The Relative Importance of Seeding Method, Soil Ripping, and Soil Variables on Seeding Success

We conducted a field experiment to examine how seeding method, soil ripping, and soil characteristics affected the initial establishment and growth of seeded species and if differences among treatments persisted into the second growing season. We planted seeds into a compacted field plot where most of the topsoil had been removed. The native seed mixture of Artemisia californica, Eschscholzia californica, Eriogonum fasciculatum, Lupinus succulentus, Nassella pulchra, and Vulpia microstachys represented different seed sizes and life histories. Three seeding methods (hydroseeding, imprinting, and drilling) and three ripping depths (0, 20, and 40 cm) were combined in a factorial experiment. Soil organic matter and NO₃ ^? were used as covariates. For two years, we measured density, percent cover, mean size, and flower production of the selected species, and weed emergence. Only seeding method and soil variation affected initial establishment of natives. Small‐seeded species had higher density in imprinted and hydroseeded than drilled treatments, whereas large‐seeded species had higher density in imprinted and drilled than hydroseeded treatments. These patterns persisted with only slight modification into the second year. Weed density in year 1 decreased with soil ripping. In year 1, Vulpia height and Lupinus height and flowering were greater with drilling or imprinting than hydroseeding but were not affected by ripping. Eschscholzia, Lupinus, and Vulpia produced seeds in the first year, but only Vulpia reestablished successfully in the second year. Vulpia had high cover in the second year that increased with increasing NO₃ ^? , but did not vary by treatment. In year 2 perennial Nassella, and to some extent Eriogonum, grew largest, produced more inflorescences, and had their highest percent cover in the 40‐cm rip treatment. Size and inflorescence production also increased with increasing NO₃ ^? ; sometimes this relationship was stronger than the effects of treatments. We found only positive associations between estimated biomass (density × height) of annuals and survival of shrubs. Potential for erosion control, as measured by total density in year 1 and total vegetative cover in year 2, was greatest in imprinted and hydroseeded treatments and increased with increasing NO₃ ^? . This relatively simple experiment yielded information critical to understanding optimum seeding methods and seedbed preparation and indicates that seeding method can be determined by seed size and germination biology. Although an experiment such as this enables some generalizations, it does not eliminate the need for site‐specific experiments prior to restoration.