USGS Science⌕ Search

SEARCH · USGS Science

Results for “Ecology and Evolution”

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 325 records · Page 18Linked to original sources
USGS2020

Response to terrestrial nest predators among endemic and introduced Hawaiian birds

Birds free from nest predators for long periods may either lose the ability to recognize and respond to predators or retain antipredator responses if they are not too costly. How these alternate scenarios play out has rarely been investigated in an avian community whose members have different evolutionary histories. We presented models of two nest predators (rat and snake) and a negative control (tree branch) to birds on Hawaiʻi Island. Endemic Hawaiian birds evolved in the absence of terrestrial predators until rats were introduced approximately 1,000 years ago. Introduced birds evolved with diverse predator communities including mammals and snakes, but since their introduction onto the island approximately one century ago have been free from snake predation. We found that (a) endemic and introduced birds had higher agitation scores toward the rat model compared with the branch, and (b) none of the endemic birds reacted to the snake model, while one introduced bird, the Red-billed Leiothrix (Leiothrix lutea), reacted as strongly to the snake as to the rat. Overall, endemic and introduced birds differ in their response to predators, but some endemic birds have the capacity to recognize and respond to introduced rats, and one introduced bird species retained recognition of snake predators from which they had been free for nearly a century, while another apparently lost that ability. Our results indicate that the retention or loss of predator recognition by introduced and endemic island birds is variable, shaped by each species' unique history, ecology, and the potential interplay of genetic drift, and that endemic Hawaiian birds could be especially vulnerable to introduced snake predators.

Hawaii↗
USGS2020

Are migratory waterfowl vectors of seagrass pathogens?

Migratory waterfowl vector plant seeds and other tissues, but little attention has focused on the potential of avian vectoring of plant pathogens. Extensive meadows of eelgrass (Zostera marina) in southwest Alaska support hundreds of thousands of waterfowl during fall migration and may be susceptible to plant pathogens. We recovered DNA of organisms pathogenic to eelgrass from environmental samples and in the cloacal contents of eight of nine waterfowl species that annually migrate along the Pacific coast of North America and Asia. Coupled with a signal of asymmetrical gene flow of eelgrass running counter to that expected from oceanic and coastal currents between Large Marine Ecosystems, this evidence suggests waterfowl are vectors of eelgrass pathogens.

Alaska↗
USGS2020

Fatty acid-based diet estimates suggest ringed seal remain the main prey of southern Beaufort Sea polar bears despite recent use of onshore food resources

Polar bears ( Ursus maritimus ) from the southern Beaufort Sea (SB) subpopulation have traditionally fed predominantly upon ice‐seals; however, as the proportion of the subpopulation using onshore habitat has recently increased, foraging on land‐based resources, including remains of subsistence‐harvested bowhead whales ( Balaena mysticetus ) and colonial nesting seabirds has been observed. Adipose tissue samples were collected from this subpopulation during the springs of 2013–2016 and analyzed for fatty acid signatures. Diet estimates were generated for the proportional consumption of ringed seal ( Pusa hispida ), bearded seal ( Erignathus barbatus ), and beluga whale ( Delphinapterus leucas ), relative to onshore foods, including bowhead whale remains and seabird, as represented by black guillemot ( Cepphus grylle mandtii ) nestlings and eggs. Quantitative fatty acid signature analysis (QFASA) estimated that the ice‐obligate prey, ringed seal, remained the predominant prey species of SB polar bears (46.4 ± 1.8%), with much lower consumption of bearded seal (19.6 ± 2.0%), seabird (17.0 ± 1.2%), bowhead whale (15.0 ± 1.4%), and hardly any beluga whale (2.0 ± 0.5%). Adult and subadult females appeared to depend more on the traditional ringed seal prey than adult and subadult males. Diet estimates of SB polar bears showed significant interannual variability for all prey ( F 12, 456 = 3.17, p < .001). Longer‐term estimates suggested that both types of onshore prey, bowhead whale remains and seabird, have represented a moderate proportion of the food resources used by SB polar bears since at least the start of the 21st Century.

