Spatial Data Science Research Articles

Using Spatial Data Science in Energy-Related Modeling of Terraforming the Martian Atmosphere

This paper proposes a methodology for numerical modeling of terraforming Mars’ atmosphere using high-energy asteroid impact and greenhouse gas production processes. The developed simulation model uses a spatial data science approach to analyze the Global Climate Model of Mars and cellular automata to model the changes in Mars’ atmospheric parameters. The developed model allows estimating the energy required to raise the planet’s temperature by sixty degrees using different variations of the terraforming process. Using a data science approach for spatial big data analysis has enabled successful numerical simulations of global and local atmospheric changes on Mars and an analysis of the energy potential required for this process.

Open Source Software for Spatial Data Science

Much progress has been made in the development of software tools for spatial analysis since the special issue of Geographical Analysis appeared in 2006, devoted to “Recent advances in software for spatial analysis in the social sciences” (Rey and Anselin 2006). The 15 some years since the publication of the issue have been marked by major changes in the spatial analytical software landscape. Arguably, three important and somewhat related phenomena can be distinguished that drove these changes: the embedding of spatial analysis into spatial data science; the growing recognition of open science/open source principles in empirical work; and the increasing adoption of a literate programming perspective.

Application of Spatial Data Science on Results of the Saber 5 Test

Application of Spatial Data Science on Results of the Saber 5 Test

Interpretable machine learning for analysing heterogeneous drivers of geographic events in space-time

ABSTRACT Machine learning (ML) interpretability has become increasingly crucial for identifying accurate and relevant structural relationships between spatial events and factors that explain them. Methodologically aspatial ML algorithms with an apparent high predictive power ignore non-stationary domain relationships in spatio-temporal data (e.g. dependence, heterogeneity), leading to incorrect interpretations and poor management decisions. This study addresses this critical methodological issue of ‘interpretability’ in ML-based modeling of structural relationships using the example of heterogeneous drivers of wildfires across the United States. Specifically, we present and evaluate a spatio-temporally interpretable random forest (iST-RF) that uses spatio-temporal sampling-based training and weighted prediction. Although the ultimate scientific objective is to derive interpretation in space-time, experiments show that iST-RF can improve predictive accuracy (76%) compared to the aspatial RF approach (70%) while enhancing interpretations of the trained model’s spatio-temporal relevance for its ensemble prediction. This novel approach can help balance prediction and interpretation with fidelity in a spatial data science life cycle. However, challenges exist for predictive modeling when the dataset is very small because in such cases locally optimized sub-model’s prediction performance can be suboptimal. With that caveat, our proposed approach is an ideal choice for identifying drivers of spatio-temporal events at country- or regional-scale studies.

The intersection of climate-related health outcomes and urban environmental characteristics using spatiotemporal data science

BACKGROUND AND AIM: Cities are home to 80% of Canadians of whom have differential vulnerabilities, in population-specific contexts, to climate change impacts on their health. To identify areas within one city - Edmonton, Alberta - experiencing current climate-susceptible health impacts, we explored spatiotemporal patterns of chronic disease and a combination of urban environmental factors. Identifying areas of convergence may help identify where climate change-related severe weather events and worsened air pollution could magnify cardiovascular, renal, respiratory, neurological, mental health, injury, and pregnancy outcomes. METHODS: We merged 1.2M records of patients (2013-2018) from the National Ambulatory Care Reporting System and the Discharge Abstract Database and classified the relevant conditions into the above seven health outcomes. All variables were aggregated/assigned to 1,196 dissemination areas (DA); environmental variables were sourced from the National Air Pollution Surveillance, National Pollutant Release Inventory, Alberta Climate Information System, vegetation “greenness,” and material-social deprivation indices. We analyzed their respective data distributions accounting for both space and time by transforming the events of each health grouping, using seasonal (3-month) time intervals for each DA, into space-time cubes. Then, we input the space-time cubes into emerging hot spot analysis to calculate statistically significant hot and cold spots. RESULTS:Geovisualization of 3-dimensional cubes and 2-dimensional maps helped us identify urban areas with extreme or unusual patterns in health outcomes. Each of the seven health groupings revealed different patterns, which typically coincided with demographics. Through spatial overlays and statistics, we associated the spatiotemporal health-exposure intersections to aid in variable selection for a future vulnerability index. CONCLUSIONS:Our space-time explorations identify where these exposed populations occur and may help target appropriate climate change adaptive capacity interventions. This research is part of a larger project that is generating deeper insight into multiple factors that affect the climate change vulnerabilities and resilience of Alberta populations. KEYWORDS: Exposures, Multi-pollutant, Air pollution, Temperature extremes and variability, Chronic diseases, Spatial data science

