2023 BTAA GIS Conference¶

The 2023 BTAA GIS Conference has concluded.

The 2023 Conference was held on November 8th and featured a keynote, presentations, lightning talks, and a map gallery.

Conference Recordings¶

View the 2023 conference recordings here or on YouTube.

Agenda¶

Welcome Remarks¶

9:00 - 9:15 am CST

Keynote: Reading Maps: Making, Searching, and Interpreting Text on Maps¶

9:45 - 10:15 am CST

Kate McDonough

Abstract

Machine learning methods are transforming the ways that anyone can interact with map collections. Using the results of a computer vision pipeline to understand past places at scale comes with new challenges and opportunities. The Machines Reading Maps project has worked with partners at the Library of Congress, the National Library of Scotland, the British Library, and also the David Rumsey Map Collection to create datasets of the text on large collections of digitized maps. This talk explores the results of the Machines Reading Maps project to date including our work in progress to, on the one hand, visualize and analyze the results for historical research, and, on the other, use text on maps data to improve map collection discovery.

Bio

Katie McDonough is a Lecturer in Digital Humanities at Lancaster University, and recently a member of Living with Machines and UK PI of Machines Reading Maps. She works at the intersection of social and intellectual history and the history of science and technology. She uses computational methods to translate large collections of eighteenth- and nineteenth-century European texts and images, maps in particular, into data that historians can interpret. Her current projects explore Enlightenment discourses of geographic knowledge, the politics of highway construction in rural France, and the impact of railway infrastructure on nineteenth-century British communities. At The Alan Turing Institute, she chairs the Computer Vision for Digital Heritage Special Interest Group.

Presentations 1: Forests and Water¶

10:20 - 11:20 CST / 11:20 - 12:20 EST

Facilitator: Laura McElfresh, University of Minnesota

Slide operator: Sue Oldenburg, Rutgers, the State University of New Jersey

Identifying potential locations for water quality wetland installation using GIS modeling¶

Annina Rupe, Ducks Unlimited

Iowa’s water quality has suffered due in part to the intense agriculture performed across the state. Wetlands naturally filter water which helps remove excess nutrients, but a majority of Iowa’s historic wetlands have been drained. Ducks Unlimited (DU) and the Iowa Department of Agriculture and Land Stewardship (IDALS) have partnered to identify where wetlands, specifically those aimed at improving water quality, would be best placed and to work with landowners to install these wetlands. This presentation delves into how GIS is used to simplify and speed up this process for the benefit of Iowans.

Plaintain Agroforestry in Uganda: Integrating Cultural Legacy into Multi-Criteria Decision Analysis¶

Joshua Russell, Penn State University

The assurance of food security and a decent livelihood for smallholder farmers in Sub-Saharan Africa is an ever-increasing challenge. Agroforestry is seen as one solution that has been promoted by numerous agencies and organizations to improve food and livelihood security without undue negative impacts on the environment. While this technology exists in many forms and has been practiced in specific locations across Africa for generations, the widespread adoption and "up-scaling" of agroforestry requires consideration of whether the agroforestry planting scheme and key specie(s) included within that scheme are not only suitable for a given ecosystem but also if cultural practices/traditions in an area will favor up-take. GIS-based Multi-Criteria Decision Analysis (MCDA) is an effective tool for structuring and evaluating criteria for decision support, through there is a noted absence in current literature illustrating the integration of cultural legacy or indigenous knowledge into MCDA, particularly in settings where field survey of individuals or societies is not an option. In such a case, the implementation of a methodological framework for gleaning cultural legacy or indigenous knowledge from ethnographic works and other pertinent research is critical for developing an output of suitability that is culturally appropriate. This project is structured to illustrate how topological, ecological, and cultural criteria can be integrated in GIS-MCDA for the assessment of plantain-based agroforestry site suitability in Uganda. By accounting for cultural criteria in MCDA suitability analysis and identifying sites where cultural traditions favor a given intervention, the potential for up-take of promoted interventions is likely to be improved.

