Geology topics

David S. Pilliod

Publications and source records attributed to David S. Pilliod.

At least 19 recordsLinked to original sources

USGS2026

Accounting for temporal variation and correlation in environmental DNA sampling can improve ecological inferences

Environmental DNA (eDNA) concentration varies through space and time, and measurements collected close together are often correlated. Ignoring this dependence can inflate the rate of incorrect ecological inferences (Type I error rate). Although spatial correlation in eDNA has received considerable attention, temporal correlation has been less well studied. Statistical models and study designs that account for temporal correlation are increasingly important to understand time-dependent effects in complex systems. We developed a hierarchical model that separates temporal ecological variation from variability stemming from sampling and laboratory processes and applied it to four single-site eDNA time series collected over 17–24 days, three of which provided sufficient information for parameter estimation. We then used the empirically estimated parameter magnitudes in a simulation study to evaluate alternative temporal sampling designs that considered (1) how a fixed number of samples are allocated across different numbers of sampling times with different levels of temporal replication and (2) equally spaced versus cluster-spaced sampling (short bursts separated by longer gaps). Across the three time series sufficient for analysis, we observed substantial sampling variability, temporal variability, and temporal correlation, although correlation was estimated imprecisely (large coefficient of variation). Simulations showed that when sampling intervals were shorter than the effective temporal correlation range, models that ignored temporal dependence produced inflated Type I error rates and frequently detected spurious temporal trends. Accounting for temporal correlation substantially reduced this inflated Type I error rate. Optimal sampling strategies depended on study objectives. Clustered sampling most effectively estimated temporal correlation. When temporal dependence was negligible, evenly spaced sampling maximized power to detect trends. Estimating sampling variability required concentrating effort into fewer sampling times with more replicates per time, whereas estimating temporal variance was most precise with intermediate levels of replication. Together, these results indicate that temporal dependence can strongly affect inference from quantitative eDNA time series when sampling intervals approach the correlation timescale. Designs that ignore this dependence risk inferring ecological change or difference where none exists. Our framework provides practical guidance for allocating sampling effort in temporally intensive eDNA monitoring and for interpreting trends from short time series.

Idaho, Florida, Montana, Wisconsin↗

USGS2026

Lead exposure of snakes near recreational and military shooting areas in a sagebrush steppe ecosystem

Lead exposure has been documented in a multitude of bird and mammal species but less frequently in reptiles. Of the studies that have evaluated lead concentrations in reptiles, few have focused on snakes. We analyzed lead concentrations in liver and two types of bone from 24 gophersnakes ( Pituophis catenifer; Pc) and 12 Great Basin rattlesnakes ( Crotalus oreganus lutosus; Col) found dead along roads in the Morley Nelson Snake River Birds of Prey National Conservation Area in southwestern Idaho, U.S.A from 2018 to 2021. Because this area has been heavily impacted by both military and recreational activity, we also quantified lead concentrations in the soil near where the dead snakes were found. All soil samples had detectable lead (median: 12.53 µg/g; range: 0.07–16.77). Within individuals of both species, dry weight lead concentrations in liver generally were lower than those in bone (n samples < LOQ Liver: 9 of 13 ( Pc ), 9 of 10 ( Col ) (only a subset of individuals were in good enough condition to provide liver samples); Bone: 0 of 24 ( Pc ), 3 of 12 ( Col ). We did not detect any differences in lead concentrations between the species for either tissue type, nor between bone from the front (anterior) or back (posterior) of the snake skeleton. Additionally, we did not detect any correlation between lead concentration in soil collected at the roadkill sites and the tissue in the snakes. These data provide baseline information for lead exposure of two snake species in southwestern Idaho and demonstrate how lead is present in higher trophic levels within this ecosystem.

