Annual Review of Ecology, Evolution, and Systematics - Volume 48, 2017

For this article, I reviewed empirical studies finding significant ecological responses to habitat fragmentation per se—in other words, significant responses to fragmentation independent of the effects of habitat amount (hereafter referred to as habitat fragmentation). I asked these two questions: Are most significant responses to habitat fragmentation negative or positive? And do particular attributes of species or landscapes lead to a predominance of negative or positive significant responses? I found 118 studies reporting 381 significant responses to habitat fragmentation independent of habitat amount. Of these responses, 76% were positive. Most significant fragmentation effects were positive, irrespective of how the authors controlled for habitat amount, the measure of fragmentation, the taxonomic group, the type of response variable, or the degree of specialization or conservation status of the species or species group. No support was found for predictions that most significant responses to fragmentation should be negative in the tropics, for species with larger movement ranges, or when habitat amount is low; most significant fragmentation effects were positive in all of these cases. Thus, although 24% of significant responses to habitat fragmentation were negative, I found no conditions in which most responses were negative. Authors suggest a wide range of possible explanations for significant positive responses to habitat fragmentation: increased functional connectivity, habitat diversity, positive edge effects, stability of predator–prey/host–parasitoid systems, reduced competition, spreading of risk, and landscape complementation. A consistent preponderance of positive significant responses to fragmentation implies that there is no justification for assigning lower conservation value to a small patch than to an equivalent area within a large patch—instead, it implies just the opposite. This finding also suggests that land sharing will usually provide higher ecological value than land sparing.

Ecological networks have a long history in ecology, and a recent increase in network analyses across environmental gradients has revealed important changes in their structure, dynamics, and functioning. These changes can be broadly grouped according to three nonexclusive mechanisms: (a) changes in the species composition of the networks (driven by interaction patterns of invaders, nonrandom extinction of species according to their traits, or differences among species in population responses across gradients); (b) changes that alter interaction frequencies via changes in search efficiency (driven by altered habitat structure or metabolic rates) or changes in spatial and temporal overlap; and (c) changes to coevolutionary processes and patterns. Taking spatial and temporal processes into account can further elucidate network variation and improve predictions of network responses to environmental change. Emerging evidence links network structure to ecosystem functioning; however, scaling up to metanetworks or multilayer networks may modify interpretations of network structure, stability, and functioning.

The use of artificial lighting to illuminate the night has provided substantial benefits to humankind. It has also disrupted natural daily, seasonal, and lunar light cycles as experienced by a diversity of organisms, and hence it has also altered cues for the timings of many biological activities. Here we review the evidence for impacts of artificial nighttime lighting on these timings. Although the examples are scattered, concerning a wide variety of species and environments, the breadth of such impacts is compelling. Indeed, it seems reasonable to conclude that the vast majority of impacts of artificial nighttime lighting stem from effects on biological timings. This adds support to arguments that artificial nighttime lighting has a quite pervasive and marked impact on ecological systems, that the rapid expansion in the global extent of both direct illuminance and skyglow is thus of significant concern, and that a widespread implementation of mitigation measures is required.

Resolving the genealogy of life—the phylogenetic relationships that describe the evolutionary history of species—remains one of the great challenges of systematic biology. The recent proliferation of DNA sequencing technologies has sparked a rapid increase in the volume of genetic data being applied to phylogenetic studies. Single nucleotide polymorphism (SNP) data, ubiquitous genetic markers once considered reserved for population genetic studies, are now being applied in phylogenetics research at deep evolutionary timescales. The potential for SNPs to resolve contentious phylogenetic problems while researchers also investigate population demographics is promising, yet serious challenges remain with respect to data collection, assembly, modeling, and analysis. The low cost and ease of collecting SNPs suggest that they will remain an important source of genetic information for inferring phylogenies across time periods ranging from the Anthropocene to the Cretaceous.

