Daniel Sheldon
A5061155671- Regional Socio-Economic Development Trends
- COVID-19, Geopolitics, Technology, Migration
- Impact of AI and Big Data on Business and Society
- Species Distribution and Climate Change
- Avian ecology and behavior
- Privacy-Preserving Technologies in Data
- Wildlife Ecology and Conservation
- Bayesian Modeling and Causal Inference
- Gaussian Processes and Bayesian Inference
- COVID-19 epidemiological studies
- Statistical Methods and Bayesian Inference
- Remote Sensing and LiDAR Applications
- Bayesian Methods and Mixture Models
- Machine Learning and Algorithms
- Statistical Methods and Inference
- Wildlife-Road Interactions and Conservation
- Cryptography and Data Security
- Ecology and Vegetation Dynamics Studies
- Complex Network Analysis Techniques
- Markov Chains and Monte Carlo Methods
- Economic and Environmental Valuation
- Animal Vocal Communication and Behavior
- Data-Driven Disease Surveillance
- Climate variability and models
- Machine Learning and Data Classification
University of Massachusetts Amherst
2015-2024
Amherst College
2017-2024
Imperial College London
2020-2024
University of Central Florida
2023-2024
University of Colorado System
2024
University of Colorado Boulder
2024
Mount Holyoke College
2014-2022
Oregon State University
2010-2021
Xylem (United States)
2017
Beaumont Hospital, Royal Oak
2016
The distributions of animal populations change and evolve through time. Migratory species exploit different habitats at times the year. Biotic abiotic features that determine where a lives vary due to natural anthropogenic factors. This spatiotemporal variation needs be accounted for in any modeling species' distributions. In this paper we introduce semiparametric model provides flexible framework analyzing dynamic patterns occurrence abundance from broad‐scale survey data. exploratory...
Short-term probabilistic forecasts of the trajectory COVID-19 pandemic in United States have served as a visible and important communication channel between scientific modeling community both general public decision-makers. Forecasting models provide specific, quantitative, evaluable predictions that inform short-term decisions such healthcare staffing needs, school closures, allocation medical supplies. Starting April 2020, US Forecast Hub ( https://covid19forecasthub.org/ ) collected,...
Abstract Background The COVID-19 pandemic has driven demand for forecasts to guide policy and planning. Previous research suggested that combining from multiple models into a single “ensemble” forecast can increase the robustness of forecasts. Here we evaluate real-time application an open, collaborative ensemble deaths attributable in U.S. Methods Beginning on April 13, 2020, collected combined one- four-week ahead cumulative jurisdictions standardized, probabilistic formats generate...
Abstract Academic researchers, government agencies, industry groups, and individuals have produced forecasts at an unprecedented scale during the COVID-19 pandemic. To leverage these forecasts, United States Centers for Disease Control Prevention (CDC) partnered with academic research lab University of Massachusetts Amherst to create US Forecast Hub. Launched in April 2020, Hub is a dataset point probabilistic incident cases, hospitalizations, deaths, cumulative deaths due county, state,...
Short-term forecasts of infectious disease burden can contribute to situational awareness and aid capacity planning. Based on best practice in other fields recent insights epidemiology, one maximise the predictive performance such if multiple models are combined into an ensemble. Here, we report ensembles predicting COVID-19 cases deaths across Europe between 08 March 2021 07 2022.
The deep image prior was recently introduced as a for natural images. It represents images the output of convolutional network with random inputs. For "inference", gradient descent is performed to adjust parameters make match observations. This approach yields good performance on range reconstruction tasks. We show that asymptotically equivalent stationary Gaussian process in limit number channels each layer goes infinity, and derive corresponding kernel. informs Bayesian inference. by...
Abstract Despite the routine nature of estimating overlapping space use in ecological research, to date no formal inferential framework for home range overlap has been available ecologists. Part this issue is due inherent difficulty comparing estimated ranges that underpin across individuals, studies, sites, species, and times. As calculated conditionally on a pair estimates, biases these estimates will propagate into estimates. Further compounding comparability estimators historical lack...
Computer and information scientists join forces with other fields to help solve societal environmental challenges facing humanity, in pursuit of a sustainable future.
Significance Collisions with built structures are an important source of bird mortality, killing hundreds millions birds annually in North America alone. Nocturnally migrating attracted to and disoriented by artificial lighting, making light pollution factor collision there is growing interest mitigating the impacts protect birds. We use two decades data show that migration magnitude, output, wind conditions predictors collisions at a large building Chicago decreasing lighted window area...
