Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia’s provinces

Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia’s provinces | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia’s provinces Fahad Almutlaq This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9418580/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract This study investigates the spatiotemporal dynamics of COVID-19 across Saudi Arabia’s administrative provinces from March 2020 to August 2022 using an integrated spatial autocorrelation and hotspot analysis approach. Spatiotemporal patterns were examined using Global and Local Moran’s I, Getis-Ord Gi* hotspot analysis, and clustering techniques to identify statistically significant spatial clusters, coldspots, and outliers, as well as their temporal evolution across distinct phases of the pandemic. The results reveal that COVID-19 transmission in Saudi Arabia was highly non-random and characterized by strong spatial dependence. Persistent hotspots were consistently identified in the Riyadh Region and the Eastern Province, reflecting the influence of population density, economic activity, and mobility networks. In contrast, southwestern provinces such as Asir, Jazan, and Al-Baha repeatedly emerged as coldspots, suggesting that geographic isolation and lower population density limited widespread transmission. The study demonstrates the effectiveness of integrating spatiotemporal analysis with spatial autocorrelation methods for understanding pandemic dynamics. The proposed framework provides valuable insights for identifying high-risk areas, optimizing resource allocation, and supporting spatially targeted interventions. This approach offers a transferable model for enhancing epidemic surveillance and preparedness in Saudi Arabia and similar geographic contexts. Health sciences/Diseases Physical sciences/Mathematics and computing Spatial Autocorrelation Anselin local Moran’s I Hot Spot Analysis COVID-19 Saudi Arabia Public Health Intervention Figures Figure 1 Figure 2 Figure 3 1. Introduction The coronavirus disease 2019 (COVID-19) pandemic has posed unprecedented challenges to global public health systems, economies, and social structures. Since its emergence in late 2019, COVID-19 has spread rapidly across countries and regions. This spread has demonstrated complex transmission dynamics shaped by human mobility, population density, socioeconomic conditions, and public health interventions [ 1 , 2 ]. Understanding how the virus propagates across space and time is therefore essential for designing effective containment strategies, allocating healthcare resources, and enhancing preparedness for future pandemics. Traditional epidemiological analyses often focus on temporal trends in case numbers or aggregate national-level statistics, which may obscure important geographic heterogeneity in disease transmission. However, infectious diseases such as COVID-19 rarely spread randomly; instead, they exhibit spatial dependence, whereby neighboring regions tend to experience similar infection patterns due to shared mobility networks, environmental conditions, and demographic characteristics [ 3 , 4 ]. Spatiotemporal analysis, combined with spatial autocorrelation techniques, offers a powerful framework for capturing these patterns by simultaneously examining where infections occur, how they cluster geographically, and how these clusters evolve over time [ 5 , 6 ]. A growing body of international research has demonstrated the effectiveness of spatial epidemiology tools—such as Moran’s I, Local Indicators of Spatial Association (LISA), and Getis-Ord Gi* hotspot analysis—in identifying COVID-19 clusters, transmission corridors, and high-risk regions [ 7 , 8 ]. These studies consistently show that metropolitan and highly connected regions act as primary hubs of viral transmission, while less populated or geographically isolated areas often experience lower incidence rates. Moreover, the intensity and configuration of spatial clustering have been shown to vary across epidemic phases, influenced by public health interventions, behavioral changes, and the emergence of new variants [ 9 , 10 ]. In Saudi Arabia, the spatial dynamics of COVID-19 warrant particular attention due to the country’s unique geographic, demographic, and socioeconomic characteristics. The Kingdom encompasses highly urbanized metropolitan centers, vast desert regions, and provinces with pronounced disparities in population density, healthcare access, and mobility patterns. Additionally, Saudi Arabia’s role as a global destination for religious pilgrimage introduces episodic surges in domestic and international movement, which can significantly influence disease transmission dynamics [ 11 , 12 ]. While several studies have examined COVID-19 trends in Saudi Arabia, many have focused on either spatial or temporal dimensions independently, limiting their ability to capture dynamic provincial-level transmission patterns [ 13 ]. These investigations are anchored in existing literature, including the works of [ 14 , 15 , 16 ], providing a comprehensive understanding of the pandemic at both local and global scales. This study addresses these limitations by applying an integrated spatiotemporal analysis framework incorporating spatial autocorrelation, hotspot detection, and clustering techniques to investigate the evolution of COVID-19 across Saudi Arabia’s administrative provinces from March 2020 to August 2022. By analyzing multi-year provincial-level data, this research aims to (i) identify statistically significant hotspots and coldspots of COVID-19 transmission, (ii) examine how spatial clustering patterns evolved across distinct phases of the pandemic, and (iii) provide spatially explicit insights to support targeted, evidence-based public health interventions. Through this approach, the study contributes to a deeper understanding of COVID-19 dynamics in Saudi Arabia and demonstrates the broader value of spatiotemporal analytics for infectious disease surveillance and pandemic preparedness. 2. Literature Survey The literature review presents a multifaceted analysis of various methodologies and findings related to the dynamics of the COVID-19 pandemic. Previous studies consistently demonstrate that spatiotemporal analysis integrated with spatial autocorrelation techniques is essential for understanding the geographic diffusion and temporal evolution of COVID-19. The core premise underlying this body of research is that COVID-19 transmission is non-random and exhibits structured spatial and temporal patterns influenced by geographic proximity, population density, human mobility, and socioeconomic and environmental factors (Alrasheed et al., 2024 [ 4 , 5 , 17 ]. Early research during the COVID-19 pandemic highlighted the value of spatiotemporal frameworks for capturing disease dynamics. In the Saudi Arabian context, [ 11 ] applied spatial–temporal analysis using scatter plots, Moran scatter plots, and ARIMA forecasting models, demonstrating high predictive accuracy with error margins below 11%. This study established the feasibility and effectiveness of spatiotemporal methods for analyzing COVID-19 diffusion at the national and provincial levels in Saudi Arabia. Subsequent work in the Makkah region further reinforced these findings, showing that Space–Time ARIMA (STARIMA) models outperformed traditional ARIMA models, particularly during periods of increased mobility, thus emphasizing the importance of incorporating spatial dependence into temporal forecasting [ 12 , 13 ]. A substantial body of literature confirms that positive spatial autocorrelation is a dominant feature of COVID-19 incidence. Studies across different geographic contexts found that regions with high infection rates tend to be adjacent to similarly affected regions, as evidenced by significant Global and Local Moran’s I statistics [ 7 , 8 ]. This clustering behavior supports the implementation of geographically coordinated public health interventions rather than isolated, location-specific responses. Moreover, research has shown that spatial autocorrelation patterns evolve across epidemic phases. For example, [ 4 ] documented a spatial shift in COVID-19 clusters across New York State over time, while [ 10 ] demonstrated that spatial autocorrelation between demographic and healthcare factors and COVID-19 incidence intensified during later epidemic waves in Thailand. Hotspot and emerging hotspot analyses have further enhanced understanding of COVID-19 spatial dynamics by identifying statistically significant concentrations of high transmission and their geographic migration. Using Getis-Ord Gi* statistics, [ 7 ] revealed persistent and shifting hotspots in Oman, with directional expansion of affected areas over time. At the global scale, [ 8 ] reported pronounced temporal fluctuations in spatial autocorrelation, with COVID-19 hotspots shifting from the Western Pacific region to Europe and the Americas during 2020. Similar regional heterogeneity was observed in Iran, where spatiotemporal hotspot analysis identified distinct provincial clusters of hospitalizations and deaths, underscoring spatial variation in disease severity [ 18 ]. Local Indicators of Spatial Association (LISA) have been particularly effective in disaggregating spatial patterns at sub-regional levels. Studies in Iran, Indonesia, and other contexts demonstrated that LISA methods can identify high-high, low-low, and spatial outlier patterns that are critical for targeted intervention planning [ 19 , 20 ]. These localized analyses revealed that COVID-19 transmission dynamics vary considerably even within single administrative units, highlighting the limitations of relying solely on global spatial statistics. More advanced spatial modeling approaches, including geographically weighted regression (GWR) and multiscale GWR (MGWR), have shown that relationships between COVID-19 outcomes and their determinants are spatially heterogeneous. Empirical studies demonstrated that population density, mobility, climate variables, and socioeconomic indicators exert varying effects across geographic space, relationships that cannot be captured by traditional global regression models [ 21 , 22 ]. These findings are particularly relevant for Saudi Arabia, where provinces differ markedly in population concentration, climate, healthcare infrastructure, and mobility patterns. Recent methodological advances have also integrated machine learning and network-based approaches with spatiotemporal analysis. Dynamic adaptive spatiotemporal graph networks and deep learning models, such as Bi-LSTM, have significantly improved forecasting accuracy by capturing complex spatial and temporal dependencies in COVID-19 data [ 23 , 24 , 25 ]. These approaches suggest promising directions for future spatiotemporal modeling at the provincial level in Saudi Arabia. Overall, the existing literature demonstrates that spatiotemporal analysis combined with spatial autocorrelation techniques provides a powerful and necessary framework for understanding COVID-19 dynamics. While foundational studies in Saudi Arabia have confirmed the applicability of these methods [ 11 , 12 ]. Despite extensive global research on COVID-19, significant gaps remain in understanding its spatiotemporal dynamics within Saudi Arabia. Existing studies often examine spatial or temporal patterns independently, with limited integration of both dimensions at the provincial level. Addressing these gaps requires an integrated spatiotemporal and spatial autocorrelation framework capable of capturing dynamic transmission patterns and regional heterogeneity, thereby supporting more effective, evidence-based public health decision-making in Saudi Arabia. 3. Materials and Methods In this section, the methodology of the research study focusing on Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia's provinces has been comprehensively discussed. The primary objective of this study is to examine the spatiotemporal patterns of COVID-19 transmission across Saudi Arabia’s provinces using spatial autocorrelation and hotspot analysis. Additionally, the study aims to utilize the findings to produce risk and social vulnerability maps to guide future interventions and public health strategies. The methodology revolves around three core components: 3.1. Study Area The study focuses on Saudi Arabia, which is divided into thirteen administrative provinces with a population of approximately thirty-five million, distributed across 114 governorates, within 13 regions (Table. 1), (Figure. 1). Table 1 shows name of 114 governorates, within 13 regions region Governorates region Governorates region Governorates region Governorates region Governorates region Governorates Ar-Riyad Alaflaj Makkah Al-Mokarramah Aljumum Eastern Region Alahsa Al-Qaseem Alasyah Jazan Alaridah Hail Alghazalah Alhariq Alkamil Addammam Albadai Abu Arish Asshinan Addiriyah Alkhurmah Aljubayl Albukayriyah Ahad almusarihah Baqa Adduwadimi Allith Alkhafji Almidhnab Addair Hail Alghat Alqunfidhah Alkhubar Annabhaniyah Alharth Aseer Abha Afif Altaif Alnuayriyah Arrass Alidabi Ahad Rifaydah Alkharj Jeddah Alqatif Ashshimasiyah AlDarb Almajardah Almajmaah Khulays Buqayq Buraydah Arrayth Annamas Almuzahimiyah Makkah almukarramah Hafar albatin Riyadh alkhabra Baysh Balqarn Alquwayiyah Rabigh Qaryah alulya Unayzah Damad Bishah Arriyad Ranyah Ras Tannurah Uyun aljiwa Farasan Khamis Mushayt Assulayyil Turubah Al-Baha Albaha Al-Madinah Al-Monawarah Alhinakiyah Jazan Muhayil Azzulfi Najran Alkharkhir Alaqiq Almadinah almunawwarah Sabya Rijal Alma Duruma Badr aljanub Almandaq Almahd Samtah Sarat Abidah Hawtat Bani Tamim Hubuna Almukhwah Alula Tabouk Alwajh Tathlith Huraymila Khubash Alqari Badr Duba Zahran aljanub Marat Najran Biljurashi Khaybar Haqil Rumah Sharurah Qilwah Yanbu albahr Tabuk Shaqra Thar Al-Jouf Alqurayyat Northern Borders Arar Tayma Thadiq Yadamah Dawamat aljandal Rafha Umluj Wadi addawasir Sakaka Turayf 3.2 Data Description Saudi Arabia is divided into 13 administrative provinces with a population of 35 million people. The research relies on daily COVID-19 data obtained from the official COVID-19 bulletin, accessible at https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/ . The first case of COVID-19 in Saudi Arabia was recorded in March 2020 by the Ministry of Health. This study relies on daily COVID-19 data retrieved from the Saudi Ministry of Health COVID-19 response bulletin, which provides several sources of data about the COVID-19 pandemic in Saudi Arabia. It includes various sources of information for use in research. The study period of the data used in this paper extends from 31 March 2020 to 31 August 2022 (three years). The data obtained in dBASE format are then converted into a spreadsheet format. This study relied on cumulative daily data for confirmed, recovered, and death cases of COVID-19 (see Table 2 ). Riyadh is the capital of Saudi Arabia that has a population of 8 million individuals, so the most COVID-19 confirmed cases were recorded within it Table 2 Cumulative numbers of Confirmed cases of Covid-19 at the end of two months of