Abstract

In this study, solar ultraviolet (UV) radiation data collected in Riyadh, Saudi Arabia, between 2015 and 2022 were analyzed to explore quasi-periodicities in the UV time series. The power spectrum density analysis revealed several local peaks that exceeded the 95% confidence interval. These peaks included periodicities of 483–490 days, 272 days, 157−162 days, 103−110 days, 64–72 days, 27 days, and 13 days. To investigate the potential influence of space weather parameters on UV radiation, data on cosmic rays, solar radio flux at 10.7 cm (F10.7 cm), the Kp index, and solar wind speed for the same time period were examined. The aim was to identify periodicities in these variables that aligned with those found in the UV radiation data. The analysis reveals that several periodicities observed in the UV radiation spectrum are also present in the spectra of the considered parameters. Prominent periodicities include a 270-day cycle in UV radiation and cosmic rays, as well as periodicities of 72 days, 27 days, and 13 days in all considered variables. Furthermore, 110-day peaks are observed in spectrum of the UV radiation, the Kp index, solar radio flux F10.7, and solar wind speed. Notably, consistent peaks at 157-day periodicity are identified in the UV spectrum, also present in the spectra of all the considered variables (cosmic rays ∼162 days, Kp index ∼162 days, solar radio flux ∼156 days, and solar wind speed ∼163 days). The identification of common periodicities between UV radiation and space weather parameters in this study provides compelling evidence of a potential direct or indirect influence of solar variations on UV radiation. This finding significantly enhances our understanding of the impact of extraterrestrial factors, particularly solar activity, on the Earth’s environment.

1. Introduction

Ultraviolet (UV) radiation, with wavelengths ranging from 200 to 400 nm, makes up a small fraction of the solar radiation reaching the Earth’s upper atmosphere. Despite its relative scarcity, UV radiation plays a crucial role in various environmental, atmospheric, biological, and industrial contexts [13]. One important aspect of UV radiation is its impact on life on Earth. Solar UV radiation serves as the primary source of vitamin D for humans, offering numerous health benefits [4, 5]. Additionally, solar UV radiation acts as the main energy source for the atmosphere, contributing significantly to its vertical, thermal, and electronic structure, as well as atmospheric chemistry (e.g., [68]).

UV radiation is influenced by a range of factors, including astronomical, environmental, and atmospheric conditions. Astronomical factors, such as seasonal variations arising from the solar zenith angle, Earth-Sun distance, and altitude, have distinct effects on UV radiation levels. For instance, during summer in the Northern Hemisphere, when sunlight strikes the Earth more directly due to the Earth’s axial tilt, UV radiation levels are typically higher. Conversely, during winter months, when sunlight is less direct, UV radiation levels tend to be lower (e.g., [9, 10]). The position of the Sun in the sky, as indicated by the solar zenith angle, also influences UV radiation. Higher solar zenith angles, such as those occurring during winter in the Northern Hemisphere, result in the Sun’s rays passing through a greater distance of the Earth’s atmosphere before reaching the surface. This increased atmospheric path leads to greater scattering and absorption of UV radiation, resulting in lower UV radiation levels at the Earth’s surface (e.g., [9, 11]).

In addition to astronomical factors, the transmission of UV radiation through the Earth’s atmosphere is affected by complex scattering and absorption processes by atmospheric gases, clouds, and aerosols [1218]. The absorption of UV radiation by atmospheric constituents such as ozone, oxygen, and water vapor leads to atmospheric heating, which affects temperature profiles and influences circulation patterns, including the formation of high and low-pressure systems, jet streams, and the movement of weather systems (e.g., [14, 15, 18, 19]).

For a considerable time, scientists have been devoted to investigating the relationship between solar activity and its impact on Earth’s climate (e.g., [2023]).

The Sun serves as the primary energy source for Earth’s climate system, and changes in solar activity can have a wide range of effects on our planet’s climate. Solar activity encompasses various processes and phenomena occurring on the Sun, including modifications in its magnetic field, sunspot activity, solar flares, and coronal mass ejections.

