High-latitude crochet: solar flare-induced magnetic disturbance independent from low-latitude

Solar flare-induced High latitude (peak at 70-75◦ geographic latitude) ionospheric current system was studied. Right after the X9.3 flare on 6 September 2017, magnetic stations at 68-77◦ geographic latitudes (GGlat) near local noon detected northward geomagnetic deviations (∆B) for more than 3 hours, with peak amplitudes >200 nT, without any accompanying substorm activities. From its location, this solar flare effect, or crochet, is different from previously studied ones, namely, subsolar crochet (seen at lower latitude), auroral crochet (pre-requires auroral electrojet in sunlight), or cusp crochet (seen only 5 in the cusp). The new crochet is much more intense and longer in duration than the subsolar crochet. The long duration matches with the period of high solar X-ray flux (more than M3-class flare level). Unlike the cusp crochet, interplanetary magnetic field (IMF) BY is not the driver with BY only 0-1 nT out of 3 nT total field. The equivalent ionospheric current flows eastward in a limited latitude range but extended at least 8 hours in local time (LT), forming a zonal current region equatorward of the polar cap on the geomagnetic closed region. 10 EISCAT radar measurements over the same region as the most intense ∆B near local noon show enhancements of electron density (and hence ion-neutral ratio) at these altitudes ( 100 km) where the background Sq ion convection (> 100 m/s) preexisted. Therefore, this new zonal current can be related to the Sq convection and the electron density enhancement, e.g., by descending E-region height. However, we have not found why the new crochet is found in a limited latitudinal range, and therefore the mechanism is still unclear compared to the subsolar crochet that is maintained by transient re-distribution of 15 electron density. The signature is sometimes seen in the Auroral Electrojet (AE) index. A quick eye-survey for X-class flares during solar cycle 23 and 24 shows clear AU increases for about half the >X2 flares during non-substorm time, although the latitudinal coverage of the AE stations is not favorable to detect this new crochet. Although some of them could be due to auroral crochet, this new crochet can be rather common feature for X flares. 20 key points (1) We found a new type of the solar flare effect on the dayside ionospheric current at high latitudes but equatorward of the cusp during quiet periods. 1 https://doi.org/10.5194/angeo-2020-48 Preprint. Discussion started: 9 July 2020 c © Author(s) 2020. CC BY 4.0 License.

However, as shown in this paper, we found that the crochet at high latitudes is not a simple extension or sub-effect of, but is independent from the subsolar crochet with larger amplitude and longer duration. We show this from a case study of X9.3 flare on 6 September 2017. We also show how these effect are seen in geomagnetic AE index using about 60 non-substorm time flares of >X2 class during past two solar cycles (cycle 23 and 24). The solar flare X-ray data observed by GOES satellites are obtained from NOAA, and the geomagnetic indices are obtained from the world data center (Kyoto and Copenhagen). The 70 other data are described in Yamauchi et al. (2018).
2 High-latitude crochet for X9.3 flare on 6 September, 2017 In the overview paper of EISCAT radar observations and geomagnetic disturbances near local noon during the 6-8 September 2017 space weather event, Yamauchi et al. (2018) briefly mentioned a sudden enhancement of ∆B (> 150 nT) at high latitudes (> 68 • GGlat) in response to the X9.3 flare on 6 September, 2017, but without special note nor detailed description on this 75 high-latitude disturbance compared to subsolar crochets, auroral crochets, or cusp crochets. 2018). This is also seen in ASY-D (63 nT at 12:04 UT) and ASY-H (77 nT at 12:05 UT) as shown in Figure 2a, with amplitude change by the flare about 60 nT. One can even recognize a crochet-like signature in ASY-D when an X2.2 flare occurred at around 9:00 UT.