three years in Saudi Arabia. NAME 2020 2021 2022 Mar. 31 Aug. 31 Mar. 31 Aug. 31 Mar. 31 Aug. 31 Ar-Riyad 568 67564 86982 129854 199562 220868 Makkah Al-Mokarramah 507 79706 92557 129457 181984 199784 Eastern Region 311 78696 93433 117278 149414 160056 Al-Madinah Al-Monawarah 76 22016 30830 39196 51439 55208 Al-Qaseem 5 11608 14997 21490 28178 29643 Aseer 32 24504 28754 40791 51037 53894 Al-Jouf 0 1027 1679 2632 3803 3896 Tabouk 2 4335 5335 8177 11496 12050 Jazan 8 10779 12474 20903 30355 32230 Hail 0 5697 7896 11586 14011 14444 Najran 5 5482 6892 10560 13150 13628 Northern Borders 2 1491 3252 5186 6769 6936 Al-Baha 13 2867 4926 7339 9616 10824 3.3 Spatiotemporal Analysis Spatiotemporal analysis examines phenomena by considering both their geographic location and their temporal evolution, focusing on where events occur, when they take place, and how they change over time. This approach integrates spatial components—such as location, spatial distribution, and inter-regional relationships—with temporal elements, including trends, sequences, rates of change, and diffusion processes. Spatiotemporal series analysis enables the monitoring of variable values across multiple locations over successive time periods, facilitating the examination of dynamic processes such as disease transmission or urban expansion. In addition, spatiotemporal clustering techniques are employed to identify statistically significant hotspots that emerge across both space and time. Distinct pandemic phases were delineated based on statistically significant change points in national COVID-19 case trends identified using the Pettitt test (p < 0.05). Within this framework, spatial autocorrelation plays a fundamental role by assessing the degree of similarity between COVID-19 case numbers in neighboring regions. Spatial statistical measures, particularly Moran’s I, are applied to determine whether observed case distributions are randomly dispersed or spatially clustered. When combined with Local Indicators of Spatial Association (LISA), this methodology allows for the detection of localized clusters and spatial outliers, thereby identifying regions that may be particularly vulnerable. In the context of Saudi Arabia, these analytical tools enhance understanding of the spatial diffusion of COVID-19 across provinces and are essential for pinpointing high-risk areas, supporting targeted and effective pandemic response strategies. 3.3.1 Anselin local Moran’s I Utilizing clustering algorithms and methods, the research classifies geographical regions that exhibit comparable COVID-19 patterns. Identifying the optimal number of clusters is a crucial component of the analysis. The objective of the research is to establish clusters in which the regions comprising each group are maximally similar, while the groups as a whole are maximally distinct. The attributes specified for the Analysis Fields parameter, which may include spatial and space-time properties, determine feature similarity within clusters. The different classes of z-values are represented as High-High (HH) or Low-Low (LL). High, positive z-values indicate that an area is surrounded by other areas with similar values. Conversely, a low negative z-score indicates a statistically significant spatial anomaly, i.e. a high-value area surrounded by low-value areas (HL), with LH representing the inverse pattern. In order to determine the spatial correlation between variables, spatial autocorrelation was utilized to match attribute similarity with location similarity. The mathematical expression for Anselin local Moran’s I, an index of spatial autocorrelation derived from cross-products, is as follows: \(\:{I}^{i}=\frac{{X}_{i}-\stackrel{-}{X}\:\:}{{S}_{i}^{2}}\:\sum\:_{j\:=1,\:j\ne\:i}^{n}{W}_{ij}\:({X}_{j}-\stackrel{-}{X})\) (Eq. 1) Where \(\:{X}_{i}\) is an attribute for feature \(\:i,\:\stackrel{-}{X}\) is the mean of the corresponding attribute, \(\:{W}_{i,j}\) is the spatial weight between feature \(\:i\) and \(\:j\) . Also, where n is the number of regions; xi the attribute value at area I; ̅x the mean value of the attribute in the study region; and wij elements of a spatial lag operator W (spatial weights of matrix W). The significance of the index is usually tested in a situation of normal distribution. Each observation/location is classified into one of the following four categories: HH - the location is part of a significant high-high correlation cluster, meaning that the location has high autocorrelated and surrounded by other high autocorrelated neighbors. LL - the location is part of a significant low-low correlation cluster, meaning that the location has low autocorrelation and surrounded by other low autocorrelation neighbors. HL - the location is an outlier with high autocorrelation but surrounded by neighbors with low autocorrelation. LH - the location is an outlier with low autocorrelation but surrounded by neighbors with high autocorrelation. 3.3.2 Hot Spot Analysis (Getis-Ord Gi*) The tool computes the Getis–Ord Gi* statistic for each spatial feature in a dataset, producing corresponding z-scores and p-values that identify statistically significant spatial clustering of high or low values. This method assesses each feature relative to its surrounding neighbors, recognizing that a feature with a high attribute value alone does not necessarily constitute a significant hotspot. Rather, statistical significance is achieved when a high-value feature is embedded within a neighborhood of similarly high values. The analysis compares the aggregated value of a feature and its neighbors with the expected aggregate across the entire study area. When the observed local sum deviates substantially from the expected value beyond what could be attributed to random variation, a statistically significant z-score is generated. To address issues related to multiple comparisons and spatial dependence, the false discovery rate (FDR) correction is applied, ensuring more reliable identification of hotspots and cold spots. The Gi(d) statistic is a distance-based measure that quantifies the concentration of a variable within a specified radius around a given location relative to its distribution across the entire study area. The statistic for location i is formally expressed as: \(\:{G}_{i}^{*}=\frac{\:{\sum\:}_{j=1}^{n}{W}_{ij}\:{X}_{i}-\stackrel{-}{X}\:{\sum\:}_{j=1}^{n}{W}_{ij}\:,j\:}{S\:\sqrt{\begin{array}{c}\\\:\frac{n\:{\sum\:}_{j=1}^{n}{{W}^{2}}_{ij}\:,j\:(\:{\sum\:}_{j=1}^{n}{W}_{ij}\:,j{)}^{2}}{n}\end{array}}}\) (Eq. 2) Where \(\:{X}_{j}\) is an attribute value for feature \(\:j,\:{W}_{ij}\) is the spatial weight between feature \(\:i\) and \(\:j\) , \(\:\text{n}\) is equal to number of features. where xj is the value of the observation at point j; wij(d) the ij element of a binary W matrix (wij = 1 if the site is within distance d or 0 if elsewhere; and n the number of observations made. The mean and the variance of this statistic can be obtained through randomization and used to derive a standard statistic. When the value of the standardized statistic is greater than the cut-off value with pre specified significance, positive or negative spatial association exists. Positive values represent spatial agglomeration, while negative values represent the opposite. The higher or lower the z-score, the higher the possibility of clustering, while a z-score close to zero means absence of obvious clusters. Thus, a positive z represents the possibility of clustering, while a negative z indicates a low possibility of clustering. The Gi* statistic returned for each feature in the dataset is a z-score. For statistically significant positive z-scores, the larger the z-score, the more intense the clustering of high values (hot spots). For statistically significant negative z-scores, the smaller the z-score, the more intense the clustering of low values (cold spots). For more information about determining statistical significance and correcting for multiple testing and spatial dependency. 4. Results In this section, the outcomes of the proposed methodology are presented and the significant observations derived from these results are discussed. The data analysis that is conducted, as previously described, offers valuable insights into the spatial distribution of variables under study. The research study acknowledges the significance of historical data to understand the progression and future dynamics of the COVID-19 pandemic. To address this, the study focuses on estimating the number of active cases over time by utilizing time-series data, which are sequences of numeric data measured at consistent time intervals. This research employs a multifaceted approach, encompassing data collection, spatial autocorrelation techniques, and spatial clustering analysis, to gain a comprehensive understanding of COVID-19 patterns and spatial relationships within Saudi Arabia's provinces. The findings from this methodology will contribute to the development of effective public health strategies and interventions in response to the pandemic. The outcomes of this spatial clustering analysis will furnish valuable insights regarding the existence of COVID-19 concentrations and regions exhibiting comparable patterns of cases. Such information can significantly assist local, state, and federal health authorities in devising targeted interventions. 4.1 Spatiotemporal Hotspot and Anselin Local Moran’s I of COVID-19 in Saudi Arabia (2020–2022) The series of hotspot and Anselin Local Moran’s I (Figure: 2) maps (A–F) illustrates the spatiotemporal progression of COVID-19 across Saudi Arabia from March 2020 to August 2022. These maps reveal how the spatial distribution of cases evolved over time—from initial localized clusters to widespread regional transmission and, eventually, to more isolated and fragmented outbreak patterns. By identifying statistically significant hotspots (High-High clusters), coldspots (Low-Low clusters), and spatial outliers, the analysis provides important insights into the geographic dynamics that shaped the pandemic’s footprint across the country. The earliest phase, in (map, A) represented in March 2020, reflects the initial introduction of COVID-19 into the Kingdom. Hotspots were highly concentrated in the Eastern Region, particularly around Al-Ahsa, Dammam, and Qatif. These areas were among the first to experience community transmission due to dense population centers, large expatriate communities, and early international travel connections. Elsewhere, only small clusters appeared sporadically. In contrast, the southwestern and northwestern regions formed clear coldspots, likely due to lower population density, limited mobility, and the rapid implementation of containment measures. This early spatial pattern marks the virus’s entry and the earliest epidemiological signals. By August 2020, the maps (map, B) indicate a clear shift from localized outbreaks to widespread regional transmission. Hotspots expanded into the Riyadh Region, Qassim, and several northern governorates. Urban centers such as Jeddah also became statistically significant hotspots. The expansion of red zones across central and eastern Saudi Arabia reflects increased mobility as restrictions eased and summer activities resumed. Coldspots, meanwhile, persisted primarily in the southwestern mountainous regions, where natural geographic isolation and lower population density contributed to lower transmission rates. This stage marks the transition from targeted outbreaks to more systemic nationwide spread. March 2021 (map, C) captures the post-winter wave, characterized by sustained and intensified clustering across major regions. Hotspots remained entrenched in Riyadh, Eastern Province, and parts of the north, demonstrating strong spatial persistence. These persistent clusters reflect areas with high population density, extensive social and economic activity, and greater interregional connectivity. Coldspots in the southwest continued to show stability, reinforcing the spatial disparity in disease dynamics. Local High-Low outliers became more prominent, indicating isolated pockets of high cases surrounded by lower-case areas—an emerging pattern consistent with complex local transmission dynamics. By August 2021, (map, D) the spread of the Delta periods amplified spatial clustering. Hotspots became more densely concentrated and statistically stronger, especially within the Riyadh metropolitan region and the Eastern Province. Spatial outliers increased, revealing more nuanced local patterns of deviation within larger clusters. The persistence of coldspots in the south and southwest again highlights regional differences in transmission potential, mobility behavior, and demographic structure. This phase represents the peak of spatial autocorrelation in the dataset, where neighboring regions exhibited highly similar and high case intensities. In March 2022, (map, E) following the Omicron periods wave, the spatial pattern began to change noticeably. Hotspots contracted in size and intensity, though they remained anchored in the central and eastern regions. More spatial outliers appeared, suggesting that transmission had become more fragmented rather than regionally widespread. The reduction in large hotspot zones corresponds with high vaccination rates, natural immunity, and the decreased severity of circulating variants. Coldspots retained their relative stability, reflecting minimal changes in historically low-transmission areas. The final map (map, F) from August 2022, represents a post-vaccination stabilization phase. Hotspots became smaller, weaker, and more localized, often linked to isolated outbreaks rather than broad regional waves. The spatial structure shifted away from extensive High-High clusters to a mixed pattern dominated by outliers and isolated pockets of transmission. This fragmentation signifies the transition toward an endemic character, where local outbreaks occur independently but no longer form sustained regional hotspots. Across all six time periods, a clear narrative emerges: COVID-19 in Saudi Arabia progressed from localized initial outbreaks to widespread regional clusters, peaking in 2021, before evolving into isolated localized patterns by 2022. The long-term persistence of hotspots in the Eastern Province and Riyadh Region underscores the influence of population density, urbanization, and human mobility in shaping COVID-19 spatial dynamics. Conversely, recurring coldspots in the southwestern regions reflect underlying socioeconomic, demographic, and geographic factors that consistently limited transmission. Overall, the hotspot and local spatial autocorrelation analysis provides a comprehensive understanding of how COVID-19 spread, intensified, and subsided across the Kingdom. These findings demonstrate the value of spatial epidemiology in guiding targeted interventions, resource allocation, and public health decision-making during a dynamic and evolving pandemic. Study area: Governorates of Saudi Arabia Hotspot colors: Red = Hotspot (High–High clusters), Blue = Coldspot (Low–Low clusters), Light red/blue = Spatial outliers (High–Low or Low–High) and Grey = Non-significant. 