Next Article in Journal
Singing-Oriented Language and Music Education (SOLME)
Previous Article in Journal
Serious Video Games: Tools for Learning, Training and Health
Previous Article in Special Issue
Visual Analogue Scale
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Encyclopedia
Volume 6
Issue 4
10.3390/encyclopedia6040084
encyclopedia-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Entry

Spatiotemporal Data Science

by
Chaowei Yang
1,*,
Anusha Srirenganathan Malarvizhi
1,
Manzhu Yu
2,
Qunying Huang
3,
Lingbo Liu
4,
Zifu Wang
4,
Daniel Q. Duffy
5,
Siqin Wang
6,
Seren Smith
1,
Shuming Bao
7 and
Nan Ding
8
1
NSF Spatiotemporal Innovation Center, George Mason University, Fairfax, VA 22030, USA
2
Department of Geography, The Pennsylvania State University, State College, PA 16802, USA
3
Department of Geography, University of Wisconsin-Madison, Madison, WI 53705, USA
4
Center for Geographical Analysis, Harvard University, Cambridge, MA 02138, USA
5
NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA
6
Spatial Sciences Institute, University of Southern California, Los Angeles, CA 90089, USA
7
Future Data Lab, Ann Arbor, MI 48106, USA
8
DistrictFirst, Dover, DE 19901, USA
*
Author to whom correspondence should be addressed.
Encyclopedia 2026, 6(4), 84; https://doi.org/10.3390/encyclopedia6040084
Submission received: 5 March 2026 / Revised: 24 March 2026 / Accepted: 25 March 2026 / Published: 6 April 2026
(This article belongs to the Collection Data Science)
Browse Figures
Versions Notes

Definition

The world evolves continuously across space and time. Massive volumes of data are generated through sensing, simulation, remote observation, and human activities, capturing dynamic processes in environmental, social, economic, and engineered systems. Critical insights are embedded within these large-scale spatiotemporal datasets. Spatiotemporal Data Science provides a conceptual and methodological framework for analyzing such data by integrating spatiotemporal thinking, computational infrastructure, artificial intelligence, and domain knowledge. The field advances methods for data acquisition, harmonization, modeling, visualization, and decision support, enabling applications in natural disaster response, public health, climate adaptation, infrastructure resilience, and geopolitical analysis. By leveraging emerging technologies—including generative Artificial Intelligence (AI), large-scale cloud platforms, Graphics Processing Unit (GPU) acceleration, and digital twin systems—Spatiotemporal Data Science enables scalable, interoperable, and solution-oriented research and innovation. It represents a critical frontier for scientific discovery, engineering advancement, technological innovation, education, and societal benefit. Spatiotemporal Data Science is a transdisciplinary field that studies and models dynamic phenomena across space and time by integrating spatial theory, temporal reasoning, artificial intelligence, and scalable computational infrastructure. It enables the development of adaptive, predictive, and increasingly autonomous systems for understanding and managing complex real-world processes.

1. Introduction

Scientific and engineering inquiry has traditionally followed an iterative cycle: observation of phenomena, formulation of hypotheses, data collection, experimentation, validation, and refinement of theory. While this cycle remains foundational, the digital transformation of science has profoundly reshaped how data are collected, stored, analyzed, and interpreted [1].
Since the invention of the computer in the mid-20th century, scientific data have shifted from analog formats to digital representations. Advances in sensing technologies, including satellite remote sensing, in situ sensor networks, mobile devices, Global Position System (GPS) technologies, and Internet-connected IoT devices, have enabled unprecedented volumes of data generation at high spatial and temporal resolutions [2]. These developments underpin transformative visions such as Digital Earth by Al Gore [3], the 4th paradigm data-intensive science initiated by Jim Gray [4], digital twins [5], autonomous systems [6], and large-scale simulation-based scientific discovery [7].
Recent digital transformations have enabled the collection of data at unprecedented spatial and temporal resolutions, thereby enabling scientific and engineering inquiries to be addressed in a spatiotemporal context. The intellectual roots of spatiotemporal thinking trace back to early human activities—estimating distances for hunting, tracking seasonal cycles for agriculture, and organizing movement across landscapes [8]. Formal academic foundations emerged through spatiotemporal statistical analyses [9], the first law of geography [10], time geography [11], environmental modeling [12], regional and geospatial science [13], spatial statistics [14,15], temporal Geographic Information System (GIS) [16], and computing simulation for scientific discovery [17].
Petabyte-level datasets are now common across disciplines, and extracting actionable information requires new methodological frameworks and computing infrastructure [18]. Recent global challenges—including COVID-19 [19], flooding [20], wildfires [21], air pollution [22], supply chain instability [23], and geopolitical conflict [24]—underscore the urgent need for self-adjusting systems capable of perceiving, modeling, and responding to dynamic phenomena as they unfold. These challenges require more than traditional spatial or temporal analysis. They demand integrated spatiotemporal intelligence architectures that combine large-scale data processing, predictive modeling, uncertainty quantification, and real-time decision support. Building upon early foundations in spatial big data science [25,26], this entry conceptualizes Spatiotemporal Data Science at the convergence of (e.g., [27]):
  • Domain sciences, which provide a mechanistic understanding of environmental, climatic, public health, engineering, economic, and social systems [28];
  • Data science and artificial intelligence, including statistical inference, spatial analytics, machine learning, and deep learning [29], enable pattern discovery, predictive reasoning, and model generalization [25];
  • Advanced computational infrastructure, such as cloud-native architectures, GPU acceleration, Field-Programmable Gate Array (FPGA) systems, and edge computing platforms, supporting scalable, real-time intelligence [30];
  • Human spatiotemporal cognition [31], which informs how individuals perceive movement, constraints, risk, and temporal evolution across personal, logistical, and societal systems [11] for analyzing and modeling movements [32];
  • Education and workforce development, which cultivate spatiotemporal reasoning skills and prepare practitioners to design, interpret, and govern intelligent, data-driven systems [33], and use spatiotemporal analytics to evaluate education and pedagogy [34].
Through this convergence, Spatiotemporal Data Science moves beyond static spatial analysis [35] toward an adaptive intelligence paradigm that integrates data-driven learning, physics-based reasoning, and scalable computational infrastructure. This shift enables continuous model updating, cross-domain integration, and increasingly autonomous decision-making in complex, nonstationary systems. The field extends beyond analytical methods to include infrastructure design, interdisciplinary integration, knowledge translation, and operational deployment—transforming descriptive mapping into predictive, feedback-driven modeling of evolving systems. It establishes the foundation for adaptive digital ecosystems that synchronize data, models, and physical environments. Spatiotemporal Data Science transforms static geospatial analysis into adaptive intelligence systems that continuously learn from, predict, and interact with dynamic real-world processes.

