Unusual Iridescent Clouds Observed Prior to the 2008 Wenchuan Earthquake and Their Possible Relation to Preseismic Disturbance in the Ionosphere

Abstract

The Wenchuan earthquake (Ms8.0), which struck Sichuan Province, China, on 12 May 2008, was one of the most devastating seismic events in recent Chinese history. It resulted in the deaths of nearly 90,000 people, left millions homeless, and caused widespread destruction of infrastructure across a vast area. In addition to the severe ground shaking and surface rupture, a variety of unusual atmospheric/ionospheric and geophysical phenomena were reported in the days and hours leading up to the earthquake. Notably, iridescent clouds were observed just before the earthquake at three distinct locations approximately 450–550 km northeast of the epicenter. These clouds appeared as fragmented rainbows located beneath the sun and were characterized by their short lifespan, lasting only 1–10 min. Moreover, they exhibited striped patterns within the iridescent regions, suggesting the influence of an external electric field. These features cannot be adequately explained by the well-known meteorological phenomenon of circumhorizontal arcs, raising the possibility of a different origin. The formation mechanism of these clouds remains unclear. In this study, we explore the hypothesis that the iridescent clouds were precursory phenomena associated with the impending earthquake. Specifically, we examine a potential causal relationship between the appearance of these clouds and the geological environment of the earthquake source. We propose a novel model in which electrical disturbances generated along the fault system immediately before the mainshock propagated upward and interacted with the ionosphere, resulting in the creation of a localized electric field. This electric field, in turn, induced electro-optic effects that altered the scattering of sunlight and projected iridescent patterns onto cirrus clouds, leading to the observed phenomena.

1. Introduction

At 14:28:01 local time on 12 May 2008, the Wenchuan earthquake, with a magnitude of Ms8.0 (Mw7.9), struck Sichuan Province, China. This earthquake occurred along the Longmenshan fault zone, which lies at the boundary between the Tibetan Plateau and the Sichuan Basin. The epicenter was located approximately 80 km northwest of Chengdu, the provincial capital of Sichuan, at a depth of 19 km. The earthquake ruptured the fault over a length of more than 240 km, causing surface displacements of several meters [1,2].

Furthermore, a wide range of anomalous phenomena related to the atmosphere, ionosphere, and geophysical processes were reported from as early as 20–30 years prior to the earthquake up to just a few hours before the event. Ma and Wu [3] conducted a comprehensive review of approximately 300 articles published during the three years following the earthquake. According to their review, numerous precursory anomalies were observed, including seismic activity, crustal deformation, changes in strain and stress, structural weakening, gravity, and broadband seismic anomalies; geomagnetic and geothermal variations; and disturbances in the atmosphere and ionosphere. Although the majority of these 300 reports were written in Chinese and remain largely unknown to the international seismological community, the sheer volume of observations reflects the significant attention and institutional effort that China has devoted to earthquake prediction, which is regarded as a national scientific priority.

Among these reports, the most noteworthy are the ionospheric disturbances that were reported to have begun one to three weeks prior to the earthquake [3]. More recently, Heki and colleagues [4] have reported that ionospheric disturbances above the epicentral area of the Wenchuan earthquake began approximately 37 min before the mainshock and reached about 5 percent of the background total electron content (TEC) value. Of particular interest in connection with the increase in ionospheric electron content just before the earthquake are the atmospheric precursor phenomena discussed below.

Namely, between 10 and 60 min before the earthquake, iridescent clouds were observed at three locations approximately 450–550 km northeast of the epicenter, and videos of these clouds spread rapidly on social media, drawing significant attention (e.g., [5,6,7,8]). Some studies have suggested that these clouds may be circumhorizontal arcs caused by the refraction of sunlight through ice crystals in the atmosphere [6]. If this is the case, the phenomenon is merely a meteorological event with no direct connection to seismic activity. Meanwhile, a French research team analyzed data from an ultra-sensitive magnetometer (SQUID) installed in an underground laboratory in southern France and found a sudden magnetic jump in the east–west direction at the exact time corresponding to the timestamp of the YouTube video showing the iridescent clouds [9]. It has been suggested that atmospheric gravity waves generated by seismic motion may have reached the ionosphere, causing disturbances that were detected by the highly sensitive SQUID magnetometer [9]. However, magnetic field fluctuations can also occur due to space weather and other external influences, necessitating careful examination of their origins. Moreover, a clear scientific causal relationship between earthquake-induced ionospheric electromagnetic disturbances and the occurrence of iridescent clouds has not been established.

