U.S. Geological Survey Chesterton, IN, United States
Abstract: Since 2010, the Great Lakes Restoration Initiative (GLRI) has provided substantial funding to strategically target the biggest threats to the Great Lakes ecosystem including protecting native habitats and restoring native species. The Great Lakes basin contains the largest freshwater bodies of water in the world embedded in a landscape that varies from intensely developed and fragmented to large natural areas, many of which exist at biome intersections and therefore are species rich. One priority of GLRI is promoting conservation of native pollinators within the Great Lakes basin, which led to the creation of the GLRI Pollinator Task Force (PTF). The PTF provides funding for restoration activities that will best support pollinator conservation. However, understanding where to apply restoration activities to best promote pollinator conservation is weakly understood. To assist the PTF in determining where restoration might be most beneficial, we performed a Zonation analysis based on how to benefit the most bee species, given various conditions and costs that affect the ability of an area, to support both rare and common bee species. We modeled inputs that affected the condition of the landscape to support bees such as habitat quality, pesticides, connectivity, costs associated with restoration, and landscape rules that favored retaining sites that are resilient to the effects of climate change. Distributional data on more than 500 native bee species were used to create restoration scenarios that protect areas with high species richness and to protect areas with rare species occurrences. Results from the Zonation analysis indicated that focusing restoration solely on protected areas will be suboptimal in conserving the entire bee biota of the Great Lakes basin. For example, mean conservation value of focal restoration areas selected by the PTF, prior to Zonation analysis, based on expert opinion and agency priorities for protected areas, was 452 units (lower values have lower conservation value) versus 575 units for non-selected areas (t = 2.1, p = 0.04). This significant difference illustrates that site selection performed without Zonation input undervalued non-protected areas.
