Intermittent Lobe Reconnection Under Prolonged Northward Interplanetary Magnetic Field Condition: Insights From Cusp Spot Event Observations

Previous studies have suggested that lobe reconnection under stable northward interplanetary magnetic field (IMF) conditions should be continuous. However, to what extent it can be considered as continuous is unknown. Auroral cusp spots have been considered a signature of lobe reconnection. Here, we show that during a cusp spot event continuously observed by the Defense Meteorological Satellite Program for approximately 4 hr under stable IMF conditions, the ground radars recorded multiple intermittent equatorward‐moving radar forms (EMRFs) near local noon and multiple intermittent poleward‐moving radar forms (PMRFs) at post‐noon, each lasting ∼20–30 min. These intermittent EMRFs and PMRFs are suggested to correspond to intermittent lobe reconnections occurring at the cusp's poleward boundary near local noon and the duskside boundary at post‐noon, respectively. These findings challenge the previously held notion of continuous lobe reconnection under stable IMF conditions.


Introduction
The interaction between the magnetosphere and the solar wind is strongly influenced by the orientation of the interplanetary magnetic field (IMF), particularly its north-south (B Z ) component.During southward IMF conditions, the IMF's orientation is opposite to that of the Earth's magnetic field at low-latitude side of the cusp, which creates favorable conditions for a low-latitude magnetic reconnection along the dayside magnetopause (Dungey, 1961;Milan et al., 2004Milan et al., , 2007)).When the IMF is directed northward, the rate of low-latitude reconnection on the dayside magnetopause decreases significantly but the reconnection does not cease entirely.The reconnection location tends to shift from the low-latitude side to the high-latitude side of the cusp, which promotes a type of reconnection known as high-latitude reconnection, or lobe reconnection (Dungey, 1961(Dungey, , 1963;;Imber et al., 2007;Lockwood et al., 2001;Milan et al., 2000;Moen et al., 2001).In the region of the magnetospheric cusp, direct deposition of particles from the magentosheath into the polar region is possible, resulting in the generation of a particular auroral form, known as the "cusp spot" (Carter et al., 2020;Frey et al., 2002;Milan et al., 2000).Fuselier et al. (2000) proposed that the site of lobe reconnection could remain stable under steady IMF conditions.Frey et al. (2003) showed an event of cusp spot that persisted for hours, leading to the inference that lobe reconnection is continuous in nature and its steadiness depends on the way that the process is driven.There are many studies, suggesting the same: the lobe reconnection is a steady or quasi-continuous process under prolonged stable IMF conditions (Dorrian et al., 2019;Fuselier, 2021;Guo et al., 2021;Marcucci et al., 2008;Retinò et al., 2005).
However, it should be noted that in addition to observations indicating continuous reconnection, there are numerous observations that imply intermittent reconnection.Low-latitude reconnection at the magnetopause has also been found to be unsteady (Lockwood et al., 2001;Wild et al., 2001;Zou et al., 2022).Intermittent reconnection typically occurs in the form of local reconnection bursts during flux transfer events (FTEs), Abstract Previous studies have suggested that lobe reconnection under stable northward interplanetary magnetic field (IMF) conditions should be continuous.However, to what extent it can be considered as continuous is unknown.Auroral cusp spots have been considered a signature of lobe reconnection.Here, we show that during a cusp spot event continuously observed by the Defense Meteorological Satellite Program for approximately 4 hr under stable IMF conditions, the ground radars recorded multiple intermittent equatorward-moving radar forms (EMRFs) near local noon and multiple intermittent poleward-moving radar forms (PMRFs) at post-noon, each lasting ∼20-30 min.These intermittent EMRFs and PMRFs are suggested to correspond to intermittent lobe reconnections occurring at the cusp's poleward boundary near local noon and the duskside boundary at post-noon, respectively.These findings challenge the previously held notion of continuous lobe reconnection under stable IMF conditions.
Plain Language Summary Under certain conditions where the solar wind magnetic field points toward the North, there is a higher chance of a specific type of magnetic interaction called "lobe reconnection" happening near the Earth's dayside boundary with space.Previous findings suggested that lobe reconnection could occur continuously under stable interplanetary magnetic field conditions.Based on coordinated satellite and ground observations of auroral cusp spot event, we revealed that the pattern of lobe reconnection is intermittent rather than continuous.This discovery holds particular significance in advancing our understanding of the solar wind-magnetosphere coupling process in the cusp region.
XIONG ET AL.

