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.

SourceResearch connectionFrom yearThrough year

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

At least 631 records · Page 35Linked to original sources

USGS2016

Detecting grizzly bear use of ungulate carcasses using global positioning system telemetry and activity data

Global positioning system (GPS) wildlife collars have revolutionized wildlife research. Studies of predation by free-ranging carnivores have particularly benefited from the application of location clustering algorithms to determine when and where predation events occur. These studies have changed our understanding of large carnivore behavior, but the gains have concentrated on obligate carnivores. Facultative carnivores, such as grizzly/brown bears ( Ursus arctos ), exhibit a variety of behaviors that can lead to the formation of GPS clusters. We combined clustering techniques with field site investigations of grizzly bear GPS locations ( n = 732 site investigations; 2004–2011) to produce 174 GPS clusters where documented behavior was partitioned into five classes (large-biomass carcass, small-biomass carcass, old carcass, non-carcass activity, and resting). We used multinomial logistic regression to predict the probability of clusters belonging to each class. Two cross-validation methods—leaving out individual clusters, or leaving out individual bears—showed that correct prediction of bear visitation to large-biomass carcasses was 78–88%, whereas the false-positive rate was 18–24%. As a case study, we applied our predictive model to a GPS data set of 266 bear-years in the Greater Yellowstone Ecosystem (2002–2011) and examined trends in carcass visitation during fall hyperphagia (September–October). We identified 1997 spatial GPS clusters, of which 347 were predicted to be large-biomass carcasses. We used the clustered data to develop a carcass visitation index, which varied annually, but more than doubled during the study period. Our study demonstrates the effectiveness and utility of identifying GPS clusters associated with carcass visitation by a facultative carnivore.

Greater Yellowstone Ecosystem↗

USGS2006

Comparison of COSPEC and two miniature ultraviolet spectrometer systems for SO 2 measurements using scattered sunlight

The correlation spectrometer (COSPEC), the principal tool for remote measurements of volcanic SO 2 , is rapidly being replaced by low-cost, miniature, ultraviolet (UV) spectrometers. We compared two of these new systems with a COSPEC by measuring SO 2 column amounts at Kīlauea Volcano, Hawaii. The two systems, one calibrated using in-situ SO 2 cells, and the other using a calibrated laboratory reference spectrum, employ similar spectrometer hardware, but different foreoptics and spectral retrieval algorithms. Accuracy, signal-to-noise, retrieval parameters, and precision were investigated for the two configurations of new miniature spectrometer. Measurements included traverses beneath the plumes from the summit and east rift zone of Kīlauea, and testing with calibration cells of known SO 2 concentration. The results obtained from the different methods were consistent with each other, with <8% difference in estimated SO 2 column amounts up to 800 ppm m. A further comparison between the COSPEC and one of the miniature spectrometer configurations, the ‘FLYSPEC’, spans an eight month period and showed agreement of measured emission rates to within 10% for SO 2 column amounts up to 1,600 ppm m. The topic of measuring high SO 2 burdens accurately is addressed for the Kīlauea measurements. In comparing the foreoptics, retrieval methods, and resultant implications for data quality, we aim to consolidate the various experiences to date, and improve the application and development of miniature spectrometer systems.`

Bulletin of Volcanology↗

USGS2022

Volcano geodesy using InSAR in 2020: The past and next decades

The study of volcano deformation has grown significantly through they year 2020 since the development of interferometric synthetic aperture radar (InSAR) in the 1990s. This relatively new data source, which provides evidence of changes in subsurface magma storage and pressure without the need for ground-based equipment, has matured during the past decade. It now provides a means to address previously inaccessible questions and offers input to increasingly complex models of magmatic processes. Here, we review how technological advances in InSAR during 2010-2020 have facilitated our ability to monitor and interpret volcanic processes, primarily through rapid and accurate observations of the changing surfaces at active volcanoes worldwide. Specifically, we examine how current systems achieve excellent resolution in time and space, provide global coverage, and generate products that are easy to use by non-specialists—factors that have often limited the practical study of volcanoes using radar measurements. We also look to the future, offering our perspective about how advancements in technology and data management in the decade to come will increase the value and accessibility of InSAR applied to the geodetic study of volcanoes and monitoring of hazardous volcanic processes. New developments will include the launch of additional satellites by both public space agencies and private companies, as well as implementation of algorithms for exploiting the growing volumes of data. To meet their full potential, these efforts will require coordination between data users and data providers so that the relevant imagery is acquired, made available to volcanologists in a timely fashion, and utilized to assess and mitigate volcanic hazards.

