GIBD members have organized a series of sessions on geospatial big data and spatial computing at 2023 AAG Annual Meeting

Zhenlong Li   University of South Carolina
​​​​Naser Lessani  University of South Carolina

Chair(s):
Zhenlong Li  University of South Carolina

Call for Participation:

Rapid onset disasters, often difficult to prepare for and respond to, make disaster management a challenging task worldwide. Disaster and emergency management effectiveness depends heavily on making good decisions in near-real time under extreme duress. These key, often life-saving, decisions are possible only with real-time data sources and the ability to timely collect, process, synthesize, and analyze these multi-sourced data. Traditional data collection practices such as remote sensing and field surveying often fail to offer timely information during or immediately following damaging events. For example, stream gauges are only useful for flood mapping while the stations are functioning properly and before they are overtopped by floodwaters and rendered inoperable.

Fortunately, sharing information such as texts, images, and videos through social media platforms enables all citizens to become part of a large sensor network and a homegrown disaster response team. Compared to traditional physical sensors, such a citizen-sensor network (social sensing) is low cost, more comprehensive, and always broadcasting situational awareness information. For example, with social sensing, massive amounts of micro-level disaster information (e.g., site specific damage) can be captured in real-time through social media platforms (e.g., Twitter, Facebook) and voluntarily reported via dedicated crowdsourcing applications (volunteered geographic information, VGI), enabling rapid assessment of evolving disaster situations. On the other hand, data collected with social sensing is often massive, heterogeneous, noisy, unreliable, and comes in continuous streams. This is inherent “Big Data”, for example, millions of microblog posts from different social media platforms can be generated in a short time right after an impactful disaster. Hence, Big Data computing methods and technologies such as cloud computing, distributed geo-information processing, spatial statistics/modeling, data mining, spatial database, and multi-source data fusion become critical components of using social sensing to understand the impact of and response to the disaster events in a timely fashion.

Along these lines, this session on “Social Sensing and Big Data Computing for Disaster Management” aims to capture the insights in and bring up the discussion of leveraging social sensing and big data computing for supporting disaster management in one or more disaster phases (mitigation, preparedness, response, and recovery).

To present a paper in this session, you will register and submit your abstract online. Then email your presenter identification number (PIN) to one of the following organizers by November 11th, 2022.
https://aag.secure-platform.com/aag2023/solicitations/39/sessiongallery/5742

Organizer(s): 
Huan Ning   University of South Carolina
​​​​​Zhenlong Li  University of South Carolina

Huan Ning  University of South Carolina

Billions of people live in cities. They organize themself to create civilization and enjoy their products that come from tremendous collaboration. Cities are complex organisms, and urban life is deeply rooted in the dynamic patterns of cities. Humans create a better life by building better cities. In this progress, there is an essential need to sense the impulse of the city matrix, i.e., urban sensing, which refers to the technologies to sense and acquire dynamic patterns of city and human behaviors in the urban space. The ultimate goal of urban sensing is to build prosperous, sustainable, and equitable cities. To serve this goal, the scientific and engineering communities have responsibility to innovate theories and practices to monitor, analyze, model, predict, and intervene the urban phenomena.

Emerging geospatial big data, such as remote sensing imagery, street view images, social media, and human mobility, are major observations in urban sensing. Analyzing these observations bring technical challenges, such as data acquisition, management, and mining. Recent progress of artificial intelligence for geospatial data (GeoAI) has proven to be powerful tool for information/knowledge extraction from big data. We believe that geospatial big data and GeoAI are among the most promising approaches to address the technical challenges in contemporary urban sensing. This session aims to capture the recent advancements in using geospatial big data/GeoAI to sense and understand urban environments, including conceptualization, knowledge framework, toolbox organization, and applications.

Potential topics include, but are not limited to, the following:
• Exploration of the definitions and sources of urban geospatial big data
• Spatiotemporal scales of geospatial big data in urban settings
• Technology on capturing, storing, processing, and analyzing urban geospatial big data
• Geo-senor network and Internet of things
• Human mobility trajectories
• Urban hazard vulnerability assessment and emergency response
• General data processing and analyzing frameworks for urban geospatial big data
• Social or physical phenomena mining, modeling, and visualization in urban areas using geospatial big data
• Data representation and fusion of multi-modality observations in urban environments, such as images, text, sound, demography, and activities in cyberspace
• Privacy policies of urban geospatial big data
• Interdisciplinary applications based on urban geospatial big data, such as public health

To present a paper in this session, you will register and submit your abstract online (https://www2.aag.org/aagannualmeeting/). Then email your presenter identification number (PIN) to one of the following organizers by November 11th, 2022.

https://aag.secure-platform.com/aag2023/solicitations/39/sessiongallery/5740

Organizer(s): 
Fengrui Jing   University of South Carolina
​​​​Zhenlong Li  University of South Carolina

Chair(s):
Fengrui Jing  University of South Carolina

Call for Participation:

Mental health issues significantly impact individuals’ life, such as physical health conditions, school or work performance, as well as relationships with family and friends. For society, mental health issues impact social security and economic development. Depression and anxiety are two of the most common mental health conditions, with a combined annual cost of $1 trillion to the global economy. Mental health issues are currently on the rise globally. According to the United Nations, mental health issues now affect approximately 20% of children and adolescents worldwide, suicide is the second leading cause of death among 15-29-year-olds, and approximately 1 in 5 people in post-conflict settings suffer from a mental health condition.
Questionnaires, census data, and medical records have traditionally been used extensively in various aspects of mental health research. However, using traditional data still presents challenges, including data update periods, dataset accessibility, and data volumes. Emerging geospatial big data addresses these drawbacks due to its large data volume, rich attributes, ease of access, and high spatial and temporal resolution. The relationship between the environment and mental health, as well as the development of place-based policies for mental health surveillance, prediction, and intervention, necessitate the use of geospatial data. Geospatial big data is thus being used in various mental health studies, such as the relationship between environmental exposure and mental health using street view imagery, regional mental health status monitoring using social media data, and mental health service utilization analysis using social media apps.

