in all months, and mean precipitation increased in most months (Fig. 2a). 68Spatial variability in climatic change (Fig. 2b,c), necessitates local matching of phenological 69 and climatic datasets rather than the use of regionally-averaged climate data (e.g. Central 70England Temperatures) or large-scale climatic indicators (e.g. North Atlantic Oscillation). 71We did not make the restrictive assumption that biological events would be related to annual CSP precip varied less among trophic levels than the upper limit (Fig. 3d,f) consumers were less than those for primary consumers (Fig. 5a). This occurred because, 195averaged across species, the opposing climate responses of primary producers and secondary 196consumers are more similar in magnitude than are those for primary consumers (Fig. 3), 197 effectively "cancelling each other out". Our models suggest greater average advances for 198 crustacea, fish and insects than for other groups, such as freshwater phytoplankton, birds and 199 mammals (Fig. 5b). However, response-variation is high for crustacea (Fig. 5b). not estimated for marine plankton data (see above), and so the second-phase LME models 441 were run twice: once to examine correlations with temperature and precipitation for all but 442 the marine plankton phenological series (9,800 series), and once to examine only correlations 443 with temperature for the whole data set (10,003 series).

Assessing species' vulnerability to climate change is a prerequisite for developing effective strategies to conserve them. The last three decades have seen exponential growth in the number of studies evaluating how, how much, why, when, and where species will be impacted by climate change. We provide an overview of the rapidly developing field of climate change vulnerability assessment (CCVA) and describe key concepts, terms, steps and considerations. We stress the importance of identifying the full range of pressures, impacts and their associated mechanisms that species face and using this as a basis for selecting the appropriate assessment approaches for quantifying vulnerability. We outline four CCVA assessment approaches, namely trait-based, correlative, mechanistic and combined approaches and discuss their use. Since any assessment can deliver unreliable or even misleading results when incorrect data and parameters are applied, we discuss finding, selecting, and applying input data and provide examples of open-access resources. Because rare, small-range, and declining-range species are often of particular conservation concern while also posing significant challenges for CCVA, we describe alternative ways to assess them. We also describe how CCVAs can be used to inform IUCN Red List assessments of extinction risk. Finally, we suggest future directions in this field and propose areas where research efforts may be particularly valuable.

Shifts in species' distribution and abundance in response to climate change have been well documented, but the underpinning processes are still poorly understood. We present the results of a systematic literature review and meta-analysis investigating the frequency and importance of different mechanisms by which climate has impacted natural populations. Most studies were from temperate latitudes of North America and Europe; almost half investigated bird populations. We found significantly greater support for indirect, biotic mechanisms than direct, abiotic mechanisms as mediators of the impact of climate on populations. In addition, biotic effects tended to have greater support than abiotic factors in studies of species from higher trophic levels. For primary consumers, the impact of climate was equally mediated by biotic and abiotic mechanisms, whereas for higher level consumers the mechanisms were most frequently biotic, such as predation or food availability. Biotic mechanisms were more frequently supported in studies that reported a directional trend in climate than in studies with no such climatic change, although sample sizes for this comparison were small. We call for more mechanistic studies of climate change impacts on populations, particularly in tropical systems.