ThePySALEcosystem: Philosophy and Implementation

PySAL is a library for geocomputation and spatial data science. Written in Python, the library has a long history of supporting novel scholarship and broadening methodological impacts far afield of academic work. Recently, many new techniques, methods of analyses, and development modes have been implemented, making the library much larger and more encompassing than that previously discussed in the literature. As such, we provide an introduction to the library as it stands now, as well as the scientific and conceptual underpinnings of its core set of components. Finally, we provide a prospective look at the library's future evolution.

Spatial data science for data driven urban planning: The youth economic discomfort index for Rome

Today, a consistent segment of the world’s population lives in urban areas, and this proportion will vastly increase in the next decades. Therefore, understanding the key trends in urbanization, likely to unfold over the coming years, is crucial to the implementation of sustainable urban strategies. In parallel, the daily amount of digital data produced will be expanding at an exponential rate during the following years. The analysis of various types of data sets and its derived applications have incredible potential across different crucial sectors such as healthcare, housing, transportation, energy, and education. Nevertheless, in city development, architects and urban planners appear to rely mostly on traditional and analogical techniques of data collection. This paper investigates the prospective of the data science field, appearing to be a formidable resource to assist city managers in identifying strategies to enhance the social, economic, and environmental sustainability of our urban areas. The collection of different new layers of information would definitely enhance planners' capabilities to comprehend more in-depth urban phenomena such as gentrification, land use definition, mobility, or critical infrastructural issues. Specifically, the research results correlate economic, commercial, demographic, and housing data with the purpose of defining the youth economic discomfort index. The statistical composite index provides insights regarding the economic disadvantage of citizens aged between 18 years and 29 years, and results clearly display that central urban zones and more disadvantaged than peripheral ones. The experimental set up selected the city of Rome as the testing ground of the whole investigation. The methodology aims at applying statistical and spatial analysis to construct a composite index supporting informed data-driven decisions for urban planning.

Don’t forget about geography

Maps are a fundamental form of human communication, and for millennia geographers have created maps that measure and describe features and phenomena on the Earth’s surface. Yet since the ‘quantitative revolution’ of the 1960s, the ancient scientific discipline of geography has become increasingly devalued within the academe and misunderstood by the general public. A review of the academic affiliations and job titles of the esteemed authors from the JOSIS 10th anniversary edition is indicative of how constant rebranding and renaming of geography has resulted in fragmentation of the discipline. While terms such as ‘Spatial Data Science’ have a cross-disciplinary appeal, other terms such as ‘geoinformation’, ‘geoinformatics’, ‘geographic data science’, and ‘geographical information science’ primarily conflate geography and computer science. Geographers have been valued for our ability to addressed complex problems and create maps that cross scientific boundaries since antiquity. To reclaim a position of centrality within the academe and the minds of public we must be unequivocal that the central value proposition for geography is the fundamental form of human communication that geographers can truly claim as their own: the map.

