.
.
.
.
.
"Air pollution, especially from wildfire smoke, has become an increasing threat to human health as drier, hotter conditions drive more frequent and larger wildfires across the United States. To reduce personal health risks, guidelines suggest staying indoors and limiting physical exertion during days with poor air quality. However, limiting physical exertion can also cause negative health outcomes, especially as the number of poor air quality days continues to increase. Quantifying when action is and isn’t taken to limit exposure is an important prerequisite for interventions designed to mitigate the negative health consequences of air pollution from wildfires, and the economic costs of these actions can inform fire management decisions. Building on prior work focused on the Northwestern United States, we estimate the extent to which the use of parks and indoor and outdoor fitness areas change with air quality across the contiguous United States. We expand the project scope by identifying sites used only for physical activity such as sports pitches and by including polygons for indoor fitness and sports centers, all obtained from ParkServe and Open Street Maps. We estimate changes in daily unique visitors to these locations over a period of four years using cellphone location data from Azira, examining heterogeneity in changes to visits using location characteristics and visitor demographics." .
"Impacts of air pollution on park and fitness area visitation across the contiguous United States." .
.
.
"tb208541@umconnect.umt.edu" .
"April 17 2026" .
"2025" .
.
.
.
.
.
.
.
"Prescribed burns help reduce wildfire risk, yet assessing post-burn vegetation recovery remains difficult due to the high dimensionality and labeling cost of hyperspectral imagery (HSI). We propose BurnSSL-DRL, a label-efficient framework that couples self-supervised learning (SSL) with deep reinforcement learning (DRL) for spectral band selection and vegetation classification. The DRL agent prioritized low-wavelength VNIR regions linked to chlorophyll degradation and soil exposure, reducing dimensionality to 30 bands while retaining key information. When combined with a 3D spectral–spatial CNN and class-balancing strategies (SMOTE + weighted loss), the BurnSSL-DRL achieved a macro-F1 ≈ 0.52—about 4–6% higher than PCA and mRMR baselines—and improved minority-class F1 (Grass 0.02 → 0.30, Soil 0.40 → 0.65). These results demonstrate that BurnSSL-DRL enables compact, interpretable, and accurate post-burn vegetation mapping, supporting scalable and near-real-time ecological monitoring from UAV platforms." .
"Hyperspectral band selection via self-supervised and reinforcement learning for prescribed burn impact analysis" .
.
.
"bradley.whitaker1@montana.edu" .
"15 Dec 2025" .
"2024" .
"Barbara Magagna" .
.
.
.
"2026-06-15T12:26:50Z"^^ .
.
"iadopt-variable-20260615T122650-92" .
"LLM-assisted I-ADOPT variable generation" .
"air temperature measured at 1.7 meter height" .
"Air temperature at 1.7 meter height" .
"temperature of air height: 1.7 meter" .
"My system is measrung air temp at 1.7 meter height" .
"Air temperature at 1.7 meter height" .
.
.
.
.
.
"height: 1.7 meter" .
.
.
"temperature" .
.
"air" .
.
.
.
.
.
.
.