New Jersey State Regulatory Riparian Zone Map¶

Kate Douthat, Rutgers, the State University of New Jersey; Larry Torok, New Jersey DEP

This map shows the locations and widths of regulatory riparian zones of the streams of New Jersey according to the Flood Hazard Area Control Act (FHA) Rules (N.J.A.C 7:13-4.1). The FHA designates riparian zones around streams, inside of which development and land disturbance is limited in order to protect target species, habitat, and water quality. The width and upstream extent of the buffer is based on state Surface Water Quality Standards and the location of habitat for endangered and threatened species. The State has a map of surface water quality standards that classifies streams by their type, designated uses, and anti-degradation statuses, but until now, the location of corresponding protective riparian zones was not mapped. Riparian zones had to be interpreted directly from the text of the rules—a complicated task for most users. Some technical issues we encountered when making this map included reconciling different sets of stream centerlines and creating isodistances with ESRI’s new Trace Network tool. Other challenges included reconciling data definitions and regulatory definitions. The final map is not able to fully translate the rules into GIS due the limitations of the current data, e.g. the location of the top of bank of streams, which is a fine-scale topographic feature that must be identified in the field. However, this map provides the public with a new understanding of the location of riparian zones and stream protections at specific areas of interest as well as landscape-level patterns.

LUNCH / COFFEE BREAK¶

11:20 - 11:50 CST / 12:20 - 12:50 EST

Lightning Talks A¶

11:50 - 12:40 CST / 12:50 - 1:40 EST -

Facilitator: Karen Majewicz, University of Minnesota; Slide Operator: Nicole Scholtz, University of Michigan

There and Back Again: An Analysis of the Movement of Avian Specimens Through Time and Space

Summer Mengarelli and Brian Weeks, University of Michigan;Shane DuBay, University of Texas-Arlington

Avian specimens, like other biological specimens housed in natural history museums and herbaria throughout the world, are useful sources of data in wide-ranging topics including environmental and ecological issues, species distribution and abundance, morphology, biomimicry, and phenology. The origins of avian specimen collection coincide with the beginnings of colonialism, and while specimens collected in centuries past retain their utility in modern research, they also retain a complicated history. This project analyzes the movement of these specimens through space in order to understand how specimen collection is situated or complicated in the ongoing legacy of colonial science. Using georeferenced specimen records from natural history collections in North America and Europe, we investigate clusters of specimen movement through history and ask whether the distance between collection and storage has lengthened or shortened. We apply our results to a discussion of whether specimen collection’s role in colonial science has faded, and pose the question of what repatriation might look like in the context of natural history museums. This geospatial analysis of the history of avian specimen collection can act as a starting place for broader exploration into the impact of colonialism in the practices of natural history knowledge work.
Analyzing Seasonal Crime Trends in Chicago: Examining the Impact of Temperature and Geography on Public Safety

Abhimanyu Hans and Milan Budhathoki, University of Maryland

This study investigates the seasonal trend of crime rates in the City of Chicago over a 20-year period. Daily crime location data were categorized into several crime types based on the classification code developed by law enforcement agencies, then spatially aggregated at the census tract level to examine spatial and temporal correlations. The study tested the hypothesis that certain crimes increase with warmer temperatures while others rise with colder temperatures, and that higher temperatures predict higher crime overall. Urban climate data from the Weather Research and Forecasting Model (WRF) and public crime data will be compared to identify census tracts sensitive to seasonal temperature changes. The findings aim to help law enforcement, urban planners, and stakeholders locate areas susceptible to heat stress and develop interventions to reduce violence and structural racism in highly crime-prone neighborhoods. This novel investigation of the overlooked links between climate, geography, and crime can inform strategies to mitigate urban heat impacts on public safety.
Mapping Heat Vulnerability in Indiana

Yuri Kim; Adriana Abreu; Jacob Alarcon; Linnea Fraser; Khoi Hua; Sungha Jo; Matthew Levy; Helena Mello; Uzoamaka Nwachukwu; Alec Siurek; Audra Stuck; and Eli VanDyke, Indiana University

