100039 research outputs found
Sort by
Compilation of a Nationwide River Image Dataset for Identifying River Channels and River Rapids via Deep Learning
No full textRemote sensing enables large-scale, image-based assessments of river dynamics, offering new opportunities for hydrological monitoring. We present a publicly available dataset consisting of 281,024 satellite and aerial images of U.S. rivers, constructed using an Application Programming Interface (API) and the U.S. Geological Survey’s National Hydrography Dataset. The dataset includes images, primary keys, and ancillary geospatial information. We use a manually labeled subset of the images to train models for detecting rapids, defined as areas where high velocity and turbulence lead to a wavy, rough, or even broken water surface visible in the imagery. To demonstrate the utility of this dataset, we develop an image segmentation model to identify rivers within images. This model achieved a mean test intersection-over-union () of 0.57, with performance rising to an actual of 0.89 on the subset of predictions with high confidence (predicted \u3e 0.9). Following this initial segmentation of river channels within the images, we trained several convolutional neural network (CNN) architectures to classify the presence or absence of rapids. Our selected model reached an accuracy and F1 score of 0.93, indicating strong performance for the classification of rapids that could support consistent, efficient inventory and monitoring of rapids. These data provide new resources for recreation planning, habitat assessment, and discharge estimation. Overall, the dataset and tools offer a foundation for scalable, automated identification of geomorphic features to support riverine science and resource management
Design of a Low-Power Magneto-Inductive Magnetometer
No full textMeasuring magnetic fields in space helps scientists understand phenomena that can affect satellite communications and navigation systems on Earth. This research develops a new low-power magnetic field sensor for spacecraft that improves upon existing designs by moving the sensitive parts away from electrical interference and using energy-efficient digital electronics for precise measurements. The sensor’s power-efficient design is particularly important for small satellites, where power is limited and must be carefully managed. It will fly on future NASA missions to study disturbances in Earth’s upper atmosphere that can disrupt radio signals and GPS. This work contributes to our ability to better predict and understand space weather events that impact our everyday technology while advancing the development of power-efficient space instrumentation
Constraint-Aware Metaheuristic Optimization for Experimental Design
No full textDesigning experiments becomes much more challenging when many variables and strict constraints are involved, as is common in modern science and engineering. This thesis introduces a new computational and mathematical framework that efficiently searches for optimal experiments in complex, high-dimensional spaces where traditional methods fail. By combining geometric techniques with flexible optimization algorithms like particle swarm optimization, our methods handle difficult constraints while scaling to real-world problems. Built in the high-performance Julia programming language and released as open-source software, this work bridges advanced theory with practical tools, offering researchers a powerful and accessible way to design better experiments under realistic conditions
Aerodynamic Parameter Estimation for a Scaled F-16: A Simulation-Based Sensitivity Analysis
No full textThe physical motion of an aircraft is determined by the resulting forces and moments acting on the aircraft. These forces and moments can be defined as a mathematical equation where the control deflections, rotation rates, and orientation of the aircraft are multiplied by individual constants and summed together to define the entire force or moment. This method results in a highly accurate model that defines the motion of the aircraft at low angles of attack and sideslip. Initial calculations of the aerodynamic model are created based on analytical estimations based on the physical properties of the aircraft. Additional refinement of the model is achieved in flight testing. During flight tests, the physical states of the aircraft, such as orientation, translational motion, and rotational motion, are recorded as well as the input commands for the aircraft. An analysis of the input commands and the output states provides the necessary data to define the aerodynamic model of the aircraft. This method of defining the motion of an object based on the input commands and the output states is called system identification. The aerodynamic model resulting from the system identification method is highly dependent on the accuracy of the input and output parameters as this is the only available information in the process. This high dependence demands a robust method to analyze and manage the error that is inherent in measurements. Most research in the area of error analysis in system identification focuses on removal or management of error. This research analyzes how much effect individual sensor error and assumptions have on the resulting aerodynamic model. This research identifies a robust maneuver input that provides sufficient measurement data for the system identification method. A best-case scenario will be tested in a simulation environment to determine what accuracy is possible with the specific maneuver case. The simulation environment provides direct control of introduced errors and assumptions, providing a platform to test the influence of each error and assumption on the resulting aerodynamic model. The results show that model assumptions need to be carefully analyzed to understand the effect on the resulting aerodynamic model. The assumption that the velocity of the aircraft can be simplified to be the forward velocity, which is the velocity measured by the pitot tube, is an erroneous assumption and highly corrupts the result. The second result that introduced significant error is using the commanded control inputs instead of the actual physical position of the control surfaces. This analysis shows the need to accurately measure or estimate the full velocity vector and the true control surface deflections. The results can provide practical guidance for designing a successful flight test, helping engineers determine what sensors are necessary and which assumptions must be avoided to build a flight test platform that can successfully find the aerodynamic model through system identification
