Geology Reports⌕ Search

SEARCH · Geology Reports

Results for “Algorithms”

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.

At least 1,477 records · Page 82Linked to original sources
USGS2011

Assimilating models and data to enhance predictions of shoreline evolution

A modeling system that considers both long- and short-term process-driven shoreline change is presented. The modeling system is integrated into a data assimilation framework that uses sparse observations of shoreline change to correct a model forecast and to determine unobserved model variables and free parameters. Application of the assimilation algorithm also provides quantitative statistical estimates of uncertainty that can be applied to coastal hazard and vulnerability assessments. Significant attention is given to the estimation of four non-observable quantities using the data assimilation framework that utilizes only one observable process (i.e. ,shoreline change). The general framework discussed here can be applied to many other geophysical processes by simply changing the model component to one applicable to the processes of interest.

Conference Paper↗
USGS2002

Monitoring black-tailed prairie dog colonies with high-resolution satellite imagery

The United States Fish and Wildlife Service has determined that the black-tailed prairie dog (Cynomys ludovicianus) warrants listing as a threatened species under the Endangered Species Act. Central to any conservation planning for the black-tailed prairie dog is an appropriate detection and monitoring technique. Because coarse-resolution satellite imagery is not adequate to detect black-tailed prairie dog colonies, we examined the usefulness of recently available high-resolution (1-m) satellite imagery. In 6 purchased scenes of national grasslands, we were easily able to visually detect small and large colonies without using image-processing algorithms. The Ikonos (Space Imaging(tm)) satellite imagery was as adequate as large-scale aerial photography to delineate colonies. Based on the high quality of imagery, we discuss a possible monitoring program for black-tailed prairie dog colonies throughout the Great Plains, using the species' distribution in North Dakota as an example. Monitoring plots could be established and imagery acquired periodically to track the expansion and contraction of colonies.

Wildlife Society Bulletin↗
USGS1985

Optimal management strategies in variable environments: Stochastic optimal control methods

Dynamic optimization was used to investigate the optimal defoliation of salt desert shrubs in north-western Utah. Management was formulated in the context of optimal stochastic control theory, with objective functions composed of discounted or time-averaged biomass yields. Climatic variability and community patterns of salt desert shrublands make the application of stochastic optimal control both feasible and necessary. A primary production model was used to simulate shrub responses and harvest yields under a variety of climatic regimes and defoliation patterns. The simulation results then were used in an optimization model to determine optimal defoliation strategies. The latter model encodes an algorithm for finite state, finite action, infinite discrete time horizon Markov decision processes. Three questions were addressed: (i) What effect do changes in weather patterns have on optimal management strategies? (ii) What effect does the discounting of future returns have? (iii) How do the optimal strategies perform relative to certain fixed defoliation strategies? An analysis was performed for the three shrub species, winterfat (Ceratoides lanata), shadscale (Atriplex confertifolia) and big sagebrush (Artemisia tridentata). In general, the results indicate substantial differences among species in optimal control strategies, which are associated with differences in physiological and morphological characteristics. Optimal policies for big sagebrush varied less with variation in climate, reserve levels and discount rates than did either shadscale or winterfat. This was attributed primarily to the overwintering of photosynthetically active tissue and to metabolic activity early in the growing season. Optimal defoliation of shadscale and winterfat generally was more responsive to differences in plant vigor and climate, reflecting the sensitivity of these species to utilization and replenishment of carbohydrate reserves. Similarities could be seen in the influence of both the discount rate and the climatic patterns on optimal harvest strategics. In general, decreases in either the discount rate or in the frequency of favorable weather patterns lcd to a more conservative defoliation policy. This did not hold, however, for plants in states of low vigor. Optimal control for shadscale and winterfat tended to stabilize on a policy of heavy defoliation stress, followed by one or more seasons of rest. Big sagebrush required a policy of heavy summer defoliation when sufficient active shoot material is present at the beginning of the growing season. The comparison of fixed and optimal strategies indicated considerable improvement in defoliation yields when optimal strategies are followed. The superior performance was attributable to increased defoliation of plants in states of high vigor. Improvements were found for both discounted and undiscounted yields.

