In lieu of an abstract, here is a brief excerpt of the content:

  • Effects of Fuel-Reduction Treatments on Pollinators in a Pinyon-Juniper Woodland (Arizona)
  • Susan Nyoka (bio)

Concern about uncontrollable wildfire in pinyon-juniper woodlands has led public land managers in the American Southwest to aggressively seek alternatives for reducing hazardous fuel loads. Currently, treatment methods are focused on reestablishment of woodland structure and reduction of fire hazards, yet restoration of functional processes such as pollination is critically important for long-term ecosystem sustainability. Insect pollinators, particularly bees, are vital for maintaining understory plant diversity and productivity in natural plant communities. The study I describe here suggests that managers can simultaneously reduce fire hazards and support insect pollinators in southwestern pinyon-juniper woodlands through overstory thinning followed by prescribed burning.

Pinyon-juniper woodlands occupy more than 19 million hectares in the western United States, representing the most common vegetation type in Arizona and New Mexico. Studies of non-pollinating insects (e.g., Kendall 2003) and butterflies (Kleintjes et al. 2004) suggest that fuels-reduction treatments may improve habitat for insect pollinators in this system. To date, however, no published studies have specifically examined the effects of restoration treatments on pollinator assemblages. Native bees, including bumblebees (Bombus) and a wide variety of solitary and semisocial species, are the principal pollinators in most ecosystems (Michener 2000). Flies, especially the Syrphidae (flower flies), Bombyliidae (bee flies), and Muscideae (house flies), are considered the second most important insect pollinator group (Larson et al. 2001), though they are often overlooked. The goal of this study was to determine how bees and flies vary in abundance and species richness in response to different fuels-reduction treatment options versus untreated controls in a pinyon-juniper woodland.

This study was conducted at an established research site in the Kaibab National Forest, just south of Grand Canyon National Park in northern Arizona. The study area ranges from 2,005 to 2,073 meters in elevation and supports an overstory of pinyon pine (Pinus edulis) and Utah juniper (Juniperus osteosperma) in upland areas, with ponderosa pine (Pinus ponderosa) and Gambel oak (Quercus gambelii) occupying canyon draws. Common floral resources include shrubs, such as Mexican cliffrose (Purshia mexicana) and broom snakeweed (Gutierrezia sarothrae), and a sparse but diverse array of forbs, such as buckwheat (Eriogonum), beardtongue (Penstemon), milkvetch (Astragalus), and a wide variety of composites (Asteraceae). [End Page 119]


Click for larger view
View full resolution
Figure 1.

Treatment effect on bees in a pinyon-juniper woodland in Arizona: a) mean (± SE) species richness (May: F3,39 = 3.19, p = 0.03; August: F3,39 = 2.73, p = 0.02); and b) mean (± SE) abundance (May: F3,24 = 2.95, p = 0.005; August: F3,24 = 2.30, p = 0.02). Means with different letters are statistically different from one another (α = 0.05).

We used a randomized complete block design for this study, with six experimental blocks encompassing approximately 770 ha. Within each block, eight 1 ha experimental treatment units were randomly located within a systematic grid of points (200 m spacing) and assigned one of four treatments (two treatment replicates per block, n = 48): no treatment (Control); tree thinning from below (Thin); prescribed fire (Burn); and thinning from below followed by prescribed fire (Thin + Burn). Thinning treatments were completed in spring 2005, and burning in fall 2006.

Insect and floral resource data were collected over the course of eight days both pre- and post-monsoon (May and August 2008). All bee and fly species observed attending flowers were considered potential pollinators. Surveys were conducted along three 50 m transects within a 10 m × 50 m belt centered on the middle of each treatment unit, and consisted of an observer walking linear transects over a 30-minute period during hours of maximum pollinator activity (10:00 a.m.–3:30 p.m.). All bees and flies detected within 1.5 m of the observer were recorded to morphospecies, along with the identity of the flowering species attended by each insect. To ensure that field identification was comparable among workers, a reference collection was made prior to field surveys and easily memorable code names were assigned to species. Insects that could not be unambiguously identified on the wing were...

pdf