Sampling
Sampling protocol
Environmental representativeness
In order to optimize sampling of adaptive variation and increase statistical power in detecting variants under spatially divergent selection, gene-environment association (GEA) analysis requires habitat variability to be comprehensively sampled across a landscape (Kawecki & Ebert 2004; Selmoni et al. 2020). Such a sampling approach (also known as 'environmental representativeness') poses a challenge to the experimental design of GEA studies, as it requires the identification of an appropriate geographic scale for sampling that will allow individuals living in different conditions to be genetically characterised in sufficient numbers (see next sections).
With this in mind, MedForAct will focus on the Italian peninsula (hereafter referred to as the study area), which is home to a representative set of climatic conditions experienced by the target species throughout their range. In order to respect the principle of ecological representativeness, particular attention will be paid to the selection of populations that ensure a balanced representation of the ecological niche of the target species (see figure below).
Sampling of pine needles through a telescopic pruning shear
Inspection of a stand of P. halepensis in Pantelleria (Sicily)
Sampling for genetic analysis and morphometric measurements
Spatial distribution of Pinus halepensis (a), Pinus pinaster (c) and Pinus pinea (e) as derived from the EUFORGEN database. Blue areas highlight the Italian distribution in both the geographic and environmental space (left and right columns, respectively). The ecological envelope is reported for Pinus halepensis (b), Pinus pinaster (d) and Pinus pinea (f) to differentiate between the Mediterranean bioclimates found across their distributions (Dufour‐Dror & Ertas 2004). This representation is based on wintriness (min: minimum temperature of the coldest month; x-axis) and a set of humidity categories defined by the Emberger’s pluviothermic quotient (y-axis). VC: very cold; C: cold; Cl: cool; T: temperate; H: hot; VH: very hot. 1: per-humid; 2: humid; 3: sub-humid; 4: semi-arid; 5: arid; 6: per-arid.
Sample size and age of individuals sampled
To increase the reliability of species distribution models and derived ecological envelopes, we aim to collect ≥100 spatial records per species (Wisz et al. 2008).
To optimise the statistical power of the GEA analysis, we aim to sample approximately 400 individuals per species, covering the full range of habitats encountered by the target species along the Italian peninsula (Selmoni et al. 2020). We also aim to sample a minimum of 10 adult trees per population, with 10-20 pine needles per individual sampled for genetic characterisation.
To match the temporal resolution between the environmental covariates used in GEA analysis and the sampled individuals, we aim to sample adult trees ≥50 years old (Dauphin at al. 2023).
Identification of natural populations
Mediterranean pines have been widely translocated by humans over the centuries to meet reforestation and food needs. The most notable cases include the movement of P. pinea across the Mediterranean basin since antiquity for the pine nut trade, and the extensive reforestation undertaken by national authorities in Italy during the last century. As a result, human-mediated migration may have altered the natural genetic landscape of the species by artificially influencing allele frequencies in some areas.
It is of the utmost importance to refrain from incorporating artificial stands into GEA analyses, as spurious correlations may emerge as a result of artefacts (e.g. due to past geographic translocations), particularly when foreign material has been introduced from populations with disparate demographic histories or carrying alleles that are adapted to the habitat of origin. Consequently, the statistical signal of local adaptation in such circumstances may be globally obscured, as GEA models are trained on a biased dataset, which both increases the rate of false detections and reduces statistical power.
Direct contact with local experts has been initiated to locate presumed natural populations of P. halepensis, P. pinaster and P. pinea throughout Italy. A literature review is also underway. We consider a population to be putatively natural if it fulfils the following characteristics:
Clear evidence from literature and/or local experts of historical occurrence in a given area with limited human intervention within the last 100 years.
Direct observation of diagnostic features in the field, such as the presence of different age classes and an uneven spatial distribution of individuals in the population.
Spatial database
Sampling activity is recorded in a spatial database showing the location of sampled stands (see interactive map below). Among other meta-information, each stand is characterised by:
The population name.
The geographical coordinates [longitude (°W) and latitude (°N)].
The number of individuals sampled and genotyped.
The current version of the spatial database (v.1.0) is freely available by clicking here.
To date, 47, 29 and 15 stands of Pinus halepensis (green dots), Pinus pinaster (yellow dots) and Pinus pinea (red dots) have been visited with a total of 768, 1065 and 245 pines sampled, respectively.
Reports from public bodies and private citizens
The search for natural populations is currently underway. We therefore welcome any feedback and suggestions from public/private bodies and private citizens. If you have any information you would like to share, please contact (email: elia.vajana@ibbr.cnr.it; mobile: +39 338 495 3752), and/or Dr. Andrea Piotti (email: andrea.piotti@cnr.it).
References
Dauphin, B., Rellstab, C., Wüest, R. O., Karger, D. N., Holderegger, R., Gugerli, F., & Manel, S. (2023). Re-thinking the environment in landscape genomics. Trends in Ecology & Evolution.
Dufour‐Dror, J. M., & Ertas, A. (2004). Bioclimatic perspectives in the distribution of Quercus ithaburensis Decne. subspecies in Turkey and in the Levant. Journal of Biogeography, 31(3), 461-474.
Kawecki, T. J., & Ebert, D. (2004). Conceptual issues in local adaptation. Ecology letters, 7(12), 1225-1241.
Selmoni, O., Vajana, E., Guillaume, A., Rochat, E., & Joost, S. (2020). Sampling strategy optimization to increase statistical power in landscape genomics: A simulation‐based approach. Molecular Ecology Resources, 20(1), 154-169.
Wisz, M. S., Hijmans, R. J., Li, J., Peterson, A. T., Graham, C. H., Guisan, A., & NCEAS Predicting Species Distributions Working Group. (2008). Effects of sample size on the performance of species distribution models. Diversity and distributions, 14(5), 763-773.