Journal of Environmental Management↗
USGS2011

Long-period earthquake simulations in the Wasatch Front, UT: misfit characterization and ground motion estimates

In this research we characterize the goodness-of-fit between observed and synthetic seismograms from three small magnitude (M3.6-4.5) earthquakes in the region using the Wasatch Front community velocity model (WCVM) in order to determine the ability of the WCVM to predict earthquake ground motions for scenario earthquake modeling efforts. We employ the goodness-of-fit algorithms and criteria of Olsen and Mayhew (2010). In focusing comparisons on the ground motion parameters that are of greatest importance in engineering seismology, we find that the synthetic seismograms calculated using the WCVM produce a fair fit to the observed ground motion records up to a frequency of 0.5 Hz for two of the modeled earthquakes and up to 0.1 Hz for one of the earthquakes. In addition to the reference seismic material model (WCVM), we carry out earthquake simulations using material models with perturbations to the regional seismic model and with perturbations to the deep sedimentary basins. Simple perturbations to the regional seismic velocity model and to the seismic velocities of the sedimentary basin result in small improvements in the observed misfit but do not indicate a significantly improved material model. Unresolved differences between the observed and synthetic seismograms are likely due to un-modeled heterogeneities and incorrect basin geometries in the WCVM. These differences suggest that ground motion prediction accuracy from deterministic modeling varies across the region and further efforts to improve the WCVM are needed.

Utah↗
USGS2012

Ensemble forecasting of potential habitat for three invasive fishes

Aquatic invasive species pose major ecological and economic threats to aquatic ecosystems worldwide via displacement, predation, or hybridization with native species and the alteration of aquatic habitats and hydrologic cycles. Modeling the habitat suitability of alien aquatic species through spatially explicit mapping is an increasingly important risk assessment tool. Habitat modeling also facilitates identification of key environmental variables influencing invasive species distributions. We compared four modeling methods to predict the potential continental United States distributions of northern snakehead Channa argus (Cantor, 1842), round goby Neogobius melanostomus (Pallas, 1814), and silver carp Hypophthalmichthys molitrix (Valenciennes, 1844) using maximum entropy (Maxent), the genetic algorithm for rule set production (GARP), DOMAIN, and support vector machines (SVM). We used inventory records from the USGS Nonindigenous Aquatic Species Database and a geographic information system of 20 climatic and environmental variables to generate individual and ensemble distribution maps for each species. The ensemble maps from our study performed as well as or better than all of the individual models except Maxent. The ensemble and Maxent models produced significantly higher accuracy individual maps than GARP, one-class SVMs, or DOMAIN. The key environmental predictor variables in the individual models were consistent with the tolerances of each species. Results from this study provide insights into which locations and environmental conditions may promote the future spread of invasive fish in the US.

Aquatic Invasions↗
USGS2007

Local search for optimal global map generation using mid-decadal landsat images

NASA and the US Geological Survey (USGS) are seeking to generate a map of the entire globe using Landsat 5 Thematic Mapper (TM) and Landsat 7 Enhanced Thematic Mapper Plus (ETM+) sensor data from the "mid-decadal" period of 2004 through 2006. The global map is comprised of thousands of scene locations and, for each location, tens of different images of varying quality to chose from. Furthermore, it is desirable for images of adjacent scenes be close together in time of acquisition, to avoid obvious discontinuities due to seasonal changes. These characteristics make it desirable to formulate an automated solution to the problem of generating the complete map. This paper formulates a Global Map Generator problem as a Constraint Optimization Problem (GMG-COP) and describes an approach to solving it using local search. Preliminary results of running the algorithm on image data sets are summarized. The results suggest a significant improvement in map quality using constraint-based solutions. Copyright ?? 2007, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.

Conference Paper↗
USGS1988

Fractional-step method for two-dimensional estuarine transport

The fractional-step method was used in this study to split the longitudinal advective transport term from the other terms in the two-dimensional, laterally-averaged equation for estuarine mass transport. The method of characteristics with spline interpolations was used to approximate the longitudinal advective transport. A general discussion of the fractional-step method, the specific algorithm developed in this investigation, and results of numerical tests are presented. Application of the fractional-step method in conjunction with the characteristic-spline scheme offers the potential for improved simulations of transport for situations in which concentration gradients are steep.

Conference Paper↗
USGS1985

MONITORING THE EARTH - TOO MANY PLAYERS?

