Geology topics

Brian J. Reich

Publications and source records attributed to Brian J. Reich.

11 recordsLinked to original sources

USGS2024

Applying mark-resight, count, and telemetry data to estimate effective sampling area and fish density with stationary underwater cameras

Accurate estimates of abundance and density for geographically open populations must account for the effective sampling area (ESA) of sampling gears. We describe a Marked N-Mixture model to estimate ESA and density (number of individuals/unit area) from repeated counts of unmarked and marked individuals, integrating mark-resight, camera counts, and telemetry data of red snapper ( Lutjanus campechanus ) at a 1.6 km 2 reef off North Carolina, USA. Cameras recorded observations of unmarked and marked individuals, whereas telemetry data indicated the number of tagged fish present on the reef. We estimated density (95 individuals/km 2 , 95%CI: 58–149), ESA (which was lower when current direction was towards the camera), detection probability (0.06, 95%CI: 0.03–0.09), and covariate relationships. Simulation studies under different scenarios of data quality and space use identified positive bias in density estimates from N-mixture models due to fish movement. In contrast, the Marked N-Mixture model returned unbiased estimates of density, ESA, and detection parameters, and appears to be a more robust method for modeling density given the data available for this analysis. This approach can be applied to other populations where count and telemetry data overlap in space and time.

North Carolina↗

USGS2024

A spatiotemporal optimization engine for prescribed burning in the Southeast US

Many ecosystems in the Southeast US are dependent upon frequent low-intensity surface fires to sustain native biodiversity, ecosystem services, and endangered species populations. Today, landscape-scale prescribed fire is required to manage these systems for conservation objectives and to mitigate wildland fire risk. Successful application of prescribed fire in this region requires careful planning and assessment of the risks and tradeoffs involved when deciding whether or not to conduct a burn. Many of these risks are closely tied to ambient environmental conditions and are reflected in sets of ‘prescription’ parameters that define safe and effective operating conditions to meet objectives or regulatory requirements. To facilitate effective decision making and acknowledge growing uncertainties related to climate change effects on wildland fire operations, we developed a spatiotemporal optimization engine to identify near-term optimal burning opportunities for prescribed fire implementation. By mining historical 3-day numerical weather forecasts and observation-based weather data for 2015–2021, we have developed a Bayesian hierarchical model for forecast verification that provides calibrated daily weather forecasts and joint uncertainty estimates on meteorological variables of interest, with the latter serving as a measure of risk associated with prescribed fire activities. Burn allocation decisions are then optimized by considering this risk jointly with the utility of burning a particular habitat parcel. The initial iteration of the optimization engine is demonstrated through a case study of short-term meteorological conditions for the Eglin Air Force Base, located in Florida, USA. Results indicate agreement between the optimization engine and the observed past decision-making, with the largest divergences likely arising primarily from differences between utility functions presumed important and used to develop the optimization engine versus the true utility functions driving management behavior in practice.

Florida↗

USGS2023

Beyond simple trend tests: Detecting significant changes in design-flood quantiles

Changes in annual maximum flood (AMF), which are usually detected using simple trend tests (e.g., Mann-Kendall test (MKT)), are expected to change design-flood estimates. We propose an alternate framework to detect significant changes in design-flood between two periods and evaluate it for synthetically generated AMF from the Log-Pearson Type-3 (LP3) distribution due to changes in moments associated with flood distribution. Synthetic experiments show MKT does not consider changes in all three moments of the LP3 distribution and incorrectly detects changes in design-flood. We applied the framework on 31 river basins spread across the United States. Statistically significant changes in design-flood quantiles were observed even without a significant trend in AMF and basins with statistically significant trend did not necessarily exhibit statistically significant changes in design-flood. We recommend application of the framework for evaluating changes in design-flood estimates considering changes in all the moments as opposed to simple trend tests.

Geophysical Research Letters↗

USGS2018

Optimal treatment allocations in space and time for online control of anemerging infectious disease

A key component in controlling the spread of an epidemic is deciding where, when and to whom to apply an intervention. We develop a framework for using data to inform these decisions in realtime. We formalize a treatment allocation strategy as a sequence of functions, one per treatment period, that map up‐to‐date information on the spread of an infectious disease to a subset of locations where treatment should be allocated. An optimal allocation strategy optimizes some cumulative outcome, e.g. the number of uninfected locations, the geographic footprint of the disease or the cost of the epidemic. Estimation of an optimal allocation strategy for an emerging infectious disease is challenging because spatial proximity induces interference between locations, the number of possible allocations is exponential in the number of locations, and because disease dynamics and intervention effectiveness are unknown at outbreak. We derive a Bayesian on‐line estimator of the optimal allocation strategy that combines simulation–optimization with Thompson sampling. The estimator proposed performs favourably in simulation experiments. This work is motivated by and illustrated using data on the spread of white nose syndrome, which is a highly fatal infectious disease devastating bat populations in North America.