Idaho↗

USGS2026

Spatial variation in effectiveness and costs of sagebrush restoration treatments across the western United States

Natural resource managers strive to improve restoration efficiency across the vast network of public lands of the United States, seeking the best return on investment through innovation, geographic prioritization, and adaptive management. This challenge is exemplified by ecosystems dominated by sagebrush ( Artemisia spp.), which occur across the western United States and have been degraded by increasing wildfire frequency driven by invasive plant spread, among other factors. However, both costs and effectiveness of sagebrush restoration can vary spatially due to management practices in addition to biotic and abiotic factors, and characterizing this variation could inform broad-scale planning. We leveraged published models of sagebrush recovery and treatment costs to predict and evaluate cost-effectiveness of aerial and drill seeding Artemisia spp. across 429 718 km 2 that overlapped 12 US states. Compared to natural recovery, effectiveness for both methods was generally low, yet effectiveness was often greater, and costs were always greater, for drill than aerial seeding, resulting in slightly higher cost-effectiveness on average for drill seeding. Cost-effectiveness for both treatments increased near major roads and in areas with more repeated burns. Cost-effectiveness also increased with predisturbance cover of sagebrush but decreased with herbaceous cover. However, we also identified areas where aerial seeding was more cost-effective than drill seeding, particularly with greater slope. Our results provide spatially explicit estimates and potential mechanisms of cost-effectiveness for two common seeding methods for sagebrush, which can help prioritize limited resources, guide land use, and improve restoration efficiency and effectiveness across public rangelands of the western United States.

western United States↗

USGS2026

Detecting bumble bees in the wild using environmental DNA: Development and validation of a qPCR assay for the endangered Franklin’s bumble bee (Bombus franklini)

Environmental DNA (eDNA) sampling is a noninvasive alternative to conventional methods of surveying insects that may be particularly useful for detecting pollinators. We developed a quantitative polymerase chain reaction (qPCR) assay to detect the DNA of Franklin’s bumble bee ( Bombus franklini ) from flower samples and conducted an initial test of the assay using samples collected within and around the historical range of the species. We further analyzed all samples using metabarcoding. Our qPCR assay successfully amplified B. franklini DNA and exhibited no cross-reactivity with nontarget bumble bee DNA during in silico and in vitro testing. We did not detect B. franklini DNA from field-collected flower samples using either qPCR or metabarcoding. However, metabarcoding analysis revealed DNA of at least 16 other bumble bee species. This finding underscores the potential utility of eDNA sampling for surveying bumble bees. Nondetection of B. franklini from field-collected flower samples may be due to the extreme rarity of the species; B. franklini is endangered and has not been observed in the wild since 2006. Our B. franklini assay is among the first bee-specific qPCR assays ever developed and provides proof of concept for additional assays that may improve detection rates of rare and endangered bees.

California, Oregon↗

USGS2025

Supporting dryland restoration success with applied ecological forecasting of seeding outcomes

Introduction Ecological restoration is increasingly used to sustain biodiversity and ecosystem services. In drylands of the western United States (US), post-disturbance restoration often involves seeding treatments to promote the recovery of native plant communities. Spatial and temporal variability in environmental conditions influences plant establishment and contributes to low restoration success in certain locations and years. Objectives Here, we discuss how forecasts for plant establishment can be developed and delivered to help land managers anticipate the impacts of near-term (months to years) environmental conditions on restoration. Methods We developed an ecological forecast system that predicts the outcome of restoration seeding by integrating weather forecasts, an ecosystem water balance model, and plant establishment models. Results In this article, we focus on a conceptual approach to developing, delivering, and applying ecological forecasts for restoration. We illustrate the potential of this approach by adapting existing ecological models to build an initial version of a decision support tool that delivers a species-specific ecological forecast for big sagebrush ( Artemisia tridentata ) establishment. Integrating ecological forecasts into plans for restoration seeding presents opportunities to anticipate and account for environmental variability. Conclusions Finally, we discuss how connecting research, forecast delivery, and management can maximize the impact of ecological forecasts on restoration success.