RESEARCH LETTER
which appear quasi-periodically in cascades of up to 10 FTEs in succession (Farrugia et al., 1987;Kullen et al., 2019).Therefore, it is puzzling why the lobe reconnection under prolonged northward IMF conditions should be always continuous.At the very least, we should establish to what extent lobe reconnection can be considered continuous.
This study aims to demonstrate that the seemingly continuous lobe reconnection, as indicated by continuous cusp auroral spots observed under prolonged northward IMF conditions, is actually comprised of intermittent and repetitive lobe reconnection processes.

Data and Event
The Defense Meteorological Satellite Program (DMSP) consists of multi satellites flying at an altitude of approximately 800 Km (Hardy et al., 1984).The Special Sensor Ultraviolet Spectrographic Imager (SSUSI) on board DMSP can generate auroral maps in five different wavelength bands.In this study, we used the SSUSI data of DMSP F16, F17, and F18 over the Northern Hemisphere on 14 December 2012.We also used data of Special Sensor J (SSJ/5) instruments of DMSP, which monitor both ion and electron particle precipitation (Redmon et al., 2017) across 19 energy channels ranging from 30 eV to 30 keV.
Ground-based radar observations from the Super Dual Auroral Radar Network (SuperDARN) are also used in this study.In normal operating mode, SuperDARN radar scans continuously in 16 beam directions, with an interval of 3.24° between adjacent beams, and a beam stop time of 7 s.A complete scan takes 2 minutes and covers about 52° of azimuth (Greenwald et al., 1995).Such spatial coverage and temporal resolution enable the radar network to monitor ionospheric plasma convection at high latitudes.We also used observations of EISCAT Svalbard 32-m radar (ESR) located at Longyearbyen (78.15°N, 16.03°E).The radar was operated with an observation mode of a fixed elevation angle of 30° and azimuth angle of 330.9°.The observable geomagnetic latitude range is about 75°-82°.Besides, the solar wind and IMF data with 1 min time from OMNI in this study were provided by the Coordinated Data Analysis Web data service (http://cdaweb.gsfc.nasa.gov).
The event examined in this study occurred on 14 December 2012 from ∼09:00 to 13:00 UT.In Figure 1, black fan-shaped area indicates a polar projection of the field of views of SuperDARN Hankasalmi from 09:00 to 12:00 UT for this event.The red straight line illustrates the scanning direction of ESR, which closely aligns with the observational direction of beam 7 of Hankasalmi.The red five-pointed star indicates the position of the ESR radar, while the shaded area illustrates the location of the auroral cusp spot observed by DMSP.

A Long-Lasting Cusp Spot Under Prolonged Northward IMF Condition
Figures 2a-2i show the auroral images in Lyman-Birge-Hopfield short (LBHS, the SSUSI 140-150 nm channel) band observed by SSUSI on DMSP F16, F17, and F18 in the Northern Hemisphere on 14 December 2012 from 08:16 to 13:34 UT.The red arrow in each image indicates the presence of a cusp spot.Throughout the observation period, the dayside auroral oval was well-documented.From 08:57 to 12:33 UT, we observed distinct bright auroral spots located at the high-latitude side of the cusp region.Such auroral spots have been considered to be a sign of lobe reconnection (Milan et al., 2000).
Figure 2j illustrates the IMF B X , B Y , and B Z , in GSE coordinates from 08:00 to 13:30 UT on the same day.Throughout this time interval, the IMF B X component remained negative, and the B Y and B Z were positive.An outstanding feature is that a strong northward IMF (B Z > 0 and B Z >  | | ) continued for about 3.5 hr before 11:25 UT.After 11:00 UT, the clock angle gradually increased in correspondence with the diminishing dominance of the northward IMF.The solar wind velocity and the dynamic pressure are shown in Figure 2l.There were no sudden changes of solar wind dynamic pressure during this time period.
Similar to the research of Frey et al. (2003), here we also observed continuous cusp spots under prolonged northward IMF conditions.However, it is important to note that the seemingly continuous observations of the cusp spot in our study based on SSUSI DMSP images have a limitation in their low time resolution, which can potentially create a misconception that the lobe reconnection indicated by these cusp spots is continuous.In addition, the variation in auroral intensity may not be sensitive enough to accurately reflect the continuity of lobe reconnection.

