Submitted to: Environmental Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 1, 2012
Publication Date: July 7, 2012
Repository URL: http://handle.nal.usda.gov/10113/56074
Citation: Boyd, C.S., Davies, K.W. 2012. Spatial variability in cost and success of revegetation in a Wyoming big sagebrush community. Environmental Management. 50:441-450. Interpretive Summary: In Wyoming big sagebrush communities, environmental factors impacting post-fire restoration success vary across space, but most restorative management practices (e.g., seeding perennial grasses) are applied uniformly within a management unit. Our objectives were to define the probability of restoration success over space using relevant soil-related environmental factors and to use this information to estimate cost of successful restoration across space. We established a 1500 m transect within a Wyoming big sagebrush landscape that had been re-seeded with perennial grasses following wildfire, and in plots spaced at 30 m along the transect monitored 1) initial environmental conditions 2) plant density of seeded species at 2-years post-seeding (i.e., restoration success); this information was used to develop models to predict probability of restoration success and cost of restoration based on initial environmental conditions. Our results indicate that restoration success (and associated costs) varied strongly over space, but success was predictable based on initial environmental conditions (i.e. soil texture and nutrient status). Our experimental protocol provides managers with a technique to identify important environmental drivers of restoration success and this process will be of value for spatially allocating logistical and capital expenditures in a variable restoration environment.
Technical Abstract: The ecological integrity of the Wyoming big sagebrush alliance is being severely interrupted by post-fire invasion of non-native annual grasses. In order to curtail this invasion, successful post-fire revegetation of perennial grasses in affected areas is required. Environmental factors impacting post-fire restoration success vary across space, however, most restorative management practices (e.g. seeding perennial grasses) are applied uniformly within a management unit. Our objectives were to define the probability of restoration success over space using relevant soil-related environmental factors and to use this information to estimate cost of successful restoration across space. Our study site was within a burned Wyoming big sagebrush landscape in southeast Oregon. The fire occurred in August 2007, and was reseeded with perennial grasses, latter that Fall. We established a 1.5 km-long transect that had 6 x 6 m plots spaced at 30 m-intervals, for which we collected data on 28 soil-related environmental factors during the spring and early summer of 2008. Density of seeded species was also estimated in the plots during July of 2008-2009. Plots were classified as successful (> 5 seedlings/m2) or unsuccessful based on density of seeded species. We used logistic regression to model the relationship between restoration success and soil correlates. Five variables representing soil nutrient content and particle size distribution were included in the final model, which correctly predicted restoration success for 88.2% of plots. The model was used to estimate probability of restoration success for each plot. We then divided an estimated seeding cost of $66.75/acre by restoration probabilities to estimate the cost outlay needed to successfully restore one acre under environmental conditions represented by each plot. Estimates for probability of restoration success ranged from 0.003 to 0.99 and cost estimates varied from $67.66 to $17,428.20/acre across the 1.5 km transect. When restoration probabilities were averaged across plots within a management-realistic scale (500m), cost estimates ranged from $100.36 to $281.65. Direct application of the correlations between soil-environmental characteristics and revegetation success from this study is limited to sites with similar ranges for important environmental variables. However, our experimental protocol provides managers with a technique to identify important environmental drivers of restoration success and this process will be of value for spatially allocating logistical and capital expenditures in a variable restoration environment.