Alaska↗
USGS2020

Timing, frequency, and duration of incubation recesses in dabbling ducks

Nest attendance is an important determinant of avian reproductive success, and identifying factors that influence the frequency and duration of incubation recesses furthers our understanding of how incubating birds balance their needs with those of their offspring. We characterized the frequency and timing (start time, end time, and duration) of incubation recesses for mallard (Anas platyrhynchos) and gadwall (Mareca strepera) hens breeding in Suisun Marsh, California, USA, and examined the influences of day of year, ambient temperature at the nest, incubation day, and clutch size on recess frequency and timing using linear mixed models. Mallard, on average, took more recesses per day (1.69 ± 0.80, mean ± standard deviation) than did gadwall (1.39 ± 0.69), and 45% of mallard nest-days were characterized by two recesses, while only 27% of gadwall nest-days were characterized by two recesses. Mallard morning recesses started at 06:14 ± 02:46, and lasted 106.11 ± 2.01 minutes, whereas mallard afternoon recesses started at 16:39 ± 02:11 and lasted 155.39 ± 1.99 minutes. Gadwall morning recesses started at 06:30 ± 02:46 and lasted 91.28 ± 2.32 minutes, and gadwall afternoon recesses started at 16:31 ± 01:57 and lasted 192.69 ± 1.89 minutes. Mallard and gadwall started recesses earlier in the day with increasing ambient temperature, but later in the day as the season progressed. Recess duration decreased as the season progressed and as clutch size increased, and increased with ambient temperature at the nest. The impending darkness of sunset appeared to be a strong cue for ending a recess and returning to the nest, because hens returned to their nests earlier than expected when recesses were expected to end after sunset. Within hens, the timing of incubation recesses was repeatable across incubation days, and was most repeatable for mallard afternoon recesses and on days in which hens took only one recess. Hens were most likely to be away from nests between 04:00 and 07:00 and between 16:00 and 19:00, therefore, investigators should search for nests between 07:00 and 16:00. Our analyses identified important factors influencing incubation recess timing in dabbling ducks, and have important implications for nest monitoring programs.

California↗
USGS2020

Paleo-metagenomics of North American fossil packrat middens: Past biodiversity revealed by ancient DNA

Fossil rodent middens are powerful tools in paleoecology. In arid parts of western North America, packrat ( Neotoma spp.) middens preserve plant and animal remains for tens of thousands of years. Midden contents are so well preserved that fragments of endogenous ancient DNA (aDNA) can be extracted and analyzed across millennia. Here, we explore the use of shotgun metagenomics to study the aDNA obtained from packrat middens up to 32,000 C 14 years old. Eleven Illumina HiSeq 2500 libraries were successfully sequenced, and between 0.11% and 6.7% of reads were classified using Centrifuge against the NCBI “nt” database. Eukaryotic taxa identified belonged primarily to vascular plants with smaller proportions mapping to ascomycete fungi, arthropods, chordates, and nematodes. Plant taxonomic diversity in the middens is shown to change through time and tracks changes in assemblages determined by morphological examination of the plant remains. Amplicon sequencing of ITS2 and rbcL provided minimal data for some middens, but failed at amplifying the highly fragmented DNA present in others. With repeated sampling and deep sequencing, analysis of packrat midden aDNA from well-preserved midden material can provide highly detailed characterizations of past communities of plants, animals, bacteria, and fungi present as trace DNA fossils. The prospects for gaining more paleoecological insights from aDNA for rodent middens will continue to improve with optimization of laboratory methods, decreasing sequencing costs, and increasing computational power.

