As the single opportunity for plants to move, seed dispersal has an important impact on plant fitness, species distributions and patterns of biodiversity. However, models that predict dynamics such as risk of extinction, range shifts and biodiversity loss tend to rely on the mean value of parameters and rarely incorporate realistic dispersal mechanisms. By focusing on the mean population value, variation among individuals or variability caused by complex spatial and temporal dynamics is ignored. This calls for increased efforts to understand individual variation in dispersal and integrate it more explicitly into population and community models involving dispersal. However, the sources, magnitude and outcomes of intraspecific variation in dispersal are poorly characterized, limiting our understanding of the role of dispersal in mediating the dynamics of communities and their response to global change. In this manuscript, we synthesize recent research that examines the sources of individual variation in dispersal and emphasize its implications for plant fitness, populations and communities. We argue that this intraspecific variation in seed dispersal does not simply add noise to systems, but, in fact, alters dispersal processes and patterns with consequences for demography, communities, evolution and response to anthropogenic changes. We conclude with recommendations for moving this field of research forward.

Under future climate change, plant species are expected to shift their ranges in response to increasing temperatures and altered precipitation patterns. As seeds represent the single opportunity for plants to move, it is critical to quantify the factors that influence reproduction. While total seed production is clearly important, seed quality is equally as critical and often overlooked. Thus, to quantify how environmental and tree‐level characteristics affect seed quality and quantity, the reproductive output of red maple (Acer rubrum) was measured along an elevation gradient in the Monongahela National Forest, WV. A variety of individual‐level characteristics were measured (e.g., DBH, canopy area, height, and tree cores were taken to quantify growth), and seed traps were placed under seed‐bearing trees to collect samaras and quantify total seed production. A random subsample of collected seeds from each tree was micro‐CT scanned to determine embryo volume, photographed for morphology measurements, and used for germination trials. The number of seeds produced was negatively affected by frost events during flowering, and stand density. The trees with the most seeds also showed reduced growth in recent years. Only 63% of scanned seeds showed embryo development, and of those seeds—only 23% germinated. The likelihood of embryo presence increased as growth rate decreased, while embryo size increased with tree height, smaller DBH, and in areas dominated by hemlock. Both larger embryo volume and larger overall seed size increased the likelihood of germination. The results highlight the importance of including seed quality in addition to seed quantity for a more complete representation of reproductive output.

Changes of grazing pressure and climate have been identified as drivers of woody encroachment in subalpine pastures, but disentangling their relative importance remains challenging. It is particularly unclear if grazing alone can mitigate potential climate change impacts on encroachment. We used the dynamic vegetation model LandClim to simulate how multiple biotic and abiotic factors influence the rate of woody encroachment (i.e. how changes in temperature and precipitation influence woody and herbaceous vegetation, and interactions with grazing). We simulated a 61‐ha subalpine pasture in Switzerland near tree line. Currently, 21% of the pasture is covered by woody plants. If current grazing levels are maintained, almost all climate change scenarios showed an increase in woody cover. Under the RCP 4.5 and RCP 8.5 scenarios, there was a mean encroachment rate of 0.24% and 0.15% per year respectively. Large variation among climate model chains with RCP 8.5 was found, due to variations in precipitation. Generally, decreasing precipitation caused woody encroachment to slow down, even under very low grazing pressures. Using a sensitivity analysis with all possible combinations of grazing pressure, temperature and precipitation changes, we found that woody encroachment rates were most sensitive to changes in grazing pressure. Precipitation change was the second most important factor, while temperature changes were least important. Synthesis and applications. Modelling shows that increasing grazing pressure in subalpine pastures can strongly reduce woody encroachment and mitigate climate change impacts on woody encroachment.