Fragmented Aurora-like Emissions (FAEs) as a new type of aurora-like phenomenon

This study analyses the observations of a new type of small-scale aurora-like feature, which is further referred to as Fragmented Aurora-like Emission(s) (FAEs). An all-sky camera captured these FAEs on three separate occasions in 2015 and 2017 at the Kjell-Henriksen Observatory near the arctic town of Longyearyben, Svalbard. A total of 305 FAE candidates were identified with varying degrees of certainty. They seem to appear in two categories randomly occurring individual FAEs and wave-like structures with regular spacing between FAEs alongside auroral arcs. FAEs show horizontal sizes typically 5 below 20 km, a lack of field-aligned emission extent and short lifetimes of less than a minute. Emissions were observed at the 557.7 nm line of atomic oxygen and at 673.0 nm (N2, first positive band system), but not at the 427.8 nm emission of N2 or the 777.4 nm line of atomic oxygen. This suggests a limit to the energy of the generating mechanism. Their lack of field-aligned extent indicates a different generation mechanism than for aurora, which is caused by particle precipitation. Instead, these FAEs could be the result of excitation by thermal ionospheric electrons. FAE observations are seemingly accompanied by elevated 10 electron temperatures between 110–120 km and increased ion temperatures at F-region altitudes. One possible explanation for this are Farley-Buneman instabilities of strong local currents. We provide an overview of the observations and discuss them as well as potential generation mechanisms in the present study.

located at the ESR. The ASC images used in the present study have a size of 2832 × 2832 pixels. The images were taken using an exposure time of 4 s and an ISO of 16000 at a cadence of 11 to 12 s, with a mean interval length of 11.8 s. This variance is due to variations of the read-out time to the attached computer, with the camera exposure time set to 55 10 s. A simple astrometry calibration was used to find the centre of the ASC images and estimate the pixel size, resulting in a scale of 16.59 pixel/degree close to centre. This is further used to determine the offset of FAEs from zenith, which can then be used to calculate the pixel sizes in km for varying elevation angles, using an equidistant projection, at an assumed FAE altitude of 110 km. This assumption was based on FAE signatures in the ESR data. and 2), consisting of 4 s (8 s) for a full 360 • scan plus another 4 s (8 s) for a background scan. Thus, scanning across the sky takes 2 s (4 s). The background measurement is achieved by tilting a narrow band-pass (∼0.5 nm) interference filter for each channel (Chen et al., 2015). 65 High temporal resolution optical observations from ASK are used to further study the movement and emission properties of the FAEs. ASK consists of three channels with individual band-pass filters for selected auroral wavelengths and lenses to adjust FOV (Ashrafi, 2007). This allows for simultaneous observations of different auroral emissions in a narrow FOV, which can be used to study the energy and flux of the precipitating electrons that produce the aurora (Lanchester et al., 2009). The temporal resolution is 20-32 Hz, and for resolutions above 5 Hz, the available 512 pixels for each camera are binned into a 256 × 256 70 pixel image (Goodbody, 2014). ASK is pointing towards the magnetic zenith and shares part of its observation region with the ESR and the MSP, which led to a finding of a FAE signature in the ESR data after observing a passing in ASK. The ASK FOV is 6.2 • and in this study we use observations of N 2 (673.0 nm, first positive band system) and atomic oxygen (777.4 nm) emissions.
Solar wind data from the Advanced Composition Explorer (ACE) and Deep Space Climate Observatory (DSCOVR) satellites 75 at the L1 Lagrangian point can provide insight into the background conditions during the observed events. For the periods preceding the two larger events (2 and 3) the ACE and DSCOVR data show average speeds of 620-640 km/s, which is above the usual threshold value for high-speed streams (Cranmer, 2002 from SuperDARN radars suggest an ionospheric plasma flow primarily in the northwest or southwest direction. For all our event times Svalbard was located in the evening cell of the convection and close to the flow reversal.

Methods
The FAE candidates appearing on the ASC images were visually identified and marked by eye, with the Fiji distribution of the freely available ImageJ software (Rueden et al., 2017;Schindelin et al., 2012), using the freehand selection tool. This resulted in a compiled database with all candidates containing their outlines, pixel coordinates and sizes. A total of 305 candidates were marked for further analysis and categorised into 4 confidence groups, depending on their intensity, size and outline characteristics. Group 1 is composed of the most well-defined candidates with clear borders and strong intensity enhancements, 90 whereas candidates in groups 2-4 are of decreasingly lower quality, meaning they are more likely to contain features that are for example part of an auroral arc. The 21 FAEs of the highest quality form group 1, whereas group 2 contains 55 candidates.
These 76 candidates are considered as the core set of observations. Group 3 contains 78 candidates and group 4 encompasses 151 candidates. FAEs in groups 3 and 4 are analysed in the same manner, but only contribute to the final conclusions if they agree with the core set findings, which would indicate that these are indeed observations of the same phenomenon.