This session seeks to capture recent advances in the use of geospatial big data and spatial analytics in mental health-related research, such as depression prevalence mapping, neighborhood-level depression surveillance, intervention, and prediction, and mechanism interpretation and analysis of mental health issues. Possible topics (but are not limited to) are the examples listed below.
• Utilizing geospatial big data (e.g., satellite images, street views, social media, mobile phone data, and traffic data) to extract geospatial contexts (e.g., environmental exposure, socioeconomic features, and human mobility) and mental health-related outcomes (e.g., subjective well-being and mental health service utilization) at various spatiotemporal scales to address theoretical questions related to mental health, such as the relationship between environmental exposure, human mobility, and mental health.
• Integrating sophisticated geostatistical techniques and geospatial big data to address practical questions related to mental health. For example, combing geospatial big data with advanced computational techniques such as machine learning (ML) and deep learning (DL) to use human mobility data and street-view environmental exposure data for mental health prevalence prediction at fine spatiotemporal levels, or to use social media data for suicide intervention.
• Developing accessible, participatory, and shareable mental health-related geovisualization approaches, data products, and online tools to serve researchers from various disciplines, the public from various backgrounds, and to aid communities, businesses, governments, and stakeholders in decision-making, such as site selection tools for mental health clinics.

To present a paper in this session, you will register and submit your abstract online (https://www2.aag.org/aagannualmeeting/). Then email your presenter identification number (PIN) to one of the following organizers by November 11th, 2022.

Multimodal Learning with Geospatial Big Data

Organizer(s): 
Meiliu Wu   University of Wisconsin-Madison
​​​​Qunying Huang  University of Wisconsin-Madison

Call for Participation:

Geospatial Big Data technology has been one of the key engines driving the new academic and industrial revolution. However, the majority of current Geospatial Big Data research efforts have been devoted to single-modal data analysis, leading to a huge gap in performance when algorithms are carried out separately. Although significant progress has been made, single-modal geospatial data is often insufficient to derive accurate and robust models in many geospatial applications.
In fact, multimodal is the most general form of geographic information representation and delivery in the real world. Using geospatial multimodal data is natural for humans to make accurate perceptions and decisions, as our digital world is essentially multimodal, combining different modalities of data (e.g., text, audio, images, and videos). Multimodal data analytics algorithms often outperform single-modal data analytics in many geospatial problems and applications. In particular, in the context of geospatial artificial intelligence (GeoAI) and machine learning (ML), we see the demand for spatially explicit multimodal learning as better ways to design AI/ML models by incorporating spatial knowledge and spatial inductive bias (e.g., spatial dependence and spatial heterogeneity) from geospatial multimodal data.
Similarly, multi-sensor geospatial information fusion has also been a topic of great interest in both academic and industrial fields. Organizations and institutions working on remote sensing applications, smart cities, urban computing, human dynamics, disaster resilience, or land use and land cover mapping have grown exponentially. They are attempting to automate processes by using a wide variety of geospatial information from various sources. Meanwhile, many geospatial problems have witnessed huge advancements with multimodal learning, such as geospatial knowledge and semantics mining, geographic question answering, and urban scene understanding.
With the rapid development of Geospatial Big Data technology and its remarkable applications in many fields, multimodal learning with Geospatial Big Data is a timely topic. This session aims to serve as a forum for researchers to share their recent advances in this promising topic, and to seek more interdisciplinary interaction and collaboration in its development.
To present your work in this session, you will register and submit your abstract to the AAG annual meeting website, and email your presenter identification number (PIN) and the abstract to Meiliu Wu (mwu233@wisc.edu) by Nov 11, 2022 along with your preference for an in-person or virtual presentation. Should you have any questions, please don’t hesitate to reach out to the session organizers.

In this session, we welcome submissions broadly contributing to the research on multimodal learning with geospatial big data. Multimodal learning algorithm design and developments tailored on geospatial big data are particularly welcome. The goal of this session is to solicit original contributions of recent findings in theory, methodologies, and applications in the field of multimodal learning with geospatial big data. The list of topics includes, but not limited to:
• Multimodal modeling with geospatial big data
• Geospatial cross-modal learning
• Contrastive learning with geospatial multimodal data
• Spatial analytics and geovisualization with multimodal big data
• Geospatial multimodal data fusion and data representation
• Geospatial multimodal big data infrastructure and management (e.g., data quality, uncertainties, and validation)
• Multimodal scene understanding
• Geospatial multimodal perception and interaction
• Geospatial multimodal benchmark datasets and evaluations
• Geospatial multimodal information tracking, retrieval and identification
• Multimodal learning for geospatial object localization, detection, classification, recognition and segmentation (e.g., remote sensing imagery processing, street view imagery analysis)
• Language and vision in the geospatial domain (e.g., geospatial knowledge and semantics mining, geographic question answering, and urban scene understanding)
• Geospatial multimodal applications (e.g., smart cities, urban computing, human dynamics, disaster resilience, land use and land cover mapping)