Perennial grass and native wildflowers: A synergistic approach to habitat management

Authors: XAVIER, SHEREEN - University Of Georgia; Olson, Dawn; Coffin, Alisa; Strickland, Timothy - Tim; SCHMIDT, JASON - University Of Georgia

Submitted to: Insects
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/20/2017
Publication Date: 9/22/2017
Citation: Xavier, S., Olson, D.M., Coffin, A.W., Strickland, T.C., Schmidt, J. 2017. Perennial grass and native wildflowers: a synergistic approach to habitat management. Insects. 8(4):104-117. https://doi.org/10.3390/insects8040104.
DOI: https://doi.org/10.3390/insects8040104

Interpretive Summary: Marginal agricultural land provides opportunities to diversify landscapes by producing biomass for biofuel, and through floral provisioning that enhances arthropod-mediated ecosystem service delivery. To examine the effects of local spatial context and irrigation on cellulosic biofeedstock production, and wildflower bloom and visitation by arthropods, a 2 x 2 design of 20 habitat buffer plots was established with two location context variables (buffer adjacent to woodland or agriculture and irrigation or no irrigation). Each buffer habitat included a subplot of Napier grass (Pennisetum perpureum Schumach) for biofeedstock, three commercial wildflower mix subplots, and a control subplot containing spontaneous weeds. Arthropods and flowers were visually observed in quadrats throughout the season, and Napier biomass was harvested at the end. Irrespective of buffer location or irrigation, pollinators were observed more frequently on wildflowers than weeds in the control plots. Natural enemies were variable and somewhat related to floral treatments. Herbivore visits were infrequent and not significantly influenced by experimental treatments. Napier grass yields were high and typical of yields reported regionally, and were not affected by location context or irrigation. Our results suggest that a landscape design integrating a bioenergy crop and floral habitat would provide marketable biomass and arthropod service provisioning.

Technical Abstract: A total of 19 buffer plots were established on University of Georgia experimental farms and lands near Tifton, GA in 2015. The buffer plots were assigned to a 2 x 2 design of local spatial context and irrigation. For local spatial context, ten plots were located adjacent to woodland (“T”) and ten in open areas between 1 and 30 m from agricultural fields (“A”). Half of each of these plots then received irrigation or no irrigation. The irrigation treatments were irrigated (“I”) weekly and the other half of the plots received no irrigation (“N”). The total buffer area was 912 m2 (36.5 m by 25 m) and the plots were separated by 150 m to 10,500 m. The buffer plots were divided into subplots with an individual area of 4.5 m2 (1.5 m by 3 m) and a single Napier grass subplot. Subplots were planted with commercially available southeast native wildflowers with 1 subplot having spontaneous weed growth as a control. The area of the Napier grass subplot was 51.75 m2 (1.5 m by 34.5 m). All subplots were spaced 1.5 m apart and 1.5 m from the buffer edge. Vegetation and arthropod sampling was carried out simultaneously on the same day. Sampling was carried on six dates from June to August of 2016. For vegetation sampling we counted the inflorescences of each species of wildflower occurring within a random area of 0.25 m2 (a pvc quadrat) in each of the subplots; we found that a quadrat of dimensions 0.5m by 0.5m accommodated the maximum number of flowering species. The arthropod counts were carried out within the same established quadrat. Each quadrat was visually observed for 3 min. for three functional groups of arthropods: pollinators, natural enemies and herbivores. Pollinators and natural enemies were observed more frequently on the wildflowers than the weeds in the control plots. No differences between the three commercial mixes planted were found in pollinator floral visitation suggesting that these floral mixes were equally attractive to pollinators. Natural enemy floral visitation was higher in two of the three of the three floral mixes. Herbivore floral visitation was low overall and not related to the floral and control treatments. Buffer location and irrigation had no effect on pollinator, natural enemy and herbivore floral visitation. Napier grass yields were often high and typical of what has been found in the region. Napier grass yields were also not affected by buffer location and irrigation.