4.2 Spatiotemporal Hotspot Dynamics of COVID-19 in Saudi Arabia (2020–2022 Getis-Ord Gi Hotspot Analysis The maps in Fig. 3 (A–F) illustrate the spatiotemporal evolution of COVID-19 hotspots and coldspots across Saudi Arabia from March 2020 to August 2022, based on Getis-Ord General G statistics. Together, they reveal how the spatial intensity of the pandemic changed across governorates over time and how transmission patterns shifted between regions. In March 2020 (Map A), during the initial outbreak phase, hotspots were highly concentrated in the Eastern Province, particularly in Qatif, Al-Ahsa, and Dammam. These areas were among the earliest to report community transmission, and their strong regional connectivity facilitated early clustering of cases. A few moderate hotspots emerged in Makkah, while large portions of the western and southern regions acted as coldspots, reflecting minimal spread during the early months of the pandemic. By August 2020 (Map B), hotspots expanded substantially. High-intensity clusters appeared across a wider portion of the Eastern Region and began to emerge in central governorates surrounding Riyadh. This period corresponds to the first major national wave of infections, driven by increased mobility and gradual relaxation of movement restrictions. Meanwhile, coldspots persisted across the southwest, consistent with low population density and reduced inter-regional movement. In March 2021 (Map C), hotspots intensified in both the Eastern Province and the Riyadh region, indicating sustained transmission in the most urbanized and economically active regions of the country. Additional clusters appeared in the western region near Jeddah, reflecting ongoing urban spread. Coldspots remained prevalent in the southwest, where demographic and geographic factors limited widespread transmission. By August 2021 (Map D), hotspot activity became more concentrated, although the same high-risk regions—Riyadh and the Eastern Province—continued to dominate. This period aligns with Saudi Arabia’s widespread vaccination rollout, which likely contributed to reduced transmission and the contraction of hotspot zones. Coldspot patterns expanded, indicating declining spatial clustering in lower-risk regions. In March 2022 (Map E), hotspot clusters fragmented further, appearing only in isolated pockets around Riyadh and the Eastern Province. Most of the country transitioned into nonsignificant or coldspot status, suggesting a notable decrease in spatial dependence of COVID-19 cases and an overall improvement in epidemiological conditions. Finally, by August 2022 (Map F), hotspot presence was minimal. Only a few governorates in Riyadh and the Eastern Province displayed statistically significant clustering, while the remaining regions indicated stable or low transmission rates. This spatial pattern reflects the stabilization of the pandemic, increased immunity levels, and the sustained effect of national health measures. Overall, the spatiotemporal hotspot analysis demonstrates a clear progression from early localized outbreaks to broader regional clustering during the pandemic’s peak, followed by a gradual dissolution of hotspots as public health interventions took effect. The Eastern Province and Riyadh emerged as persistent high-intensity clusters throughout the study period, whereas the southwestern regions consistently appeared as coldspots. This dynamic evolution underscores the importance of spatial analytics in understanding epidemic progression and supporting targeted health interventions. 5. Discussion This study is subject to several data limitations. First, COVID-19 case counts likely underrepresent true infection levels due to asymptomatic cases, limited testing capacity during early pandemic phases, and variations in reporting practices across provinces. Second, testing policies evolved over time, particularly during 2020–2021, which may have influenced observed temporal trends. Third, potential differences in case definitions and reporting delays across provinces could introduce spatial bias. These factors should be considered when interpreting the results, and future studies should incorporate excess mortality data or seroprevalence surveys to refine spatial estimates. The spatiotemporal analysis of COVID-19 in Saudi Arabia from 2020 to 2022 reveals a clear, non-random progression of the pandemic shaped by population density, mobility patterns, and public health interventions. Persistent hotspots in the Eastern Province and Riyadh Region highlight the dominant role of urbanization, economic activity, and interconnected travel networks in sustaining transmission. In contrast, the repeated identification of southwestern governorates as coldspots reflects the protective influence of lower population density, geographic isolation, and reduced mobility, underscoring the importance of localized rather than uniform national responses. The spatiotemporal findings of the present study both confirm and extend these established insights by providing a long-term, province-level analysis of COVID-19 dynamics across Saudi Arabia. In agreement with global and regional research, the results reveal strong positive spatial autocorrelation, with early hotspots concentrated in the Eastern Province—reflecting international connectivity, expatriate labor concentration, and industrial activity—followed by spatial diffusion toward central and western urban centers as mobility restrictions were relaxed [ 4 , 11 ]. The persistent hotspot status of Riyadh and the Eastern Province mirrors findings from other national and international studies identifying major metropolitan regions as enduring transmission nodes due to high population density and transportation connectivity [ 6 , 8 ]. Conversely, the sustained coldspot patterns observed in southwestern provinces align with prior evidence that geographic isolation, lower population density, and reduced mobility can significantly limit disease spread [ 10 , 20 ]. The persistent coldspot status of Asir, Jazan, and Al-Baha is likely related to structural geographic and demographic factors—including mountainous topography, lower population density, climatic conditions, and more limited transport and industrial infrastructure—which together reduced mobility and interregional transmission, resulting in consistently lower spatial clustering of COVID-19 cases. Crucially, the temporal evolution of spatial clustering observed in this study addresses key gaps identified in previous research. The intensification of spatial autocorrelation during variant-driven waves—particularly the Delta phase in mid-2021—and the subsequent fragmentation of hotspots following widespread vaccination and the emergence of Omicron periods illustrate the dynamic nature of spatial dependence over time [ 5 , 26 ]. These findings reinforce earlier conclusions that static spatial models are inadequate for capturing pandemic dynamics and underscore the necessity of integrated spatiotemporal frameworks capable of identifying shifting hotspots, emerging spatial outliers, and phase-specific transmission mechanisms [ 21 , 22 ]. From a methodological perspective, the combined application of global and local spatial autocorrelation measures with temporal analysis advances prior work by revealing nuanced sub-regional patterns that would remain obscured under single-method approaches. The identification of local clusters and spatial outliers through LISA and Getis-Ord statistics supports previous assertions that localized spatial analysis is essential for effective public health planning and geographically targeted interventions [ 2 , 19 ]. Overall, the convergence between previous studies and the present findings strengthens the evidence that COVID-19 transmission is governed by enduring place-based characteristics and evolving mobility networks, while demonstrating the added value of longitudinal, context-specific spatiotemporal analysis for informing adaptive and regionally tailored pandemic response strategies in Saudi Arabia. Compared with other Middle Eastern studies, Saudi Arabia exhibited similar spatial patterns to Oman and Iran, where early hotspots emerged in economically active coastal or industrial regions [ 7 , 18 ]. However, unlike some Gulf states, Saudi Arabia displayed greater spatial persistence of hotspots in its central metropolitan region (Riyadh), likely due to its larger population size and stronger interprovincial mobility networks. This suggests that national urban structure plays a critical role in shaping pandemic spatial dynamics across the region. 6. Conclusion This study applied an integrated spatiotemporal and spatial autocorrelation framework to examine the evolution of COVID-19 across Saudi Arabia’s administrative provinces between March 2020 and August 2022. By combining hotspot analysis, Anselin Local Moran’s I, Getis-Ord statistics, and clustering techniques, the research provides a comprehensive understanding of how COVID-19 transmission unfolded across space and time within the Kingdom. The findings confirm that the spread of COVID-19 in Saudi Arabia was highly non-random, exhibiting strong spatial dependence and clear temporal phases that reflect both epidemiological and behavioral dynamics. The results consistently identified the Eastern Province and the Riyadh Region as persistent hotspots throughout multiple stages of the pandemic. These areas, characterized by high population density, economic activity, and extensive domestic and international mobility, played a central role in sustaining transmission. In contrast, southwestern provinces such as Asir, Jazan, and Al-Baha repeatedly emerged as coldspots, highlighting the influence of geographic isolation, lower population density, and reduced mobility in limiting disease spread. This persistent spatial disparity demonstrates that structural geographic and demographic characteristics strongly condition pandemic outcomes, often independently of short-term policy interventions. Temporally, the analysis revealed a clear progression from localized outbreaks during the early phase of the pandemic to widespread regional clustering during peak transmission periods in 2020 and 2021. The intensification of spatial clustering during the Delta periods wave, followed by a marked fragmentation of hotspots after the rollout of mass vaccination and the emergence of less severe variants, illustrates the dynamic interaction between viral evolution, human behavior, and public health responses. The transition toward localized and sporadic transmission patterns in 2022 signals a shift toward endemicity within the Saudi Arabian context. Overall, this research demonstrates the critical value of spatial epidemiology and spatiotemporal analytics in pandemic assessment and response. The integrated methodological approach adopted in this study offers a robust framework for identifying high-risk regions, monitoring evolving transmission patterns, and supporting evidence-based, regionally tailored public health strategies. The insights generated emphasize the necessity of adaptive, spatially differentiated interventions rather than uniform nationwide measures. Beyond COVID-19, the analytical framework and findings of this study provide a transferable model for managing future infectious disease outbreaks and strengthening public health preparedness in Saudi Arabia and comparable settings. In summary, the findings support a targeted, spatially differentiated public health strategy that prioritizes enhanced surveillance and testing in persistent hotspots, implements mobility management along high-risk urban corridors, strengthens healthcare capacity in transitional zones, adopts flexible province-specific vaccination approaches, and utilizes real-time spatial dashboards to inform timely decision-making. Declarations Supplementary Materials: The research relies on daily COVID-19 data obtained from the official COVID-19 bulletin, accessible at https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/. Author Contributions: Fahad Almutlaq. Funding: The author extends his appreciation to the Deanship of Scientific Research at King Saud ongoing research funding program (ORF-2026-896). Data Availability Statement: all Data is open source: COVID-19 bulletin, accessible at https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/. Acknowledgments This author expresses ongoing research funding program (ORF-2026-896), King Saud University, Riyadh, Saud Arabia Conflicts of Interest: The authors declare no conflict of interest. References World Health Organization (WHO). (2020). Coronavirus disease (COVID-19) pandemic. WHO. Ye, L., & Hu, L. (2020). Spatiotemporal distribution and trend of COVID-19 in the Yangtze River Delta region of the People's Republic of China. Geospatial Health, 15(1). ‏ Anselin, L. (1995). Local indicators of spatial association—LISA. Geographical Analysis, 27(2), 93–115. Zheng, A., Wang, T., & Li, X. (2021). Spatiotemporal characteristics and risk factors of the COVID-19 pandemic in New York State: Implication of future policies. ISPRS International Journal of Geo-Information, 10(9), 627. Phang, P., Aslam, S., Labadin, J., & Jayaraj, V. J. (2025). Spatial autocorrelation analysis of infectious disease incidence rates at state and district level using supra-adjacency weights matrix. Universal Journal of Public Health, 13 (2), 456-470. Vilinová, K., & Petrikovičová, L. (2023). Spatial autocorrelation of COVID-19 in Slovakia. Tropical Medicine and Infectious Disease, 8(6), 298. ‏ Al-Kindi, K. M., Alkharusi, A., Alshukaili, D., Al Nasiri, N., Al-Awadhi, T., Charabi, Y., & El Kenawy, A. M. (2020). Spatiotemporal assessment of COVID-19 spread over Oman using GIS techniques. Earth Systems and Environment, 4(4), 797-811. ‏ Fatima, M., Arshad, S., Butt, I., & Arshad, S. (2021). Geospatial clustering and hot spot detection of COVID-19 incidence in 2020: A global analysis. International Journal of Geospatial and Environmental Research, 8(1). ‏ Liu, M., Liu, M., Li, Z., Zhu, Y., Liu, Y., Wang, X., ... & Guo, X. (2021). The spatial clustering analysis of COVID-19 and its associated factors in mainland China at the prefecture level. Science of the Total Environment, 777, 145992. Sandar, E., Laohasiriwong, W., & Sornlorm, K. (2023). Spatial autocorrelation and heterogenicity of demographic and healthcare factors in the five waves of COVID-19 epidemic in Thailand. Geospatial Health , 18 (1). ‏ Almobarak, A. S., Almohammadi, H. R., Aboalnaser, S. A., & Syed, L. (2020, December). Spatio-temporal analysis of the spread COVID-19 in Saudi Arabia. In 2020 13th International Conference on Developments in eSystems Engineering (DeSE) (pp. 341-346). IEEE. ‏ Awwad, F. A., Mohamoud, M. A., & Abonazel, M. R. (2021). Estimating COVID-19 cases in Makkah region of Saudi Arabia: Space-time ARIMA modeling. PLoS One, 16(4), e0250149. ‏ Al-Turaiki, I., Almutlaq, F., Alrasheed, H., & Alballa, N. (2021). Empirical evaluation of alternative time-series models for COVID-19 forecasting in Saudi Arabia. International Journal of Environmental Research and Public Health, 18(16), 8660. Alharbi, M. M., Rabbani, S. I., Asdaq, S. M. B., Alamri, A. S., Alsanie, W. F., Alhomrani, M., ... & Alajlan, S. A. (2021). Infection spread, recovery, and fatality from coronavirus in different provinces of Saudi Arabia. Healthcare, 9, 931. Abdel-Aal, M. A., Eltoukhy, A. E., Nabhan, M. A., & AlDurgam, M. M. (2022). Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia. Environmental Science and Pollution Research, 29(14), 20449–20462. Faisal, K., Alshammari, S., Alotaibi, R., Alhothali, A., Bamasag, O., Alghanmi, N., & Bin Yamin, M. (2022). Spatial analysis of COVID-19 vaccine centers distribution: A case study of the city of Jeddah, Saudi Arabia. International Journal of Environmental Research and Public Health, 19(6), 3526. Alrasheed, H., Alballa, N., Al-Turaiki, I., Almutlaq, F., & Alabduljabbar, R. (2024). City transmission networks: Unraveling disease spread dynamics. ISPRS International Journal of Geo-Information, 13(8). Mounesan, L., Farhadi, E., Eybpoosh, S., Hosseini, A., Parsaeian, M., Gharibzadeh, S., Ahmadinezhad, M., Bahari, F., Gouya, M., Haghdoost, A., & Mostafavi, E. (2025). Detecting the seasonal and spatial patterns of COVID-19 hospitalization and deaths in Iran: Insights from a spatiotemporal and hotspot analysis. International Journal of Preventive Medicine. Isazade, V., Qasimi, A. B., Dong, P., Kaplan, G., & Isazade, E. (2023). Integration of Moran’s I, geographically weighted regression (GWR), and ordinary least square (OLS) models in spatiotemporal modeling of COVID-19 outbreak in Qom and Mazandaran Provinces, Iran. Modeling Earth Systems and Environment, 9(4), 3923-3937. ‏ Syetiawan, A., Harimurti, M., & Prihanto, Y. (2022). A spatiotemporal analysis of COVID-19 transmission in Jakarta, Indonesia for pandemic decision support. Geospatial Health; volume 17(s1):1042 Chen, M., Chen, Y., Wilson, J. P., Tan, H., & Chu, T. (2022). Using an eigenvector spatial filtering-based spatially varying coefficient model to analyze the spatial heterogeneity of COVID-19 and its influencing factors in mainland China. ISPRS International Journal of Geo-Information, 11(1), 67. ‏ Fan, Z., Zhan, Q., Yang, C., Liu, H., & Zhan, M. (2020). How did distribution patterns of particulate matter air pollution (PM2. 5 and PM10) change in China during the COVID-19 outbreak: A spatiotemporal investigation at Chinese city-level. International journal of environmental research and public health, 17(17), 6274. ‏ Aldhyani, T. H., & Alkahtani, H. (2021). A bidirectional long short-term memory model algorithm for predicting COVID-19 in gulf countries. Life, 11(11), 1118. ‏ Pu, X., Zhu, J., Wu, Y., Leng, C., Bo, Z., & Wang, H. (2024). Dynamic adaptive spatio–temporal graph network for COVID‐19 forecasting. CAAI Transactions on Intelligence Technology, 9(3), 769-786. ‏ Foruzandeh, M., Neysani Samany, N., & Khodakaramian, B. (2024). A Machine Learning Approach for Modeling the Spatial-temporal Propagation Pattern of COVID-19. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 48, 183-190. ‏ . Liu, L., Hu, T., Bao, S., Wu, H., Peng, Z., & Wang, R. (2021). The spatiotemporal interaction effect of COVID-19 transmission in the United States. ISPRS International Journal of Geo-Information, 10(6), 387. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 05 May, 2026 Reviewers agreed at journal 01 May, 2026 Reviewers agreed at journal 30 Apr, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers invited by journal 29 Apr, 2026 Editor assigned by journal 27 Apr, 2026 Editor invited by journal 22 Apr, 2026 Submission checks completed at journal 18 Apr, 2026 First submitted to journal 18 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9418580","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":634618333,"identity":"8400c3fd-e25d-42b1-bb8c-34f77a5747d7","order_by":0,"name":"Fahad Almutlaq","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIiWNgGAWjYBAC9gbmBiBlAWS1gQUYGwhp4TkAUpMgAWQdI1mLRBqxWtgPtm7m/SEhz3fzWfJnHgYb2Q0H2B9+wKuFJ7HtNk+ChOHM22nHpHkY0ow3HOAxlsCnxZ4BooVxw+30NmYehsOJQC0MeLXw8D8Ea7HfcPN4M9Bh/4Fa2B//wKtFAmJL4oYbbAeADjsA1MJght8WiYdtN+ekSSTPPJOWJjnHINl45mEeMwv8Dks+duONjY1t3/Fjxh/eVNjJ9h1vf3wDnxYEOAAiDICYmTj1MC2jYBSMglEwCrAAALHHTOpZg6DjAAAAAElFTkSuQmCC","orcid":"","institution":"King Saud University","correspondingAuthor":true,"prefix":"","firstName":"Fahad","middleName":"","lastName":"Almutlaq","suffix":""}],"badges":[],"createdAt":"2026-04-14 17:53:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9418580/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9418580/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108685897,"identity":"700fd63d-1d1b-4f2e-bdbc-e9c4c70214e3","added_by":"auto","created_at":"2026-05-07 10:01:33","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":265710,"visible":true,"origin":"","legend":"\u003cp\u003eThe 114 governorates in Saudi Arabia.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9418580/v1/2681248471621632538b0dbc.jpeg"},{"id":108805902,"identity":"97e4b93d-3e0e-4b40-bdff-5e6d43aeb932","added_by":"auto","created_at":"2026-05-08 15:27:09","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":594541,"visible":true,"origin":"","legend":"\u003cp\u003eSpatiotemporal hotspot analysis and Anselin Local Moran’s I of COVID-19 cases in Saudi Arabia’s provinces (2020–2022).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9418580/v1/a8b61928577c2376c22657b9.jpeg"},{"id":108685895,"identity":"455331a1-33e6-4113-8ed0-8dfe36993f76","added_by":"auto","created_at":"2026-05-07 10:01:33","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":604689,"visible":true,"origin":"","legend":"\u003cp\u003eSpatiotemporal Getis-Ord Gi* hotspot analysis of COVID-19 cases\u003c/p\u003e\n\u003cp\u003eacross Saudi Arabia’s provinces. (2020–2022).\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9418580/v1/676d9640a07592eb70605dac.jpeg"},{"id":108809559,"identity":"27b486e1-0bdb-44b9-bf21-7b00de46e869","added_by":"auto","created_at":"2026-05-08 15:53:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1845115,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9418580/v1/6ff3e6e5-2130-486e-b75f-2aa86163d1e6.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia’s provinces","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe coronavirus disease 2019 (COVID-19) pandemic has posed unprecedented challenges to global public health systems, economies, and social structures. Since its emergence in late 2019, COVID-19 has spread rapidly across countries and regions. This spread has demonstrated complex transmission dynamics shaped by human mobility, population density, socioeconomic conditions, and public health interventions [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Understanding how the virus propagates across space and time is therefore essential for designing effective containment strategies, allocating healthcare resources, and enhancing preparedness for future pandemics.\u003c/p\u003e \u003cp\u003eTraditional epidemiological analyses often focus on temporal trends in case numbers or aggregate national-level statistics, which may obscure important geographic heterogeneity in disease transmission. However, infectious diseases such as COVID-19 rarely spread randomly; instead, they exhibit spatial dependence, whereby neighboring regions tend to experience similar infection patterns due to shared mobility networks, environmental conditions, and demographic characteristics [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Spatiotemporal analysis, combined with spatial autocorrelation techniques, offers a powerful framework for capturing these patterns by simultaneously examining where infections occur, how they cluster geographically, and how these clusters evolve over time [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA growing body of international research has demonstrated the effectiveness of spatial epidemiology tools\u0026mdash;such as Moran\u0026rsquo;s I, Local Indicators of Spatial Association (LISA), and Getis-Ord Gi* hotspot analysis\u0026mdash;in identifying COVID-19 clusters, transmission corridors, and high-risk regions [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. These studies consistently show that metropolitan and highly connected regions act as primary hubs of viral transmission, while less populated or geographically isolated areas often experience lower incidence rates. Moreover, the intensity and configuration of spatial clustering have been shown to vary across epidemic phases, influenced by public health interventions, behavioral changes, and the emergence of new variants [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn Saudi Arabia, the spatial dynamics of COVID-19 warrant particular attention due to the country\u0026rsquo;s unique geographic, demographic, and socioeconomic characteristics. The Kingdom encompasses highly urbanized metropolitan centers, vast desert regions, and provinces with pronounced disparities in population density, healthcare access, and mobility patterns. Additionally, Saudi Arabia\u0026rsquo;s role as a global destination for religious pilgrimage introduces episodic surges in domestic and international movement, which can significantly influence disease transmission dynamics [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. While several studies have examined COVID-19 trends in Saudi Arabia, many have focused on either spatial or temporal dimensions independently, limiting their ability to capture dynamic provincial-level transmission patterns [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. These investigations are anchored in existing literature, including the works of [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], providing a comprehensive understanding of the pandemic at both local and global scales.\u003c/p\u003e \u003cp\u003eThis study addresses these limitations by applying an integrated spatiotemporal analysis framework incorporating spatial autocorrelation, hotspot detection, and clustering techniques to investigate the evolution of COVID-19 across Saudi Arabia\u0026rsquo;s administrative provinces from March 2020 to August 2022. By analyzing multi-year provincial-level data, this research aims to (i) identify statistically significant hotspots and coldspots of COVID-19 transmission, (ii) examine how spatial clustering patterns evolved across distinct phases of the pandemic, and (iii) provide spatially explicit insights to support targeted, evidence-based public health interventions. Through this approach, the study contributes to a deeper understanding of COVID-19 dynamics in Saudi Arabia and demonstrates the broader value of spatiotemporal analytics for infectious disease surveillance and pandemic preparedness.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"2. Literature Survey","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe literature review presents a multifaceted analysis of various methodologies and findings related to the dynamics of the COVID-19 pandemic. Previous studies consistently demonstrate that spatiotemporal analysis integrated with spatial autocorrelation techniques is essential for understanding the geographic diffusion and temporal evolution of COVID-19. The core premise underlying this body of research is that COVID-19 transmission is non-random and exhibits structured spatial and temporal patterns influenced by geographic proximity, population density, human mobility, and socioeconomic and environmental factors (Alrasheed et al., 2024 [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEarly research during the COVID-19 pandemic highlighted the value of spatiotemporal frameworks for capturing disease dynamics. In the Saudi Arabian context, [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] applied spatial\u0026ndash;temporal analysis using scatter plots, Moran scatter plots, and ARIMA forecasting models, demonstrating high predictive accuracy with error margins below 11%. This study established the feasibility and effectiveness of spatiotemporal methods for analyzing COVID-19 diffusion at the national and provincial levels in Saudi Arabia. Subsequent work in the Makkah region further reinforced these findings, showing that Space\u0026ndash;Time ARIMA (STARIMA) models outperformed traditional ARIMA models, particularly during periods of increased mobility, thus emphasizing the importance of incorporating spatial dependence into temporal forecasting [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA substantial body of literature confirms that positive spatial autocorrelation is a dominant feature of COVID-19 incidence. Studies across different geographic contexts found that regions with high infection rates tend to be adjacent to similarly affected regions, as evidenced by significant Global and Local Moran\u0026rsquo;s I statistics [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. This clustering behavior supports the implementation of geographically coordinated public health interventions rather than isolated, location-specific responses. Moreover, research has shown that spatial autocorrelation patterns evolve across epidemic phases. For example, [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] documented a spatial shift in COVID-19 clusters across New York State over time, while [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] demonstrated that spatial autocorrelation between demographic and healthcare factors and COVID-19 incidence intensified during later epidemic waves in Thailand.\u003c/p\u003e \u003cp\u003eHotspot and emerging hotspot analyses have further enhanced understanding of COVID-19 spatial dynamics by identifying statistically significant concentrations of high transmission and their geographic migration. Using Getis-Ord Gi* statistics, [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] revealed persistent and shifting hotspots in Oman, with directional expansion of affected areas over time. At the global scale, [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] reported pronounced temporal fluctuations in spatial autocorrelation, with COVID-19 hotspots shifting from the Western Pacific region to Europe and the Americas during 2020. Similar regional heterogeneity was observed in Iran, where spatiotemporal hotspot analysis identified distinct provincial clusters of hospitalizations and deaths, underscoring spatial variation in disease severity [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eLocal Indicators of Spatial Association (LISA) have been particularly effective in disaggregating spatial patterns at sub-regional levels. Studies in Iran, Indonesia, and other contexts demonstrated that LISA methods can identify high-high, low-low, and spatial outlier patterns that are critical for targeted intervention planning [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. These localized analyses revealed that COVID-19 transmission dynamics vary considerably even within single administrative units, highlighting the limitations of relying solely on global spatial statistics.