Bulletin of Volcanology↗

USGS2026

Detecting volcanic deformation in Hawaii using trustworthy multimodal deep learning techniques

Monitoring volcanoes involves a variety of data sources and methods to maintain complete continuity of coverage. Global navigation satellite system (GNSS) and interferometric synthetic aperture radar (InSAR) are commonly used complementary methods to assess the deformation state of a volcano as magma migrates beneath the surface. The amount of data these methods produce, however, is growing rapidly beyond human analysis capabilities and is becoming difficult to manage. Here, we create a novel multimodal deep learning framework to ingest InSAR and GNSS data simultaneously and classify the deformation state of the system. We apply this methodology to Mauna Loa, Hawai‘i given its wealth of InSAR and GNSS data as well as its propensity to deform on multiple timescales. Our model performs with high accuracy and is able to identify both slow and fast deformation from 2015 to 2023. The multimodal nature of our model also allows us to identify the presence of atmospheric noise in InSAR data. Furthermore, we employ explainability algorithms to show that our model is making decisions for the right reasons and to connect complex black-box machine learning mappings to current real-world geodetic interpretations of the Mauna Loa magmatic system.

Hawaii↗

USGS2026

Thermal infrared and ultraviolet remote sensing of sulfur dioxide gas emitted during the 2018 lower East Rift Zone eruption of Kilauea, Hawaiʻi

Ultraviolet (UV) remote sensing is widely used to detect volcanic sulfur dioxide (SO₂) due to its high sensitivity and favorable spatial and temporal resolution. However, significant discrepancies have been reported between ground-based and satellite-based UV observations of dense volcanic plumes. A notable example is the 2018 lower East Rift Zone eruption of Kīlauea, where SO₂ emission rates derived from ground-based Differential Optical Absorption Spectroscopy (DOAS) measurements differed substantially from those obtained by the spaceborne Tropospheric Monitoring Instrument (TROPOMI). In this study, we investigate these differences by applying thermal infrared (TIR) satellite retrievals using a modified version of the SO₂-ALTA algorithm to Moderate Resolution Imaging Spectroradiometer (MODIS) observations. The resulting TIR-derived SO₂ fluxes are compared with ground-based DOAS data, satellite UV observations, and petrological estimates of gas emissions. Our results show strong agreement between TIR-derived fluxes, ground-based DOAS measurements, and petrological estimates, particularly during the peak and plateau phases of the eruption. In contrast, satellite UV-derived SO₂ emissions are systematically lower. We find that TIR observations are more effective in quantifying high-concentration SO₂ plumes in the near-vent region, while UV measurements are more sensitive under lower-concentration conditions but more affected by scattering in optically dense plumes. These findings highlight the complementary strengths of UV and TIR remote sensing techniques and emphasize the importance of accounting for plume density and observation geometry when interpreting satellite SO₂ retrievals.

Hawaii↗

USGS2019

Self-organizing maps for compositional data: coal combustion products of a Wyoming power plant

A self-organizing map (SOM) is a non-linear projection of a D-dimensional data set, where the distance among observations is approximately preserved on to a lower dimensional space. The SOM arranges multivariate data based on their similarity to each other by allowing pattern recognition leading to easier interpretation of higher dimensional data. The SOM algorithm allows for selection of different map topologies, distances and parameters, which determine how the data will be organized on the map. In the particular case of compositional data (such as elemental, mineralogical, or maceral abundance), the sample space is governed by Aitchison geometry and extra steps are required prior to their SOM analysis. Following the principle of working on log-ratio coordinates, the simplicial operations and the Aitchison distance, which are appropriate elements for the SOM, are presented. With this structure developed, a SOM using Aitchison geometry is applied to properly interpret elemental data from combustion products (bottom ash, fly ash, and economizer fly ash) in a Wyoming coal-fired power plant. Results from this effort provide knowledge about the differences between the ash composition in the coal combustion process.