Idaho, Nevada↗

USGS2025

Environmental drivers of Greater Sage-grouse population trends over 25 years in Idaho, USA

Greater Sage-grouse ( Centrocercus urophasianus ) populations have been in decline for decades across much of the US Intermountain West. However, findings from 25 years of lek counts in Idaho indicate that some populations are stable or even increasing. After accounting for potential biases in past lek count data, we sought to explain the variability in population trends among all 70 lek clusters (i.e., populations) we identified in the state. For each population, we identified lek count troughs, or low-point years, that occurred between the mid-1990s and 2021 and used a regression slope of those abundance low points to quantify each population's trend over the 25-year time span. We related the 70 populations' slopes to climate, fire, topographic, vegetation, and landcover variables. Our analyses revealed that populations with negative trends tend to occur toward the ends of climate gradients (i.e., extremes of occupied habitats) and in locations with more wildfire, agriculture, and riparian landcover. Populations with positive trends generally occur in landscapes toward the middle of the climate gradient, with high amounts of low sagebrush ( Artemisia arbuscula ) landcover and intermediate amounts of riparian and agricultural landcover. Post hoc analysis indicated that the latter two drivers were strongly associated with high raven occupancy rates, which may contribute to the negative sage-grouse population trends we observed in areas with high riparian or agricultural landcover. When modeled separately for different regions however, various region-specific drivers were identified, including tree cover, annual herbaceous cover, and human development. This information can help guide sage-grouse habitat management decisions and set expectations for population recovery, given the diversity of habitats occupied by the species and the cyclic nature of sage-grouse populations.

Idaho↗

USGS2025

Field test of the self-preserving eDNA filter and the importance of calibration when changing methods during long-term monitoring

Filtering water is currently the primary field method used for collecting aquatic environmental DNA (eDNA). One of the drawbacks of filtering is the need to transfer the filter from the filter housing to a preservative-filled container in the field. New products are being developed to avoid this handling step, but comparative studies are needed to ensure that the results produced by new protocols are transferable within and across eDNA monitoring programs. To meet this need, we evaluated two filter preservation methods (self-preserving filter housing vs. ethanol) of the 5.0-μm polyethersulfone (PES) filter membrane in a field trial typical of stream fisheries eDNA sampling. We compared DNA detection and yield for free-swimming rainbow trout, Oncorhynchus mykiss (Walbaum, 1792), from streams in Washington, United States, and British Columbia, Canada, while accounting for the effects of two environmental covariates: stream discharge and water temperature. As these streams were part of an ongoing fisheries eDNA monitoring program, we also compared these methods to the original protocol, which used a 0.45-μm cellulose nitrate (CN) filter membrane and ethanol preservative. We found that the self-preserving filter housings collected and preserved eDNA well and provided similar results to identical filters preserved in ethanol. The 5.0-μm PES filters preserved in ethanol significantly outperformed the original protocol in terms of both DNA detection and yield, highlighting the importance of calibration of eDNA results when changing sampling methods during an ongoing monitoring program.

british Columbia, Washington↗

USGS2025

A systematic review and meta-analysis of post-fire seeding and herbicide treatment effectiveness for controlling exotic annual grasses in the sagebrush biome

Introduction Outcomes of ecological restoration treatments can be highly variable and challenging to generalize, even for the same treatment type applied in similar ecological communities at different times and places. Notable examples are the herbicide and seeding treatments that have been extensively applied across the perennial sagebrush steppe of the United States to reduce impacts of fire-promoting exotic annual grasses (EAGs) such as Cheatgrass ( Bromus tectorum ). Objectives We asked if statistically based generalizations about the effects of pre-emergent herbicide and drill seeding of perennials, implemented following wildfires when threats of annual-grass invasion are greatest, could be made from the available literature. Methods We conducted a meta-analysis of these treatment effects for 1228 treatment-control comparisons from 35 studies published from 1990 to 2023 that met basic criteria for topical relevance and repeatability. Results EAGs and forbs were each reduced by herbicides and by seeding perennial grasses. The combination of herbicide and seeding reduced annuals and led to the largest increases in perennials. Although these outcomes support the intended effects of the treatments, there was high variability in outcomes among studies. Conclusions Combined use of pre-emergent herbicides and seeding can increase the success of restoration interventions that are aimed at reducing the invasion of exotic annual grasses and increasing perennials after fire in sagebrush steppe. Our analysis revealed, however, that the available literature was not suited to answering more specific questions, in spite of the massive amount of post-fire herbicide and seedings that have been applied in burned sagebrush steppe. Specifically, there were too few topically relevant studies with adequate scientific reporting to properly evaluate differences among specific treatment methodologies, including specific herbicides, that affect restoration success.

