Data files for "Eccles KM, Karmaus AL, Kleinstreuer NC, Parham F, Rider CV, Wambaugh JF, Messier KP. A geospatial modeling approach to quantifying the risk of exposure to environmental chemical mixtures via a common molecular target. Sci Total Environ. 2023 Jan 10;855:158905. doi: 10.1016/j.scitotenv.2022.158905. Epub 2022 Sep 21. PMID: 36152849"

The Spatial Analysis Software market is experiencing robust growth, driven by the increasing adoption of cloud-based solutions, the expanding use of drones and other data acquisition technologies for precise geographic data collection, and the rising demand for advanced analytics across diverse sectors. The market's expansion is fueled by the need for efficient geospatial data processing and interpretation in applications such as urban planning, infrastructure development, environmental monitoring, and precision agriculture. Key trends include the integration of Artificial Intelligence (AI) and Machine Learning (ML) for automating analysis and improving accuracy, the proliferation of readily available satellite imagery and sensor data, and the growing adoption of 3D modeling and visualization techniques. While data security concerns and the high initial investment costs for advanced software solutions pose some restraints, the overall market outlook remains positive, with a projected compound annual growth rate (CAGR) exceeding 10% (a reasonable estimate based on the rapid technological advancements and market penetration observed in related sectors). This growth is expected to be particularly strong in the North American and Asia-Pacific regions, driven by substantial government investments in infrastructure projects and burgeoning private sector adoption. The segmentation by application (architecture, engineering, and other sectors) reflects the versatility of spatial analysis software, enabling its use across various industries. Similarly, the choice between cloud-based and locally deployed solutions caters to specific organizational needs and technical capabilities. The competitive landscape is characterized by both established players and emerging technology companies, showcasing the dynamic nature of the market. Major players like Autodesk, Bentley Systems, and Trimble are leveraging their existing portfolios to integrate advanced spatial analysis capabilities, while smaller companies are focusing on niche applications and innovative analytical techniques. The ongoing advancements in both hardware and software, coupled with increasing data availability and affordability, are set to further fuel the market's growth in the coming years. The historical period (2019-2024) likely witnessed moderate growth as the market matured, laying the foundation for the accelerated expansion expected during the forecast period (2025-2033). Continued innovation and industry convergence will be key drivers shaping the future trajectory of the Spatial Analysis Software market.

ArcGIS Map Packages and GIS Data for Gillreath-Brown, Nagaoka, and Wolverton (2019)

**When using the GIS data included in these map packages, please cite all of the following:

Gillreath-Brown, Andrew, Lisa Nagaoka, and Steve Wolverton. A Geospatial Method for Estimating Soil Moisture Variability in Prehistoric Agricultural Landscapes, 2019. PLoSONE 14(8):e0220457. http://doi.org/10.1371/journal.pone.0220457

Gillreath-Brown, Andrew, Lisa Nagaoka, and Steve Wolverton. ArcGIS Map Packages for: A Geospatial Method for Estimating Soil Moisture Variability in Prehistoric Agricultural Landscapes, Gillreath-Brown et al., 2019. Version 1. Zenodo. https://doi.org/10.5281/zenodo.2572018

OVERVIEW OF CONTENTS

This repository contains map packages for Gillreath-Brown, Nagaoka, and Wolverton (2019), as well as the raw digital elevation model (DEM) and soils data, of which the analyses was based on. The map packages contain all GIS data associated with the analyses described and presented in the publication. The map packages were created in ArcGIS 10.2.2; however, the packages will work in recent versions of ArcGIS. (Note: I was able to open the packages in ArcGIS 10.6.1, when tested on February 17, 2019). The primary files contained in this repository are:

