Geospatial Data Science Research Articles

Towards practical artificial intelligence in Earth sciences

AbstractAlthough Artificial Intelligence (AI) projects are common and desired by many institutions and research teams, there are still relatively few success stories of AI in practical use for the Earth science community. Many AI practitioners in Earth science are trapped in the prototyping stage and their results have not yet been adopted by users. Many scientists are still hesitating to use AI in their research routine. This paper aims to capture the landscape of AI-powered geospatial data sciences by discussing the current and upcoming needs of the Earth and environmental community, such as what practical AI should look like, how to realize practical AI based on the current technical and data restrictions, and the expected outcome of AI projects and their long-term benefits and problems. This paper also discusses unavoidable changes in the near future concerning AI, such as the fast evolution of AI foundation models and AI laws, and how the Earth and environmental community should adapt to these changes. This paper provides an important reference to the geospatial data science community to adjust their research road maps, find best practices, boost the FAIRness (Findable, Accessible, Interoperable, and Reusable) aspects of AI research, and reasonably allocate human and computational resources to increase the practicality and efficiency of Earth AI research.

Scaling Geospatial Data from the Perspective of Complexity: Exploring the Scaling Behavior of the Entropogram

A fundamental challenge in geospatial data science is to determine how a property, or its characterization, changes with a change in the scale of measurement. Except for geostatistical regularization of the variogram, which is theoretically well established, the scaling behaviors of a wide range of alternative measures of spatial association remain unclear. This limits the ability to make inferences at scales beyond the scale of measurement. The scaling behavior of the recently introduced entropogram function also remains unclear. Because the entropogram is essentially the generalization of the variogram to categorical spatial variables, the possibility to derive a scaling model for the entropogram exists. Here, the scaling behavior of the entropogram based on the scale effect of Shannon entropy is derived, providing a theoretical basis for the regularization of the entropogram. To validate the developed regularization model for the entropogram, a series of multiscale data was generated. Both theoretical derivation and experimental results showed that the entropogram is scale-invariant, under certain conditions for the generation of the categorical data. This research, thus, generalizes the entropogram to changes in measurement scale, thereby increasing our ability to characterize spatial data and make inferences about the underlying dynamic process. It also provides a reference for the interactions between patterns and processes at different scales.

From hearing to seeing: Linking auditory and visual place perceptions with soundscape-to-image generative artificial intelligence

People experience the world through multiple senses simultaneously, contributing to our sense of place. Prior quantitative geography studies have mostly emphasized human visual perceptions, neglecting human auditory perceptions at place due to the challenges in characterizing the acoustic environment vividly. Also, few studies have synthesized the two-dimensional (auditory and visual) perceptions in understanding human sense of place. To bridge these gaps, we propose a Soundscape-to-Image Diffusion model, a generative Artificial Intelligence (AI) model supported by Large Language Models (LLMs), aiming to visualize soundscapes through the generation of street view images. By creating audio-image pairs, acoustic environments are first represented as high-dimensional semantic audio vectors. Our proposed Soundscape-to-Image Diffusion model, which contains a Low-Resolution Diffusion Model and a Super-Resolution Diffusion Model, can then translate those semantic audio vectors into visual representations of place effectively. We evaluated our proposed model by using both machine-based and human-centered approaches. We proved that the generated street view images align with our common perceptions, and accurately create several key street elements of the original soundscapes. It also demonstrates that soundscapes provide sufficient visual information places. This study stands at the forefront of the intersection between generative AI and human geography, demonstrating how human multi-sensory experiences can be linked. We aim to enrich geospatial data science and AI studies with human experiences. It has the potential to inform multiple domains such as human geography, environmental psychology, and urban design and planning, as well as advancing our knowledge of human-environment relationships.