"Hyperspectral image (HSI) analysis plays a central role in remote sensing tasks requiring fine-grained material discrimination, vegetation health assessment, and post-disturbance monitoring. Yet, the high dimensionality and strong spectral redundancy in HSIs often reduce the efficiency and reliability of machine learning models. These challenges are especially important in wildfire science and prescribed-fire monitoring, where spectral responses vary due to burn severity, char deposition, canopy structure, and early vegetation recovery. Benchmark datasets such as Indian Pines and Pavia University and others provide controlled environments for algorithms’ evaluation, but real-world post-fire forest conditions pose additional complexity. This study presents a unified and comprehensive evaluation of five dimensionality reduction strategies: Principal Component Analysis (PCA), Spatial–Spectral Edge Preservation (SSEP), Spectral-Redundancy Penalized Attention (SRPA), and a Deep Reinforcement Learning (DRL)-based selector together with a clustering based baseline, K-Means Clustering-Based Band Selection (KMCBS). These strategies are combined with classical machine learning and deep learning classifiers: Random Forest (RF), Support Vector Machines (SVMs), K-Nearest Neighbors (KNNs), and 3D Convolutional Neural Networks (3D-CNN). The full pipeline includes exploratory data analysis, preprocessing, patch-based spatial–spectral modeling, consistent train–validation protocols, and multi-dataset evaluation across Indian Pines, Pavia University, and a new custom VNIR hyperspectral dataset collected after prescribed burns at the Lubrecht Experimental Forest in Montana, USA. By systematically comparing statistical, edge-aware, attention-guided, and reinforcement learning-based band-selection strategies, this work identifies compact yet informative spectral subsets that enhance classification performance while reducing computational cost. Importantly, the inclusion of the Montana prescribed-burn dataset provides a unique real-world testbed for understanding band selection behavior in fire-affected forest environments. Overall, this study contributes a generalizable and extensible framework for HSI dimensionality reduction and classification, laying the groundwork for future applications in wildfire assessment, vegetation recovery monitoring, and remote sensing.\r\nKeywords: hyperspectral imaging; band selection; machine learning; prescribed fire" .
"Hyperspectral Band Selection for Ground Fuel Classification for Prescribed Fires" .
.
.
"mahmadisaq.karankot@student.montana.edu" .
"6 May 2026" .
"2024" .
.
.
.
.
.
.
.
"Finding potential research collaborators is a challenging task, especially in today’s fast-growing, interdisciplinary research landscape. While traditional methods rely on observable ties like co-authorships and citations, we focus solely on publication content to build a topic-based research network using BERTopic with a fine-tuned SciBERT model that connects and recommends researchers across disciplines based on shared topical interests. A key challenge we address is publication imbalance, where some researchers publish much more than others, often across several topics. Without careful handling, their less frequent interests are hidden under dominant topics, limiting the network’s ability to capture their full research scope. To tackle this, we introduce a cloning strategy that clusters a researcher’s publications and treats each cluster as a separate node. This allows researchers to belong to multiple communities, improving the detection of interdisciplinary links. Evaluation shows that the cloned network leads to more meaningful communities and uncovers broader collaboration opportunities." .
"Handling Publication Imbalance for Effective Community Detection in Scholarly Networks" .
.
.
"mdasaduzzamannoor@montana.edu" .
"Feburary 3 2026" .
"2024" .
.
.
.
.
.
.
.
.
"Many resources and professionals reference the 3-2-1 backup rule as an effective strategy to prevent active research data loss. However, the changes in storage technology and the pace of research data growth have outgrown the 3-2-1 rule. Objectives: The authors want to contribute background information and invite community input to evolve the 3-2-1 rule to fit modern research data and storage better. This evolution would provide better information to research data management professionals and researchers for more resilient research data. Methods: The authors facilitated a workshop at the Research Data Access and Preservation (RDAP) Summit in 2025 to present the necessary information for understanding the current storage and backup landscape. Backups were reframed as failure modes for data loss and corresponding preventative data protection measures. Results: The workshop resulted in an overview and summary of data protection methods and the ways in which they mitigate different “failures,” which allows for a more nuanced discussion of data protection that is not enabled by use of the 3-2-1 rule nor the term “backup” alone. Workshop participants brainstormed ways that the information presented in the workshop could be synthesized and incorporated into various learning materials, including materials for data professionals and researchers." .
"Evolving the 3-2-1 backup rule for more resilient data" .
.
.
"dmccaffrey@starfishstorage.com" .
"January 2026" .
"2025" .
.
.
.
.
.
.
"This Project generated data in the proces of replication of Weber et al 2025, paper validated on UK Biobank subset of 15000people, by reuse of MIMIC-IV-ECG Physionet clinical dataset, on close to 300000 records, in creation of ECG-extracted enriched cluster biomarkers acting as autonomic profiles for estimation of depression severity and potential suicide risk." .
"FAIR Mind" .