• Raw DEM and Soils data
  • Digital Elevation Model Data (Map services and data available from U.S. Geological Survey, National Geospatial Program, and can be downloaded from the National Elevation Dataset)
    • DEM_Individual_Tiles: Individual DEM tiles prior to being merged (1/3 arc second) from USGS National Elevation Dataset.
    • DEMs_Merged: DEMs were combined into one layer. Individual watersheds (i.e., Goodman, Coffey, and Crow Canyon) were clipped from this combined DEM.
  • Soils Data (Map services and data available from Natural Resources Conservation Service Web Soil Survey, U.S. Department of Agriculture)
    • Animas-Dolores_Area_Soils: Small portion of the soil mapunits cover the northeastern corner of the Coffey Watershed (CW).
    • Cortez_Area_Soils: Soils for Montezuma County, encompasses all of Goodman (GW) and Crow Canyon (CCW) watersheds, and a large portion of the Coffey watershed (CW).
• ArcGIS Map Packages
  • Goodman_Watershed_Full_SMPM_Analysis: Map Package contains the necessary files to rerun the SMPM analysis on the full Goodman Watershed (GW).
  • Goodman_Watershed_Mesa-Only_SMPM_Analysis: Map Package contains the necessary files to rerun the SMPM analysis on the mesa-only Goodman Watershed.
  • Crow_Canyon_Watershed_SMPM_Analysis: Map Package contains the necessary files to rerun the SMPM analysis on the Crow Canyon Watershed (CCW).
  • Coffey_Watershed_SMPM_Analysis: Map Package contains the necessary files to rerun the SMPM analysis on the Coffey Watershed (CW).

For additional information on contents of the map packages, please see see "Map Packages Descriptions" or open a map package in ArcGIS and go to "properties" or "map document properties."

LICENSES

Code: MIT year: 2019
Copyright holders: Andrew Gillreath-Brown, Lisa Nagaoka, and Steve Wolverton

CONTACT

Andrew Gillreath-Brown, PhD Candidate, RPA
Department of Anthropology, Washington State University
andrew.brown1234@gmail.com – Email
andrewgillreathbrown.wordpress.com – Web

Well-established conservation planning principles and techniques framed by geodesign were used to assess the restorability of areas that historically supported coastal wetlands along the U.S. shore of the connecting rivers (Detroit River and St. Clair River). The resulting analysis supported planning efforts to identify, prioritize, and track wetland restoration opportunity and investment in the region. To accomplish this, publicly available data, criteria derived from the regional managers and local stakeholders, and geospatial analysis were used to form an ecological model for spatial prioritization.

This dataset is the output of a python script/ArcGIS model that identifes dikes as having a difference in elevation above a certain threshold. If the elevation difference was below a certain threshold the area was not considered a dike; however, if the difference in elevation between two points was significantly high then the area was marked as a dike. Areas continuous with eachother were considered part of the same dike. Post processing occured. Users examined the data output, comparing the proposed dike locations to aerial imagery, flowline data, and the DEM. Dikes that appeared to be false positives were deleted from the data set.

This seminar is an applied study of deep learning methods for extracting information from geospatial data, such as aerial imagery, multispectral imagery, digital terrain data, and other digital cartographic representations. We first provide an introduction and conceptualization of artificial neural networks (ANNs). Next, we explore appropriate loss and assessment metrics for different use cases followed by the tensor data model, which is central to applying deep learning methods. Convolutional neural networks (CNNs) are then conceptualized with scene classification use cases. Lastly, we explore semantic segmentation, object detection, and instance segmentation. The primary focus of this course is semantic segmenation for pixel-level classification.

The associated GitHub repo provides a series of applied examples. We hope to continue to add examples as methods and technologies further develop. These examples make use of a vareity of datasets (e.g., SAT-6, topoDL, Inria, LandCover.ai, vfillDL, and wvlcDL). Please see the repo for links to the data and associated papers. All examples have associated videos that walk through the process, which are also linked to the repo. A variety of deep learning architectures are explored including UNet, UNet++, DeepLabv3+, and Mask R-CNN. Currenlty, two examples use ArcGIS Pro and require no coding. The remaining five examples require coding and make use of PyTorch, Python, and R within the RStudio IDE. It is assumed that you have prior knowledge of coding in the Python and R enviroinments. If you do not have experience coding, please take a look at our Open-Source GIScience and Open-Source Spatial Analytics (R) courses, which explore coding in Python and R, respectively.

After completing this seminar you will be able to:

1. explain how ANNs work including weights, bias, activation, and optimization.
2. describe and explain different loss and assessment metrics and determine appropriate use cases.
3. use the tensor data model to represent data as input for deep learning.
4. explain how CNNs work including convolutional operations/layers, kernel size, stride, padding, max pooling, activation, and batch normalization.
5. use PyTorch, Python, and R to prepare data, produce and assess scene classification models, and infer to new data.
6. explain common semantic segmentation architectures and how these methods allow for pixel-level classification and how they are different from traditional CNNs.
7. use PyTorch, Python, and R (or ArcGIS Pro) to prepare data, produce and assess semantic segmentation models, and infer to new data.
8. explain how object and instance segmentation are different from traditional CNNs and semantic segmentation and how they can be used to generate bounding boxes and feature masks for each instance of a class.
9. use ArcGIS Pro to perform object detection (to obtain bounding boxes) and instance segmentation (to obtain pixel-level instance masks).

