High-altitude wetlands of the Central Andes, locally known as bofedales, provide important ecosystem services, particularly carbon storage, forage provisioning, and water regulation. Local communities have artificially expanded bofedales by irrigating surrounding grasslands to maximise areas for alpaca grazing. Despite their importance, biophysical processes of both natural and artificial bofedales are still poorly studied, which hinders the development of adequate management and conservation strategies. We analyse and compare the vegetation composition, hydrological variables, groundwater chemistry, and soil characteristics of a natural and an artificial bofedal of at least 10 years old in southern Peru, to understand their interrelations and the consequences for ecosystem service provisioning. We do not find statistically significant differences in the soil, water, and vegetation characteristics. Soil organic carbon (SOC) content, which we use as a proxy for carbon storage, is negatively correlated to dissolved oxygen, pH, and soil water temperature. In addition, Non-Metric Multidimensional Scaling analysis shows a positive relation between plant community composition, SOC content, and water electric conductivity. Our results suggest a three-way interaction between hydrological, soil, and vegetation characteristics in the natural bofedal, which also holds for the artificial bofedal. Vegetation cover of two of the most highly nutritious species for alpaca, Lachemilla diplophylla and Lilaeopsis macloviana with 19-22% of crude protein, are weakly or not correlated to environmental variables, suggesting grazing might be obscuring these potential relationships. Given the high economic importance of alpaca breeding for local communities, expanding bofedales artificially appears an effective strategy to enhance their ecosystem services with minimal impact on the ecohydrological properties of bofedales.