In recent decades, heat waves have been growing as global issues due to climate change. Heat vulnerability is critical, especially for minority people living in cities. In the Advance GIS Class (GEOG-G438/539) at Indiana University Bloomington in Fall 2023, we analyzed heat vulnerability in the top twelve high-populated counties, focusing on social and environmental equity. We tried to build a heat vulnerability index with socioeconomic data (e.g., age, race, ethnicity, income, education attainment, housing status) and physical environment variables (e.g., surface temperature, canopy cover, and imperviousness). The results highlight diverse disparities in heat vulnerability levels associated with demographic and socioeconomic status across Indiana. While further analysis is required to enhance the heat vulnerability index, this analysis can alleviate extreme heat by prioritizing the areas for strategic plans such as regulating development, increasing tree cover, setting emergency cooling shelters, etc.
Identifying the Optimal Approach for Developing a 1-km Downscaled Soil Moisture Product for CONUS

Eshita Eva, The Ohio State University; and Steven M. Quiring, The Ohio State University

Developing improved methods for visualizing soil moisture at a fine resolution could prove useful in climatology, hydrology, the agricultural sector, drought monitoring, remote sensing, and other impacts. The recent evolution of remote sensing technology has provided additional tools for monitoring the near-surface moisture of soil across the continental United States (CONUS). The aim of this study is to determine the importance of different environmental variables on downscaled soil moisture products by leveraging remote sensing soil moisture. This study analyzes soil moisture data derived from the NASA Soil Moisture Active Passive Level 3 (SMAP L3) satellite mission. For downscaling soil moisture data to a 1-kilometer resolution, ancillary variables including precipitation, antecedent precipitation index (API), mean temperature, maximum temperature, minimum temperature, daytime range temperature (DTR), dew point temperature, soil texture, normalized difference vegetation index (NDVI), leaf area index (LAI), land use land cover, elevation, slope, and aspect is taken into account. To determine the optimal soil moisture value at any location, a machine learning model is used called random forest. In addition, examining the significance of random forest variables provides an opportunity to examine the association between soil moisture and environmental variables.
Making Open-Source Georeferencing Technology Collections-Ready with the Allmaps Platform

Stephen Appel, UW-Milwaukee Libraries - American Geographical Society Library; and Ian Spangler, Leventhal Map and Education Center, Boston Public Library

This presentation will introduce the NEH/IMLS Digital Humanities Advancement Grant funded project Making Open-Source Georeferencing Technology Collections-Ready with the Allmaps Platform, a partnership between the Leventhal Map & Education Center at Boston Public Library, the American Geographical Society Library at the University of Wisconsin-Milwaukee, and the Allmaps Project to build an open-source platform for georeferencing digitized cartographic collections, based on the Allmaps software library already under development. By building a patron-facing digital interface for georeferencing, hundreds of cultural institutions around the world can make their digitized map collections newly vibrant for scholars, educators, and the public. The project involves four major goals: (1) creating an interface for institutions to manage georeferencing project workflows and building inter-institutional collections; (2) writing documentation & educational modules that help both institutions and individuals get started using Allmaps; (3) creating two framework integrations—one for the open-source software GeoBlacklight used by LMEC, and one for the commercial software CONTENTdm used by AGSL—that allow Allmaps to be easily incorporated into widely-used collections management systems; and (4) establishing the Consortium for Public Geography, a coalition of institutions and individuals tasked with the maintenance of Allmaps, including long-term data management, scholarly engagement, and general governance.

Break¶

12:40 - 12:45 CST / 1:40 - 1:45 EST

Presentations 2: Urban*¶

12:45 - 1:45 CST / 1:45 - 2:45 EST

Facilitator: Laura McElfresh, University of Minnesota; Slide Operator: Nicole Scholtz, University of Michigan

An interactive biodiversity model for seven taxa in urban New York City¶

Marci Meixler, Rutgers, the State University of New Jersey; Kim Fisher, Wildlife Conservation Society; Eric W. Sanderson, New York Biological Garden