In this study, we hypothesize that these iridescent clouds were linked to the earthquake and investigate the possibility that electrical coupling occurred between the lithosphere, atmosphere, and ionosphere (referred to as LAI coupling), mediated by subsurface gas emissions from the fault fracture zone just before the earthquake. Consequently, we explore the possibility that an electro-optic effect associated with ionospheric anomalies dispersed sunlight, leading to the formation of iridescent clouds.

2. Materials and Methods

2.1. Video Information of the Iridescent Clouds

The video footage of the iridescent clouds analyzed in this study was recorded 10 to 60 min before the Ms8.0 earthquake that struck Sichuan Province, China, in 2008. These clouds were observed at three locations: Tianshui, Baoji, and Meixian, approximately 450 to 550 km from the epicenter. The spatial relationship between the earthquake source region and the location where the iridescent clouds were observed is shown in Figure 1. The videos were uploaded by their respective photographers to video-sharing platforms such as YouTube, Youku, YouMaker, etc. However, direct access to YouTube is currently restricted in China, and the original sources of these videos on Chinese video-sharing sites remain unclear. Consequently, the identities of those who filmed and uploaded the footage are unknown. Nevertheless, secondary sources supporting the investigation into the cause of these iridescent clouds are available (e.g., [5,10,11]). Additionally, these videos are included in Figure 7 of the article by Waysand et al. [9]. This study inevitably relies on these video sources, which are provided in Appendix A.

2.2. Experimental Material for Electro-Optic Effects

To examine the possibility that the ionosphere disturbed by seismic activity exhibits electro-optic effects, we conducted a solar light dispersion experiment using a KTB (K-Ta-Nb oxide) single crystal developed by Nippon Telegraph and Telephone Corporation (NTT) [12]. This crystal employs Ti as the electrode material, allowing electrons to be injected into the crystal from the cathode. As a result, an electric field gradient proportional to the square root of the distance from the cathode is formed, causing incident light, entering perpendicularly, to disperse according to its wavelength. A schematic diagram illustrating the dispersion of sunlight due to the electro-optic effect using a KTN crystal has already been reported elsewhere [13].

3. Results

3.1. Investigation Results of Iridescent Cloud Footage

On the internet, where footage of iridescent clouds just before the Wenchuan earthquake was introduced, there is an opinion that the iridescent clouds were likely a circumhorizon arc [8]. A typical rainbow is formed by the reflection and refraction of sunlight through raindrops and appears with the sun behind the observer. In contrast, a circumhorizon arc occurs when sunlight is refracted by tiny ice crystals suspended in the atmosphere. It appears in the same direction as the sun as a horizontal band when the solar altitude is 58° or higher. At the time when the iridescent clouds appeared, the solar altitude was approximately 68°, and the phenomenon was observed in the same direction as the sun. Figure 2a shows a circumhorizon arc observed in Nagano Prefecture, Japan, around 11:00 a.m. (L.T.) on 28 April 2019. The solar altitude at that time was 64.7°, and the circumhorizon arc was projected onto thin clouds located 46° below the sun along the horizontal line. A parhelic circle is also visible, with the sun positioned at its center at an altitude of 64.7°, making the elevation angle of the circumhorizon arc 18.7°.

