Predviđanje jakosti 25. Sunčevog ciklusa ARMA modelom i procjena geomagnetske aktivnosti u tom razdoblju

master's thesis

• Title (english): The forecast of the strength of the solar cycle 25 using ARMA model and estimation of geomagnetic activity in that period
• Author: Patrik Krčelić
• Mentor: Giuliana Verbanac (mentor)
• Committee member: Giuliana Verbanac (predsjednik povjerenstva)
• Committee member: Ivica Sović (član povjerenstva)
• Committee member: Željko Večenaj (član povjerenstva)
• Committee member: Zoran Pasarić (član povjerenstva)
• Committee member: Josip Stipčević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Geophysics); Zagreb
• Defense date and country: 2019-09-26, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Geophysics; Seismology and Physics of the Earth's Interior

Abstract

U radu je predstavljeno predviđanje 25. Sunčevog ciklusa analizom broja Sunčevih pjega ARMA modelom. Trenutno se nalazimo na kraju 24. Sunčevog cikusa, a 25. Sunčev ciklus počet će s minumumom Sunčeve aktivnosti koja se očekuje do kraja 2019. godine. Korišteni red metode bio je ARMA(3,3). Predviđanja su dosta dobra ukoliko imamo trend pada makismuma. Razlog tomu je što je ARMA statistička metoda, a statistički je više bilo padova Sunčevih maksimuma od rasta. Osim samog predviđanja, ispitana je i veza Sunčeve aktivnosti s geomagnetskim indeksima aa, Dst i AE. Spomenuti indeksi se dobro koreliraju s brojem Sunčevih pjega te je pretpostavljena linearna veza među parametrima. Za aa indeks je ta veza djelomično dobra pa se može iskoristiti u sklopu nekog budućeg istraživanja s obzirom da je trend dosta dobro opisan. AE indeks ima preveliku raspršenost te se linearna veza ne može primijeniti. Dst indeks daje vrlo pouzdane rezultate. Linearna veza ovdje je poprilično dobra i može se koristiti za procjenu aktivnosti. Također je napravljeno i predviđanje geomagnetskih indeksa iz rezultata predviđanja ARMA modelom i linearne regresije Sunčevih pjega i geomagnetskih indeksa.

Abstract (english)

The forecast of the 25. Solar cycle was made using the ARMA model analysis of the sunspot number. We are currently at the end of 24. solar cycle and 25. solar cycle will start with minimum of the solar activity, which is expected by the end of year 2019. The order of the model was ARMA(3,3). This forecast gives very good results for the trend of the downgoing maximum of the cycles. Reason for this behaviour is the fact that the ARMA is statistic model and statistically there are more downgoing than upgoing trends of maximum of the cycle. Aside from the forecast, the connection between sunspot number and geomagnetic indices aa, Dst and AE, was also analysed. The mentioned indices correlate well with sunspot number and the linear dependence between this parameters was assumed. For aa index this dependence is partially good, and can be used as a part of another research, considering the fact that the trend of index matches very well. AE index has too big standard deviation and linear function can not be used. Dst index gives very reliable results and can be used to estimate geomagnetic activity. Linear function matches very well and can be used. The forecast of the geomagnetic indices was also made using the results of the ARMA model and linear regression of geomagnetic indices.

• Keywords: Svemirsko vrijeme; Sunčevi ciklusi; ARMA model; geomagnetski indeksi
• Keywords (english): Solar weather; Solar cycles; ARMA model; geomagnetic indices
• Language: croatian
• URN:NBN: urn:nbn:hr:217:021556
• Study programme: Title: Graduate university study pf Physics - Geophysics; specializations in: Seismology and Physics of the Earth's Interior, Meteorology and Physical Oceanography; Course: Seismology and Physics of the Earth's Interior; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra fizike – geofizike (magistar/magistra fizike – geofizike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2019-10-10 09:14:17