How to Build a Prairie

            Less than 1 percent of the Midwest’s original tallgrass prairie remains. Iowa has been hit particularly hard, the vast majority of tallgrass prairie having been converted to agricultural land. One way conservationists are trying to deal with the loss of tallgrass prairie and other degraded ecosystems (and their ecosystem services) is ecological restoration, which involves re-creating the community structures that existed before human degradation. It’s more complicated than just putting some plants in the ground; restoration ecology draws upon fields such as wildlife biology, landscape ecology, and invasion ecology in order to re-create viable ecosystems. Restoration projects of varying sizes -including the large-scale Neal Smith National Wildlife Refuge - have become important for maintaining biodiversity and ecosystem functions in Iowa.

            In order to better restore tallgrass prairie in Iowa, researchers need to know what factors influence restoration outcomes. Ecologists disagree on the extent to which the outcomes of restoration can be determined by land management decisions. For instance, factors such as soil conditions, land-use legacy, and site history - which are largely out of manager’s control - have been proven in the past to affect species establishment and community assembly.

            A recent study (Grman, Bassett, & Brudvig, 2013) looked at which factors are most important in determining restoration outcomes. The researchers gathered data on the restoration and management of 27 different prairies in southwest Michigan. They studied four main factors:

•    Management - seed-mix species richness, grass and forb (non-grass herbaceous plant) planting density, and burn frequency

•    Historical factors - land-use history, site age, and precipitation during the planting year.

•    Site characteristics - soil type, soil heterogeneity, shade, site area, and perimeter: area ratio

•    Landscape context - what kind of habitats (forest, grassland, cropland) are surrounding the restoration?

            The researchers found that management decisions are actually quite important in determining the plant community structure of prairie restorations. Overall, management practices accounted for 65% of the variation in sown species richness. Higher forb seeding density was shown to increase species richness and decrease the richness of non-sown species (species not intentionally planted in the restoration). Using seed mixes with lower numbers of species and a higher proportion of dry prairie grasses, however, led to a higher plant beta diversity. Together, seed mix composition and forb seeding density were shown to have a significant effect on plant community composition.

            Historical factors that are out of managers’ control were also highly important in determining the restoration outcome. Older sites had lower numbers of non-sown species, and lower numbers of species in general, probably due to the increased abundance of Andropogon gerardii (a dominant C4 grass) in older sites. In formerly tilled sites, sown-species beta diversity was generally lower, and former old-fields tend to have lower diversity of non-sown species while having more exotic clonal C3 grasses.

          Surprisingly, site characteristics did not have a major effect on species richness. They shouldn’t be completely discounted, however, as sites with more heterogeneous soil had higher abundance of Poa pratensis, an exotic clonal C3 grass. It should also be pointed out that site characteristics, such as soil type, have been shown in other studies to have bigger effects on restoration outcome. The sites used in this study may have all exhibited conditions “suitable” for restoration, leading to the lack of variation based on this category. Landscape features, including the habitat types surrounding the restored prairie, were also less important, but restorations in areas dominated by row crops did have lower non-sown species beta diversity.

Figure 1. This graphic shows how much of the variation in species richness, beta diversity, and community composition among the sites is explained by each of the main factors (Management decisions, site characteristics, landscape features, site history, and unknown/residual factors). As seen in the graphic, management was a very important determinant of species richness and beta diversity.

So, what does this study in Michigan have to do with the natural history of the Iowa flora? The results of this study can be applied to tallgrass prairie restoration in Iowa, a state which has undergone similar land use changes.

If the goal of the restoration is to maximize the diversity and richness of native species restored, then the results of the paper suggest using seed mixes with a high diversity of native species, a high density of forbs, and a lower density of grasses, which is something relatively feasible for managers to accomplish. 

            However, if the ultimate goal of a restoration is to accurately restore tallgrass prairie to their conditions before human settlement, then things become more complicated. One study (Carter & Blair, 2012) compared community structure in six restored prairies (with similar landscapes and seeding regimes to the Michigan restorations) to 3 nearby remnants in Iowa. The study found that while the restorations were similar to remnants in their native species richness and abundance, species diversity, and late-appearing C3 plant abundance, they differed in many other aspects. For instance, restorations had a greater richness of late-appearing C3 species, and a lower abundance and richness of early-appearing native plants than remnants. So, while the restorations were able to create viable ecosystems, they were not able to necessarily re-create ones resembling those of Iowa’s past. The authors suggest that this discrepancy is due to management practices and not the inability of these plants to establish, as the seed mixtures most commonly used in restorations often under-represent early-appearing C3 plants.

            Restoration can have many different goals, whether it is re-creating a vanishing, historic habitat, restoring biological diversity, or reestablishing ecosystem services. The Michigan and Iowa studies suggest that the success of these goals, and which of them are most prioritized, can often depend on the management choices we make. Whether we are re-creating a historic habitat or actually creating a slightly new one can depend on something as simple as the type and amount of seeds planted in a restoration.

References

Carter, D.L. and Blair, J.M. 2012.  Recovery of Native Plant Community Characteristics on a Chronosequence of Restored Prairies Seeded into Pastures in West-Central Iowa. Ecological Restoration. 20(2): 170-179.

Grman, E. Tyler Bassett, T., and Brudvig, L.A. 2013. Confronting contingency in restoration: management and site history determine outcomes of assembling prairies, but site characteristics and landscape context have little effect. Journal of Applied Ecology. 50: 1234–1243.