Journal of the Royal Statistical Society. Series C↗

USGS2018

Avian response to shade‐layer restoration in coffee plantations in Puerto Rico

Documenting the evolving processes associated with habitat restoration and how long it takes to detect avian demographic responses is crucial to evaluate the success of restoration initiatives and to identify ways to improve their effectiveness. The importance of this endeavor prompted the U.S. Fish and Wildlife Service and the USDA Natural Resources Conservation Service to evaluate their sun‐to‐shade coffee restoration program in Puerto Rico initiated in 2003. We quantified the responses of 12 resident avian species using estimates of local occupancy and extinction probabilities based on surveys conducted in 2015–2017 at 65 restored farms grouped according to time‐since‐initial‐restoration (TSIR): new (2011–2014), intermediate (2007–2010), and old (2003–2006). We also surveyed 40 forest sites, which served as reference sites. Vegetation complexity increased with TSIR, ranging between 35 and 40% forest cover in farms 6–9 years TSIR. Forest specialists (e.g. Loxigilla portoricencis ) exhibited highest average occupancy in farms initially classified as intermediate (6–9 years) and old (>10 years), paralleling occupancy in secondary forests. Occupancy of open‐habitat specialists (e.g. Tiaris olivaceus ) was more variable, but higher in recently restored farms. Restoring the shade layer has the potential to heighten ecological services derived from forest specialists (e.g. frugivores) without losing the services of many open‐habitat specialists (e.g. insectivores). Annual local extinction probability for forest specialists decreased with increasing habitat complexity, strengthening the potential value of shade restoration as a tool to enhance habitat for avifauna that evolved in forested landscapes.

Restoration Ecology↗

USGS2017

Integrating multiple data sources in species distribution modeling: A framework for data fusion

The last decade has seen a dramatic increase in the use of species distribution models (SDMs) to characterize patterns of species’ occurrence and abundance. Efforts to parameterize SDMs often create a tension between the quality and quantity of data available to fit models. Estimation methods that integrate both standardized and non-standardized data types offer a potential solution to the tradeoff between data quality and quantity. Recently several authors have developed approaches for jointly modeling two sources of data (one of high quality and one of lesser quality). We extend their work by allowing for explicit spatial autocorrelation in occurrence and detection error using a Multivariate Conditional Autoregressive (MVCAR) model and develop three models that share information in a less direct manner resulting in more robust performance when the auxiliary data is of lesser quality. We describe these three new approaches (“Shared,” “Correlation,” “Covariates”) for combining data sources and show their use in a case study of the Brown-headed Nuthatch in the Southeastern U.S. and through simulations. All three of the approaches which used the second data source improved out-of-sample predictions relative to a single data source (“Single”). When information in the second data source is of high quality, the Shared model performs the best, but the Correlation and Covariates model also perform well. When the information quality in the second data source is of lesser quality, the Correlation and Covariates model performed better suggesting they are robust alternatives when little is known about auxiliary data collected opportunistically or through citizen scientists. Methods that allow for both data types to be used will maximize the useful information available for estimating species distributions.

Ecology↗

USGS2017

Characterizing sources of uncertainty from global climate models and downscaling techniques

In recent years climate model experiments have been increasingly oriented towards providing information that can support local and regional adaptation to the expected impacts of anthropogenic climate change. This shift has magnified the importance of downscaling as a means to translate coarse-scale global climate model (GCM) output to a finer scale that more closely matches the scale of interest. Applying this technique, however, introduces a new source of uncertainty into any resulting climate model ensemble. Here we present a method, based on a previously established variance decomposition method, to partition and quantify the uncertainty in climate model ensembles that is attributable to downscaling. We apply the method to the Southeast U.S. using five downscaled datasets that represent both statistical and dynamical downscaling techniques. The combined ensemble is highly fragmented, in that only a small portion of the complete set of downscaled GCMs and emission scenarios are typically available. The results indicate that the uncertainty attributable to downscaling approaches ~20% for large areas of the Southeast U.S. for precipitation and ~30% for extreme heat days (> 35°C) in the Appalachian Mountains. However, attributable quantities are significantly lower for time periods when the full ensemble is considered but only a sub-sample of all models are available, suggesting that overconfidence could be a serious problem in studies that employ a single set of downscaled GCMs. We conclude with recommendations to advance the design of climate model experiments so that the uncertainty that accrues when downscaling is employed is more fully and systematically considered.

