Heating at the remote footpoints as a brake on jet flows along loops in the solar atmosphere

We report on observations of a solar jet propagating along coronal loops taken by the Solar Dynamics Observatory (SDO), the Interface Region Imaging Spectragraph (IRIS) and 1-m New Vacuum Solar Telescope (NVST). The ejecta of the jet consist of multi-thermal components and propagate with a speed greater than 100 km/s. Brightenings are found in the remote footpoints of the coronal loops having compact and round-shape in the Halpha images. The emission peak of the remote brightening in the Atmospheric Imaging Assembly (AIA) 94 \AA passband lags 60 s behind that in the jet base. The brightenings in the remote footpoints are believed to be consequences of heating by nonthermal electrons, MHD waves and/or conduction front generated by the magnetic reconnection processes of the jet. The heating in the remote footpoints leads to extension of the brightening along the loops toward the jet base, which is believed to be the chromospheric evaporation. This apparently acts as a brake on the ejecta, leading to a deceleration in the range from 1.5 to 3 km s$^{-2}$ with an error of $\sim1.0$\,km s$^{-2}$ when the chromospheric evaporation and the ejecta meet at locations near the loop apexes. The dynamics of this jet allows a unique opportunity to diagnose the chromospheric evaporation from the remote footpoints, from which we deduce a velocity in the range of 330--880 km/s.


INTRODUCTION
Solar jets are ejections of solar plasma in a manner that are forced out of a small opening compared to their lengths. They are ubiquitous in the solar atmosphere, such as chromosphere spicules (e.g. Sterling 2000;De Pontieu et al. 2007;Zhang et al. 2012;Pereira et al. 2012;Samanta et al. 2019, etc.), transition region network jets (e.g. Tian et al. 2014;Narang et al. 2016;Qi et al. 2019, etc.) and coronal jets (see a review by Raouafi et al. 2016). These jets are not only one of the basic components of the solar atmosphere, but also play a crucial role in mass and energy transformation and transportation in the solar atmosphere. For example, spicules and network jets are believed to play an important role in heating the corona and balancing the mass loss by solar wind (see e.g. De Pontieu et al. 2011;Tian et al. 2014).
In particular, coronal jets have been studied intensively since they were first identified in the early 1990s by the soft X-ray Telescope onboard Yohkoh (Shibata et al. 1992). In Yohkoh soft X-ray observations, coronal jets normally have lengths ranging from a few 10 4 km to 4×10 5 km, widths from 5 × 10 3 km to 10 5 km and lifetimes from a few minutes to a few hours (Shimojo et al. 1996). The kinetic energy of the coronal jets is estimated to be 10 25 -10 28 ergs (Shibata et al. 1992). The observations taken by subsequent missions, such as SOHO, TRACE, Hinode, STEREO and SDO, have revealed that coronal jets can also be seen in the EUV passbands and they contain both hot and cool components (see Raouafi et al. 2016, and the references therein).
Coronal jets are believed to be consequences of magnetic reconnection (Yokoyama & Shibata 1995;Wyper et al. 2017). Many studies have focused on observations of the evidence and the dynamic processes of magnetic reconnection in coronal jets. Many coronal jets tend to have an inverted-Y shape (see e.g. Shibata et al. 1992;Madjarska 2011), suggesting magnetic reconnection between small loops and (quasi-) open field lines (Yokoyama & Shibata 1995). Moreover, magnetic cancellations, which are one of the signatures of magnetic reconnection (e.g. Priest et al. 2018;Syntelis et al. 2019;Syntelis & Priest 2020), have also been found to be closely associated with the occurrences of coronal jets (see e.g. Huang et al. 2012;Panesar et al. 2017;McGlasson et al. 2019, etc.). In some coronal jets, magnetic reconnections are also evidenced by the presence of blobs in current sheets (Zhang & Ji 2014;Zhang et al. 2016a;Zhang & Ni 2019). Recent studies reveal that the magnetic reconnection in coronal jets is triggered by mini-filament eruptions (e.g. Sterling et al. 2015Sterling et al. , 2016Hong et al. 2016;Huang et al. 2018;Moore et al. 2018;Sterling et al. 2019;Hong et al. 2019;Shen et al. 2019;Yang et al. 2019, etc.). This suggests that coronal jets involve physical processes that couple the solar atmosphere from the chromosphere to the corona.