Radar Observations Associated With the Cusp Spots
Figures 3a-3c show the range-time plots of the electron density, ion temperature, and ion-line-of-sight velocity (positive away from radar) measured by the ESR between 08:00 and 13:00 UT on 14 December 2012.Meanwhile, Figures 3d and 3e depict the range-time plots of power and velocity (negative away from radar) observed by beam 7 of the Hankasalmi radar in channel A during the same time period.We noticed that these observations can be divided into two distinct segments, namely before and after 11:00 UT.
From 09:00 and 10:00 UT, Figures 1 and 2 illustrate that the scanning range of both the beam 7 of Hankasalmi and ESR covered the aurora cusp spot.In Figure 3, we used the first two red dotted lines to depict the initiation times of two repeated equatorward-moving radar forms (EMRFs), which are visible in both the ESR and Hankasalmi radar observations shown in Figures 3c and 3e.Because the ESR data has a gap around 9:15 UT, the first EMRFs was not detected by ESR.The first EMRF was associated with the ion temperature increase and the 2nd one shows enhancements in electron density.In previous studies, such observations have been generally accepted as a signature of lobe reconnection (Freeman et al., 1993;Hu et al., 2006).From 10:00 to 11:00 UT, three strong equatorward flows were observed propagating toward higher latitudes in Figure 3c.These poleward displacements might have to do with the poleward displacement of the cusp aurora during that time period, as seen in Figures 2d and 2e.This suggests a potential movement of the entire cusp region toward higher latitudes.
From 11:00 to 12:00 UT, the radar's scanning ranges changed from ∼14:00 to 15:00 magnetic local time (MLT) and were just duskward of the duskward boundary of the auroral cusp spot, as shown in Figure 1.During this time period, Figure 3 shows that three poleward-moving radar forms (PMRFs) were observed.Their initiation times are indicated by the three black dotted lines.We argue that these recurrent PMRFs are also a result of lobe reconnection, which will be further discussed later.

Inverse Ion Energy Dispersion Associated With EMRFs
Figure 4a shows an auroral image captured by the SSUSI instrument aboard DMSP F17.From 09:52 to 09:54 UT, as the satellite traversed the cusp region and moved to lower latitudes (pink curve), a distinct and vibrant auroral spot was observed within the cusp area.Figures 4b and 4c depict the differential electron energy flux and ion energy flux spanning from 30 eV to 30 keV during this event, respectively.Within this interval (red box), the ion energy diminishes in tandem with the reduction in magnetic latitude, a phenomenon referred to as inverse ion energy dispersion.This phenomenon has been elucidated as a sign of lobe reconnection (Hu et al., 2006).Figure 4d illustrates that the predominant direction of the magnetic field aligns sunward, coinciding with the direction of the flow burst.Same as Figure 4a, Figure 4e shows a bright cusp spot observed by DMSP F16 from 10:04 to 10:06 UT.The presence of soft electron deposition and inverse ion energy dispersion characteristics further confirms the occurrence of lobe reconnection.The observation times of the two cusp spots coincide with the occurrence of the second and third equatorward flows as observed by both the SuperDARN and EISCAT radars.This provides additional reinforcement to the evidence that the previously noted EMRFs are sign of lobe reconnection.