Big issues for big data: challenges for critical spatial data analytics

In this paper we consider some of the issues of working with big data and big spatial data and highlight the need for an open and critical framework. We focus on a set of challenges underlying the collection and analysis of big data. In particular, we consider 1) inference when working with usually biased big data, challenging the assumed inferential superiority of data with observations, n, approaching N, the population n -> N. We also emphasise 2) the need for analyses that answer questions of practical significance or with greater emphasis on the size of the effect, rather than the truth or falsehood of a statistical statement; 3) the need to accept messiness in your data and to document all operations undertaken on the data because of this, in support of openness and reproducibility paradigms; and 4) the need to explicitly seek to understand the causes of bias, messiness etc in the data and the inferential consequences of using such data in analyses, by adopting critical approaches to spatial data science. In particular we consider the need to place individual data science studies in a wider social and economic contexts, along with the role of inferential theory in the presence of big data, and issues relating to messiness and complexity in big data.

Spatial Analysis and Modeling: Advances and Evolution

This review is based on a plenary lecture delivered at a recent annual meeting of the American Association of Geographers, sponsored by the Spatial Analysis and Modeling specialty group. This group began around 1979 as Mathematical Models & Quantitative Methods, with a subsequent name change in 2000 to better reflect the maturing interests and concerns of members. Some twenty years down the road, advances in computing, methods, and models across academic disciplines have brought about many advances and evolution in capabilities. This article discusses a number of important spatial analytical methods with the intent of highlighting understanding and solution. Further, this is based upon development efforts rooted in the fundamentals of GIScience and spatial data science. Three spatial analytical approaches are reviewed, offering insight regarding evolved capabilities. Of particular emphasis is the ability to take advantage of derived geographic knowledge in order to structure, solve, and apply these methods.

Discovering the Unexpected & Explicable: Scientific Reasoning and Research Design for Spatial Data Analysis

Abstract. Methods for exploratory data analysis and exploratory spatial data analysis (or ESDA) are useful to identify outliers, clusters, skewed distributions and correlations (see Figure 1 for examples implemented in our GeoDa software). Researchers routinely use these methods to find insights.However, what motivated this project is that, by default, it is easier to find insights that confirm the expected. Often, in fields like geography, statistics or computer science, available software and data drive the choice of research questions and the process of how we explore data. Typical insights gained this way are descriptive, like where a cluster is located or whether variables are correlated.To be sure, expected insights are reassuring. And descriptive insights are important. But instead of stopping here we want to build on them to also find insights that are new and relevant – we want to discover the unexpected (Anselin, 1998; Kielman & May, 2009). And, while we do need to know where clusters and outliers are – as researchers, we also want to go further and explain why these patterns exist (Good, 1983). In this project we presume that there are ways for structuring the process of data exploration that make it more likely to discover unexpected and explanatory insights (Platt, 1964).This presentation summarizes results from a summer 2020 lab where we started experimenting with how to do this using our Center’s GeoDa software. The summer lab was directed by Julia Koschinsky of the University of Chicago’s Center for Spatial Data Science. Marcos Falcone helped mentor five young University of Chicago and high school students for 7–10 weeks (majoring in statistics, computation, geography, political science, and economics).Our approach was to draw on philosophy of science and scientific reasoning to understand how the discovery of unexpected and explicable insights can work. We then tried to translate this to research designs for ESDA. Finally, we implemented the designs in replicable prototype examples for teaching and learning spatial research at the undergraduate or high school level.For instance, in terms of scientific reasoning, classic work on causal explanations (Mill, 1843) augments the typical current focus on correlations by also highlighting the need to assess the plausibility of your own explanation versus alternatives. This requires a mindset and practice of rigorously testing how our explanations might be wrong (Popper, 1959) rather than confirming that they're right (Nuzzo, 2015; Kahneman, 2011). To do this requires an iterative exchange between data and explanations – referred to as abductive reasoning, as it combines inductive and deductive approaches (Peirce, 1878; Heckman and Singer, 2017). We used Sherlock Holmes stories and the famous John Snow cholera case to illustrate the structure of these scientific reasoning concepts for a high school context (Coleman, 2019; cf Konnikova, 2013; Vinten-Johansen, 2020).Scientific reasoning goes back hundreds of years. Our challenge this summer and from here on has been to translate this reasoning to research designs that are applicable to modern interactive ESDA tools. Each of us developed four prototype resources for teaching and learning ESDA in GeoDa that we will develop further (Fig. 2): 1) protocols for how this could be done; 2) case examples to apply and revise the protocol; 3) GeoDa demo scripts to make the examples replicable; and 4) cleaned data and documentation. These resources will be released as part of a GeoDa Cookbook in the near future.Fig. 3 illustrates one of the protocols that differs from how ESDA is typically navigated. The starting point is the exploration of patterns in the outcome variable of interest. Next is the formulation of alternative explanations whose patterns plausibly match those of the outcome variable. Then we draw on quasi-experimental research designs to structure the testing of this match (Shadish et al., 2002). Finally, data about the hypothesized explanations are analyzed with ESDA and regressions to test or reformulate the hypotheses as part of an abductive process.