Declines in biodiversity have prompted efforts aimed at promoting resilience and producing sound ecosystem management plans particularly in urban coastal ecosystems vulnerable to the impacts of climate change. Our focus in this study was on development, demonstration, and testing of a biodiversity model within Visionmaker,nyc, an interactive online tool for resilience decision-making in New York City (NYC). Our biodiversity model utilizes information on habitat type, area and latitude (from user input), species-habitat preferences and species minimum habitat areas for 4,147 possible species (from the literature), and established ecological principles (latitude-enhanced species-area curves) to calculate species richness and composition for amphibians, birds, freshwater fish, mammals, marine fish, plants, and reptiles. Model validation for combined species and sites by taxon showed that all taxa had significantly small (P≤0.05) sign test probabilities indicating good agreement between observed and predicted presence/absence. Looking by individual species and across sites by taxon, 70-86% of species had significantly small (P≤0.05) sign test probabilities with freshwater fish the fewest significant results and plants the most. Our biodiversity model has proven effective at predicting species richness and species composition across a variety of habitats in a complex urban environment. Ecologists, planners and interested citizens can use our biodiversity model within Visionmaker.nyc to create sustainable and resilient visions of NYC now and into the future.

Association between COVID-19 policies and spatial distribution of burglaries and robberies in New York City*¶

Nina Yin, Penn State University

This project examines the associations between COVID-19 policies and the spatial distribution of burglaries and robberies in New York City. In response to the COVID-19 pandemic, mitigation policies were implemented at both a state (New York) and city (New York City) level. These policies, including lockdowns, phased re-openings, and mask policies, impacted human mobility within the city, and changed urban dynamics. These changes may have inadvertently affected the spatial distributions of crime in the city, resulting in differences in distributions before, during, and after the lifting of these policies. By analyzing spatial data on robberies and burglaries using a temporal analysis framework, this project gauges the correlations between specific COVID-19 measures, and fluctuations in the frequency and spatial distributions of robberies and burglaries in New York City. Uncovering such correlations, and their changes over space and time, would help shed light on the impact of COVID-19 policies on these crimes, and could guide policy makers, law enforcement, and urban planners in how to best address issues of crime during crises such as pandemics. Additionally, results from this project can help researchers and policy makers better understand the impact of the pandemic on crime in New York City, and therefore help them make more informed policy decisions should similar crises occur in the future.

LUNCH / COFFEE BREAK¶

1:45 - 2:15 CST / 2:45 - 3:15 EST

Presentations 3: Technical¶

2:15 - 3:15 CST / 3:15 - 4:15 EST

Facilitator: Kathleen Weessies, Michigan State University; Slide Operator: Karen Majewicz, University of Minnesota

The mapKurator System: A Complete Pipeline for Extracting and Linking Text from Historical Maps¶

Zekun Li; Jina Kim; Yijun Lin; Min Namgung; Leeje Jang; and Yao-Yi Chiang, University of Minnesota

Scanned historical maps in libraries and archives are valuable repositories of geographic data that often do not exist elsewhere. Despite the potential of machine learning tools like the Google Vision APIs for automatically transcribing text from these maps into machine-readable formats, they do not work well with large-sized images (e.g., high-resolution scanned documents), cannot infer the relation between the recognized text and other datasets, and are challenging to integrate with post-processing tools. This paper introduces the mapKurator system, an end-to-end system integrating machine learning models with a comprehensive data processing pipeline. mapKurator empowers automated extraction, post-processing, and linkage of text labels from large numbers of large-dimension historical map scans. The output data, comprising bounding polygons and recognized text, is in the standard GeoJSON format, making it easily modifiable within Geographic Information Systems (GIS). The proposed system allows users to quickly generate valuable data from large numbers of historical maps for in-depth analysis of the map content and, in turn, encourages map findability, accessibility, interoperability, and reusability (FAIR principles). We deployed the mapKurator system and enabled the processing of over 60,000 maps and over 100 million text/place names in the David Rumsey Historical Map collection. We also demonstrated a seamless integration of mapKurator with a collaborative web platform to enable accessing automated approaches for extracting and linking text labels from historical map scans and collective work to improve the results.

Python and ArcGIS Notebooks Improve Reproducibility of County Health Indicator¶

Jess Hoffelder and Ganhua Lu, University of Wisconsin Madison Population Health Institute

Access to exercise opportunities is a form of environmental justice and can impact community health. The County Health Rankings uses the creation of buffers around parks and recreational facilities to determine the proportion of county residents within census blocks near locations for exercise. The manual process was updated by implementing Python code in an ArcGIS Notebook to automate the workflow for buffer creation, spatial intersect, data summation and export, saving 100 plus hours annually and limiting human error in interaction with standard geoprocessing tool interfaces. The modified approach resulted from collaboration between staff with expertise in GIS and Python. The ArcPy Python package was utilized, in addition to external Python libraries. Additionally, this approach allowed the use of a single software combining steps previously completed between ArcGIS Pro and SAS; standardization and technical documentation of the annual calculation process between years; and flexibility in use of team time through improved accessibility of the process for analysts without high levels of GIS expertise.

