Alessandro Ferrarini

Molecular characterization and identification of a group of local Olea europaea L. varieties

There is an urgent need for a rapid and accurate procedure to evaluate the degree of genetic diversity in Olea europaea L. In this research work, we used simple sequence repeat markers for the characterization and identification of the genetic profiles of a group of ancient olive trees, using clustering analysis (dendrogram analysis, Bayesian method) to estimate the genetic distance and relationships among individuals and “random forests” to evaluate the importance of the applied markers and create the differential profiles. As a result, by the use of ten microsatellite loci, we were able to separate 142 samples into homogeneous groups. Our results indicate a high genetic diversity within the group of local accessions. Most accessions seem to have a foreign origin, particularly from neighbouring zones, but a discrete number of them appear to be of unknown origin. We have expressed the differential genetic profiles of the identified groups in terms of “if-then-else” rules. This paper, after a comparison with classical methods, proposes a rigorous methodological approach to the purpose of characterizing olive trees. It also introduces for the olive the concept of differential genetic profiles as complementary to classic ones.

Planning for assisted colonization of plants in a warming world

Assisted colonization is one way of facilitating range shifts for species that are restricted in their ability to move in response to climate change. Here we conceptualize and apply a new decision framework for modelling assisted colonization of plant species prior to in situ realization. Three questions were examined: a) Is species translocation useful in a certain area? b) where, and c) how long will it be successful in the future? Applying our framework to Carex foetida in Italy at the core of its distribution and its southern edge revealed that assisted colonization could be successful in short-term (2010–2039) climate conditions, partially in medium (2040–2069) but not in long-term (2070–2099) scenarios. We show that, for some species, it is likely that assisted colonization would be successful in some portions of the recipient site under current and short-term climate conditions, but over the mid- and long-term, climate changes will make species translocation unsuccessful. The proposed decision framework can help identify species that will need different conservation actions (seed banks and/or botanical gardens) when assisted colonization is unlikely to be successful. Furthermore it has broad applicability, as it can support planning of assisted migration in mountainous areas in the face of climate change.

Braking effect of climate and topography on global change-induced upslope forest expansion

Forests are expected to expand into alpine areas due to global climate change. It has recently been shown that temperature alone cannot realistically explain this process and that upslope tree advance in a warmer scenario may depend on the availability of sites with adequate geomorphic/topographic characteristics. Here, we show that, besides topography (slope and aspect), climate itself can produce a braking effect on the upslope advance of subalpine forests and that tree limit is influenced by non-linear and non-monotonic contributions of the climate variables which act upon treeline upslope advance with varying relative strengths. Our results suggest that global climate change impact on the upslope advance of subalpine forests should be interpreted in a more complex way where climate can both speed up and slow down the process depending on complex patterns of contribution from each climate and non-climate variable.

Improving niche projections of plant species under climate change: Silene acaulis on the British Isles as a case study

Empirical works to assist in choosing climatically relevant variables in the attempt to predict climate change impacts on plant species are limited. Further uncertainties arise in choice of an appropriate niche model. In this study we devised and tested a sharp methodological framework, based on stringent variable ranking and filtering and flexible model selection, to minimize uncertainty in both niche modelling and successive projection of plant species distributions. We used our approach to develop an accurate, parsimonious model of Silene acaulis (L.) presence/absence on the British Isles and to project its presence/absence under climate change. The approach suggests the importance of (a) defining a reduced set of climate variables, actually relevant to species presence/absence, from an extensive list of climate predictors, and (b) considering climate extremes instead of, or together with, climate averages in projections of plant species presence/absence under future climate scenarios. Our methodological approach reduced the number of relevant climate predictors by 95.23% (from 84 to only 4), while simultaneously achieving high cross-validated accuracy (97.84%) confirming enhanced model performance. Projections produced under different climate scenarios suggest that S. acaulis will likely face climate-driven fast decline in suitable areas on the British Isles, and that upward and northward shifts to occupy new climatically suitable areas are improbable in the future. Our results also imply that conservation measures for S. acaulis based upon assisted colonization are unlikely to succeed on the British Isles due to the absence of climatically suitable habitat, so different conservation actions (seed banks and/or botanical gardens) are needed.