2. Historical Development and Key Advances

The conceptual foundation of spatiotemporal analysis can be traced back to ancient times when humans identified hotspots for hunting and agriculture [8], but the formal origin came from geography, time geography, environmental modeling, and spatial statistics, with applications in many science domains (Figure 1).

2.1. Milestones of Spatiotemporal Data Science Evolution

Spatiotemporal Data Science synthesizes concepts from various domains of GIScience, data science, and big data analytics while leveraging advances in computer science and cyberinfrastructure. Driven by grand challenge applications—including climate change [36], climate and environmental disasters, public health and social crises, infrastructure resilience, and global stability—Spatiotemporal Data Science has evolved through a series of foundational breakthroughs that progressively enabled dynamic intelligence and increasingly autonomous decision systems (Figure 1). Key milestones across the domains in this evolution include:
  • Spatial databases and GIS architectures [37], which established the computational foundations for digital representation, querying, and structured management of spatial information. They laid the groundwork for machine-readable geospatial knowledge [38]. Analytical methods such as spatial autocorrelation [9], modifiable area unit [39], and geostatistics [40] were further introduced to enrich the functionality. The success of GIS in the past decades has expanded the architecture from single workstations to cloud-computing-based GIS, which paved one of the architectural foundations for Spatiotemporal Data Science.
  • Time-geographic and temporal modeling frameworks [11], which introduced formal representations of movement, constraints, and dynamic processes, enabling reasoning about change and evolution rather than static patterns [41]. This research has been further enhanced in the past decade to analyze dynamic patterns in various applications, such as detecting GIS objects [42] and human motion [43].
  • Spatiotemporal indexing and scalable storage mechanisms [44], which made it possible to manage and retrieve massive, high-velocity datasets—an essential prerequisite for real-time analytics and adaptive learning [45]. This key advancement not only enabled us to store big spatiotemporal data, but also enabled the efficient analyses and mining of the big data [26,46].
  • Big data frameworks [47] for Earth observation and environmental monitoring, which enabled distributed processing, data exchange, and large-scale model integration across heterogeneous sensing systems. These frameworks paved the path for us to leverage distributed big spatiotemporal data assets in a federated fashion for interoperable access and analyses [48,49].
  • Cloud-native and distributed geospatial computing [18], which introduced elastic, on-demand computational infrastructure capable of supporting continuous model updating and scalable simulation. Cloud and distributed computing provide the on-demand and sometimes real-time processing capabilities for big spatiotemporal data to address emergencies such as flooding [20] and global conflict [24].
  • AI-enabled spatiotemporal prediction and pattern discovery [50], which shifted the field from descriptive analysis toward predictive intelligence, enabling systems to learn evolving dynamics from complex data streams. The last decadal advancement in AI and machine learning helped provide the surrogate models to process big spatiotemporal data for patterns and principles toward prediction and discovery [51,52].

2.2. Relationship to GIScience and Data Science

Though with origins in GIScience [53] and Data Science [54], Spatiotemporal Data Science is a convergence and the latest evolution of relevant technologies and sciences. GIScience provides theoretical foundations for spatial representation and analysis [55]. Data science uses statistical or machine learning methods to extract patterns for insight and prediction [54]. Spatiotemporal Data Science extends the framework toward large-scale data integration, AI-enabled analytics, and evolving intelligence infrastructures (Table 1). Spatiotemporal Data Science emphasizes scalable data-driven modeling and cross-domain integration across dynamic spatial and temporal environments (Table 1).
The convergence of large-scale data ecosystems, cloud infrastructure, and artificial intelligence has transformed Spatiotemporal Data Science from a computational analytics discipline into a foundation for adaptive, increasingly autonomous spatiotemporal intelligence systems. These systems are capable of continuous learning, uncertainty quantification, and decision support in complex, evolving environments. However, challenges remain in data interoperability, model validation, uncertainty quantification, and governance of AI-driven decision systems.

3. Foundations of Spatiotemporal Data Science

Spatiotemporal Data Science is a cross-disciplinary field with contributions from different domains (such as illustrated in Figure 2). The foundations of Spatiotemporal Data Science consist of interconnected components that collectively enable perception, data acquisition, modeling, and decision support for dynamic phenomena across space and time. From a technological perspective, this can be structured into the following components: conceptual architecture, data collection, data management and infrastructure, analytics, visualization and human interaction, and application domains. The following illustrates the technological aspects of the Spatiotemporal Data Science workflow in leveraging the functionalities contributed by the six supporting domains, as also labeled in section numbers in the figure.

