Summary1. Schedules of survival, growth and reproduction are key life-history traits. Data on how these traits vary among species and populations are fundamental to our understanding of the ecological conditions that have shaped plant evolution. Because these demographic schedules determine population *Correspondence author. E-mails: salguero@demogr.mpg.de; compadre-contact@demogr.mpg.de † Joint senior author. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 2015, 103, 202-218 doi: 10.1111/1365-2745.12334 growth or decline, such data help us understand how different biomes shape plant ecology, how plant populations and communities respond to global change and how to develop successful management tools for endangered or invasive species. Journal of Ecology2. Matrix population models summarize the life cycle components of survival, growth and reproduction, while explicitly acknowledging heterogeneity among classes of individuals in the population. Matrix models have comparable structures, and their emergent measures of population dynamics, such as population growth rate or mean life expectancy, have direct biological interpretations, facilitating comparisons among populations and species. 3. Thousands of plant matrix population models have been parameterized from empirical data, but they are largely dispersed through peer-reviewed and grey literature, and thus remain inaccessible for synthetic analysis. Here, we introduce the COMPADRE Plant Matrix Database version 3.0, an opensource online repository containing 468 studies from 598 species world-wide (672 species hits, when accounting for species studied in more than one source), with a total of 5621 matrices. COMPADRE also contains relevant ancillary information (e.g. ecoregion, growth form, taxonomy, phylogeny) that facilitates interpretation of the numerous demographic metrics that can be derived from the matrices. 4. Synthesis. Large collections of data allow broad questions to be addressed at the global scale, for example, in genetics (GENBANK), functional plant ecology (TRY, BIEN, D3) and grassland community ecology (NUTNET). Here, we present COMPADRE, a similarly data-rich and ecologically relevant resource for plant demography. Open access to this information, its frequent updates and its integration with other online resources will allow researchers to address timely and important ecological and evolutionary questions.

We synthesized evidence for unintended consequences and trade‐offs associated with the passage of fishes. Provisioning of fish passageways at dams and dam removals are being carried out increasingly as resource managers seek ways to reduce fragmentation of migratory fish populations and restore biodiversity and nature‐like ecosystem services in tributaries altered by dams. The benefits of provisioning upstream passage are highlighted widely. Possible unwanted consequences and trade‐offs of upstream passage are coming to light, but remain poorly examined and underappreciated. Unintended consequences arise when passage of native and desirable introduced fishes is delayed, undone (fallback), results in patterns of movement and habitat use that reduce Darwinian fitness (e.g. ecological traps), or is highly selective taxonomically and numerically. Trade‐offs arise when passage decisions intended to benefit native species interfere with management decisions intended to control the unwanted spread of non‐native fishes and aquatic invertebrates, or genes, diseases and contaminants carried by hatchery and wild fishes. These consequences and trade‐offs will vary in importance from system to system and can result in large economic and environmental costs. For some river systems, decisions about how to manage fish passage involve substantial risks and could benefit from use of a formal, structured process that allows transparent, objective and, where possible, quantitative evaluation of these risks. Such a process can also facilitate the design of an adaptive framework that provides valuable insights into future decisions.

Significant declines in the abundance of lake sturgeon Acipenser fulvescens across most of its North American range have led to abundances less than 10% of estimated minimum sustainable population sizes and, in some locations, less than 1% of historic abundances. These precipitous declines in abundance have resulted in most lake sturgeon populations being considered a conservation concern and have prompted management actions toward recovery. Here we present modeling in support of a recovery potential assessment, using stage-structured matrix models and population viability analysis to quantitatively assess allowable harm, recovery efforts, probabilities of recovery, and recovery time frames. From this assessment, we conclude that lake sturgeon populations are most sensitive to harm on adult survival and that some designatable units are highly sensitive to any level of harm. However, the scope for recovering lake sturgeon by improving adult survival is limited; instead, larger proportional increases in population growth rates can be achieved by focusing recovery efforts on age-0 and juvenile survival. Finally, based on a recovery target of 1,188 spawning females/year for each discrete population and current abundances from 1% to 10% of this target, long-term projections indicate recovery time frames ranging from 20 to over 100 years, depending on the recovery actions implemented.

