神经网络方法用于太阳质子事件警报

尝试用神经网络方法作太阳质子事件警报。首先介绍太阳质子事件警报的发展,神经网络方法的特点,其次讨论选取预报因子的物理考虑,然后介绍试验结果,在两种试验情况下均可以有〉８５％的报准率。最后讨论改进警报的问题。     

太阳活动23周上升阶段质子事件的 预报分析

以２２周太阳活动低年（１９９３－１９９５）质子事件及其对应活动区的综合分析结果为判据,预报２３周太阳活动上升阶段的质子事件．从１９９７年１１月开始到１９９８年１２月,用该方法预报的质子事件共６个,报准３个,不确定一个,虚报１个,漏报１个（太阳背面产生的事件）．本文对用该方法预报的结果进行了分析讨论,并与世界警报中心的预报结果进行了比对,结果表明,该方法对于质子事件的短期预报是有效的．     

太阳活动23周上升阶段质子事件的预报分析

以２２周太阳活动低年（１９９３－１９９５）质子事件及其对应活动区的综合分析结果为判据,预报２３周太阳活动上升阶段的质子事件。从１９９７年１１月开始到１９９８年１２月,用该方法预报的质子事件共６个,报准３个,不确定一个,虚报１个,漏报１个（太阳背面产生的事件）。本对用该方法预报的结果进行了分析讨论,并与世界警报中心的预报结果进行了比对,结果表明,该方法对于质子事件的短期预报是有效的。     

太阳质子事件的性质和定量预报

本文介绍了太阳质子事件的性质、能谱特征以及国外质子事件定量预报的方法,并作了一定的评述。     

太阳射电爆发U形谱与质子事件能谱

本文根据第20太阳活动周内地面和空间观测的有关资料,按文[1]中探讨太阳射电Ⅳ型爆发U形谱产生机制的理论和模型,经分析统计,发现累积流量U形谱与质子事件能谱(幂律谱)的物理量之间存在着密切的相关关系,从而提出了一个完整而较有效的警报质子事件能谱的方法。     

太阳质子事件耀斑的短期预报

回顾产生太阳质子事件耀斑的短期预报,讨论短期预报在近期应做的研究．给出以下结论：（１）在６０年代和７０年代,质子事件耀斑的预报有相当大的进展;（２）新预报方法的探索和质子流在日冕与行星际的传播问题,是当前改进短期预报的关键;（３）对实际应用的短期预报工作的改进,可能需要从空间天气预报的角度,研究太阳活动区的分类．     

太阳质子事件预报研究进展

就太阳质子事件预报研究的重要意义,产生太阳质子事件的太阳活动区的一般特征,质子耀斑的辐射特征,质子事件几个重要参数预报方法简述了目前的研究进展。还给出了当前为满足用户需要改进预报应着重研究的方面。     

太阳中微子流变化的周期性与太阳活动

本文对1970年到1979年的太阳中微子流和太阳活动(太阳黑子、耀斑和质子事件)的数据作最大熵谱分析,并求其互相关函数和初相,得到:太阳中微子流和太阳活动均有11年的长周期;太阳中微子流和太阳质子事件还有共同的近3年、2年和1年左右的周期。太阳中微子流的3年周期占支配地位,质子事件中的3年周期亦占有重要地位。二级以上的耀斑事件亦有近两年的周期。它们的互相关函数和初相表明:太阳中微子流与太阳活动有正相关;对近11年周期的数据,太阳黑子和太阳耀斑相对于太阳中微子均有延迟时间47个月,对质子事件有延迟时间41个月。对于约3年的周期,质子的延迟时间为10个月。结合他人的太阳半径和太阳磁场的测量与分析结果,得到一个符合标准太阳模型中物理过程的太阳中微子流变化与太阳活动间的因果关系,并对这个因果关系的可能机制进行了讨论。     

太阳质子辐射和软X射线耀斑与射电爆发间的相关研究

统计分析了太阳质子事件与微波爆发和软Ｘ射线（ＳＸＲ）耀斑间的关系．结果表明：质子事件的峰值流量与微波爆发和ＳＸＲ耀斑的峰值流量、能通量间呈正的对数线性相关,相关系数０．７—０．８．根据这一统计结果和观测的微波爆发、ＳＸＲ耀斑的有关物理量,可以估算伴随的质子事件峰值流量．太阳质子辐射、ＳＸＲ耀斑和微波爆发三者间的共生关系,可以用磁环中耀斑产生的磁流体动力学过程来解释．大约３３％的质子事件没有对应的Ⅱ型爆发,这表明高能质子的加速有随机ＭＨＤ湍流加速（有Ⅱ型暴）和低频快磁声波湍动加速（无Ⅱ型暴,但有γ射线耀斑）２种不同的加速机制     

