Response of non-target species and their interactions to vegetation restoration: Pollinator communities along the Sacramento River

Attempts to restore degraded landscapes often focus on replanting key vegetation elements while presuming that other desirable species will colonize on their own. Despite their potentially key roles for ecological functioning, non-target species, such as pollinators, are seldom included in assessments of restoration. To investigate the success of current approaches along the Sacramento River at restoring non-target species, I compared bee and flowering plant communities at six restored sites and riparian remnants. Remnants contained mixtures of riparian forest and open grass/gravel.  Restored areas were mid-succession stands planted with riparian trees and shrubs in 1997. Each site was surveyed every six weeks from late February through August 2003. Bees were netted at flowers and trapped in water-filled pans.  Floral abundance was measured using quadrats.  Sampling yielded 125 bee species, 91 at remnant sites and 90 at restored sites. Average richness and abundance did not differ between sites types.  Thus, the restored habitats supported abundant and diverse pollinator communities within five to ten years. The composition of bee communities at restored sites, however, differed significantly from those in remnants (multi-response permutation procedure, A =0.12, P =0.003).  Communities at restored sites were not nested subsets of those in the remnant fragments suggesting that many pollinators found at restored sites arrived from habitats other than forests.  Flowering plants were also distinct between restored and remnant sites, which may contribute to the differences in pollinator communities.  In addition, I use survey data to explore methods for assessing variation in the structure of plant-visitor webs.