New crochet after X9.3 flare
The new finding is the subsequent geomagnetic disturbances: a large positive ∆H deviation (northward ∆B) also started right 90 after or even during the negative ∆H spike (south-weastward ∆B) of the subsolar crochet, with much higher amplitudes as shown in Figure 1b. This positive ∆H continued for hours with the peak at around 13:00 UT at Bear Island (BJN) at 74.5 • GGLat (> 200 nT), 13:20 UT at 70 • GGlat (SOR and TRO: 180 nT) and 68 • GGlat (KIR: 140 nT), and 14:50 UT at 65 • GGlat (ROR: 70 nT). With larger amplitude and longer duration than the subsolar crochet, this geomagnetic signature is visible even in the AU index, as shown in Figure 2b, although the baseline is as large as 100 nT due to the previous substorm activity and 95 can hide the subsolar crochet if any exists.
The development is also quick. At BJN (74.5 • GGlat), ∆H reached ∆H>65 nT at already 12:05 UT, i.e., at the peak time of the subsolar crochet, and reached 130 nT at already 12:20 UT. Since the long duration already indicates that the generation mechanism is different from that of the subsolar crochet (re-distribution of electron), the positive ∆H of this crochet with diminishing amplitude toward lower latitude should cancel the negative ∆H of the subsolar crochet at high-latitude. In fact, 100 ∆H exceeded the value before the flare at 12:10 UT at 70 • GGlat, 12:12 UT at 68 • GGlat, and 12:17 UT at 65 • GGlat. 5 https://doi.org/10.5194/angeo-2020-48 Preprint. Discussion started: 9 July 2020 c Author(s) 2020. CC BY 4.0 License.
These large ∆H, however, is observed in a limited latitudinal range, diminishing toward higher latitudes with 140 nT at 76.5 • GGlat (Hopen: HOP, at 12:50 UT) and not visible at 78.2 • GGlat (Longyearbyn: LYB). Since the geomagnetic latitude of LYB is only 75.3 • and IMF is weak with B Y = 0 nT, the positive ∆H is limited to the geomagnetic closed region outside the cusp nor polar cap. The closed geometry is also indicated by the EISCAT Svalbard radar data (Yamauchi et al., 2018). The

105
PC index (Polar Cap index), which corresponds to the polar cap activity, shows enhanced values in the same period, but not as prominent as in AU, as shown in Figure 2.
On the other hand, positive ∆B at around 75 • GGlat was observed in a wide local time range, as shown in Figures 1c and 1d (∆X is nearly the same as ∆H in both stations). Danmarkshavn (DMH) at 19 • W and Dikson Island (DIK) at 81 • E showed ∆X about 120 nT and 100 nT from before the flare at around 13:00 UT, respectively. Together with the zero IMF B Y condition, 110 the observed large ∆H cannot be a cusp crochet. Considering its location and pre-existing activity, this crochet is neither the subsolar crochet nor auroral crochet, although some part of the observed ∆H could be affected by the auroral crochet.
For example, DIK is located near the evening terminator (it is still under sunlight near horizon) and the geomagnetic activity before 12 UT indicates some auroral activity. Therefore the first peak at around 12:20 UT, which is larger than that of BJN or HOP and with more westward ∆B, can be the auroral crochet rather than the extension of the new crochet. However, the 115 second and third peaks are in the same direction (northward ∆B) as near local noon, and multiple peaks are not expected for an auroral crochet does not under smoothly declining X-ray flux. Therefore, positive ∆X at DIK at around 13:00 UT and 13:40 UT can be interpreted as a part of the new high-latitude crochet rather than the auroral crochet, although we cannot dismiss the possibility of the auroral crochet.
With such large amplitude, the crochet is visible even in AU although the AU stations are not located at favorable GGlat.

120
During 12:00-14:00 UT, AU has three positive peaks (provisional values of 240 nT at 12:12 UT, 190 nT at 12:55 UT, and 160 nT at 13:43 UT but further baseline subtraction might be needed). The timing of these peaks corresponds to the subsolar crochet and the high-latitude one at BJN, but the provisional AU value reflects DIK data (DIK is one of the AE station) as shown in Figures 1c and 2b. Although the amplitude is larger at BJN than DIK for the second and third peaks, BJN is located far poleward of the AE stations and does not contribute to AU.