Ecology and Evolution↗
USGS2020

Removal of chronic Mycoplasma ovipneumoniae carrier ewes eliminates pneumonia in a bighorn sheep population

Chronic pathogen carriage is one mechanism that allows diseases to persist in populations. We hypothesized that persistent or recurrent pneumonia in bighorn sheep ( Ovis canadensis ) populations may be caused by chronic carriers of Mycoplasma ovipneumoniae ( Mo ). Our experimental approach allowed us to address a conservation need while investigating the role of chronic carriage in disease persistence. We tested our hypothesis in two bighorn sheep populations in South Dakota, USA. We identified and removed Mo chronic carriers from the Custer State Park (treatment) population. Simultaneously, we identified carriers but did not remove them from the Rapid City population (control). We predicted removal would result in decreased pneumonia, mortality, and Mo prevalence. Both population ranges had similar habitat and predator communities but were sufficiently isolated to preclude intermixing. We classified chronic carriers as adults that consistently tested positive for Mo carriage over a 20‐month sampling period ( n = 2 in the treatment population; n = 2 in control population). We failed to detect Mo or pneumonia in the treatment population after chronic carrier removal, while both remained in the control. Mortality hazard for lambs was reduced by 72% in the treatment population relative to the control (CI = 36%, 91%). There was also a 41% reduction in adult mortality hazard attributable to the treatment, although this was not statistically significant (CI = 82% reduction, 34% increase). Synthesis and Applications : These results support the hypothesis that Mo is a primary causative agent of persistent or recurrent respiratory disease in bighorn sheep populations and can be maintained by a few chronic carriers. Our findings provide direction for future research and management actions aimed at controlling pneumonia in wild sheep and may apply to other diseases.

South Dakota↗
USGS2020

Preliminary analysis to estimate the spatial distribution of benefits of P load reduction: Identifying the spatial influence of phosphorus loading from the Maumee River (USA) in western Lake Erie

Since the early 2000s, Lake Erie has been experiencing annual cyanobacterial blooms that often cover large portions of the western basin and even reach into the central basin. These blooms have affected several ecosystem services provided by Lake Erie to surrounding communities (notably drinking water quality). Several modeling efforts have identified the springtime total bioavailable phosphorus (TBP) load as a major driver of maximum cyanobacterial biomass in western Lake Erie, and on this basis, international water management bodies have set a phosphorus (P) reduction goal. This P reduction goal is intended to reduce maximum cyanobacterial biomass, but there has been very limited effort to identify the specific locations within the western basin of Lake Erie that will likely experience the most benefits. Here, we used pixel‐specific linear regression to identify where annual variation in spring TBP loads is most strongly associated with cyanobacterial abundance, as inferred from satellite imagery. Using this approach, we find that annual TBP loads are most strongly associated with cyanobacterial abundance in the central and southern areas of the western basin. At the location of the Toledo water intake, the association between TBP load and cyanobacterial abundance is moderate, and in Maumee Bay (near Toledo, Ohio), the association between TBP and cyanobacterial abundance is no better than a null model. Both of these locations are important for the delivery of specific ecosystem services, but this analysis indicates that P load reductions would not be expected to substantially improve maximum annual cyanobacterial abundance in these locations. These results are preliminary in the sense that only a limited set of models were tested in this analysis, but these results illustrate the importance of identifying whether the spatial distribution of management benefits (in this case P load reduction) matches the spatial distribution of management goals (reducing the effects of cyanobacteria on important ecosystem services).