3.1. Conceptual Architecture

A modern Spatiotemporal Data Science system operates not as a linear investigation workflow, but as an evolving intelligence lifecycle. It is an iterative, self-improving architecture that integrates perception, reasoning, prediction, and action within a continuous feedback loop. The lifecycle consists of the following interconnected phases:
  • Dynamic problem framing and contextual reasoning: Domain knowledge, physical principles, and policy objectives are formalized into computational representations that define system goals, constraints, and performance metrics. Rather than a static problem definition, this phase enables evolving objective refinement as new evidence emerges in a dynamic manner. For example, wildfire intelligence systems dynamically update risk thresholds and response strategies as environmental conditions change [56].
  • Multi-modal data perception and acquisition: Heterogeneous spatiotemporal data streams—including environmental, infrastructural, and social signals—are continuously ingested to support real-time situational awareness. Data requirements are adaptively adjusted based on evolving model states and uncertainty levels.
  • Cross-scale harmonization and knowledge integration: Data are aligned across spatiotemporal scales, resolutions, and semantic frameworks to construct coherent system representations. This integration enables consistent reasoning across heterogeneous domains and supports modeling and decision synchronization across space and time.
  • Hybrid analytical modeling and predictive reasoning: AI-driven learning models and physics-based simulations interact to generate forecasts, detect anomalies, and evaluate alternative solutions. Models are continuously validated and refined as new spatiotemporal data streams become available, enabling learning-based and self-improving intelligence.
  • Cognitive visualization and human–AI interactive analytics: Multi-dimensional visualization environments and interactive interfaces translate predictive outputs into interpretable insights. They facilitate collaborative reasoning between human experts and intelligent systems in real time.
  • Decision support, proactive response, and feedback integration: Analytical outputs continuously inform policy, operational, or automated responses. Outcomes of these actions are fed back into the system, enabling continuous learning, performance calibration, and refinement of predictive and decision models for dynamic systems.
Together, these phases form a closed-loop intelligence architecture. Rather than terminating at prediction, the system evolves through iterative perception–reasoning–action cycles, enabling progressively adaptive and semi-autonomous decision capabilities within dynamic environments. This lifecycle aligns with emerging paradigms in autonomous systems [57] and digital twin architectures [5], where perception, modeling, and decision-making are continuously coupled through feedback-driven intelligence loops.

3.2. Data Collection and Generation Mechanisms

Spatiotemporal intelligence systems rely on diverse and continuously evolving data streams that function as mechanisms of computational perception. Rather than isolated datasets, these sources collectively form a multi-modal sensing and generation ecosystem through which nonstationary systems are observed, modeled, and acted upon. Key data acquisition and generation modalities include:
  • In situ sensor networks, deployed near observed phenomena to provide high-spatiotemporal-resolution, ground-based measurements such as rainfall, air quality, soil moisture, and infrastructure stress. These sensors enable real-time environmental awareness and localized reactive response [58].
  • Satellite and airborne remote sensing platforms, which provide large-scale, synoptic observation of Earth systems continuously, including wildfire detection, land surface temperature monitoring, atmospheric dynamics, and ocean circulation. These systems enable continuous global surveillance and cross-scale monitoring.
  • IoT devices and edge computing systems, embedded in smartphones, wearable technologies, smart infrastructure, and industrial systems. These devices support distributed sensing and on-device processing in high-velocity, continuous streams, enabling low-latency perception and localized decision-making.
  • Crowdsourced and participatory sensing, capturing near-real-time human-generated signals such as mobility traces, social media activity, and public opinion indicators. These data streams extend environmental perception into social and behavioral domains.
  • Simulation outputs and computational forecasts, including high-resolution weather prediction, wildfire spread modeling, and conflict evolution simulations. These synthetic high-velocity data streams act as predictive extensions of observational systems, enabling anticipatory intelligence.
  • Administrative and transactional datasets, representing economic activities, infrastructure projects, policy interventions, and supply chain dynamics [59]. When integrated with physical observations, these datasets enable cross-domain reasoning about coupled human–environment systems.
Together, these heterogeneous spatiotemporal data streams form a multi-scale, multi-resolution perception layer. Effective spatiotemporal intelligence requires harmonization across spatial scales, temporal resolutions, and semantic structures—enabling coherent fusion, continuous model updating, and adaptive system synchronization.

3.3. Data Management and Infrastructure

Efficient management of large-scale spatiotemporal data [25] is no longer solely a matter of storage and retrieval. It forms the foundational infrastructure for responsive, real-time intelligence systems. Modern spatiotemporal cyberinfrastructure must support continuous data ingestion, dynamic model updating, distributed reasoning, and cross-domain interoperability to enable semi-autonomous and autonomous decision environments. Key components of this intelligence-oriented infrastructure include:
  • Distributed and resilient storage architectures, capable of ingesting streaming data from geographically dispersed sources while ensuring redundancy, fault tolerance, and persistent knowledge representation for high-value datasets.
  • High-speed networking and low-latency communication systems, enabling rapid data synchronization, model updating, and coordination across distributed computational nodes and digital twin instances.
  • Cloud-native and hybrid computing platforms, providing elastic, on-demand computational capacity for large-scale simulation, self-learning, and event-driven analytics, while maintaining long-term operational stability.
  • GPU-accelerated and parallel processing frameworks, supporting large-scale simulation, real-time inference, and the training of foundation and generative models within evolving spatiotemporal environments.
  • Advanced spatiotemporal indexing, compression, and retrieval mechanisms, enabling efficient querying of massive, multi-resolution datasets and supporting near real-time analytical responsiveness [60].
  • Interoperable and federated cyberinfrastructure ecosystems, facilitating secure data exchange, distributed learning, and coordinated model execution across institutions, regions, and domains.
Together, these scalable architectures transform data management systems into agile intelligence substrates. They enable real-time analytics, cross-domain data fusion, continuous model refinement, and the coordinated operation of digital twin ecosystems—forming the computational backbone of increasingly autonomous spatiotemporal intelligence systems.