In most coronal jets, part of the ejecta tend to fall back toward the solar surface. In these cases, the plasma flows may have a parabolic trajectory (see examples in Lu et al. 2019, though the falling back motions are not their topics). Here, we report on observations of a solar jet flowing along closed loops, in which the deceleration of the ejecta was associated with heating in the remote footpoints of the loops. In what follows, we give the description of the data in Section 2, the data analysis and results in Section 3 and the discussion and conclusions in Section 4.

DATA DESCRIPTION
The observations analysed here were taken by the Atmospheric Imaging Assembly (AIA, Lemen et al. 2012)  The AIA data used in this study include images taken in the EUV passbands around 94 Å, 131 Å, 171 Å, 193 Å, 211 Å and 335 Å. The temperature response functions of these passbands peaks at 6.3×10 6 K, 1.0×10 7 K, 6.3×10 5 K, 1.3×10 6 K, 2.0 × 10 6 K and 2.5 × 10 6 K, respectively (Lemen et al. 2012) The cadences of the images of both the passbands are 12 s and the pixel sizes are 0.6 ′′ ×0.6 ′′ . The line-of-sight magnetograms taken by HMI are used, which have a cadence of 45 s and a pixel size of 0.6 ′′ ×0.6 ′′ . Both AIA and HMI data have been reduced by the standard procedure of aia prep.pro in the solarsoft.
The IRIS spectrograph was running in a "large coarse 8step raster" mode with a 2 ′′ step size and 8 s exposure time. It obtained 80 rasters of the spectral data in total and the cadence between each raster is 72 s. The spectrograph slit has 0.35 ′′ width and the pixel size along the slit axis is 0.33 ′′ because the data have been binned by a factor of 2 aboard before being transferred to the ground. The IRIS slit-jaw (SJ) imager took observations in the passbands of 1330 Å and 2796 Å with a cadence of 18 s and a pixel size of 0.17 ′′ ×0.17 ′′ . We use the IRIS level 2 data that have been calibrated by the operation team and are suitable for science analyses. To do the radiometric calibration, we follow the procedures given in the IRIS technical note 24.
The analysed data also include observations from the NVST, which were taken in Hα line centre with a bandpass of 0.25 Å. The Hα data have a pixel size of 0.17 ′′ ×0.17 ′′ . The data have been reduced and reconstructed using Speckle technics (see details in Xiang et al. 2016). The reconstructed data have a cadence of about 12 s.
The data taken from different instruments are aligned by using the images of passbands having close representative temperatures. For this purpose, the images taken with the AIA 1600 Å and 304 Å passbands have also been used. Moreover, the calibrated full-disc microwave flux (intensity) at 17 GHz and 35 GHz taken by the Nobeyama Radio Polarimeters (NoRP, Nakajima et al. 1985) are also used to determine the variations of the nonthermal radiations. The NoRP data have a cadence of 1 s.

DATA ANALYSES AND RESULTS
We first derive the emission measures (EM) of the jet based on the observations of the AIA 94 Å, 131 Å, 171 Å, 193 Å, 211 Å and 335 Å passbands with the differential emission measure (DEM) code developed and optimized by Cheung et al. (2015) and Su et al. (2018). To reduce noise while constructing an EM image of the region, we have binned the AIA data by 2 pixels×2 pixels, which leaves a spatial sampling of 1.2 ′′ ×1.2 ′′ for the constructed EM images. The evolution of the region where the jet took place as seen in IRIS SJ 1330 Å, the NVST Hα, the AIA 171 Å and the AIA 94 Å images and the constructed emission measures at a few ranges of temperatures are shown in Figure 1 and the associated animation.