Supplementary files for article Groundwater vulnerability to pollution assessment: an application of geospatial techniques and integrated IRN-DEMATEL-ANP decision model

This study evaluated the susceptibility to groundwater pollution using a modified DRASTIC model. A novel hybrid multi-criteria decision-making (MCDM) model integrating Interval Rough Numbers (IRN), Decision Making Trial and Evaluation Laboratory (DEMATEL), and Analytical Network Process (ANP) was used to investigate the interrelationships between critical hydrogeologic factors (and determine their relative weights) via a novel vulnerability index based on the DRASTIC model. The flexibility of GIS in handling spatial data was employed to delineate thematic map layers of the hydrogeologic factors and to improve the DRASTIC model. The hybrid MCDM model results show that Net Recharge (a key hydrogeologic factor) had the highest priority with a weight of 0.1986. In contrast, the Topography factor had the least priority, with a weight of 0.0497. A case study validated the hybrid model using Anambra State, Nigeria. The resultant vulnerability map shows that 12.98 % of the study area falls into a very high vulnerability class, 31.90 % falls into a high vulnerability, 23.52 % falls into the average vulnerability, 21.75 % falls into a low vulnerability, and 9.85 % falls into very low vulnerability classes, respectively. In addition, nitrate concentration was used to evaluate the degree of groundwater pollution. Based on observed nitrate concentration, the modified DRASTIC model was validated and compared to the traditional DRASTIC model; interestingly, the spatial model of the modified DRASTIC model performed better. This study is thus critical for environmental monitoring and implementing appropriate management interventions to protect groundwater resources against indiscriminate sources of pollution.

The global 3D mapping and modeling market is expected to grow significantly in the next few years as demand increases for detailed and accurate representations of physical environments in three-dimensional space. Estimated to be valued at USD 38.62 billion in the year 2025, the market was expected to grow at a CAGR of 14.5% from 2025 to 2033 and was estimated to reach an amount of USD 90.26 billion by the end of 2033. The high growth rate is because of improvement in advanced technologies with the development of high-resolution sensors and methods of photogrammetry that make possible higher-resolution realistic and immersive 3D models.Key trends in the market are the adoption of virtual and augmented reality (VR/AR) applications, 3D mapping with smart city infrastructure, and increased architecture, engineering, and construction utilization of 3D models. Other factors are driving the growing adoption of cloud-based 3D mapping and modeling solutions. The solutions promise scalability, cost-effectiveness, and easy access to 3D data, thus appealing to business and organizations of all sizes. Recent developments include: Jun 2023: Nomoko (Switzerland), a leading provider of real-world 3D data technology, announced that it has joined the Overture Maps Foundation, a non-profit organization committed to fostering collaboration and innovation in the geospatial domain. Nomoko will collaborate with Meta, Amazon Web Services (AWS), TomTom, and Microsoft, to create interoperable, accessible 3D datasets, leveraging its real-world 3D modeling capabilities., May 2023: The Sanborn Map Company (Sanborn), an authority in 3D models, announced the development of a powerful new tool, the Digital Twin Base Map. This innovative technology sets a new standard for urban analysis, implementation of Digital Cities, navigation, and planning with a fundamental transformation from a 2D map to a 3D environment. The Digital Twin Base Map is a high-resolution 3D map providing unprecedented detail and accuracy., Feb 2023: Bluesky Geospatial launched the MetroVista, a 3D aerial mapping program in the USA. The service employs a hybrid imaging-Lidar airborne sensor to capture highly detailed 3D data, including 360-degree views of buildings and street-level features, in urban areas to create digital twins, visualizations, and simulations., Feb 2023: Esri, a leading global provider of geographic information system (GIS), location intelligence, and mapping solutions, released new ArcGIS Reality Software to capture the world in 3D. ArcGIS Reality enables site, city, and country-wide 3D mapping for digital twins. These 3D models and high-resolution maps allow organizations to analyze and interact with a digital world, accurately showing their locations and situations., Jan 2023: Strava, a subscription-based fitness platform, announced the acquisition of FATMAP, a 3D mapping platform, to integrate into its app. The acquisition adds FATMAP's mountain-focused maps to Strava's platform, combining with the data already within Strava's products, including city and suburban areas for runners and other fitness enthusiasts., Jan 2023: The 3D mapping platform FATMAP is acquired by Strava. FATMAP applies the concept of 3D visualization specifically for people who like mountain sports like skiing and hiking., Jan 2022: GeoScience Limited (the UK) announced receiving funding from Deep Digital Cornwall (DDC) to develop a new digital heat flow map. The DDC project has received grant funding from the European Regional Development Fund. This study aims to model the heat flow in the region's shallower geothermal resources to promote its utilization in low-carbon heating. GeoScience Ltd wants to create a more robust 3D model of the Cornwall subsurface temperature through additional boreholes and more sophisticated modeling techniques., Aug 2022: In order to create and explore the system's possibilities, CGTrader worked with the online retailer of dietary supplements Hello100. The system has the ability to scale up the generation of more models, and it has enhanced and improved Hello100's appearance on Amazon Marketplace.. Key drivers for this market are: The demand for 3D maps and models is growing rapidly across various industries, including architecture, engineering, and construction (AEC), manufacturing, transportation, and healthcare. Advances in hardware, software, and data acquisition techniques are making it possible to create more accurate, detailed, and realistic 3D maps and models. Digital twins, which are virtual representations of real-world assets or systems, are driving the demand for 3D mapping and modeling technologies for the creation of accurate and up-to-date digital representations.