3.4. Analytical Methods

Analytical methodologies in Spatiotemporal Data Science increasingly function as components of flexible intelligence systems rather than isolated modeling techniques. Leveraging advances in high-performance computing, distributed architectures, and AI acceleration, these approaches integrate statistical inference, machine learning, and physics-based reasoning. They enable predictive, self-updating, and uncertainty-aware decision environments. Core analytical capabilities include:
  • Spatiotemporal data mining, which extracts evolving patterns, anomalies, and dependencies from large-scale and streaming datasets, forming the foundation for dynamic situational awareness.
  • Machine learning and deep learning architectures, capable of learning nonlinear spatiotemporal representations and continuously refining predictive models as new data streams become available.
  • Generative AI and foundation models, which generalize knowledge across domains and scales, enabling transfer learning, scenario simulation, and adaptive forecasting within complex environments.
  • Physics-based and geophysical simulations [61], which encode a mechanistic understanding of natural and engineered systems—providing structured constraints and interpretability to complement data-driven learning.
  • Network and time-geographic models, which capture movement, interaction, and constraint structures across social, economic, and infrastructure systems—supporting anticipatory modeling of cascading effects.
  • Uncertainty quantification and explainable AI frameworks, which provide transparency, probabilistic reasoning, and trust calibration—essential for semi-autonomous and autonomous decision support systems.
A defining characteristic of Spatiotemporal Data Science is the integration of domain-driven mechanistic models with data-driven AI systems. This hybrid intelligence paradigm allows physics-based understanding and machine learning to inform and correct one another. It enables context-aware, self-improving modeling ecosystems that evolve in synchrony with real-world dynamics.

3.5. Visualization and Human Interaction

Effective interpretation of spatiotemporal patterns increasingly functions not merely as visualization, but as a cognitive interface between computational intelligence systems and human decision-makers. Modern spatiotemporal analytics environments integrate visualization, interaction, and adaptive reasoning to enable collaborative human–AI decision processes. Key components include:
  • Multi-dimensional visualization (2D, 3D, and 4D) environments that dynamically represent evolving phenomena, enabling users to explore patterns, simulate alternative scenarios, and evaluate trade-offs across space and time.
  • Virtual and augmented reality platforms, which provide immersive environments for interacting with complex application systems, supporting intuitive spatial reasoning, training, and operational planning in real time.
  • Immersive analytics frameworks, combining visualization, interaction, and real-time data streams to support deeper cognitive engagement with dynamic systems.
  • Interactive and intelligent dashboards, capable of integrating high-velocity streaming data, predictive models, and uncertainty quantification, allowing users to query, test, and refine hypotheses across spatial and temporal scales within specific spatiotemporal contexts.
  • Intelligent decision-support interfaces, customized for operational contexts, that prioritize actionable insights, compress cognitive load, and enable rapid response in high-stakes environments such as disaster management, conflict monitoring, and public health emergencies.
In this evolving paradigm, visualization is no longer a passive reporting mechanism. It becomes an integral component of autonomous spatiotemporal intelligence systems—facilitating human–AI collaboration, enhancing situational awareness, analyzing high-velocity data streams, and supporting semi-autonomous and autonomous decision workflows.

4. Applications

Spatiotemporal Data Science enables adaptive intelligence systems across diverse and rapidly expanding domains. Rather than merely supporting analysis, the field increasingly underpins predictive, anticipatory, and semi-autonomous to autonomous decision environments, including:
  • Climate and environmental intelligence systems, enabling continuous monitoring, risk forecasting, and self-adjusting insurance and policy decisions for coastal resilience, wildfire-driven air quality, and climate-induced hazards.
  • Autonomous disaster risk management systems, fusing multi-source spatiotemporal data to anticipate flooding, hurricane impacts, water quality degradation, and cascading infrastructure failures through predictive modeling and dynamic risk assessment.
  • Transportation and infrastructure intelligence networks, continuously assessing safety zones, aging assets, traffic dynamics, and climate-induced vulnerabilities to enable proactive maintenance, resilience planning, and intelligent mobility management.
  • Urban and smart city ecosystems, integrating sensor networks, mobility data, environmental indicators, and digital twin simulations to optimize land use, energy systems, human movement, and sustainable development strategies [62,63].
  • Public health and pandemic intelligence platforms, integrating real-time surveillance, predictive epidemiological modeling, and scenario simulation to support responsive interventions, policy optimization, and resilient governance during crises such as COVID-19.
  • Conflict monitoring and geopolitical intelligence systems, fusing satellite imagery, social media signals, and ground observations to update evolving front lines and detect emerging tensions in near real time [24].
  • Economic and supply chain intelligence architectures, modeling spatially distributed production networks and trade flows to anticipate policy impacts, mitigate disruptions, and support resilient logistics under uncertainty.
Across these domains, Spatiotemporal Data Science is transitioning from descriptive analytics toward the development of autonomous ecosystems and adaptive intelligence infrastructures capable of continuous learning, real-time prediction, and increasingly autonomous system coordination. As dynamic global systems grow more interconnected, Spatiotemporal Data Science provides the computational foundation for anticipatory governance, resilient engineering, and data-driven societal adaptation.