Vines in the Landscape: Virginia Creeper
No full textVirginia creeper (Parthenocissus quinquefolia) is a vigorous and aggressive deciduous vine in the grape family (Vitaceae), known for its rapid growth, dense foliage, and striking fall color. Some landscapers consider this plant a nuisance because of its aggressive growth, potential damage to weak structures, and difficulty in removing, but others love it for its potential for erosion control, rapid coverage, fall color, and as a bird and pollinator habitat
Effects of Beaver Dams on Components of Riverscape Health and Biodiversity in the Great Basin
No full textNetworks of rivers and their floodplains (i.e., low-lying areas adjacent to rivers and streams that could plausibly flood) have been degraded, reducing the availability of freshwater and habitat for people and wildlife. Therefore, land managers are pursuing cost-effective restoration strategies, including partnering with beavers (Castor canadensis), to promote the health of these ecosystems and associated benefits. However, our understanding of beaver-mediated restoration benefits remains incomplete. Here, we compared locations along streams with and without beaver dams to measure the benefits of beavers in the Great Basin of the Western United States. We measured the effects of beaver dams on i) stream depths and depth variability, interactions between the stream and surrounding floodplains, and water movement as ecosystem health indicators; and ii) number of unique species, diversity, and activity levels for various fish, bat, and bird taxa as biodiversity indicators. We further investigated how the benefits of beavers varied by dam status (i.e., actively maintained or abandoned), and across valley settings, from steep and narrow valleys to low-gradient and wide valleys. Our results showed deeper and more variable stream habitats, stronger connections between streams and floodplains, and slower-moving water in stream segments with beaver dams than in segments without. Importantly, we found that the benefits of beaver activity are influenced by valley setting and that many benefits persisted even after dam abandonment. Although we found little difference in biodiversity indicators, the benefits of beaver dams to other components of ecosystem health likely contribute to diverse habitats that benefit biodiversity beyond individual stream segments. As interest in nature-based restoration grows, understanding the variable effects that beavers have on ecosystem processes within and across regions can inform the feasibility, potential effectiveness, and benefits of beaver-mediated ecosystem health restoration
Irrigation, Crop, and Soil Management Impacts on Water Use in Pivots/Laterals
No full textIn this fact sheet, we describe evapotranspiration (ET), introduce a new tool, the soil moisture evapotranspiration (SMET) model, and share results from multi-year trials across Utah evaluating different center-pivot irrigation technologies, deficit irrigation strategies, crop types, drought-tolerant crop genetics, and conservation agriculture practices. These findings highlight how Utah growers can improve water productivity and maintain profitable production under limited water supplies
Improving Ground Cover Crop Fractional Vegetation Mapping via Causality-Based Deep Representation Learning
No full textSemantic segmentation and deep learning methods have rarely been applied to fractional vegetation cover (FVC) segmentation tasks due to the lack of publicly available datasets for training deep learning models. FVC is a key indicator for assessing vegetation distribution, crop density, and crop responses to water availability and fertilizer application, yet conventional field-based measurement methods are time consuming, costly, labor intensive, and may lack the accuracy required for critical applications such as drought stress evaluation and water productivity. In this paper, we introduced causality-based deep learning techniques for FVC segmentation on a publicly available RGB dataset that consists of four ground cover crops: Phyla nodiflora L., Cynodon dactylon, Frankenia thymifolia Desf., and Oxalis stricta L. By separating causal from spurious correlations in pretrained features, using the stepwise intervention and reweighting (SIR) method at different encoder stages reduced confounding bias and enabled the models to learn more generalizable and task-relevant features. Extensive experiments on the FVC dataset, conducted with and without causality learning, showed that the proposed FCN + ResNet-50 model with causality learning and data augmentation achieved an accuracy of 94.80%, a precision of 94.97%, a recall of 94.35%, and an F1-score of 94.62%, which outperformed non-causal baselines and state-of-the-art transformer-based models including SegFormer and Mask2Former
Children and Youth with Special Health Care Needs (CYSHCN) Regional Support Centers (RSC): A framework for effectively engaging stakeholders
No full textThe three University Centers for Excellence in Developmental Disabilities (UCEDDs) in New York served as Children and Youth with Special Health Care Needs (CYSHCN) Regional Support Centers (RSC). The RSCs aimed to increase the capacity of local CYSHCN programs to connect with and support CYSHCN and their families in their community. RSCs conducted activities in the areas of family engagement, training and educational materials development, and technical assistance. RSCs engaged in ongoing dialogue with CYSHCN and their families to learn about experiences to inform systemic improvements through focus groups, interviews, and surveys. This article describes a statewide framework to serve and support CYSHCN and their families, which promotes data-driven quality improvement that incorporates family engagement at all levels. Data from conversations with families of CYSHCN informed systemic improvements through the following themes: 1) a disability diagnosis impacts and puts demands on the entire family; 2) parents of CYSHCN must develop and communicate detailed understanding of their children’s needs regarding health care, education, child care, and other supports; 3) parents often experience difficulty navigating systems when seeking services and supports; and, 4) community inclusion is difficult to achieve due to limited understanding and support. Informed by feedback from families, RSCs provided local CYSHCN programs with ongoing, individualized technical assistance to develop outreach strategies and resources. RSCs are a viable option to engage parents and improve services to CYSHCN in local communities