Diverse ecosystems are changing in response to shifts in climate and species composition; and understanding how communities, populations, and species will respond to these changes is a major challenge in ecology. Progress has been complicated because some climate changes are directional while others are cyclic in nature. Furthermore, responses to climate can be affected by species interactions and by time lags in how species respond to changes in climate. This project will use a long-term study of small mammals near Portal, Arizona, to understand whether recent changes in desert communities result from directional anthropogenic global change or from a natural, decadal-scale climate cycle, the Pacific Decadal Oscillation. Long-term experimental manipulations removing a keystone group of species, kangaroo rats, will be used to assess whether and how species composition influences ecological responses to climate.
This project will further scientific understanding of climate change impacts on desert ecosystems, and whether shifts in deserts are due to anthropogenic global change or natural decadal scale climate oscillations. Results will also indicate whether species composition can direct ecosystem-level responses to climate change. This project will integrate research and education by training undergraduate and graduate students and by disseminating research results broadly through web-based fora such as blogs.
Nature is always changing. We often think about the importance of non-biological events in causing nature to change; meteors, droughts, floods, and fires cause immediate and dramatic changes in the populations of species and the community of species that live in places affected by those events. While clearly non-biological events impact on the dynamics of an ecosystem, a major focus in ecology is on how species interact with each other. The primary focus of this grant was to understand how species interacting with each other can influence the dynamics of an ecosystem. We used data from a long-term experiment in southeastern Arizona that manipulated desert rodents to gain a better understanding of the influence of interactions among species in driving how nature changes. We focused on desert rodents because they are a very diverse and abundant group of organisms in deserts and one of the dominant consumer of seeds. Through their consumption of seeds, desert rodents can have strong effects on the abundance and diversity of plants. Because this study has been running since 1977, we have a long record of changes in both the climate and the species that we could use. Using this data, we found evidence that shifts in species interactions, in combination with climate fluctuations, can result in the population explosion of an invasive plant. When an important rodent seed consumer was negatively impacted by drought, this created a window of opportunity for an invasive plant to increase its population. Once the invasive plant increased, it had negative effects on the abundance and richness of native species. This demonstrates that species interactions can drive important changes in ecosystems. While reducing populations of seed consumers had a strong impact on identity of plant species in an ecosystem, additional research showed that changing who the dominant seed consumer is has little impact on structure of the plant community (i.e. the number of common and rare species, and how those species accumulate over space and time). Only removing seed consumers almost completely impacted the structure of the plants community. While minor changes in species interactions can have major impacts on what species are present, only major changes in the interactions among species (i.e. major reduction in seed consumption pressures) impact the overall structure of the ecosystem. In addition to studying when changes in species interactions might drive changes in ecosystems, we also studied how differences among species might affect cause species to respond differently to the same changes in their environment. Rare species are often thought to be more likely to go extinct in response to random fluctuations in the environment because they have such small population sizes. Using a combination of computer modelling and long-term data, we found that some rare species have population growth rates that are extremely sensitive to how common or rare that species is in the ecosystem. At low abundances, these species exhibit high growth rates that can rescue them from extinction. However, when they become abundant, they suffer severe declines in their growth rates, causing their population to shrink. This combination results in some rare species being able to maintain stable, but small populations, in the face of a fluctuating environment. More common species, however, seem to lack this tight regulation of their populations. We do not know why some rare species exhibit this phenomenon, but looking at data across many communities we found evidence for this behavior in many different vertebrate, invertebrate and plant species across many ecosystems. In another study, we also found that rare species might have a correlated set of life history traits (traits related to the timing of growth, reproduction, and mortality). At our study site, rodent species that tend to be rare also tend to have larger movement distances, higher mortality rates, and higher reproduction rates than common species. Common species appear to stay in one location for most of their lives, but rare species are constantly moving. These differences in how much common and rare species move could indicate that common and rare species use the same environment in different ways. Continuing to understand how and why common and rare species differ is an important step to being able to predict how these different species may respond to the same changes in their environment. We accomplished tasks to ensure that work from this grant not only had scientific impact, but also broader impacts on training scientists and public awareness. In addition to training three graduate students, we partnered with organizations that train scientists to use computer programming more effectively in data management and analysis and created special versions of our long-term data for use in their workshops. We also communicated our work to the broader public by writing about our work on a publicly accessible blog.
