Statistical properties of H$\alpha$ jets in the polar coronal hole and their implications in coronal activities

The chromosphere is dynamic and highly-structured, which links the photosphere and the corona. In the photosphere, the magnetic flux tubes build the magnetic network, which expands upward and appears as bright patches in the chromosphere, namely, chromospheric network. Therefore, the network boundaries in the chromosphere spatially coincide with the magnetic network concentrations in the photosphere. The chromospheric network boundaries are abundant with jet-like, fine-scale features, which are referred as spicules at the limb, mottles in the quiet sun and fibrils in active regions (e.g., Beckers 1968; Rouppe van der Voort et al. 2007; De Pontieu et al. 2007a, etc.). Spicules have been studied for more than a century and are still one of the most popular topics at the present days (see e.g., Beckers 1968; Sterling 2000; Tsiropoula et al. 2012, for extensive overviews)

Thanks to high spatio-temporal resolution observations, De Pontieu et al. (2007c) pointed out that there are two different types of spicules ( Type-I $\&$ Type-II ). The typical characteristic of Type-I spicules ( also known as traditional spicules ) is that they move up and down with the ascending speeds in the range of $15-40$ km s^-1. The Type-I spicules have lifetimes of $150-400$ s and heights of $4-8$ Mm, and they are believed to result from magnetoacoustic shocks. In contrast, Type-II spicules rise with much higher speeds of $\sim 100$ km s^-1and disappear rapidly with shorter lifetimes of $50-150$ s. Type II spicules are suggested to be driven by magnetic reconnections. De Pontieu et al. (2007d) found that the Type-II spicules also undergo sideways swaying motion which is signature of Alfvén waves permeating with enough energy to heat the corona and accelerate the solar wind. With high-quality observations, the Type-II spicules have been found to combine three different types of motions: field-aligned flows, swaying motions and torsional motions, of which swaying and torsional motions are considered as a sign that spicules propagate along their axis with Alfvén waves at high speeds ranging from 100 – 300 km s^-1. With multi-wavelength observations, Type-II spicules often appeared to heat to the transition region temperatures, even up to the coronal temperatures (e.g., De Pontieu et al. 2011, 2014; Rouppe van der Voort et al. 2015). In the advanced 2.5D radiative MHD simulations, Martínez-Sykora et al. (2017) reproduced many observed properties of type II spicules with $\sim 100$ km s^-1speeds and $\sim 10$ Mm heights, and also revealed that spicules could heat plasma to the corona by dissipation of current which is due to ambipolar diffusion. The further numerical study showed that the dissipation through ambipolar diffusion of electrical currents could also explain why the observed spicules could reach a speed exceeding 100 km s^-1and heat to coronal temperatures (De Pontieu et al. 2017).

In active region plage regions, Hansteen et al. (2006) and De Pontieu et al. (2007a) analyzed H$\alpha$ observations and found that fibrils appear as short, dynamic and spicule-like features which have lifetimes of $3-6$ minutes and show quasi-periodicity. These dynamic fibrils undergo ascent and descent motions with velocities of $10-35$ km s^-1 at the initial phase, which are similar to that of Type-I spicules. Moreover, the numerical simulations demonstrated that these fibril-like features also show wave-behaviors. That could be formed when flows and oscillations leak from magnetic flux concentrations into chromosphere (e.g., Hansteen et al. 2006; De Pontieu et al. 2007a; Heggland et al. 2007, 2011).

In quiet-Sun regions, mottles show as dark and spicule-like features against the disk in the wings of H$\alpha$. Similar to fibrils, the motions of some mottles also follow a parabolic path and they can be driven by shock waves (e.g., Rouppe van der Voort et al. 2007; De Pontieu et al. 2007b).

