Species distribution models predict genetic isolation of Hetaerina vulnerata Hagen in Selys, 1853 (Odonata, Calopterygidae).

Understanding how past and current environmental conditions shape the demographic and genetic distributions of organisms facilitates our predictions of how future environmental patterns may affect populations. The Canyon Rubyspot damselfly (Odonata: Zygoptera: Hetaerina vulnerata) is an insect with a range distribution from Colombia to the arid southwestern United States, where it inhabits shaded mountain streams in the arid southwestern United States. Past spatial fragmentation of habitat and limited dispersal capacity of H. vulnerata may cause population isolation and genetic differentiation, and projected climate change may exacerbate isolation by further restricting the species' distribution. We constructed species distribution models (SDMs) based on occurrences of H. vulnerata and environmental variables characterizing the species' niche. We inferred seven current potential population clusters isolated by unsuitable habitat. Paleoclimate models indicated habitat contiguity in past conditions; projected models indicated some habitat fragmentation in future scenarios. Seventy-eight H. vulnerata individuals from six of the current clusters were sequenced via ddRADseq and processed with Stacks. Principal components and phylogeographic analyses resolved three subpopulations; Structure resolved four subpopulations. F ST values were low (<0.05) for nearby populations and >0.15 for populations separated by expanses of unsuitable habitat. Isolation by distance was an existing but weak factor in determining genomic structure; isolation by environment and the intervening landscape explained a significant proportion of genetic distance. Hetaerina vulnerata populations were shown to be isolated by a lack of tree canopy coverage, an important habitat predictor for oviposition and territoriality. Thus, H. vulnerata populations are likely separated and are genetically isolated. Integrating SDMs with landscape genetics allowed us to identify populations separated by distance and unsuitable habitat, explaining population genetic patterns and probable fates for populations under future climate scenarios.

Watershed-Mediated Ecomorphological Variation: A Case Study with the Twin-Striped Clubtail Dragonfly (Hylogomphus geminatus).

Anthropogenic land-cover change is modifying ecosystems at an accelerating rate. Changes to ecomorphologically variable taxa within those ecosystems serve as early-warning signs that resources on which humans and other animals depend are being altered. One known ecomorphologically variable taxon is Hylogomphus geminatus, a species of dragonfly in the southeastern United States that shows pronounced variation in total body length across its limited geographic range. We measured total length of live as well as preserved museum specimens of H. geminatus and the sympatric species Progomphus obscurus (as a means for comparison). Both species showed significant size differences linked to HUC-8 watersheds in which they occur. H. geminatus showed additional significant differences on either side of the Apalachicola River, Florida, for all comparisons by sex. In overlapping watersheds, the species tended to show the same trends in length relative to their respective averages. Smaller body length was associated with more urban and agricultural land cover. These findings indicate that ecomorphological variation is tied to the watershed scale and point to significant variations on either side of the Apalachicola River. More thorough future analyses would be needed to verify trends in body length and identify the drivers behind them.

Urban Areas Create Refugia for Odonates in a Semi-Arid Region.

In western Texas, most wetlands are fed from precipitation runoff, making them sensitive to drought regimes, anthropogenic land-use activities in their surrounding watersheds, and the interactive effect between these two factors. We surveyed adult odonates in 133 wetlands (49 in grassland settings, 56 in cropland, and 28 in urban areas) in western Texas from 2003-2020; 33 species were recorded. Most species were widespread generalists, but urban wetlands had the highest species richness, as well as the most unique species of any of the three wetland types. Non-metric, multidimensional scaling ordination revealed that the odonate community in urban wetlands was distinctly different in composition than the odonates in non-urban wetlands. Urban wetlands were smaller in surface area than the other wetland types, but because they were fed from more consistently available urban runoff rather than seasonal precipitation, they had longer hydroperiods, particularly during a multi-year drought when wetlands in other land-cover contexts were dry. This anthropogenically enhanced water supply was associated with higher odonate richness despite presumably impaired water quality, indicating that consistent and prolonged presence of water in this semi-arid region was more important than the presence of native land cover within which the wetland existed. Compared to wetlands in the regional grassland landscape matrix, wetlands in agricultural and urban areas differed in hydroperiod, and presumably also in water quality; these effects translated to differences in the regional odonate assemblage by surrounding land-use type, with the highest richness at urban playas. Odonates in human environments may thus benefit through the creation of a more reliably available wetland habitat in an otherwise dry region.

Identifying structural connectivity priorities in eastern Paraguay's fragmented Atlantic Forest.

