Grasslands are prone to shrub and tree invasion where natural mechanisms (e.g. fire or grazing) which retard their establishment are reduced or lost. In southwest USA, large areas of semi-arid grasslands have become scrub-invaded in response to reduction of livestock grazing. Two shrubs in particular, burroweed Isocoma tenuisecta and broom snakeweed Gutierrezia sarothrae have invaded millions of hectares. Whilst prescribed burning may assist grassland restoration, there is a risk of increasing erosion and soil degradation as vegetation cover is greatly reduced for some time post-burn. In this study within Cibola National Forest (New Mexico), effects of burning semi-arid grasslands were investigated.
The study was undertaken at two sites where broom snakeweed was the dominant shrub:
1) Bernalillo Watershed (1,660 m; livestock grazing ended 1947); prescribed fire 15-16 November 1995, about 168 ha burned;
2) West Mesa (1,820 m; livestock grazing ended early1970s); prescribed fire 14 February 1996, burns confined to treatment plots (4 ha).
At each site, eight, 1 ha plots (4 control, 4 burned) were established. Along three 60-m transects (central plot area to avoid edge effects) vegetation cover, bare soil, litter, gravel and rock was recorded on four occasions (before burning, within 3 months (Bernalillo) or one month (West Mesa) post-burn, and after the first and second growing seasons). Soil samples were collected (at these times) according to cover-type (shrub, grass or bare soil) and analysed for various properties.
Runoff-sediment collectors (3 x 10 m) were installed in plots at both sites.
Before burning, vegetation cover (total 36-38%; grass c. 20%; shrubs 12-14%) was fairly sparse and patchy; all cover-types declined from May 1995 to February 1996 (during a period of drought). Grass cover reduction was greatest in burned plots, but not significantly less than controls. With the return of the rains, total vegetation and grass cover increased from February 1996 to January 1997. One year post-burn, grass cover had recovered to almost that of the controls; after 2-seasons cover was similar.
In 1997, shrub cover was slightly less than prior to treatment commencement but appeared unrelated to burning as values were similar for control (8%) and burn plots (7%).
Over the 2 years, potentially mineralizable soil N was similar on control and burn plots (lack of effect may be due to ash being blown away); it was greatest in soils under shrubs, slightly lower under grass and lowest in bare soils. Sediment transport was also similar in control and burn plots (there was a non-significant trend toward higher sediment yields in burn plots)
The potential for adverse effects from fire was enhanced due to the drought following burning. However, any effects of burning were overall minor compared to controls.
If using or referring to this published study, please read and quote the original paper, this can be viewed at: https://www.uair.arizona.edu/holdings/journal/issue?r=http://jrm.library.arizona.edu/Volume53/Number1/