FAE characteristics
FAEs can be categorised into two distinct categories, the first being individually occurring FAEs. These occur seemingly randomly across the sky, sometimes with a significant offset to the closest auroral arc. The second type are periodic structures with regular spacing between FAEs, which appear close to and generally northwards of auroral arcs. The category 2 FAE group shown in Figure 3 is a typical example.

Distribution, sizes and movement
For the three observed events, most FAEs (73.1%) occurred west of zenith. This is the case for both high-and low-quality candidates, with the dashed kernel density estimation (KDE) in Figure 1 for FAEs of groups 1 and 2 agreeing with the overall distribution KDE. Due to the observational bias caused by the vast majority (262) of FAEs occurring during event 3, this asymmetry in FAE location on the sky might simply be explained by the underlying space weather and ionospheric convection 105 conditions being biased towards westward convection during this period. The low number of FAEs close to zenith (see Figure 1) is possibly explained by observational bias, since FAEs near zenith are harder to identify. Their lack of field-aligned emission extent is not visible when viewed from directly underneath. In addition, most FAEs occurred close to auroral arcs, which rarely appeared close to zenith during the analysed events. The location of category 1 FAEs appears to be fairly random and not necessarily close to auroral arcs, whereas category 2 FAE groups generally appear within the vicinity northwards of an arc, 110 typically with an offset on the scale of the fragment size, corresponding to a few kilometres. Visual inspection of all events shows that FAEs appear mostly elliptical, thus fitting an ellipse to follow the marked outline of each FAE provides a more robust estimate of its size. As shown in Figure  has a mean value of 2.04. Most FAEs seem to have fairly regular, rounded shapes with few indents, with a mean circularity value of c = 0.705 (c = 1 being perfectly circular), which is determined using the formula c = 4π · [Area]/[Perimeter] 2 . This determination is of course affected by their size, with deviations from rounded shapes being harder to identify in smaller FAEs, with an added general operator bias to outline regular shapes compared to complex indents. It should be noted that due to the 4 s integration time of the ASC, any fast-moving object will appear somewhat elliptical. Nevertheless, this is not true for 120 the high-framerate data from ASK, which also show FAEs to be elliptical. The described trends are observable in both highand low-quality candidates, as KDEs for high-quality FAEs are in good agreement with the entire data set in Figure 2. This suggests that most of the marked candidates of groups 3 and 4 are indeed FAEs. Category 2 FAEs can be seen moving along the auroral arc in Figure 3. The distance between these FAEs does not vary significantly as they move eastward over a period of 35 seconds. A spatial intensity variation is visible in the grouped structure, where FAEs appear dim towards the edges of 125 the group and become more intense the closer they move towards the centre. Some of the variation in intensity seems to be caused by fragments appearing and disappearing at the ends of the group. Using an average distance of 45 pixels between the FAEs and their approximate elevation angle of ∼ 65 • , we can roughly estimate the spacing between FAEs for this group to be around ∼ 6 km. Visual inspection of the ASC images shows a general westward movement of the FAEs for the observed events, which might originate from the underlying convection pattern. No obvious eastward motion was observed. A few FAEs were 130 observed in the ASK high-framerate images (see Figure 5), with some remaining stable for multiple seconds while they drift, whereas others appeared and vanished within a second. The ASK FOV corresponds to 10×10 km 2 at an altitude of 100 km, which FAEs passed within ∼10-14 s. This results in an estimated drift speed on the order of ∼1 km/s.  moving nature of FAEs, the respective MSP red channel measurements are unlikely to show any distinct FAE signatures, with any potential emissions "smeared" over the temporal axis. Figure 4 shows a clear peak at the FAE elevation of ∼100 • in the  At least one FAE was passing through the ASK FOV during event 2 on 2015-12-07 (for the corresponding video file see

Observed emissions
Whiter (2020)), which provides much higher temporal and spatial resolution observations. It shows N 2 emission signatures at the energy of the generation mechanism between ∼8-11 eV.