The Atlantic Forest of eastern Paraguay has experienced extensive recent deforestation. Less than one-third of the region is forested, and the remaining forest largely consists of isolated remnants with potentially disrupted connectivity for forest fauna. We used a graph theory approach to identify those forest remnants that are important in maintaining landscape structural connectivity for mammals in this fragmented forest. We quantified structural connectivity for forest remnants over the period 2000-2019 at three levels: the entire network of Atlantic Forest remnants in eastern Paraguay; at 10 smaller, nested spatial scales (40-10,000 m) encompassing a range of potential mammalian dispersal abilities; and at the level of individual remnants. We used 10 graph theory metrics to assess aspects of network complexity, dispersal-route efficiency, and individual remnant importance in supporting structural connectivity. We identified forest remnants that serve as important structural connectivity roles as stepping stones, hubs, or articulation points and that should be prioritized for connectivity conservation. Structural connectivity was constrained for organisms incapable of travelling at least 9-12 km (farthest distances between nearest-neighboring forest remnants depending on whether smaller remnants were included or not) and was particularly limited for area-sensitive forest-specialist mammals. With the increased forest loss and fragmentation that is occurring, the connectivity of this system will likely be further compromised, but most of the remnants that we identified as playing important roles for structural connectivity were outside of the country's proposed "green corridor," indicating additional areas where conservation action can be directed.

Exposure of Foraging Bees (Hymenoptera) to Neonicotinoids in the U.S. Southern High Plains.

Exposure to pesticides is a major threat to insect pollinators, potentially leading to negative effects that could compromise pollination services and biodiversity. The objectives of this study were to quantify neonicotinoid concentrations among different bee genera and to examine differences attributable to body size and surrounding land use. During the period of cotton planting (May-June), 282 wild bees were collected from habitat patches associated with cropland, grassland, and urban land cover and analyzed for three neonicotinoids (thiamethoxam, clothianidin, and imidacloprid). Twenty bees among eight genera contained one or more of the neonicotinoid compounds and detections occurred in all landscape types, yet with the most detections occurring in cropland-associated habitats. Apis Linnaeus (Hymenoptera: Apidae), Melissodes Latreille (Apidae), Perdita Smith (Andrenidae), and Lasioglossum Curtis (Halictidae) had multiple individuals with neonicotinoid detections. Two of the largest bees (Apis and Melissodes) had the greatest number of detections within genera, yet the relatively small-bodied genus Perdita had the three highest neonicotinoid concentrations reported. The number of detections within a genus and average generic body mass showed a marginally significant trend towards larger bees having a greater frequency of neonicotinoid detections. Overall, the relatively low percentage of detections across taxa suggests practices aimed at conserving grassland remnants in intensified agricultural regions could assist in mitigating exposure of wild bees to agrochemicals, while differences in bee traits and resource use could in part drive exposure. Further work is needed to address variable agrochemical exposures among pollinators, to support strategies for conservation and habitat restoration in affected landscapes.

Linking microbial community structure and function to seasonal differences in soil moisture and temperature in a Chihuahuan desert grassland.

Global and regional climate models predict higher air temperature and less frequent, but larger precipitation events in arid regions within the next century. While many studies have addressed the impact of variable climate in arid ecosystems on plant growth and physiological responses, fewer studies have addressed soil microbial community responses to seasonal shifts in precipitation and temperature in arid ecosystems. This study examined the impact of a wet (2004), average (2005), and dry (2006) year on subsequent responses of soil microbial community structure, function, and linkages, as well as soil edaphic and nutrient characteristics in a mid-elevation desert grassland in the Chihuahuan Desert. Microbial community structure was classified as bacterial (Gram-negative, Gram-positive, and actinomycetes) and fungal (saprophytic fungi and arbuscular mycorrhiza) categories using (fatty acid methyl ester) techniques. Carbon substrate use and enzymic activity was used to characterize microbial community function annually and seasonally (summer and winter). The relationship between saprophytic fungal community structure and function remained consistent across season independent of the magnitude or frequency of precipitation within any given year. Carbon utilization by fungi in the cooler winter exceeded use in the warmer summer each year suggesting that soil temperature, rather than soil moisture, strongly influenced fungal carbon use and structure and function dynamics. The structure/function relationship for AM fungi and soil bacteria notably changed across season. Moreover, the abundance of Gram-positive bacteria was lower in the winter compared to Gram-negative bacteria. Bacterial carbon use, however, was highest in the summer and lower during the winter. Enzyme activities did not respond to either annual or seasonal differences in the magnitude or timing of precipitation. Specific structural components of the soil microbiota community became uncoupled from total microbial function during different seasons. This change in the microbial structure/function relationship suggests that different components of the soil microbial community may provide similar ecosystem function, but differ in response to seasonal temperature and precipitation. As soil microbes encounter increased soil temperatures and altered precipitation amounts and timing that are predicted for this region, the ability of the soil microbial community to maintain functional resilience across the year may be reduced in this Chihuahuan Desert ecosystem.

