Natural disturbances influence wetland carbon cycling, and fire is a key driver of terrestrial carbon stocks. However, the influence of fire on wetland carbon cycling remains poorly understood. Here, we investigated how prescribed fire and wildfire impact soil carbon storage in a forested floodplain of south-eastern Australia. We sampled four areas within Murray Valley National Park, the world’s largest river red gum (Eucalyptus camaldulensis) stand, and compared soil carbon (C), nitrogen (N) and C:N ratios between control (unburnt in the 50 years prior to sampling), prescribed burn and wildfire-impacted floodplain areas. Mean soil C and N concentrations were 4.7% ± 0.32% and 0.36% ± 0.02%, respectively, and mean C:N ratios were 14.23 ± 0.33. Carbon concentrations and C:N were highest in control areas of the floodplain, while N concentrations were highest at wildfire-impacted areas. However, flood frequency was a stronger driver of soil C than fire disturbance. Soils at more frequently flooded areas had higher C concentrations compared to less frequently flooded areas, suggesting that resilience to C loss through fire could be enhanced through hydrological restoration. We believe this warrants further research as a potential nature-based climate measure. Mean C density data indicate soil C stocks of 9.4 Tg across Barmah-Millewa Forest, highlighting the significant carbon storage value of this ecosystem.

Medusahead is an introduced, winter-annual grass covering millions of hectares of the semiarid West. It forms exclusive stands and has a dense thatch cover that resists the establishment of desirable species. Prescribed fire can remove medusahead litter and improve plant establishment. Medusahead control is fundamental to establishing desirable species that will, in turn, resist further invasion. Imazapic is an effective herbicide for control of medusahead, but more information is needed on its effects on desirable species. Our objectives were to test how imazapic application rate and timing affected medusahead, seeded desirable species, and other nontarget vegetation on burned and unburned rangeland in southeast Oregon. We burned existing medusahead infestations at two different sites in June 2003. Following the burn, imazapic was applied at rates of 0, 35, 70, 105, 140, 175, and 210 g ai/ha between July and October of 2003 in a randomized strip-plot design. In November 2003, monocultures of seven desirable species were drill-seeded across the imazapic treated areas. Data on cover and density of medusahead and seeded species were collected in 2004 and 2005. Cover data of nontarget species were collected in the summer of 2005. Medusahead cover was highest in control plots and lowest in plots that received the highest herbicide application rates. Medusahead cover was lower in burned plots. The effect of imazapic on nontarget vegetation was less clear. Seeded species established in the study plots, but their response to herbicide rate showed few consistent patterns; some of the seeded species showed little response to herbicide, whereas others appeared to establish best at different herbicide rates, depending on site and whether the plots were burned or unburned. Site and burn treatment also affected how imazapic rate or application month influenced cover of perennial or annual grasses or forbs.

Desert grasslands of the southwestern United States have experienced an increase in the abundance and distribution of woody plant species over the past century. Shrub encroachment has caused a substantial loss of grasslands in the northern Chihuahuan Desert. The Chihuahuan Desert has also been invaded by Lehmann lovegrass, a fire-adapted species from southern Africa. In 1999, the U.S. Department of Agriculture–Agricultural Research Service burned a remnant desert grassland to determine the effects of prescribed fire on shrub–perennial grass dynamics. The grassland also contained the nonnative perennial grass Lehmann lovegrass. I am reporting on a study that was initiated to determine whether prescribed burning would alter the vegetative community within and proximal to a Lehmann lovegrass–dominated patch. Cover of Lehmann lovegrass showed no significant response to the burn treatment. Of the dominant native species, only black grama and broom snakeweed had a significant year by treatment interaction. No species or growth form had a significant vegetation type by year by treatment interaction. After 6 yr, differences between burned and unburned transects were not significant for any species or growth form.