Evaluating the Role of Resource Heterogeneity in Restoring Grasslands

Alan K. Knapp, John M. Blair, Scott L. Collins and Sara G. Baer

Project Summary

Although there are many examples of restoration projects in which a subset of the characteristics of the native system have been successfully restored, typically there are critical ecosystem attributes that recover very slowly. Plant species diversity is one such attribute. Native plant species richness and diversity in restored ecosystems is usually much lower than in comparable undisturbed ecosystems, and may take decades or longer to approach undisturbed ecosystem levels. This is particularly true in grasslands, where it has proven relatively easy to restore the dominant C4 grasses, but it is difficult to establish the diverse C3 forb populations that characterize most undisturbed grasslands. Most grassland restoration efforts occur on former agricultural fields, where soil resources have been homogenized. In contrast, most natural communities are characterized by a high degree of spatial heterogeneity in resources, which is assumed to enhance plant species diversity. Indeed, much recent evidence indicates that individual plant species may be able to alter and maintain levels of resources (such as soil N) that differ from unvegetated areas or from beneath other species. We contend that the problem of low plant species diversity that plagues most restoration efforts is due to the slow conversion of an ecosystem from a spatially homogeneous to a heterogeneous state of resource availability. We propose to test the hypotheses that recovery of plant species richness, diversity and community heterogeneity will occur more rapidly during the restoration process if greater resource heterogeneity is imposed on the system from the beginning. We will test this hypothesis by conducting a replicated restoration project in tallgrass prairie in which soil N and soil depth (water availability) are altered in a spatially explicit fashion within restoration plots. Soil N availability will be altered in strips within plots either by adding N to increase N availability or recalcitrant C (sawdust) to reduce soil N availability. Soil water availability, will be altered in strips perpendicular to the N manipulation strips by reducing the available rooting depth of the plants to 30 cm. Buried limestone layers (abundant locally) will be used to essentially create alternating areas of shallow and deep soils in the plots, which will in turn affect soil rooting depth and water availability. Control plots, N manipulation plots, soil depth manipulation plots, and combined soil depth and N manipulation plots will provide a range of resource heterogeneity. These plots will be established in former agricultural fields located within the bounds of the Konza Prairie LTER, and reseeded with native prairie plants using standard restoration techniques. Our goal is not to assess the restoration technique per se, but to evaluate the impact of different degrees and forms of resource heterogeneity on the recovery trajectories of plant species diversity and richness during the restoration process. We will quantify resource availability (N, water and light) and plant community responses at a range of spatial scales. We feel that this replicated, experimental approach to restoration will provide unique insights into the role that resource heterogeneity may play in the restoration of ecosystems.