Remote sensing from satellites provides a unique tool to measure the parameters of the Earth on a worldwide scale. A number of organizations are currently engaged in, or proposing to embark on, worldwide measurement/monitoring programs. Program objectives vary in type and complexity, including a form of technical library and an ambitious experiment to validate algorithms to derive land surface climatological parameters. Modeling and understanding the Earth as an integrated system is an immense undertaking and will require at least as many measurements as currently acquired. Development of a long-term program of Earth system monitoring/modeling, such as the International Global Change Program proposed by the International Council of Scientific Unions, could provide the necessary focus and structure for effective international coordination and cooperation.

Conference Paper↗
USGS1984

HYDRAULIC RESEARCH - U. S. GEOLOGICAL SURVEY.

Research at the Gulf Coast Hydroscience Center near Bay St. Louis, Mississippi, is being conducted in an indoor laboratory, and an outdoor laboratory. Much of the current indoor lab research is directed at improved methods of measuring flow. A towing tank and submerged jet tank are used for calibrating velocity meters. The outdoor laboratory consists of a flood-plain simulation facility. Much emphasis has been placed on applying results to the development of numerical simulation models. Laboratory data are used to develop and validate the models. These models are also validated with actual field data. Various algorithms for one- and two-dimensional flow and transport models based on finite difference and finite element schemes are being developed.

Conference Paper↗
USGS1988

Simulation and video animation of canal flushing created by a tide gate

A tide-gate algorithm was added to a one-dimensional unsteady flow model that was calibrated, verified, and used to determine the locations of as many as five tide gates that would maximize flushing in two canal systems. Results from the flow model were used to run a branched Lagrangian transport model to simulate the flushing of a conservative constituent from the canal systems both with and without tide gates. A tide gate produces a part-time riverine flow through the canal system that improves flushing along the flow path created by the tide gate. Flushing with no tide gates and with a single optimally located tide gate are shown with a video animation.

Conference Paper↗
USGS1984

TECHNIQUE FOR ENHANCING DIGITAL COLOR IMAGES BY CONTRAST STRETCHING IN MUNSELL COLOR SPACE.

The Munsell color system can be used to further enhance the appearance of high-quality digital color-composite images. A color-balanced 'standard' color-composite image is first produced using any desired contrast stretching algorithm. The stretched digital data are then transformed into the cylindrical Munsell color space. An enhanced version of a color-composite image is produced by stretching the saturation parameter over the full digital range and inverting the modified Munsell coordinates to red-blue-green (tristimulus) data space. The resulting image has greater color-saturation contrast than the original image, without hue change. Contrast stretching in Munsell color space reduces the correlation between individual bands or ratios and is similar to decorrelation processing based on principal-components transforms. However, principal components are based on data variance, with less variance being explained by each higher order component.

Conference Paper↗
USGS1987

GEOMETRIC PROCESSING OF DIGITAL IMAGES OF THE PLANETS.

New procedures and software have been developed for geometric transformations of images to support digital cartography of the planets. The procedures involve the correction of spacecraft camera orientation of each image with the use of ground control and the transformation of each image to a Sinusoidal Equal-Area map projection with an algorithm which allows the number of transformation calculations to vary as the distortion varies within the image. When the distortion is low in an area of an image, few transformation computations are required, and most pixels can be interpolated. When distortion is extreme, the location of each pixel is computed. Mosaics are made of these images and stored as digital databases.

Photogrammetric Engineering and Remote Sensing↗
USGS1986

Processing techniques for digital sonar images from GLORIA.

Image processing techniques have been developed to handle data from one of the newest members of the remote sensing family of digital imaging systems. This paper discusses software to process data collected by the GLORIA (Geological Long Range Inclined Asdic) sonar imaging system, designed and built by the Institute of Oceanographic Sciences (IOS) in England, to correct for both geometric and radiometric distortions that exist in the original 'raw' data. Preprocessing algorithms that are GLORIA-specific include corrections for slant-range geometry, water column offset, aspect ratio distortion, changes in the ship's velocity, speckle noise, and shading problems caused by the power drop-off which occurs as a function of range.-from Author

Photogrammetric Engineering and Remote Sensing↗
USGS1990

Relative efficiency of four parameter-estimation methods in steady-state and transient ground-water flow models

Parameters in numerical ground-water flow models have been successfully estimated using nonlinear-optimization methods such as the modified Gauss-Newton (GN) method and conjugate-direction methods. This paper investigates the relative efficiency of GN and three conjugate-direction parameter-estimation methods on two-dimensional, steady-state and transient ground-water flow test cases. The steady-state test cases are included to compare the performance of the algorithm with published examples. The three conjugate-direction methods are the Fletcher-Reeves (FR) and quasi-Newton (QN) regression methods, and combination Fletcher-Reeves quasi-Newton (FR-QN). All three are combined with Newton's method of calculating step size. The numerical ground-water flow model is described by McDonald and Harbaugh.