However, there are several doubts about interpreting the iridescent clouds observed before the Wenchuan earthquake as a circumhorizon arc:

• If the phenomenon were a circumhorizon arc, the rainbow-like band should have extended more horizontally. However, the iridescent clouds observed at three locations appeared only as patches in the sky.
• While the iridescent cloud observed in Tianshui is not very distinct, those in Baoji and Meixian exhibit a striped structure, suggesting the presence of an electric field [9].
• The duration of a circumhorizon arc is generally short when meteorological conditions are unstable, such as when clouds obscure the sun. However, based on the videos of the iridescent clouds shown in Appendix A, shadows can be seen, indicating that the weather was sunny and meteorologically stable. Under such weather conditions, a circumhorizon arc can last for up to an hour, but the iridescent clouds shown in Appendix A lasted only for a short duration of about 1 to 10 min.

For these reasons, rather than being a meteorological phenomenon, we hypothesize that these iridescent clouds were caused by a physicochemical phenomenon related to earthquake precursors, as previously reported [13]. A possible factor is the increase in the total electron content (TEC) in the ionosphere above the epicenter, a phenomenon frequently observed before large earthquakes [14,15]. This phenomenon was also pronounced in the case of the 2008 Wenchuan earthquake [4]. That is, when sunlight passes through the ionosphere at an altitude of 300 km, it is hypothesized that some electro-optic effect causes dispersion. Figure 2b shows the iridescent clouds observed in Tianshui [10]. Although Figure 2a,b are not taken from the same perspective, comparing them suggests that the elevation angle of the iridescent cloud in Tianshui was approximately 20°. Assuming that the dispersed solar spectrum in a region of the ionosphere, where some electro-optic effect was transiently generated at an altitude of 300 km, was projected onto cirrus clouds at an altitude of 5 km, the refraction angle α at the dispersion onset altitude (300 km) was estimated based on the geometric relationship shown in Figure 2c, yielding α = 3.5°.

3.2. Electro-Optic Effect Experiment

Due to variations in the refractive index within the KTN crystal caused by the electric field generated by charges that have penetrated the crystal, light undergoes dispersion. Considering the possibility that a similar phenomenon might occur in the ionosphere, we conducted a demonstration experiment to observe the dispersion of sunlight induced by the electro-optic effect of a KTN crystal. Specifically, we mounted a KTN crystal device onto a universal head mount capable of automatically tracking the sun. While tracking sunlight, we varied the voltage applied to the crystal and captured photographs of the dispersed light projected onto a screen (Figure 3a,b). As a result, as shown in Figure 3c–e, clear light dispersion was observed when a voltage of +500 V was applied. The distance between the device and the screen is 260 mm. When the applied voltage is +500 V, the distance from the center of the undeviated light to the center of the dispersed light is ∼16 mm. Therefore, from the relation tan α = 16/260, the refraction angle α is calculated to be 3.5°, which matches the angle estimated in Figure 2c. This finding suggests that if a phenomenon similar to the behavior of electrons in the KTN crystal occurred in the ionosphere just before an earthquake, the electro-optic effect could have contributed to the formation of iridescent clouds.

3.3. Ionospheric Electron Contents Anomalies Before the 2008 Wenchuan Earthquake

As mentioned earlier, the increase in the total electron content (TEC) in the ionosphere, observed just before the 2008 Wenchuan earthquake, began several tens of minutes prior to the event and is referred to as positive TEC (TEC enhancement) [4]. This time period coincides exactly with the period when iridescent clouds appeared. Positive TEC originated in the equatorial anomaly region, where the electron density is high, both north and south of the magnetic equator of the ionosphere [4] (Figure 4a). Figure 5b shows the temporal changes in vertical positive TEC (VTEC) before and after the earthquake at various receiving points [4]. The occurrence times of iridescent clouds at the three locations coincide with the bending points, marked by red circles in Figure 4b, where positive TEC begins at one or more of the receiving points. In other words, it is hypothesized that during this timing, a region similar to the electro-optic effect in a KTN crystal may have formed near the earthquake’s epicenter in the ionosphere.

