. ca/proc_files/khabarova. pdf 3. Khabarova O. V. , and Yu. I. Yermolaev, Solar wind parameters' behavior before and after magnetic storms. , J. of Atm. and Sol. -Ter. Phys. , 70 (2 -4), pp. 384 -390, 2008