The solar jet initiates at around 06:35 UT (see its base region denoted by the red arrow in Figure 1d). The system seems to be initially destabilised by a GOES B9.7 flare (see its main ribbons as the bright region at coordinates of X=−560 ∼ −520 ′′ and Y=−220 ∼ −180 ′′ in Figure 1, and the flare started from 06:31:20 UT, peaked at 06:37:14 UT and ended at 06:48:16 UT as given in the RHESSI flare list 1 ). In the base region of the jet, we can see the eruption of a dark filament-like feature in AIA 171 Å and Hα images (see the animation associated with Figure 1b&c). In the AIA 94 Å images, we can see that multiple coronal loops are also rooted in the base of the jet (see Figure 1d and one example is denoted by yellow dotted line that is about 70 ′′ in length). Hereafter, these loops are termed "jet loops", and their footpoints other than the base of the jet are termed "remote footpoints". The HMI magnetogram reveals a magnetic structure with a patch   Figure 1. The image is artificially saturated at −800 Mx cm −2 (black) and 800 Mx cm −2 (white). The contours in red mark the places around the remote footpoints where the HMI line-of-sight magnetic flux density is −500 Mx cm −2 . The white dotted lines enclose the jet base region and remote footpoint region as shown in Figure 1d. The yellow dotted line outlines the path of the jet loops as shown in Figure 1d.  Figure 2). The coronal loops seen in the AIA 94 Å channel connect the base of the jet and the negative polarities spread in the regions toward the east (see the red contours shown in Figure 1d and Figure 2).
The jet activity begins with appearance of two bright cores in the base of the jet as seen in the IRIS SJ 1330 Å image at 06:34:53 UT. The brightening in the base evolves into a circular ribbon seen in the Hα images (see Figure 1b) and arcade-like feature in the AIA 94 Å images (see Figure 1d). The increase of the bright region in the base of the jet is then followed by plasma ejections along the coronal loops (see the animation associated with Figure 1). Along with the plasma ejections, we can also see that dark mini-filament threads are lifted off in the Hα images, indicating of an eruption of mini-filaments that agrees with the previous studies (see e.g. Sterling et al. 2015). In the Hα images, we also see that the jet front appears to be brighter than the background. This indicates that there is a cool counterpart (as seen in Hα) of the jet. The multi-thermal nature of the jet is also confirmed by the constructed EMs, which show clear structures with enhanced emission in the temperatures from log(T/K)=5.8 to log(T/K)=6.7 (see the structures denoted by the arrows in Figures 1e-g).
Almost at the same time when the jet was leaving the bright base, we see brightenings in the remote footpoints of the jet loops (see the sub-image shown in each panel of Figure 1 and the associated animation). In the Hα images, we can see multiple round-shaped bright kernels with a subarcsecond size, which are isolated from each other (see three clear examples denoted by purple arrows in Figure 1b). The flaring time of each bright kernel appears to be different from the others. In the IRIS 1330 Å images, bright features with similar structure are also occurring at the same location (see Figure 1a). However, they are not as isolated as that shown in the Hα images because the saturation in the IRIS observations. Similar bright features can also be seen in the AIA 171 Å images (see Figure 1c). The morphology of the bright feature in the AIA 94 Å images is a contrast to that shown in the others diplayed in Figure 1. The bright features in the AIA 94 Å images appears to have only one bright kernel that is located in the place surrounded by those seen in the other three passbands (see Figure 1d). We further compare the light-curves of the AIA 94 Å emissions at the jet base and the remote footpoints ( Figure 3). There is a time lag of 60 s between the peak in the emission of the remote footpoints and that of the jet base (see Figure 3). Moreover, in the AIA 94 Å images, we can see that the dynamics in the remote footpoints is followed by sudden extension of brightness along the jet loop toward the jet base (see the animation associated with Figure 1). The bright feature shows response only in the AIA 94 Å and 131 Å passbands, suggesting its high temperature nature. The DEM analyses reveal enhanced emissions in the temperatures from log(T/K)=6.7 to log(T/K)=7.3 (see the structures denoted by the arrows in Figures 1h-i). These EM images indicate that the feature could be heated to a temperature as high as 10 7 K.
