Geology Reports⌕ Search

SEARCH · Geology Reports

Results for “Studies in Computational Intelligence”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

8 recordsLinked to original sources
USGS2008

Information fusion in regularized inversion of tomographic pumping tests

In this chapter we investigate a simple approach to incorporating geophysical information into the analysis of tomographic pumping tests for characterization of the hydraulic conductivity (K) field in an aquifer. A number of authors have suggested a tomographic approach to the analysis of hydraulic tests in aquifers - essentially simultaneous analysis of multiple tests or stresses on the flow system - in order to improve the resolution of the estimated parameter fields. However, even with a large amount of hydraulic data in hand, the inverse problem is still plagued by non-uniqueness and ill-conditioning and the parameter space for the inversion needs to be constrained in some sensible fashion in order to obtain plausible estimates of aquifer properties. For seismic and radar tomography problems, the parameter space is often constrained through the application of regularization terms that impose penalties on deviations of the estimated parameters from a prior or background model, with the tradeoff between data fit and model norm explored through systematic analysis of results for different levels of weighting on the regularization terms. In this study we apply systematic regularized inversion to analysis of tomographic pumping tests in an alluvial aquifer, taking advantage of the steady-shape flow regime exhibited in these tests to expedite the inversion process. In addition, we explore the possibility of incorporating geophysical information into the inversion through a regularization term relating the estimated K distribution to ground penetrating radar velocity and attenuation distributions through a smoothing spline model. ?? 2008 Springer-Verlag Berlin Heidelberg.

Studies in Computational Intelligence↗
USGS2002

An Intelligent Systems Approach to Automated Object Recognition: A Preliminary Study

Attempts at fully automated object recognition systems have met with varying levels of success over the years. However, none of the systems have achieved high enough accuracy rates to be run unattended. One of the reasons for this may be that they are designed from the computer's point of view and rely mainly on image-processing methods. A better solution to this problem may be to make use of modern advances in computational intelligence and distributed processing to try to mimic how the human brain is thought to recognize objects. As humans combine cognitive processes with detection techniques, such a system would combine traditional image-processing techniques with computer-based intelligence to determine the identity of various objects in a scene.

Open-File Report↗
USGS2000

Endocranial volume of mid-late eocene archaeocetes (order: cetacea) revealed by computed tomography: Implications for cetacean brain evolution

The large brain of modern cetaceans has engendered much hypothesizing about both the intelligence of cetaceans (dolphins, whales, and porpoises) and the factors related to the evolution of such large brains. Despite much interest in cetacean brain evolution, until recently there have been few estimates of brain mass and/or brain–body weight ratios in fossil cetaceans. In the present study, computed tomography (CT) was used to visualize and estimate endocranial volume, as well as to calculate level of encephalization, for two fully aquatic mid-late Eocene archaeocete species, Dorudon atrox and Zygorhiza kochii. The specific objective was to address more accurately and more conclusively the question of whether relative brain size in fully aquatic archaeocetes was greater than that of their hypothesized sister taxon Mesonychia. The findings suggest that there was no increase in encephalization between Mesonychia and these archaeocete species.

Journal of Mammalian Evolution↗
USGS2023

Improving ecological data science with workflow management software

Pressing environmental research questions demand the integration of increasingly diverse and large-scale ecological datasets as well as complex analytical methods, which require specialized tools and resources. Computational training for ecological and evolutionary sciences has become more abundant and accessible over the past decade, but tool development has outpaced the availability of specialized training. Most training for scripted analyses focuses on individual analysis steps in one script rather than creating a scripted pipeline, where modular functions comprise an ecosystem of interdependent steps. Although current computational training creates an excellent starting place, linear styles of scripting can risk becoming labor- and time-intensive and less reproducible by often requiring manual execution. Pipelines, however, can be easily automated or tracked by software to increase efficiency and reduce potential errors. Ecology and evolution would benefit from techniques that reduce these risks by managing analytical pipelines in a modular, readily parallelizable format with clear documentation of dependencies. Workflow management software (WMS) can aid in the reproducibility, intelligibility and computational efficiency of complex pipelines. To date, WMS adoption in ecology and evolutionary research has been slow. We discuss the benefits and challenges of implementing WMS and illustrate its use through a case study with the targets r package to further highlight WMS benefits through workflow automation, dependency tracking and improved clarity for reviewers. Although WMS requires familiarity with function-oriented programming and careful planning for more advanced applications and pipeline sharing, investment in training will enable access to the benefits of WMS and impart transferable computing skills that can facilitate ecological and evolutionary data science at large scales.

Methods in Ecology and Evolution↗
USGS2022

Development of projected depth-duration frequency curves (2050–89) for south Florida