Transient solar activity refers to abrupt and unpredictable events like solar flares and coronal mass ejections, which have the potential to generate geomagnetic storms. Solar events can have significant consequences on Earth’s magnetosphere, leading to disturbances such as geomagnetic storms and disruptions in the upper atmosphere, particularly the ionosphere. These disturbances pose potential hazards to various terrestrial processes, technologies, and human health (see [24, 25] and references therein).

On the other hand, periodic solar activity involves regular variations in solar output, such as changes in the number of sunspots, the interplanetary magnetic field, and the amount of solar radiation reaching Earth [26]. To quantify and assess solar activity, scientists employ various indices and parameters. The sunspot number is a commonly used index that quantifies the abundance of sunspots on the solar surface. Other parameters, such as solar wind speed, interplanetary magnetic field strength, and F10.7 cm radio flux, can also be utilized to evaluate solar activity. In addition to solar indices, scientists utilize several geophysical parameters to study the effects of solar activity on Earth’s environment. Parameters like the Kp index, Dst index, and aa index serve as examples. These parameters measure variations in Earth’s magnetic field and indicate the level of geomagnetic activity caused by interactions with the Sun.

Extensive scientific research has focused on examining the periodic patterns observed in solar data, uncovering both long-term and short-term cycles. The long-term cycle, which lasts roughly 11 years, is closely linked to solar magnetic activity and is characterized by fluctuations in sunspots and other solar phenomena. On the other hand, the short-term cycle, spanning approximately 27 days, arises from the modulation of solar features due to the Sun’s rotation, resulting in periodic changes in the appearance and behavior of solar structures. In addition to these well-established long and short-term cycles, scientists have identified several quasi-periodic patterns in solar and cosmic ray data. These quasi-periodicities, including durations of approximately 4.2 years, 2 years, 1.7 years, 1.3 years, 157 days, 127 days, 78 days, 57 days, and 27 days, have been established in the spectra of solar activity parameters by several investigators [2740]. It is worth noting that these periodic patterns may display variations depending on the specific dataset and the time interval being analyzed. For instance, Chowdhury et al. [41] utilized the wavelet power technique to analyze cosmic rays, sunspot number, the coronal green line, and the 10.7 cm solar radio flux for the period between 1996 and 2003. They identified periodicities ranging from 16–30 days to 240–260 days. Singh and Badruddin [40] employed wavelet analysis techniques on sunspot number, 10.7 cm solar radio flux, Ap index, and cosmic ray data from 1968 to 2014, identifying periodicities including 1.3 years, 1.7 years, 27 days, 13.5 days, 9.0 days, 51 days, 76 days, and 101 days. Maghrabi et al. [42] analyzed cosmic ray data collected using the KACST muon detector and identified periodicities at various frequency scales, including approximately 1.3 years, 290 days, 185 days, 153 days, 65–73 days, 45−53 days, 25–27 days, and 13 days.

While anthropogenic greenhouse gases (GHGs) and natural phenomena such as El Niño-Southern Oscillation (ENSO) and volcanic activity are widely recognized as the primary drivers of climate, solar activity also exerts an influence on climate [21, 4345]. These alterations involve variations in solar radiative output, such as ultraviolet (UV) radiation and total solar irradiance (TSI), as well as fluctuations in the solar wind and interplanetary magnetic fields. The Sun’s magnetic field, for instance, plays a critical role in regulating the flow of cosmic rays, and variations in solar activity directly or indirectly influence atmospheric ionization. Consequently, these changes in ionization levels result in modifications to atmospheric chemistry, affecting the composition and behavior of gases and aerosols in the atmosphere (e.g., [22, 26]). Furthermore, solar activity can influence various atmospheric processes, including temperature distribution, circulation patterns, and cloud formation. These atmospheric dynamics, in turn, have broader implications for weather patterns and long-term climate trends [46]. Therefore, comprehending the complex relationships between solar activity and Earth’s climate is essential for understanding the effects of solar variability on our planet.