Restoration Ecology↗

USGS2025

Metabarcoding analysis of arthropod pollinator diversity: A methodological comparison of eDNA derived from flowers and DNA derived from bulk samples of insects

Limitations of traditional insect sampling methods have motivated the development and optimisation of new non-lethal methods capable of quantifying diverse arthropod communities. Environmental DNA (eDNA) metabarcoding using arthropod-specific primers has recently been investigated as a novel way to characterise arthropod communities from the DNA they deposit on the surface of plants. This sampling method has had demonstrated success, but pollinators—especially bees—are oddly underrepresented in these studies. To evaluate this inconsistency, we investigated the limitations of eDNA metabarcoding for bees and other pollinators. We compared pollinator diversity derived from eDNA extracted from flowers and DNA extracted from pulverised bulk samples of insects collected from vane traps deployed at the same sites using three metabarcoding primers, two of which target arthropods generally (COI-Jusino and 16S-Marquina) and one that targets bumblebees ( Bombus spp., COI-Milam). Across methods, we detected 77 insect families from 9 orders. The COI-Jusino marker amplified the highest taxonomic diversity compared to 16S-Marquina and COI-Milam. More amplicon sequence variants (ASVs) were recovered from vane traps (blue: 1357, yellow: 1542) than flowers (245), but only 23% of families and 13% of genera were shared among methods, indicating that flowers and blue and yellow vane traps may each sample different parts of the available arthropod community. Of 29 flower samples with known bee visitations, only 10 samples had bee detections from eDNA, and incomplete reference databases hindered assignment to species. Although our study provides additional evidence for the usefulness of eDNA metabarcoding for characterising arthropod communities, significant challenges remain when using eDNA metabarcoding methods to identify and quantify pollinator communities, especially bees.

Molecular Ecology↗

USGS2025

Costs of land treatments on public lands in the western United States

Public land managers often conduct rehabilitation and restoration actions to achieve desired conditions or specific natural resource objectives. These “land treatments” include a variety of techniques, such as biomass removal or manipulation, seeding, and herbicide application. Limited information exists on the costs of conducting many common types of land treatments, but such information can be paired with treatment effectiveness data to prioritize application of limited resources where they may have the greatest benefit and improve efficiency. Here, we investigated cost information recorded in the Land Treatment Digital Library, a catalog of legacy land treatment information on public lands managed by the U.S. Department of the Interior's Bureau of Land Management. Based on 1,701 treatment records across eleven western U.S. states, we developed empirical per-acre cost estimates for representative land treatments in eight categories: three seeding categories (aerial seeding, drill seeding, and seedling planting), prescribed burning, soil disturbance, soil stabilization, vegetation disturbance, and weed control. We evaluated spatio-temporal factors that may be associated with variation in treatment costs and found strong evidence for nonlinear decreases in per-acre costs as treatment areas increased and that per-acre treatment costs have increased in real terms in recent decades. We also found evidence that per-acre costs for drill seeding, prescribed burns, and soil stabilization increased with the average slope of the terrain of a treated area and that per-acre costs for prescribed burns, seedling planting, and soil stabilization were influenced by distance to urban areas or major roads. These results can inform planning, prioritization, and assessment of common land treatments on public lands in the western United States, in particular supporting greater consideration of costs and cost effectiveness.