With the spectroscopic data, Langangen et al. (2008) firstly used Ca ii $854.2$ nm line to investigate the rapid blueshifted excursions (RBEs) on disk, which are a sudden widening of line profile in the blue wing without an associated redshift. Combined with SJ images of H$\alpha$, these RBEs show as narrow streaks emanating from network (Rouppe van der Voort et al. 2009) . The RBEs show velocities of $15-20$ km s^-1and a mean lifetime of 45 s. There are $\sim 10^{5}$ RBEs on the surface of disk at any moment. Recently, Sekse et al. (2012, 2013) found that RBEs also undergo three kinds of motions like spicules. Owing to the similarity, RBEs are interpreted as the on-disk counterparts of Type-II spicules and suggested to result from magnetic reconnection (Rouppe van der Voort et al. 2009).

Jet-like features are not only seen in the chromosphere, but also in the transition region and corona. In the transition region, network jets are small-scale structures, having lifetimes of $20-80$ s and speeds of $80-250$ km s^-1. A number of them appear as “inverse-Y” shape, which should be driven by magnetic reconnection (Tian et al. 2014). In the corona, coronal jets appear as transient collimated beams. More findings suggest that coronal jets are driven by magnetic reconnection and may be the source of mass and energy input to the upper solar atmosphere and the solar wind (see Raouafi et al. 2016, and the references therein). Plumes belong to another type of collimated structures, which are relatively large and stable.  (see Wilhelm et al. 2011, and the references therein) and might supply plasma and energy for solar wind  (e.g., Tian et al. 2010, 2011; Fu et al. 2014; Liu et al. 2015; Huang et al. 2021).

There are PDs in coronal plumes with 10% - 20% intensity variations, periods of 10 - 30 minutes and speeds of 75 - 170 km s^-1 (e.g., DeForest & Gurman 1998; Krishna Prasad et al. 2011). Furthermore,  Tian et al. (2011) discovered that high-speed quasi-periodic outflows could be found in both plumes and inter-plume regions, and their speeds are similar to that of PDs. These outflows were suggested to be responsible for the generation of PDs (McIntosh et al. 2010). By following the evolution of spicules, Jiao et al. (2015) and Pant et al. (2015) concluded that the spicules can trigger PDs.

In the base of coronal plumes, Raouafi et al. (2008) found that coronal jets could contribute to rise and change brightness in the pre-existed plumes. Given that those coronal jets are rooted in the chromospheric network and are mainly triggered by magnetic reconnection, coronal plumes could also be powered by magnetic reconnection  (for an extensive overview, see Raouafi et al. 2016). Furthermore,  Raouafi & Stenborg (2014) and Panesar et al. (2018) found that a great number of jets and transient bright points frequently occurred in the bases of coronal plumes and these features could be the main energy source for coronal plumes. In fact, the scenario that coronal plumes are driven by the magnetic reconnection between the unipolarity magnetic features and nearby small-scale bipolar has been proposed much earlier (e.g., Wang & Sheeley 1995; Wang et al. 1997; Wang & Muglach 2008).

Since there are many kinds of jet-like features rooted in network regions, it is natural to ask whether there is any connection among them or not. Aiming at this question, Qi et al. (2019, hereafter Paper I) developed an automatic method to identify and track network jets. They statistically analyzed a set of transition region network jets in the same network region, and they found that network jets in the root of coronal plumes have lifetime, height and speed averaging at 45.6 s, 8.1${}^{{}^{\prime\prime}}$ and 131 km s^-1, significantly more dynamic than those in the region without discernible plume, which have longer lifetime of 50.2 s, smaller height of 5.5${}^{{}^{\prime\prime}}$ and lower speed of 89 km s^-1 in average. Paper I suggested that only more energetic network jets (likely with speeds more than 100 km s^-1) can feed plasma to the corona. In the present study, we further investigate the possible relationship between chromospheric jet-like features (e.g., spicules, fibrils, mottles or RBEs) and coronal activities. To achieve this, we analyzed data obtained with the ground-based 1-m NVST (Liu et al. 2014; Xiang et al. 2016) installed at the Fuxian Lake Solar Observing Stations operated by Yunnan Astronomical Observatories in China and the space-borne AIA (Lemen et al. 2012) and the Helioseismic and Magnetic Imager (HMI, Schou et al. 2012) aboard the Solar Dynamics Observatory (SDO, Pesnell et al. 2012).

