Conditions For Coronal Mass Ejections To Originate From Solar Active Region

Coronal mass ejections(CMEs)are the major driving sources of hazardous space weather,while solar active regions(ARs)are the major sources of the CMEs.More than 60%CMEs originate from the ARs.So what are needed for an AR to produce CMEs,and how does a CME-rich AR produce CMEs successively?The thesis try to answer these questions.To answer the first one,we compared the recent super flare-rich but CME-poor AR 12192,with other four ARs;two were productive in both flares and CMEs and the other two were inert to produce any M-class or intenser flares or CMEs.By investigating the photospheric parameters based on the SDO/HMI vector magnetogram,we find the three productive ARs have larger magnetic flux,current and free magnetic energy than the inert ARs,which means that those ARs have larger sizes,and contain strong current system and free energy.Because enough magnetic free energy is a necessory condition for an AR to power a flare,thus,this conclusion is reasonable.Furthermore,we find that the combination of mean current helicity density and total unsigned current helicity can be used to distinguish the ARs' CME productivity.ARs producing more CMEs have larger mean current helicity(|Hc|),but the ARs only producing flares or pro-ducing no eruptions has smaller mean current helicity.Considering the total unsigned current helicity(Hctotai),NOAA 12192 has as large value as the CME-productive AR,NOAA 11158 and 11429,which means ARNOAA 12192 contains current helicity with opposite handedness but equivalent amount.Furthermore,the two ARs productive in both flares and CMEs contain strong cur-rent helicity concentrating along both sides of the flaring neutral lines,indicating the presence of a seed flux rope;they also have higher decay index in the low corona,showing weak constraint.The results suggest that whether an AR is able to power a CME is seemingly related to(1)if there is significant twisted core field at the erupt-ing position serving as the seed of the CME flux rope and(2)if the constraint of the overlying arcades is weak enough.To answer the second question,we firstly do a statistical investigation of the wait-ing times(time difference between a CME and its proceeding one)of 281 quasi-homologous CMEs(i.e.,QH-CMEs,the CMEs successively originating from the same ARs)from 28 super ARs in solar cycle 23.The waiting time distribution consists of two compo-nents with a separation at about 18 hours.The first component looks like a Gaussian distribution and peaks at 7 hours.The peak waiting time probably characterize the time scale of the gradual release of the free energy or the growth of instabilities triggered by the preceding CMEs.The QH-CMEs in the first component are more likely to be phys-ically related.We further add two super ARs,NOAA 11158 and 11429 in solar cycle 24 into our sample,and analyze waiting time distribution of all the CMEs.We found that the distribution still consists of two component,in which 188 ones has waiting time less than 18 hours,showing a Gussian-like distribution with a peak at 7.5 hours.We define the same location of one PIL as the same magnetic source location,the different parts of one PIL or different PILs as different magnetic source locations,and further identified the precise magnetic source locations(i.e.,CME-related flux systems)of 142 QH-CMEs with waiting times less than 18 hours to check their possible mech-anisms.Among those CMEs,90 ones(63%)originate from the same location as their preceding ones,which are defined as S type CMEs;while 52(37%)ones originate from the different locations compared with their preceding ones,which are defined as D type CMEs.Waiting time distributions of the two types of QH CMEs have different peaks,7.5 hours for S-types while 1.5 hours for D-types.We furtherly selected two cases:the S-type QH CME and its predecessor are from the same bipolar system in AR NOAA 11158,the D type QH CME and its predessor originate from the different flux systems in AR NOAA 11429.Detailed analysis,in-cluding the decay index n,squashing factor Q and twist number Tw,of two CMEs of each type indicates:during the S-type eruption,the seed flux rope experienced a partial eruption process,part of the flux rope erupted,forming the first CME,the other part survived and erupted later;the whole process could be desribed as a multiple-stage energy release scenario,during which the free energy of the AR was refilled by contin-uous shear motion and flux emergence.During the D-type CME,a flux rope along one PIL erupted as the first CME,influenced the magnetic source of the second CME,made the upper flux rope along the second PIL erupted.The upper flux rope along the second PIL had the opposite handedness as the lower flux rope along the same PIL,exerting a downward force to the lower flux rope.After the eruption of the upper flux rope,the lower one lost its constraing force,started to expand,rise and finally erupted out as the second CME.In conclusion,the S type CME and its preceding one may experience a process of recurring release of free magnetic energy;the D type CME is more likely to be promoted by a sequence of disturbances caused by its preceding one.Different peaks of the two types of QH CMEs may represent the charasteristic time scales of the two physicla process.