Skip to main content
U.S. flag

An official website of the United States government

All Publications

Access all publications and filter by type, location, and search for keywords to find specific science and data information conducted by our scientists. 

Filter Total Items: 171126

Physics-guided architecture (PGA) of LSTM models for uncertainty quantification in lake temperature modeling

This chapter focuses on meeting the need to produce neural network outputs that are physically consistent and also express uncertainties, a rare combination to date. It explains the effectiveness of physics-guided architecture - long-short-term-memory (PGA-LSTM) in achieving better generalizability and physical consistency over data collected from Lake Mendota in Wisconsin and Falling Creek Reserv
Authors
Arka Daw, R. Quinn Thomas, Cayelan C. Carey, Jordan Read, Alison P. Appling, Anuj Karpatne

Heat budget of lakes

This article gives an overview of the heat fluxes between lakes and their environment. The heat budget of most lakes is dominated by heat fluxes at the lake surface, especially shortwave radiation, incoming and outgoing longwave radiation, and the latent heat flux. The seasonality of these fluxes is the most important driver for seasonal mixing processes in lakes. Changes in heat fluxes and the re
Authors
Martin Schmid, Jordan Read

Physics-guided neural networks (PGNN): An application in lake temperature modeling

This chapter introduces a framework for combining scientific knowledge of physics-based models with neural networks to advance scientific discovery. It explains termed physics-guided neural networks (PGNN), leverages the output of physics-based model simulations along with observational features in a hybrid modeling setup to generate predictions using a neural network architecture. Data science ha
Authors
Arka Daw, Anuj Karpatne, William Watkins, Jordan Read, Vipin Kumar

Physics-guided recurrent neural networks for predicting lake water temperature

This chapter presents a physics-guided recurrent neural network model (PGRNN) for predicting water temperature in lake systems. Standard machine learning (ML) methods, especially deep learning models, often require a large amount of labeled training samples, which are often not available in scientific problems due to the substantial human labor and material costs associated with data collection. M
Authors
Xiaowei Jia, Jared Willard, Anuj Karpatne, Jordan Read, Jacob Aaron Zwart, Michael Steinbach, Vipin Kumar

Planetary-scale change to the biosphere signalled by global species translocations can be used to identify the Anthropocene

We examine three distinctive biostratigraphic signatures associated with: hunting and gathering, landscape domestication, and globalisation. All three signatures have significant fossil records of regional importance that can be correlated inter-regionally and help describe the developing pattern of human expansion and appropriation of resources. While none have individual first or last appearance
Authors
Mark A. Williams, Reinhold Leinfelder, Anthony D. Barnosky, Martin J Head, Francine M G McCarthy, Cearreta. Alejandro, Stephen J Himson, Rachael Holmes, Colin N. Waters, Jan Zalasiewicz, Simon Turner, Mary McGann, Elizabeth A. Hadly, M. Allison Stegner, Paul Michael Pilkington, Jérôme Kaiser, Juan Carlos Berrio, Ian P. Wilkinson, Jens Zinke, Kristine L. DeLong

Daily surface temperatures for 185,549 lakes in the conterminous United States estimated using deep learning (1980–2020)

The dataset described here includes estimates of historical (1980–2020) daily surface water temperature, lake metadata, and daily weather conditions for lakes bigger than 4 ha in the conterminous United States (n = 185,549), and also in situ temperature observations for a subset of lakes (n = 12,227). Estimates were generated using a long short-term memory deep learning model and compared to exist
Authors
Jared D. Willard, Jordan Read, Simon Nemer Topp, Gretchen J. A. Hansen, Vipin Kumar

Predictive accuracy of post-fire conifer death declines over time in models based on crown and bole injury

A key uncertainty of empirical models of post-fire tree mortality is understanding the drivers of elevated post-fire mortality several years following fire, known as delayed mortality. Delayed mortality can represent a substantial fraction of mortality, particularly for large trees that are a conservation focus in western US coniferous forests. Current post-fire tree mortality models have undergon
Authors
Timothy M. Shearman, J. Morgan Varner, Sharon M. Hood, Phillip J. van Mantgem, C. Alina Cansler, Micah C. Wright

Identifying nutrient sources and sinks to the South Platte River and Cherry Creek, Denver, CO, during low-flow conditions in 2019–2020

Elevated concentrations and loads of nutrients in the South Platte River and Cherry Creek in Denver, Colorado, may have adverse effects on those streams and downstream water bodies, including increased production of algae, eutrophication, and decreased recreational opportunities. This article describes streamflow and concentrations and loads of nutrients for the South Platte River and Cherry Creek
Authors
William A. Battaglin, Tanner William Chapin

Phylogenetic risk assessment is robust for forecasting the impact of European insects on North American conifers

Some introduced species cause severe damage, although the majority have little impact. Robust predictions of which species are most likely to cause substantial impacts could focus efforts to mitigate those impacts or prevent certain invasions entirely. Introduced herbivorous insects can reduce crop yield, fundamentally alter natural and managed forest ecosystems, and are unique among invasive spec
Authors
Daniel R. Uden, Angela M. Mech, Nathan P. Havill, Ashley N. Schulz, Matthew P Ayers, Daniel A. Herms, Angela Marie Hoover, Kamal JK Gandhi, Ruth A. Hufbauer, Andrew M. Liebhold, Travis D Marisco, Kenneth F. Raffa, Kathryn A. Thomas, Patrick C. Tobin, Craig R. Allen

Decision support for aquatic restoration based on species-specific responses to disturbance

Disturbances to aquatic habitats are not uniformly distributed within the Great Lakes and acute effects can be strongest in nearshore areas where both landscape and within lake effects can have strong influence. Furthermore, different fish species respond to disturbances in different ways. A means to identify and evaluate locations and extent of disturbances that affect fish is needed throughout t
Authors
James E. McKenna, Catherine Riseng, Kevin Wehrly

Comparing imidacloprid, clothianidin, and azoxystrobin runoff from lettuce fields using a soil drench or treated seeds in the Salinas Valley, California

Neonicotinoid insecticide use has increased over the last decade, including as agricultural seed treatments (application of chemical in a coating to the seed prior to planting). In California, multiple crops, including lettuce, can be grown using neonicotinoid treated seeds or receive a direct neonicotinoid soil application (drenching) at planting. Using research plots, this study compared pestici
Authors
Emily Woodward, Michelle Hladik, Anson Main, Michael Cahn, James Orlando, Jennifer Teerlink

Pleistocene–Holocene vicariance, not Anthropocene landscape change, explains the genetic structure of American black bear (Ursus americanus) populations in the American Southwest and northern Mexico

The phylogeography of the American black bear (Ursus americanus) is characterized by isolation into glacial refugia, followed by population expansion and genetic admixture. Anthropogenic activities, including overharvest, habitat loss, and transportation infrastructure, have also influenced their landscape genetic structure. We describe the genetic structure of the American black bear in the Ameri
Authors
Matthew J. Gould, James W. Cain, Todd C. Atwood, Larisa E. Harding, Heather E. Johnson, Dave P. Onorato, Frederic S. Winslow, Gary W. Roemer