. Potential restraints include: The acquisition and processing of 3D data can be expensive, especially for large-scale projects. There is a lack of standardization in the 3D mapping modeling industry, which can make it difficult to share and exchange data between different software and systems. There is a shortage of skilled professionals who are able to create and use 3D maps and models effectively.. Notable trends are: 3D mapping and modeling technologies are becoming essential for a wide range of applications, including urban planning, architecture, construction, environmental management, and gaming. Advancements in hardware, software, and data acquisition techniques are enabling the creation of more accurate, detailed, and realistic 3D maps and models. Digital twins, which are virtual representations of real-world assets or systems, are driving the demand for 3D mapping and modeling technologies for the creation of accurate and up-to-date digital representations..

Well-established conservation planning principles and techniques framed by geodesign were used to assess the restorability of areas that historically supported coastal wetlands along the U.S. shore of western Lake Erie. The resulting analysis supported planning efforts to identify, prioritize, and track wetland restoration opportunity and investment in the region. To accomplish this, publicly available data, criteria derived from the regional managers and local stakeholders, and geospatial analysis were used to form an ecological model for spatial prioritization (Western Lake Erie Restoration Assessmente (WLERA)). Within the 192,618 ha study area that was bounded by the mouths of the Detroit River, MI to the north and the Black River, OH to the south, the model identified and prioritized 6,600 hectares of land most suitable for coastal wetland habitat restoration.