The variations of the NoRP microwave flux show clear increase in both frequencies during the time when the remote footpoints are getting brighter in AIA 94 Å (see the peaks of the NoRP curves around 06:38 UT in Figure 3). This indicates the enhanced nonthermal radiations of the Sun due to nonthermal particles produced in magnetic reconnections (e.g. Guo et al. 2015). In order to determine whether the enhancement of nonthermal radiation is from the studied event, we carefully check the AIA fulldisc observations between 06:20 UT and 06:50 UT using Helioviewer 2 and the RHESSI flare list. We found that the peaks of NoRP curves around 06:32 UT are corresponding to the impulsive phase of the B9.7 flare mentioned previously and the peaks around 06:47 UT are associated with another weak flare occurring in a different active region. During the period between these two small flares, we do not see any other obvious activities on the near side of the Sun, neither in the AIA EUV images nor in the RHESSI records. Therefore, we believe that the physical processes associated with the jet are responsible for the enhancement in the NoRP flux around 06:38 UT.
After 06:35:37 UT, we observe that the plasma ejecta of the jet are moving along the jet loops toward the remote footpoints (hereafter, we name the motions toward the remote footpoints as upward motion). It starts to fall back toward the jet base around 06:40 UT (hereafter, we name the motions toward the jet base as downward motions). To investigate the dynamics of the plasma ejecta, we produce timedistance plots along the trajectory of the jet loops (see the yellow dotted line in Figure 1d) using the AIA 171 Å observations ( Figure 4a) and the AIA 94 Å observations (Figure 4b). The upward motions of the ejecta have speed more than 130 km s −1 , and the downward motions reach a speed as high as 100 km s −1 (both with 1σ errors of 10-30 km s −1 ). Both the upward and downward motions of the plasma can be clearly seen in the AIA 171 Å (see the dotted line in Figure 4a), but in the AIA 94 Å channel the downward one cannot be detected (see Figure 4b and also the animation associated with Figure 1). We can see that the upward ejecta are much brighter than the downward ones. These facts indicate that the upward ejecta are multi-thermal with plasma as hot as 6.3 × 10 6 K, whereas the downward ones are much cooler.
In the AIA 94 Å channel, we can see that the brightening in the remote footpoints also leads to extension of brightness along the loops (see the region indicated by the con-2 https://helioviewer.org tour in Figure 4b and the animation associated with Figure 1). This extension of bright features cannot be seen in the cooler temperatures, suggesting that its temperature can reach 6.3 × 10 6 K (i.e. representative temperature of the AIA 94 Å channel). At the points where the extension of the brightness from the remote footpoints approaches the jet ejecta, we see that the jet ejecta start changing their directions from upward to downward (see Figure 4a&b). This suggests a connection between the brake on the ejecta and the extension of the brightness from the remote footpoints. We manually select three trajectories of the ejecta based on the AIA 171 Å time-distance plots (see "R1-R3" in Figure 4a). In Figure 4c-e, we show how the velocities change with time along the three trajectories. It shows that the velocities rapidly change near the points where the velocities are close to zero (i.e. when the ejecta are approaching the extending brightness from the remote footpoints). This can also be demonstrated by the accelerations derived from the velocity curves in Figure 4f-h. The accelerations have positive value if it has direction toward the remote footpoints. We can see that the decelerations near the point of zero velocity are in the range of 1.5-3 km s −2 with a 1σ error of 0.8-1 km s −2 (see the points near that marked by the dotted lines in Figure 4f-h). These are much larger than the gravitational that is less than 0.3 km s −2 . It is clear that the velocities of the ejecta change directions in a very short period of time that the gravity itself cannot explain. This indicates that there are other forces pushing the ejecta in the direction from the remote footpoints toward the jet base.