Stochastic Environmental Research and Risk Assessm↗

USGS2013

Uncertainty in simulated groundwater-quality trends in transient flow

In numerical modeling of groundwater flow, the result of a given solution method is affected by the way in which transient flow conditions and geologic heterogeneity are simulated. An algorithm is demonstrated that simulates breakthrough curves at a pumping well by convolution-based particle tracking in a transient flow field for several synthetic basin-scale aquifers. In comparison to grid-based (Eulerian) methods, the particle (Lagrangian) method is better able to capture multimodal breakthrough caused by changes in pumping at the well, although the particle method may be apparently nonlinear because of the discrete nature of particle arrival times. Trial-and-error choice of number of particles and release times can perhaps overcome the apparent nonlinearity. Heterogeneous aquifer properties tend to smooth the effects of transient pumping, making it difficult to separate their effects in parameter estimation. Porosity, a new parameter added for advective transport, can be accurately estimated using both grid-based and particle-based methods, but predictions can be highly uncertain, even in the simple, nonreactive case.

Hydrogeology Journal↗

USGS2003

Evaluation of airborne thermal infrared imagery for locating mine drainage sites in the Lower Kettle Creek and Cooks Run Basins, Pennsylvania, USA

High-resolution airborne thermal infrared (TIR) imagery data were collected over 90.6 km 2 (35 mi 2 ) of remote and rugged terrain in the Kettle Creek and Cooks Run Basins, tributaries of the West Branch of the Susquehanna River in north-central Pennsylvania. The purpose of this investigation was to evaluate the effectiveness of TIR for identifying sources of acid mine drainage (AMD) associated with abandoned coal mines. Coal mining from the late 1800s resulted in many AMD sources from abandoned mines in the area. However, very little detailed mine information was available, particularly on the source locations of AMD sites. Potential AMD sources were extracted from airborne TIR data employing custom image processing algorithms and GIS data analysis. Based on field reconnaissance of 103 TIR anomalies, 53 sites (51%) were classified as AMD. The AMD sources had low pH (<4) and elevated concentrations of iron and aluminum. Of the 53 sites, approximately 26 sites could be correlated with sites previously documented as AMD. The other 27 mine discharges identified in the TIR data were previously undocumented. This paper presents a summary of the procedures used to process the TIR data and extract potential mine drainage sites, methods used for field reconnaissance and verification of TIR data, and a brief summary of water-quality data.

Pennsylvania↗

USGS2010

Parameter-expanded data augmentation for Bayesian analysis of capture-recapture models

Data augmentation (DA) is a flexible tool for analyzing closed and open population models of capture-recapture data, especially models which include sources of hetereogeneity among individuals. The essential concept underlying DA, as we use the term, is based on adding "observations" to create a dataset composed of a known number of individuals. This new (augmented) dataset, which includes the unknown number of individuals N in the population, is then analyzed using a new model that includes a reformulation of the parameter N in the conventional model of the observed (unaugmented) data. In the context of capture-recapture models, we add a set of "all zero" encounter histories which are not, in practice, observable. The model of the augmented dataset is a zero-inflated version of either a binomial or a multinomial base model. Thus, our use of DA provides a general approach for analyzing both closed and open population models of all types. In doing so, this approach provides a unified framework for the analysis of a huge range of models that are treated as unrelated "black boxes" and named procedures in the classical literature. As a practical matter, analysis of the augmented dataset by MCMC is greatly simplified compared to other methods that require specialized algorithms. For example, complex capture-recapture models of an augmented dataset can be fitted with popular MCMC software packages (WinBUGS or JAGS) by providing a concise statement of the model's assumptions that usually involves only a few lines of pseudocode. In this paper, we review the basic technical concepts of data augmentation, and we provide examples of analyses of closed-population models ( M 0, M h , distance sampling, and spatial capture-recapture models) and open-population models (Jolly-Seber) with individual effects.