The paper is organized as follows. The observations and methodology are described in Sect. 2. The results are shown in Sect. 3. The discussion and conclusions are given in Sect. 4 and in Sect. 5, respectively.

The data set analyzed in this study was taken on 2018 September 15 targeting on a polar coronal hole. The observations include filtergrams taken by NVST at the H$\alpha$ –0.6 Å with a bandpass of 0.25 Å, EUV images taken at 171 Å and 193 Å passbands by AIA and line-of-sight magnetograms by HMI.

The NVST data were taken from 09:29 UT to 10:03 UT including a series of H$\alpha$ images with a cadence of 6 s and a spatial scale of 0.136 arcsec/pixel. The field-of-view covers a 144${}^{{}^{\prime\prime}}$ $\times$ 144${}^{{}^{\prime\prime}}$ region. The stabilization of the H$\alpha$ image series is performed with a fast sub-pixel image registration algorithm by the instrument team (Yang et al. 2015).

The AIA and HMI data were downloaded from JSOC. The cadences of the AIA 171 Å data and HMI line-of-sight magnetograms are 12 s and 45 s, respectively. The spatial resolutions of the AIA and HMI data are 1.2${}^{{}^{\prime\prime}}$. The AIA and HMI data are prepared with standard procedures provided by the instrument teams, and the level 1.5 data are analyzed.

The H$\alpha$ images have been transposed and rotated clockwise by 90 degrees to fit into the Helioprojective-Cartesian coordinate system as used in AIA and HMI data. Bad images in H$\alpha$ data have been manually removed and replaced by artificial ones obtained from interpolation of nearby good frames. The network lanes that present as magnetic concentrations on HMI magnetograms and clusters of bright dots on H$\alpha$ images are used to align the corresponding images. We make use of AIA 1600 Å as reference to align the H$\alpha$ images and 171 Å images. And we manually check the alignment between 1600 Å and 171 Å using referent features, despite images from different passbands of AIA have been aligned with each other by the data processing pipeline.

In Fig. 1, we show the studied regions seen with AIA, HMI and H$\alpha$ –0.6 Å around 09:28 UT. The region is part of the polar coronal hole (see Fig. 1(a)). In Fig. 1(b), even though the region is part of the polar region, we can see typical magnetic structures of network regions (e.g., Xia et al. 2003, 2004). The network regions also could be clearly seen in NVST H$\alpha$ –0.6 Å images (see the bright lanes in Fig. 1(d)). We could clearly see lots of jet-like features rooting in the network lanes in the H$\alpha$ –0.6 Å images. Given that this region is close to the north polar, these jets exhibit upward motions from the H$\alpha$ off-bands filter, so we defined these jet-like features as the H$\alpha$ jets in this paper. In the coronal images (AIA 171 Å), we also see plume-like and/or jet-like features rooted in coronal bright points at some locations in the coronal hole (Figs. 1(a)&(e)). Figure 1(c) shows PDs with a simple bandpass filter method on the AIA 171 Å images.

In H$\alpha$ images, jets appear as dark features. In Fig. 1 and the associated animation, we can see that H$\alpha$ jets are abundant in the network regions. To identify these jets, we upgrade the automatic algorithm that was described in Paper I and some key definitions are given in Huang et al. (2017). Note that “ local peaks” as defined in Huang et al. (2017) are replaced by “local trough”, or in other words, multiply the H$\alpha$ intensity by –1 to identify “local peaks”. In practical operations, we trace them from their bases to tops to obtain their full trajectories and then deduce their lifetimes and speeds.