The existence of additional forces is also evidenced by the spectral profiles of the ejecta near the loop top. In Figure 5ac, we show the slit images at the wavelengths near 1400 Å at three points of observing time, which are taken for the places near the loop top (see the white dotted lines in Figure 1b&c). From the animation associated with Figure (Figure 5c). It clearly shows that the Si iv 1403 Å profiles are blue-shifted in all the three cases. The profiles at 06:40:18 UT and 06:42:43 UT can be blueshifted to a velocity of 130 km s −1 , and that at 06:41:30 UT is only blue-shifted to less than 100 km s −1 . Since the apparent motions of the ejecta at 06:40:18 UT and 06:42:43 UT have opposite directions, the blue-shifted Si iv profiles in both cases suggest that the field-of-view of the slit covers a section of the loop at both sides of its apex. That means the ejecta have passed through the loop apex. It again indicates existence of forces other than the gravity that push the ejecta toward the jet base.
The ejecta also emit O iv 1399.8 Å& 1401.2 Å above the instrumental noise level (see Figure 5d-f). This allows a diagnostics of the electron densities of the ejecta. Based on the line ratio vs. electron density curve given by the version 9 of the CHIANTI database (Dere et al. 1997;Dere et al. 2019, see our Figure 5g), the electron densities of the ejecta are found to be 4×10 10 cm −3 , 1.5×10 11 cm −3 and 2.1×10 11 cm −3 at 06:40:18 UT, 06:41:30 UT and 06:42:43 UT, respectively. The variation of the electron densities suggests a compressing process in the ejecta near the loop apex.

DISCUSSION AND CONCLUSIONS
In the present work, we report on coordinated observations of a solar jet taken by SDO, IRIS and NVST. The jet was triggered by the destabilization of a mini-filament. The observations reveal that the ejecta of the jet are propagating along coronal loops that are as long as ∼70 ′′ . The propagation of the ejecta away from the jet base can reach a velocity of >100 km s −1 seen in the AIA 171 Å, which is also confirmed by the IRIS spectral observations in Si iv 1403 Å. The ejecta can be seen in IRIS 1330 Å, NVST Hα, AIA 171 Å and AIA 94 Å channels, suggesting both cool and hot components. The emission measures of the ejecta based on the AIA EUV images show enhancements in the temperature range from log(T/K)=5.8 to log(T/K)=6.7. Using the line pair of O iv 1399.8 Å and 1401.2 Å, we find that the electron densities of the ejecta near the loop apex vary from 4 × 10 10 cm −3 to 2.1 × 10 11 cm −3 with a hint of occurrence of compressing processes.
Before the ejecta leave the jet base region, dynamics in the remote footpoints of the coronal loops have started, including multiple compact bright kernels in Hα emission and similar features in IRIS 1330 Å and AIA 171 Å passbands and a brightening in AIA 94 Å. The peak of the AIA 94 Å brightenings in the remote footpoints lags 60 s, comparing to that in the jet base. The NoRP microwave flux at 17 GHz and 35 GHz also show increases in phase with the AIA 94 Å brightenings in the remote footpoints. The occurrence of the remote brightening is followed by a sudden extension of brightness from the remote footpoint toward the jet base. The emission measures of this bright feature based on the AIA EUV images show enhancements in the temperature range from log(T/K)=6.7 to log(T/K)=7.3.
