Predicting and mitigating weed invasions to
restore natural post-fire succession
in Mesa Verde National Park , Colorado , USA
M. Lisa Floyd 1, David Hanna 2, William H. Romme 3, Tim Crews 2
1 Environmental Studies Program, Prescott College, 220 Grove Ave. Prescott, AZ, USA 81301 Telephone +1 928 350-2220; fax 970 588-3383;
email: lfloyd-hanna@prescott.edu
2 Environmental Studies Program, Prescott College, 220 Grove Ave. Prescott, AZ, USA 81301
3 Department of Forest Sciences, Colorado State University, Ft. Collins, CO, USA 80523
Summary : Recently, wildfires in Colorado have become vulnerable to invasion by non-native plants. We investigated post-fire weed patterns and created a model to predict vulnerable sites in Mesa Verde National Park . Communities that lacked re-sprouting species on particular soils and those of high biodiversity were most vulnerable. Restoration using native grasses was effective in reducing weeds. A weed risk model displayed on the Mesa Verde website research page predicts the most vulnerable sites in a spatial display. Additional details to modify this model can be obtained (dhanna@prescott.edu) for other management units with similar fire and weed concerns.
 |
Abstract. Six large wildfires burned in Mesa Verde National Park during the last 15 years, and extensive portions of burns were invaded by non-native plant species. The most threatening weed species include Carduus nutans, Cirsium arvense, and Bromus tectorum, and if untreated, they persist at least 13 years. We investigated patterns of weed distribution to identify plant communities most vulnerable to post-fire weed invasion and created a spatially explicit model (displayed on the Mesa Verde National Park Website) to predict the most vulnerable sites. At the scale of the entire park, mature piñon-juniper woodlands growing on two soil types were most vulnerable to post-fire weed invasion; mountain shrublands were the least vulnerable. At a finer scale, greater richness of native species was correlated with greater numbers of non-native species, indicating that habitats with high native biodiversity are at the greatest risk of weed invasion. In unburned areas, weed density increased with increased soil nitrogen, phosphorus, and salinity and in burned areas with soil nitrogen and texture. We also evaluate the effectiveness of a variety of weed mitigation methods; aerial seeding of targeted high-risk areas with native grasses was the most effective treatment tested. We recommend a conservative mitigation plan using natives grass seed on only the most invasible sites.