Equivalent ionospheric current
From the Norwegian and Greenland magnetometer data, we also calculated the ionospheric equivalent currents, using the Spherical Elementary Current System (SECS) technique (Amm, 1998;Amm and Viljanen, 1999). Here we obtained epsilon In Figure 3, one can see a sudden appearance in the ionospheric current in a wide region at around 12:00 UT when the X-ray flux from the X9.3 flare increased at the Earth. The enhancement is westward at lower latitudes (< 70 • GGlat at 20 • E or 13 LT, and <72 • W at 50 • W or 9 LT), and eastward at higher latitudes in the both meridians. They correspond to the subsolar crochet current in the northern hemisphere (Curto et al., 2018) and the new high-latitude crochet mentioned above, respectively. Figures 4a and 4b show the 2D vector directions, corresponding to this timing: right before the flare (11:50 UT), at the subsolar crochet peak (12:04 UT). The low-latitude side composes the counter-clockwise current that agrees with the high-latitude return current direction of the subsolar crochet (Curto et al., 2018;Annadurai et al., 2018). The high-latitude side forms another independent counter-clockwise current, with strong eastward current near BJN(c. Figure 1) as mentioned above.
The resultant shear poleward of BJN corresponds to upward field-aligned current, i.e., afternoon Region 1 field-aligned current.

140
The eastward current expanded quickly in both longitudinal direction and toward lower latitude as soon as the subsolar crochet diminished. At the same time, the current density gradually increased toward its peak. Figures 4c and 4d show the 2D vector directions at these peaks: at the first minor peak of ∆H at around 75 • GGlat (12:20 UT) after the subsolar crochet disappeared, and at the major peak of ∆H at around 70 • GGlat (13:20 UT), respectively. By 12:20, region is this westward current with peak at around 72-74 • GGlat expanded in a wide local time from Greenland to eventually Siberia (DIK at 81 • E as 145 shown in Figure 1d), i.e., more than 130 • in longitude. The entire current lies on the geomagnetic closed region as mentioned above, and its peak latitude gradually moved equatorward. By the peak time at around 13:20 UT, all region over 5 • in latitude and >100 • in longitude are intensified, with much higher intensity than the subsolar crochet current.
The current direction of eastward (or positive ∆H) is the same as the ionospheric current in the evening auroral oval (we here mean that the region between the upward region 1 field-aligned current and downward region 2 field-aligned current (Iijima 150 and Potemra, 1976;Akasofu, 1977;Yamauchi and Slapak, 2018)), and the observed eastward current continued until the next substorm onset took place at 15:00 UT (cf. Figure 2b). However, no outstanding substorm is visible in AE (Figure 2b) or 7 https://doi.org/10.5194/angeo-2020-48 Preprint. Discussion started: 9 July 2020 c Author(s) 2020. CC BY 4.0 License. DIK data (Figure 1c) before this substorm with continued weak IMF condition. The eastward current patch is even started at Greenland meridian at 9 LT and continued toward the afternoon sector although IMF B Y = 0 nT. Since there was no auroral current signature at BJN before this crochet, this current system is not the auroral crochet current. Rather the question is how 155 much does this new crochet contribute to ∆B in the evening sector, including the auroral crochet. In this sense, we cannot judge moment if the crochet detected in DIK is the evening extension of this crochet or auroral crochet or both effects mixed. However, the Sq is driven by daily neutral convection starting from subsolar region, and hence it has long been expected to be small at high-latitude (Yamasaki and Maute, 2017). Therefore, we need direct evidence with the Sq scenario. For that purpose, Figures 5 shows ionospheric electron density and ion velocity, respectively, observed by EISCAT VHF radar at Tromsø. Figure 5b shows that the electron densities in the 100-200 km altitude range were significantly enhanced by the enhanced 165 X-ray flux, starting at around 12 UT (doubling at 100 km altitude and seen up to 200 km altitude). Figure 5a shows that northward ion convection was also enhanced, and more importantly that background Sq ions convection starting from around 9 UT is already strongly northward (away from the sub-solar region) with large values as much as 150 m/s at 71 • Mlat (at 100 km altitude).

EISCAT data
Since the increase in the electron density means also the increase of the ion-neutral density ratio, the altitude of ionospheric 170 current, is expected to decrease, reaching close to the altitude where the Sq convection is strong. Such changes can enhance the pre-existing Sq current significantly, although this scenario does not easily explain why it is found in a limited latitudinal range and wide longitude.
The time development of the electron density enhancement together with the elevated ion velocity at 100 km matches with the ∆H time profile at BJN that is located under the area observed by the EISCAT VHF radar. Note that ion velocity direction 175 (northward) is the electric field direction at this altitude where only ions are collisional with neutrals but not electrons, and hence collision-free electrons drift westward, resulting in eastward an Hall current (that causes ∆H>0 geomagnetic disturbance).
The EISCAT data in Figure 5d also revealed a decrease of electron density above 300 km after 12 UT. Since photochemistry predicts density increase at all altitudes, this density decrease must be caused by ionospheric dynamics, such as 3D distribution of ionospheric current, while this decrease does not affect the total current because the ion velocity at > 300 km is not changing 180 very much compared to the ion velocity at < 200 km, and not contributing to the ionospheric current.