Michigan, Ohio↗
USGS2020

Assessment of spatial genetic structure to identify populations at risk for infection of an emerging epizootic disease

Understanding the geographic extent and connectivity of wildlife populations can provide important insights into the management of disease outbreaks but defining patterns of population structure is difficult for widely distributed species. Landscape genetic analyses are powerful methods for identifying cryptic structure and movement patterns that may be associated with spatial epizootic patterns in such cases. We characterized patterns of population substructure and connectivity using microsatellite genotypes from 2,222 white-tailed deer ( Odocoileus virginianus ) in the Mid-Atlantic region of the United States, a region where chronic wasting disease was first detected in 2009. The goal of this study was to evaluate the juxtaposition between population structure, landscape features that influence gene flow, and current disease management units. Clustering analyses identified four to five subpopulations in this region, the edges of which corresponded to ecophysiographic provinces. Subpopulations were further partitioned into 11 clusters with subtle ( F ST ≤ 0.041), but significant genetic differentiation. Genetic differentiation was lower and migration rates were higher among neighboring genetic clusters, indicating an underlying genetic cline. Genetic discontinuities were associated with topographic barriers, however. Resistance surface modeling indicated that gene flow was diffuse in homogenous landscapes, but the direction and extent of gene flow were influenced by forest cover, traffic volume, and elevational relief in subregions heterogeneous for these landscape features. Chronic wasting disease primarily occurred among genetic clusters within a single subpopulation and along corridors of high landscape connectivity. These results may suggest a possible correlation between population substructure, landscape connectivity, and the occurrence of diseases for widespread species. Considering these factors may be useful in delineating effective management units, although only the largest features produced appreciable differences in subpopulation structure. Disease mitigation strategies implemented at the scale of ecophysiographic provinces are likely to be more effective than those implemented at finer scales.

Maryland, Pennsylvania, Virginia↗
USGS2020

Thermal heterogeneity, migration, and consequences for spawning potential of female bull trout in a river-reservoir system

The likelihood that fish will initiate spawning, spawn successfully, or skip spawning in a given year is conditioned in part on availability of energy reserves. We evaluated the consequences of spatial heterogeneity in thermal conditions on the energy accumulation and spawning potential of migratory bull trout ( Salvelinus confluentus ) in a regulated river–reservoir system. Based on existing data, we identified a portfolio of thermal exposures and migratory patterns and then estimated their influence on energy reserves of female bull trout with a bioenergetics model. Spawning by females was assumed to be possible if postspawning energy reserves equaled or exceeded 4 kJ/g. Given this assumption, results suggested up to 70% of the simulated fish could spawn each year. Fish that moved seasonally between a cold river segment and a warmer reservoir downstream had a greater growth rate and higher propensity to spawn in a given year (range: 40%–70%) compared with fish that resided solely in the cold river segment (25%–40%). On average, fish that spawned lost 30% of their energy content relative to their prespawn energy. In contrast, fish that skipped spawning accumulated, on average, 16% energy gains that could be used toward future gamete production. Skipped spawning occurred when water temperatures were relatively low or high, and if upstream migration occurred relatively late (mid-July or later) or early (early-May or earlier). Overall, our modeling effort suggests the configuration of thermal exposures, and the ability of bull trout to exploit this spatially and temporally variable thermal conditions can strongly influence energy reserves and likelihood of successful spawning.

Idaho↗
USGS2020

Ringed seal (Pusa hispida) seasonal movements, diving, and haul-out behavior in the Beaufort, Chukchi, and Bering Seas (2011–2017)

Continued Arctic warming and sea-ice loss will have important implications for the conservation of ringed seals, a highly ice-dependent species. A better understanding of their spatial ecology will help characterize emerging ecological trends and inform management decisions. We deployed satellite transmitters on ringed seals in the summers of 2011, 2014, and 2016 near Utqiaġvik (formerly Barrow), Alaska to monitor their movements, diving, and haul-out behavior. We present analyses of tracking and dive data provided by 17 seals that were tracked until at least January of the following year. Seals mostly ranged north of Utqiaġvik in the Beaufort and Chukchi Seas during summer before moving into the southern Chukchi and Bering Seas during winter. In all seasons, ringed seals occupied a diversity of habitats and spatial distributions; from near shore and localized, to far offshore and wide-ranging in drifting sea-ice. Continental shelf waters were occupied for >96% of tracking-days, during which repetitive-diving (suggestive of foraging) primarily to the seafloor was the most frequent activity. From mid-summer to early-fall, 12 seals made ~ one-week forays off-shelf to the deep Arctic Basin, most reaching the retreating pack-ice, where they spent most of their time hauled out. Diel activity patterns suggested greater allocation of foraging efforts to midday hours. Haul-out patterns were complementary, occurring mostly at night until April-May when midday hours were preferred. Ringed seals captured in 2011—concurrent with an unusual mortality event (UME) that affected all ice seal species—differed morphologically and behaviorally from seals captured in other years. Speculations about the physiology of molting and its role in energetics, habitat use, and behavior are discussed; along with possible evidence of purported ringed seal ecotypes.