Intermittent Lobe Reconnection Reflected by EMRFs
The key feature of lobe reconnection occurring under a northward IMF condition is that it takes place on the high-latitude side of the cusp, where the IMF and the geomagnetic field are approximately anti-parallel, as indicated in Figure 5. Satellites are capable of observing distinct bright auroral cusp spots near magnetic noon resulting from the lobe reconnection (Frey et al., 2002;Phan et al., 2003).Previous studies (Frey et al., 2003;Fuselier, 2021;Fuselier et al., 2000) have suggested that under prolonged stable northward IMF conditions, lobe reconnection could occur continuously or exhibit quasi-stable behavior.However, it is important to note that this description is somewhat rough and imprecise, as it does not clearly define the extent to which lobe reconnection can be considered continuous or quasi-steady.
Auroral cusp spots have been generally accepted as the signature of lobe reconnection.Figure 2 demonstrates a continuous sequence of auroral cusp spots observed from ∼09:00 to ∼12:00 UT.Based on these observations, it seems that we can draw a conclusion that lobe reconnection occurred continuously during this period, which is consistent with the conclusion of Frey et al. (2003).However, here we argue that this conclusion is inaccurate at all. Figure 3 demonstrates that from 09:00 to 10:00 UT, the ESR and Hankansalmi radars observed two intermittent EMRFs.From 10:00 to 11:00 UT, three strong poleward-moving equatorward flow structure were observed in Figure 3. Two of the equatorward flows are also associated with inverse energy dispersion of ions, as shown in Figure 4.These observational features associated with EMRF have been widely accepted as the results of lobe reconnection (Freeman et al., 1993;Hu et al., 2006;Zhang et al., 2012).Therefore, we argue that the presence of 2 intermittent EMRFs and 3 intermittent poleward-moving equatorward flow structures implies 5 intermittent lobe reconnections, namely the lobe reconnection occurs intermittently instead of continuously.
Although the cascade of FTEs may occur during a northward IMF (Kullen et al., 2019) and under stable IMF conditions, it is worth noting that the observed EMRFs during this period were not accompanied by correlated variations in solar wind and IMF conditions.This implies that the intermittent lobe reconnections were not directly driven by changes in solar wind or IMF.In addition, during this period, both ESR and Hankansalmi radars were running continuously with high time resolution, providing reliable observations to demonstrate the intermittent nature of these processes.Based on these observations, we suggest that under prolonged and stable northward IMF conditions when cusp aurora occurs, lobe reconnection should be more accurately described as occurring intermittently rather than continuously.Nevertheless, Figure 3 shows that the duration of each EMRF is ∼20-30 min.If we consider the lobe reconnection within this time scale, specifically shorter than 20 min, it is reasonable to regard that the lobe reconnection is continuous.

Intermittent Lobe Reconnection Reflected by PMRFs
In general, poleward-moving signatures have been observed near the cusp region, such as poleward-moving auroral forms (PMAFs) (Fasel et al., 1994), PMRF (Wild et al., 2001), and poleward-moving geomagnetic impulses (Kokubun et al., 1988).The PMAFs, typically split from the dayside oval boundary and move within minutes poleward, have been generally believed to be caused by FTE (Lockwood & Smith, 1989;Xing et al., 2012).Similarly, PMRFs are often observed together with the PMAF and have also been regarded as signature of low-latitude reconnection (Lockwood et al., 2000;Zhang et al., 2012).These signatures are believed to result from the poleward draping of newly reconnected field lines.However, there is an exception.Feng et al. (2022) suggested that when the IMF has a positive and predominant B Y component, the IMF field lines tend to reconnect with the geomagnetic field lines on the duskside boundary of the cusp.Since these geomagnetic field lines are open field lines with only one end rooted in the ionosphere, this type of reconnection should be referred to as lobe reconnection or high-latitude reconnection.The newly reconnected field lines from this type of reconnection will drape poleward instead of equatorward due to the special field line topology, as indicated in Figure 5.In addition, this type of reconnection can result in the formation of a distinct polar cap arc known as the 15MLT-PCA (Han et al., 2020), which is connected to the auroral oval at around 15:00 MLT and extends toward the polar cap. Figure 3 displays the occurrence of three intermittent PMRFs between 11:00 and 12:00 UT.During this time period, the radar's observation region was approximately at ∼14:00-15:00 MLT and aligned with the duskside edge of the auroral cusp spot, as depicted in Figure 1.After 11:00 UT, the IMF B Y component became predominant, while the IMF B Z component was positive, both of which are favorable for lobe reconnection between the IMF and the geomagnetic field lines on the duskside boundary of the cusp.Hence, we propose that these PMRFs shown in Figure 3 are a result of lobe reconnection occurring between the IMF and the geomagnetic field lines on the duskside boundary of the cusp.It is important to note that these reconnections are also intermittent in nature.