On the interpretability of predictors in spatial data science: the information horizon

Two important theories in spatial modelling relate to structural and spatial dependence. Structural dependence refers to environmental state-factor models, where an environmental property is modelled as a function of the states and interactions of environmental predictors, such as climate, parent material or relief. Commonly, the functions are regression or supervised classification algorithms. Spatial dependence is present in most environmental properties and forms the basis for spatial interpolation and geostatistics. In machine learning, modelling with geographic coordinates or Euclidean distance fields, which resemble linear variograms with infinite ranges, can produce similar interpolations. Interpolations do not lend themselves to causal interpretations. Conversely, with structural modeling, one can, potentially, extract knowledge from the modelling. Two important characteristics of such interpretable environmental modelling are scale and information content. Scale is relevant because very coarse scale predictors can show nearly infinite ranges, falling out of what we call the information horizon, i.e. interpretation using domain knowledge isn’t possible. Regarding information content, recent studies have shown that meaningless predictors, such as paintings or photographs of faces, can be used for spatial environmental modelling of ecological and soil properties, with accurate evaluation statistics. Here, we examine under which conditions modelling with such predictors can lead to accurate statistics and whether an information horizon can be derived for scale and information content.

Transition to the New Normal: Lens on the Urban Spatial Data Science Agenda

Transition to the New Normal: Lens on the Urban Spatial Data Science Agenda

Developing Lesson Plans for Teaching Spatial Data Management in Academic Libraries through a Lens of Popular Culture

ABSTRACT Understanding data management in spatial data science is fundamentally important when ensuring a project’s data is appropriately collected, analyzed, stored, and communicated. However, these concepts are often introduced to learners through compliance (i.e., we do this because we are required to or it is the right thing to do), and as a result can mean learning experiences around spatial data management are viewed less than enthusiastically. Academic libraries are places where engaging learning experiences around spatial data management can be explored and implemented through unique pedagogical techniques. For example, activities teaching spatial data management in academic libraries through a popular culture lens can make these topics more approachable and reduce learning barriers. In this paper, I present three lesson plans for academic librarians which approach teaching spatial data management through popular culture. The following concepts are represented: (1) understanding spatial metadata through Dungeons & Dragons; (2) learning to write good documentation through drawing Pokémon; and (3) developing good file naming habits with 1980s album covers. Alongside the lesson plans, I provide a narrative outlining why popular culture pedagogy is a useful technique to spatial data management education to increase the inclusivity and engagement around geospatial learning experiences in academic libraries.

Opening practice: supporting reproducibility and critical spatial data science

This paper reflects on a number of trends towards a more open and reproducible approach to geographic and spatial data science over recent years. In particular, it considers trends towards Big Data, and the impacts this is having on spatial data analysis and modelling. It identifies a turn in academia towards coding as a core analytic tool, and away from proprietary software tools offering ‘black boxes’ where the internal workings of the analysis are not revealed. It is argued that this closed form software is problematic and considers a number of ways in which issues identified in spatial data analysis (such as the MAUP) could be overlooked when working with closed tools, leading to problems of interpretation and possibly inappropriate actions and policies based on these. In addition, this paper considers the role that reproducible and open spatial science may play in such an approach, taking into account the issues raised. It highlights the dangers of failing to account for the geographical properties of data, now that all data are spatial (they are collected somewhere), the problems of a desire for n = all observations in data science and it identifies the need for a critical approach. This is one in which openness, transparency, sharing and reproducibility provide a mantra for defensible and robust spatial data science.