Plasma characteristics measured with the ESR
To further understand the underlying plasma properties of FAEs, an attempt was made to find signatures within incoherent scatter data of the ESR. The auroral arc visible south of the FAE in Figure 5 extended across the entire FOV of ASK (partially shared with the ESR) shortly before the FAE occurrence at 18:23 UT, and is visible in Figure 6 as a general increase in electron 165 density across the entire altitude range. The density decreases across most altitudes as the arc moves out of the FOV towards 18:23 UT. It remains high at 113 km at the time of the FAE occurrence. No associated increase in electron temperatures is visible in Figure 6 for the period and altitudes of the arc signature in the electron density panel.
The FAE visible in Figure 6 shows as a local increase in electron temperature to ∼2300 K at 113 km around 18:23 UT.
This increase seems to be confined to a narrow altitude range, which is further established by the time series at four successive 170 altitudes shown in Figure 7. The increase at the time of the FAE passing is limited to altitudes below 119 km and strongest at 113 km. For the period directly after the FAE occurrence, multiple increases in electron temperature are visible at low altitudes, which indicates an unstable lower ionosphere. Simultaneous increases in ion temperatures are visible at higher altitudes, with significant increases around 190 km, up to ∼4500 K.
The background conditions during these analysed events might be able to further provide some insight into the underlying 175 generation mechanism. For the entire duration of event 3, significant intermittent increases in electron temperatures were observed at altitudes in the E-region, as well as elevated ion temperatures (mostly) in the F-region. This seems to indicate a connection between FAEs and elevated electron temperatures at low altitudes, which we will discuss below.

Discussion
From the presented measurements it is clear that a new aurora-like phenomenon has been observed. As the FAEs were found 180 by manual inspection of images, there is some bias in which features were selected and how they were classified. The data set could contain other auroral small-scale forms or diffuse patches, which is the reason for the classification into four confidence groups. As the general properties of candidates between high-and low-confidence groups agree well, we are confident that  in ion temperatures at higher altitudes can provide some clues towards the origin of FAEs.
One possible group of generation mechanisms for the required energy to excite FAEs are Farley-Buneman instabilities, which are streaming instabilities typically occurring at altitudes of 90-120 km (Oppenheim et al., 1996). The proposed FAE altitude  falls within this region. They become significant when the difference between electron and ion drift speeds exceeds the ion acoustic speed (Liu et al., 2016), which is generally the case in geomagnetically disturbed conditions, typically also resulting in aurora. This would explain why FAEs are observed alongside aurora. Particularly at high latitudes, these instabilities can result in significant local electron heating. This is consistent with the low-altitude elevated electron temperatures observed during the FAE events, for which Farley-Buneman instabilities are the most likely explanation.

200
The observed large ion temperatures in the F-region around 190 km height are caused by Joule heating from strong electric fields, or ion-neutral friction. The measurements are used to estimate the electric field strength below assuming that E = 0, i.e. the magnetic field-lines are equipotentials. We neglect the effect of the slightly different magnetic field strengths between balance, neglecting also thermal energy transfer to/from electrons (whose temperatures are generally not enhanced above the 205 E-region, especially preceding the FAE occurrence at 18:23 UT) is (Alcayde et al., 1983, Equation (4)): Here T i and T n are the ion and neutral temperatures, V i and V n the ion and neutral drifts, respectively. m i is the mean ion mass, k B the Boltzmann constant, ν in the ion-neutral collision frequency, and N n and N e the neutral and electron densities. In the steady state Q in = 0, and for the F-region we insert (V i − V n ) = E ⊥ × B/B 2 with E ⊥ the electric field in the frame of 210 the neutral gas and B the geomagnetic field. We are only interested in the magnitude of E ⊥ , E ⊥ . It can be estimated as  (Williams et al., 1992). It should be noted that this is an approximation and the filtering for average values is based on somewhat arbitrary choices, but the derived E ⊥

220
is not all that dependent on the inserted T i and T n and should exceed the typical limit for Farley-Buneman instabilities by a significant margin regardless of the exact filtering values. The threshold may be already exceeded in the arc, before 18:23 UT, but T e was perhaps not high enough to excite optical emissions. Buchert et al. (2008) showed an example with the ESR where temperatures around 110-120 km, for which Farley-Buneman instabilities are the only known cause at these low altitudes.
Simultaneously, increased ion temperatures are visible at altitudes in the F-region, which enables us to estimate the strength of the E-field. The derived estimate of E ⊥ ≈ 70 mVm −1 exceeds the typical Farley-Buneman threshold of 30 mVm −1 . Category 2 FAE groups show a fairly regular and stable spacing and appear to be modulated by some kind of wave.
Open questions are the exact nature of the generation mechanism, whether FAEs of categories  Author contributions. JD analysed the data set and wrote the present study. NP contributed towards the entire writing and analysis process.

285
DW suggested the FAE name and contributed towards the writing and data analysis process, especially regarding the ASK data. PGE originally discovered the FAEs in ASC images and contributed towards the writing and data analysis process, especially regarding the ASC and MSP data. LB contributed towards the ESR data analysis and respective section. SB suggested Farley-Buneman instabilities as a potential generation mechanism and contributed the respective discussion section.
Competing interests. NP and DW are editors for the special issue this paper is submitted to.

290
Disclaimer. This study is based on J. Dreyer's master's thesis (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-388546), which in parts contains some additional information that might be of interest.