Brightenings at remote footpoints are not rare for energetic events in the corona that involve larger scale closed magnetic fields. Strong et al. (1992) reported that a short-lived X-ray bright point triggers a brightening at the remote footpoint of a loop with a seperation of 340 Mm between the two footpoints, and that also generates a jet at the remote footpoint moving reversely toward the X-ray bright point. Similar phenomena appear to be common in coronal jets if they flow along closed magnetic field (Shimojo et al. 2007). Using Hinode X-ray observations, Shimojo et al. (2007) analysed in detail one such case that generates reverse jet with a speed of 90 km s −1 in the remote footpoint that is 480 Mm away from the main jet. They found a time lag of 700 s between the peak time of the main jet and that of the reverse jet, and that indicates the energy is transported with a speed of 680 km s −1 . Shimojo et al. (2007) suggested a physical mechanism for the occurrence of remote brightening and formation of reverse jet by heat conduction and/or MHD waves. This kind of activities in remote footpoints has also been frequently seen in flares (see e.g. Wang & Liu 2012;Sun et al. 2013;Wang et al. 2014;Yang et al. 2014), and interpreted as direct heating by nonthermal electrons at relativistic speed traveling along the large scale loops connecting the main flaring to the remote sites (Liu et al. 2006, and references therein).
What is the mechanism of the remote brightening seen in the present case? The 60 s time lag between the peak time of the jet base and that of the remote brightening indicates a speed of about 800 km s −1 for the energy transportation, which is comparable to that found in the case reported by Shimojo et al. (2007). This speed is comparable with the typical coronal Alfvén speed and the propagation speed of the conduction front from flare site is also about the same speed (∼500-1000 km s −1 ) for high temperature plasma (∼ 10 7 K). The sudden extension of brightness from the remote footpoint could be a consequence of the conduction front that may excite another evaporation flow or the Alfvén wave that may trigger another flare (reconnection) or directly heat coronal plasma, causing counter evaporation flow. Alternatively, the bright kernels in the remote site in the present case have a compact round shape, which agrees with the nonthermal electron heating scenario (see similar examples but larger scale in Wang & Liu 2012). Evidence of nonthermal electrons produced in this coronal jet has been given by radio emissions at 17 GHz and 35 GHz observed by NoRP, and have also been reported in the other jets recently (Chen et al. 2018;Paraschiv & Donea 2019). The time lag between the peak times of the jet base and that of the remote footpoint is relatively large for the scenario of nonthermal electrons. This can be interpreted as the time required to reconfigure the magnetic field in both the jet base and the remote site that allows nonthermal electrons escape and produce the brightenings (Wang & Liu 2012). The nonthermal heating in the remote footpoints can then lead to chromospheric evaporation (see e.g. Tian et al. 2015;Dudík et al. 2016;Zhang et al. 2016b;Li et al. 2018;Polito et al. 2019;Li et al. 2019). How such a phenomenon takes place and the role of the nonthermal electrons would be interesting subjects for detailed theoretical modelings, but that is beyond the scope of the present study. In any case, the extension of the brightness from the remote footpoint seen in AIA 94 Å channel contains only hot plasma with temperature of about 10 7 K, suggesting that they are results of the chromospheric evaporation. The very fast propagation of the brightness could be due to the heating front that can travel much faster than the materials (De Pontieu et al. 2017).
Based on the discussion above, we have the following scenario for the event studied here. Initially, a mini-filament destabilization process triggers magnetic reconnection in the corona at the jet base, which is the same as that described in Sterling et al. (2015). Then, the magnetic reconnection produces high speed nonthermal electrons, MHD waves and/or conduction front that move along the large coronal loops and the jet ejecta containing materials from the mini-filaments that move much slower in the same loops. Next, chromospheric evaporations are generated at the remote footpoints as the consequences of the interactions between the chromospheric plasma and the nonthermal phenomena produced by the magnetic reconnection processes of the jet (such as nonthermal electrons, MHD waves and/or conduction front).
After that, the chromospheric evaporations move along the coronal loops toward the jet bases and meet the jet ejecta near the loop tops. Finally, the jet ejecta are braked by the chromospheric evaporation and reversed their moving directions.
