Research

Modeling tools for agricultural decision support

My current research supports the development of science-based, ecologically-informed site and spatialized models that can help agricultural decision makers with managing and monitoring pests, their crop hosts, and their natural enemies. Most of my work has focused on developing an open-source population modeling platform that can estimate where (potential distribution) an invasive species could potentially establish as well as when (phenology) developmental stages are expected to occur, in order to support and improve strategic and tactical pest management decisions. The Degree-Day, establishment Risk, and Phenological event mapping system (DDRP) uses a process-based (mechanistic) model approach with a daily time-step to estimate the potential distribution, number of generations, and phenological events (e.g., when adults first emerge) over a user-specified time period (Barker et al. 2020). I have co-developed DDRP models for 16 high-priority invasive insect species (see here) to support surveillance activities of the USDA's Cooperative Agricultural Pest Survey (CAPS) program, in addition to collaborating on model development for three biological control insects (Grevstad et al. 2022). For example, the video showing forecasts of emerald ash borer adult emergence for each month of the year in 2022 (upper right panel) could help with planning surveillance and monitoring activities, most of which target the adult stage.

Currently I am expanding the DDRP system's capabilities to accurately model moisture-sensitive organisms such as weeds and plant pathogens. This includes work to predict risk of infection and the potential distribution of an invasive fungal pathogen that has devastated boxwood (Buxus spp.), a major evergreen shrub crop and keystone forest species (lower right panel). My recent climatic suitability modeling work on the pathogen at regional and world scales (Barker et al. 2022) helped with developing the DDRP model for this species.

Additionally, I am working with collaborators at OSU and the USA National Phenology Network to build models for a wider range of biosecurity threats, use ground-based observation data to validate mode forecasts, offer forecasts in user-friendly and interactive formats, and engage with stakeholders to promote forecast use and gather citizen science data (NIFA AFRI funded project, 2022-2025).

EAB 2022

Video of DDRP forecasts of adult emerald ash borer emergence for 2022 (click to play). Forecasts are shown for each month of the year, where the potential for long-term establishment is indicated by areas not experiencing moderate (excl.-moderate) or severe (excl.-severe) temperature stress. The map for December 2022 depicts the potential distribution based on climate stress accumulation over the entire year.

DDRP forecasts of cumulative infection risk for boxwood blight for 2022, where the potential for long-term establishment is indicated by areas not experiencing moderate (excl.-moderate) or severe (excl.-severe) temperature stress and dry stress for the entire year.

Measuring drivers of ecological change in the Great Basin

Understanding the factors that influence vegetation responses to disturbance is important because vegetation is the foundation of food resources, wildlife habitat, and ecosystem properties and processes. My research in the Pilliod Lab at the USGS Forest and Rangeland and Ecosystem Science Center (FRESC) in Boise, ID, integrated vegetation cover data derived from field plots and remotely sensed Landsat images over a 37-year period (1979-2016) to investigate the factors that influence the responses of vegetation to fire in sagebrush ecosystems in the Great Basin (Barker et al. 2018, 2019).

I found that areas burned by fire since the 1980s had higher annual herbaceous cover than unburned areas both historically and contemporarily, and that plots with historically high herbaceous cover were more susceptible to burning. The results suggest that burned areas historically occupied by sagebrush-dominated plant communities may have been invaded by exotic annuals prior to burning, possibly because of prior land uses, and after burning, have now transitioned to a persistent herbaceous-dominated state.

Map of a study area where field plot data were integrated with Landsat-derived component data to measure long-term vegetation trends (Barker et al. 2019).

Types and sources of adaptive genetic variation for colonizers

Introduced species offer excellent opportunities to investigate the factors that shape species ranges because their introduction history (i.e., source population, founding population size, and sites of introduction) is often well documented. As a PERT postdoctoral fellow at the University of Arizona, I conducted a population genomic study of the hairy weevil (Eustenopus villosus), an insect which is native to Europe and introduced to the western US to control invading yellow starthistle (Centaurea solstitialis; right). Introduced weevils are descended from a single source population in Greece, and may therefore have insufficient genetic variation to adapt to climate in the western US.

I am using landscape genomics approaches to explore genetic adaptation to climate in the native range of hairy weevils, and to assess how much of this variation is present in introduced populations. The results will improve our understanding of the evolutionary potential (i.e., ability to adapt) of small populations in novel environments, as well as the role of adaptive genetic variation in shaping species distributions. In collaboration with Katrina Dlugosch, I also reconstructed colonization pathways and identified sources of adaptive genetic variation for invading yellow starthistle (Barker et al. 2017). Additionally, we reported on one of the first empirical tests of the joint effects of expansion dynamics and environment on effective population size variation during invasive range expansion (Barker et al. 2019).

The hairy weevil (E. villosus) and its host plant, yellow starthistle (C. solstitialis; top). A genomic analysis of starthistle (Barker et al. 2017), revealed the origins of invading starthistle populations in the western US and the history of colonization in Eurasia (right, a discriminant analysis of principal components).


Historical and topographic drivers of tropical insular diversity

Island ecosystems, or those which are geographically isolated, are “natural laboratories” for studying evolution. Despite a rich history of research on island evolution and biogeography, the effects of topography and past climate change in generating island biodiversity are not well understood. My PhD dissertation explored the role of topography and range shifts resulting from past sea-level changes and climate suitability in shaping genetic diversity of two frogs in the Puerto Rican Bank, an archipelago in the eastern Caribbean Sea.