5. Future Directions: From Analytics to Autonomous Intelligence

Spatiotemporal Data Science continues to advance scientific discovery and engineering innovation toward autonomous intelligence [26]. Future developments are expected in:
  • Autonomous spatiotemporal reasoning systems that integrate domain knowledge, physical principles, and data-driven learning to enable self-adjusting understanding of dynamic environments across scales.
  • Explainable, trustworthy, and self-improving AI architectures capable of continuously discovering patterns from streaming spatiotemporal data, updating models in real time, and quantifying uncertainty to support reliable autonomous decision-making.
  • Real-time digital twin ecosystems that evolve in synchrony with physical systems, enabling simulation, prediction, and intelligent intervention for complex real-world challenges through closed-loop feedback mechanisms.
  • Cross-domain interoperable intelligence infrastructures that integrate heterogeneous spatiotemporal datasets across regional, national, and global systems—supporting coordinated, autonomous responses to interconnected crises such as pandemics, climate extremes, and infrastructure disruptions [19].
  • Open, collaborative intelligence platforms that enable shared data, models, computing resources, and AI agents to co-evolve through federated learning, distributed analytics, and interoperable cyberinfrastructure.
  • Ethical governance frameworks for autonomous spatiotemporal intelligence systems, ensuring transparency, accountability, fairness, and human oversight in digital twin–enabled decision environments.
  • Self-optimizing, resilient societal systems empowered by continuously learning spatiotemporal intelligence, enabling proactive risk mitigation, anticipatory governance, and sustainable system optimization.
  • Risk reduction and trust research, focusing on exploring algorithmic bias [64], protecting ethics and privacy [65]), disparate spatial impacts [66], quantifying uncertainty [67], and filling accountability gaps.
As global systems become increasingly interconnected and dynamic, the ability to model and predict spatiotemporal evolution will become foundational to governance, engineering, and scientific discovery. Spatiotemporal Data Science represents both a methodological consolidation and a transformative frontier for navigating complexity in the Anthropocene.

Author Contributions

Conceptualization, C.Y., M.Y., S.B., D.Q.D. and L.L.; methodology, C.Y. and A.S.M.; formal analysis, C.Y., A.S.M., Q.H., Z.W. and S.W.; investigation, C.Y., S.S. and N.D.; resources, C.Y., D.Q.D., S.B. and N.D.; data curation, C.Y., A.S.M., M.Y., Z.W., L.L. and S.S.; writing—original draft preparation, C.Y., A.S.M., M.Y. and S.S.; writing—review and editing, C.Y., Q.H., L.L., S.S. and S.W.; funding acquisition, C.Y., D.Q.D., S.B. and N.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the NSF I/UCRC Program (1841520) and the NASA AIST Program (80NSSC23K1023) and NASA Goddard CISTO (80NSSC21P2373).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created for this entry.

Acknowledgments

Ideas are formed from research funded by the NSF Spatiotemporal I/UCRC, such as Microsoft, NASA, NOAA, Google, Harris, Northrop Grumman, and others.

Conflicts of Interest

Nan Ding is the Chief Executive Officer at DistrictFirst. All authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
IoTInternet of Things
AIArtificial Intelligence
GPUGraphics Processing Unit
GPSGlobal Positioning System
COVID-19Coronavirus Disease 2019
FPGAField-Programmable Gate Array
2DTwo-Dimensional
3DThree-Dimensional
4DFour-Dimensional
NSFNational Science Foundation
NASANational Aeronautics and Space Administration
NOAANational Oceanic and Atmospheric Administration
CISTOComputational and Information Sciences and Technology Office
AISTAdvanced Information Systems Technology
IUCRCIndustry-University Cooperative Research Center