4. Discussion

4.1. Geological Environment Inducing Ionospheric Anomalies

The 2008 Wenchuan earthquake occurred along the Longmenshan fault zone, which forms the boundary between the Tibetan Plateau and the Sichuan Basin. After the earthquake, the release of gases such as CH₄, CO₂, and O₂, as well as CO derived from C-H-O fluids and the radioactive element ²²²Rn, was measured from this fault fracture zone [16,17,18]. Although these subterranean gas emissions were observed after the earthquake, it is considered that, just before the earthquake, during the quasi-static nucleation process in which microcracks within the fault began to coalesce, gases migrated from deeper regions through the crack openings, accompanied by electromagnetic changes.

Several mechanisms have been proposed to explain these electrical anomalies. These include the emanation and decay of ²²²Rn [19], charge migration due to stress acting on the crust [20], atmospheric gravity waves [21], electrical interactions between rock fracturing and subterranean gases [13,22,23], and electrostatic coupling mediated by supercritical water [24]. These processes are thought to have induced electrical fluctuations at the Earth’s surface, which propagated through the atmosphere and eventually disturbed the ionosphere, leading to changes in TEC enhancement. There are multiple hypotheses regarding how the lithosphere, atmosphere, and ionosphere (LAI) are electrically coupled. The main theories, as illustrated in Figure 5a and Figure 6c, include the formation of conductive pathways in the atmosphere [20,25,26,27], electromagnetic induction coupling between the lithosphere and ionosphere [13,23], and capacitive coupling [24].

Figure 5. Geological environment model of the Longmenshan fault zone [28]. Partially modified and LAI coupling models: (a) electric conduction path model [20,25,26,27]; (b) magnetic induction coupling model [22]; (c) capacitive coupling model [24].

Figure 5. Geological environment model of the Longmenshan fault zone [28]. Partially modified and LAI coupling models: (a) electric conduction path model [20,25,26,27]; (b) magnetic induction coupling model [22]; (c) capacitive coupling model [24].

While this study does not determine which LAI coupling mechanism is dominant, the Longmenshan fault zone, which triggered the Wenchuan earthquake, exhibited gas emissions such as ²²²Rn, methane, and CO [16,17,18]. This suggests that it possessed the necessary physicochemical potential to facilitate LAI coupling, regardless of the specific mechanism involved.

4.2. Iridescent Cloud Phenomenon Associated with Earthquakes

The iridescent clouds photographed just before the Wenchuan Earthquake in May 2008 attracted attention as a rare atmospheric optical phenomenon. The iridescent clouds associated with the Wenchuan Earthquake were recorded on video from multiple locations. The footage reveals aspects that are difficult to explain by meteorological conditions alone, suggesting that this phenomenon should be considered as a potential natural precursor to earthquakes. Given the volume of information available, it is insufficient to dismiss these occurrences as mere coincidences, and they deserve careful investigation.

Interestingly, similar phenomena can be found in historical earthquake records in Japan. Specifically, approximately one to two hours before the magnitude 7.9 Tonankai Earthquake, which struck at 13:35 (L.T.) on 7 December 1944 (epicenter: 136°11′ E, 33°34′ N; focal depth approximately 40 km), an eyewitness account was reported from Shimizu Ward in Shizuoka City, located about 210 km northeast of the epicenter. The account reads: “Around noon that day, there was no wind, the air felt stagnant, and spectral bands of sunlight could be seen through the thin clouds. I don’t clearly remember how long it lasted, but there were two or three small areas where it appeared” [29].

Although these records suggest the possibility that iridescent clouds may appear as precursors to earthquakes, they often lack the detailed information necessary for scientific verification. As a result, they have long been excluded from research as so-called “macroscopic anomalous precursor phenomena”. Nevertheless, in the case of the 2008 Wenchuan Earthquake—repeating what has been noted above—multiple video recordings provide detailed data, making these phenomena worthy of scientific investigation. The hypothesis proposed in this paper is that some kind of electro-optical effect in the ionosphere may have caused the dispersion of sunlight. This hypothesis is examined in the next section.