In Paper I, we studied the relation between transition region network jets and coronal plumes. In the present work, we focus on chromospheric jet-like features observed in H$\alpha$ –0.6 Å, namely H$\alpha$ jets. In agreement with Paper I, we found that the network regions where the coronal counterparts are active with coronal bright points and plumes tend to produce H$\alpha$ jets with more dynamic characteristics (i.e., larger heights and faster speeds) than those from the other network regions. Given that H$\alpha$ jets, network jets and coronal plumes may be associated with magnetic reconnection, it is possible that the energy produced by magnetic activities could transfer from fine structures in the low solar atmosphere to those in the corona.

In this paper, we find a few examples that H$\alpha$ jets have a one-to-one correspondence with PDs in spatial and temporal. The H$\alpha$ jets associated with PDs are usually faster and higher than the others. What’s more, Wang et al. (1998) found that He ii 304 Å jets are always associated with H$\alpha$ jets, but not all the H$\alpha$ jets are associated with He ii 304 Å jets. We find that the faster and higher H$\alpha$ jets have a certain relationship with coronal jets. Perhaps, only parts of H$\alpha$ jets which are more energetic can reach the low corona and also be seen in He ii 304 Å.

Why the H$\alpha$ jets are higher and faster at the roots of coronal plumes and jets? We will discuss the following possible mechanisms.

The first interpretation is associated with the generation of transverse wave induced KHi in H$\alpha$ jets. It has been proved that chromospheric jets usually accompany transverse motions with amplitudes up to $10-25$ km s^-1 (De Pontieu et al. 2007d), which are generally interpreted as kink waves  (He et al. 2009). Such waves are possible to induce the KHi due to velocity shear between magnetic structures and background medium (e.g., Terradas et al. 2008; Antolin et al. 2018, etc.). Antolin et al. (2018) considered the KHi induced by kink motions and attributed some observational properties of spicules to the development of KHi eddies. The small spatial structures induced by the KHi can help the dissipation of wave energy, and thus playing an important role in an understanding temperature increase of magnetic structures (e.g., Guo et al. 2019; Karampelas et al. 2019). In addition, Lau & Liu (1980) pointed out that if a longitudinal flow exceeds a criterion, the magnetized plasma is unstable. With the IRIS observations, Li et al. (2018b) found that the KHi could develop due to the strong shear (more than 204 km s^-1) between two blowout jets. In the case of kink waves with longitudinal flows, the criterion for developing the KHi has been discussed in many literatures, (e.g., Andries & Goossens 2001; Zhelyazkov et al. 2016; Li et al. 2018b). The threshold was derived by Andries & Goossens (2001) from eigen-mode analysis in a cylindrical model. It reads $U=V_{\rm Ai}(1+B_{\rm e}/B_{\rm i}\sqrt{\rho_{\rm i}/\rho_{\rm e}})$, where $U$ ($V_{\rm Ai}$) represents the flow (Alfvén) speed inside a jet, $B_{\rm i}$($\rho_{\rm i}$) and $B_{e}$($\rho_{\rm e}$) are the magnetic field strength (density) in a jet and the ambient region, respectively. In spicules, the density and magnetic field of a chromospheric jet is about 3 $\times 10^{10}{\rm cm}^{-3}$ and 40 G, respectively. Assume those for the ambient regions are 1 $\times 10^{9}{\rm cm}^{-3}$ and 10 G (see Tsiropoula et al. 2012; Sterling 2000; Centeno et al. 2010, and references therein). Generally, the Alfvén speed in the choromosphere is 40 km s^-1(De Pontieu et al. 2007d), indicating that the KHi can be induced in spicules when the upflow speed exceeds 98 km s^-1. Furthermore, Zhelyazkov et al. (2016) studied the propagating of kink waves in EUV jets in vertical magnetic tubes. They found that kink waves could become unstable when flow velocity exceeds 112 km s^-1. In Fig. 4, we can see that 28.4% H$\alpha$ jets have higher speed (more than 110 km s^-1) in R1–R5, while only 8.3% H$\alpha$ jets with such high speed in other regions. This means that in aforementioned observations, the KHi is possible to develop in faster H$\alpha$ jets in R1–R5. Due to the development of the KHi, the faster H$\alpha$ jets are possible to dissipate energy into higher atmospheres. Moreover, Cavus & Kazkapan (2013) proposed that the KHi might be a mechanism for the formation of magnetic reconnection. This means that the H$\alpha$ jets with KHi might be able to heat the corona. However, more detailed simulations should be considered in future studies.

