Genetic diversity is the raw material for species’ persistence over time, providing the potential to survive stochastic events, as well as climate and/or human-induced environmental changes. Biodiversity in dry rangelands is decreasing due to intensification of livestock production, but its effects on the genetic diversity of the consumed biota have seldom been assessed. We examined livestock effects on the genetic diversity of two dominant creosote species of the Patagonian Monte Desert, Larrea divaricata and Larrea cuneifolia. We deployed competing hierarchical regression models to assess the relationship between genetic variation within natural populations as a function of increasing stocking rates on ten arid rangelands. These species exhibit similar levels and patterns of genetic structure, with high levels of both inbreeding and divergence among locations. We found that increased stocking reduces genetic variation and increases genetic subdivision between populations. Our results indicate that grazing pressures are impoverishing the gene pool of these dominant native species of the Monte Desert, decreasing the evolutionary potential of the primary plant producers and increasing the desertification risk for a vulnerable habitat. We highlight the importance of considering livestock as a major driver of genetic losses in dry rangelands under overgrazing pressure, especially given current forecasts of climate change.

The pattern of seed dispersal in time and space can affect plant fitness and the soil seed bank, and thus information is needed on this aspect of the seed biology of a species before it is selected for use in habitat restoration projects. Zygophyllum xanthoxylon is a super-xerophilous shrub that is a potential pioneer species for use in revegetating highly disturbed areas of the cold deserts of northwest China. We studied fruit release and soil seed banks of Z. xanthoxylon for 3 years in two cold desert habitats characterized by different degrees of drought and wind velocity. In our study, fruit (a three-winged capsule) release began in summer (June 2010, August 2011, July 2012) and extended for 9–10 months, but plants can be found in the population with previous- and current-year fruits attached to them. More than 50% of the fruits were released in the first 3–4 months after maturity, while the others were released gradually over a 7–8 month period. The temporal pattern of fruit dispersal varied with habitat but not with amount of precipitation during summer. The pattern of fruit deposition on the soil surface was affected by neighbouring plants, wind velocity, wind direction and topography. In both habitats, >90% of the fruits were deposited beside large and small clusters of plants, mainly Ephedra przewalskii. To facilitate plant community development, we suggest that E. przewalskii should be planted (as a wind break) together with Z. xanthoxylon when native pioneer species are used for restoration of cold desert shrublands.