Spatial data science for sustainable mobility

The constant rise of urban mobility and transport has led to a dramatic increase in greenhouse gas emissions. In order to ensure livable environments for future generations and counteract climate change, it will be necessary to reduce our future CO2 footprint. Spatial data science contributes to this effort in major ways, also fuelled by recent progress regarding the availability of spatial big data, computational methods and geospatial technologies. This paper demonstrates important contributions from Spatial data science to mobility pattern analysis and prediction, context integration, and the employment of geospatial technologies for changing people's mobility behavior. Among the interdisciplinary research challenges that lie ahead of us are an enhanced public availability of mobility studies and their data sets, improved privacy protection strategies, spatially-aware machine learning methods, and evaluating the potential for people's long-term behavior change towards sustainable mobility.

An Ontology-Driven Cyberinfrastructure for Intelligent Spatiotemporal Question Answering and Open Knowledge Discovery

The proliferation of geospatial data from diverse sources, such as Earth observation satellites, social media, and unmanned aerial vehicles (UAVs), has created a pressing demand for cross-platform data integration, interoperation, and intelligent data analysis. To address this big data challenge, this paper reports our research in developing a rule-based, semantic-enabled service chain model to support intelligent question answering for leveraging the abundant data and processing resources available online. Four key techniques were developed to achieve this goal: (1) A spatial and temporal reasoner resolves the spatial and temporal information in a given scientific question and enables place-name disambiguation based on support from a gazetteer; (2) a spatial operation ontology categorizes important spatial analysis operations, data types, and data themes, which will be used in automated chain generation; (3) a language-independent chaining rule defines the template for input, spatial operation, and output as well as rules for embedding multiple spatial operations for solving a complex problem; and (4) a recursive algorithm facilitates the generation of executive workflow metadata according to the chaining rules. We implement this service chain model in a cyberinfrastructure for online and reproducible spatial analysis and question answering. Moving the problem-solving environment from a desktop-based environment onto a geospatial cyberinfrastructure (GeoCI) offers better support to collaborative spatial decision-making and ensures science replicability. We expect this work to contribute significantly to the advancement of a reproducible spatial data science and to building the next-generation open knowledge network.

Towards Spatial Data Science: Bridging the Gap between GIS, Cartography and Data Science