The United States geospatial analytics market is experiencing robust growth, projected to reach a significant size within the forecast period (2025-2033). The market's Compound Annual Growth Rate (CAGR) of 10.04% from 2019-2033 indicates a consistently expanding demand for geospatial data analysis across diverse sectors. Key drivers include the increasing availability of high-resolution satellite imagery, advancements in data processing capabilities (cloud computing, AI), and the growing need for data-driven decision-making in various industries. Specific sectors like agriculture, utilizing geospatial analytics for precision farming, and the defense and intelligence sectors, leveraging it for surveillance and strategic planning, are major contributors to market growth. Further fueling expansion are trends like the rising adoption of Internet of Things (IoT) devices generating location-based data, and the increasing sophistication of geospatial analytics software, incorporating advanced visualization and predictive modeling techniques. While data security concerns and the high cost of implementation pose some restraints, the overall market outlook remains positive, driven by the substantial benefits offered by geospatial analytics in improving efficiency, optimizing resource allocation, and enhancing situational awareness across a wide spectrum of applications. The market segmentation reveals significant opportunities across different types of geospatial analytics (surface analysis, network analysis, and geovisualization) and end-user verticals. While the provided data indicates a significant presence of companies like Harris Corporation, Bentley Systems Inc., and ESRI Inc., the market's competitive landscape is dynamic, with both established players and emerging technology companies vying for market share. The United States' dominance in geospatial technology and data infrastructure further supports the market's projected growth trajectory. The substantial investments in R&D and the prevalence of skilled professionals in the country further contribute to the market's expansion. Looking ahead, the integration of geospatial analytics with other technologies like blockchain and big data is expected to unlock new possibilities, further driving market growth and innovation in the coming years. Recent developments include: May 2023 : CAPE Analytics, a player in AI-powered geospatial property intelligence, has extended its partnership with The Hanover Insurance Group, which provides independent agents with the best insurance coverage and prices. Integrating geospatial analytics and inspection and rating models into Hanover's underwriting procedure is the central component of the partnership expansion. The company's rating plans will benefit from this strategic move, which will improve workflows, new and renewal underwriting outcomes, and pricing segmentation., March 2023 : Carahsoft Technology Corp., The Trusted Government IT Solutions Provider, and Orbital Insight, a player in geospatial intelligence, announced a partnership. By the terms of the agreement, Carahsoft will act as Orbital Insight's Master Government Aggregator, making the leading AI-powered geospatial data analytics available to the public sector through Carahsoft's reseller partners and contracts for Information Technology Enterprise Solutions - Software 2 (ITES-SW2), NASA Solutions for Enterprise-Wide Procurement (SEWP) V, National Association of State Procurement Officials (NASPO) ValuePoint, National Cooperative Purchasing.. Key drivers for this market are: Increasing in Demand for Location Intelligence, Advancements of Big Data Analytics. Potential restraints include: Increasing in Demand for Location Intelligence, Advancements of Big Data Analytics. Notable trends are: Network Analysis is Expected to Hold Significant Share of the Market.

According to Cognitive Market Research, the global geospatial analytics artificial intelligence market size is USD 100.5 million in 2024 and will expand at a compound annual growth rate (CAGR) of 28.60% from 2024 to 2031.

North America held the major market of more than 40% of the global revenue with a market size of USD 40.20 million in 2024 and will grow at a compound annual growth rate (CAGR) of 26.8% from 2024 to 2031.
Europe accounted for a share of over 30% of the global market size of USD 30.15 million.
Asia Pacific held the market of around 23% of the global revenue with a market size of USD 23.12 million in 2024 and will grow at a compound annual growth rate (CAGR) of 30.6% from 2024 to 2031.
Latin America market of more than 5% of the global revenue with a market size of USD 5.03 million in 2024 and will grow at a compound annual growth rate (CAGR) of 28.0% from 2024 to 2031.
Middle East and Africa held the major market of around 2% of the global revenue with a market size of USD 2.01 million in 2024 and will grow at a compound annual growth rate (CAGR) of 28.3% from 2024 to 2031.
The remote sensing held the highest geospatial analytics artificial intelligence market revenue share in 2024.

Market Dynamics of Geospatial analytics artificial intelligence Market

Key Drivers for Geospatial analytics artificial intelligence Market

Advancements in AI and Machine Learning to Increase the Demand Globally

The global demand for geospatial analytics is significantly driven by advancements in AI and machine learning, technologies that are revolutionizing how spatial data is analyzed and interpreted. As AI models become more sophisticated, they enhance the capability to automate complex geospatial data processing tasks, leading to more accurate and insightful analyses. Machine learning, particularly, enables systems to improve their accuracy over time by learning from vast datasets of geospatial information, including satellite imagery and sensor data. This leads to more precise predictions and better decision-making across multiple sectors such as environmental management, urban planning, and disaster response. The integration of AI with geospatial technologies not only improves efficiency but also opens up new possibilities for innovation, making it a critical driver for increased global demand in the geospatial analytics market.

Government Initiatives and Support for Smart Cities to Propel Market Growth

Government initiatives supporting the development of smart cities are propelling the growth of the geospatial analytics market. As urban areas around the world transform into smart cities, there is a significant increase in demand for advanced technologies that can analyze and interpret geospatial data to enhance urban planning, infrastructure management, and public safety. Geospatial analytics, powered by AI, plays a crucial role in these projects by enabling real-time data processing and insights for traffic control, utility management, and emergency services coordination. These technologies ensure more efficient resource allocation and improved quality of urban life. Government funding and policy support not only validate the importance of geospatial analytics but also stimulate innovation, attract investments, and foster public-private partnerships, thus driving the market forward and enhancing the capabilities of smart city initiatives globally.