Wildlife Refuges Support High Bee Diversity on the Southern Great Plains.

The native prairie of the southern Great Plains has been especially modified by two related forces: conversion of native prairie to agricultural forms of land use and removal of black-tailed prairie dogs (Rodentia: Sciuridae, Cynomys ludovicianus (Ord, 1815)) that act as ecosystem engineers via their burrowing and grazing activities. It is unknown how these changes have affected the native bee community. We surveyed the bee communities in relatively intact native prairie at two National Wildlife Refuges in Texas, quantifying bee community structure as a function of the presence/absence of grazing by prairie dogs. Over a 5-mo sampling period in spring-summer 2013, we found high overall bee diversity (180 species, mostly solitary ground-nesters), with differences detected in diversity between Muleshoe and Buffalo Lake National Wildlife Refuges as well as on and off prairie dog colonies. Although the same three species dominated the bee community at both refuges, most species were represented by relatively few individuals, leading to overall differences in diversity (richness, evenness, and effective number of species) by refuge. Bee diversity differed between sites on and off prairie dog colonies, but in trends that differed by refuge and by index, suggesting that location was more important than prairie dog presence. These results represent a reference fauna against which other regional bee communities in other land-cover types can be compared, but the high spatial heterogeneity we found indicates that detecting effects of landscape change on native bees will be challenging.

Seasonal variation of North American form of Gigantolaelaps mattogrossensis (Acari: Laelapidae) on marsh rice rat in southern coastal Texas.

The ectoparasites of a small mammal community within an intertidal zone in the upper Gulf coast region of Texas were studied to assess the seasonal variation in abundances of the mite Gigantolaelaps mattogrossensis (Fonseca) (Acari: Laelapidae) on the marsh rice rat, Oryzomys palustris (Harlan). Further study into the ecology and dynamics of this parasite-host relationship was deemed to be necessary to expand the understanding of these potential participants in the ecology of Bayou Hantavirus, an important causative agent of Hantavirus pulmonary syndrome. The objective of this study was to determine the effects of five predictor variables on mite abundance: prevalences of hosts, relative humidity, precipitation, temperature, and length of daylight. Mite abundance was modeled as a function of the five variables with analyses of variance and multiple regressions; however, because the predictor variables pertain to the sampling period rather than to the individual rodent host, the effective sample size was small and thus the sums of squares and cross products matrix was singular. We therefore developed and used a new method for estimating regression coefficients based on the "noise-addition method" (random residual variation) combined with a bootstrap step converting the reduced rank data to full rank, providing realistic estimates of confidence intervals for the regression statistics. The population abundances of mites fluctuated significantly across collecting periods. Humidity and precipitation were the most influential variables in explaining the variation in abundances of mites. Model interpretation suggests that G. mattogrossensis is a nidicolous parasite. These results provide a baseline understanding of the seasonal interactions between parasite and host.

Quantifying the degree of bias from using county-scale data in species distribution modeling: Can increasing sample size or using county-averaged environmental data reduce distributional overprediction?

Citizen-science databases have been used to develop species distribution models (SDMs), although many taxa may be only georeferenced to county. It is tacitly assumed that SDMs built from county-scale data should be less precise than those built with more accurate localities, but the extent of the bias is currently unknown. Our aims in this study were to illustrate the effects of using county-scale data on the spatial extent and accuracy of SDMs relative to true locality data and to compare potential compensatory methods (including increased sample size and using overall county environmental averages rather than point locality environmental data). To do so, we developed SDMs in maxent with PRISM-derived BIOCLIM parameters for 283 and 230 species of odonates (dragonflies and damselflies) and butterflies, respectively, for five subsets from the OdonataCentral and Butterflies and Moths of North America citizen-science databases: (1) a true locality dataset, (2) a corresponding sister dataset of county-centroid coordinates, (3) a dataset where the average environmental conditions within each county were assigned to each record, (4) a 50/50% mix of true localities and county-centroid coordinates, and (5) a 50/50% mix of true localities and records assigned the average environmental conditions within each county. These mixtures allowed us to quantify the degree of bias from county-scale data. Models developed with county centroids overpredicted the extent of suitable habitat by 15% on average compared to true locality models, although larger sample sizes (>100 locality records) reduced this disparity. Assigning county-averaged environmental conditions did not offer consistent improvement, however. Because county-level data are of limited value for developing SDMs except for species that are widespread and well collected or that inhabit regions where small, climatically uniform counties predominate, three means of encouraging more accurate georeferencing in citizen-science databases are provided.

A longitudinal study of Bayou virus, hosts, and habitat.