Conference Paper↗
USGS1990

Adaptive box filters for removal of random noise from digital images

We have developed adaptive box-filtering algorithms to (1) remove random bit errors (pixel values with no relation to the image scene) and (2) smooth noisy data (pixels related to the image scene but with an additive or multiplicative component of noise). For both procedures, we use the standard deviation (??) of those pixels within a local box surrounding each pixel, hence they are adaptive filters. This technique effectively reduces speckle in radar images without eliminating fine details. -from Authors

Photogrammetric Engineering and Remote Sensing↗
USGS1992

Comparison of three newton-like nonlinear least-squares methods for estimating parameters of ground-water flow models

Three methods of solving nonlinear least-squares problems were compared for robustness and efficiency using a series of hypothetical and field problems. A modified Gauss-Newton/full Newton hybrid method (MGN/FN) and an analogous method for which part of the Hessian matrix was replaced by a quasi-Newton approximation (MGN/QN) solved some of the problems with appreciably fewer iterations than required using only a modified Gauss-Newton (MGN) method. In these problems, model nonlinearity and a large variance for the observed data apparently caused MGN to converge more slowly than MGN/FN or MGN/QN after the sum of squared errors had almost stabilized. Other problems were solved as efficiently with MGN as with MGN/FN or MGN/QN. Because MGN/FN can require significantly more computer time per iteration and more computer storage for transient problems, it is less attractive for a general purpose algorithm than MGN/QN.

Conference Paper↗
USGS1997

Generation of coastal marsh topography with radar and ground-based measurements

A topographic surface of a low lying coastal marsh was created by using three flood extent vectors digitized from ERS-1 SAR images and two elevation contours from U.S. Geological Survey topographic quadrangles. Point measurement of water depth at the times of the SAR collections allowed conversion of the radar measured flood extent vectors to topographic contours. Generation of the topographic surface was accomplished with a surface gridding algorithm. SAR and on-site measures. Errors in the generated topography were mainly associated with the lack of input contours covering narrow to broad plateaus and topographic highs and lows. The misplacement of SAR derived flood extent vectors also caused errors in sparsely vegetated high marsh at convoluted marsh-forest boundaries, and at topographic depressions. Overall, the standard deviation of differences between measured and predicted elevations at 747 points was 19 cm. Excluding the above mentioned abrupt boundaries and topographic highs and lows outside the range of available contours, standard deviation differences averaged about 14 cm, but most often averaged about 8 cm. This suggested a 5 to 9 factor improvement over the 150 cm topographic resolution currently available for this area.

Journal of Coastal Research↗
USGS1998

Generation of coastal marsh topography with radar and ground-based measurements

A topographic surface of a low lying coastal marsh was created by using three flood extent vectors digitized from ERS-1 SAR images and two elevation contours from U.S. Geological Survey topographic quadrangles. Point measurement of water depth at the times of the SAR collections allowed conversion of the radar measured flood extent vectors to topographic contours. Generation of the topographic surface was accomplished with a surface gridding algorithm, SAR and on-site measures. Errors in the generated topography were mainly associated with the lack of input contours covering narrow to broad plateaus and topographic highs and lows. The misplacement of SAR derived flood extent vectors also caused errors in sparsely vegetated high marsh, at convoluted marsh-forest boundaries, and at topographic depressions. Overall, the standard deviation of differences between measured and predicted elevations at 747 points was 19 cm. Excluding the above mentioned abrupt boundaries and topographic highs and lows outside the range of available contours, standard deviation differences averaged about 14 cm, but most often averaged about 8 cm. This suggested a 5 to 9 factor improvement over the 150 cm topographic resolution currently available for this area.

Journal of Coastal Research↗