Numerous extensive research studies have been conducted to explore the connections between meteorological and atmospheric parameters, such as temperature, pressure, humidity, cloud cover, ozone concentrations, and rainfall, and variables related to solar activity. These investigations have utilized different analytical methods, including correlation analysis and power spectral analysis, to investigate and comprehend the relationships between these variables [50, 51]; [49]; [22, 43, 5056]. For instance, Lean et al. [57] discovered strong correlation (R = 0.86) between Northern Hemisphere summer temperature anomalies and a reconstruction of solar irradiance based on faculae and sunspots from 1610 to 1800. Moreover Svensmark and Christensen [49] found a correlation between cosmic ray flux and global cloud coverage, indicating that cosmic rays may play a role in cloud formation. Mendoza et al. [58] examined minimum extreme temperature variability from five meteorological stations in the central part of Mexico from ∼1920 to ∼1990 and found moderate correlation (R = 0.65) between these temperature records and geomagnetic activity. Laurenz et al. [59] studied the influence of the changes in solar activity on rainfall over Europe and found that February rainfall has the strongest relationship with solar activity in Western and Central Europe. Nazari-Sharabian and Karakouzian [60] utilized cross wavelet transform to explore the periodicity between sunspot number (SSN) and Iran’s annual precipitation (1950–2018) and discovered an 8–12-year periodicity in rainfall that aligns with SSN periodicities. Mostafa et al. [61] used the rainfall data from 19 stations in Sudan for the period 1910–2018 and confirmed a negative correlation between rainfall and the SSN over certain stations. Maghrabi et al. [62] conducted a study examining the periodicities of downward longwave atmospheric radiation data within the wavelength range of 4–100 μm. The investigation explored the relationship between these periodicities and various solar and interplanetary parameters during the time span of 2014–2020. The findings of the study showed that several common periodicities were observed across multiple variables. Specifically, the periodicities of 154–157 days, 25–27 days, and 21 days were identified in the downward longwave atmospheric radiation, as well as in cosmic rays, solar radio flux, Dst index, and solar wind speed.

Despite the acknowledged significance of ultraviolet (UV) radiation in influencing Earth’s climate and atmospheric conditions, the existing body of research concerning the cyclic variations of UV radiation and its association with solar activity has been relatively limited [6366]. Investigations in this area have the potential to provide valuable insights into the cyclic variations of UV radiation and their correlation with solar activity. However, additional research is required to enhance our comprehension of these relationships and their implications for Earth’s climate and atmospheric dynamics. To address the current gaps, the objective of this research is to examine the quasi-periodicities present in UV radiation data obtained from Riyadh, Saudi Arabia, during the period 2015 to 2022 and to explore potential shared periodicities between UV radiation and solar and interplanetary data.

2. Data and Methodology

2.1. Experimental Data

The daily mean data of solar UV radiation collected at the King Abdulaziz City for Science and Technology (KACST) campus in Riyadh (latitude of 24.43, longitude of 46.40, and altitude of 613), Saudi Arabia, during the period from 2015 to 2022 were employed for the purpose of this study.

The UV radiation measurements were conducted using a SKU421 UV sensor provided by Skye Instruments (Skye Instruments, 2015). This sensor has a waveband with a spectral range of measurements between 315 and 400 nm. To ensure the quality of the collected data, the UV data were limited to solar elevation angles greater than 10°, as the cosine law is only valid for solar elevation angles greater than 10° [10]. Additionally, the UV radiation was excluded if its value exceeded the corresponding extraterrestrial radiations (UV0) at the site of observation, as this indicates errors in the measurement or calibration of the instrument (e.g., [67, 68]). The extraterrestrial radiation of UV radiation was calculated using the standard formula as described in previous studies (e.g., [9, 17, 67]).

Furthermore, daily mean cosmic ray data for the same period as the UV measurements were utilized from KACST muon detector located at the same site as the UV sensor. The technical specifications and the calibration procedures of this detector are discussed in several publications (e.g., [3, 69, 70]). Basically, the detector consists of four sheets of an NE10-type plastic scintillator and a Hamamatsu Photomultiplier Tube (PMT) housed in a light-tight container. Cosmic ray muons passing through the detector excite the scintillator material, which emits fluorescent light. This light is detected by the PMT and converted into an electrical signal. The signals are then amplified, digitized, and recorded. The muon detector has been in operation since 2002 for cosmic ray and atmospheric research activities. However, due to technical and calibration procedures, the detector did not have sufficient data for the period between mid-2014 and early 2015. Therefore, the study focused on analyzing cosmic ray data from the period starting from mid-2015 until 2022.