Conclusions
In this study, we have examined the characteristics of lobe reconnection using observations from the DMSP/ SSUSI and ground-based radars.In conjunction with a cusp spot event continuously observed by DMSP/SSUSI under prolonged stable northward IMF conditions, ground-based radars recorded intermittent EMRFs and PMRFs near local noon and postnoon, respectively.Both the EMRFs and PMRFs have been suggested to correspond to lobe reconnections.These findings strongly indicate the intermittent nature of lobe reconnection under prolonged northward IMF conditions, thereby challenging the previously established concept of continuous lobe reconnection under such conditions.However, if we consider this process in a short period, for example, less than 20 min, it may be reasonable to regard it as continuous.

Figure 1 .
Figure 1.A polar projection of the field of views (FOVs) of SuperDARN and ESR from 09:00 to 12:00 UT on 14 December 2012.The black fan-shaped area indicates the FOV of the SuperDARN radar of Hankasalmi.The red straight line is the FOV of EISCAT Svalbard radar, and the red five-pointed star is the position of the EISCAT Svalbard radar.The shaded area illustrates the location of the auroral cusp spot observed by Defense Meteorological Satellite Program.

Figure 2 .
Figure 2. The auroral observations from Defense Meteorological Satellite Program/Special Sensor Ultraviolet Spectrographic Imager and solar wind conditions on 14 December 2012.(a-i) Images of the auroral oval and spot indicated by the red arrows; (j) the variations of interplanetary magnetic field B Z (blue curve), B Y (red curve), and B Z (black curve) are presented in different colors; (k) the black curves represent variations in the clock angle; and (l) the solar wind velocity and dynamic pressure are displayed in red and black curves, respectively.

Figure 3 .
Figure 3. (a-c) Are the range-time plots of the electron density, ion temperature, and ion-line-of-sight velocity (positive away from radar) measured by the ESR between 08:00 and 13:00 UT on 14 December 2012.(d, e) Depict the range-time plots of power and velocity (negative away from radar) observed by beam 7 of the Hankasalmi radar in channel A during the same period.

Figure 4 .
Figure 4.In situ observations from Defense Meteorological Satellite Program (DMSP) F16 and F17.(a, e) A polar projection plot from the DMSP F17 and F16 Special Sensor Ultraviolet Spectrographic Imager instrument of auroral emissions in Lyman-Birge-Hopfield short channel.Magnetic local time (MLT) is also marked from 06:00 to 18:00 MLT.The footprint of the satellite (projected down to ionospheric altitudes) is shown as a pink line, and red dots represent footprints of satellite at different times.The satellite path, showing cusp ion signatures, is just equatorward of the visible cusp brightening.(b, f) Differential electron energy flux.(c, g) Differential ion energy flux.(d, h) Magnetic field recorded by satellite.The time in the red box corresponds to the time between the red dots in the aurora on the left.

Figure 5 .
Figure 5.A schematic illustrating lobe reconnection at local noon producing a cusp spot and equatorward-moving radar form, while lobe reconnection at post-noon may result in a poleward-moving radar form.