Rangeland Ecology & Management↗

USGS2025

A 21st Century butterfly net: Using eDNA to detect the imperiled Dakota skipper

The development of environmental DNA (eDNA) methods for terrestrial arthropods could be transformative for the difficult task of assessing the status of species of conservation concern. The primary goal of this study was to investigate the efficacy of detecting the Dakota skipper ( Hesperia dacotae ) from its DNA left behind on inflorescences as a means of inferring species presence. We developed and tested a novel qPCR assay and validated the assay in both controlled and field contexts. Using captive animals at the Minnesota Zoo, we found that the number of skippers in an enclosure increased the probability of skipper DNA detection. In the field, Dakota skipper DNA was found on 14% (11 of 81) of inflorescences collected. All detections were from narrowleaf purple coneflower ( Echinacea angustifolia ). Known visitation of an inflorescence by Dakota skipper prior to sample collection was not a strong predictor of either skipper DNA presence or amount of DNA, but skipper eDNA was detected at 60% (3 of 5) of sites where skippers were observed and 33% (1 of 3) of sites where skippers were not observed. These findings demonstrate successful application of a targeted-species approach to eDNA sampling for butterflies in the field. Taken together, our findings indicate that this method could provide a novel and useful source of data for assessing occupancy trends of butterflies without capturing or even observing them in the wild.

Minnesota, North Dakota, South Dakota↗

USGS2025

Evaluation of the gap intercept method to measure rangeland connectivity

Characterizing the connectivity of materials, organisms, and energy on rangelands is critical to understanding and managing ecosystem response to disturbances. For over twenty years, scientists and rangeland managers have used the gap intercept method to monitor connectivity. However, using gap intercept measurements to infer ecosystem processes or inform management actions and conservation practices on rangelands has been limited because there are few tools and syntheses to help managers pragmatically interpret gap measurements. In this synthesis, we review the different ways the gap intercept method is conducted and highlight considerations when collecting and using gap intercept data. We discuss what is known about how gap intercept data can be used to assess wind and water erosion, biocrust abundance and composition, soil fertility, plant community dynamics, wildlife habitat characteristics, and fuel connectivity and fire behavior. Finally, we identify emerging opportunities regarding the collection and use of gap intercept data to assess rangeland connectivity, ecosystem function, and ecological processes. Through this synthesis we demonstrate the value of the gap intercept method, particularly “all-plant” (i.e., whole-community) canopy gap, in supporting assessments of rangeland condition and vulnerability, and in planning future management actions.

Rangeland Ecology & Management↗

USGS2025

Wildfire and postfire restoration treatments have lasting effects on rodent habitat and community composition

Rangeland fires can quickly change the structure of wildlife habitat and cause changes that persist for years to decades. To facilitate habitat recovery, postfire restoration actions often involve sowing seeds of native and nonnative perennial grasses and shrubs. Empirical information on whether such restoration activities are effective and how wildlife will respond is unknown. We evaluated the effects of wildfire and postfire seeding on rodent communities in sagebrush steppe by measuring environmental characteristics and live-trapping rodents at three wildfire locations in the northern Great Basin that burned 2–14 yr prior. Sampled plots were either 1) burned (control), 2) burned and seeded with nonnative species, 3) burned and seeded with native species, or 4) not burned (reference). Unburned reference plots had 95.7% more shrub cover and less cover of nonnative annual grasses than all burned plots. Burned plots seeded with native species had greater cover of native perennial grasses and less bare ground compared with burned plots seeded with nonnative species or control plots. Wildfire and postfire seeding significantly changed rodent community composition but not species richness. American deermouse ( Peromyscus maniculatus ) and least chipmunk ( Tamias minimus ) were less likely to occur in burned plots than in unburned reference plots. Ord’s kangaroo rat ( Dipodomys ordii ) and Piute ground squirrel ( Urocitellus mollis ) were more likely to occur in burned plots than in unburned reference plots. Sagebrush vole ( Lemmiscus curtatus ) was more likely to occur in burned plots with native seeding than with all other treatments. Species’ responses to wildfire and seeding reflected their relationships to cover of shrubs, perennial grasses, and nonnative annual grasses. These results suggest that wildfire and postfire seeding have long-lasting effects on wildlife communities. Landscape-level rodent diversity may be enhanced when small-scale disturbances or restoration actions increase landscape heterogeneity.

northern Great Basin↗

USGS2025

Amphibian diversity of the western Colorado canyonlands including potential threats from nonnative bullfrogs and disease