Journal of Ornithology↗

USGS2010

Challenges and opportunities for integrating lake ecosystem modelling approaches

A large number and wide variety of lake ecosystem models have been developed and published during the past four decades. We identify two challenges for making further progress in this field. One such challenge is to avoid developing more models largely following the concept of others (‘ reinventing the wheel ’). The other challenge is to avoid focusing on only one type of model, while ignoring new and diverse approaches that have become available (‘ having tunnel vision ’). In this paper, we aim at improving the awareness of existing models and knowledge of concurrent approaches in lake ecosystem modelling, without covering all possible model tools and avenues. First, we present a broad variety of modelling approaches. To illustrate these approaches, we give brief descriptions of rather arbitrarily selected sets of specific models. We deal with static models (steady state and regression models), complex dynamic models (CAEDYM, CE-QUAL-W2, Delft 3D-ECO, LakeMab, LakeWeb, MyLake, PCLake, PROTECH, SALMO), structurally dynamic models and minimal dynamic models. We also discuss a group of approaches that could all be classified as individual based: super-individual models (Piscator, Charisma), physiologically structured models, stage-structured models and trait-based models. We briefly mention genetic algorithms, neural networks, Kalman filters and fuzzy logic. Thereafter, we zoom in, as an in-depth example, on the multi-decadal development and application of the lake ecosystem model PCLake and related models (PCLake Metamodel, Lake Shira Model, IPH-TRIM3D-PCLake). In the discussion, we argue that while the historical development of each approach and model is understandable given its ‘leading principle’, there are many opportunities for combining approaches. We take the point of view that a single ‘right’ approach does not exist and should not be strived for. Instead, multiple modelling approaches, applied concurrently to a given problem, can help develop an integrative view on the functioning of lake ecosystems. We end with a set of specific recommendations that may be of help in the further development of lake ecosystem models.

Aquatic Ecology↗

USGS2023

Biophysical factors control invasive annual grass hot spots in the Mojave Desert

Invasive annual grasses can promote ecosystem state changes and habitat loss in the American Southwest. Non-native annual grasses such as Bromus spp . a nd Schismus spp. have invaded the Mojave Desert and degraded habitat through increased fire occurrence, severity, and shifting plant community composition. Thus, it is important to identify and characterize the areas where persistent invasion has occurred, identifying where subsequent habitat degradation has increased. Previous plot and landscape-scale analyses have revealed anthropogenic and biophysical correlates with the establishment and dominance of invasive annual grasses in the Mojave Desert. However, these studies have been limited in spatial and temporal scales. Here we use Landsat imagery validated using an extensive network of plot data to map persistent and productive populations of invasive annual grass, called hot spots , across the entire Mojave Desert ecoregion over 12 years (2009–2020). We also identify important variables for predicting hot spot distribution using the Random Forest algorithm and identifying the most invaded subregions. We identified hot spots in over 5% of the Mojave Desert mostly on the western and eastern edges of the ecoregion, and invasive grasses were detected in over 90% of the Mojave Desert at least once in that time. Across the entire Mojave Desert, our results indicate that soil texture, aspect, winter precipitation, and elevation are the highest-ranking predictive variables of invasive grass hot spots , while anthropogenic variables contributed the least to the accuracy of the predictive model. The total area covered by hot spots varied significantly among subregions of the Mojave Desert. We found that anthropogenic variables became more important in explaining invasive annual establishment and persistence as spatial scale was reduced to the subregional level. Our findings have important implications for informing where land management actions can prioritize reducing invasive annual persistence and promoting restoration efforts.

California↗

USGS2007

Disentangling the role of hybridization in the evolution of the endangered Arizona cliffrose (Purshia subintegra; Rosaceae): A molecular and morphological analysis

Hybridization may threaten the conservation status of rare species through genetic assimilation and may confound the ability to distinguish among taxa. We studied these issues in an endangered shrub, Purshia subintegra (Rosaceae), known from four populations growing on limestone outcrops in central Arizona (USA). Using amplified fragment length polymorphisms (AFLP) and the Bayesian clustering algorithm implemented in STRUCTURE, we identified three distinct genetic lineages among Arizona Purshia subintegra and P. stansburiana. An initial split divided San Carlos Basin P. subintegra (considered P. pinkavae by Schaack) from northern P. stansburiana populations (FST = 0.394). A subsequent split separated northern P. stansburiana from two P. subintegra populations at Horseshoe Lake and Burro Creek (FST = 0.207), which comprised a nearly perfect admixture of the two lineages identified in the initial analysis. In the Verde River Valley P. subintegra is sympatric with P. stansburiana and exhibited an average 27% P. stansburiana genes for 5 of 6 stands analyzed, indicating ongoing hybridization and backcrossing with P. subintegra. Individuals carrying >90% P. subintegra markers are identifiable 68% of the time based on morphology, with leaf lobing, leaf size, and leaf length acting as the most reliable indicators of taxonomic status. However, the genetic and morphological distance correlation among individuals was low (r = 0.17, P = 0.0002), indicating that morphology cannot always accurately predict genetic admixture or taxonomy. Overall, our study confirmed the genetic distinctiveness of the San Carlos Basin population, an ancient natural hybrid origin of P. subintegra, and the presence of a hybrid swarm in the Verde Valley, whose conservation value may lie in its heightened genetic diversity. ?? 2007 Springer Science+Business Media B.V.