Preliminary survey results
We further conducted quick survey of geomagnetic AXY/AE indices (cf. Figures 3a and 3b) during past two solar cycles. There are 73 flare events with >X2 class since 1996. For all these events, we examined web-interfaced quick plot for AE and ASY, by adding marks that indicate the X-ray flux level and start timing, as shown in Figure 6 for 15 July, 2002 event. We here show 185 raw plot that is also found as the supplemental materials. For auroral crochets, the precondition requirement is severe: a substantial auroral electrojet must pre-exist in the sunlit hemisphere. This removes more than half the cases, and therefore we expect that the new high-latitude crochet can also be observed in the AE index, as seen in Figure 3b and Figure 6. In Figure 6, even AL deviation started simultaneously as the To examine it further, Figure 7 shows the Scandinavian and Green land geomagnetic data at the same stations as Figure 1. 200 We also added Kullorsuaq (KUV) data from the west coast of Greenland that is 18 MLT at the time of crochet. Then the new high-latitude crochet extends toward the evening quite wide near the summer solstice. Even DIK data (which is 01 LT past the midnight but still under sunlight) showed minor signature. This suggest that AU signature is most likely caused by this crochet rather than auroral crochet.
On the other hand, unique bipolar signature with short-live ∆H>0 is seen at BJN. This is a candidate for the auroral crochet.
a signature of small auroral electrojet is seen at BJN, SOR, TRO, and DIK before the crochet (starting at around 19:40 UT).
Although the signature is not visible at Kiruna and the value at BJN returned normal, weak aurora existed in this narrow region before the crochet.
However, the auroral electric field before the crochet must be very weak compared to what was reported as the auroral 215 crochet (Pudovkin and Sergeev, 1977), and at least the ∆H>0 signature, that is consistently observed at many stations with quiet pre-condition, is better interpreted as the new crochet. Then we can even wonder if the interaction between the new crochet and the auroral oval accelerated the large ∆B<0 bay.
3.1 Discussion and future tasks 3.2 Why not found in the past?

220
Although magnetic stations have been extended toward higher latitudes beyond 68 • GGlat since 1980's, this new crochet has never been reported, at least not to our knowledge. One possible reason is that the phenomenon is limited to a relatively small range in geographic latitude (68 • -75 • GGlat) while the station should be completely outside the geomagnetic cusp (< 75 • Mlat). This criterion excludes many geomagnetic stations over Greenland and Canada from finding the new crochet.

225
Although we made a survey using quick plots, we need to make more solid analyses. For example, we need to include amplitudes (we examined only yes or no), examine the closure of the ionospheric current system at high latitudes, analyze global magnetometer network data, and take such statistics. Such a global perspective would also tell for which conditions the effect can be detected in AU or AL.
One important note is that the difference between the GGlat and Mlat (i.e., UT dependence) must be considered. Also, we 230 have to note that the current system might be different between different events (size of flare may affect the intensity and size, season may affect the distribution pattern). In addition, if intensification of the Sq current is important, the new crochet might be the equinox phenomenon (Yamasaki and Maute, 2017). Therefore, it might be difficult to obtain consistent results, but at least a common feature can be obtained.

What is the main driver of the new crochet?
235 As shown in Figure 4, this current system might be related to enhancement of the Sq current through the enhancement of both the ion/electron density and ion velocity (Pederson electric field). Then the question is the relative importance of these mechanisms. In addition, we need to understand the relation to the nightside crochet that we found a substantial number, but we have not even found the criterion how to classify the observed crochet as the auroral crochet or nightside extension of the new crochet. To answer this, we need similar radar data for different events. Since the availability of radar data in favorable 240 observation modes and geometry is limited (such as the investigated EISCAT data), we need other radar data including future facility such as EISCAT 3D for a solid answer. Such a work also probably give some hints why the electron density decreased at > 300 km in Figure 4. Future satellite missions that cover ionospheric E-region, such as Daedalus (Sarris, et al., 2020), are strongly needed.
4.2 Can crochet trigger a substorm or M-I coupling?