频率为8800MHz的射电爆发与质子事件强度的警报

本文考察了1967—1972年期间46个较大的0级以上的质子事件与所对应的8800MHz的太阳微波爆发之间的统计关系。 研究表明,将射电爆发按总持续时间T进行分类后,可以在不同的持续时间范围内分别找到同太阳质子流强度紧密相关的射电爆发参数(相关系数为0.83—0.91)。应用这些爆发参数得出的警报方案,预报质子流强度的误差范围为半级或一级(S—S分级法)时,报准率可达70％-94％,虚报率为52％。 此外,统计的结果还表明,Croom提出的平均持续时间T_M并不和质子流强度N_p相关,至少在此波长上不能用来预报质子流强度。     

On forecasting solar proton events using a ground-based neutron monitor

Results of applying the method developed for early warning about arrival of ∼10–100-MeV proton fluxes to the Earth after powerful eruptive events on the Sun, which uses the real time observable data received by the global network of ground-based neutron monitors (Mavromichalaki et al., 2009), are discussed. The retrospective analysis and comparison to the 2001–2006 observations indicate that more than 50% of solar proton events were omitted in such a forecasting method. For higher reliability, it is necessary to use additional data on the state of solar and heliospheric activity.     

Pulsations of type IV radio bursts as an indicator of protonarility of solar flares

The analysis of observational data has shown that the duration of a pulse train in type IV radio bursts decreases with increasing hardness of the spectrum of high-energy protons and increases with decreasing proton fluxes from the Sun. It is shown that such a correlation corresponds to a magnetohydrodynamic (MHD) model of pulsations and is inexplicacable within the framework of a nonlinear periodical regime of plasma instabilities. The pulse train duration is determined by proton pitch-angle diffusion caused by Alfvén waves in coronal magnetic loops. A method of predicting solar proton hardness and proton fluxes using type IV radio burst pulsations is proposed.     

Velocity-Space Proton Diffusion in the Solar Wind Turbulence

We study the velocity-space quasi-linear diffusion of the solar wind protons driven by oblique Alfvén turbulence at proton kinetic scales. Turbulent fluctuations at these scales possess the properties of kinetic Alfvén waves (KAWs) that are efficient in Cherenkov-resonant interactions. The proton diffusion proceeds via Cherenkov kicks and forms a quasi-linear plateau – the nonthermal proton tail in the velocity distribution function (VDF). The tails extend in velocity space along the mean magnetic field from 1 to (1.5?–?3) V _A, depending on the spectral break position, on the turbulence amplitude at the spectral break, and on the spectral slope after the break. The most favorable conditions for the tail generation occur in the regions where the proton thermal and Alfvén velocities are about equal, V _Tp/V _A≈1. The estimated formation times are within 1?–?2 h for typical tails at 1 AU, which is much shorter than the solar wind expansion time. Our results suggest that the nonthermal proton tails, observed in situ at all heliocentric distances >?0.3 AU, are formed locally in the solar wind by the KAW turbulence. We also suggest that the bump-on-tail features – proton beams, often seen in the proton VDFs, can be formed at a later evolutional stage of the nonthermal tails by the time-of-flight effects.     

Short-term Forecast of Solar Proton Events With Characteristic Phyiscal Quantities of Photospheric Magnetic Fields

Via the three physical quantities (i.e., the maximal horizontal gradient of longitudinal magnetic field |Δ_hB_z|_m, the length of neutral line with a large gradient L, and the number of isolated singular points η), which are used to represent the characteristics of the complexity and non-potentiality of the photospheric magnetic fields in solar active regions, a model of the shortterm forecast of proton events is built. The effectivity of the short-term forecast of proton events by means of the characteristic physical quantities of magnetic fields is verified. In the nowadays commonly used models of short-term forecast of solar proton events, until present the characteristic physical quantituieas of magnetic fields are not formally taken to be the factors of forecast. Because the solar proton events are low probability events, the physical mechanism of their occurrence is still not well understood. In the models of their prediction, the problems of high rates of false alarm or low rates of right alarm often exist. In this paper the traditional factors used in the existing models of forecast of proton events and the characteristic physical quantities of magnetic fields are combined together. By using the method of neural network, a more effective method of the short-term prediction of proton events is established. With the 1871 sample data in 1997-2001, we have set up Model A with the traditional forecast factors as the input layer, and also Model B with the traditional forecast factors plus the characteristic physical quantities of magnetic fields as the input layer. Via the set of 973 sample data of the years 2002 and 2003, we have carried out a simulative forecast, and found that under the condition that these two models possess the same rate of accuracy in the forecast of proton events, the rate of false alarm of Model B becomes evidently lower. This has further verified the effectiveness of the characteristic physical quantities of magnetic fields in shortterm prediction. Furthermore, this may improve the actual ability of forecast of solar proton events.     