Throughout the canyons of the Colorado and Uncompahgre Plateaus, water is a limited resource for wildlife, with patchy distribution and seasonal availability. Tributary creeks within these canyons drain into mainstem rivers, providing habitat and breeding sites for native amphibians. Yet, little is known about the diversity and distribution of amphibians that live in these harsh, dynamic environments. In addition, the rivers that border these canyon tributaries may serve as corridors for nonnative species and disease. The American Bullfrog ( Lithobates catesbeianus ) is a nonnative species in western Colorado known to prey on native amphibians and act as a reservoir for pathogens such as Batrachochytrium dendrobatidis ( Bd ). From 2019 to 2022, we surveyed for amphibians using visual encounter surveys (VES) and environmental DNA (eDNA) surveys throughout the McInnis Canyons National Conservation Area (MCNCA), the Dominguez–Escalante National Conservation Area (DENCA), and the Dolores River Canyon Wilderness Study Area (DRCWSA). Our primary goals were to document the diversity and distribution of native amphibians in the canyonlands and evaluate potential threats to these species from bullfrogs and Bd . We confirmed that sensitive species, such as the Great Basin Spadefoot ( Spea intermontana ) and the Northern Leopard Frog ( Lithobates pipiens ), inhabit these protected areas. In most cases, bullfrogs were not detected within ephemeral tributaries, but bullfrog DNA was detected in some tributaries at the confluence with the mainstem rivers. In Mee Canyon (MCNCA), however, bullfrogs were found within the tributary, up to 3 km from the Colorado River. A bullfrog individual removed from this canyon tested positive for Bd , and diet contents suggested that native amphibians are potential prey in this system. Nonnative predators and disease pose a threat to native amphibians, alongside environmental changes such as drought and hydrological shifts driven by ongoing climate change.

Colorado↗

USGS2024

Field trials of an autonomous eDNA sampler in lotic waters

Environmental DNA (eDNA) analysis has become a transformative technology, but sample collection methods lack standardization and sampling at effective frequencies requires considerable field effort. Autonomous eDNA samplers that can sample water at high frequencies offer potential solutions to these problems. We present results from four case studies using a prototype autonomous eDNA sampler as part of the U.S. Geological Survey’s Rapid Environmental eDNA Assessment and Deployment Initiative & Network (READI-Net) project. These case studies involved short-term deployments of an eDNA autosampler (Smith-Root) across a range of riverine habitats with the objectives of (a) identifying what insights could be gained from high-frequency autosampling and (b) benchmarking these autosamples against manually collected samples. The high frequency autosampling revealed high temporal variability of eDNA concentrations and provided valuable insights about eDNA associations with environmental covariates, such as discharge and turbidity. Benchmarking assessments indicated autosamples had similar detection rates to manual samples and obtained similar or greater eDNA quantities. We did find minimal carryover contamination in autosampler field controls. We conclude that eDNA autosamplers have potential to improve freshwater biosurveillance by reducing logistical sampling barriers, standardizing collection methods, and clarifying the influence of environmental covariates on eDNA results.

Idaho, Missouri, Montana, New York↗

USGS2024

Leveraging extensive soil, vegetation, fire, and land treatment data to inform restoration across the sagebrush biome

Context Widespread ecological degradation has prompted calls for massive global investments in ecological restoration, yet limited resources necessitate efficient application of restoration efforts. In western North America, altered fire regimes are increasing the scale of restoration needed to preserve the sagebrush ( Artemisia species) biome but prioritizing and implementing effective restoration is complicated by the vast and heterogeneous sagebrush landscape, which includes gradients in climate, disturbance, and species composition. Objectives To develop spatially explicit and context-dependent estimates of treatment efficacy and sagebrush recovery rates. Methods We leveraged a suite of spatio-temporally extensive datasets to evaluate the influence of restoration treatments and environmental conditions on trends in post-disturbance sagebrush cover, with an emphasis on understanding differences between sites recovering naturally and sites receiving restoration treatments. We used estimates from these models to develop spatially explicit projections for sagebrush recovery, conditional on disturbance, restoration practice, and environmental conditions. Results We found seeding Artemisia spp. increased sagebrush cover over time relative to natural recovery, but this relationship depended on spring soil moisture availability and treatment methods. Natural recovery was positively influenced by soil moisture and sagebrush cover and negatively influenced by cumulative burns and annual herbaceous cover, while the influence of perennial herbaceous cover varied with soil moisture. Conclusions Our results provide biome-wide insights and spatially explicit tools that can inform economic cost-effectiveness analyses, restoration prioritization tools, and other scientific endeavors to ensure managers have the tools and information needed to effectively steward the sagebrush biome in a rapidly changing world.