Structured Land Domain Modeling for Sustainable Land Administration in Sri Lanka

Developing countries face challenges in identifying suitable land due to insufficient administration systems, leading to governance issues and economic decline. A well-functioning land administration system is crucial for development. This technical paper elaborates an easy pathway to develop a local land administration Domains toward an effective and efficient Land Administration. Introduction of Land Fabric Domain (LFD) and the Localized Land Domains (LLD), are new concept for structuring of land parcels at comparatively low accuracy but well-defined topologically adjusted feature classes to manipulate as electronic information base to sustain the national land administration, real estate management and spatial data infrastructure development. The system architecture designs and the descriptive feature class overview elaborate the land administration system developers to begin their own LLD systems and interactive collaboration with national domains in order to manage overall system compatibility and consistency. The research focuses on the land information system (LIS) currently maintained and operated by the Survey Department and national issues in sustainable land administration of the country. Research focuses to analyze the prevailing process and outcomes of the LIS and introduces possible successive system architectures for implementation. The introductory LFD will be discussed against the feature topology on the basis of geospatial data science and then detailed feature classification will further be discussed accommodating entity relation diagrams. Proposed LLD with process architecture models will be conceptualized for successive implementation. The localized land domains will be structured and interactively linked through national LADM ISO-19152 guidelines for sustainable land administration of the country.

Importance of Community Engagement in Data Decision Making.

Importance of Community Engagement in Data Decision Making.

Multivariate analysis in data science for the geospatial distribution of the breast cancer mortality rate in Colombia.

This research is framed in the area of biomathematics and contributes to the epidemiological surveillance entities in Colombia to clarify how breast cancer mortality rate (BCM) is spatially distributed in relation to the forest area index (FA) and circulating vehicle index (CV). In this regard, the World Health Organization has highlighted the scarce generation of knowledge that relates mortality from tumor diseases to environmental factors. Quantitative methods based on geospatial data science are used with cross-sectional information from the 2018 census; it's found that the BCM in Colombia is not spatially randomly distributed, but follows cluster aggregation patterns. Under multivariate modeling methods, the research provides sufficient statistical evidence in terms of not rejecting the hypothesis that if a spatial unit has high FA and low CV, then it has significant advantages in terms of lower BCM.

Can green roofs help with stormwater floods? A geospatial planning approach

Increasing urbanization, impervious space, and the impact of climate change are threatening the future of cities. Nature-based solutions, specifically urban green infrastructures, are seen as a sustainable strategy to increase resilience against extreme weather events, including the escalating occurrence of stormwater runoff flooding. Consequently, urban planners and decision-makers have pushed their efforts toward implementing green infrastructure solutions to reduce the impact of stormwater floods. Among others, green roofs help store water and decrease stormwater runoff impacts on a local scale. This research aims to investigate the effect of surface permeability and green roof implementation on reducing stormwater flooding and subsequently provide urban planners with evidence-based geospatial planning recommendations to improve urban resilience in Helsinki. First, we modeled the current impact of stormwater flooding using the Arc-Malstrom model in Helsinki. The model was used to identify districts under high stormwater flood risk. Then, we zoomed in to a focus area and tested a combination of scenarios representing four levels of green roof implementation, two levels of green roof infiltration rates under 40-, 60-, 80-, 100 mm precipitation events on the available rooftops. We utilized open geographic data and geospatial data science principles implemented in the GIS environment to conduct this study. Our results showed that low-level implementation of green roofs with low retention rates reduces the average flood depth by only 1 %. In contrast, the maximum green roof scenario decreased most of the average flood depth (13 %) and reduced the number of vulnerable sites. The proposed methodology can be used for other cities to develop evidence-based plans for green roof implementations.

Interactive analysis of big geospatial data with high‐performance computing: A case study of partisan segregation in the United States

AbstractResearchers are increasingly working with large geospatial datasets that contain hundreds of millions of records. At this scale, desktop GIS systems typically fall short and so new approaches and methods are needed. The objective of this work is to develop new approaches to interactively analyze large datasets and then to demonstrate the usefulness of those approaches using a case study looking at voter, or partisan segregation. Historically, the measurement of partisan segregation has been limited to comparing large geographic areas such as counties or states because researchers only had access to aggregated data. In this case study, however, we measure partisan segregation down to the individual for 180 million U.S. voters using advanced geospatial data science and high‐performance computing. This article discusses interactive method development for big geospatial data analysis including techniques used, solutions developed, and processing time statistics.

