Полярные геомагнитные возмущения и связанные с ними явления : материалы международного симпозиума «Полярные геомагнитные явления», 25-31 мая 1986 г., Суздаль / Междунар. геодез. и геофиз. союз, междунар. ассоциация геомагнетизма и аэрономии, АН СССР, КНЦ, ПГИ. – Апатиты : [б. и.], 1989.

V upon the process of FEM and u p o n U T a n d D s^ -index. H owever in the IPh all the members of this cluster become a lready r e l atively independent one fr o m another processes, but pairs v + v ^ a n d F E M + U T are r e tained and m a n i f est themselves at the next phase - MPh, wh e r e th e y a g a i n unite into cluster w i t h , but already i n more complicated c o m p osition which, slightly changed, is preserved at RPh. F o r storm 2 (Pig.4) the role of D st is also different at different phases, but the c h a racter of thia role ia noticeably different, in particular Dst is practically not connected wi t h pair FEM + U T i n all the phases and Dst connection wi t h w h i c h is c learly seen at M P h 1 and M P h 2 phases, is fu lly broken in RPh. 2. These particularities of D s-j. role change at QM S phases lead to the possibility of the w r o n g interpretation of the generalising role of this pa r ameter through the whole period. Thus, i n st o r m 2 D st manifests itself i n PP as a r e l ativ ely independent process wh e r e a s D s1; has o l o s e i n t e r c o n n e c t i o n with different groups of parameters at each phaee, 3. The role of the leading base physical process ca n be played by the other parameters playing an important but different role in each separate GMS phase. Fo r example, temperature T turned out to be such parameter in storm 2. 4. It has b e e n r e vealed that none of the proposed "theoretical pa r a m e ters" ai m e d to e xp l ain GMS, such аз E y , 6 o r F M , i a able to ex p l a i n GMS as a phenomenon in the whole due to its sharp time non-uniformity. Thus, for storm 2 parameters ( * z - E < V Z ) and ( ec(. , Fp ) w h i c h were thought to be g e o effective, in reality turned out to be not effective at all. 5. None of the G M S separate phases, e v e n such lo n g ones as RPh, cannot p retend to be a phase wh i c h properties ca n fully determine the physical pecul iarities revealed t hrough the whole process. All of t h e m contribute ess e n t i a lly to the whole picture. H owever the c o n n e c t i o n oan be revealed he r e wh i c h is not seen at different phases due to their short duration. Fo r example, for storm 2 the natural connection of ( D R , FEM , U T ) i n the full phase is revealed, i.e. the c o n n ection of the r i n g - current f i e l d w i t h the injecting and dissipating energy, though at different phases (MPh 1, M P h 2, RPh) the only existing c o n nection is that of pair FEM +Uj, , an d D R is i ncluded in other clusters a n d is p r a o tically independent fr o m it. 6. At 1 ho u r shift of the inte r p l a n e t a r y parameters relative to g eomag- n e t o s p h e r ic-ionospheric ones, dendritic structure is changed, so the physical picture of tha p h e n o m e n o n i n whole too. 3. C o n c l u s i o n . The obtained in the a n alysis data could not, i n principle be obtained during standard procedures of the mathematical statistics. T h e r e fore, the proposed m e t h o d of G M S sequential - cluster analysis can be consider ed as a positive al ternative to these procedures, R E F E R E N C E S 1. V e r h a g e n C . J . D . M . , D u i n R.P.W., G r o e n F . C . A . , Joosten J.C. and Verbeek P.W. P rogress report on pattern recongnltion. - Reports on P r ogress i n Physics, 1980, v.43, No 6, p . 785-831. 2. Har m u t h H.F. Sequency theory. Fou n d a t i o n s a n d applications. Academic Press, N e w - Y ork-San-Prancisco-London, 1977. 3. K i n g J.H. C omposite inte r p l a n e t a r y magnetic field and plasma tape. - S M 4 1 B , N3SDC-A, G r e e n b e l t , 1979. 87