Abstract Short-term probabilistic forecasts of the trajectory COVID-19 pandemic in United States have served as a visible and important communication channel between scientific modeling community both general public decision-makers. Forecasting models provide specific, quantitative, evaluable predictions that inform short-term decisions such healthcare staffing needs, school closures, allocation medical supplies. Starting April 2020, US Forecast Hub ( https://covid19forecasthub.org/ )...
During the COVID-19 pandemic, forecasting trends to support planning and response was a priority for scientists decision makers alike. In United States, coordinated by large group of universities, companies, government entities led Centers Disease Control Prevention US Forecast Hub ( https://covid19forecasthub.org ). We evaluated approximately 9.7 million forecasts weekly state-level cases predictions 1–4 weeks into future submitted 24 teams from August 2020 December 2021. assessed coverage...
Abstract Redd (nest) surveys for resident brook trout ( S alvelinus fontinalis ) were conducted annually in a mountain lake northern New York 11 years with multiple during the spawning season eight of those years. Repeated throughout allowed us to fit an individually based parametric model and estimate day year on which was initiated, reached its midpoint, ended each year. Spawning phenology then assessed relative (1) mean maximum daily air temperature (2) water at bottom summer using linear...
Solar archival tags (henceforth called geolocators) are tracking devices deployed on animals to reconstruct their long-distance movements the basis of locations inferred post hoc with reference geographical and seasonal variations in timing speeds sunrise sunset. The increased use geolocators has created a need for analytical tools produce accurate objective estimates migration routes that explicit uncertainty about position estimates. We developed hidden Markov chain model analysis...
Abstract Home‐range estimation is an important application of animal tracking data that frequently complicated by autocorrelation, sampling irregularity, and small effective sample sizes. We introduce a novel, optimal weighting method accounts for temporal bias in autocorrelated data. This corrects irregular missing data, such oversampled times are downweighted undersampled upweighted to minimize error the home‐range estimate. also computationally efficient algorithms make this feasible with...
Summary Migration is a common strategy used by birds that breed in seasonal environments, and multiple environmental biological factors determine the timing of migration. How these operate combination during autumn migration, which considered to be under weaker time constraints relative spring not clear. Here, we examine patterns determinants migration for nocturnal migrants north‐eastern USA using reflectivity data from 12 weather surveillance radar stations modelled diurnal probability...
Abstract Large networks of weather radars are comprehensive instruments for studying bird migration. For example, the US WSR‐88D network covers entire continental and has archived data since 1990s. The can quantify both broad fine‐scale movements to address a range migration ecology questions. However, problem automatically discriminating precipitation from biology significantly limited ability conduct biological analyses with historical radar data. We develop M ist N et , deep convolutional...
Abstract Migrating birds require en route habitats to rest and refuel. Yet, habitat use has never been integrated with passage understand the factors that determine where when stopover during spring autumn migration. Here, we introduce stopover‐to‐passage ratio (SPR), percentage of migrants stop in an area, 8 years data from 12 weather surveillance radars estimate over 50% SPR through Gulf Mexico Atlantic coasts south‐eastern US, most prominent corridor for North America’s migratory birds....
Summary 1. Species monitoring is an essential component of assessing conservation status, predicting effects habitat change and establishing management priorities. The pervasive access to the Internet has led development several extensive projects that engage massive networks volunteers who provide observations following relatively unstructured protocols. However, value these data largely unknown. 2. We develop a novel cross‐data validation method for measuring survey from one source (e.g....
Search engines can automatically reformulate user queries in a variety of ways, often leading to multiple that are candidates replace the original. However, selecting replacement be risky: reformulation may more effective than original or significantly worse, depending on nature query, source candidates, and corpus. In this paper, we explore methods mitigate risk by issuing several versions query (including original) merging their results. We focus reformulations generated random walks click...
Abstract Billions of birds migrate at night over North America each year. However, few studies have described the phenology these movements, such as magnitudes, directions, and speeds, for more than one migration season regional scales. In this study, we characterize density, direction, speed nocturnally migrating using data from 13 weather surveillance radars in autumns 2010 2011 northeastern USA . After screening radar to remove precipitation, applied a recently developed algorithm...
Determining the distribution of stopover and overwintering areas migratory animals is essential for understanding population dynamics building predictive models. Tree Swallows (Tachycineta bicolor) are small songbirds that breed across North America. Data from Doppler weather radar eBird indicate Swallow numbers increase throughout October November in southeastern Louisiana, but then decrease during December. We thus hypothesized Louisiana a area used by fall migration before they move to...