Sub-lethal effects of energy development on a migratory mammal—The enigma of North American pronghorn

摘要

Abstract To meet societal demands for energy, ～ ～ 50,000 gas sites are developed annually in North America, among which many are in western less-developed and wildlife rich areas. To evaluate effects of increasing energy infrastructure requires sufficiently robust study designs, an onerous issue given the vastness of scale, limited funds, and an abject dearth of baseline data. Here we address these issues, first by discussion of the type of approaches needed to develop proper inference about potential effects of energy footprints, and subsequently through an empirical approach by examining the biological performances of more than 370 GPS radio-collared adult female pronghorn ( Antilocapra americana ). A rigorous attempt to examine if industrial development has any impact on pronghorn is based on three assumptions: (i) late-winter body mass reflects a period of inadequate food availability because winter habitat is altered; (ii) variation between population segments reflects spatial differences in food availability, increased energetic costs, or varying survival rates between gas field and non-developed sites; and (iii) reproductive correlates including physiological and immunological markers and adult survival are lower at sites varying in habitat quality. Our study area situated in one of the world’s premier wildlife regions–the southern Greater Yellowstone Ecosystem–harbors approximately 100,000 wintering ungulates, some of North America’s longest migrations, and two of the continent’s largest gas fields. We compared the response of five variables between wild pronghorn in control (not disturbed) and experimental (developed gas fields with well pads, roads, and traffic) sites—pregnancy, chronic stress, immune function, body mass, and adult survival. Despite shifts in animal movements, which included avoidance of energy infrastructure where development is occurring at the highest densities inside two of the largest natural gas fields in North America (Pinedale Anticline Project Area [PAPA] and Jonah fields) and other behavioral or ecological observations of sub-lethal effects, we failed to reject the null hypothesis that development is unrelated to parity in pronghorn biological responses. Studies intent on producing knowledge to assess whether energy development is inimical or not to ungulates will increasingly require appropriate time scales and understanding whether populations are below an expected food ceiling. Further, as with pronghorn in our study region, knowing if individuals are at the limits of their biological range (e.g. altitude) where stressful winter conditions may mask impacts of development is important. Graphical abstract Figure options Download full-size image Download as PowerPoint slide Keywords Migration ; Natural gas ; Pronghorn ; Survival ; Yellowstone prs.rt("abs_end"); 1. Introduction As the human footprint continues to expand globally, governments will be the final arbiters of strategies implemented on behalf of people and the planet’s biodiversity. This is especially true of relatively intact geographies such as the neotropics, Arctic, and savannas where hydrocarbon development presses into wildlife-rich areas ( Copeland et?al., 2009 , Berger and Beckmann, 2010 and Naugle, 2010 ). The establishment of current and future conservation practices is partially dependent on credible science which, in turn requires appropriate study designs to detect if change occurs, and if so the magnitude of change from habitat alteration and other disturbance on population performances. Ideally, such designs would include baseline data on species abundance and distribution, biological attributes, abiotic factors (e.g.?temperature and precipitation) and demographic trends ( Beckmann et?al., 2011 ). Such information is however rarely available ( Northrup and Wittemyer, 2013 and Lynch et?al., 2015 ). The most detailed studies of mammals in petroleum-rich areas have concentrated on four species in North America—caribou ( Rangifer tarandus ), elk ( Cervus elpahus ), mule deer ( Odocoileus hemionus ), and pronghorn ( Antilocapra americana ). While abundant GPS data are available ( Hebblewhite and Haydon, 2010 and Hebblewhite, 2011 ) much remains unknown about biological impacts of development at the individual or population level ( Sawyer et?al., 2009a , Sawyer et?al., 2009b , Lendrum et?al., 2012 and Lendrum et?al., 2013 ). Most studies have addressed questions about habitat fragmentation and road avoidance ( Beckmann et?al., 2011 , Beckmann et?al., 2012 and Seidler et?al., 2015 ) with responses broadly classified as sub-lethal which may include behavior and ecological components such as movement, density alterations, and habitat shifts ( Bayne and Dale, 2011 ). Since most work has been in areas where baseline data lack, projects have generally been observational and correlative, although at times Before–After-Control-Impact (BACI) approaches have been deployed. Less frequent are studies with replica