Another possible interpretation is the interaction strength among small-scaled magnetic elements in the network region. For R1-R5, there are dominant network fields and some small-scale opposite-polarity fields in Fig. 1(b). Wang et al. (2016) proposed that the convergence of the base flux could trigger coronal plumes. The interaction of magnetic flux concentrations with constantly weak fields could result in strong magnetic tension. The strong magnetic tension which is amplified and transported upward through ambipolar diffusion could produce numerical faster spicules (more than 100 km s^-1) in 2.5D simulation (Martínez-Sykora et al. 2017). In addition, same as the 2.5D numerical simulation, Ding et al. (2011) also showed that the stronger magnetic field could produce higher up-flows velocities. These simulations mean that there are a great possibility of rapid jets in network field with weak opposite magnetic elements, and these jets might have coronal connection. Using high-spatial-resolution and high-time-cadence observations of the Goode Solar Telescope (GST, Goode et al. 2010; Cao et al. 2010),  Samanta et al. (2019) provided observations from the photosphere to the chromosphere and the corona and supported that these chromospheric jets ($\sim$50 km s^-1) are driven by the interaction between strong network fields with small-scale opposite-polarity fields, of which have a good coronal connection. By performed order-of-magnitude estimation of energies, the authors estimated that these spicules supply enough energy into the corona. These speeds of jets are smaller than that in our results because the FOV in  Samanta et al. (2019) is close to the center of disk. With respect to the relation between the chromospheric jets and coronal activities, the mixed-polarity magnetic fields could result in shuffling and further braiding of small-scale magnetic field lines, which produce magnetic reconnection at braiding boundaries and power the corona. By the same,  Li et al. (2018a) proposed that the dissipation of electric currents due to the effect of small-scale weak magnetic activities (included spicules) could effectively heat corona at low heights. In addition, Priest et al. (2002) showed that the formation and dissipation of current sheets along these separatrices resulting from highly fragmented photospheric magnetic configurations could be an important contribution to coronal heating. This means that the activities of mixed-polarity on small scales might have enough power to coronal activities. In Fig. 1(d), we see that H$\alpha$ jets locate near the dominant network fields, but the lower resolution of magnetic fields measurements affect the detection of weak magnetic elements at the footpoints of H$\alpha$ jets. Further studies using high spatial-temporal resolution observations of the photosphere (such as the Goode Solar Telescope (GST), the forthcoming Daniel K. Inouye Solar Telescope (DKIST) and the planned Chinese Advanced Solar Observatories–Ground-based (ASO-G)) should shed more light on this topic.

Our analyses here have shown that H$\alpha$ jets with coronal responses tend to be higher and faster. Whether there are critical height and speed for a H$\alpha$ jet having coronal response should be investigated both theoretically and observationally. As discussed above, it may have different answers based on different mentioned mechanisms. Based on the present observations, we cannot find evidence of KHi due to the spatial and temporal resolutions. HMI magnetograms show clearly that a number of small-scale magnetic elements appear and disappear in the footpoints of these H$\alpha$ jets (see the animation associated with Fig. 1). Although the observed region is pretty much closed to the limb, the observations still give a hint of a degree of magnetic dynamics. Therefore, with the present observations, we are more favorable with the scenario of that more powerful interaction between small-scale magnetic elements tends to generate more dynamic phenomena. However, the KHi scenario cannot be ruled out within the present circumstance. In the near future, we will study in-depth how much initial energy of chromospheric jets is required to have coronal responses based on the above-mentioned mechanism.