References

  1. Kraus, S.; Jones, P.; Kailer, N.; Weinmann, A.; Chaparro-Banegas, N.; Roig-Tierno, N. Digital Transformation: An Overview of the Current State of the Art of Research. SAGE Open 2021, 11, 21582440211047576. [Google Scholar] [CrossRef]
  2. Goodchild, M.F. Citizens as sensors: The world of volunteered geography. GeoJournal 2007, 69, 211–221. [Google Scholar] [CrossRef]
  3. Gore, A. The Digital Earth: Understanding our planet in the 21st century. Aust. Surv. 1998, 43, 89–91. [Google Scholar] [CrossRef]
  4. Hey, A.J.; Tansley, S.; Tolle, K.M. The Fourth Paradigm: Data-Intensive Scientific Discovery; Microsoft research: Redmond, WA, USA, 2009; Volume 1. [Google Scholar]
  5. Grieves, M.; Vickers, J. Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems. In Transdisciplinary Perspectives on Complex Systems: New Findings and Approaches; Springer International Publishing: Cham, Switzerland, 2016; pp. 85–113. [Google Scholar]
  6. Geiger, A.; Lenz, P.; Urtasun, R. Are we ready for autonomous driving? The KITTI vision benchmark suite. In Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI, USA, 16–21 June 2012; pp. 3354–3361. [Google Scholar]
  7. De Jong, T.; Van Joolingen, W.R. Scientific discovery learning with computer simulations of conceptual domains. Rev. Educ. Res. 1998, 68, 179–201. [Google Scholar] [CrossRef]
  8. Anderson, J.K. Hunting in the Ancient World; University of California Press: Berkeley, CA, USA, 1985; p. 192. [Google Scholar]
  9. Moran, P.A. Notes on continuous stochastic phenomena. Biometrika 1950, 37, 17–23. [Google Scholar] [CrossRef] [PubMed]
  10. Tobler, W.R. A computer movie simulating urban growth in the Detroit region. Econ. Geogr. 1970, 46, 234–240. [Google Scholar] [CrossRef]
  11. Hägerstrand, T. What about people in regional science. Transp. Sociol. Soc. Asp. Transp. Plan. 1970, 35, 143–158. [Google Scholar] [CrossRef]
  12. Jones, J.W.; Colwick, R.F.; Threadgi, E.D. A simulated environmental model of temperature, rainfall, evaporation, and soil moisture. Agric. Eng. 1970, 51, 291. [Google Scholar]
  13. Isard, W.; Bramhall, D.F.; Carrothers, G.A.P.; Cumberland, J.H.; Moses, L.N.; Price, D.O.; Schooler, E.W. Methods of Regional Analysis: An Introduction to Regional Science; The MIT Press: Cambridge, MA, USA, 1966. [Google Scholar]
  14. Griffith, D.A. Towards a theory of spatial statistics. Geogr. Anal. 1980, 12, 325–339. [Google Scholar] [CrossRef]
  15. Cressie, N.; Wikle, C.K. Statistics for Spatio-Temporal Data; John Wiley & Sons: Hoboken, NJ, USA, 2011. [Google Scholar]
  16. Yuan, M. Temporal GIS and spatio-temporal modeling. In Proceedings of the Third International Conference Workshop on Integrating GIS and Environment Modeling, Santa Fe, NM, USA, 21–26 January 1996; Volume 33. [Google Scholar]
  17. Bradshaw, G.F.; Langley, P.W.; Simon, H.A. Studying scientific discovery by computer simulation. Science 1983, 222, 971–975. [Google Scholar] [CrossRef] [PubMed]
  18. Yang, C.; Huang, Q. Spatial Cloud Computing: A Practical Approach; CRC Press: Boca Raton, FL, USA, 2013. [Google Scholar]
  19. Yang, C.; Bao, S.; Guan, W.; Howell, K.; Hu, T.; Lan, H.; Li, Y.; Liu, Q.; Smith, J.; Srirenganathan, A.; et al. Challenges and opportunities of the spatiotemporal responses to the global pandemic of COVID-19. Ann. GIS 2022, 28, 425–434. [Google Scholar] [CrossRef]
  20. Cian, F.; Marconcini, M.; Ceccato, P. Normalized Difference Flood Index for rapid flood mapping: Taking advantage of EO big data. Remote. Sens. Environ. 2018, 209, 712–730. [Google Scholar] [CrossRef]
  21. Westerling, A.L.; Hidalgo, H.G.; Cayan, D.R.; Swetnam, T.W. Warming and earlier spring increase western US forest wildfire activity. Science 2006, 313, 940–943. [Google Scholar] [CrossRef]
  22. Dockery, D.W.; Pope, C.A.; Xu, X.; Spengler, J.D.; Ware, J.H.; Fay, M.E.; Ferris, B.G., Jr.; Speizer, F.E. An association between air pollution and mortality in six US cities. New Engl. J. Med. 1993, 329, 1753–1759. [Google Scholar] [CrossRef] [PubMed]
  23. Ivanov, D. Two views of supply chain resilience. Int. J. Prod. Res. 2023, 62, 4031–4045. [Google Scholar] [CrossRef]
  24. Wang, Z.; Masri, Y.; Malarvizhi, A.S.; Stover, T.; Ahmed, S.; Wong, D.; Jiang, Y.; Li, Y.; Bere, M.; Rothbert, D.; et al. Optimizing context-based location extraction by tuning open-source LLMs with RAG. Int. J. Digit. Earth 2025, 18, 2521786. [Google Scholar] [CrossRef]
  25. Jiang, Z.; Shekhar, S. Spatial and spatiotemporal big data science. In Spatial Big Data Science: Classification Techniques for Earth Observation Imagery; Springer Nature: Berlin/Heidelberg, Germany, 2017; pp. 15–44. [Google Scholar]
  26. Yang, C.; Clarke, K.; Shekhar, S.; Tao, C.V. Big Spatiotemporal Data Analytics: A research and innovation frontier. J. Geogr. Inf. Sci. 2020, 34, 1075–1088. [Google Scholar] [CrossRef]
  27. Amato, F.; Lombardo, L.; Tonini, M.; Marvuglia, A.; Castro-Camilo, D.; Guignard, F. Spatiotemporal data science: Theoretical advances and applications. Stoch Env. Res Risk Assess 2022, 36, 2027–2029. [Google Scholar] [CrossRef]
  28. Doshi-Velez, F.; Kim, B. Towards a rigorous science of interpretable machine learning. arXiv 2017, arXiv:1702.08608. [Google Scholar]
  29. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef]
  30. Dikaiakos, M.D.; Katsaros, D.; Mehra, P.; Pallis, G.; Vakali, A. Cloud Computing: Distributed Internet Computing for IT and Scientific Research. IEEE Internet Comput. 2009, 13, 10–13. [Google Scholar] [CrossRef]
