Posts

Abstract: The 2022 Buffalo blizzard was a catastrophic winter storm that struck Buffalo, New York in the week of Christmas in 2022. It claimed 47 lives and left much of the region stranded for the holiday week. In this disaster, the 311 call service was used by many residents to request help for issues due to the blizzard. This study examines these 311 help requests and their potential disparities across communities. Specifically, we aim to: (1) understand the spatial and temporal distributions of different types of 311 help requests; (2) identify the physical and social vulnerability factors, as well as human behavior factors, that are associated with the use of 311 calls. Methodologically, we leverage both explainable geospatial artificial intelligence (GeoAI) methods and statistical analysis to analyze 311 help requests and their associated factors. Our analysis shows significant spatial disparities in 311 help requests across communities. Results from explainable GeoAI and statistical analysis also reveal complementary insights on key factors associated with 311 help requests, such as historical 311 request behavior and percentage of minority population. These results could inform future disaster management decisions and help mitigate the negative impacts of winter storm disasters.

Full paper is available at: Zhou, R.Z., Hu, Y., Sun, K., Muldoon, R., Clark, S., & Joseph, K., (2025): Explainable GeoAI and statistical analysis reveal complementary insights about disparities of 311 help requests during the 2022 Buffalo blizzard. International Journal of Disaster Risk Reduction, 105635. [PDF]

Dr. Hu introduced geospatial technologies and AI, and the use of these technologies for disaster response in Encyclopedia Britannica.

He did an interview with Encyclopaedia Britannica previously:

In the 2024-2025 Department Award Ceremony, GeoAI Lab member Ryan Zhenqi Zhou received the 2025 Hugh Calkins GIS Award. This award is given annually to a graduate student from the Department of Geography at the University at Buffalo (UB) who demonstrates the principles of Dr. Calkins in using GIS to address real-world challenges. Congratulations, Ryan!

Lab member Jiyeon Kim receives the Michael Trapasso Climate Impact Award. This award recognizes and supports a graduate student from UB Geography, who uses meteorological or climatological data in their research to make a positive impact on the society. Congratulations, Jiyeon!

Three of our lab members, Ryan Zhou, Jiyeon Kim, and Yingjie Hu, attended and presented in this year’s AAG Annual Conference at Detroit.

Jiyeon presented her work on assessing the ability of deep learning models for wildfire spread prediction.

Ryan presented his work on the 2022 winter storm at Buffalo using explainable GeoAI

Yingjie presented the PyGRF work led by Kai Sun:

We had a great conference at AAG and Detroit! Looking forward to the conference next time!

Dr. Hu was recently interviewed by Channel 2 WGRZ for NSF-funded research on winter storm disasters focusing on the 2022 Buffalo blizzard.

– Link to the interview: https://www.wgrz.com/video/news/local/as-seen-on-tv/facebooks-impact-during-22-blizzard-studied-by-ub/71-b9a52ee6-b58d-44dc-ba70-20a324734215
– Link to the story: https://www.wgrz.com/article/news/local/buffalo/study-social-medias-role-connecting-community-2022-blizzard/71-19a08cec-9ebb-4ef7-a442-4d98d404f845
– Another story covered by Buffalo News: https://buffalonews.com/news/local/education/facebook-groups-saved-lives-during-22-blizzard-ub-studying-how-they-kept-communities-connected/article_7c13b1a0-be46-11ef-a327-778c60013a65.html

As the year 2024 comes to an end, our members at the GeoAI Lab wish you a Merry Christmas and a wonderful new year ahead in 2025!

With the support of the community, Dr. Hu has been elected as the Academic Director of the AAG Geographic Information Science and Systems (GISS) Specialty Group. He looks forward to continuing contributing to the research and education of GIScience.

Abstract: Geographical random forest (GRF) is a recently developed and spatially explicit machine learning model. With the ability to provide more accurate predictions and local interpretations, GRF has already been used in many studies. The current GRF model, however, has limitations in its determination of the local model weight and bandwidth hyperparameters, potentially insufficient numbers of local training samples, and sometimes high local prediction errors. Also, implemented as an R package, GRF currently does not have a Python version which limits its adoption among machine learning practitioners who prefer Python. This work addresses these limitations by introducing theory-informed hyperparameter determination, local training sample expansion, and spatially weighted local prediction. We also develop a Python-based GRF model and package, PyGRF, to facilitate the use of the model. We evaluate the performance of PyGRF on an example dataset and further demonstrate its use in two case studies in public health and natural disasters.

Full paper link: https://doi.org/10.1111/tgis.13248
Preprint PDF: https://www.acsu.buffalo.edu/~yhu42/papers/2024_TGIS_PyGRF.pdf

Our lab receives a new NSF grant to study disrupted human mobility, help requests, and voluntary support during natural disasters. Focusing on a type of understudied disasters, i.e., winter storms, this project examines the impacts of natural disasters on community resilience from a geographical perspective. It examines three aspects of the human responses to natural disasters. First, anonymized mobile phone location data and GeoAI methods will be used to investigate spatial variation in the disrupted patterns of mobility. Second, we investigate spatial disparities in the extent to which residents request assistance from municipal authorities. Third, we study the geography of help-seeking behavior among social media users and the roles of digital platforms in facilitating voluntary support. We will also examine the socioeconomic and demographic factors associated with the variation of human behaviors and community resilience. This project is empirically grounded on a complement of datasets, engages theories in disaster resilience and disparities, and leverages a combination of methods in geographical and statistical analysis, GeoAI, and network modeling. Insights from this project and the developed methodological frameworks have the potential to inform future disaster studies on disaster-caused human mobility disruptions, community-initiated responses, and the roles that social media play in shaping these responses and spatial disparities.

NSF Award Link: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2416886

At the 2024 CaGIS+UCGIS Symposium, lab member Kai Sun received the ICA Scholarship from the International Cartographic Association (ICA). The ICA Scholarship is awarded to “support early career scholars and professionals in advancing their career in cartography and GIScience.” Kai also made a poster presentation, “GALLOC: A GeoAnnotator for Labeling LOCation descriptions from disaster-related text messages”, at UCGIS.

Lab member Jiyeon Kim made an oral presentation, “Assessing the ability of deep learning models integrated with environmental and weather variables for predicting fire spread: A case study of the 2023 Maui wildfires.” Jiyeon’s presentation received the 3rd Place Award in the Student Presentation Competition.