Basins and past climate change: impacts on genetic variation in montane island endemics

To better understand lineage diversification of tropical montane species within islands, I studied the historical biogeography and consequences of palaeoenvironmental change on patterns of genetic diversity in the mountain coquí (above), Eleutherodactylus portoricensis, a frog endemic to montane rain forests of Puerto Rico. I found that the low-elevation Caguas Basin in eastern Puerto Rico is a long-term barrier to gene flow between populations in the Luquillo and Cayey Mountains, with population divergence beginning more than 75 ka (Barker et al. 2011). Stable population sizes over time indicate a lack of demographic response to climatic changes during the last glacial period. The results highlight the importance of topographic complexity in promoting within-island vicariant speciation in the Greater Antilles, and indicate long-term persistence and lineage diversification despite Quaternary climatic oscillations.

The mountain coqui (E. portoricensis) of Puerto Rico has experienced widespread declines and is currently restricted to only two mountain ranges (Barker and Rios-Franceschi, 2014). These populations have been isolated for thousands of years, as demonstrated by a phylogenetic tree (right) that shows two divergent clades (from Barker et al. 2015).

Sea level, topography, and human introductions: impacts on genetic variation in widespread island endemics

Quaternary climatic oscillations caused changes in sea level that altered the size, number, and degree of isolation of islands, particularly in land-bridge archipelagos such as the Puerto Rican Bank. To increase our understanding of the role of climate change in shaping evolutionary processes in tropical insular systems, I used multiple DNA loci to examine demographic phenomena and assess the effect of sea-level changes on populations of the red-eyed coquí, E. antillensis across the Puerto Rican Bank. A hypothesis of colonization of the Eastern Islands from sources in eastern Puerto Rico during the penultimate and last glacial period, when a land-bridge united the Puerto Rican Bank, was supported (Barker et al. 2012). The Río Grande de Loíza Basin in eastern Puerto Rico delineates a phylogeographic break. Haplotypes shared between the Puerto Rican Bank and St. Croix (an island ca. 105 km southeast of this archipelago) represent human-mediated introductions (Barker et al. 2017).

I identified three genetically distinctive population groups of the red-eyed coqui (Eleutherodactylus antillensis) in the Puerto Rican Bank (Barker et al. 2012). However, populations in the long-isolated island of St. Croix were likely introduced recently from sources in the Virgin Islands and/or eastern Puerto Rico (Barker et al. 2017).

Climatic stability and genetic endemism in tropical islands

The ecological requirements of species largely determine what geologic and climatic factors constitute barriers to gene flow, and ultimately play a role in divergence. There is remarkably little known, however, about the interplay of historical events and ecological requirements driving intra-specific diversification within tropical island systems. Using multi-locus sequence data and spatially explicit models of population history developed from ecological niche models, I explored the role of late Quaternary climate change in shaping population structure and historical demographics of E. portoricensis, a narrowly distributed montane forest specialist, and E. antillensis, a widespread habitat generalist. I found that areas of high habitat stability fostered population persistence and the accumulation of genetic endemism, with E. antillensis experiencing greater climate-driven fluctuations in population size than E. portoricensis (Barker et al. 2015). Unstable habitat in the Río Grande de Loíza Basin is a historical barrier to dispersal and levels of genetic isolation in each species likely track species-specific physiological requirements. Understanding how late Quaternary climatic oscillations and topography affect gene flow and past population dynamics provides a framework for identifying evolutionary processes that underlie speciation in tropical islands, and can inform predictions of potential responses of endemic tropical insular biodiversity to future habitat alterations.

Frog tissues and both traditional and photographic voucher specimens associated with my dissertation research are archived at the Museum of Southwestern Biology, University of New Mexico, Albuquerque or the Museum of Vertebrate Zoology, University of California, Berkeley. Collaborators and co-authors Joe Cook (University of New Mexico), Bob Waide (University of New Mexico), and Javier Rodríguez-Robles (University of Nevada Las Vegas) were instrumental in completing this research. Multiple colleagues, undergraduates, and friends in both New Mexico and Puerto Rico also made this work possible!

Areas in Puerto Rico that were climatically stable during the late Quaternary promoted the accumulation of unique genetic diversity in populations of the endemic mountain coqui (left) and red-eyed coqui (right) (Barker et al. 2015).

Long-term bird population dynamics

Tropical ecosystems are changing rapidly in response to deforestation, climate change, urbanization, and other drivers. Unfortunately, a lack of long-term data on the effects of these changes hinders our ability to predict how ecosystems will respond to future perturbations. I collaborated with Principal Investigator Bob Waide at the University of New Mexico on a project investigating how change in climate and disturbance regimes affect long-term population dynamics of birds in the Luquillo Mountains of Puerto Rico. This work, which is part of the Luquillo Long-Term Ecological Research Program (Luquillo LTER), involves estimating the density and frequency of 54 bird species across a 16 hectare study plot known as the "Big Grid" over a 24 year time period.

Collecting mist-net data for a long-term bird monitoring project in the Luquillo Experimental Forest in Puerto Rico (middle Puerto Rican tody). Disturbances such as hurricanes and landslides (right) can strongly influence population dynamics.

Testing the role of selection in stabilizing and shaping the G-matrix

The additive genetic variance-covariance matrix, or G-matrix, plays a central role in evolutionary theory in predicting deterministic responses to selection as well as the stochastic consequences of finite population size. Yet, only a few studies have formulated and tested predictions about how multivariate selection might affect the evolution and stability of the G-matrix. In collaboration with Steve Arnold (Oregon State University) and Pat Phillips (University of Oregon), I formulated a prediction about how multivariate selection might affect the evolution and stability of the G-matrix and tested it with a comparison of G-matrices estimated in three populations of the garter snake Thamnophis elegans (Barker et al. 2010).

Thamnophis elegans