Plant recruitment in drylands varies by site, year, and seeding technique

Restoration in dryland ecosystems is hindered by low establishment of seeded species. As such, evaluations of current seeding methods are critical to understanding limitations and barriers to seeding success. Drill seeding is perceived as an optimal seeding strategy in many dryland ecosystems, but broadcast seeding is more commonly used as a seeding method due to physical and logistical constraints. For example, broadcast seeding may be conducted by aerial drops where other methods are limited by topography or obstructive features in the landscape. Few studies have quantified the differences between drill and broadcast seeding through space and time. We compare 2‐year recruitment of emergent Pseudoroegneria spicata (bluebunch wheatgrass) seedlings in the sagebrush steppe biome for drill versus broadcast seeding methods across three seeding years, three landscape aspects and two soil types using a 95% confidence interval approach to avoid the penalty of multiplicity. We found drill seeding had 2.7 times greater recruitment of seedlings after 2 years compared with broadcast seeding. However, differences were highly subject to seeding year, aspect and soil type, likely because of soil moisture and temperature variations. Drill seeding had an advantage on clay soils with flat and north aspects (10.1 and 4.6 times greater for drill than broadcast seeding, respectively). In most conditions, drill seeding had greater recruitment than broadcast seeding, though in 2014 on south aspects broadcast seeding had 2.7 times greater recruitment than drill seeding. The results of this study demonstrate a need for restoration plans that account for spatiotemporal variation in seeding success.