Planning stormwater projects requires estimates of current and future extreme precipitation depths for events with specified return periods and durations. In this study, precipitation data from four downscaled climate datasets are used to determine changes in precipitation depth-duration-frequency curves from the period 1966–2005 to the period 2050–89 primarily on the basis of Representative Concentration Pathways 4.5 and 8.5 emission scenarios from the Coupled Model Intercomparison Project Phase 5. The four downscaled climate datasets are (1) the Coordinated Regional Downscaling Experiment (CORDEX) dataset, (2) the Localized Constructed Analogs (LOCA) dataset, (3) the Multivariate Adaptive Constructed Analogs (MACA) dataset, and (4) the Jupiter Intelligence Weather Research and Forecasting Model (JupiterWRF) dataset. Change factors—multiplicative changes in expected extreme precipitation magnitude from current to future period—were computed for grid cells from the downscaled climate datasets containing National Oceanic and Atmospheric Administration Atlas 14 stations in central and south Florida. Change factors for specific durations and return periods may be used to scale the National Oceanic and Atmospheric Administration Atlas 14 historical depth-duration-frequency values to the period 2050–89 on the basis of changes in extreme precipitation derived from downscaled climate datasets. Model culling was implemented to select downscaled climate models that best captured observed historical patterns of precipitation extremes in central and south Florida. Overall, a large variation in change factors across downscaled climate datasets was found, with change factors generally greater than one and increasing with return period. In general, median change factors were higher for the south-central Florida climate region (1.05–1.55 depending on downscaled climate dataset, duration, and return period) than for the south Florida climate region (1–1.4 depending on downscaled climate dataset, duration, and return period) when considering best performing models for both areas, indicating a projected overall increase in future extreme precipitation events.

Florida↗
USGS2023

Challenges and solutions for automated avian recognition in aerial imagery

Remote aerial sensing provides a non-invasive, large geographical-scale technology for avian monitoring, but the manual processing of images limits its development and applications. Artificial Intelligence (AI) methods can be used to mitigate this manual image processing requirement. The implementation of AI methods, however, has several challenges: (1) imbalanced (i.e., long-tailed) data distribution, (2) annotation uncertainty in categorization, and (3) dataset discrepancies across different study sites. Here we use aerial imagery data of waterbirds around Cape Cod and Lake Michigan in the United States to examine how these challenges limit avian recognition performance. We review existing solutions and demonstrate as use cases how methods like Label Distribution Aware Marginal Loss with Deferred Re-Weighting, hierarchical classification, and FixMatch address the three challenges. We also present a new approach to tackle the annotation uncertainty challenge using a Soft-fine Pseudo-Label methodology. Finally, we aim with this paper to increase awareness in the ecological remote sensing community of these challenges and bridge the gap between ecological applications and state-of-the-art computer science, thereby opening new doors to future research.

Remote Sensing in Ecology and Conservation↗
USGS2022

Manatee population traits elucidated through photo-identification

Data on the demography and distribution of wildlife populations are important for informing conservation and management decisions; however, determination of life history traits and population trends often are elusive. All four extant species in the order Sirenia are deemed vulnerable to extinction; therefore, determining the demography and distribution for populations worldwide is crucial. Aerial surveys, radio-tagging and tracking, genetic sampling and analyses, health assessments, carcass examination, and photographic documentation are all techniques used to study sirenian populations. A 40 +-year computer-aided catalog of images and demography data collected on Florida manatees enables searches of individuals by descriptions of feature (scar) types and has enabled estimates of annual survival and reproductive rates, documented extra-limital movements, and advanced modeling designs. Photography is discussed as a method for the documentation of unique and acquired features specifically on Florida manatees. By means of these features, individual Florida manatees have been re-identified as far from their established range as Cape Cod, Massachusetts, Houston, Texas, and in Cuba, The Bahamas, and Mexico. The length of gestation (11–13 months) and calf dependency (1–3 years), and potential longevity in the wild (> 50 years), have been verified. To meet the challenge of an increasing number of images collected with the advent of digital photography, there has been an increasing interest and potential for new techniques to assist with individual identification. Several researchers are utilizing drones and artificial intelligence to find, photograph, and streamline the individual identification of sirenians as well as other marine mammal species. New techniques have potential to simplify the photographic identification of Florida manatees. Photographic documentation could be a model for demographic and distribution research of sirenian populations outside of Florida and as a tool to monitor the viability of sirenian populations, particularly as threats emerge due to anthropogenic pressures and global climate change.

Mammalian Biology↗
USGS2024

A multi-sensor approach to characterize winter water-level drawdown patterns in lakes

Artificial manipulation of lake water levels through practices like winter water-level drawdown (WD) is prevalent across many regions, but the spatiotemporal patterns are not well documented due to limited in situ monitoring. Multi-sensor satellite remote sensing provides an opportunity to map and analyze drawdown frequency and metrics (timing, magnitude, duration) at broad scales. This study developed a cloud computing framework to process time series of synthetic aperture radar (Sentinel 1-SAR) and optical sensor (Landsat 8, Sentinel 2) data to characterize WD in 166 lakes across Massachusetts, USA, during 2016–2021. Comparisons with in situ logger data showed that the Sentinel 1-derived surface water area captured relative water-level fluctuations indicative of WD. A machine learning approach classified lakes as WD versus non-WD based on seasonal water-level fluctuations derived from Sentinel 1-SAR data. The framework mapped WD lakes statewide, revealing prevalence throughout Massachusetts with interannual variability. Results showed WDs occurred in over 75% of lakes during the study period, with high interannual variability in the number of lakes conducting WD. Mean WD magnitude was highest in the wettest year (2018) but % lake area exposure did not show any association with precipitation and varied between 8% to 12% over the 5-year period. WD start date was later and duration was longer in wet years, indicating climate mediation of WD implementation driven by management decisions. The data and tools developed provide an objective information resource to evaluate ecological impacts and guide management of this prevalent but understudied phenomenon. Overall, the results and interactive web tool developed as part of this study provide new hydrologic intelligence to inform water management and policies related to WD practices.

Massachusetts↗