Sexual selection plays several intricate and complex roles in the related processes of local adaptation and speciation. In some cases sexual selection can promote these processes, but in others it can be inhibitory. We present theoretical and empirical evidence supporting these dual effects of sexual selection during local adaptation, allopatric speciation, and speciation with gene flow. Much of the empirical evidence for sexual selection promoting speciation is suggestive rather than conclusive; we present what would constitute strong evidence for sexual selection driving speciation. We conclude that although there is ample evidence that sexual selection contributes to the speciation process, it is very likely to do so only in concert with natural selection.

Ongoing increases in air temperature and changing precipitation patterns are altering water temperatures and flow regimes in lotic freshwater systems, and these changes are expected to continue in the coming century. Freshwater taxa are responding to these changes at all levels of biological organization. The generation of appropriate hydrologic and water temperature projections is critical to accurately predict the impacts of climate change on freshwater systems in the coming decade. The goal of this review is to provide an overview of how changes in climate affect hydrologic processes and how climate-induced changes in freshwater habitat can impact the life histories and traits of individuals, and the distributions of freshwater populations and biodiversity. Projections of biological responses during the coming century will depend on accurately representing the spatially varying sensitivity of physical systems to changes in climate, as well as acknowledging the spatially varying sensitivity of freshwater taxa to changes in environmental conditions.

Mating in seed plants arises from interactions between plant traits and the environmental and demographic context in which individuals reside. These interactions commonly cause nonrandom mating, including selfing and promiscuous outcrossing within local neighborhoods. Shared features of seed plants, specifically immobility, hermaphroditism, and modularity, shape the essential character of mating mediated by animals, wind, and water. In addition, diverse floral strategies promote cross- and self-mating, depending on environmental circumstances. Extrinsic ecological factors influence all stages of the mating process—pollination, pollen-tube growth, ovule fertilization—as well as seed development, determining offspring quantity and quality. Traditionally, measures of plant mating systems have focused on a single axis of variation, the maternal outcrossing rate. Instead, we argue for an expanded perspective encompassing mating portfolios, which include all offspring to which individuals contribute genetically as maternal or paternal parents. This approach should expose key ecological determinants of mating-system variation and their evolutionary consequences.

Phenology is a key aspect of plant and animal life strategies that determines the ability to capture seasonally variable resources. It defines the season and duration of growth and reproduction and paces ecological interactions and ecosystem functions. Phenology models have become a key component of models in agronomy, forestry, ecology, and biogeosciences. Plant and animal process-based phenology models have taken different paths that have so far not crossed. Yet, they share many features because plant and animal annual cycles also share many characteristics, from their stepwise progression, including a resting period, to their dependence on similar environmental factors. We review the strengths and shortcomings of these models and the divergences in modeling approaches for plants and animals, which are mostly due to specificities of the questions they tackle. Finally, we discuss the most promising avenues and the challenges phenology modeling needs to address in the upcoming years.

How ecological niche breadth evolves is central to adaptation and speciation and has been a topic of perennial interest. Niche breadth evolution research has occurred within environmental, ecological, evolutionary, and biogeographical contexts, and although some generalities have emerged, critical knowledge gaps exist. Performance breadth trade-offs, although long invoked, may not be common determinants of niche breadth evolution or limits. Niche breadth can expand or contract from specialist or generalist lineages, and so specialization need not be an evolutionary dead end. Whether niche breadth determines diversification and distribution breadth and how niche breadth is partitioned among individuals and populations within a species are important but particularly understudied topics. Molecular genetic and phylogenetic techniques have greatly expanded understanding of niche breadth evolution, but field studies of how niche breadth evolves are essential for providing mechanistic details and allowing the development of comprehensive theory and improved prediction of biological responses under global change.

Hybrid zones provide a powerful opportunity to analyze ecological and evolutionary interactions between divergent lineages. As such, research on hybrid zones has played a prominent role in the fields of evolutionary biology and systematics. Herein, we clarify what hybrid zones are, what is (and is not) known about them, and how different types of genomic data contribute to our understanding of hybrid zones. We then review two key topics, namely, what genomic analyses of hybrid zones have revealed about the basis and dynamics of speciation and how hybrid zones directly affect evolutionary processes. In the latter case, we emphasize the importance of contingency and ecological and genomic context in outcomes of hybridization. Throughout, we highlight limitations and key unknowns, and suggest approaches most likely to advance our understanding of hybrid zones and evolutionary processes in general.

