Shrub proliferation is a common consequence of land degradation in desert grasslands that has adversely affected ecosystem processes. It is often assumed to be the result of livestock grazing reducing the ability of grasses to competitively exclude shrubs. However, there is a growing body of literature showing this is not necessarily the case. What then may be driving grassland-to-shrubland shifts? It is well-known that disruption of ground cover by grazers amplifies aeolian transport in wind-erodible arid lands. We hypothesize that this intensifies ‘sandblasting’ effects, with grasses being more susceptible to damage than shrubs. To test this hypothesis, we conducted a controlled experiment at the Chihuahuan Desert Jornada LTER site. Three grass growthforms (Bouteloua eriopoda (strong perennial, stoloniferous), Sporobolus airoides (strong perennial, bunchgrass), and Aristida purpurea (weak perennial)) and three shrub growthforms (Prosopis glandulosa (fruticose, deciduous), Atriplex canescens (suffrutescent, semi-deciduous), and Larrea tridentata (fruticose, evergreen)) species commonly found in the Southwestern USA were targeted. Potted individuals were subjected to a factorial combination of two intensities (5 min with 60 and 120 g soil cm-1 min-1) and two frequencies (7 and 14 day intervals) of sandblasting in a portable wind tunnel and leaf gas exchange and chlorophyll fluorescence were quantified.
Results/Conclusions
Stomatal conductance (gs) increased in all species immediately after sandblasting, but more so in grasses (200-600%) than shrubs (~100%). Nighttime gs also significantly increased (~100-200%) in grasses, but gs in shrubs was unchanged. Leaf transpiration rates also increased in grasses following sandblasting, particularly in B. eriopoda, while it remained steady in shrubs. The potential maximum quantum yield of photosystem II and net assimilation rate decreased after sandblasting on two of the grass species (B. eriopoda and A. purpurea), but generally stayed relatively stable on S. airoides and the shrub species. Accordingly, sandblasting substantially reduced leaf water use efficiency (WUE) of the grasses but had no effect on shrub WUE. The first sandblasting event had more significant effects on plants than the subsequent events and there were no consistent intensity-frequency effects. Overall, our results suggest grasses are more ecophysiologically vulnerable (e.g., reduced photosynthesis, increased water loss) to sandblasting than shrubs. Under field conditions, sandblasting could therefore amplify grazing-induced stresses on grasses and hasten their demise, while having little impact on shrubs. Over time, this differential lifeform response to sandblasting would promote grassland-to-shrubland transitions in wind-erodible drylands.