Predicting lead water service lateral locations: Geospatial data science in support of municipal programming

We present and discuss machine learning predictions of customers' service line materials in Pittsburgh, PA and demonstrate the degree to which these predictions and the supporting data can and cannot improve municipal lead programming. Like previous work, predictive features reflect a combination of property characteristics, administrative spatial data, and tap water quality samples. Our work also includes labels of service line materials diagnosed by photographs taken at the curb box, which prove to boost predictions but are imperfect exclusive diagnostic methods. We use sample weighting and spatial cross validation in an effort to overcome the oversampling of lead service line characteristics of data collected for regulatory compliance. Cross-validation demonstrates precise predictions (precision >90%) for only 13% of customers, suggesting that predictions could improve short-term replacement decisions in avoiding unnecessary excavations. However, model precision declines when expanding predictions to more customers, limiting the degree to which predictions can estimate system-wide inventories and inform the regulatory decisions requiring complete inventories. We discuss the necessary trade-offs between biased sampling for regulatory compliance, which favors finding and replacing lead, and predictive modeling, which improves with unbiased sampling. We present a flow diagram that can help municipalities balance biased and unbiased sampling when integrating predictive modeling into compliance with federal regulations.

GeoAI at ACM SIGSPATIAL

Geospatial Artificial Intelligence (GeoAI) is an interdisciplinary field enjoying tremendous adoption. However, the efficient design and implementation of GeoAI systems face many open challenges. This is mainly due to the lack of non-standardized approaches to artificial intelligence tool development, inadequate platforms, and a lack of multidisciplinary engagements, which all motivate domain experts to seek a shared stage with scientists and engineers to solve problems of significant impact on society. Since its inception in 2017, the GeoAI series of workshops has been co-located with the Association for Computing Machinery International Conference on Advances in Geographic Information Systems. The workshop series has fostered a nexus for geoscientists, computer scientists, engineers, entrepreneurs, and decision-makers, from academia, industry, and government to engage in artificial intelligence, spatio-temporal data computing, and geospatial data science research, motivated by various challenges. In this article, we revisit and discuss the state of GeoAI open research directions, the recent developments, and an emerging agenda calling for a continued cross-disciplinary community engagement.

Google Earth Engine for geo-big data applications: A meta-analysis and systematic review

Google Earth Engine (GEE) is a cloud-based geospatial processing platform for large-scale environmental monitoring and analysis. The free-to-use GEE platform provides access to (1) petabytes of publicly available remote sensing imagery and other ready-to-use products with an explorer web app; (2) high-speed parallel processing and machine learning algorithms using Google’s computational infrastructure; and (3) a library of Application Programming Interfaces (APIs) with development environments that support popular coding languages, such as JavaScript and Python. Together these core features enable users to discover, analyze and visualize geospatial big data in powerful ways without needing access to supercomputers or specialized coding expertise. The development of GEE has created much enthusiasm and engagement in the remote sensing and geospatial data science fields. Yet after a decade since GEE was launched, its impact on remote sensing and geospatial science has not been carefully explored. Thus, a systematic review of GEE that can provide readers with the “big picture” of the current status and general trends in GEE is needed. To this end, the decision was taken to perform a meta-analysis investigation of recent peer-reviewed GEE articles focusing on several features, including data, sensor type, study area, spatial resolution, application, strategy, and analytical methods. A total of 349 peer-reviewed articles published in 146 different journals between 2010 and October 2019 were reviewed. Publications and geographical distribution trends showed a broad spectrum of applications in environmental analyses at both regional and global scales. Remote sensing datasets were used in 90% of studies while 10% of the articles utilized ready-to-use products for analyses. Optical satellite imagery with medium spatial resolution, particularly Landsat data with an archive exceeding 40 years, has been used extensively. Linear regression and random forest were the most frequently used algorithms for satellite imagery processing. Among ready-to-use products, the normalized difference vegetation index (NDVI) was used in 27% of studies for vegetation, crop, land cover mapping and drought monitoring. The results of this study confirm that GEE has and continues to make substantive progress on global challenges involving process of geo-big data.