Artificial Photosynthesis for Hydrogen Fuel: Feasibility and Siting¶

Brice Richardson; Will Hope; and Eli Ganong, University of Wisconsin-Madison

We explored the feasibility of artificial photosynthesis (AP) as applied to hydrogen fuel production in two modules. In the first, we sought to understand the perceptions of AP in academia, industry, and the government as well as the challenges and opportunities facing the technology. In the second, we conducted a site suitability analysis for AP facilities across Wisconsin. To accomplish the first objective, we conducted semi-structured interviews with stakeholders in academia and government and examined industry press releases. While AP faces some challenges, these do not detract from the technology’s feasibility to be implemented on a short timescale. We then moved to the second portion of the study using ArcGIS to find suitable sites for installing production facilities based on criteria including co-location with existing infrastructure and proximity to natural gas pipelines and traffic centers. We found the Fox Valley and Madison areas to be best suited for AP facilities, prompting us to zoom into Madison to find suitable locations on a local scale.

Break¶

3:15 - 3:20 CST / 4:15 - 4:20 EST

Lightning Talks B¶

3:20 - 4:10 CST / 4:20 - 5:10 EST

Facilitator: Kathleen Weessies, Michigan State University; Slide Operator: Karen Majewicz, University of Minnesota

Close Beside the Winding Cedar: Demonstrating a Local History StoryMap

Eric Tans, Michigan State University Libraries

While there are numerous works detailing Michigan State University’s history from a wide range of perspectives, none focus solely on arguably the campus’ most notable and beloved feature: the Red Cedar River. This is a considerable oversight, taking into consideration the significant role the river has played in university life, campus development, and teaching and research. The StoryMap “Close Beside the Winding Cedar” builds on historical and archival research into the cultural and institutional history of the Red Cedar River and its relationship to Michigan State University and aims to fill this gap in telling a part of Michigan State University’s story. Produced as part of the MSU Library Digital Scholarship Lab’s Project Incubator and weaving together a cohesive narrative employing maps, images, documents, and timelines, “Close Beside the Winding Cedar” acknowledges the crucial role the Red Cedar has played in the history of a major institution in Michigan’s history.
Integration of Satellite and UAV Imagery for Assessing Corn Nitrogen Status at Early Vegetative Growth Stages

Ana Gabriela Morales-Ona; Robert Nielsen; James Camberato; and Daniel Quinn, Purdue University

Nitrogen fertilizer accounts for 20-25% of the variable cost of production for rotation maize in Indiana. Spatial variability within fields and variable, unpredictable rainfall patterns make N a challenging nutrient to manage, with up to 65% of the nitrogen applied being lost to the environment. Post-emergence sidedress applications of N fertilizer can reduce N loss and improve plant uptake, so efficient and practical ways to identify maize N status at early maize growth stages is key to assessing plant N needs. The objectives of this study were to 1) compare spatial and spectral metrics for predicting biomass, 2) assess aerial imagery derived metrics for estimating N uptake and concentration, and 3) identify if integration of spatial metrics derived from UAV imagery integrated with spectral metrics from satellite imagery can improve N uptake prediction at early growth stages. To accomplish this, two large scale field trials were used for the study. Multispectral UAV (0.05 m resolution) and satellite imagery (3 m resolution) was acquired at early maize growth stages prior to the sidedress application of fertilizer treatments. Imagery was post-processed to calculate multiple vegetative indices and extract canopy cover fraction. Biomass samples were collected from pre-determined sampling areas to obtain dry matter weight and calculate N uptake. Regression analysis determined the relationship between biomass, nitrogen uptake, and metrics derived from UAV and satellite imagery. The results suggest that the integration of satellite and UAV imagery derived metrics can be used to assess maize N status in a time efficient way.
Leveraging GIS Resources at Indiana University

