| Summary: | Reducing the cover of non-native species is one of the challenges of ecosystem
restoration. The goal of this study is to identify native species traits that will increase
native species cover and reduce non-native species cover in the first growing season at
upland and wetland prairie restoration sites.
Native and non-native prairie species were planted in the fall and harvested the
following summer at both an upland and a wetland site. Native species traits, such as
plant weight, leaf area, relative growth rate, leaf area ratio, leaf weight ratio, net
assimilation rate, and specific leaf area were measured under laboratory conditions for 7-
and 21-day old seedlings. Germination rate (laboratory) and phenology (field) were also
measured.
At the upland site, species with a large 7-day plant weight and a high germination
rate also had high native cover in the field (P<0.00l, R��=0.83). At the wetland site, high
21-day leaf area, low 21-day leaf weight ratio, and high net assimilation rate predicted
increased native cover in the field (P<0.00l, R��=0.87). An abundance of natives, as
measured by native cover, native biomass, and number of individuals, likely results in
fewer resources (light, nutrients, and water) available for the non-native species growth
resulting in a reduction in the non-native cover.
Intrinsic traits of native species also predicted the field performance of non-native
species, although the amount of variation explained was lower than the amount of
variation explained in the models that predicted native cover. In the upland site, native
species with high leaf weight ratio (21-day) tended to have lower non-native cover in
their field plots (P=0.087, R��=0.23). In the wetland site, the native species traits that
predict non-native cover were low 21-day leaf area and high 21-day leaf weight ratio
(P<0.00l, R��=0.46). These traits were similar to those that predicted native species cover
at the wetland site.
This study demonstrates the ability of species traits to predict field performance.
Predictive models were generated using native species traits to select species for
restoration that will increase native cover and decrease non-native cover in the first
growing season. Traits can be measured for species not included in this study and the
models generated can be used to predict the field performance of species at similar sites. === Graduation date: 2002
|
|---|