In this paper, we analyzed observations taken by the 1-m New Vacuum Solar Telescope at a passband H$\alpha$ –0.6 Å. Thanks to the higher spatial and temporal resolution of the data, we were able to trace the evolutions of H$\alpha$ jets over time. Using an upgraded automated detection algorithm, we carried out a statistical study of H$\alpha$ jets. Using the algorithm, we identified and traced 1 320 H$\alpha$ jets in the data set. We found that the average lifetime, height and ascending speed of the H$\alpha$ jets are 75.38 s, 2.67 Mm, 65.60 km s^-1, respectively. These characteristics of the events agree well with those of Type-II spicules. In addition, we found that the H$\alpha$ jets occur in network regions where magnetic concentrations are present. We believe that the H$\alpha$ jets are corresponding to spicules or the on-disk counterparts of spicules.

We found that H$\alpha$ jets rooted in the footpoint regions of coronal plumes have an average lifetime of 76.1$\pm$35.0 s, an average height of 3.01$\pm$1.07 Mm and an average ascending velocity of 81.2$\pm$48.3 km s^-1. While the H$\alpha$ jets in the rest regions are having an average lifetime of 75.0$\pm$32.0 s, an average height of 2.48$\pm$0.95 Mm and an average ascending velocity of 56.9$\pm$36.3 km s^-1, these results show that the H$\alpha$ jets associated with coronal plumes are on average higher and more dynamic than those not with coronal plumes. Can these jets possibly provide energy to sustain a coronal plume? In corona, radiation cooling time is expressed as $\tau_{rad}(n_{0},T_{0})=\frac{9}{5}\frac{k_{B}T^{5/3}}{n\Lambda_{0}}$ in the corona (Aschwanden & Terradas 2008), taking in typical parameters of corona and plumes, where the radiative loss rate $\Lambda_{0}\sim 10^{-17.73}ergs/cm^{3}$, temperature T$\sim$1.0 MK and density n$\sim 10^{9}cm^{-3}$, then we found the radiation cooling time scale is in the order of 1000 s. While the H$\alpha$ jets in the footpoint of a plume studied here present repeatedly in a time scale of about 80 s, much less than the radiation cooling time in the corona, it suggests that H$\alpha$ jets may provide energy continuously to sustain coronal plumes.

To determine exactly what type of H$\alpha$ jets might link to coronal activities, we further investigated the connection between H$\alpha$ jets and the PDs in plumes. We found that there are always H$\alpha$ jets appearing before the initiation of PDs and extending in the same directions of the PDs. Almost all these jets (28 out of 29) have speeds $\gtrsim$50 km s^-1, thus we suggest that only those H$\alpha$ jets exceeding a certain velocity can be precursors of PDs.

A fade coronal jet with a speed of about 160 km s^-1 was also observed. We found that this coronal jet was accompanied by a H$\alpha$ jet that has a speed comparable to that of the coronal jet, suggesting that the coronal jet might be the extension of the part of H$\alpha$ jets.

In agreement with Paper I, present studies confirmed that small-scaled jet-like features in the solar lower atmosphere can directly connect to coronal activities. Given that most H$\alpha$ jets observed here are likely spicules in the chromosphere, we suggest that H$\alpha$ jets, if dynamic to a certain degree, can sufficiently power the corona. Combined the results shown in Paper I, we suggest that spicule-like features in the chromosphere, transition region network jets and ray-like features in the corona are coherent phenomena, and they are important tunnels for cycling energy and mass in the solar atmosphere.

Acknowledgments: We are grateful to the anonymous reviewer for the constructive comments and suggestions. L.X. and Y.Q. are supported by National Natural Science Foundation of China (NSFC) under contract No. 41974201. Z.H. thanks financial supports from NSFC under contract No. U1831112 and the Young Scholar Program of Shandong University, Weihai (2017WHWLJH07). H.F. is supported by NSFC contract No. U1931105. W.L. is supported by NSFC contract No. U1931122. M.S. is supported by NSFC contract No. 41627806. We would like to thank the NVST operation team for their hospitality during the observing campaign and preparation of the data. The data used are also courtesy of NASA/SDO, the AIA and HMI teams and JSOC.