Summary
In the northern Great Plains of North America, Japanese brome Bromus japonicus is a non-native, annual which is now common in many mixed prairie communities. In an attempt to control it, plots dominated by Japanese brome and western wheatgrass Agropyron smithii were burned in Badlands National Park, South Dakota (central USA) in April 1983 and/or 1984.
In the study area, wheatgrass (273 tillers/m²) and Japanese brome (1,500 tillers/m²) were co-dominants. Green needlegrass Stipa viridula, sand dropseed Sporobolus cryptandrus, threadleaf sedge Carex filifolia and buffalograss Buchloe dactylodies were common.
Twelve 5 x 5 m plots were established comprising three replications of four treatments: i) unburned, ii) burned 20 April 1983, iii) burned 16 April 1984, and iv) burned 20 April 1983 and 16 April 1984.
Fuel weight and water content were determined prior to burns (clipping 5, 0.1 m² quadrats, drying and weighing). Air temperature, relative humidity and wind speed were measured immediately before, during and after each fire. Litter frequency was determined using a point frame.
Western wheatgrass and Japanese brome tiller densities, and standing crop (annual above-ground production) of all species, were estimated in July 1983, 1984, 1985 and 1986 (in 5, 0.1 m² quadrats/plot).
April burning favoured western wheatgrass (e.g. from 300 in 1983 to 507 tillers/m² in July 1984 after burning in April 1984) and reduced Japanese brome tiller density (e.g. from 2,250 in 1983 to 23 tillers/m² in July 1984) and standing crop for at least one growing season post-burn.
In years of no burning, Japanese brome standing crop and tiller density were positively correlated with frequency of surface litter e.g. tiller density 500/m² at around 10% litter frequency, and 2,500/m² at around 80% litter frequency. Burning also reduced subsequent generations by reducing surface litter (and presumably seed production); this was most apparent when autumn precipitation was below average.
Effects of burning on other species varied. It reduced the standing crop of green needlegrass for at least 3 years post-burn (e.g. from 4 g/m² in 1983 to 0 g, 1 g and 1 g in the 3 years following burning in 1984; burning in 2 consecutive years reduced standing crop further); green needlegrass should thus be considered intolerant of spring burning.
Conversely, standing crop of buffalograss and sand dropseed increased for 3-4 years post-burn. Threadleaf sedge was not significantly affected by any of the burn treatments.