Restraint Factor for the Geospatial analytics artificial intelligence Market

Complexity of Data Integration to Limit the Sales

The complexity of data integration poses a significant barrier to the adoption and effectiveness of geospatial analytics AI systems, potentially limiting sales in this market. Geospatial data, inherently diverse and sourced from various collection methods like satellites, UAVs, and ground sensors, comes in multiple formats and resolutions. Integrating such disparate data into a cohesive, usable format for AI analysis is a challenging process that requires advanced data processing tools and expertise. This complexity not only increases the time and costs associated with project implementation but also raises the risk of errors and inefficiencies in data analysis. Furthermore, the difficulty in achieving seamless integration can deter organizations, particularly those with limited IT capabilities, from investing in geospatial analytics solutions. Overcoming these integration challenges is crucial for enabl...

Geoparsing with Large Language Models

The .zip file included in this repository contains all the code and data required to reproduce the results from our paper. Note, however, that in order to run the OpenAI models, users will required an OpenAI API key and sufficient API credits.

Data

The data used for the paper are in the datasetst and results folders.

**Datasets: **This contains the XML files (LGL and Geovirus) and Json files (News2024) used to benchmark the models. It also contains all the data used to fine-tune the gpt-3.5 model, the prompt templates sent to the LLMs, and other data used for mapping and data creation.

**Results: **This contains the results for the models on the three datastes. The folder is separated by dataset, with a single .csv file giving the results for each model on each dataset separately. The .csv file is structured so that each row contains either a predicted toponym and an associated true toponym (along with assigned spatial coordinates), if the model correctly identified a toponym; otherwise the true toponym columns are empty for false positives and the predicted columns are empty for false negatives.

Code

The code is split into two seperate folders gpt_geoparser and notebooks.

**GPT_Geoparser: **this contains the classes and methods used process the XML and JSON articles (data.py), interact with the Nominatim API for geocoding (gazetteer.py), interact with the OpenAI API (gpt_handler.py), process the outputs from the GPT models (geoparser.py) and analyse the results (analysis.py).

Notebooks: This series of notebooks can be used to reproduce the results given in the paper. The file names a reasonably descriptive of what they do within the context of the paper.

Code/software

Requirements

Numpy

Pandas

Geopy

Scitkit-learn

lxml

openai

matplotlib

Contextily

Shapely

Geopandas

tqdm

huggingface_hub

Gnews

Access information

Other publicly accessible locations of the data:

The LGL and GeoVirus datasets can also be obtained here (opens in new window).

Abstract

Geoparsing- the process of associating textual data with geographic locations - is a key challenge in natural language processing. The often ambiguous and complex nature of geospatial language make geoparsing a difficult task, requiring sophisticated language modelling techniques. Recent developments in Large Language Models (LLMs) have demonstrated their impressive capability in natural language modelling, suggesting suitability to a wide range of complex linguistic tasks. In this paper, we evaluate the performance of four LLMs - GPT-3.5, GPT-4o, Llama-3.1-8b and Gemma-2-9b - in geographic information extraction by testing them on three geoparsing benchmark datasets: GeoVirus, LGL, and a novel dataset, News2024, composed of geotagged news articles published outside the models' training window. We demonstrate that, through techniques such as fine-tuning and retrieval-augmented generation, LLMs significantly outperform existing geoparsing models. The best performing models achieve a toponym extraction F1 score of 0.985 and toponym resolution accuracy within 161 km of 0.921. Additionally, we show that the spatial information encoded within the embedding space of these models may explain their strong performance in geographic information extraction. Finally, we discuss the spatial biases inherent in the models' predictions and emphasize the need for caution when applying these techniques in certain contexts.

Methods

This contains the data and codes required to reproduce the results from our paper. The LGL and GeoVirus datasets are pre-existing datasets, with references given in the manuscript. The News2024 dataset was constructed specifically for the paper.

To construct the News2024 dataset, we first created a list of 50 cities from around the world which have population greater than 1000000. We then used the GNews python package https://pypi.org/project/gnews/ (opens in new window) to find a news article for each location, published between 2024-05-01 and 2024-06-30 (inclusive). Of these articles, 47 were found to contain toponyms, with the three rejected articles referring to businesses which share a name with a city, and which did not otherwise mention any place names.