4.3. Hypothesis: Verification: Ionospheric Electro-Optic Effect

Dielectric crystals exhibiting the electro-optic effect can be classified into two types: the first-order electro-optic (Pockels) effect, in which the refractive index changes linearly with the applied electric field, and the second-order electro-optic (Kerr) effect, where the refractive index varies in proportion to the square of the electric field. In the demonstration experiment on solar light dispersion, the KTN crystal used exhibited the second-order electro-optic effect [12]. This occurred because, when titanium was used as the electrode material, electrons were injected from the cathode, forming a potential gradient proportional to the square root of the distance from the cathode between the electrodes [12].

The electro-optic effect is not limited to solid materials; it is also known to occur significantly in the electric double layer formed at the solid–liquid interface in electrolyte solutions, manifesting as a prominent first-order electro-optic effect [30]. In this case, when positive charges are fixed to the solid surface, the electric field of the electric double layer reaches its maximum at the solid surface and exponentially decays toward the diffuse layer in the liquid.

Can the electro-optic effect manifest in the ionosphere? In the ionosphere, the electric fields that cause variations in TEC are generally weak, and the ionosphere is a physical system vastly different from dielectric crystals. Therefore, it is unlikely that primary or secondary electro-optic effects would occur directly. However, to explore this possibility, it is necessary to consider the conditions under which a non-centrosymmetric plasma structure could temporarily form within the ionosphere, as well as situations where locally strong electric fields might arise.

A noteworthy phenomenon in this context is the equatorial anomaly, as shown in Figure 5a. This anomaly results from the upward ExB drift during daytime and diffusion along magnetic field lines, forming regions of high electron density at low latitudes in both hemispheres, away from the magnetic equator. The occurrence of TEC variations in these equatorial anomaly regions suggests the possible involvement of locally strong electric field formation.

Furthermore, in the temporal variation in positive VTEC shown in Figure 4b, the time at which a bending in the curve occurred closely coincided with the time of iridescent cloud formation. Given this fact, it might be difficult to dismiss the possibility that an electro-optic effect similar to conventional ones may have manifested as a result of electric field variations during TEC increases in conjunction with interactions with the Earth’s magnetic field.

Additionally, three-dimensional analyses of TEC variations just before earthquakes have shown that a region of negative TEC (TEC depletion) tends to appear slightly higher and to the north of a positive TEC region [31,32]. Since TEC anomalies are thought to accompany electron transport within the ionosphere, a similar negative TEC region was likely to have appeared adjacent to the northward upper side of the positive TEC region during the 2008 Wenchuan earthquake, as illustrated in Figure 6. If this was the case, an electric field may have temporarily formed between the negative and positive TEC regions within the equatorial anomaly, oriented perpendicular to the sunlight, potentially leading to an electro-optic effect. This effect could have contributed to the appearance of iridescent clouds. However, further examination is necessary regarding the intensity and distribution of the electric field.

Elucidating the occurrence of such anomalous electro-optic phenomena in the ionosphere could offer a new perspective in ionospheric physics and space weather research. Future studies in this area are anticipated to provide further insights.

5. Conclusions

In this study, we examined the causal relationship between iridescent clouds observed at three locations 450–550 km northeast of the epicenter just before the 2008 Wenchuan earthquake and seismic activity based on the hypothesis that these clouds were related to the earthquake. Around the same time as the appearance of the iridescent clouds, fluctuations in positive TEC were observed in the equatorial anomaly region above the epicenter.

We proposed a model in which a transient electric field, oriented perpendicular to sunlight, was generated between the positive TEC structure within the equatorial anomaly region and the negative TEC region that likely formed at a slightly higher altitude to its north. This electric field could have caused anomalous dispersion due to the electro-optic effect. However, further verification is needed to determine whether the intensity of this electric field was sufficient to create a non-centrosymmetric plasma structure capable of inducing the electro-optic effect. Future research advancing the understanding of unusual electro-optic phenomena in the ionosphere is expected to contribute to new developments in ionospheric physics and space weather studies.