NSF I-GUIDE

NEXT WEEK! Learn how to get started with the I-GUIDE Platform! Wednesday November 20 at 11:00 am CT join Anand P. of CyberGIS Center for Advanced Digital and Spatial Studies at Geography and GIS, University of Illinois for part of the NSF I-GUIDE VCO Webinar Series. ?? Registration and Information: https://lnkd.in/gi-pWVuy

?? New publication from the NSF I-GUIDE team! "Toward Reproducible and Interoperable Environmental Modeling: Integration of HydroShare with Server-side Methods for Exposing Large-Extent Spatial Datasets to Models" published in Environmental Modelling and Software By authors: Young-Don Choi of K-water (Korea Water Resources Corporation) and the University of Virginia Iman Maghami of the University of Virginia and Brigham Young University Jonathan Goodall of the University of Virginia Larry Band of the University of Virginia Ayman Nassar of the Utah State University YuJu (Laurence) LIN of the University of Virginia Linnea Saby, Ph.D. of the University of Virginia ZHIYU (Drew) LI of the University of Utah Shaowen Wang of the University of Illinois Urbana-Champaign Chris Calloway of the University of North Carolina at Chapel Hill Hong Yi of the University of North Carolina at Chapel Hill Martin Seul of the Consortium of Universities for the Advancement of Hydrologic Science, Inc. (CUAHSI) Dan Ames of Brigham Young University David Tarboton of the Utah Water Research Laboratory at the University of Utah Read the full paper here ?? https://lnkd.in/gvTJcRhG This study presents a method for automating the distribution of state-scale spatial datasets for GIS-based environmental modeling, using GeoServer and THREDDS Data Server connected to HydroShare. By applying this approach with the RHESSys model across three U.S. watersheds, the study demonstrates minimal errors compared to traditional methods, enhancing reproducibility. This method can be adapted for various scales and other online data repositories, streamlining the process for researchers. Read the full paper here ?? https://lnkd.in/gvTJcRhG Abstract Reproducible environmental modelling often relies on spatial datasets as inputs, typically manually subset for specific areas. Yet, models can benefit from a data distribution approach facilitated by online repositories, and automating processes to foster reproducibility. This study introduces a method leveraging diverse state-scale spatial datasets to create cohesive packages for GIS-based environmental modelling. These datasets were generated and shared via GeoServer and THREDDS Data Server Connected to HydroShare, contrasting with conventional distribution methods. Using the Regional Hydro-Ecologic Simulation System (RHESSys) across three U.S. catchment-scale watersheds, we demonstrate minimal errors in spatial inputs and model streamflow outputs compared to traditional approaches. This spatial data-sharing method facilitates consistent model creation, fostering reproducibility. Its broader impact allows scientists to tailor the method to various use cases, such as exploring different scales beyond state-scale or applying it to other online repositories using existing data distribution systems, eliminating the need to develop their own.

Participate in our current Spatial AI Challenge! Join Shaowen Wang from the CyberGIS Center for Advanced Digital and Spatial Studies and the Department of Geography and GIS at University of Illinois Urbana-Champaign for an upcoming discussion on our NSF I-GUIDE 2024 Spatial AI Challenge! Coming up on Tuesday December 10 at 12:00 PM Central. ?? Registration and Information: https://lnkd.in/g9QxE8sz

Join Maryam Sahraei from South Dakota State University and Nitheshnirmal Sadhasivam from Virginia Tech for an upcoming session in the NSF I-GUIDE VCO series, "Estimating the Blue Water Footprint of a Reservoir in the Transboundary US-Mexico Region" coming up December 4th at 11:00am Central. Other team members for this research include Kanak Kanti Kar, Ruixuan Ding, Mohamed Awaad, Pravin Gamage, and team leader Santosh S. Palmate, Ph.D.! This will be the final presentation from our 2024 I-GUIDE Summer School. ?? Registration and Information: https://lnkd.in/e2aY9QYA

