Passive restoration potential of riparian ecosystems following the manual removal of invasive giant reed (Arundo donax)

Background/Question/Methods

Giant reed (Arundo donax L.) is an agressive woody grass native to the Old World that has invaded the southwestern US and Mexico.  Considered one of the greatest threats to riparian areas, giant reed is firmly established on throughout the bi-national Rio Grande Basin, and as a consequence, these landscapes have been severely fragmented.  Discussions of restoration of this rare riparian ecosystem center around the control of giant reed, although to date, little is known regarding the potential of passive restoration, or natural recruitment and re-emergence of other native vegetation, after giant reed management efforts.  To examine this, we conducted two sequential studies: (1) an analysis of reemerging vegetation after periodic manual removal of Arundo donax, and (2) a soil core analysis to look at the potential soil seed bank.  Four sites were chosen along a 35 mile stretch of the Rio Grande in Webb County near Laredo, TX.  In each site, we randomly placed and established up to four 25m² plots, in which all existing above ground biomass was identified, measured, and removed. Each plot was revisited each month for 16 months to remove all Arundo donax stems greater than one meter in length.  Any emergent transitional vegetation was also identified and documented.  Soil cores were taken at the end of the 16 months, seeds were keyed out to morphospecies and viability tested using 1% triphenyltetrazolium chloride (TTC).

Results/Conclusions

Initial diversity indices were nil, with the only plant species across all four sites being giant reed.   Plant succession increased geometrically over time of giant reed removal, and after 16 months of selective removal, species abundance in the different sites ranged from 6 to 32 different species. Collectively 46 non-Arundo donax species were identified, most of which were native forbs, vines, and saplings.   These results suggest that a persistent reduction in the competitiveness of the clonal Arundo donax can significantly increase species richness over time. Our soil bank analysis across all sites yielded total of 5,850 seeds, of which 21% were viable. Most seeds were found in the first 5 cm of the soil and leaf litter, illustrating the presence of a potentially significant, but possibly transient, propugule source for restoration of these riparian areas. However, this data confirms the recalcitrant nature of the potential succession of native vegetation, centered on the suppression of the dominance of giant reed.