Landscape Ecology↗

USGS2024

Climate futures for lizards and snakes in western North America may result in new species management issues

We assessed changes in fundamental climate-niche space for lizard and snake species in western North America under modeled climate scenarios to inform natural resource managers of possible shifts in species distributions. We generated eight distribution models for each of 130 snake and lizard species in western North America under six time-by-climate scenarios. We combined the highest-performing models per species into a single ensemble model for each scenario. Maps were generated from the ensemble models to depict climate-niche space for each species and scenario. Patterns of species richness based on climate suitability and niche shifts were calculated from the projections at the scale of the entire study area and individual states and provinces, from Canada to Mexico. Squamate species' climate-niche space for the recent-time climate scenario and published known ranges were highly correlated ( r = 0.81). Overall, reptile climate-niche space was projected to move northward in the future. Sixty-eight percent of species were projected to expand their current climate-niche space rather than to shift, contract, or remain stable. Only 8.5% of species were projected to lose climate-niche space in the future, and these species primarily occurred in Mexico and the southwestern U.S. We found few species were projected to lose all suitable climate-niche space at the state or province level, although species were often predicted to occupy novel areas, such as at higher elevations. Most squamate species were projected to increase their climate-niche space in future climate scenarios. As climate niches move northward, species are predicted to cross administrative borders, resulting in novel conservation issues for local landowners and natural resource agencies. However, information on species dispersal abilities, landscape connectivity, biophysical tolerances, and habitat suitability is needed to contextualize predictions relative to realized future niche expansions.

Ecology and Evolution↗

USGS2024

Propagating observation errors to enable scalable and rigorous enumeration of plant population abundance with aerial imagery

Estimating and monitoring plant population size is fundamental for ecological research, as well as conservation and restoration programs. High-resolution imagery has potential to facilitate such estimation and monitoring. However, remotely sensed estimates typically have higher uncertainty than field measurements, risking biased inference on population status. We present a model that accounts for false negative (missed plants) and false positive (misclassified or double-counted plants) error in counts from high-resolution imagery via integration with ground data. We apply it to estimate the abundance of a foundational shrub species in post-wildfire landscapes in the western United States. In these landscapes, plant recruitment is crucial for ecological recovery but locally patchy, motivating the use of spatially extensive measurements from unoccupied aerial systems (UAS). Integrating >16 ha of UAS imagery with >700 georeferenced field plots, we fit our model to generate insights into the prevalence and drivers of observation errors associated with classification algorithms used to distinguish individual plants, relationships between abundance and landscape context, and to generate spatially explicit maps of shrub abundance. Raw counts of plant abundance in high-resolution imagery resulted in substantial false negative and false positive observation errors. The probability of detecting ( p ) adult plants (≥0.25 m tall) varied between sites within 0.52 < p̂adult < 0.82, whereas the detection of smaller plants (<0.25 m) was lower, 0.03 < p̂small < 0.3. On average, we estimate that 19% of all detected plants were false positive errors, which varied spatially in relation to topographic predictors. Abundance declined toward the interior of previous wildfires and was positively associated with terrain roughness. Our study demonstrates that integrated models accounting for imperfect detection improve estimates of plant population abundance derived from inherently imperfect UAS imagery. We believe such models will further improve inference on plant population dynamics—relevant to restoration, wildlife habitat and related objectives—and echo previous calls for remote sensing applications to better differentiate between ecological and observational processes.

Methods in Ecology and Evolution↗