Enhanced emission of Alfvén waves from sunspots during proton flares

Thomas (1978) has shown that, if Alfvén waves exist in a sunspot umbra, they are normally reflected so strongly by the temperature minimum as to be essentially undetectable in the upper solar atmosphere. However, it is known that in many proton flares, chromospheric emission overlies the umbra of a sunspot, indicating that the transition region (TR) between chromosphere and corona in the umbral flux tube has moved down to lower altitudes. As a result of this lowering, umbral Alfvén waves have readier access to the corona: the coronal leakage depends exponentially on the altitude of the TR. We find that the Alfvén wave flux which leaks out of the umbra into the corona can exceed 10⁷ ergs cm^-2 s^-1. A flux of this magnitude is expected to dissipate rapidly in the corona, thereby contributing to a positive feedback loop which ensures prolonged (1 hr) leakage of the umbral Alfvén waves into the corona. We propose that these Alfvén waves may contribute significantly to prolonged energization of proton flares in which umbral coverage occurs.     

Real-time abnormal light curve detection based on a Gated Recurrent Unit network

Targeting the problem of high real-time requirements in astronomical data processing, this paper proposes a real-time early warning model for light curves based on a Gated Recurrent Unit(GRU) network.Using the memory function of the GRU network, a prediction model of the light curve is established, and the model is trained using the collected light curve data, so that the model can predict a star magnitude value for the next moment based on historical star magnitude data. In this paper,we calculate the difference between the model prediction value and the actual observation value and set a threshold. If the difference exceeds the set threshold, the observation value at the next moment is considered to be an abnormal value,and a warning is given. Astronomers can carry out further certification based on the early warning and in combination with other means of observation. The method proposed in this paper can be applied to real-time observations in time domain astronomy.     

A Study of the Frequency of Occurrence of Large-Fluence Solar Proton Events and the Strength of the Interplanetary Magnetic Field

It has been shown previously that the number of very-large-fluence solar proton events inferred for the period since 1561 were more frequent at times of low solar activity (e.g., following the recovery from the Maunder minimum), than in the present epoch of high solar activity. An inverse dependence is demonstrated between the probability of observation of the very large-fluence solar proton events and the strength of the interplanetary magnetic field derived from empirical predictions. Using the observed dependence, it is predicted and demonstrated that large-fluence solar proton events have been observed at Earth more frequently near the recurrent minima of the solar activity cycle in the past than during the present epoch. We show that these results are explicable in terms of the linear dependence of the Alfvén velocity upon the strength of the interplanetary magnetic field, leading to higher shock compression ratios in the past. These results indicate that this aspect of “solar weather” will be significantly influenced by the prevailing strength of the interplanetary magnetic field, and that recurrence of solar conditions similar to those of the solar activity minimum of solar cycles 12–14 (1878.9–1913.6) would be accompanied by a factor of ∼4 increase in the occurrence of large-fluence solar proton events.     

Modification of Proton Velocity Distributions by Alfvénic Turbulence in the Solar Wind

In the present paper, the proton velocity distribution function (VDF) in the solar wind is determined by numerically solving the kinetic evolution equation. We compare the results obtained when considering the effects of external forces and Coulomb collisions with those obtained by adding effects of Alfvén wave turbulence. We use Fokker–Planck diffusion terms to calculate the Alfvénic turbulence, which take into account observed turbulence spectra and kinetic effects of the finite proton gyroradius. Assuming a displaced Maxwellian for the proton VDF at the simulation boundary at 14 solar radii, we show that the turbulence leads to a fast (within several solar radii) development of the anti-sunward tail in the proton VDF. Our results provide a natural explanation for the nonthermal tails in the proton VDFs, which are often observed in-situ in the solar wind beyond 0.3 AU.