Renewable resources of the northern half of the United States: potential for 100% renewable electricity

We previously presented a model for deep penetration of renewables in the electricity sector of the southern half of the United States (Khoie and Yee in Clean Technol Environ Policy 17(4):957–971, 2015). In this paper, we present a strategy for the northern half of the United States to utilize its available renewable resources to gradually decrease its reliance on fossil fuels in electricity generation and develop energy portfolios with increasing share of renewables. Using the electricity generation data from the US EIA (Electricity, electric power monthly, 2019d. https://www.eia.gov/electricity/monthly/ ), and the renewable resource maps produced by NREL (Geospatial data science, 2018a. http://www.nrel.gov/gis/solar.html ; Geospatial data science, 2018b. http://www.nrel.gov/gis/wind.html ), we develop strategies for the states in the northern half of the USA. We group these states into seven regions: West Coast, Mountain States, Middle West States, Lake States, Mid-Atlantic, South Atlantic, and New England states. For each region, we determine when and if the electricity generation from renewables will meet the region’s electricity need while accounting for a 1% annual increase in electricity demand. The renewable resources included in our models are solar (rural, urban, and rooftop photovoltaic), wind (onshore), hydro, biomass, and geothermal which vary greatly from region to region. We also include nuclear, coal, natural gas, and petroleum. Our results show that the West Coast, Mountain, and Middle West regions have the potential to become 100% renewable in the years 2041, 2039, and 2038, respectively. Additionally, our results show that the four regions of Lake, Mid-Atlantic, South Atlantic, and New England states may not be able to produce 100% of their electricity demand from their renewable resources.

How Do Cities Flow in an Emergency? Tracing Human Mobility Patterns during a Natural Disaster with Big Data and Geospatial Data Science

Understanding human movements in the face of natural disasters is critical for disaster evacuation planning, management, and relief. Despite the clear need for such work, these studies are rare in the literature due to the lack of available data measuring spatiotemporal mobility patterns during actual disasters. This study explores the spatiotemporal patterns of evacuation travels by leveraging users’ location information from millions of tweets posted in the hours prior and concurrent to Hurricane Matthew. Our analysis yields several practical insights, including the following: (1) We identified trajectories of Twitter users moving out of evacuation zones once the evacuation was ordered and then returning home after the hurricane passed. (2) Evacuation zone residents produced an unusually large number of tweets outside evacuation zones during the evacuation order period. (3) It took several days for the evacuees in both South Carolina and Georgia to leave their residential areas after the mandatory evacuation was ordered, but Georgia residents typically took more time to return home. (4) Evacuees are more likely to choose larger cities farther away as their destinations for safety instead of nearby small cities. (5) Human movements during the evacuation follow a log-normal distribution.

A spatio-contextual probabilistic model for extracting linear features in hilly terrains from high-resolution DEM data

ABSTRACTThis article introduces our research in developing a probabilistic model to extract linear terrain features from high resolution Digital Elevation Models (DEMs). The proposed model takes full advantage of spatio-contextual information to characterize terrain changes. It first derives a quantifiable measure of spatio-contextual patterns of linear terrain features, such as ridgelines, valley lines and crater boundaries, and then adopts multiple neighborhood analysis and a probability model to address data uncertainty in terrain surface modeling. Different from traditional approaches, the proposed model has the ability to achieve near-automated processing. It also supports effective extraction of terrain features in both smooth and rough surfaces. Through a series of experiments, we demonstrate that the proposed approach outperforms existing techniques, including thresholding, stream/drainage network analysis, visual descriptor detection, object-based image analysis and edge detection. This work contributes to both the geospatial data science and geomorphology communities with a new way of utilizing high-resolution imagery in terrain analysis.

The Internet of Things and fast data streams: prospects for geospatial data science in emerging information ecosystems

ABSTRACTThis paper surveys the rapid development of the Internet of Things, the massive data streams that are only now beginning to be generated from it, and the resulting opportunities and challenges that these data streams bring to geographic information analysis. These challenges arise because streaming data volumes cannot bt subjected to analysis using the standard repertoire of methods that have been designed to analyze static geospatial datasets. New approaches are needed, not to supplant, but to supplement, these existing tools. A focus is placed on the concept of data velocity (fast data) and its effects on sampling and inference. Innovative data ingestion strategies based on principles related to reservoir sampling and sketching are described. Dynamic temporal data flows present significant challenges to load balancing in distributed (e.g. cloud) parallel environments, even at exascale levels of performance. Further advances in the exploitation of data locality based on geographical concepts, as well as advanced processing methods based on edge and approximate computing, require further elucidation. Concepts are illustrated using a database compiled from a distributed sensor network of mobile radioactivity detectors.