\u003c/p\u003e \u003cp\u003eMore advanced spatial modeling approaches, including geographically weighted regression (GWR) and multiscale GWR (MGWR), have shown that relationships between COVID-19 outcomes and their determinants are spatially heterogeneous. Empirical studies demonstrated that population density, mobility, climate variables, and socioeconomic indicators exert varying effects across geographic space, relationships that cannot be captured by traditional global regression models [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. These findings are particularly relevant for Saudi Arabia, where provinces differ markedly in population concentration, climate, healthcare infrastructure, and mobility patterns.\u003c/p\u003e \u003cp\u003eRecent methodological advances have also integrated machine learning and network-based approaches with spatiotemporal analysis. Dynamic adaptive spatiotemporal graph networks and deep learning models, such as Bi-LSTM, have significantly improved forecasting accuracy by capturing complex spatial and temporal dependencies in COVID-19 data [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. These approaches suggest promising directions for future spatiotemporal modeling at the provincial level in Saudi Arabia.\u003c/p\u003e \u003cp\u003eOverall, the existing literature demonstrates that spatiotemporal analysis combined with spatial autocorrelation techniques provides a powerful and necessary framework for understanding COVID-19 dynamics. While foundational studies in Saudi Arabia have confirmed the applicability of these methods [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Despite extensive global research on COVID-19, significant gaps remain in understanding its spatiotemporal dynamics within Saudi Arabia. Existing studies often examine spatial or temporal patterns independently, with limited integration of both dimensions at the provincial level. Addressing these gaps requires an integrated spatiotemporal and spatial autocorrelation framework capable of capturing dynamic transmission patterns and regional heterogeneity, thereby supporting more effective, evidence-based public health decision-making in Saudi Arabia.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"3. Materials and Methods","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn this section, the methodology of the research study focusing on Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia's provinces has been comprehensively discussed. The primary objective of this study is to examine the spatiotemporal patterns of COVID-19 transmission across Saudi Arabia\u0026rsquo;s provinces using spatial autocorrelation and hotspot analysis. Additionally, the study aims to utilize the findings to produce risk and social vulnerability maps to guide future interventions and public health strategies. The methodology revolves around three core components:\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Study Area\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe study focuses on Saudi Arabia, which is divided into thirteen administrative provinces with a population of approximately thirty-five million, distributed across 114 governorates, within 13 regions (Table. 1), (Figure. 1).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eshows name of 114 governorates, within 13 regions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"12\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eregion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eGovernorates\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"20\" rowspan=\"21\"\u003e \u003cp\u003e\u003cb\u003eAr-Riyad\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlaflaj\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"11\" rowspan=\"12\"\u003e \u003cp\u003e\u003cb\u003eMakkah Al-Mokarramah\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAljumum\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003e\u003cb\u003eEastern Region\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlahsa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003e\u003cb\u003eAl-Qaseem\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlasyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\" morerows=\"13\" rowspan=\"14\"\u003e \u003cp\u003e\u003cb\u003eJazan\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAlaridah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e\u003cb\u003eHail\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAlghazalah\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlhariq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlkamil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAddammam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlbadai\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAbu Arish\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAsshinan\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAddiriyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlkhurmah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAljubayl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlbukayriyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAhad almusarihah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eBaqa\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAdduwadimi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAllith\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlkhafji\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlmidhnab\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAddair\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eHail\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlghat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlqunfidhah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlkhubar\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAnnabhaniyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAlharth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\" morerows=\"16\" rowspan=\"17\"\u003e \u003cp\u003e\u003cb\u003eAseer\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAbha\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAfif\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAltaif\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlnuayriyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eArrass\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAlidabi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAhad Rifaydah\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlkharj\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eJeddah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlqatif\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAshshimasiyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAlDarb\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAlmajardah\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlmajmaah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKhulays\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eBuqayq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBuraydah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eArrayth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eAnnamas\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlmuzahimiyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMakkah almukarramah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eHafar albatin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eRiyadh alkhabra\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eBaysh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eBalqarn\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlquwayiyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRabigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eQaryah alulya\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eUnayzah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eDamad\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eBishah\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eArriyad\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRanyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRas Tannurah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eUyun aljiwa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eFarasan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eKhamis Mushayt\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAssulayyil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTurubah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"6\" rowspan=\"7\"\u003e \u003cp\u003e\u003cb\u003eAl-Baha\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlbaha\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"6\" rowspan=\"7\"\u003e \u003cp\u003e\u003cb\u003eAl-Madinah Al-Monawarah\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlhinakiyah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eJazan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eMuhayil\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAzzulfi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"8\" rowspan=\"9\"\u003e \u003cp\u003e\u003cb\u003eNajran\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlkharkhir\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlaqiq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlmadinah almunawwarah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSabya\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eRijal Alma\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDuruma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBadr aljanub\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlmandaq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlmahd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSamtah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eSarat Abidah\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHawtat Bani Tamim\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHubuna\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlmukhwah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAlula\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003e\u003cb\u003eTabouk\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAlwajh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eTathlith\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHuraymila\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKhubash\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlqari\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBadr\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eDuba\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003eZahran aljanub\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMarat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNajran\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eBiljurashi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eKhaybar\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eHaqil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRumah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSharurah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eQilwah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eYanbu albahr\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eTabuk\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShaqra\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThar\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eAl-Jouf\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAlqurayyat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eNorthern Borders\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eArar\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eTayma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eThadiq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYadamah\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDawamat aljandal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eRafha\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003eUmluj\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWadi addawasir\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSakaka\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eTurayf\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Data Description\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eSaudi Arabia is divided into 13 administrative provinces with a population of 35\u0026nbsp;million people. The research relies on daily COVID-19 data obtained from the official COVID-19 bulletin, accessible at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/\u003c/span\u003e\u003cspan address=\"https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The first case of COVID-19 in Saudi Arabia was recorded in March 2020 by the Ministry of Health. This study relies on daily COVID-19 data retrieved from the Saudi Ministry of Health COVID-19 response bulletin, which provides several sources of data about the COVID-19 pandemic in Saudi Arabia. It includes various sources of information for use in research. The study period of the data used in this paper extends from 31 March 2020 to 31 August 2022 (three years). The data obtained in dBASE format are then converted into a spreadsheet format. This study relied on cumulative daily data for confirmed, recovered, and death cases of COVID-19 (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Riyadh is the capital of Saudi Arabia that has a population of 8\u0026nbsp;million individuals, so the most COVID-19 confirmed cases were recorded within it\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCumulative numbers of Confirmed cases of Covid-19 at the end of two months of three years in Saudi Arabia.