?? New publication from the NSF I-GUIDE team! "InstructG2I: Synthesizing Images from Multimodal Attributed Graphs" by Bowen (Peter) Jin, Ziqi Pang, Bingjun Guo, Yu-Xiong Wang, Jiaxuan You, and Jiawei Han of University of Illinois Urbana-Champaign and published in arXiv. Read the full paper here ?? https://lnkd.in/g-Mq9yaC This paper introduces a novel approach, Graph2Image, which focuses on generating images from multimodal attributed graphs (MMAGs), a complex task due to graph size, entity dependencies, and the need for control. The authors propose InstructG2I, a diffusion model that uses graph structure and multimodal information for effective neighbor sampling, along with a Graph-QFormer encoder to guide the generation process. Experiments on three datasets show that this method effectively enhances image generation control. Read the full paper here ?? https://lnkd.in/g-Mq9yaC Abstract In this paper, we approach an overlooked yet critical task Graph2Image: generating images from multimodal attributed graphs (MMAGs). This task poses significant challenges due to the explosion in graph size, dependencies among graph entities, and the need for controllability in graph conditions. To address these challenges, we propose a graph context-conditioned diffusion model called InstructG2I. InstructG2I first exploits the graph structure and multimodal information to conduct informative neighbor sampling by combining personalized page rank and re-ranking based on vision-language features. Then, a Graph-QFormer encoder adaptively encodes the graph nodes into an auxiliary set of graph prompts to guide the denoising process of diffusion. Finally, we propose graph classifier-free guidance, enabling controllable generation by varying the strength of graph guidance and multiple connected edges to a node. Extensive experiments conducted on three datasets from different domains demonstrate the effectiveness and controllability of our approach..

Next week join Anand Padmanabhan of the University of Illinois for the next NSF I-GUIDE VCO Webinar "I-GUIDE Platform 101: Exploring the Essentials of this Knowledge Sharing Environment." Coming up Wednesday, November 20 at 11:00am CT. ?? Registration and Information at https://lnkd.in/gi-pWVuy

TODAY join Yuanquan Li of UNICEF, Ruowei Liu of The University of Georgia, Maria Gorret Nabuwembo of the University of Arkansas, and Wenxin Yang of the UC Santa Barbara for the next NSF I-GUIDE VCO Webinar, "A Machine Learning Approach To Study The Cooling Effects Of Urban Green Spaces (UGS) in the Southeast U.S." coming up at 11:00 am CT!! ?? Registration and Information: https://lnkd.in/g-p4TK8N

?? New publication from the NSF I-GUIDE team! "Increasing Dam Failure Risk in the USA Due to Compound Rainfall Clusters as Climate Changes" by Jeongwoo Hwang and Upmanu Lall of Columbia University and npj Natural Hazards. Read the full paper here ?? https://lnkd.in/eRYQhXHE This paper analyzes rainfall patterns associated with the failure of 552 dams in the United States, finding that total rainfall over several days combined with maximum 1-day rainfall contributed to many failures. The study highlights that these compound precipitation risks are becoming more frequent due to changing atmospheric patterns, raising concerns about the future reliability of aging dams. Read the full paper here ?? https://lnkd.in/eRYQhXHE Abstract A changing climate, with intensifying precipitation may contribute to increasing failures of dams by overtopping. We present the first analysis of rainfall sequences and events associated with recent hydrologic failures of 552 dams in the United States. We find that the maximum 1-day rainfall associated with failure was often not extreme compared to dam spillway design criteria, even when accounting for rainfall statistics changing with time at each site. However, the combination of the total rainfall 5 to 30 days prior and the maximum 1-day rainfall associated with dam failure is rare. Persistent atmospheric circulation patterns that lead to recurrent rainfall events, rather than just more moisture in the atmosphere is a possible reason. The probability of these compound precipitation risks has increased across much of the country. With over 90,000 aging dams still in service, the increasing likelihood of intense rainfall sequences raises concerns about future dam failures.

Next week join Yuanquan Li of UNICEF, Ruowei Liu of The University of Georgia, Maria Gorret Nabuwembo of the University of Arkansas and Wenxin Yang of the UC Santa Barbara for the next NSF I-GUIDE VCO Webinar "A Machine Learning Approach To Study The Cooling Effects Of Urban Green Spaces (UGS) in the Southeast U.S." Coming up Wednesday, November 13 at 11:00am CT. ?? Registration and Information: https://lnkd.in/g-p4TK8N

?Coming up TODAY November 6 at 1:00pm CT, join Zhenlei Song of Texas A&M University and Jinyi Cai of the University of Iowa for the next NSF I-GUIDE VCO "Assessing the Economic Impact of Wildfire on Building Losses in Texas"! ?? Registration and Information: https://lnkd.in/eCeBezfc