Conservation Genetics↗

USGS2012

The genetic structure of a relict population of wood frogs

Habitat fragmentation and the associated reduction in connectivity between habitat patches are commonly cited causes of genetic differentiation and reduced genetic variation in animal populations. We used eight microsatellite markers to investigate genetic structure and levels of genetic diversity in a relict population of wood frogs ( Lithobates sylvatica ) in Rocky Mountain National Park, Colorado, where recent disturbances have altered hydrologic processes and fragmented amphibian habitat. We also estimated migration rates among subpopulations, tested for a pattern of isolation-by-distance, and looked for evidence of a recent population bottleneck. The results from the clustering algorithm in Program STRUCTURE indicated the population is partitioned into two genetic clusters (subpopulations), and this result was further supported by factorial component analysis. In addition, an estimate of FST (FST = 0.0675, P value \0.0001) supported the genetic differentiation of the two clusters. Estimates of migration rates among the two subpopulations were low, as were estimates of genetic variability. Conservation of the population of wood frogs may be improved by increasing the spatial distribution of the population and improving gene flow between the subpopulations. Construction or restoration of wetlands in the landscape between the clusters has the potential to address each of these objectives.

Colorado↗

USGS2026

Colonization of southern flying squirrels (Glaucomys volans) to urban Nebraska: Range expansion or human assisted translocation?

Southern flying squirrels ( Glaucomys volans; Linnaeus 1758) were first observed in Lincoln, Nebraska, in 2018, 80 km north of their described range. Given that southern flying squirrels are a species of concern within Nebraska, determining the origin of this new population (natural expansion or pet-trade) garnered interest from state biologists. Further, the recent colonization of Lincoln by southern flying squirrels presents a unique opportunity to investigate the genetic implications of a founding event on a small arboreal mammal. The Lincoln population had genetic characteristics suggestive of a single-event colonization with fewer rare alleles and lower genetic diversity than potential source populations and a high genetic variation between populations. Sample size and absence of other geographically close populations in our data set make it difficult to ascertain the origin of the Lincoln population. Based on shared co-ancestry and membership assignment clustering algorithms, the Lincoln population had greater genetic associations with an individual sampled from the native south-eastern Nebraska population relative to other studied locations, suggesting that Lincoln was colonized by a native population.

Nebraska↗

USGS2019

Bayesian analysis of the impact of rainfall data product on simulated slope failure for North Carolina locations

In the past decades, many different approaches have been developed in the literature to quantify the load-carrying capacity and geotechnical stability (or the Factor of Safety, F_s) of variably saturated hillslopes. Much of this work has focused on a deterministic characterization of hillslope stability. Yet, simulated F_s values are subject to considerable uncertainty due to our inability to characterize accurately the soil mantle’s properties (hydraulic, geotechnical and geomorphologic) and spatiotemporal variability of the moisture content of the hillslope interior. This is particularly true at larger spatial scales. Thus, uncertainty-incorporating analyses of physically based models of rain-induced landslides are rare in the literature. Such landslide modeling is typically conducted at the hillslope scale using gauge-based rainfall forcing data with rather poor spatiotemporal coverage. For regional landslide modeling, the specific advantages and/or disadvantages of gauge-only, radar-merged and satellite-based rainfall products are not clearly established. Here, we compare and evaluate the performance of the Transient Rainfall Infiltration and Grid-based Regional Slope-stability analysis (TRIGRS) model for three different rainfall products using 112 observed landslides in the period between 2004 and 2011 from the North Carolina Geological Survey database. Our study includes the Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis Version 7 (TMPA V7), the North American Land Data Assimilation System Phase 2 (NLDAS-2) analysis, and the reference ‘truth’ Stage IV precipitation. TRIGRS model performance was rather inferior with the use of literature values of the geotechnical parameters and soil hydraulic properties from ROSETTA using soil textural and bulk density data from SSURGO (Soil Survey Geographic database). The performance of TRIGRS improved considerably after Bayesian estimation of the parameters with the DiffeRential Evolution Adaptive Metropolis (DREAM) algorithm using Stage IV precipitation data. Hereto, we use a likelihood function that combines binary slope failure information from landslide event and ‘null’ periods using multivariate frequency distribution-based metrics such as the False Discovery and False Omission Rates. Our results demonstrate that the Stage IV-inferred TRIGRS parameter distributions generalize well to TMPA and NLDAS-2 precipitation data, particularly at sites with considerably larger TMPA and NLDAS-2 rainfall amounts during landslide events than null periods. TRIGRS model performance is then rather similar for all three rainfall products. At higher elevations, however, the TMPA and NLDAS-2 precipitation volumes are insufficient and their performance with the Stage IV-derived parameter distributions indicate their inability to accurately characterize hillslope stability.

