Prairie restorations can be difficult due to presence of non-native grasses. An experiment was undertaken at Grasslands National Park (49°22′N, 107°53′W) in Saskatchewan, Canada. It investigated emergence of crested wheatgrassAgropyron cristatum (a widespread non-native perennial) and native blue grama Bouteloua gracilis, in a former cultivated field (surrounded by native prairie) sown with wheatgrass in about 1947 (and still dominated by it), in herbicide and non-herbicide treated plots with two levels of watering (simulating years with high rainfall).
Ten (3 x 10 m) plots were established in A. cristatum-dominated areas and five randomly assigned a glyphosate (Round-up) treatment (applied each May 1994-1997) and five untreated (controls). In April 1997, 10 PVC tubes (10 cm diameter x 15 cm long) were sunk into each plot, with the top 4 cm left protruding (to stop seeds from blowing away and to prevent added water from flowing out). Five tubes in each plot were randomly assigned a watering level (high or low). For each level, two tubes were used to measure soil moisture and three to record grass seedling emergence from broadcast sown seeds or seeds sown at 1 cm depth (200 seeds/tube), and the seed bank.
Water addition significantly increased emergence of seedlings from sown seed. For broadcast seed, Agropyron emergence was around 20% at the higher watering level and 5% at the lower (regardless of whether herbicide applied). At the higher watering level Bouteloua emergence was around 20% in herbicide treated plots but only 14% in no herbicide plots; at the lower, emergence was around 2% in herbicide plots but 6% in no herbicide plots.
Emergence was greater for buried seed (80%); herbicide did not increase emergence but significantly more Bouteloua than Agropyron germinated from buried seeds.
Only one Bouteloua seedling emerged from the seed bank, whilst the average density of Agropyron seed bank derived seedlings was 930/m2; this was not reduced even after 4 years of herbicide treatment (hence Agropyron maintained dominance) and remained a problem in restoration efforts.
If using or referring to this published study, please read and quote the original paper, this can be viewed at: http://onlinelibrary.wiley.com/doi/10.1046/j.1526-100X.2003.00020.x/full