Journal of Applied Meteorology and Climatology↗

USGS2017

Occupancy and abundance of Eleutherodactylus frogs in coffee plantations in Puerto Rico

Shaded coffee plantations are of conservation value for many taxa, particularly for resident avifauna in the face of extensive landscape changes. Yet, little is known about the value of coffee plantations for amphibians because there are scant demographic data to index their value among species with different habitat preferences. We estimated the probability of occupancy of three frog species: Eleutherodactylus wightmanae, a forest species; E. brittoni, a grassland species; and E. antillensis, an open habitat species. Occupancy was estimated in sun and shaded plantations, and in secondary forest, in the west-central mountains of Puerto Rico. We also estimated the probability that a survey station was occupied by no individuals, one, or >1 individual, as a proxy of abundance. The aforementioned parameters, and local colonization and extinction probability, were modeled as a function of weather conditions (temperature, humidity) and vegetation cover at the sampling station (5 m) and contextual (100 m) scales. Encounter histories were obtained with passive acoustic recorders between February and July in 2015. Consistent with known habitat preferences, the highest occupancies were associated with secondary forests for E. wightmanae and sun plantations for E. brittoni . Occupancy probability for E. antillensis was similar across habitat types, indicating no aversion to shaded–forested habitats. Shaded plantations harbored moderate levels of occupancy for all species, indicating their potential value for multispecies conservation. Local colonization rates increased with forest cover for E. wightmanae, and with open habitats for E. brittoni and E. antillensis . Open habitats harbored a higher abundance of E. brittoni and E antillensis, but lower values for E. wightmanae . Sun and shaded plantations could provide quality habitat for Eleutherodactylus spp. if managed for features that promote local colonization and abundance.

Puerto Rico↗

USGS2017

Occupancy and abundance of Eleutherodactylus wightmanae and E. brittoni along elevational gradients in west-central Puerto Rico

Populations of Eleutherodactylus species in Puerto Rico have declined in recent decades due to habitat loss and long-term climatic changes. The conservation of these habitat specialists requires an understanding of factors influencing their abundance and distribution, which at present is scant. We estimated occupancy probability and the probability of encountering ≥2 individuals of E. wightmanae (Melodius Coqui or Wightman's Robber Frog) and E. brittoni (Grass Coqui), species with contrasting habitat affinities, using multi-season, multi-state occupancy models. These parameters also served as an index of abundance (non-presence, 1, and ≥2 individuals). We modeled parameters as a function of seasonal temperature and humidity, long-term average monthly precipitation, and habitat covariates measured at survey sites along 2 elevation gradients in the southern slopes of west-central Puerto Rico. We collected survey data using passive acoustic recorders during 3 seasonal periods between February and July 2015. Occupancy patterns of both species was unimodal, containing higher probabilities (e.g., ≥0.5) at elevations between 400 m and 700 m, where long-term monthly precipitation varied between 120 mm and 160 mm. Chances of encountering ≥2 individuals increased with ground cover for E. brittoni , and decreased with increasing canopy cover for E. wightmanae . Seasonal temperature and relative humidity did not influence occupancy or the probability of encountering ≥2 individuals, likely because covariates varied within known tolerance levels for Eleutherodactylus . Our findings help reduce local extinction probability through management of habitat conditions that increase the likelihood of encountering ≥2 individuals. We also detailed an analytical framework suitable to test hypotheses aimed at predicting potential impacts from land use and climatic changes, and species responses to conservation actions.

Puerto Rico↗

USGS2016

Uncertainty quantification and propagation for projections of extremes in monthly area burned under climate change: A case study in the coastal plain of Georgia, USA

Human‐caused climate change is predicted to affect the frequency of hazard‐linked extremes. Unusually large wildfires are a type of extreme event that is constrained by climate and can be a hazard to society but also an important ecological disturbance. This chapter focuses on changes in the frequency of extreme monthly area burned by wildfires for the end of the 21st century for a wildfire‐prone region in the southeast United States. Predicting changes in area burned is complicated by the large and varied uncertainties in how the climate will change and in the models used to predict those changes. The chapter characterizes and quantifies multiple sources of uncertainty and propagate the expanded prediction intervals of future area burned. It illustrates that while accounting for multiple sources of uncertainty in global change science problems is a difficult task, it will be necessary in order to properly assess the risk of increased exposure to these society‐relevant events.

Georgia↗

USGS2015

Occupancy estimation for rare species using a spatially-adaptive sampling design

Summary 1. Spatially clustered populations create unique challenges for conservation monitoring programmes. Advances in methodology typically are focused on either the design or the modelling stage of the study but do not involve integration of both. 2. We integrate adaptive cluster sampling and spatial occupancy modelling by developing two models to handle the dependence induced by cluster sampling. We compare these models to scenarios using simple random sampling and traditional occupancy models via simulation and data collected on a rare plant species, Tamarix ramosissima , found in China. 3. Our simulations show a marked improvement in confidence interval coverage for the new models combined with cluster sampling compared to simple random sampling and traditional occupancy models, with greatest improvement in the presence of low detection probability and spatial correlation in occupancy. 4. Accounting for the design using the simple cluster random-effects model reduces bias considerably, and full spatial modelling reduces bias further, especially for large n when the spatial covariance parameters can be estimated reliably. Both new models build on the strength of occupancy modelling and adaptive sampling and perform at least as well, and often better, than occupancy modelling alone. 5. We believe our approach is unique and potentially useful for a variety of studies directed at patchily distributed, clustered or rare species exhibiting spatial variation.

Methods in Ecology and Evolution↗