The daily average space weather data for the study period, including the solar radio flux at 10.7 cm (F10.7 cm), solar wind speed (SWS), and Kp index, were obtained from the National Oceanic Atmospheric Administration, the National Geophysical Data Center, USA, and the National Space Weather Prediction Center, USA.

2.2. Methodology

Various statistical procedures have been employed by researchers to investigate the periodicities of time series variables. These procedures include discrete and fast Fourier transforms, spherical harmonic decomposition, wavelet transform analysis, periodogram analysis, passband filters (such as time-smoothing), maximum entropy method, and the Huang–Hilbert transform (e.g., [3, 38, 71, 76], and references therein). Although these methods may differ in their calculation approaches, they are consistent within the accuracy of the processing.

In this study, the daily measurements of the variables under investigation were subjected to analysis using the fast Fourier transform (FFT) technique. The Welch window method was employed to obtain the power spectra of the variables. The AutoSignal© software version 1.7 was utilized for the calculations, as it provides accurate tools for time series analysis and power spectral investigations. This software incorporates all the necessary functions and procedures required for precise calculations.

To address data gaps, various approaches were employed. For small gaps, the missing values were extrapolated from neighboring days. In cases of unevenly spaced data, a technique was employed that accounted for the spectra of data with slightly longer gaps. This methodology enabled the identification of dominant periodicities across the frequency spectrum that may have implications for other periodicities.

To uncover local quasi-periodicities near the prominent peaks, the minimum variance technique, as described by Kay [77], was employed. The power spectral density (PSD) typically decreases with frequency, so multiple frequency windows were selected around these peaks, and the data were detrended. Fourier filtering was then applied within the chosen frequency range to reconstruct the data. The minimum-variance technique was subsequently used to derive the PSD for the region surrounding the prominent peaks. This approach aided in the detection of local quasi-periodicities that might have been masked by the dominant peaks (e.g., [37, 38]).

The obtained peaks were evaluated for significance using the peak-based critical limit significance level method. These confidence limits allow for the assessment of the significance of the largest spectral component and can help reject the null hypothesis (e.g., [78]), which assumes either a white noise signal (AR(1) = 0.0) or a red noise signal (AR(1) > 0.0). In this study, to assess the significance of spectral peaks, a red-noise spectrum was generated for the UV and other time series. The analysis primarily focused on considering a 99% confidence level above the red-noise baseline. Spectral peaks that exceed the 99% confidence interval threshold for red noise are considered to be statistically significant. These peaks stand out from the red‐noise model with a high degree of certainty, with a 99% confidence level suggesting that their occurrence is not due to random fluctuations.

3. Results and Discussion

Figure 1 illustrates the time series of the daily mean values for the UV radiation and other variables considered throughout the study period. The UV radiation exhibits noticeable seasonal variations, characterized by a peak during summer and a minimum during winter. Over this period, the average UV value was 4.92 ± 1.50 W/m2, with the minimum and maximum values recorded as 0.67 W/m2 and 8.83 W/m2, respectively.

Similarly, the cosmic rays also exhibit seasonal variations, influenced by complex atmospheric effects affecting the muon measurements [69]; [75]. During this period, the average cosmic ray count rate was 160.22 ± 20 counts/s, ranging from 153.48 to 166.67 counts/s.

Considering that the data cover the declining phase of solar cycle 24 and the early years of the ascending phase of solar cycle 25, it is evident that solar activity parameters display distinct patterns. This is particularly evident in the F10.7 cm index, with partial indications observed in the Kp index and solar wind speed (SWS) time series data. The mean F10.7 value was 90 ± 20 sfu, with minimum and maximum values of 63.40 sfu and 211 sfu, respectively. The SWS ranged from a minimum of 269.13 km/s to a maximum of 752.96 km/s, with an average value of 430 ± 90 km/s. Lastly, the Kp index displayed an average value of 20 ± 11 nT, with a maximum value of 60.86 nT and a minimum value of zero.

