Can a Combination of Grazing, Herbicides, and Seeding Facilitate Succession in Old Fields?

Abandoned agricultural lands called “old fields” currently cover millions of hectares of land on nearly every continent. Studies of old fields indicate the effects of past cultivation on soils and vegetation may persist for hundreds of years (Dupouey et al. 2002, Fuhlendorf et al. 2002). Additionally, exotic species are often planted after final tillage, making it difficult for native species to recolonize.

In arid shrub-steppes of the Pacific Northwest, many old fields remain dominated by exotic species decades after abandonment (Kulmatiski 2006). The Zumwalt Prairie in northeastern Oregon is one of the last large relicts of Pacific Northwest bunchgrass prairie. However, old fields comprise approximately 10% of the prairie. Zumwalt old fields are dominated by exotic perennial grasses that were bred for rapid reproduction, tolerance to high levels of livestock herbivory, and disease resistance. Zumwalt old fields are low in native plant species richness and lack biological soil crusts. Improving the ecological condition of old fields is a priority of local natural resource management entities (Wallowa Resources 2005) and The Nature Conservancy (Shephard and Taylor 2009).

Augmentative restoration recognizes the positive ecological attributes intact within a plant community while seeking to restore key aspects of composition, structure, and function (Sheley et al. 2009). An augmentative approach can be used to meet specific land management goals while also improving cost effectiveness. Although Zumwalt old fields are dominated by exotic grasses, several factors favor their recolonization by native species, including their small size (<100 ha), proximity to unaltered native prairie, and intact hydrologic function. The potential for improving these old fields using augmentative approaches is high. We investigated the effectiveness of three treatments—grazing, herbicide, and seeding—in an augmentative restoration approach on Zumwalt old fields. The treatments had to meet several criteria: low cost, ease of application for typical landowners, and low impact to existing native species. Our goal was not to restore an unaltered native species prairie nor was it to eradicate non-native species. Rather, we sought to significantly reduce dominance of exotic perennial grasses while increasing the abundance of native perennial grasses.

The study was conducted in an old field on The Nature Conservancy’s Zumwalt Prairie Preserve of northeastern Oregon (lat 117°39N, long 45°319W) at 1600 m elevation. Soils are deep to very deep, well drained, silt loams of the Tippett and Zumwalt series. Climate is characterized by cold, dry winters (average -3 °C), cool, moist springs (4 °C), and warm, dry summers (average 15 °C). Total yearly precipitation averages 35 cm. Abundant plant species include intermediate wheatgrass (Thinopyrum intermedium), timothy (Phleum pratense), and exotic Festuca species. Native grass species comprise <1% of the floral density.

Treatments were replicated four times and arranged in a randomized block split-plot design, with grazing as the whole plot and seeding and herbicide as split-plots. One contiguous half of each block (100 m × 200 m) was assigned a grazing treatment: summer grazing at 80% utilization using cow-calf pairs for three consecutive years (2004– 2006), or non-grazed. The seeding and herbicide treatments were randomly assigned to four 50 m × 50 m plots in each of the grazing treatments, resulting in eight total treatments (grazing, grazing+herbicide, grazing+herbicide+seeding, grazing+seeding, herbicide, herbicide+seeding, seeding, and a non-treated control). In July 2006, immediately following grazing, the herbicide treatment, 2.3 L/ha (1 qt/ac) glyphosate (N-[phosphonomethyl] glycine) was applied using a vehicle with a boom sprayer. Five species of locally-sourced native perennial grasses were seeded using a rangeland drill at a rate of 10.1 kg PLS/ha in the following proportions: bluebunch wheatgrass (Pseudoroegneria spicata), 30%; Idaho fescue (Festuca idahoensis), 32%; basin wildrye (Leymus cinereus), 12%; Sandberg’s bluegrass (Poa secunda), 15%; and prairie Junegrass (Koeleria macrantha), 11%. Seeding began in November 2006 but inclement weather delayed completion until April 2007.

In June 2004, pre-treatment density by species (number of culms of grasses) was measured in ten randomly located 20 cm × 50 cm quadrats per 50 m × 50 m treatment plot...