Alaska↗
USGS2020

Diet of a rare herbivore based on DNA metabarcoding of feces: Selection, seasonality, and survival

In herbivores, survival and reproduction are influenced by quality and quantity of forage, and hence, diet and foraging behavior are the foundation of an herbivore's life history strategy. Given the importance of diet to most herbivores, it is imperative that we know the species of plants they prefer, especially for herbivorous species that are at risk for extinction. However, it is often difficult to identify the diet of small herbivores because: (a) They are difficult to observe, (b) collecting stomach contents requires sacrificing animals, and (c) microhistology requires accurately identifying taxa from partially digested plant fragments and likely overemphasizes less-digestible taxa. The northern Idaho ground squirrel ( Urocitellus brunneus ) is federally threatened in the United States under the Endangered Species Act. We used DNA metabarcoding techniques to identify the diet of 188 squirrels at 11 study sites from fecal samples. We identified 42 families, 126 genera, and 120 species of plants in the squirrel's diet. Our use of three gene regions was beneficial because reliance on only one gene region (e.g., only trnL ) would have caused us to miss >30% of the taxa in their diet. Northern Idaho ground squirrel diet differed between spring and summer, frequency of many plants in the diet differed from their frequency within their foraging areas (evidence of selective foraging), and several plant genera in their diet were associated with survival. Our results suggest that while these squirrels are generalists (they consume a wide variety of plant species), they are also selective and do not eat plants relative to availability. Consumption of particular genera such as Perideridia may be associated with higher overwinter survival.

Idaho↗
USGS2020

Robust age estimation of southern sea otters from multiple morphometrics

Reliable age estimation is an essential tool to assess the status of wildlife populations and inform successful management. Aging methods, however, are often limited by too few data, skewed demographic representation, and by single or uncertain morphometric relationships. In this study, we synthesize age estimates in southern sea otters Enhydra lutris nereis from 761 individuals across 34 years of study, using multiple noninvasive techniques and capturing all life stages from 0 to 17 years of age. From wild, stranded, and captive individuals, we describe tooth eruptions, tooth wear, body length, nose scarring, and pelage coloration across ontogeny and fit sex‐based growth functions to the data. Dental eruption schedules provided reliable and identifiable metrics spanning 0.3–9 months. Tooth wear was the most reliable predictor of age of individuals aged 1–15 years, which when combined with total length, explained >93% of observed age. Beyond age estimation, dental attrition also indicated the maximum lifespan of adult teeth is 13‒17 years, corresponding with previous estimates of life expectancy. Von Bertalanffy growth function model simulations of length at age gave consistent estimates of asymptotic lengths (male L oo = 126.0‒126.8 cm, female L oo = 115.3‒115.7 cm), biologically realistic gestation periods ( t 0 = 115 days, SD = 10.2), and somatic growth (male k = 1.8, SD = 0.1; female k = 2.1, SD = 0.1). Though exploratory, we describe how field radiographic imaging of epiphyseal plate development or fusions may improve aging of immature sea otters. Together, our results highlight the value of integrating information from multiple and diverse datasets to help resolve conservation problems.