Figure 2 displays the periodogram, which offers insights into the prominent peaks observed in the UV radiation data analyzed throughout the study period. The power is quantified using the unit (mag2/n/var), where mag2 represents the squared magnitude of the spectrum normalized for the spectrum size at a specific frequency. The analysis encompasses a dataset of size (n) and incorporates the variance (var) of the data series [62]. Evidently, several significant signals with different amplitudes are found in the spectrum. The strongest peak was 272 days, and other significant peaks, such as the 483–490 days (1.3 yr.) and 196–210 days, were also found across the spectrum. It can be clearly seen that several periodicities existed in the spectra that are masked by the effect of the strong magnitude of the 272-day​ peak.

To reveal the other periodicities in the spectrum of UV radiation data, a frequency filter using the technique of minimum variance was applied [34, 51, 62, 77, 7679].

Figure 3 depicts a periodogram presenting the periodicities of the UV radiation throughout the study period for frequencies below 0.005 1/day. The application of frequency filters enhances certain periodicities at specific frequencies, surpassing the 95% confidence limit. Notable enhanced periodicities include 157–162 days, 103–107 days, 64–72 days, 27 days, and 13 days.

In order to investigate potential similarities in periodicities between the UV spectrum and other variables, power spectral procedures were applied to cosmic rays (CRs) and three solar activity parameters using data from the same time period as the UV data. The aim was to identify any potential matches or overlaps in the periodic patterns observed in the UV radiation and these variables. The power spectral analysis included the utilization of the minimum variance technique, as described by Kay [77]; Kudela et al. [79]; and Joshi [77]. Figure 4 displays selected periodograms that reveal some significant peaks associated with the analyzed variables. These periodograms visually represent the detected periodic patterns and provide evidence for potential correlations or shared periodicities between UV radiation, CRs, and the solar activity parameters.

The cosmic ray data spectrum exhibits prominent peaks at 1.7 years, 1.3 years (with confidence interval CI = 95%), 270 days, 162 days, and 90 days. In the spectrum of the Kp index, several peaks are observed, including 162 days, 128 days, 90 days, 36−27 days, and 13 days. Similarly, the spectrum of the solar radio flux F10.7 displays significant peaks at 1.3 years, 265 days 190 days, 156 days, 110 days, 70 days, 42 days, 27 days, and 13 days. Peaks at 270 days, 163 days, 127 days, 110 days, 21–27 days, and 13 days are identified in the solar wind speed spectrum.

The analysis of the current study reveals that several periodicities observed in the UV radiation spectrum also appear in the spectra of the other considered variables. The 270-day cycle is a prominent periodicity found in both the UV radiation and cosmic ray data. Additionally, periodicities of 72 days, 27 days, and 13 days are identified in the UV radiation spectrum, as well as in the spectra of all the considered variables. The solar wind spectrum also exhibits peaks at 270 days, 163 days, 127 days, 27 days, and 17−13 days, which align with the periodicities observed in the UV radiation spectrum. The 157-day periodicity identified in the UV spectrum is also found in the spectra all the considered variables (cosmic rays ∼158 days, Kp index ∼161 days, solar radio flux ∼156 days, and solar wind ∼163 days).

The periodicities identified in the UV radiation spectrum in this study are consistent with previous findings reported by various investigators (e.g., [51, 6466, 80, 81]). For instance, Pap et al. [80] analyzed data from the Solar Backscatter Ultraviolet experiment on the Nimbus-7 satellite for the period from 1980 to 1990. They identified periodicities of 51 days and 150–157 days in total and UV irradiances, F10.7 cm radio flux, plage index, and sunspot blocking function. Alamodi and Abdelbasset [63] analyzed daily UV radiation data from four stations in Egypt and found dominant waves with periods of approximately 13, 13.4, 13.4, and 13.5 months for the respective stations.

The presence of similar periodicities in the spectra of cosmic rays (CRs) and solar activity parameters is a noteworthy finding, as it aligns with previous reports by various researchers. This consistency can be attributed to the influence of solar disturbances on the flux of primary cosmic rays, which subsequently impacts atmospheric ionization levels. Changes in atmospheric ionization, in turn, have the potential to affect the physical and chemical properties of the atmosphere. For instance, variations in ionization can alter cloud formation rates, precipitation patterns, water vapor content, and the presence of atmospheric aerosols [46, 51, 82, 83].