We used a semi autonmous approach to geotagging the articles. The articles were first processed using a Distil-BERT model, fine tuned for named entity recognicion. This provided a first estimate of the toponyms within the text. A human reviewer then read the articles, and accepted or rejected the machine tags, and added any tags missing from the machine tagging process. We then used OpenStreetMap to obtain geographic coordinates for the location, and to identify the toponym type (e.g. city, town, village, river etc). We also flagged if the toponym was acting as a geo-political entity, as these were reomved from the analysis process. In total, 534 toponyms were identified in the 47 news articles.

D = Discrete; C = Continuous; GLiM = Global Lithological Map; CCSM.3 = Community Climate System Model 3; SRTM = Shuttle Radar Topography Mission.

In this study, an integrated risk-modelling approach for estimating tropical cyclone impacts under present and future climate change scenarios using remote sensing, field data and spatial analysis was developed. The geospatial linear storm-surge models, and the resultant storm-surge models were used in the risk-modelling procedures. The storm-surge models were validated by a two-dimensional hydrodynamic model. The study demonstrated a simple and effective way for modelling detailed tropical cyclone risk information using geospatial techniques under present and future climate change scenarios at the local scale. The geospatial techniques were found potentially effective for providing essential input spatial information under the current approach.

As a result of extreme weather conditions, such as heavy precipitation, natural hillslopes can fail dramatically; these slope failures can occur on a dry day due to time lags between rainfall and pore-water pressure change at depth, or even after days to years of slow-motion. While the pre-failure deformation is sometimes apparent in retrospect, it remains challenging to predict the sudden transition from gradual deformation (creep) to runaway failure. We use a network science method–multilayer modularity optimization–to investigate the spatiotemporal patterns of deformation in a region near the 2017 Mud Creek, California landslide. We transform satellite radar data from the study site into a spatially-embedded network in which the nodes are patches of ground and the edges connect the nearest neighbors, with a series of layers representing consecutive transits of the satellite. Each edge is weighted by the product of the local slope (susceptibility to failure) measured from a digital elevation model and ground surface deformation (current rheological state) from interferometric synthetic aperture radar (InSAR). We use multilayer modularity optimization to identify strongly-connected clusters of nodes (communities) and are able to identify both the location of Mud Creek and nearby creeping landslides which have not yet failed. We develop a metric, community persistence, to quantify patterns of ground deformation leading up to failure, and find that this metric increases from a baseline value in the weeks leading up to Mud Creek's failure. These methods promise as a technique for highlighting regions at risk of catastrophic failure.

AI foundation models have demonstrated some capabilities for the understanding of geospatial semantics. However, applying such pre-trained models directly to geospatial datasets remains challenging due to their limited ability to represent and reason with geographical entities, specifically vector-based geometries and natural language descriptions of complex spatial relations. To address these issues, we investigate the extent to which a well-known-text (WKT) representation of geometries and their spatial relations (e.g., topological predicates) are preserved during spatial reasoning when the geospatial vector data are passed to large language models (LLMs) including GPT-3.5-turbo, GPT-4, and DeepSeek-R1. Our workflow employs three distinct approaches to complete the spatial reasoning tasks for comparison, i.e., geometry embedding-based, prompt engineering-based, and everyday language-based evaluation. Our experiment results demonstrate that both the embedding-based and prompt engineering-based approaches to geospatial question-answering tasks with GPT models can achieve an accuracy of over 0.6 on average for the identification of topological spatial relations between two geometries. Among the evaluated models, GPT-4 with few-shot prompting achieved the highest performance with over 0.66 accuracy on topological spatial relation inference. Additionally, GPT-based reasoner is capable of properly comprehending inverse topological spatial relations and including an LLM-generated geometry can enhance the effectiveness for geographic entity retrieval. GPT-4 also exhibits the ability to translate certain vernacular descriptions about places into formal topological relationships, and adding the geometry-type or place-type context in prompts may improve inference accuracy, but it varies by instance. The performance of these spatial reasoning tasks unveils the strengths and limitations of the current LLMs in the processing and comprehension of geospatial vector data and offers valuable insights for the refinement of LLMs with geographical knowledge towards the development of geo-foundation models capable of geospatial reasoning.