  31. Northoff, G.; Wolman, A.; Zhang, J. Brain dynamics shape cognition–Spatiotemporal Neuroscience. Phys. Life Rev. 2025, 54, 173–201. [Google Scholar] [CrossRef]
  32. Sun, L.; Jia, K.; Yeung, D.Y.; Shi, B.E. Human action recognition using factorized spatio-temporal convolutional networks. In Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December 2015; pp. 4597–4605. [Google Scholar]
  33. Lu, J. Empowering Community College Students’ Geospatial Education Through Virtual Practical Training. J. Geogr. 2024, 123, 123–128. [Google Scholar] [CrossRef]
  34. Lingard, B. Relations and locations: New topological spatio-temporalities in education. Eur. Educ. Res. J. 2022, 21, 983–993. [Google Scholar] [CrossRef]
  35. Anselin, L. Spatial Econometrics: Methods and Models; Springer Science & Business Media: Berlin/Heidelberg, Germany, 1988; Volume 4. [Google Scholar]
  36. Laurini, M.P. A spatio-temporal approach to estimate patterns of climate change. Environmetrics 2019, 30, e2542. [Google Scholar] [CrossRef]
  37. Smith, D.A.; Tomlinson, R.F. Assessing costs and benefits of geographical information systems: Methodological and implementation issues. Int. J. Geogr. Inf. Syst. 1992, 6, 247–256. [Google Scholar] [CrossRef]
  38. Shekhar, S.; Xiong, H. (Eds.) Encyclopedia of GIS; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2007. [Google Scholar]
  39. Fotheringham, A.S.; Wong, D.W.S. The Modifiable Areal Unit Problem in Multivariate Statistical Analysis. Environ. Plan. A Econ. Space 1991, 23, 1025–1044. [Google Scholar] [CrossRef]
  40. Journel, A.G. Geostatistics for conditional simulation of ore bodies. Econ. Geol. 1974, 69, 673–687. [Google Scholar] [CrossRef]
  41. Chen, B.Y.; Yuan, H.; Li, Q.; Shaw, S.L.; Lam, W.H.; Chen, X. Spatiotemporal data model for network time geographic analysis in the era of big data. Int. J. Geogr. Inf. Sci. 2016, 30, 1041–1071. [Google Scholar] [CrossRef]
  42. Ardeshir, S.; Zamir, A.R.; Torroella, A.; Shah, M. GIS-assisted object detection and geospatial localization. In Proceedings of the European Conference on Computer Vision, Zurich, Switzerland, 6–12 September 2014; Springer International Publishing: Cham, Switzerland, 2014; pp. 602–617. [Google Scholar]
  43. Liu, J.; Mahdavi-Amiri, A.; Savva, M. Paris: Part-level reconstruction and motion analysis for articulated objects. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Paris, France, 2–6 October 2023; pp. 352–363. [Google Scholar]
  44. Theodoridis, Y.; Sellis, T.; Papadopoulos, A.N.; Manolopoulos, Y. Specifications for efficient indexing in spatiotemporal databases. In Proceedings of the Tenth International Conference on Scientific and Statistical Database Management, Capri, Italy, 1–3 July 1998; IEEE: New York, NY, USA, 1998; pp. 123–132. [Google Scholar]
  45. Xia, J.; Yang, C.; Li, Q. Building a spatiotemporal index for earth observation big data. Int. J. Appl. Earth Obs. Geoinf. 2018, 73, 245–252. [Google Scholar] [CrossRef]
  46. Vatsavai, R.R.; Ganguly, A.; Chandola, V.; Stefanidis, A.; Klasky, S.; Shekhar, S. Spatiotemporal data mining in the era of big spatial data: Algorithms and applications. In Proceedings of the 1st ACM SIGSPATIAL international workshop on analytics for big geospatial data, Redondo Beach, CA, USA, 6 November 2012; pp. 1–10. [Google Scholar]
  47. Han, D.; Stroulia, E. Hgrid: A data model for large geospatial data sets in hbase. In Proceedings of the 2013 IEEE sixth international conference on cloud computing, Santa Clara, CA, USA, 28 June–3 July 2013; pp. 910–917. [Google Scholar]
  48. Alam, M.M.; Torgo, L.; Bifet, A. A survey on spatio-temporal data analytics systems. ACM Comput. Surv. 2022, 54, 219. [Google Scholar] [CrossRef]
  49. Mohamed, A.; Najafabadi, M.K.; Wah, Y.B.; Zaman, E.A.K.; Maskat, R. The state of the art and taxonomy of big data analytics: View from new big data framework. Artif. Intell. Rev. 2020, 53, 989–1037. [Google Scholar] [CrossRef]
  50. Xu, L.; Chen, N.; Chen, Z.; Zhang, C.; Yu, H. Spatiotemporal forecasting in earth system science: Methods, uncertainties, predictability and future directions. Earth-Sci. Rev. 2021, 222, 103828. [Google Scholar] [CrossRef]
  51. Shekhar, S.; Jiang, Z.; Ali, R.Y.; Eftelioglu, E.; Tang, X.; Gunturi, V.M.; Zhou, X. Spatiotemporal data mining: A computational perspective. ISPRS Int. J. Geo-Inf. 2015, 4, 2306–2338. [Google Scholar] [CrossRef]
  52. Wang, J.; Tang, J.; Xu, Z.; Wang, Y.; Xue, G.; Zhang, X.; Yang, D. Spatiotemporal modeling and prediction in cellular networks: A big data enabled deep learning approach. In Proceedings of the IEEE Infocom 2017—IEEE Conference on Computer Communications, Atlanta, GA, USA, 1–4 May 2017; IEEE: New York, NY, USA, 2017; pp. 1–9. [Google Scholar]
  53. Goodchild, M.F. Geographical information science. Int. J. Geogr. Inf. Syst. 1992, 6, 31–45. [Google Scholar] [CrossRef]
  54. Cleveland, W.S. Data science: An action plan for expanding the technical areas of the field of statistics. Int. Stat. Rev. 2001, 69, 21–26. [Google Scholar] [CrossRef]
  55. Longley, P.A.; Goodchild, M.F.; Maguire, D.J.; Rhind, D.W. Geographic Information Science and Systems; John Wiley & Sons: Hoboken, NJ, USA, 2015. [Google Scholar]
  56. Koh, J.; Pimont, F.; Dupuy, J.L.; Opitz, T. Spatiotemporal wildfire modeling through point processes with moderate and extreme marks. Ann. Appl. Stat. 2023, 17, 560–582. [Google Scholar] [CrossRef]
