Grazing on arid and semi-arid lands is often the only livelihood for many farming families in these adverse environments [1–3]. Although grazing on natural vegetation significantly alters its structure and functionality [4,5], the effects of herder-monitored grazing on ecosystems in high Andean regions, such as the Argentine Puna, remain poorly understood. Therefore, it is crucial to generate data that enriches the global theoretical framework of grazing ecology, enabling the development of management strategies that enhance the sustainability of pastoral systems and can ultimately be integrated into public policies.

Like many high-altitude environments [6], extreme conditions such as intense solar radiation, temperature fluctuations, strong winds, and limited rainfall prevail in the Argentine Puna. Despite these challenges, pastoral communities have long inhabited the area, sometimes by generating exchange relations with other neighboring communities, agro-pastoralists, or pure agriculturalists, within a framework of mutual benefits [7,8]. Their livelihood relies on the mobile and fluctuating breeding of domestic animals, primarily sheep and llamas (L. glama), which graze on the native vegetation. These native pastoralists operate within a production logic that integrates various environmental, socio-economic, and historical factors, probably resulting in highly efficient production systems tailored to these harsh conditions [9,10]. However, the relationship between historical livestock grazing practices and the high Andes plant communities, which form the backbone of pastoral activity, remains uncertain.

Various theoretical and methodological approaches have examined the relationship between natural vegetation and grazing patterns. In arid and semi-arid environments, livestock congregates around water sources and corrals, creating a gradient of grazing pressure known as the piosphere pattern. This pattern entails decreasing grazing pressure as the distance from the focal point increases [11,12]. While the piosphere pattern approach has been used in different rangelands across Argentina, such as the Chaco Seco and the Monte phytogeographical provinces [12,13], its applicability in the Puna remains unexplored. However, given that this approach overlooks other potentially significant factors, such as livestock species, herd size, grazing rotation frequency, duration of grazing, and seasonal variations, it may be necessary to explore the relationship between livestock pressure and vegetation using alternative approaches that incorporate this variability in grazing regimes [14]. To quantify the potential effect of grazing pressure on vegetation, the increasing recognition of species with similar responses to disturbances coexisting in communities has led to a growing use of the functional classification of plants [15,16]. This approach considers that phylogenetically distant organisms subjected to similar selective pressures develop comparable morphological, anatomical, and physiological traits [17]. Consequently, changes in the proportion of different plant functional types are expected to reflect alterations in ecosystem functionality in response to grazing pressure [18–20]. However, research on the response of plant functional types to grazing pressure in Argentine Puna rangelands is currently lacking.

In this study, we investigate the relationship between grazing pressure and plant functional types in two ecological sites within the aboriginal community of Suripujio (Jujuy province, Puna of Argentina). We employed two methodological approaches to address the following questions: (a) How does the plant community (plant functional types richness and cover) vary in proximity to or remoteness from the corrals, which function as high-pressure points? (b) What is the relationship between plant functional types richness and cover and a Grazing Pressure Index that encompasses the diversity of grazing regimes? (c) Do the general outcomes derived from employing the piosphere pattern approach (a) and the Grazing Pressure Index characterizing grazing regimes (b) exhibit consistency? Addressing these questions is essential for enhancing our understanding of the dynamics between grazing pressure and plant functional types, thereby informing more effective management strategies for sustainable pastoral systems in the high Andean region of Argentina.

Our study pioneers the investigation of plant functional type diversity and abundance about grazing in the Puna highland ecosystems. The results varied depending on the ecological site and methodological approach. The piosphere pattern approach showed few vegetation changes about grazing pressure only in the Tolar ecological site. Using the GPI approach, we found more sensitivity, but in general, the results were contrary to expectations.

Based on the results obtained regarding the piosphere pattern and plant functional types, our findings indicate a few minimal differences between sectors near and far from the corral. It suggests that this approach does not robustly detect changes in vegetation cover in response to grazing pressure, although in Tolar, it shows a slightly significant variation. This uniformity in the plant community structure, irrespective of proximity to the grazing point, hints at a certain level of resilience or adaptation to grazing pressure in the Ciénego ecological site. Conversely, in the Tolar ecological site, we noted a trend toward higher total cover in sectors closer to the corral than those farther away. Although this observed increase is marginally significant, it hints at the potential influence of grazing pressure on vegetation composition and distribution within this site. Contrary to what was expected by the piosphere pattern, this result could be explained by a positive effect of fertilization from animal dung, as we found a higher frequency of dung in plots near the corrals in Tolar (Fig. S1). Similar findings support this explanation, as it has been found that the presence of manure from both exotic and native animals, can increase plant diversity and cover [12,26–29]. However, it is important to highlight that while total cover tended to augment in closer sectors, the remaining response variables in Tolar exhibited no discernible changes between distances from the corral.