This dataset is the output of a python script/ArcGIS model that identifes dikes as having a difference in elevation above a certain threshold. If the elevation difference was below a certain threshold the area was not considered a dike; however, if the difference in elevation between two points was significantly high then the area was marked as a dike. Areas continuous with eachother were considered part of the same dike. Post processing occured. Users examined the data output, comparing the proposed dike locations to aerial imagery, flowline data, and the DEM. Dikes that appeared to be false positives were deleted from the data set.

This dataset contains 1) results of a series of model runs that explore the sensitivity coastal groundwater discharge to hydrogeological parameters, 2) results of a large series of numerical models of coastal groundwater discharge that cover parameter space for topographic gradients, recharge and permeability of coastal groundwater systems and 3) the results of a global geospatial data analysis of relief, watershed geometry, recharge and permeability of coastal watersheds, and values for coastal groundwater discharge that are based on a combination of the model experiments and the geospatial data analysis. A description of the methods that were used to generate these datasets can be found in a preprint on eartharxiv, see linked publication below.

Land degradation from gully erosion poses a significant threat to the Erer watershed in Eastern Ethiopia, particularly due to agricultural activities and resource exploitation. Identifying erosion-prone areas and underlying factors using advanced machine learning algorithms (MLAs) and geospatial analysis is crucial for addressing this problem and prioritizing adaptive and mitigating strategies. However, previous studies have not leveraged machine learning (ML) and GIS-based approaches to generate susceptibility maps identifying these areas and conditioning factors, hindering sustainable watershed management solutions. This study aimed to predict gully erosion susceptibility (GES) and identify underlying areas and factors in the Erer watershed. Four ML models, namely, XGBoost, random forest (RF), support vector machine (SVM), and artificial neural network (ANN), were integrated with geospatial analysis using 22 geoenvironmental predictors and 1,200 inventory points (70% used for training and 30% for testing). Model performance and robustness were validated through the area under the curve (AUC), accuracy, precision, sensitivity, specificity, kappa coefficient, F1 score, and logarithmic loss. The relative slope position is most influential, with 100% importance in SVM and RF and 95% importance in XGBoost, while annual rainfall (AR) dominated ANN (100% importance). Notably, XGBoost demonstrated robustness and superior prediction/mapping, achieving an AUC of 0.97, 91% accuracy, 92% precision, and 81% kappa while maintaining a low logloss (0.0394). However, SVM excelled in classifying gully resistant/susceptible areas (97% sensitivity, 98% specificity, and 91% F1 score). The ANN model predicted the most areas with very high gully susceptibility (13.74%), followed by the SVM (11.69%), XGBoost (10.65%), and RF (7.85%) models, while XGBoost identified the most areas with very low susceptibility (70.19%). The ensemble technique was employed to further enhance GES modeling, and it outperformed the individual models, achieving an AUC of 0.99, 93.5% accuracy, 92.5% precision, 97.5% sensitivity, 95.4% specificity, 85.8% kappa, and 94.9% F1 score. This technique also classified the GES of the watershed as 36.48% very low, 26.51% low, 16.24% moderate, 11.55% high, and 9.22% very high. Furthermore, district-level analyses revealed the most susceptible areas, including the Babile, Fedis, Harar, and Meyumuluke districts, with high GES areas of 32.4%, 21.3%, 14.3%, and 13.6%, respectively. This study offers robust and flexible ML models with comprehensive validation metrics to enhance GES modeling and identify gully prone areas and factors, thereby supporting decision-making for sustainable watershed conservation and land degradation prevention.

This data release publishes datasets within and surrounding the Piney Branch watershed located in the Washington, D.C. metropolitan suburb of Vienna, Virginia. This dataset was utilized in studies that investigated the accuracy and application of geospatial modeling techniques, structure-from-motion (SfM) photogrammetric methods, and digital elevation model (DEM) differencing to assess and quantify geomorphic and anthropogenic landform change. The United States Geological Survey’s (USGS) three-dimensional digital elevation program (3DEP) light detection and ranging (LiDAR) digital terrain models (DTMs) were used together with and as a means for comparison to DTMs created from historical aerial imagery. The creation and usage of both historical and current elevation datasets allows for the mapping of landscape change over time. Such mapping and assessment of geomorphic and anthropogenic change provides critical information for land management, hazard identification, and the management of challenges related to urbanization