Ecology and Evolution↗
USGS2020

Using a structured decision analysis to evaluate bald eagle vital signs monitoring in Southwest Alaska National Parks

1. Monitoring programs can benefit from an adaptive monitoring approach, where key decisions about why, where, what, and how to monitor are revisited periodically in order to ensure programmatic relevancy. 2. The National Park Service (NPS) monitors status and trends of Vital Signs to evaluate compliance with the NPS mission. Although abundant, The Southwest Alaska Network (SWAN) monitors bald eagles because of their inherent importance to park visitors and role as an important ecological indicator. Our goal is to identify an optimal monitoring program that may be standardized among participating parks. 3. We gathered an expert panel of scientists and managers, and implemented a Delphi Process to gather information about the bald eagle monitoring program. Panelists generated a list of means objectives for the monitoring program: minimizing cost, minimizing effort, maximizing the ability to detect change in bald eagle populations, and maximizing the amount of accurate information collected about bald eagles. 4. We used a swing-weighting technique to assign importance to each objective. Collecting accurate information about bald eagles was considered the most important means objective. 5. Combining panelist-generated information with objective importance, we analyzed the scenarios and defined the optimal decision using linear value modeling. Through our analysis, we found that a “Comprehensive” monitoring scenario, comprised of all feasible monitoring metrics is the optimal monitoring scenario. Even with greatly increased cost, the Comprehensive monitoring scenario remains the best solution. 6. We suggest further exploration of the cost and effort required for the Comprehensive scenario, to determine if it is in the parks’ best interest to begin monitoring additional metrics.

Alaska↗
USGS2020

Avian eggshell thickness in relation to egg morphometrics, embryonic development, and mercury contamination

Eggshell thickness is important for physiological, ecological, and ecotoxicological studies on birds; however, empirical eggshell thickness measurements for many species and regions are limited. We measured eggshell thickness at the equator and the egg poles for 12 avian species and related eggshell thickness to egg morphometrics, embryonic development, egg status, and mercury contamination. Within an egg, eggshells were approximately 5.1% thicker at the equator than the sharp pole of the egg, although this difference varied among species (0.6%–9.8%). Within Forster's tern ( Sterna forsteri ), where eggshell thickness was measured at 5 equally spaced positions along the longitude of the egg, eggshell thickness changed more rapidly near the sharp pole of the egg compared to near the blunt pole of the egg. Within species, eggshell thickness was related to egg width and egg volume for six of the 12 species but was not related to egg length for any species. Among species, mean eggshell thickness was strongly related to species mean egg width, egg length, egg volume, and bird body mass, although species mean body mass was the strongest predictor of species mean eggshell thickness. Using three species (American avocet [ Recurvirostra americana ], black‐necked stilt [ Himantopus mexicanus ], and Forster's tern), whose nests were carefully monitored, eggshell thickness (including the eggshell membrane) did not differ among viable, naturally abandoned, dead, or failed‐to‐hatch eggs; was not related to total mercury concentrations of the egg content; and did not decrease with embryonic age. Our study also provides a review of all existing eggshell thickness data for these 12 species.

Ecology and Evolution↗
USGS2020

Tropical understory herbaceous community responds more strongly to hurricane disturbance than to experimental warming

The effects of climate change on tropical forests may have global consequences due to the forests’ high biodiversity and major role in the global carbon cycle. In this study, we document the effects of experimental warming on the abundance and composition of a tropical forest floor herbaceous plant community in the Luquillo Experimental Forest, Puerto Rico. This study was conducted within Tropical Responses to Altered Climate Experiment (TRACE) plots, which use infrared heaters under free‐air, open‐field conditions, to warm understory vegetation and soils + 4°C above nearby control plots. Hurricanes Irma and María damaged the heating infrastructure in the second year of warming, therefore, the study included one pretreatment year, one year of warming, and one year of hurricane response with no warming. We measured percent leaf cover of individual herbaceous species, fern population dynamics, and species richness and diversity within three warmed and three control plots. Results showed that one year of experimental warming did not significantly affect the cover of individual herbaceous species, fern population dynamics, species richness, or species diversity. In contrast, herbaceous cover increased from 20% to 70%, bare ground decreased from 70% to 6%, and species composition shifted pre to posthurricane. The negligible effects of warming may have been due to the short duration of the warming treatment or an understory that is somewhat resistant to higher temperatures. Our results suggest that climate extremes that are predicted to increase with climate change, such as hurricanes and droughts, may cause more abrupt changes in tropical forest understories than longer‐term sustained warming.