The archipelagos that form the transition between Asia and Australia were immortalized by Alfred Russel Wallace's observations on the connections between geography and animal distributions, which he summarized in what became the first major modern biogeographic synthesis. Wallace traveled the island region for eight years, during which he noted the marked faunal discontinuity across what has later become known as Wallace's Line. Wallace was intrigued by the bewildering diversity and distribution of life he discovered. But even today we ask ourselves how species formed within the region and why they are not evenly distributed. Biogeography, phylogeny, dispersal, biotic interactions, and abiotic factors affect the assembly of diversity. On the basis of a decade of research on the ecology, evolution, and systematics of corvoid passerine birds, we summarize what we have learned about the biogeography and assembly of island bird diversity. Corvoid passerine birds include nearly 800 species and 2,300 named taxa and thus represent a large, well-described, and globally distributed clade. Understanding the processes influencing biodiversity in this group is certain to deepen our general understanding of ecology and evolution in the context of biogeography and faunal assembly.

It seems improbable that a thin veneer of attached algae coating submerged surfaces in lakes and rivers could be the foundation of many freshwater food webs, but increasing evidence from chemical tracers supports this view. Attached algae grow on any submerged surface that receives enough light for photosynthesis, but animals often graze attached algae down to thin, barely perceptible biofilms. Algae in general are more nutritious and digestible than terrestrial plants or detritus, and attached algae are particularly harvestable, being concentrated on surfaces. Diatoms, a major component of attached algal assemblages, are especially nutritious and tolerant of heavy grazing. Algivores can track attached algal productivity over a range of spatial scales and consume a high proportion of new attached algal growth in high-light, low-nutrient ecosystems. The subsequent efficient conversion of the algae into consumer production in freshwater food webs can lead to low-producer, high-consumer biomass, patterns that Elton (1927) described as inverted trophic pyramids. Human perturbations of nutrient, sediment, and carbon loading into freshwaters and of thermal and hydrologic regimes can weaken consumer control of algae and promote nuisance attached algal blooms.

The trophic cascade has emerged as a key paradigm in ecology. Although ecologists have made progress in understanding spatial variation in the strength of trophic cascades, temporal variation remains relatively unexplored. Our review suggests that strong trophic cascades are often transient, appearing when ecological conditions support high consumer abundance and rapidly growing, highly edible prey. Persistent top-down control is expected to decay over time in the absence of external drivers, as strong top-down control favors the emergence of better-defended resources. Temporal shifts in cascade strength—including those driven by contemporary global change—can either stabilize or destabilize ecological communities. We suggest that a more temporally explicit approach can improve our ability to explain the drivers of trophic cascades and predict the impact of changing cascade strength on community dynamics.

The extensive postglacial mammal losses in the West Indies provide an opportunity to evaluate extinction dynamics, but limited data have hindered our ability to test hypotheses. Here, we analyze the tempo and dynamics of extinction using a novel data set of faunal last-appearance dates and human first-appearance dates, demonstrating widespread overlap between humans and now-extinct native mammals. Humans arrived in four waves (Lithic, Archaic, Ceramic, and European), each associated with increased environmental impact. Large-bodied mammals and several bats were extinct by the Archaic, following protracted extinction dynamics perhaps reflecting habitat loss. Most small-bodied rodents and lipotyphlan insectivores survived the Ceramic, but extensive landscape transformation and the introduction of invasive mammals following European colonization caused further extinctions, leaving a threatened remnant fauna. Both large- and small-bodied nonvolant mammals disappeared, reflecting complex relationships between body size, ecology, and anthropogenic change. Extinct bats were generally larger species, paralleling declines from natural catastrophes.

