The utilization of species distribution model(s) (SDM) for approximating, explaining, and predicting changes in species' geographic locations is increasingly promoted for proactive ecological management. Although frameworks for modeling non-invasive species distributions are relatively well developed, their counterparts for invasive species-which may not be at equilibrium within recipient environments and often exhibit rapid transformations-are lacking. Additionally, adaptive ecological management strategies address the causes and effects of biological invasions and other complex issues in social-ecological systems. We conducted a review of biological invasions, species distribution models, and adaptive practices in ecological management, and developed a framework for adaptive, niche-based, invasive species distribution model (iSDM) development and utilization. This iterative, 10-step framework promotes consistency and transparency in iSDM development, allows for changes in invasive drivers and filters, integrates mechanistic and correlative modeling techniques, balances the avoidance of type 1 and type 2 errors in predictions, encourages the linking of monitoring and management actions, and facilitates incremental improvements in models and management across space, time, and institutional boundaries. These improvements are useful for advancing coordinated invasive species modeling, management and monitoring from local scales to the regional, continental and global scales at which biological invasions occur and harm native ecosystems and economies, as well as for anticipating and responding to biological invasions under continuing global change.

Core Ideas Field‐based conservation CUREs can engage more students in authentic research. Model‐based pedagogy in CUREs allows students to grapple with complexity of scientific research. Post CURE, student assessment shows science skill gains and clarity in professional goals. Post CURE, students are more likely to talk with friends, family, or the public about wildlife conservation. Course‐based undergraduate research experiences (CUREs) have been developed to overcome barriers including students in research. However, there are few examples of CUREs that take place in a conservation and natural resource context with students engaging in field research. Here, we highlight the development of a conservation‐focused CURE integrated to a research program, research benefits, student self‐assessment of learning, and perception of the CURE. With the additional data, researchers were able to refine species distribution models and facilitate management decisions. Most students reported gains in their scientific skills, felt they had engaged in meaningful, real‐world research. In student reflections on how this experience helped clarify their professional intentions, many reported being more likely to enroll in graduate programs and seek employment related to science. Also interesting was all students reported being more likely to talk with friends, family, or the public about wildlife conservation issues after participating, indicating that courses like this can have effects beyond the classroom, empowering students to be advocates and translators of science. Field‐based, conservation‐focused CUREs can create meaningful conservation and natural resource experiences with authentic scientific teaching practices.

Home range estimation is an important analytical method in applied spatial ecology, yet best practices for addressing the effects of spatial variation in detection probability on home range estimates remain elusive. We introduce the R package "DiagnoseHR," simulation tools for assessing how variation in detection probability arising from landscape, animal behavior, and methodological processes affects home range inference. We demonstrate the utility of simulation methods for home range analysis planning by comparing bias arising from three home range estimation methods under multiple detection scenarios. We simulated correlated random walks in three landscapes that varied in detection probability and constructed home ranges from locations filtered through a range of sampling protocols. Home range estimates were less biased by reduced detection probability when sampling effort was increased, but the effects of sampling day distribution were minimal. Like others, we found that kernel density estimates were the least affected by variation in detection probability, while minimum convex polygons were most affected. Our results illustrate the value of quantifying uncertainty in home range estimates and suggest that field biologists working in environments with low detection may wish to weight sample-size greater than concerns about temporal autocorrelation when designing sampling protocols.

No abstract

Camera traps are an increasingly popular means to monitor wildlife populations. However, like other techniques for measuring populations, camera traps are subject to sources of error that may bias population estimates. Past studies accounting for detection error have failed to account for a simple but potentially widely pervasive source of environmental error: weather conditions. Using 5,108,416 photographs from 804 scent-lured camera traps deployed in western Nebraska, USA, during spring and autumn of 2014 and 2015, we analyzed the relationship between weather conditions (barometric pressure, wind speed, precipitation, and temperature) and coyote (Canis latrans) detection probability. Using binomial generalized linear mixed-effects models, we showed that detection probability was affected by all weather conditions examined. Weather effects on detection suggests that either weather alters coyote behavior or decreases trap efficacy. Detection probability also decreased over the exposure period, indicating that coyotes either avoided traps after initial exploration or that lure efficacy decreased over time. Our findings suggest that to achieve accurate population indices, camera-trap studies need to incorporate effects of weather conditions and sampling duration into population models to account for detection bias in estimates.