North Carolina↗

USGS2007

Application of two tests of multivariate discordancy to fisheries data sets

The generalized (Mahalanobis) distance and multivariate kurtosis are two powerful tests of multivariate discordancies (outliers). Unlike the generalized distance test, the multivariate kurtosis test has not been applied as a test of discordancy to fisheries data heretofore. We applied both tests, along with published algorithms for identifying suspected causal variable(s) of discordant observations, to two fisheries data sets from Lake Erie: total length, mass, and age from 1,234 burbot, Lota lota; and 22 combinations of unique subsets of 10 morphometrics taken from 119 yellow perch, Perca flavescens. For the burbot data set, the generalized distance test identified six discordant observations and the multivariate kurtosis test identified 24 discordant observations. In contrast with the multivariate tests, the univariate generalized distance test identified no discordancies when applied separately to each variable. Removing discordancies had a substantial effect on length-versus-mass regression equations. For 500-mm burbot, the percent difference in estimated mass after removing discordancies in our study was greater than the percent difference in masses estimated for burbot of the same length in lakes that differed substantially in productivity. The number of discordant yellow perch detected ranged from 0 to 2 with the multivariate generalized distance test and from 6 to 11 with the multivariate kurtosis test. With the kurtosis test, 108 yellow perch (90.7%) were identified as discordant in zero to two combinations, and five (4.2%) were identified as discordant in either all or 21 of the 22 combinations. The relationship among the variables included in each combination determined which variables were identified as causal. The generalized distance test identified between zero and six discordancies when applied separately to each variable. Removing the discordancies found in at least one-half of the combinations (k=5) had a marked effect on a principal components analysis. In particular, the percent of the total variation explained by second and third principal components, which explain shape, increased by 52 and 44% respectively when the discordancies were removed. Multivariate applications of the tests have numerous ecological advantages over univariate applications, including improved management of fish stocks and interpretation of multivariate morphometric data. ?? 2007 Springer Science+Business Media B.V.

Environmental Biology of Fishes↗

USGS2016

Bayesian analysis of Jolly-Seber type models

We propose the use of finite mixtures of continuous distributions in modelling the process by which new individuals, that arrive in groups, become part of a wildlife population. We demonstrate this approach using a data set of migrating semipalmated sandpipers ( Calidris pussila ) for which we extend existing stopover models to allow for individuals to have different behaviour in terms of their stopover duration at the site. We demonstrate the use of reversible jump MCMC methods to derive posterior distributions for the model parameters and the models, simultaneously. The algorithm moves between models with different numbers of arrival groups as well as between models with different numbers of behavioural groups. The approach is shown to provide new ecological insights about the stopover behaviour of semipalmated sandpipers but is generally applicable to any population in which animals arrive in groups and potentially exhibit heterogeneity in terms of one or more other processes.

Environmental and Ecological Statistics↗

USGS2017

Harmonization of forest disturbance datasets of the conterminous USA from 1986 to 2011

Several spatial forest disturbance datasets exist for the conterminous USA. The major problem with forest disturbance mapping is that variability between map products leads to uncertainty regarding the actual rate of disturbance. In this article, harmonized maps were produced from multiple data sources (i.e., Global Forest Change, LANDFIRE Vegetation Disturbance, National Land Cover Database, Vegetation Change Tracker, and Web-Enabled Landsat Data). The harmonization process involved fitting common class ontologies and determining spatial congruency to produce forest disturbance maps for four time intervals (1986–1992, 1992–2001, 2001–2006, and 2006–2011). Pixels mapped as disturbed for two or more datasets were labeled as disturbed in the harmonized maps. The primary advantage gained by harmonization was improvement in commission error rates relative to the individual disturbance products. Disturbance omission errors were high for both harmonized and individual forest disturbance maps due to underlying limitations in mapping subtle disturbances with Landsat classification algorithms. To enhance the value of the harmonized disturbance products, we used fire perimeter maps to add information on the cause of disturbance.

Environmental Monitoring and Assessment↗