Persistent Establishment of Outplanted Seedlings in the Mojave Desert

Owing to short-term monitoring durations, the persistence of plants restored to disturbed sites is frequently uncertain. This is particularly true in deserts, where extreme climates display high variability among years, with some years hardly receiving any precipitation. Restoration performed during favorable weather could seem successful in the short term, only to fail later. While short-term studies are important for understanding the fast effectiveness of restoration for initiating recovery, longer-term studies are required to assess the persistence of restoration efforts (Wortley et al. 2013).

Starting in 2008, the National Park Service and University of Nevada Las Vegas conducted an experiment with a goal of assessing the persistence over six years of 12 native perennial species outplanted on a disturbed site in the Mojave Desert. The 12 species represented a range of growth forms: the forbs Baileya multiradiata (desert marigold), Penstemon bicolor (pinto beardtongue), and Sphaeralcea ambigua (desert globemallow); the grasses Oryzopsis hymenoides (Indian ricegrass), Aristida purpurea (purple threeawn), and Muhlenbergia porteri (bush muhly); and the shrubs Bebbia juncea (sweetbush), Encelia farinosa (brittlebush), Ambrosia salsola (cheesebush), Ambrosia dumosa (white bursage), Eriogonum fasciculatum (buckwheat), and Larrea tridentata (creosote bush). Seed was sourced from the Mojave Desert region, with most from National Park Service collections within 30 km of the experimental restoration site. Seedlings were grown in local greenhouses for one year in 4-L, round, plastic pots filled with 2:1 sand:organic potting soil. During this growout, pots were in natural lighting and soil water content was kept near field capacity.

The experimental restoration site was a 0.2-ha cleared area, 16 km east of Las Vegas, Nevada, US, within the 563,000-ha Lake Mead National Recreation Area managed by the National Park Service. Vegetation and soil on the site had been disturbed through blading by heavy equipment, and was mostly surrounded by creosote bush-white bursage shrubland. A fence was installed to enclose the site. The fence had mesh openings 0.6 cm in diameter, was 1.25 m tall, and included buried aluminum flashing to a height of 0.45 m aboveground (Figure 1). The upper 25 cm of soil was a sandy loam low in organic carbon (0.2% by weight) and nitrogen (0.01%). The site was at an elevation of 381 m. A climate station at McCarran Airport 28 km away (but at a higher elevation of 662 m) reported the following 1949–2015 averages: 40°C July daily high temperature, 1°C January daily low temperature, and 10.5 cm of rain/year. During the study from 2009 through 2014, annual precipitation was 79% of average and ranged from lows of 38% (2009) and 44% (2014), to highs of 128% (2012) and 142% (2010).

Figure 1.

Main photo (May 2010): experimental restoration site (community plots portion) where persistence of 12 native perennial species was assessed for six years adjacent to the native plant nursery within Lake Mead National Recreation Area, Nevada, US. Inset (April 2014): large, flowering individuals of Encelia farinosa (brittlebush; center foreground) and Sphaeralcea ambigua (desert globemallow; left side). Photo credit: S.R. Abella.

Two types of plots were established with different planting arrangements to evaluate species performance: community plots within a 0.05-ha area and species plots within a 0.01-ha area. Community plots were each 2 m × 2 m and separated from each other by a 0.5-m buffer. Species plots were 1 m × 1 m and also separated by 0.5 m. Strips of aluminum flashing, buried 0.2 m into the ground and extending to 0.45 m aboveground, were installed around the perimeter of each plot to further separate plots and deter herbivory. Treatments were completely randomly assigned to plots, and the planting density was 9 seedlings/m² for both types of plots. There were 60 community plots, including 12 replicates each of five types of plantings (forbs, grasses, two sets of shrubs, and a no-planting control). Within each community plot, 12 individuals of each of the three species within the assigned species group were randomly...