The imaging and spectral data reveal that the ejecta turn back to the jet base after crossing the apexes of the loops and meet the extension of brightness from the remote footpoints. The resulting decelerations are found to be in the range from 1.5 ± 0.8 km s −2 to 3.0 ± 1.0 km s −2 while the ejecta are in the places with velocities approaching zero. This indicates that in addition to gravity, there must be other forces that brake the ejecta and turn them backward. The additional forces are very likely the pressures of the chromospheric evaporation in the remote footpoints. If we ignore any hydrodynamics effects in the interface of the ejecta and the evaporation materials, we can have a simplified cylinder model. With this simple model, based on the one-dimensional momentum equations for compressible fluid we can estimate the dynamic pressure (P evp dy ) of the evaporated materials near the loop tops, where we can ignore the gravity, using the following formula: where k is Boltzmann constant, n evp and T evp are the number density and temperature of the chromospheric evaporation respectively and the term of n evp kT evp is the corresponding thermal pressure, n jet and T jet are the number density and temperature of the jet ejecta respectively and the term of n jet kT jet is the corresponding thermal pressure, m jet is the mass of the ejecta of the jet, S jet is the area of the crosssection of the ejecta, a is the deceleration of the ejecta, ρ jet is the mass density of the ejecta, L jet is the length of the ejecta that is about 35 ′′ (i.e. 2.5 × 10 7 m) in the present case, and P jet dy is the dynamic pressure of the jet. The electron density of the ejecta is about 4 × 10 10 cm −3 and thus the mass density ρ jet is about 7 × 10 −11 kg m −3 if we assume the ejecta contain only electrons and protons. P jet dy expresses 1 2 ρ jet v 2 jet where v jet is the speed of the jet that is about 130 km s −1 , and thus P jet dy is given 0.6 Pa. Based on the analyses of emission measures, here we use 10 6 K as the temperature of the ejecta and 10 7 K as the temperature of the chromospheric evaporation. Because the number density of the chromospheric evaporation cannot be achieved with the present data, we adopt the value of 10 10 cm −3 , which is the same order of that found in the chromospheric evaporations of C class flares (Milligan 2011;Polito et al. 2017). While a is given 1.5 ± 0.8 km s −2 , the dynamic pressure of the chromospheric evaporation (P evp dy ) is 2.4 ± 1.4 Pa. While a is given 3.0 ± 1.0 km s −2 , the dynamic pressure of the chromospheric evaporation (P evp dy ) is 5.1 ± 1.8 Pa. P dy expresses 1 2 ρ evp v 2 evp , where ρ evp is the mass density of the chromospheric evaporation and assumed to be 1.8×10 −11 kg m −3 using the number density given above, v evp is the velocity of chromospheric evaporation. The velocity of chromospheric evaporation (v evp ), which is normally difficult to obtain due to the insufficient temporal resolution, can be deduced here via the calculations of the dynamic pressure. For a = 1.5 ± 0.8 km s −2 , v evp is given as 490 ± 160 km s −1 . For a = 3.0 ± 1.0 km s −2 , v evp is given as 750 ± 130 km s −1 . These values are consistent with those reported previously for flares based on spectroscopic data, in which the line-ofsight velocity of chromospheric evaporations are found to be a few hundred kilometers per second (see Tian et al. 2015, and the references therein). Please note that the change of cross section of the flux tube and the effect of MHD waves should also play a role, but that should involve an MHD simulation to understand the physics thus could be an interesting subject for a future study.
In summary, the heating in the remote footpoints by nonthermal electrons may act as a brake on the jet flows along coronal loops. This effect prevents the transportation of plasma from the jet base to a remote site along a coronal loop. The changes in the motions of the ejecta caused by the brake can then allow a diagnostics to the dynamics of the plasma heated in the remote footpoints.