Papers

Akinboyewa, Temitope: Estimating floodwater depth from on-site flood photos (paper)

Information on the depth of floodwater is crucial for rapid mapping of areas affected by floods. However, previous approaches for estimating floodwater depth, including field surveys, remote sensing, and machine learning techniques, can be time-consuming and resource-intensive. This paper presents an automated and fast approach for estimating floodwater depth from on-site flood photos. A pre-trained large multimodal model, Generative pre-trained transformers (GPT-4) Vision, was used specifically for estimating floodwater. The input data were flood photos that contained referenced objects, such as street signs, cars, people, and buildings. Using the heights of the common objects as references, the model returned the floodwater depth as the output. Results show that the proposed approach can rapidly provide a consistent and reliable estimation of floodwater depth from flood photos. Such rapid estimation is transformative in flood inundation mapping and assessing the severity of the flood in near-real time, which is essential for effective flood response strategies.
Keywords: flood mapping, large multimodal model, ChatGPT, GeoAI, disaster management.

Dedeaux, Danielle: WHY WE DON’T ALWAYS TURN AROUND: A Look into Why People Drive in Floods (lightning paper)

Although it’s not currently regarded as the deadliest, there is no doubt that flooding is a major cause of concern, injury, and even death in the United States. And while most people know not to go outside during a flood, there’s still a significant number of people who decide to drive through high-standing water (even after numerous warnings have been issued)- especially in car-centric areas. Thus, this paper seeks to review literature related to this topic to better understand the different reasons why a person would choose to drive through a flood. As this paper tries to understand the thought process behind this dangerous behavior, weather forecasters and first responders can use the information to update and improve how they communicate the very real risks of driving in a flood.

Mugeni, Maria: Understanding the integration of climate and urban development goals: An early exploration of policies in East Africa (paper)

Many African countries are grappling with integrating imperative climate actions and sustainable development goals as a mainstreaming-oriented approach toward climate-resilient development (Werners et al., 2021). National actors face the challenge of swiftly transitioning from climate change commitments to implementing measures that significantly address the risks of a changing climate (Taylor et al., 2023). This study focuses on understanding how climate policies in Kenya, Uganda, and Tanzania integrate development goals, in particular urban development goals focused on informal settlements that are prone to flooding and other climate change risks. Using content analysis of policy documents, the research examines synergies and incoherences between Nationally Determined Contributions (NDCs), National Adaptation Plans (NAPs), National Urban Policies (NUPs), and the SDGs 11 and 13 in these three developing countries. Preliminary findings indicate potential synergies between NDCs, NAPs, NUPs, and SDGs 11 and 13 for Kenya and synergies between NDCs, NUPs, and reviewed SDGs 11 and 13 for Uganda and Tanzania. This review offers critical insights for governments striving to operationalize climate-resilient development pathways. It contributes to ongoing discourses on sustainable development in the context of climate change, underscoring the importance of harmonizing policies to create resilient, sustainable, and inclusive urban environments in Africa. While there were no direct incoherence in the policy documents, ensuring that urban development and climate resilience become mutually reinforcing rather than conflicting goals is important during implementation. Understanding potential trade-offs at the implementation stages needs empirical research to better inform both climate and urban development actors in Kenya, Uganda, and Tanzania.

Ning, Huan: Measuring street tree trunk diameters based on street view images (paper)

Street trees play an important role in green space and infrastructure. They can increase biodiversity, provide wildlife habitat, improve air quality, regulate climate, and benefit pedestirans’ physical and mental health. Municipal administrations need detailed street tree inventories to maintain and plan green space better. Many studies have been conducted to mitigate the workload of the traditional resource-intensive field survey for tree inventories, such as automatic tree presence and species detection using street view images or laser sensors. However, little research investigated the scalable methods to measure the tree diameter, which is a critical attribute for inventory, providing valuable information for tree height and canopy estimation, biodiversity and habitat assessment, and urban planning. This research presents a practical approach to measuring street tree trunk diameters using street view imagery, addressing the need for efficient street tree inventories by combining photogrammetric triangulation and advanced computer vision techniques such as object detection, semantic segmentation, and monocular depth estimation. This methodology enables the extraction of tree locations and diameters at the city level, showing an average error of 0.05 meters and a relative error of 7%  in diameter measurement. This study offers a scalable, cost-effective solution for municipal administrations to manage urban green infrastructure.

Posters

Dedeaux, Danielle: Veteran Population and Mental Health Facilities in Pennsylvania

Lessani, M. Naser: SGWR: Similarity and Geographically Weighted Regression

Geographically weighted regression (GWR) offers a local approach to modeling spatial data, considering geographical location and spatial relationships between observations. A salient feature of GWR is the emphasis on geographical proximity, in accordance with Tobler’s First Law of Geography, which assumes that closer entities have a greater influence on the target location. Traditional GWR models have been augmented to consider various forms of physical distances aimed at enhancing model performance, and they often disregarded the potential influence of other data attributes, a shortcoming that extends to most GWR extensions. In this study, we introduce a novel weight matrix construction, which integrates data attribute similarity alongside the conventional geographically weighted matrix. The two weights are integrated in a manner that results in improved model performance. The proposed model, called Similarity and Geographically Weighted Regression or SGWR, was applied to five distinct datasets: housing prices, crime rates, and three health outcomes including mental health, depression, and HIV. Results show that SGWR significantly improved model performance based on several statistical measures, outperforming the global regression model and the traditional GWR.

Turnage-Barney, Finan: Assessing Surface Runoff Pathways: A Comparison of LIDAR and UAS Structure from Motion Digital Elevation Models in the Context of Excess Agricultural Nutrient Application from Manure and Synthetic Fertilizer