Katie Chapman, Indiana University

At Indiana University, GIS is an essential component of the work done by many of our students, faculty, and staff. From the classroom to administration, research, and public outreach, the IU community draws on the GIS resources available from our Esri Education subscription to visualize and analyze GIS data both close to home and across the globe to support public service, community building, information sharing, and our students’ career development. Research Technologies, a division of University Information and Technical Services, provides access to Esri’s GIS software for all members of the IU community. In addition to supporting GIS work at all of IU’s campuses, we also support the Indiana Spatial Data Portal (ISDP), a publicly accessible and annually increasing repository of GIS data from Indiana. The Indiana GIS data, primarily collected by the Indiana Geographic Information Office (GIO), includes downloadable orthoimagery and LiDAR data for the State. Here, we showcase some of the recent products and workflows developed by the IU GIS community.
Participatory GIS and Place-Network Investigations of Neglected Places: Exploring Nature Crime in Vietnam*

Elle Jingjing Xu; Judith Rakowski, UMD; and Meredith L. Gore, University of Maryland; Laure Joanny and Luan Van Nguyen, Fauna & Flora; and James Slade, Re:wild

Integrating participatory knowledge into efforts to resolve nature crime is appealing to decision makers because it offers inclusion of neglected spaces and thus more optimized and sustainable solutions. Participatory mapping has been widely used in rural conservation contexts and place-network investigations are offering promising applications to urban street crime problems at the neighborhood level. We explore the application of place-network investigations of wildlife poaching and trafficking, a globally distributed type of nature crime, in and around Pu Mat National Park in Vietnam in 2023. Using repeated measures and participatory mapping with local stakeholders and crime data from conservation rangers and nongovernmental organizations, we used QGIS to map the distribution of crime sites, corruption spots, comfort spaces, and convergent settings. We then used descriptive statistics and point pattern analysis to identify spatial patterns of each type of location, LISA statistics for hotspot detection, and geographically weighted regression model to explain how attributes such as facility type, distance to road affect the clustering of crime associated locations. Preliminary results identify a range of neglected spaces and spatial network trends: corruption spots facilitate wildlife crime outside the protected area concentrated in towns and markets; comfort spaces are more widely dispersed within protected areas in comparison to convergence settings and corruption spots. Participatory spatial data collection methods can help identify neglected spaces, places, and their spatial relationships. With a more holistic spatial data landscape, nature crime prevention strategies may be optimized for social legitimacy, cost efficiency, safety and sustainability.
Habitat Suitability and Behavioral Responses to Anthropogenic Disturbances of a Local Endangered Bear Species in Taiwan*

Fang Chen and Neil Carter, University of Michigan; Mei-Hsiu Hwang, National Pingtung University of Science and Technology

Asiatic black bears (Ursus thibetanus) are locally endangered in Taiwan due to habitat loss and poaching. Moreover, the recent proliferation of illegal snare traps has escalated threats to this species, leading to higher mortality rates and physical injuries. Despite the severity of the situation and the threats to this endangered species, there remains a lack of information regarding the favored habitat traits of bears and the behavioral impacts of injuries caused by snare traps. To address this knowledge gap, we integrated high-resolution movement data from 15 GPS-collared bears with various environmental covariates to (i) model habitat suitability and (ii) determine the influence of snare-induced injuries on space-use patterns. We utilized Resource Selection Function to identify environmental variables impacting bears’ habitat selection. The results were ranked on a scale from 1 to 10 and visualized as a habitat suitability map. Furthermore, our results revealed that injured bears show distinct habitat selection patterns compared to their non-injured counterparts. Injured bears tended to select less rugged terrains, likely attributed to mobility limitations resulting from snare-induced injuries. In addition, injured bears show stronger preference for areas closer to trails, potentially leading to increased future human-bear encounters and conflicts. Through this study, we have identified critical areas warranting prioritized conservation efforts and underscored the behavioral impact of snaring on Asiatic black bears. This groundwork establishes a crucial foundation for future conservation plans benefiting both Asiatic black bears and the ecosystem.

*Presentations listed with an asterisk were not recorded at the request of the presenter