Deschampsia antarctica has paracytic stomas, a structure that is common in Poaceae (Abid et al. 2007; López and Devesa 1991; Sanchez Anta et al. 1988; Dahlgren et al. 1985; Finot et al. 2006; Zarinkamar 2006). Previous works on this species have shown that D. antarctica is susceptible to factors such as snow cover, soil type, and temperature rise (Lewis Smith 1994; Parnikoza et al. 2007, 2011; Romero et al. 1999). In all the studied sites, our observations showed that the stomas were distributed on both sides of the leaves (amphystomatic), with high numbers on the adaxial side, as were observed by Romero et al. (1999) and Giełwanowska et al. (2005) in plants from other sites of the Antarctic Peninsula. However, in some sampled areas, plants showed leaves with stomas only present on the adaxial side (Barcikowski et al. 2003).
The results obtained in the Peñón 7 plants (Table 2) differ from those observed by other authors in other areas of the Antarctic Peninsula, for example, in Robert Island the SD of the adaxial side epidermis was: 170.10 (number of stomas per mm2) and in the abaxial side epidermis was: 382.50 (number of stomas per mm2) (Alberdi et al. 2004); while in the vicinity of the Polish Station Arctowski (Admiralty Bay, 25 de Mayo Island), the SD in the adaxial epidermis was: 5.56 (number of stomas per mm2) (Barcikowski et al. 2003). These differences could be due to morphological variations between populations and subpopulations. Therefore, we consider that SD values should only be used as a reference for the specific sites studied. Morales Rodríguez et al. (2016) observed that the decrease in SD was related to the expansion of the leaflet as the leaf ages. Peñón 7 is an area free of anthropogenic impact where leaves can age and expand and therefore a low SD is observed. The plants in the areas near and at Carlini Station were exposed to different types of stress (fuel, vapors, trampling) and, as a form of protection, they expanded their leaves less.
Our results showed a clear trend towards an increase in SD, mainly on the abaxial side of the leaves (Table 1). The increase of stomas could cause a failure in the protection of the leaves as it would make them more susceptible favoring, for example, the pathogens entry (Morales Rodríguez et al. 2016), a fact that was also reported for other species (Husen and Iqbal 1999; Bruno et al. 2007; Fusaro et al. 2015). The morphological differences, observed among the plant leaves from the different sites, would not be of genetic origin, since the populations and subpopulations had a low rate of genetic differentiation and were distinguished using specific molecular markers (Holderegger et al. 2003; Chwedorzewska et al. 2004, 2008; Rabokon et al. 2019). In summary, the variation in SD among the impacted sites would then form a plastic response to the present stress conditions (Alberdi et al. 2004; Giełwanowska et al. 2005).
The increase in SD recorded in the plant leaves of the Supply Zone was due to the effect of continuous trampling through the periods of unloading food and fuel during the summer months, a fact that continuously occurred since the installation year of the Jubany (Carlini) Scientific Station (1953). This effect was similar to that observed by Lewis Smith (1988) on Signy Island (South Orkney) in response to trampling by a natural biological agent, the Sea Lion Arctocephalus gazella. As a result of the trampling in the supply area, the vegetation cover has disappeared and, besides, the effect of the melting snow further erodes this transit area. Jägerbrand and Alatalo (2015) reported that trampling, even with low frequency, produces alterations in the ecosystem.
Exposure to anthropogenic activity produces effects on plants that are reflected in altered SD and SI values, which, also, depending on leaf age, plant type, and species, among other parameters (Pourkhabbaz et al. 2010; Kardel et al. 2010). Occasional fuel spills, which may occur mainly when the Scientific Station is restocked, would be directly responsible for the damage caused on the leaves. Therefore, an increase in SD, chlorosis, and the formation of small mats was observed in the leaves, but without epidermal rupture or reduction in the size of the stomas, as was observed in other species such as Sorghum bicolor L. (Komolafe et al. 2015).
We did not find significant differences in SI among all the sampled sites (Table 1). This parameter would be influenced by the incidence of sunlight during leaf development (Schoch et al. 1980) and, therefore, was not affected by anthropogenic impact. Although leaf size was not measured, the high positive correlation between SD and SI (r = 0.92, p < 0.00) recorded in Peñón 7 indicated that leaves were large in that area. The decrease in correlation at the impacted sites would be associated with a decrease in leaf size which would provide an adaptive advantage in D. antarctica by protecting its leaves from non-specific cell damage (Ferriol et al. 2004).
The somatic aperture is associated with physiological factors such as K+ and Ca2+ ion regulation, xylem pH variations, CO2 concentration and osmotic potential regulation (Ψω) among others (Dayanandan and Kaufman 1975; Suarez Moya and Fernández González 1984; Cadena Iñiguez et al. 2001; Eisenach and De Angeli 2017). We did not find differences in the opening of the leaves´ stomas from plants collected at the four studied sites. This would be explained by the phenotypic plasticity of the leaves of D. antarctica leaves that allowed it to acclimatize to a micro-environment by varying the number and distribution of stomas.
The present work showed that the correlation between SD and SI on the adaxial side of the leaves could be a good biomarker for the anthropogenic impact estimation since they did not experience morphological changes. Therefore, the data obtained could be used as reference values to adopt, in time, the appropriate strategies to avoid vegetation loss and changes in the surveyed areas.