Abstract. Data is regarded as the oil of the 21st century, and the concept of data science has received increasing attention in the last years. These trends are mainly caused by the rise of big data – data that is big in terms of volume, variety and velocity. Consequently, data scientists are required to make sense of these large datasets. Companies have problems acquiring talented people to solve data science problems. This is not surprising, as employers often expect skillsets that can hardly be found in one person: Not only does a data scientist need to have a solid background in machine learning, statistics and various programming languages, but often also in IT systems architecture, databases, complex mathematics. Above all, she should have a strong non-technical domain expertise in her field (see Figure 1).As it is widely accepted that 80% of data has a spatial component, developments in data science could provide exciting new opportunities for GIS and cartography: Cartographers are experts in spatial data visualization, and often also very skilled in statistics, data pre-processing and analysis in general. The cartographers’ skill levels often depend on the degree to which cartography programs at universities focus on the “front end” (visualisation) of a spatial data and leave the “back end” (modelling, gathering, processing, analysis) to GIScientists. In many university curricula, these front-end and back-end distinctions between cartographers and GIScientists are not clearly defined, and the boundaries are somewhat blurred.In order to become good data scientists, cartographers and GIScientists need to acquire certain additional skills that are often beyond their university curricula. These skills include programming, machine learning and data mining. These are important technologies for extracting knowledge big spatial data sets, and thereby the logical advancement to “traditional” geoprocessing, which focuses on “traditional” (small, structured, static) datasets such shapefiles or feature classes.To bridge the gap between spatial sciences (such as GIS and cartography) and data science, we need an integrated framework of “spatial data science” (Figure 2).Spatial sciences focus on causality, theory-based approaches to explain why things are happening in space. In contrast, the scope of data science is to find similar patterns in big datasets with techniques of machine learning and data mining – often without considering spatial concepts (such as topology, spatial indexing, spatial autocorrelation, modifiable area unit problems, map projections and coordinate systems, uncertainty in measurement etc.).Spatial data science could become the core competency of GIScientists and cartographers who are willing to integrate methods from the data science knowledge stack. Moreover, data scientists could enhance their work by integrating important spatial concepts and tools from GIS and cartography into data science workflows. A non-exhaustive knowledge stack for spatial data scientists, including typical tasks and tools, is given in Table 1.There are many interesting ongoing projects at the interface of spatial and data science. Examples from the ArcGIS platform include:Integration of Python GIS APIs with Machine Learning libraries, such as scikit-learn or TensorFlow, in Jupyter NotebooksCombination of R (advanced statistics and visualization) and GIS (basic geoprocessing, mapping) in ModelBuilder and other automatization frameworksEnterprise GIS solutions for distributed geoprocessing operations on big, real-time vector and raster datasetsDashboards for visualizing real-time sensor data and integrating it with other data sourcesApplications for interactive data explorationGIS tools for Machine Learning tasks for prediction, clustering and classification of spatial dataGIS Integration for HadoopWhile the discussion about proprietary (ArcGIS) vs. open-source (QGIS) software is beyond the scope of this article, it has to be stated that a.) many ArcGIS projects are actually open-source and b.) using a complete GIS platform instead of several open-source pieces has several advantages, particularly in efficiency, maintenance and support (see Wilkening et al. (2019) for a more detailed consideration). At any rate, cartography and GIS tools are the essential technology blocks for solving the (80% spatial) data science problems of the future.

Urban analytics in the context of public safety

Given a collection of geolocated activities, the goal of urban analytics in the context of public safety is to discover the underlying motives of people that affect their movement/activity patterns in space and time. Understanding the spatial patterns from urban mobility/activity datasets is an important task in public safety, city planning and sociology since these may reveal the underlying causes of crimes and safety issues, as well as behavior changes of individuals. Avoidance patterns are a type of behavioral change characterized by a lack of movement contrary to expectation. Avoidance pattern detection is a challenging task due to the lack of observations (e.g. lack of movement), defining the expected "normal" movement and large datasets (i.e. high number of GPS trajectories which are spread across the study area and large road network graphs). In addition, these challenges are exacerbated by the complicated and often hidden drivers of human activities and the complex relationships and dependencies between the spatially associated features. In this paper, we will provide a brief overview of the state-of-the-art spatial data science approaches in the context of avoidance patterns. First, we introduce the background from the domain (i.e. public safety) perspective, followed by an overview of the current state-of-the-art work. Then we will discuss possible future directions that may help shape future research on the topic.

Automated terrain feature identification from remote sensing imagery: a deep learning approach

ABSTRACTTerrain feature detection is a fundamental task in terrain analysis and landscape scene interpretation. Discovering where a specific feature (i.e. sand dune, crater, etc.) is located and how it evolves over time is essential for understanding landform processes and their impacts on the environment, ecosystem, and human population. Traditional induction-based approaches are challenged by their inefficiency for generalizing diverse and complex terrain features as well as their performance for scalable processing of the massive geospatial data available. This paper presents a new deep learning (DL) approach to support automatic detection of terrain features from remotely sensed images. The novelty of this work lies in: (1) a terrain feature database containing 12,000 remotely sensed images (1,000 original images and 11,000 derived images from data augmentation) that supports data-driven model training and new discovery; (2) a DL-based object detection network empowered by ensemble learning and deep and deeper convolutional neural networks to achieve high-accuracy object detection; and (3) fine-tuning the model’s characteristics and behaviors to identify the best combination of hyperparameters and other network factors. The introduction of DL into geospatial applications is expected to contribute significantly to intelligent terrain analysis, landscape scene interpretation, and the maturation of spatial data science.