Constraints to restoring fire and grazing ecological processes to optimize grassland vegetation structural diversity

• Bison extirpation and fire suppression have homogenized grassland habitat.
• Specifications for managers to engineer cattle*fire interactions are needed.
• Constraints inhibiting patch-burn grazing for ecological restoration were modeled.
• Most informative model included stocking rate, burn completion and precipitation.
• Suggests patch-burn grazing as an ecosystem engineering strategy.

Extirpation of the ecosystem engineer (bison) and its interaction with fire, coupled with the utilitarian concept of moderate grazing, have contributed to homogenization of grassland habitat in North America. Although cattle may serve as a proxy for bison, combining fire with cattle grazing has been uncommon and to date managers have not always successfully applied cattle and controlled burns as tools to manipulate grassland vegetation heterogeneity and increase habitat diversity. Using an information-theoretic approach, we assessed factors constraining the fire-grazing interaction ecological process to engineer habitat structure of grasslands via patch-burn grazing. We assessed how grazing, fire, and biotic and abiotic features in tallgrass prairie influenced establishment and maintenance of low vegetative structure in burned patches, the positive feedback driving the fire-grazing interaction, and subsequent structural heterogeneity across a pasture. Four pastures were divided into three patches with a different patch burned annually in March/April from 2007 to 2013. Cattle were stocked from light to heavy (1.1–4.4 AUM/ha) from May to October (∼150 days) with access to the entire pasture. We hypothesized that the exotic C3 grass tall fescue (Schedonorus arundinaceus), lag-time between burning date and the date cattle were put into experimental pastures, and burn date would be the constraining factors. However, the most informative model included stocking rate, date of burn completion, and precipitation. The lightest cattle stocking rate did not establish low vegetative structure in the burn patch, which resulted in the lowest heterogeneity among patches. The heaviest cattle stocking rate established but did not maintain low vegetative structure in the burn patch. The intermediate cattle stocking rate maintained the lowest vegetative structure in the burn patch and the greatest heterogeneity among patches, i.e., the best efficacy of patch-burn grazing to engineer habitat structural heterogeneity. The relationships of stocking rate to burn patch vegetative structure and to landscape heterogeneity were both quadratic and were both optimized at intermediate stocking rate.