The relationships became more complex when other factors related to the grazing regime were incorporated into the GPI. At the Ciénego ecological site, grazing pressure seems to be positively correlated with the total cover and the cover of grasses and graminoids. Similar to the explanation provided for the piosphere pattern, these results could be attributed to the effect of organic fertilization provided by animal manure (as higher GPI values are expected to correspond to greater dung frequency) and/or the removal of soil crusts in these deserts or semidesert environments, creating microsites that are subsequently colonized by native vegetation, especially grasses, and graminoids [26]. Conversely, at the Tolar ecological site, total cover and cover of herbaceous dicotyledons decreased towards intermediate values of the GPI, suggesting a potential threshold effect where moderate grazing pressure may lead to reduced vegetation cover. Low vegetation cover values at both ends of the index could be explained by reduced plant consumption by livestock at low GPI values (less pressure), and likely by dung fertilization at higher GPI values [26–30]. The only negative linear response regarding the GPI was for the cover of clonal shrubs at the Tolar ecological site. However, since the shrubs comprising this plant functional type have low grazing preference, especially for sheep, another yet undisclosed mechanism should be responsible for this pattern rather than grazing pressure itself. These shrubs are frequently used as firewood, and it is expected that where there is a higher GPI (ultimately indicating greater use), the harvest of these shrubs would also increase [31]. These findings emphasize the need to enhance the construction of the GPI to include, in future studies, other aspects related to livestock management (such as collection of firewood and dung density estimation, among others).

By incorporating the heterogeneity of the grazing regime through the Grazing Pressure Index (GPI), we identified more associations than with the piosphere pattern approach. Our results highlight the complex interaction between heterogeneous forms of livestock management and vegetation dynamics. Further investigation into the underlying mechanisms driving these variations is needed to enhance our understanding of ecosystem responses to grazing in highland environments.

In summary, our study in the vicinity of Suripujio, in the highlands of the Argentinean Puna, did not find a clear and significant negative impact of grazing pressure on vegetation cover and plant functional type richness. Similarly, no significant differences were found in functional diversity between sites with different grazing intensities in semi-arid Mediterranean grasslands [32]. Although our study did not quantify functional diversity, plant functional type richness also showed no relationship with grazing pressure. This partially contrasts with the findings of other authors, who analyzed plant functional types about stocking rates and detected a decrease in the abundance of dwarf palms and short grasses with higher grazing pressure in arid or semi-arid environments in Spain [18].

The lack of response to grazing pressure for most of the variables studied could be explained by various reasons, not mutually exclusive. Firstly, grazing pressure may be moderated enough not to modify vegetation structure and functionality [5,33–35]. Secondly, given that Puna environments have evolved under a long grazing history [36], it is expected that vegetation diversity is not drastically altered by grazing pressure and that livestock grazing is compatible with biodiversity conservation [32,37–40]. Moreover, arid and semi-arid sites often exhibit minimal responses to herbivory pressure, possibly because the vegetation has evolved traits to escape drought that are often similar to traits that avoid grazing [23,41]. Additionally, the daily grazing pressure away from the corral, monitored by herders, prevents animals from remaining near the corral for extended periods. Managing grazing pressure through daily, seasonal, and inter-annual practices helps avoid the formation of overgrazed or undergrazed areas.

Furthermore, the longstanding co-evolution of pastures and herds in the Puna highlands could also explain the absence of negative responses mentioned above. This ancestral environmental management, livestock husbandry, and pastoral economy, passed down through generations, likely increases the compatibility of pastoral activities and biodiversity conservation by dispersing grazing pressure across space and time [42,43]. In this sense, the subdivision of land into units of management based on inherent landscape productivity and biophysical characteristics seems to be a conservative use of low-productive landscapes in the Puna region as other authors postulate for low-productive grasslands in Australia [44]. Thus, as in traditional Andean grazing management, the intensity and moments of grazing should be varied and utilized in a vast menu of management strategies [45].