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNAME\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMar. 31\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAug. 31\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMar. 31\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAug. 31\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMar. 31\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAug. 31\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAr-Riyad\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e568\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e67564\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e86982\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e129854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e199562\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e220868\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMakkah Al-Mokarramah\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e507\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e79706\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e92557\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e129457\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e181984\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e199784\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEastern Region\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e311\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e78696\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e93433\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e117278\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e149414\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e160056\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAl-Madinah Al-Monawarah\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e22016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e30830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e39196\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e51439\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e55208\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAl-Qaseem\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e11608\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e14997\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e21490\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28178\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e29643\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAseer\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24504\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e28754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e40791\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e51037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e53894\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAl-Jouf\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1679\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2632\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3803\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3896\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTabouk\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4335\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5335\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8177\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e11496\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e12050\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eJazan\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10779\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e12474\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e20903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e30355\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e32230\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHail\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5697\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7896\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e11586\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e14011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e14444\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNajran\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5482\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6892\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e10560\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e13150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e13628\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNorthern Borders\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1491\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3252\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e6769\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e6936\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAl-Baha\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4926\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7339\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9616\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10824\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Spatiotemporal Analysis\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eSpatiotemporal analysis examines phenomena by considering both their geographic location and their temporal evolution, focusing on where events occur, when they take place, and how they change over time. This approach integrates spatial components\u0026mdash;such as location, spatial distribution, and inter-regional relationships\u0026mdash;with temporal elements, including trends, sequences, rates of change, and diffusion processes. Spatiotemporal series analysis enables the monitoring of variable values across multiple locations over successive time periods, facilitating the examination of dynamic processes such as disease transmission or urban expansion. In addition, spatiotemporal clustering techniques are employed to identify statistically significant hotspots that emerge across both space and time. Distinct pandemic phases were delineated based on statistically significant change points in national COVID-19 case trends identified using the Pettitt test (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003eWithin this framework, spatial autocorrelation plays a fundamental role by assessing the degree of similarity between COVID-19 case numbers in neighboring regions. Spatial statistical measures, particularly Moran\u0026rsquo;s I, are applied to determine whether observed case distributions are randomly dispersed or spatially clustered. When combined with Local Indicators of Spatial Association (LISA), this methodology allows for the detection of localized clusters and spatial outliers, thereby identifying regions that may be particularly vulnerable. In the context of Saudi Arabia, these analytical tools enhance understanding of the spatial diffusion of COVID-19 across provinces and are essential for pinpointing high-risk areas, supporting targeted and effective pandemic response strategies.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 Anselin local Moran\u0026rsquo;s I\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eUtilizing clustering algorithms and methods, the research classifies geographical regions that exhibit comparable COVID-19 patterns. Identifying the optimal number of clusters is a crucial component of the analysis. The objective of the research is to establish clusters in which the regions comprising each group are maximally similar, while the groups as a whole are maximally distinct. The attributes specified for the Analysis Fields parameter, which may include spatial and space-time properties, determine feature similarity within clusters. The different classes of z-values are represented as High-High (HH) or Low-Low (LL). High, positive z-values indicate that an area is surrounded by other areas with similar values. Conversely, a low negative z-score indicates a statistically significant spatial anomaly, i.e. a high-value area surrounded by low-value areas (HL), with LH representing the inverse pattern. In order to determine the spatial correlation between variables, spatial autocorrelation was utilized to match attribute similarity with location similarity. The mathematical expression for Anselin local Moran\u0026rsquo;s I, an index of spatial autocorrelation derived from cross-products, is as follows:\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{I}^{i}=\\frac{{X}_{i}-\\stackrel{-}{X}\\:\\:}{{S}_{i}^{2}}\\:\\sum\\:_{j\\:=1,\\:j\\ne\\:i}^{n}{W}_{ij}\\:({X}_{j}-\\stackrel{-}{X})\\)\u003c/span\u003e \u003c/span\u003e (Eq.\u0026nbsp;1)\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{i}\\)\u003c/span\u003e\u003c/span\u003e is an attribute for feature \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i,\\:\\stackrel{-}{X}\\)\u003c/span\u003e\u003c/span\u003e is the mean of the corresponding attribute, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{W}_{i,j}\\)\u003c/span\u003e\u003c/span\u003e is the spatial weight between feature \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:j\\)\u003c/span\u003e\u003c/span\u003e. Also, where n is the number of regions; xi the attribute value at area I; ̅x the mean value of the attribute in the study region; and wij elements of a spatial lag operator W (spatial weights of matrix W). The significance of the index is usually tested in a situation of normal distribution. Each observation/location is classified into one of the following four categories:\u003c/p\u003e\u003cp\u003eHH - the location is part of a significant high-high correlation cluster, meaning that the location has high autocorrelated and surrounded by other high autocorrelated neighbors.\u003c/p\u003e\u003cp\u003eLL - the location is part of a significant low-low correlation cluster, meaning that the location has low autocorrelation and surrounded by other low autocorrelation neighbors.\u003c/p\u003e\u003cp\u003eHL - the location is an outlier with high autocorrelation but surrounded by neighbors with low autocorrelation.\u003c/p\u003e\u003cp\u003eLH - the location is an outlier with low autocorrelation but surrounded by neighbors with high autocorrelation.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Hot Spot Analysis (Getis-Ord Gi*)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe tool computes the Getis\u0026ndash;Ord Gi* statistic for each spatial feature in a dataset, producing corresponding z-scores and p-values that identify statistically significant spatial clustering of high or low values. This method assesses each feature relative to its surrounding neighbors, recognizing that a feature with a high attribute value alone does not necessarily constitute a significant hotspot. Rather, statistical significance is achieved when a high-value feature is embedded within a neighborhood of similarly high values. The analysis compares the aggregated value of a feature and its neighbors with the expected aggregate across the entire study area. When the observed local sum deviates substantially from the expected value beyond what could be attributed to random variation, a statistically significant z-score is generated. To address issues related to multiple comparisons and spatial dependence, the false discovery rate (FDR) correction is applied, ensuring more reliable identification of hotspots and cold spots.\u003c/p\u003e \u003cp\u003eThe Gi(d) statistic is a distance-based measure that quantifies the concentration of a variable within a specified radius around a given location relative to its distribution across the entire study area. The statistic for location \u003cem\u003ei\u003c/em\u003e is formally expressed as:\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{G}_{i}^{*}=\\frac{\\:{\\sum\\:}_{j=1}^{n}{W}_{ij}\\:{X}_{i}-\\stackrel{-}{X}\\:{\\sum\\:}_{j=1}^{n}{W}_{ij}\\:,j\\:}{S\\:\\sqrt{\\begin{array}{c}\\\\\\:\\frac{n\\:{\\sum\\:}_{j=1}^{n}{{W}^{2}}_{ij}\\:,j\\:(\\:{\\sum\\:}_{j=1}^{n}{W}_{ij}\\:,j{)}^{2}}{n}\\end{array}}}\\)\u003c/span\u003e \u003c/span\u003e (Eq.\u0026nbsp;2)\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{X}_{j}\\)\u003c/span\u003e\u003c/span\u003e is an attribute value for feature \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:j,\\:{W}_{ij}\\)\u003c/span\u003e\u003c/span\u003e is the spatial weight between feature \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:j\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{n}\\)\u003c/span\u003e\u003c/span\u003e is equal to number of features. where xj is the value of the observation at point j; wij(d) the ij element of a binary W matrix (wij\u0026thinsp;=\u0026thinsp;1 if the site is within distance d or 0 if elsewhere; and n the number of observations made. The mean and the variance of this statistic can be obtained through randomization and used to derive a standard statistic. When the value of the standardized statistic is greater than the cut-off value with pre specified significance, positive or negative spatial association exists. Positive values represent spatial agglomeration, while negative values represent the opposite. The higher or lower the z-score, the higher the possibility of clustering, while a z-score close to zero means absence of obvious clusters. Thus, a positive z represents the possibility of clustering, while a negative z indicates a low possibility of clustering. The Gi* statistic returned for each feature in the dataset is a z-score. For statistically significant positive z-scores, the larger the z-score, the more intense the clustering of high values (hot spots). For statistically significant negative z-scores, the smaller the z-score, the more intense the clustering of low values (cold spots). For more information about determining statistical significance and correcting for multiple testing and spatial dependency.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4. Results","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn this section, the outcomes of the proposed methodology are presented and the significant observations derived from these results are discussed. The data analysis that is conducted, as previously described, offers valuable insights into the spatial distribution of variables under study. The research study acknowledges the significance of historical data to understand the progression and future dynamics of the COVID-19 pandemic. To address this, the study focuses on estimating the number of active cases over time by utilizing time-series data, which are sequences of numeric data measured at consistent time intervals. This research employs a multifaceted approach, encompassing data collection, spatial autocorrelation techniques, and spatial clustering analysis, to gain a comprehensive understanding of COVID-19 patterns and spatial relationships within Saudi Arabia's provinces. The findings from this methodology will contribute to the development of effective public health strategies and interventions in response to the pandemic. The outcomes of this spatial clustering analysis will furnish valuable insights regarding the existence of COVID-19 concentrations and regions exhibiting comparable patterns of cases. Such information can significantly assist local, state, and federal health authorities in devising targeted interventions.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Spatiotemporal Hotspot and Anselin Local Moran\u0026rsquo;s I of COVID-19 in Saudi Arabia (2020\u0026ndash;2022)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe series of hotspot and Anselin Local Moran\u0026rsquo;s I (Figure: 2) maps (A\u0026ndash;F) illustrates the spatiotemporal progression of COVID-19 across Saudi Arabia from March 2020 to August 2022. These maps reveal how the spatial distribution of cases evolved over time\u0026mdash;from initial localized clusters to widespread regional transmission and, eventually, to more isolated and fragmented outbreak patterns. By identifying statistically significant hotspots (High-High clusters), coldspots (Low-Low clusters), and spatial outliers, the analysis provides important insights into the geographic dynamics that shaped the pandemic\u0026rsquo;s footprint across the country.\u003c/p\u003e \u003cp\u003eThe earliest phase, in (map, A) represented in March 2020, reflects the initial introduction of COVID-19 into the Kingdom. Hotspots were highly concentrated in the Eastern Region, particularly around Al-Ahsa, Dammam, and Qatif. These areas were among the first to experience community transmission due to dense population centers, large expatriate communities, and early international travel connections. Elsewhere, only small clusters appeared sporadically. In contrast, the southwestern and northwestern regions formed clear coldspots, likely due to lower population density, limited mobility, and the rapid implementation of containment measures. This early spatial pattern marks the virus\u0026rsquo;s entry and the earliest epidemiological signals.