245
In the preliminary survey, we found many "coincidence" cases in which large gradient of ∆H occurred at the same time as substorm onsets or sudden intensifications of substorms when X-flares took place. Figure 8 shows one such example when X6.2 flare took place at 19:13 UT, and substorm onset took place immediately after. Solar wind density and velocity were stable after CME arrival 7 hours before, and traditional explanation of the trigger is IMF changes.
However, the crochet mechanism can trigger a substorm through sudden intensification of ionospheric density and electric 250 field through magnetosphere-ionosphere (MI) coupling (e.g., Kan et al., 1988). Furthermore, this substorm is peculiar with larger AU than AL right after the onset, in agreement with crochet whereas substorm onset is characterized large negative ∆H and AL. Since several different onset mechanism may causes the substorm, it is quite possible that crochet may also trigger a substorms, as one of minor onset mechanisms (Yamauchi, 2019). The investigation of such a scenario is a future task.
The same question arises with respect to the Magnetosphere-Ionosphere (M-I) coupling. The latitude range of the new 255 crochet is inside the closed field-line region (near local noon), but yet close to the dayside Region 1 field-aligned current (Iijima and Potemra, 1976;Yamauchi and Slapak, 2018). This means that the new crochet might influence the field-aligned current system. Such a study requires satellite observation at right location and right timing.

Modulation of Pc5?
If the long lasting high X-ray flux influences the ionospheric current, some effect might arise from the X2.2 flare at 9 UT 260 on the same day. One possibility is the density moderation synchronizing with the Pc5 pulsation during the recovery phase of the substorm that started at 09:37 UT with peak at 05:58 UT with AL=-666 nT ( Figure 2). In fact, ∆H showed largeamplitude oscillations with periodicity about 30 min during 09:30-11:20 UT ( Figure 1b). However, electron density at 150-200 km altitude showed irregular oscillations with a different periodicity ( 15 min) although no mediation is seen at 100 km altitude.
The periodicity in the ion convection is also about 15 min for 150-200 km altitude. The only candidate that may match with 265 30 min periodicity is irregularity in the ion velocity at 100 km, but profile does not really match with the ∆H variation. This suggests that the Pc5 pulsation can be modulated by the density variation at 150-200 km altitude.

Relation to space weather
In addition to opening up new science case that is concentrated on 68-75 • GGlat, large extra ∆H by X-ray during substorms might have some connection with space weather researches. Therefore, although this work is not currently related to any other 270 researches, this has potential that may collaborate with other researches.
Using the Norwegian geomagnetic chain and EISCAT data, we found a new type of solar flare effect on the geomagnetic disturbance (SFE, or crochet) in response to the X9.3 flare on 6 September 2017 at high latitudes (65-75 • GGlat). The new crochet is found over a wide local time range including local noon but outside the cusp, i.e., in the geomagnetic closed region.

275
It lasted for a longer duration with higher peak amplitude than the subsolar crochet. The equivalent ionospheric current flows eastward in a limited latitude range but extended at least 8 hours in local time (LT), forming a zonal current region at around 70-75 • Geographic latitude (equatorward of the polar cap in at least in dayside). Considering its location and duration, this crochet is different from previously studied crochets (subsolar, auroral, and cusp).
Ionospheric parameters at local noon during this crochet shows strong background ion convection before the crochet as well 280 as sharp enhancement of electron density (and hence the ion-neutral density ratio). Thus the new crochet can be related to increase electron density at 100-150 km altitudes, where strong Sq ion convection exists. For example, change in the E-layer height can actually cause the ionospheric current at high-latitude, but such scenario does not easily explain why it is found in a limited latitudinal range, and therefore the mechanism is still unclear.
We also examined the crochet signatures in AE and ASY indices for all X flares (> X2.0) over past two solar cycles. While 285 the subsolar crochet is well recognized in ASY indices, crochet signatures that represent the new crochet or auroral crochet are also recognized in AU for half the cases, and even in AL index sometimes. However, AE alone cannot distinguish between this new crochet and the auroral crochet in the evening sector, and further studies is needed to understand the current system related to these crochets.
Data availability. The X-flare list is obtained from National Oceanic and Atmospheric Administration Space Weather Prediction Center