Puerto Rico↗
USGS2020

Applications of deep convolutional neural networks to predict length, circumference, and weight from mostly dewatered images of fish

Simple biometric data of fish aid fishery management tasks such as monitoring the structure of fish populations and regulating recreational harvest. While these data are foundational to fishery research and management, the collection of length and weight data through physical handling of the fish is challenging as it is time consuming for personnel and can be stressful for the fish. Recent advances in imaging technology and machine learning now offer alternatives for capturing biometric data. To investigate the potential of deep convolutional neural networks to predict biometric data, several regressors were trained and evaluated on data stemming from the FishL™ Recognition System and manual measurements of length, girth, and weight. The dataset consisted of 694 fish from 22 different species common to Laurentian Great Lakes. Even with such a diverse dataset and variety of presentations by the fish, the regressors proved to be robust and achieved competitive mean percent errors in the range of 5.5 to 7.6% for length and girth on an evaluation dataset. Potential applications of this work could increase the efficiency and accuracy of routine survey work by fishery professionals and provide a means for longer‐term automated collection of fish biometric data.

Illinois, Michigan, Ohio↗
USGS2020

Spatial grain of adaptation is much finer than ecoregional-scale common gardens reveal

Adaptive variation among plant populations must be known for effective conservation and restoration of imperiled species and predicting their responses to a changing climate. Common‐garden experiments, in which plants sourced from geographically distant populations are grown together such that genetic differences may be expressed, have provided much insight on adaptive variation. Common‐garden experiments also form the foundation for climate‐based seed‐transfer guidelines. However, the spatial scale at which population differentiation occurs is rarely addressed, leaving a critical information gap for parameterizing seed‐transfer guidelines and assessing species’ climate vulnerability. We asked whether adaptation was evident among populations of a foundational perennial within a single “empirical” seed‐transfer zone (based on previous common‐garden findings evaluating very distant populations) but different “provisional” seed zones (groupings of areas of similar climate and are not parameterized from common‐garden data). Seedlings from three populations originating from similar conditions within an intermediate elevation were planted into gardens nearby at the same elevation, or 250–450 m higher or lower in elevation and 0.4–25 km away. Substantial variation was observed between gardens in survival (ranging 2%–99%), foliar crown volume (7.8–22.6 dm 3 ), and reproductive effort (0%–65%), but not among the three transplanted populations. The between garden variation was inversely related to climatic differences between the gardens and seed‐source populations, specifically the site differences in maximum–minimum annual temperatures. Results suggest that substantial site‐specificity in adaptation can occur at finer scales than is accounted for in empirical seed‐transfer guidance when the guidance is derived from broadscale common‐garden studies. Being within the same empirical seed zone, geographic unit, and even within 10 km distance may not qualify as “local” in the context of seed transfer. Moving forward, designing common‐garden experiments so that they allow for testing the scale of adaptation will help in translating the resulting seed‐transfer guidance to restoration projects.

Idaho↗
USGS2020

Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

Motion‐activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter‐out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the “species model,” and one that determines if an image is empty or if it contains an animal, the “empty‐animal model.” Our species model and empty‐animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out‐of‐sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%–91% across all out‐of‐sample datasets) and the empty‐animal model achieved an accuracy of 91%–94% on out‐of‐sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty‐animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.

Ecology and Evolution↗