It is predicted that climate change will cause species extinctions and distributional shifts in coming decades, but data to validate these predictions are relatively scarce. Here, we compare recent and historical surveys for 48 Mexican lizard species at 200 sites. Since 1975, 12% of local populations have gone extinct. We verified physiological models of extinction risk with observed local extinctions and extended projections worldwide. Since 1975, we estimate that 4% of local populations have gone extinct worldwide, but by 2080 local extinctions are projected to reach 39% worldwide, and species extinctions may reach 20%. Global extinction projections were validated with local extinctions observed from 1975 to 2009 for regional biotas on four other continents, suggesting that lizards have already crossed a threshold for extinctions caused by climate change.

Climate warming is known to have effects on population dynamics through variations in survival, fecundity and density. However, the impacts of climate change on population composition are still poorly documented. Morphotypes are powerful markers to track changes in population composition. In the common lizard, Lacerta vivipara, individuals display two types of dorsal patterns: reticulated (R individuals) and linear (L individuals). We examined how local warming affected intrapopulation frequencies of these morphotypes across 11 years. We observed changes in morph frequency of dorsal patterns across years, paralleling the rise of spring temperatures. The proportion of R individuals increased with June temperatures in juveniles, yearlings, and adult males and females. Three mechanisms could explain these changes: phenotypic plasticity, microevolution and/or dispersal between populations. We investigated the ontogenetic determinism, fitness and recruitment rates associated with dorsal morphotypes. Dorsal pattern ontogeny showed temperature dependence but this relationship was not associated with the warming trend during this study. We found variation by morphotype in survival and clutch size, but these factors did not explain R frequency increases. Among all the parameters considered in this study, only a decrease of immigration, which was more pronounced in the L morphotype, could explain the change in population composition. To our knowledge, this provides the first evidence of the impact of climate warming on population composition due to its effects on immigration.

Climate change affects all levels of biology and is a major threat for biodiversity. Hence, it is fundamental to run biodiversity monitoring programs to understand the effects of climate change on the biota and to be able to adjust management and conservation accordingly. So far, however, very few existing monitoring programs allow for the detection of climate change effects, as shown by a survey undertaken by the European project EuMon. Despite this shortcoming, several methods exist which allow to make inferences from existing data by integrating data across different monitoring programs: correlative analyses, meta-analyses and models. In addition, experiments are thought to be useful tools to understand the effects of climate change on plants and animals. Here, we evaluate the utility of these four main approaches. All these methods allow to evaluate long term effects of climate change and make predictions of species' future development, but they are arguable. We list and compare their benefits and inconveniences, which can lead to uncertainties in the extrapolation of species responses to global climate change. Individual characteristics and population parameters have to be more frequently monitored. The potential evolution of a species should be also modelled, to extrapolate results across spatial and temporal scales as well as to analyse the combined effects of different climatic and biotic factors, including intra but also interspecific relationships. We conclude that a combination of methodologies would be the most promising tool for the assessment of biological responses to climate change, and we provide some thoughts about how to do so. Particularly, we encourage long-term studies along natural gradients (altitudinal or latitudinal) on the same species/habitats to be able to extrapolate to large geographic scales, and to have more complete data sets, necessary to understand the mechanisms of responses. Such data may provide a more accurate base for simulations across spatial and temporal scales, especially if they are publicly available in a common database. These recommendations could allow the adaptation of species management and the development of conservation tools to climate change which threatens species.