We applied elasticity analysis to 88 North American freshwater fishes to assess the relative impacts of changes in the vital rates on asymptotic population growth. Variance in vital rates was summarized for four distinct functional groups: (i) species with population growth rates strongly sensitive to perturbations in adult survival; (ii) species with population growth rates sensitive to perturbations in overall survival; (iii) species with population growth rates most sensitive to perturbations in juvenile survival; and (iv) species with population growth rates sensitive to perturbations in juvenile survival and fecundity. The results of the present study also showed that (a) elasticity patterns cannot be inferred in a straightforward manner from trade-offs between life-history traits, (b) the sensitivity of a population's growth rate to changes in adult survival and fecundity can be predicted empirically from life span and age at maturity, respectively, (c) elasticities are highly conserved among genera within the same taxonomic family, and (d) there are key divergences between elasticity patterns of freshwater fish and other vertebrate taxa.

Pacific Northwest Chinook, Oncorhynchus tshawytscha, have exhibited a high degree of variability in smoltto-adult survival over the past three decades. This variability is summarized for 22 Pacific Northwest stocks and analyzed using generalized linear modeling techniques. Results indicate that survival can be grouped into eight distinct regional clusters: (1) Alaska, Northern BC and North Georgia Strait; (2) Georgia Strait; (3) Lower Fraser River and West Coast Vancouver Island; (4) Puget Sound and Hood Canal; (5) Lower Columbia Tules; (6) Columbia Upriver Brights, Willamette and Cowlitz; (7) Oregon and Washington Coastal; and (8) Klamath River and Columbia River Summers. Further analysis for stocks within each of the eight regions indicates that local ocean conditions following the outmigration of smolts from freshwater to marine areas had a significant effect on survival for the majority of the stock groups analyzed. Our analyses of the data indicate that Pacific Northwest Chinook survival covaries on a spatial scale of 350-450 km. Lagged time series models are presented that link large-scale tropical Pacific conditions, intermediate-basin scale northeastern Pacific conditions, and local sea surface temperatures to survival of Pacific Northwest stocks.

American eel, Anguilla rostrata, is a catadromous fish that spawns in the Sargasso Sea and migrates to coastal waters and freshwater systems ranging from Greenland down the Atlantic coast to South America and has been regarded as comprising a panmictic population. American eel is in decline across much of its range. Research and management is primarily conducted at local to regional scales, yielding inconsistent research results and management recommendations that may be inhibited by large-scale processes. We review the research on American eel ecology demonstrating that its variable life history and movement patterns can be explained based on: (1) latitudinal productivity gradients; (2) ideal free habitat selection; (3) conditional evolutionarily stable life history strategies; (4) size at arrival to the coast (correlated with distance from the spawning grounds); and, (5) temperature variance and annual degree-day effects on somatic growth. Using these ecological processes, we outline how local and largescale effects on American eel dynamics can be integrated in a panmictic (or quasi-panmictic) modelling framework to enable defensible predictions of population responses to anthropogenic and oceanic phenomena. Due to its widespread distribution and existing knowledge gaps, the conservation and management of American eel will require international cooperation.

No abstract

Most introduced non-native species fail to establish as a result of mortality or reproductive failure. An established population can increase the probability of survival and reproductive success of newly introduced individuals by reducing both Allee effects and demographic stochasticity. Previously, attention has been paid to the establishment phase of the invasion process and its probability modelled as a stochastic process, while the spread phase has received less attention. By analyzing data collected during the spread phase of an invasion of the round goby, Neogobius melanostomus, in Hamilton Harbour, Lake Ontario, we develop an analytical approach to backcalculate the time to establishment and to determine the time to habitat saturation. Our modelling shows that: (1) during the transition between arrival and establishment, propagule pressure in the form of new adults entering the area can be very low and still represent a significant probability of establishment; (2) much higher concentrations of juveniles would be needed to pose a significant risk of invasion; (3) the demographic contribution of propagule pressure during the spread phase is low and its total elimination will not halt population growth and spread; (4) a short elapsed time between arrival and establishment indicated that the transition between these two phases can be characterized as a deterministic process with high propagule pressure and low adult mortality rates; and, (5) very aggressive management actions would be needed to halt population growth after population establishment, suggesting that preventative measures are the most effective management options available to reduce risk of future invasions.