\u003c/p\u003e \u003cp\u003eBy August 2020, the maps (map, B) indicate a clear shift from localized outbreaks to widespread regional transmission. Hotspots expanded into the Riyadh Region, Qassim, and several northern governorates. Urban centers such as Jeddah also became statistically significant hotspots. The expansion of red zones across central and eastern Saudi Arabia reflects increased mobility as restrictions eased and summer activities resumed. Coldspots, meanwhile, persisted primarily in the southwestern mountainous regions, where natural geographic isolation and lower population density contributed to lower transmission rates. This stage marks the transition from targeted outbreaks to more systemic nationwide spread.\u003c/p\u003e \u003cp\u003eMarch 2021 (map, C) captures the post-winter wave, characterized by sustained and intensified clustering across major regions. Hotspots remained entrenched in Riyadh, Eastern Province, and parts of the north, demonstrating strong spatial persistence. These persistent clusters reflect areas with high population density, extensive social and economic activity, and greater interregional connectivity. Coldspots in the southwest continued to show stability, reinforcing the spatial disparity in disease dynamics. Local High-Low outliers became more prominent, indicating isolated pockets of high cases surrounded by lower-case areas\u0026mdash;an emerging pattern consistent with complex local transmission dynamics.\u003c/p\u003e \u003cp\u003eBy August 2021, (map, D) the spread of the Delta periods amplified spatial clustering. Hotspots became more densely concentrated and statistically stronger, especially within the Riyadh metropolitan region and the Eastern Province. Spatial outliers increased, revealing more nuanced local patterns of deviation within larger clusters. The persistence of coldspots in the south and southwest again highlights regional differences in transmission potential, mobility behavior, and demographic structure. This phase represents the peak of spatial autocorrelation in the dataset, where neighboring regions exhibited highly similar and high case intensities.\u003c/p\u003e \u003cp\u003eIn March 2022, (map, E) following the Omicron periods wave, the spatial pattern began to change noticeably. Hotspots contracted in size and intensity, though they remained anchored in the central and eastern regions. More spatial outliers appeared, suggesting that transmission had become more fragmented rather than regionally widespread. The reduction in large hotspot zones corresponds with high vaccination rates, natural immunity, and the decreased severity of circulating variants. Coldspots retained their relative stability, reflecting minimal changes in historically low-transmission areas.\u003c/p\u003e \u003cp\u003eThe final map (map, F) from August 2022, represents a post-vaccination stabilization phase. Hotspots became smaller, weaker, and more localized, often linked to isolated outbreaks rather than broad regional waves. The spatial structure shifted away from extensive High-High clusters to a mixed pattern dominated by outliers and isolated pockets of transmission. This fragmentation signifies the transition toward an endemic character, where local outbreaks occur independently but no longer form sustained regional hotspots.\u003c/p\u003e \u003cp\u003eAcross all six time periods, a clear narrative emerges: COVID-19 in Saudi Arabia progressed from localized initial outbreaks to widespread regional clusters, peaking in 2021, before evolving into isolated localized patterns by 2022. The long-term persistence of hotspots in the Eastern Province and Riyadh Region underscores the influence of population density, urbanization, and human mobility in shaping COVID-19 spatial dynamics. Conversely, recurring coldspots in the southwestern regions reflect underlying socioeconomic, demographic, and geographic factors that consistently limited transmission.\u003c/p\u003e \u003cp\u003eOverall, the hotspot and local spatial autocorrelation analysis provides a comprehensive understanding of how COVID-19 spread, intensified, and subsided across the Kingdom. These findings demonstrate the value of spatial epidemiology in guiding targeted interventions, resource allocation, and public health decision-making during a dynamic and evolving pandemic.\u003c/p\u003e \u003cp\u003eStudy area: Governorates of Saudi Arabia Hotspot colors: Red\u0026thinsp;=\u0026thinsp;Hotspot (High\u0026ndash;High clusters), Blue\u0026thinsp;=\u0026thinsp;Coldspot (Low\u0026ndash;Low clusters), Light red/blue\u0026thinsp;=\u0026thinsp;Spatial outliers (High\u0026ndash;Low or Low\u0026ndash;High) and Grey\u0026thinsp;=\u0026thinsp;Non-significant.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e4.2 Spatiotemporal Hotspot Dynamics of COVID-19 in Saudi Arabia (2020\u0026ndash;2022 Getis-Ord Gi Hotspot Analysis\u003c/em\u003e\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe maps in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (A\u0026ndash;F) illustrate the spatiotemporal evolution of COVID-19 hotspots and coldspots across Saudi Arabia from March 2020 to August 2022, based on Getis-Ord General G statistics. Together, they reveal how the spatial intensity of the pandemic changed across governorates over time and how transmission patterns shifted between regions.\u003c/p\u003e \u003cp\u003eIn March 2020 (Map A), during the initial outbreak phase, hotspots were highly concentrated in the Eastern Province, particularly in Qatif, Al-Ahsa, and Dammam. These areas were among the earliest to report community transmission, and their strong regional connectivity facilitated early clustering of cases. A few moderate hotspots emerged in Makkah, while large portions of the western and southern regions acted as coldspots, reflecting minimal spread during the early months of the pandemic.\u003c/p\u003e \u003cp\u003eBy August 2020 (Map B), hotspots expanded substantially. High-intensity clusters appeared across a wider portion of the Eastern Region and began to emerge in central governorates surrounding Riyadh. This period corresponds to the first major national wave of infections, driven by increased mobility and gradual relaxation of movement restrictions. Meanwhile, coldspots persisted across the southwest, consistent with low population density and reduced inter-regional movement.\u003c/p\u003e \u003cp\u003eIn March 2021 (Map C), hotspots intensified in both the Eastern Province and the Riyadh region, indicating sustained transmission in the most urbanized and economically active regions of the country. Additional clusters appeared in the western region near Jeddah, reflecting ongoing urban spread. Coldspots remained prevalent in the southwest, where demographic and geographic factors limited widespread transmission.\u003c/p\u003e \u003cp\u003eBy August 2021 (Map D), hotspot activity became more concentrated, although the same high-risk regions\u0026mdash;Riyadh and the Eastern Province\u0026mdash;continued to dominate. This period aligns with Saudi Arabia\u0026rsquo;s widespread vaccination rollout, which likely contributed to reduced transmission and the contraction of hotspot zones. Coldspot patterns expanded, indicating declining spatial clustering in lower-risk regions.\u003c/p\u003e \u003cp\u003eIn March 2022 (Map E), hotspot clusters fragmented further, appearing only in isolated pockets around Riyadh and the Eastern Province. Most of the country transitioned into nonsignificant or coldspot status, suggesting a notable decrease in spatial dependence of COVID-19 cases and an overall improvement in epidemiological conditions.\u003c/p\u003e \u003cp\u003eFinally, by August 2022 (Map F), hotspot presence was minimal. Only a few governorates in Riyadh and the Eastern Province displayed statistically significant clustering, while the remaining regions indicated stable or low transmission rates. This spatial pattern reflects the stabilization of the pandemic, increased immunity levels, and the sustained effect of national health measures.\u003c/p\u003e \u003cp\u003eOverall, the spatiotemporal hotspot analysis demonstrates a clear progression from early localized outbreaks to broader regional clustering during the pandemic\u0026rsquo;s peak, followed by a gradual dissolution of hotspots as public health interventions took effect. The Eastern Province and Riyadh emerged as persistent high-intensity clusters throughout the study period, whereas the southwestern regions consistently appeared as coldspots. This dynamic evolution underscores the importance of spatial analytics in understanding epidemic progression and supporting targeted health interventions.\u003c/p\u003e \u003c/div\u003e "},{"header":"5. Discussion","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThis study is subject to several data limitations. First, COVID-19 case counts likely underrepresent true infection levels due to asymptomatic cases, limited testing capacity during early pandemic phases, and variations in reporting practices across provinces. Second, testing policies evolved over time, particularly during 2020\u0026ndash;2021, which may have influenced observed temporal trends. Third, potential differences in case definitions and reporting delays across provinces could introduce spatial bias. These factors should be considered when interpreting the results, and future studies should incorporate excess mortality data or seroprevalence surveys to refine spatial estimates.\u003c/p\u003e \u003cp\u003eThe spatiotemporal analysis of COVID-19 in Saudi Arabia from 2020 to 2022 reveals a clear, non-random progression of the pandemic shaped by population density, mobility patterns, and public health interventions. Persistent hotspots in the Eastern Province and Riyadh Region highlight the dominant role of urbanization, economic activity, and interconnected travel networks in sustaining transmission. In contrast, the repeated identification of southwestern governorates as coldspots reflects the protective influence of lower population density, geographic isolation, and reduced mobility, underscoring the importance of localized rather than uniform national responses.\u003c/p\u003e \u003cp\u003eThe spatiotemporal findings of the present study both confirm and extend these established insights by providing a long-term, province-level analysis of COVID-19 dynamics across Saudi Arabia. In agreement with global and regional research, the results reveal strong positive spatial autocorrelation, with early hotspots concentrated in the Eastern Province\u0026mdash;reflecting international connectivity, expatriate labor concentration, and industrial activity\u0026mdash;followed by spatial diffusion toward central and western urban centers as mobility restrictions were relaxed [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. The persistent hotspot status of Riyadh and the Eastern Province mirrors findings from other national and international studies identifying major metropolitan regions as enduring transmission nodes due to high population density and transportation connectivity [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Conversely, the sustained coldspot patterns observed in southwestern provinces align with prior evidence that geographic isolation, lower population density, and reduced mobility can significantly limit disease spread [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The persistent coldspot status of Asir, Jazan, and Al-Baha is likely related to structural geographic and demographic factors\u0026mdash;including mountainous topography, lower population density, climatic conditions, and more limited transport and industrial infrastructure\u0026mdash;which together reduced mobility and interregional transmission, resulting in consistently lower spatial clustering of COVID-19 cases.\u003c/p\u003e \u003cp\u003eCrucially, the temporal evolution of spatial clustering observed in this study addresses key gaps identified in previous research. The intensification of spatial autocorrelation during variant-driven waves\u0026mdash;particularly the Delta phase in mid-2021\u0026mdash;and the subsequent fragmentation of hotspots following widespread vaccination and the emergence of Omicron periods illustrate the dynamic nature of spatial dependence over time [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. These findings reinforce earlier conclusions that static spatial models are inadequate for capturing pandemic dynamics and underscore the necessity of integrated spatiotemporal frameworks capable of identifying shifting hotspots, emerging spatial outliers, and phase-specific transmission mechanisms [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFrom a methodological perspective, the combined application of global and local spatial autocorrelation measures with temporal analysis advances prior work by revealing nuanced sub-regional patterns that would remain obscured under single-method approaches. The identification of local clusters and spatial outliers through LISA and Getis-Ord statistics supports previous assertions that localized spatial analysis is essential for effective public health planning and geographically targeted interventions [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Overall, the convergence between previous studies and the present findings strengthens the evidence that COVID-19 transmission is governed by enduring place-based characteristics and evolving mobility networks, while demonstrating the added value of longitudinal, context-specific spatiotemporal analysis for informing adaptive and regionally tailored pandemic response strategies in Saudi Arabia.\u003c/p\u003e \u003cp\u003eCompared with other Middle Eastern studies, Saudi Arabia exhibited similar spatial patterns to Oman and Iran, where early hotspots emerged in economically active coastal or industrial regions [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, unlike some Gulf states, Saudi Arabia displayed greater spatial persistence of hotspots in its central metropolitan region (Riyadh), likely due to its larger population size and stronger interprovincial mobility networks. This suggests that national urban structure plays a critical role in shaping pandemic spatial dynamics across the region.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThis study applied an integrated spatiotemporal and spatial autocorrelation framework to examine the evolution of COVID-19 across Saudi Arabia\u0026rsquo;s administrative provinces between March 2020 and August 2022. By combining hotspot analysis, Anselin Local Moran\u0026rsquo;s I, Getis-Ord statistics, and clustering techniques, the research provides a comprehensive understanding of how COVID-19 transmission unfolded across space and time within the Kingdom. The findings confirm that the spread of COVID-19 in Saudi Arabia was highly non-random, exhibiting strong spatial dependence and clear temporal phases that reflect both epidemiological and behavioral dynamics.