Seed release in a changing climate: initiation of movement increases spread of an invasive species under simulated climate warming

Abstract

Aim Reliable estimates of how species will move under future climatic conditions are lacking, and the ability of species to track suitable habitat under climate change is strongly predicated on dispersal. In wind‐dispersed plants, initiation of movement by maternal seed release characteristics can affect the ultimate shape of the dispersal kernel. If seed dispersal traits respond to altered local abiotic conditions induced by climate change during development, the effects could have major impacts on projections of future population spread. Location Central Pennsylvania, USA. Methods To quantify how seed release characteristics might be affected by climate change, we experimentally manipulated temperature in a field study of the wind‐dispersed, invasive plant Carduus nutans (Asteraceae). We quantified seed release in a series of trials conducted in a full‐size experimental wind tunnel, and we examined the effect of the probability of seed release on the projected population spread rate (c*). Results Experimental warming over the growing season significantly increased the probability of seed release in wind tunnel trials by 7%. When these data were incorporated into matrix integro‐differential equation models of population spread for this species, the measured increase in seed release, combined with previously reported increases in plant height, resulted in a projected increase in the population spread rate of C. nutans of 38% per year due to increased temperature. Main conclusions Ignoring altered seed release dynamics significantly underestimates the increase in potential spread rate by 11%. This increase could allow this species to reach suitable habitat more quickly than expected from existing models of population spread, and could require management timing to shift in the future. More generally, our results suggest that models of future distributions of other wind‐dispersed species that do not include information about species' dispersal responses to climate may incur significant errors and could result in poorly applied management or inadequate spatial risk assessment.

Pervasive human‐mediated large‐scale invasion: analysis of spread patterns and their underlying mechanisms in 17 of China's worst invasive plants

“…Common to most theoretical and empirical studies is the basic assumption portraying invasive spread as a spatially continuous process, characterized by the (very) low probabilities of extreme LDD (or jump dispersal) events at the dispersal kernel tail . This basic assumption prevails in both classic (Skellam 1951;Okubo et al 1989;Allen et al 1991) and contemporary literature (Adams et al 2015;Uden et al 2015;Teller, Zhang & Shea 2016). However, the findings of the current study suggest that Fig.…”

“…; Uden et al . ; Teller, Zhang & Shea ). However, the findings of the current study suggest that the occurrence of such long jumps cannot be consistently estimated based on a continuous fat‐tailed dispersal kernel.…”

How are the phenologies of ripening and seed release affected by species' ecology and evolution?

“…Likewise, Jongejans et al () and Soons and Bullock () found morphological mechanisms that facilitate short‐term synchronisation of seed release with optimal wind conditions showing low air humidity and high wind speeds (respectively). Such synchronisations of seed release timing and optimal conditions for long‐distance dispersal are increasingly considered important for the assessment of species’ ability to reach suitable habitats (Teller et al ). Our result of the existence of enhanced conditions for long‐distance seed dispersal by wind during seed release can be cautiously interpreted as a phenological adaptation i.e.…”

Moving forward: insights and applications of moving‐habitat models for climate change ecology

“…Estimating dispersal kernels may involve measuring seed shadows (Clark 1998), identifying and quantifying multiple dispersal vectors (Nathan et al 2008), measuring average lifetime displacement or converting from a measured distribution of dispersal distances to a dispersal kernel (Cousens & Rawlinson 2001). Recent advanced approaches to parameterizing IDE dispersal kernels include using climate projections to fit wind dispersal models (Bullock et al 2012), collecting onsite data on wind speed and tree canopy heights (Caplat, Nathan & Buckley 2012), incorporating complex movement of dispersal vectors such as birds (Neupane & Powell 2015), and using experimental warming to anticipate changes in seed release rates (Teller, Zhang & Shea 2016). For MHMs, it is common to select a flexible family of dispersal kernels or to rely on the type of kernel commonly used for a particular taxon.…”