Periodicities have been observed in solar, cosmic ray, and geophysical data, with some having clear explanations while others remain enigmatic. The 13-day and 27-day periodicities directly correspond to the solar rotation cycle, representing half and full rotations, respectively [26, 84, 85]. These periodicities are evident in UV observations due to their association with solar activity.

Subharmonics of the 27-day period, such as 72 days and 64–67 days, have been identified in the UV radiation spectrum, cosmic ray spectra, solar radio flux F10.7, and the Kp index [27]; [80]. These subharmonics, along with other periods like 51 days, 78 days, 104 days, and 129 days, are thought to be linked to increased flare activity along specific longitude bands [86]. A pronounced peak with a period of about 1.3 years consistently appears in the UV radiation spectrum, as well as in the solar radio flux and cosmic ray spectra. This periodicity is believed to be associated with the solar rotation rate near the base of the convection layer, indicating its fundamental nature in relation to the solar dynamo [87]. A period of approximately 150–160 days, considered the third harmonic of the 1.3-year period, has been identified in the UV spectrum, cosmic rays, the Kp index, solar radio flux, and solar wind speed [88].

The UV spectrum exhibits unique periodicities of 196–210 days, which correspond to the seventh and eighth harmonics of the 27-day solar synodic rotation. These periodicities are specific to the UV spectrum and are not consistently observed in other solar activity parameters. Additionally, a peak around 272 days has been detected in both the UV spectrum and solar wind data, which is close to the 260-day period found in other solar activity parameters like the solar radio flux and cosmic rays [34, 41]. The 260-day period may be associated with changes in active solar regions related to the boundaries of coronal holes.

While some periodicities align with previous research, there are variations in others, such as the 272-day period. These differences may be attributed to medium to short-term variations during different epochs, instrument uncertainties, and variations in analysis techniques.

4. Conclusions

In conclusion, the study investigated the quasi-periodicities in the daily averages of UV radiation data recorded in Riyadh, Saudi Arabia, from 2015 to 2022. The time series analysis revealed significant peaks at various periods, including 490−483 days, 272 days, 157–162 days, 103–107 days, 64–72 days, 27 days, and 13 days. These peaks exceeded the 95% confidence limit, indicating their statistical significance.

Furthermore, the analysis conducted in this study reveals that several periodicities observed in the UV radiation spectrum are also apparent in the spectra of cosmic rays, solar radio flux F10.7, the Kp index, and solar wind. The cycle of 270 days stands out as a prominent periodicity observed in both the UV radiation and cosmic ray data. Additionally, periodicities of 72 days, 64–67 days, 27 days, and 13 days are identified not only in the UV radiation spectrum but also in the spectra of cosmic rays, solar radio flux F10.7, and the Kp index. Moreover, peaks at 13 days are consistently observed in the UV radiation, the Kp index, and solar radio flux F10.7. The solar wind spectrum also displays peaks at 270 days, 163 days, and 17−13 days, aligning with the periodicities found in UV radiation. Notably, a consistent peak at a 157-day periodicity is identified in the UV spectrum, which is also present in the spectra of all the considered variables (cosmic rays ∼158 days, Kp index ∼161 days, solar radio flux ∼156 days, and solar wind ∼163 days). These observed periodicities in UV radiation are consistent with previous findings in solar, interplanetary, and cosmic ray parameters. However, it is crucial to acknowledge that these findings represent preliminary insights, and further research is necessary to delve into the underlying causes and relationships associated with these observed periodicities. This ongoing investigation will contribute to a more comprehensive understanding of the complexities involved in UV radiation and its broader impact [8587].

Data Availability

The data are the property of King Abdulaziz City for Science and Technology; the author is not authorized to distribute the data without obtaining consent from the aforementioned organization.

Conflicts of Interest

The author declares that there are no conflicts of interest.

Acknowledgments

The author would like to thank the King Abdulaziz City for Science and Technology (KACST) for supporting this work. The solar indices, F10.7 cm, Dst index, and solar wind speed, were obtained from the GSFC/SPDF OMNIWeb interface at https://omniweb.gsfc.nasa.gov. The author acknowledges the SPDF OMNIWeb database as the source of data used in this research study. The cosmic ray data for Oulu neutron monitor were obtained from http://cosmicrays.oulu.fi/. The author would also like to thank the people of this station who made the data available.