What is currently known about the ecology of North American hantaviruses has come largely from studies on Sin Nombre virus (SNV). We conducted a longitudinal study of Bayou virus (BAYV), the second-leading agent of hantavirus pulmonary syndrome in the United States. Antibodies to hantavirus were detected from Oryzomys palustris (most commonly infected species), Sigmodon hispidus, Peromyscus leucopus, Reithrodontomys fulvescens, and Baiomys taylori. However, only O. palustris had viral RNA in tissues and excreta, suggesting that antibodies detected in other species may have resulted from spill-over infection. Seroprevalence rates averaged around 16% for O. palustris and varied seasonally. The heaviest males exhibited the highest levels of seroprevalence. Seroprevalence was higher in coastal prairie (20.0%) than old-fields (10.5%) and was associated with host abundance. These patterns are similar to those of SNV and can be used in identification of potentially at-risk areas.

Socio-ecology of the marsh rice rat (Oryzomys palustris) and the spatio-temporal distribution of Bayou virus in coastal Texas.

Along the southeastern coast of the United States of America (USA), the marsh rice rat (Oryzomys palustris) is the primary host for the hantavirus, genotype Bayou. According to the socio-ecological model for a territorial, polygamous species, females should be distributed across space and time by habitat resources and predation risks, whereas males should space themselves according to the degree of female aggregation and reproductive synchrony. To investigate how females affect the male-male transmission paradigm of Bayou virus, rodents were captured, marked, released in two macrohabitat types and followed across a 30-month period. Microhabitat cover variables were quantified around the individual trap stations. A geodatabase was created from habitat and rodent capture data and analysed in a geographical information system. The ratio of breeding to non-breeding females was ~1:1, with breeding females overly dispersed and non-breeding females randomly dispersed. Spatial analyses revealed both macro- and microhabitat preferences in females. Compared to seronegatives, higher proportions of seropositive adult males were found consistently within closer proximities to breeding females but not to non-breeding females, indicating that male locations were not driven simply by habitat selection. Activities to acquire dispersed receptive females could be an important driver of Bayou virus transmission among male hosts. Herein, we describe an interdisciplinary effort providing a novel approach to elucidate the complexity of hantavirus trafficking and maintenance in rodent populations of a coastal marsh ecosystem.

Bayou virus detected in non-oryzomyine rodent hosts: an assessment of habitat composition, reservoir community structure, and marsh rice rat social dynamics.

In the United States, Bayou virus (BAYV) ranks second only to Sin Nombre virus (SNV) in terms of hantavirus pulmonary syndrome (HPS) incidents, having been confirmed in cases from Texas and Louisiana since its discovery in 1994. This study on BAYV infection among sympatric, non-oryzomyine rodents ("spillover") in Freeport, TX, is the first to link patterns of hantavirus interspecific spillover with the spatiotemporal ecology of the primary host (marsh rice rat, Oryzomys palustris). Mark-recapture and/or harvest methods were employed from March 2002 through May 2004 in two macrohabitat types. Rodent blood samples were screened for the presence of IgG antibody to BAYV antigen by IFA after which Ab-positive blood, saliva, and urine were analyzed for the presence of viral RNA by nested RT-PCR. From 727 non-oryzomyine captures, five seropositive (but not viral RNA positive) individuals were detected: one each of Baiomys taylori, Peromyscus leucopus, and Reithrodontomys fulvescens; and two Sigmodon hispidus. Spillover hosts were not associated with macrohabitat where O. palustris abundance, density, or seroprevalence was highest. Rather, spillover occurred in the macrohabitat indicative of greater overall disturbance (as indicated by grazing and exotic plant diversity) and overall biodiversity. Spillover occurred during periods of high seroprevalence detected elsewhere within the study region. Spillover locations differed significantly from all other capture locations in terms of percent water, shrub, and grass cover. Although greater habitat and mammal diversity of old-fields may serve to reduce seroprevalence levels by tempering intraspecific contacts between rice rats, greater diversity also may create an ecologically opportunistic setting for BAYV spillover. Impacts of varying levels of disturbance and biodiversity on transmission dynamics represent a vastly uncharacterized component of the evolutionary ecology of hantaviruses.

From patterns to emerging processes in mechanistic urban ecology.

Rapid urbanization has become an area of crucial concern in conservation owing to the radical changes in habitat structure and loss of species engendered by urban and suburban development. Here, we draw on recent mechanistic ecological studies to argue that, in addition to altered habitat structure, three major processes contribute to the patterns of reduced species diversity and elevated abundance of many species in urban environments. These activities, in turn, lead to changes in animal behavior, morphology and genetics, as well as in selection pressures on animals and plants. Thus, the key to understanding urban patterns is to balance studying processes at the individual level with an integrated examination of environmental forces at the ecosystem scale.