  57. Li, Z.; Ning, H.; Gao, S.; Janowicz, K.; Li, W.; Arundel, S.T.; Yang, C.; Bhaduri, B.; Wang, S.; Hodgson, M.E. Giscience in the era of artificial intelligence: A research agenda towards autonomous gis. Ann. GIS 2025, 31, 501–536. [Google Scholar] [CrossRef]
  58. Smith, S.; Trefonides, T.; Srirenganathan Malarvizhi, A.; LaGarde, S.; Liu, J.; Jia, X.; Wang, Z.; Cain, J.; Huang, T.; Pourhomayoun, M. A Systematic Study of Popular Software Packages and AI/ML Models for Calibrating In Situ Air Quality Data: An Example with Purple Air Sensors. Sensors 2025, 25, 1028. [Google Scholar] [CrossRef]
  59. Wang, X.; Wang, L.; Zhang, X.; Fan, F. The spatiotemporal evolution of COVID-19 in China and its impact on urban economic resilience. China Econ. Rev. 2022, 74, 101806. [Google Scholar] [CrossRef] [PubMed]
  60. Li, Z.; Hu, F.; Schnase, J.L.; Duffy, D.Q.; Lee, T.; Bowen, M.K.; Yang, C. A spatiotemporal indexing approach for efficient processing of big array-based climate data with MapReduce. Int. J. Geogr. Inf. Sci. 2017, 31, 17–35. [Google Scholar] [CrossRef]
  61. Rohrlich, F. Computer simulation in the physical sciences. In PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association; Cambridge University Press: Cambridge, UK, 1990; Volume 19, pp. 507–518. [Google Scholar]
  62. Batty, M. The New Science of Cities; MIT Press: Cambridge, MA, USA, 2013. [Google Scholar]
  63. Tang, L.; Gao, J.; Ren, C.; Zhang, X.; Yang, X.; Kan, Z. Detecting and evaluating urban clusters with spatiotemporal big data. Sensors 2019, 19, 461. [Google Scholar] [CrossRef] [PubMed]
  64. Franklin, G.; Stephens, R.; Piracha, M.; Tiosano, S.; Lehouillier, F.; Koppel, R.; Elkin, P.L. The sociodemographic biases in machine learning algorithms: A biomedical informatics perspective. Life 2024, 14, 652. [Google Scholar] [CrossRef]
  65. Chen, Y.; Esmaeilzadeh, P. Generative AI in Medical Practice: In-Depth Exploration of Privacy and Security Challenges. J. Med. Internet Res. 2024, 26, e53008. [Google Scholar] [CrossRef]
  66. Török, I. Qualitative Assessment of Social Vulnerability to Flood Hazards in Romania. Sustainability 2018, 10, 3780. [Google Scholar] [CrossRef]
  67. Malarvizhi, A.S.; Smith, K.; Yang, C. Uncertainty quantification in geospatial AI/ML applications: Methods, metrics, and open-source support with an air quality use case. Big Earth Data 2026, 1–34. [Google Scholar] [CrossRef]
Encyclopedia 06 00084 g001
Figure 1. Spatiotemporal Data Science is driven by grand challenges and enabled by various advancements in spatiotemporal studies, and research initiatives such as the NSF Spatiotemporal Innovation Center have further accelerated the development of the field with publications, workforce training, and popular research results [19,30].
Figure 1. Spatiotemporal Data Science is driven by grand challenges and enabled by various advancements in spatiotemporal studies, and research initiatives such as the NSF Spatiotemporal Innovation Center have further accelerated the development of the field with publications, workforce training, and popular research results [19,30].
Encyclopedia 06 00084 g001
Encyclopedia 06 00084 g002
Figure 2. Conceptual framework of Spatiotemporal Data Science as a convergence of computer science, cyberinfrastructure, geospatial analytics, and data science, supporting application domains spanning environmental, climate, health, and social systems.
Figure 2. Conceptual framework of Spatiotemporal Data Science as a convergence of computer science, cyberinfrastructure, geospatial analytics, and data science, supporting application domains spanning environmental, climate, health, and social systems.
Encyclopedia 06 00084 g002
Table 1. The relationship between Data Science, GIScience, and Spatiotemporal Data Science.
Table 1. The relationship between Data Science, GIScience, and Spatiotemporal Data Science.
DimensionData ScienceGIScienceSpatiotemporal Data Science
Primary FocusPatterns in dataSpatial representation and analysisDynamic systems across space and time
SpaceOptionalCentralFundamental
TimeOften simplifiedLimited/secondaryCore (continuous, evolving)
ScaleDataset-levelMap/layer-levelMulti-scale, real-time, streaming
MethodsStatistics, MLSpatial analysis, cartographyAI + physics + spatial-temporal modeling
GoalInsight and predictionUnderstanding spatial relationshipsPrediction, simulation, and decision support
System TypeAnalytical workflowsGIS systemsIntelligent, adaptive infrastructures
Intelligence LevelAnalyticalSpatial reasoningAdaptive/predictive/autonomous
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yang, C.; Malarvizhi, A.S.; Yu, M.; Huang, Q.; Liu, L.; Wang, Z.; Duffy, D.Q.; Wang, S.; Smith, S.; Bao, S.; et al. Spatiotemporal Data Science. Encyclopedia 2026, 6, 84. https://doi.org/10.3390/encyclopedia6040084

AMA Style

Yang C, Malarvizhi AS, Yu M, Huang Q, Liu L, Wang Z, Duffy DQ, Wang S, Smith S, Bao S, et al. Spatiotemporal Data Science. Encyclopedia. 2026; 6(4):84. https://doi.org/10.3390/encyclopedia6040084

Chicago/Turabian Style

Yang, Chaowei, Anusha Srirenganathan Malarvizhi, Manzhu Yu, Qunying Huang, Lingbo Liu, Zifu Wang, Daniel Q. Duffy, Siqin Wang, Seren Smith, Shuming Bao, and et al. 2026. "Spatiotemporal Data Science" Encyclopedia 6, no. 4: 84. https://doi.org/10.3390/encyclopedia6040084

APA Style

Yang, C., Malarvizhi, A. S., Yu, M., Huang, Q., Liu, L., Wang, Z., Duffy, D. Q., Wang, S., Smith, S., Bao, S., & Ding, N. (2026). Spatiotemporal Data Science. Encyclopedia, 6(4), 84. https://doi.org/10.3390/encyclopedia6040084

Article Metrics

Back to TopTop