Tackling local ecological homogeneity: Finding intraspecific trait variability in local populations of Mediterranean plants

Local homogeneity, in ecology, is the often undisclosed assumption that variability within populations is negligible or mostly distributed evenly. In large areas, this can lead to the aggregation of different populations without regard for their unique needs and characteristics, such as drought sensitivity and functional traits distributions. Here we discuss whether this assumption can be justified, and we hypothesize that discerning the source of variation between plasticity and adaptation could be a feasible approach to formulate an informed decision. We test this hypothesis on plants, resorting to a common garden experiment to determine the source of variation of several plant functional traits at a local scale (~60 Km) of three wild species: Quercus ilex, Pistacia lentiscus and Cistus salviifolius. Individuals of each species were sourced from three key sites chosen along a local aridity gradient. Our approach led to the rejection of the local homogeneity assumption for Q. ilex and C. salviifolius at this scale due to the adaptive divergence observed among neighbouring populations. This case study provides evidence that addressing local homogeneity can highlight diverging populations in a relatively simple way. We conclude that gathering empirical evidence on intraspecific variability is a feasible approach that can provide researchers with solid bases to decide whether to adopt the local homogeneity assumption or not.

Methods
This data were obtained over the course of two sampling campaigns.

The first campaign happened in June-August 2021 in situ, as samples were collected directly from adult individuals from three populations of the Latium region in central Italy, namely Castel Fusano (41°43'23.6" N, 12°19'55.7" E), La Farnesiana (42°11'38.9" N, 11°52'33.1" E) and Tivoli (41°57'51.5" N, 12°48'54.9" E).

The second campaign happened in June-August 2022 ex situ, as measurements were obtained from seedlings grown in a common garden from seeds collected at the original provenance sites.

We followed the same standard procedures described by Perez-Harguindeguy et al. (2016), "Corrigendum to: New handbook for standardised measurement of plant functional traits worldwide." (Australian Journal of botany, 64(8), 715-716), to directly measure plant height (H), fresh leaf area (LA), and leaf thickness (LT). Leaf mass was measured at collection (FM), after 48h at 5 °C in the dark (SM) and after being kept at 90 °C until costant weight was reached (DM). Stomatal density (SD) was obtained using leaf gel impressions and digital microscopy.

All the aforementioned measures were used to obtain the traits presented in this dataset: LMA  (mgcm-2, LMA = DM/LA), LDMC (mgmg-1, LDMC= DM/SM), LTD (mgcm-3 LTD= LMA/LT), SAI (n° of stomata*mm-2, SAI= SD*stomatal length). H (cm) and SD (n° of stomata*mm-2) were also directly included in the dataset.

Dixon test was used to highlight potential outliers, which are flagged with the red fields in the dataset file.