Spatial processes underlie major species coexistence mechanisms. A range of spatial analysis techniques are increasingly applied to data of fully mapped communities to quantify spatial structures in species and phylogenetic and functional diversity at some given spatial scale with the goal of gaining insights into processes of community assembly and dynamics. We review these techniques, including spatial point pattern analysis, quadrat-based analyses, and individual-based neighborhood models, and provide a practical roadmap for ecologists in the analysis of local spatial structures in species and phylogenetic and functional diversity. We show how scale-dependent metrics of spatial diversity can be used in concert with ecological null models, statistical models, and dynamic community simulation models to detect spatial patterns, reveal the influence of the biotic neighborhood on plant performance, and quantify the relative contribution of species interactions, habitat heterogeneity, and stochastic processes to community assembly across scale. Future works should integrate these approaches into a dynamic spatiotemporal framework.

By facilitating plant reproduction, pollinators perform a crucial ecological function that supports the majority of the world's plant diversity, and associated organisms, and a significant fraction of global agriculture. Thus, pollinators are simultaneously vital to supporting both natural ecosystems and human food security, which is a unique position for such a diverse group of organisms. The past two decades have seen unprecedented interest in pollinators and pollination ecology, stimulated in part by concerns about the decline of pollinator abundance and diversity in some parts of the world. This review synthesizes what is currently understood about the taxonomic diversity of organisms that are known to act as pollinators; their distribution in both deep time and present space; the importance of their diversity for ecological function (including agro-ecology); changes to diversity and abundance over more recent timescales, including introduction of non-native species; and a discussion of arguments for conserving their diversity.

Nervous systems are among the most spectacular products of evolution. Their provenance and evolution have been of interest and often the subjects of intense debate since the late nineteenth century. The genomics era has provided researchers with a new set of tools with which to study the early evolution of neurons, and recent progress on the molecular evolution of the first neurons has been both exciting and frustrating. It has become increasingly obvious that genomic data are often insufficient to reconstruct complex phenotypes in deep evolutionary time because too little is known about how gene function evolves over deep time. Therefore, additional functional data across the animal tree are a prerequisite to a fuller understanding of cell evolution. To this end, we review the functional modules of neurons and the evolution of their molecular components, and we introduce the idea of hierarchical molecular evolution.

Evolutionary biologists often predict the outcome of natural selection on an allele by measuring its effects on lifetime survival and reproduction of individual carriers. However, alleles affecting traits like sex, evolvability, and cooperation can cause fitness effects that depend heavily on differences in the environmental, social, and genetic context of individuals carrying the allele. This variability makes it difficult to summarize the evolutionary fate of an allele solely on the basis of its effects on any one individual. Attempts to average over this variability can sometimes salvage the concept of fitness. In other cases, evolutionary outcomes can be predicted only by considering the entire genealogy of an allele, thus limiting the utility of individual fitness altogether. We describe a number of intriguing new evolutionary phenomena that have emerged in studies that explicitly model long-term lineage dynamics and discuss implications for the evolution of infectious diseases.

Soil organic matter (SOM) anchors global terrestrial productivity and food and fiber supply. SOM retains water and soil nutrients and stores more global carbon than do plants and the atmosphere combined. SOM is also decomposed by microbes, returning CO2, a greenhouse gas, to the atmosphere. Unfortunately, soil carbon stocks have been widely lost or degraded through land use changes and unsustainable forest and agricultural practices. To understand its structure and function and to maintain and restore SOM, we need a better appreciation of soil organic carbon (SOC) saturation capacity and the retention of above- and belowground inputs in SOM. Our analysis suggests root inputs are approximately five times more likely than an equivalent mass of aboveground litter to be stabilized as SOM. Microbes, particularly fungi and bacteria, and soil faunal food webs strongly influence SOM decomposition at shallower depths, whereas mineral associations drive stabilization at depths greater than ∼30 cm. Global uncertainties in the amounts and locations of SOM include the extent of wetland, peatland, and permafrost systems and factors that constrain soil depths, such as shallow bedrock. In consideration of these uncertainties, we estimate global SOC stocks at depths of 2 and 3 m to be between 2,270 and 2,770 Pg, respectively, but could be as much as 700 Pg smaller. Sedimentary deposits deeper than 3 m likely contain >500 Pg of additional SOC. Soils hold the largest biogeochemically active terrestrial carbon pool on Earth and are critical for stabilizing atmospheric CO2 concentrations. Nonetheless, global pressures on soils continue from changes in land management, including the need for increasing bioenergy and food production.