\u003c/p\u003e \u003cp\u003eThe results consistently identified the Eastern Province and the Riyadh Region as persistent hotspots throughout multiple stages of the pandemic. These areas, characterized by high population density, economic activity, and extensive domestic and international mobility, played a central role in sustaining transmission. In contrast, southwestern provinces such as Asir, Jazan, and Al-Baha repeatedly emerged as coldspots, highlighting the influence of geographic isolation, lower population density, and reduced mobility in limiting disease spread. This persistent spatial disparity demonstrates that structural geographic and demographic characteristics strongly condition pandemic outcomes, often independently of short-term policy interventions.\u003c/p\u003e \u003cp\u003eTemporally, the analysis revealed a clear progression from localized outbreaks during the early phase of the pandemic to widespread regional clustering during peak transmission periods in 2020 and 2021. The intensification of spatial clustering during the Delta periods wave, followed by a marked fragmentation of hotspots after the rollout of mass vaccination and the emergence of less severe variants, illustrates the dynamic interaction between viral evolution, human behavior, and public health responses. The transition toward localized and sporadic transmission patterns in 2022 signals a shift toward endemicity within the Saudi Arabian context.\u003c/p\u003e \u003cp\u003eOverall, this research demonstrates the critical value of spatial epidemiology and spatiotemporal analytics in pandemic assessment and response. The integrated methodological approach adopted in this study offers a robust framework for identifying high-risk regions, monitoring evolving transmission patterns, and supporting evidence-based, regionally tailored public health strategies. The insights generated emphasize the necessity of adaptive, spatially differentiated interventions rather than uniform nationwide measures. Beyond COVID-19, the analytical framework and findings of this study provide a transferable model for managing future infectious disease outbreaks and strengthening public health preparedness in Saudi Arabia and comparable settings.\u003c/p\u003e \u003cp\u003eIn summary, the findings support a targeted, spatially differentiated public health strategy that prioritizes enhanced surveillance and testing in persistent hotspots, implements mobility management along high-risk urban corridors, strengthens healthcare capacity in transitional zones, adopts flexible province-specific vaccination approaches, and utilizes real-time spatial dashboards to inform timely decision-making.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eSupplementary Materials:\u0026nbsp;\u003c/strong\u003eThe research relies on daily COVID-19 data obtained from the official COVID-19 bulletin, accessible at https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u0026nbsp;\u003c/strong\u003eFahad Almutlaq.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e The author extends his appreciation to the Deanship of Scientific Research at King Saud ongoing research funding program (ORF-2026-896).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement:\u003c/strong\u003e all Data is open source: COVID-19 bulletin, accessible at https://data.kapsarc.org/explore/assets/saudi-arabia-coronavirus-disease-covid-19-situation/.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e This author expresses ongoing research funding program (ORF-2026-896), King Saud University, Riyadh, Saud Arabia\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest:\u003c/strong\u003e The authors declare no conflict of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWorld Health Organization (WHO). (2020). Coronavirus disease (COVID-19) pandemic. WHO.\u003c/li\u003e\n\u003cli\u003eYe, L., \u0026amp; Hu, L. (2020). Spatiotemporal distribution and trend of COVID-19 in the Yangtze River Delta region of the People\u0026apos;s Republic of China. Geospatial Health, 15(1).\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAnselin, L. (1995). Local indicators of spatial association\u0026mdash;LISA. Geographical Analysis, 27(2), 93\u0026ndash;115.\u003c/li\u003e\n\u003cli\u003eZheng, A., Wang, T., \u0026amp; Li, X. (2021). Spatiotemporal characteristics and risk factors of the COVID-19 pandemic in New York State: Implication of future policies. ISPRS International Journal of Geo-Information, 10(9), 627.\u003c/li\u003e\n\u003cli\u003ePhang, P., Aslam, S., Labadin, J., \u0026amp; Jayaraj, V. J. (2025). Spatial autocorrelation analysis of infectious disease incidence rates at state and district level using supra-adjacency weights matrix. Universal Journal of Public Health, \u003cem\u003e13\u003c/em\u003e(2), 456-470.\u003c/li\u003e\n\u003cli\u003eVilinov\u0026aacute;, K., \u0026amp; Petrikovičov\u0026aacute;, L. (2023). Spatial autocorrelation of COVID-19 in Slovakia. Tropical Medicine and Infectious Disease, 8(6), 298.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAl-Kindi, K. M., Alkharusi, A., Alshukaili, D., Al Nasiri, N., Al-Awadhi, T., Charabi, Y., \u0026amp; El Kenawy, A. M. (2020). Spatiotemporal assessment of COVID-19 spread over Oman using GIS techniques. Earth Systems and Environment, 4(4), 797-811.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eFatima, M., Arshad, S., Butt, I., \u0026amp; Arshad, S. (2021). Geospatial clustering and hot spot detection of COVID-19 incidence in 2020: A global analysis. International Journal of Geospatial and Environmental Research, 8(1).\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eLiu, M., Liu, M., Li, Z., Zhu, Y., Liu, Y., Wang, X., ... \u0026amp; Guo, X. (2021). The spatial clustering analysis of COVID-19 and its associated factors in mainland China at the prefecture level. Science of the Total Environment, 777, 145992.\u003c/li\u003e\n\u003cli\u003eSandar, E., Laohasiriwong, W., \u0026amp; Sornlorm, K. (2023). Spatial autocorrelation and heterogenicity of demographic and healthcare factors in the five waves of COVID-19 epidemic in Thailand. \u003cem\u003eGeospatial Health\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(1).\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAlmobarak, A. S., Almohammadi, H. R., Aboalnaser, S. A., \u0026amp; Syed, L. (2020, December). Spatio-temporal analysis of the spread COVID-19 in Saudi Arabia. In 2020 13th International Conference on Developments in eSystems Engineering (DeSE) (pp. 341-346). IEEE.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAwwad, F. A., Mohamoud, M. A., \u0026amp; Abonazel, M. R. (2021). Estimating COVID-19 cases in Makkah region of Saudi Arabia: Space-time ARIMA modeling. PLoS One, 16(4), e0250149.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAl-Turaiki, I., Almutlaq, F., Alrasheed, H., \u0026amp; Alballa, N. (2021). Empirical evaluation of alternative time-series models for COVID-19 forecasting in Saudi Arabia. International Journal of Environmental Research and Public Health, 18(16), 8660.\u003c/li\u003e\n\u003cli\u003eAlharbi, M. M., Rabbani, S. I., Asdaq, S. M. B., Alamri, A. S., Alsanie, W. F., Alhomrani, M., ... \u0026amp; Alajlan, S. A. (2021). Infection spread, recovery, and fatality from coronavirus in different provinces of Saudi Arabia. Healthcare, 9, 931.\u003c/li\u003e\n\u003cli\u003eAbdel-Aal, M. A., Eltoukhy, A. E., Nabhan, M. A., \u0026amp; AlDurgam, M. M. (2022). Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia. Environmental Science and Pollution Research, 29(14), 20449\u0026ndash;20462.\u003c/li\u003e\n\u003cli\u003eFaisal, K., Alshammari, S., Alotaibi, R., Alhothali, A., Bamasag, O., Alghanmi, N., \u0026amp; Bin Yamin, M. (2022). Spatial analysis of COVID-19 vaccine centers distribution: A case study of the city of Jeddah, Saudi Arabia. International Journal of Environmental Research and Public Health, 19(6), 3526.\u003c/li\u003e\n\u003cli\u003eAlrasheed, H., Alballa, N., Al-Turaiki, I., Almutlaq, F., \u0026amp; Alabduljabbar, R. (2024). City transmission networks: Unraveling disease spread dynamics. ISPRS International Journal of Geo-Information, 13(8).\u003c/li\u003e\n\u003cli\u003eMounesan, L., Farhadi, E., Eybpoosh, S., Hosseini, A., Parsaeian, M., Gharibzadeh, S., Ahmadinezhad, M., Bahari, F., Gouya, M., Haghdoost, A., \u0026amp; Mostafavi, E. (2025). Detecting the seasonal and spatial patterns of COVID-19 hospitalization and deaths in Iran: Insights from a spatiotemporal and hotspot analysis. International Journal of Preventive Medicine.\u003c/li\u003e\n\u003cli\u003eIsazade, V., Qasimi, A. B., Dong, P., Kaplan, G., \u0026amp; Isazade, E. (2023). Integration of Moran\u0026rsquo;s I, geographically weighted regression (GWR), and ordinary least square (OLS) models in spatiotemporal modeling of COVID-19 outbreak in Qom and Mazandaran Provinces, Iran. Modeling Earth Systems and Environment, 9(4), 3923-3937.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eSyetiawan, A., Harimurti, M., \u0026amp; Prihanto, Y. (2022). A spatiotemporal analysis of COVID-19 transmission in Jakarta, Indonesia for pandemic decision support. Geospatial Health; volume 17(s1):1042\u003c/li\u003e\n\u003cli\u003eChen, M., Chen, Y., Wilson, J. P., Tan, H., \u0026amp; Chu, T. (2022). Using an eigenvector spatial filtering-based spatially varying coefficient model to analyze the spatial heterogeneity of COVID-19 and its influencing factors in mainland China. ISPRS International Journal of Geo-Information, 11(1), 67.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eFan, Z., Zhan, Q., Yang, C., Liu, H., \u0026amp; Zhan, M. (2020). How did distribution patterns of particulate matter air pollution (PM2. 5 and PM10) change in China during the COVID-19 outbreak: A spatiotemporal investigation at Chinese city-level. International journal of environmental research and public health, 17(17), 6274.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eAldhyani, T. H., \u0026amp; Alkahtani, H. (2021). A bidirectional long short-term memory model algorithm for predicting COVID-19 in gulf countries. Life, 11(11), 1118.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003ePu, X., Zhu, J., Wu, Y., Leng, C., Bo, Z., \u0026amp; Wang, H. (2024). Dynamic adaptive spatio\u0026ndash;temporal graph network for COVID‐19 forecasting. CAAI Transactions on Intelligence Technology, 9(3), 769-786.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e\u003c/li\u003e\n\u003cli\u003eForuzandeh, M., Neysani Samany, N., \u0026amp; Khodakaramian, B. (2024). A Machine Learning Approach for Modeling the Spatial-temporal Propagation Pattern of COVID-19. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 48, 183-190.\u003cspan dir=\"RTL\"\u003e\u0026rlm;\u003c/span\u003e.\u003c/li\u003e\n\u003cli\u003eLiu, L., Hu, T., Bao, S., Wu, H., Peng, Z., \u0026amp; Wang, R. (2021). The spatiotemporal interaction effect of COVID-19 transmission in the United States. ISPRS International Journal of Geo-Information, 10(6), 387.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Spatial Autocorrelation, Anselin local Moran’s I, Hot Spot Analysis, COVID-19, Saudi Arabia, Public Health Intervention","lastPublishedDoi":"10.21203/rs.3.rs-9418580/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9418580/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study investigates the spatiotemporal dynamics of COVID-19 across Saudi Arabia\u0026rsquo;s administrative provinces from March 2020 to August 2022 using an integrated spatial autocorrelation and hotspot analysis approach. Spatiotemporal patterns were examined using Global and Local Moran\u0026rsquo;s I, Getis-Ord Gi* hotspot analysis, and clustering techniques to identify statistically significant spatial clusters, coldspots, and outliers, as well as their temporal evolution across distinct phases of the pandemic. The results reveal that COVID-19 transmission in Saudi Arabia was highly non-random and characterized by strong spatial dependence. Persistent hotspots were consistently identified in the Riyadh Region and the Eastern Province, reflecting the influence of population density, economic activity, and mobility networks. In contrast, southwestern provinces such as Asir, Jazan, and Al-Baha repeatedly emerged as coldspots, suggesting that geographic isolation and lower population density limited widespread transmission. The study demonstrates the effectiveness of integrating spatiotemporal analysis with spatial autocorrelation methods for understanding pandemic dynamics. The proposed framework provides valuable insights for identifying high-risk areas, optimizing resource allocation, and supporting spatially targeted interventions. This approach offers a transferable model for enhancing epidemic surveillance and preparedness in Saudi Arabia and similar geographic contexts.\u003c/p\u003e","manuscriptTitle":"Spatiotemporal Analysis within Spatial Autocorrelation of COVID-19 in Saudi Arabia’s provinces","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-07 10:01:25","doi":"10.21203/rs.3.rs-9418580/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-05T08:02:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"137041116448081916120778486979395206492","date":"2026-05-01T09:42:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"110353033210584612170258171624319010164","date":"2026-04-30T13:51:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81570502623510013023787763669561964971","date":"2026-04-29T11:22:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"62385539077057825027152761382403201618","date":"2026-04-29T10:20:55+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-29T09:39:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-27T07:04:51+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-22T10:18:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-18T11:39:36+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-18T11:34:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"066fdea7-f418-4ac8-8b35-a617c24268cf","owner":[],"postedDate":"May 7th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-05T08:02:47+00:00","index":63,"fulltext":""},{"type":"reviewerAgreed","content":"137041116448081916120778486979395206492","date":"2026-05-01T09:42:29+00:00","index":62,"fulltext":""},{"type":"reviewerAgreed","content":"110353033210584612170258171624319010164","date":"2026-04-30T13:51:30+00:00","index":61,"fulltext":""},{"type":"reviewerAgreed","content":"81570502623510013023787763669561964971","date":"2026-04-29T11:22:20+00:00","index":58,"fulltext":""},{"type":"reviewerAgreed","content":"62385539077057825027152761382403201618","date":"2026-04-29T10:20:55+00:00","index":57,"fulltext":""},{"type":"reviewersInvited","content":"15","date":"2026-04-29T09:39:29+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67534064,"name":"Health sciences/Diseases"},{"id":67534065,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-07T10:01:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-07 10:01:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9418580","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9418580","identity":"rs-9418580","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}