Influence of domestic livestock grazing on American pika (Ochotona princeps) haypiling behavior in the eastern Sierra Nevada and Great Basin

—In a pilot study, I observed a relationship between domestic livestock grazing and location of American pika (Ochotona princeps) haypiles in the eastern Sierra Nevada and several Great Basin mountain ranges. Where vegetation communities adjacent to talus bases (forefields) were grazed, mean distance from the talus borders to the closest fresh hay piles was 30.1 m (SD = 18.9 m, n = 27), and haypiles were found only high in the talus. In ungrazed forefields, mean distance was 1.8 m (SD = 0.9 m, n = 57), and haypiles were found along the low-elevation talus–vegetation border. Where grazing was active, haypiles consistently contained vegetation gathered from plants growing within the talus. Talus vegetation appeared to be of lower diversity and the plant species of lower nutritional value than forefield plants. This difference, if real, would compromise quality of forage for summer browsing and winter haypile storage. This condition, combined with potentially less favorable summer and winter thermal conditions of upper talus locations relative to lower talus borders, suggests that grazing might be a factor compromising population conditions and status of pikas. Recent studies have reported higher extirpation rates of pika populations in Great Basin ranges (primarily in Nevada) than in adjacent regions. Because domestic livestock grazing is widely permitted on public lands throughout pika habitat in the Great Basin but not permitted (or much more restricted) in pika habitat of the Sierra Nevada, California, grazing effects might be contributing to observed regional differences in viability of pikas. RESUMEN.—En un estudio preliminar, observé la relación entre el pastoreo por ganado doméstico y la ubicación de las reservas de comida de la pica americana (Ochotona princeps) en el este de la Sierra Nevada y de varias cordilleras de la Gran Cuenca. En las zonas de vegetación adyacentes a bases del talud (antepaís) que han sido pastoreadas, la distancia promedio de los límites del talud a las reservas de alimento fresco más cercanas fue de 30.1 m (DE = 18.9 m, n = 27); a su vez, sólo se encontraron tales reservas en las partes altas del talud. En el antepaís que no ha sido pastoreado, la distancia promedio fue de 1.8 m (DE = 0.9 m, n = 57), y se encontraron reservas de alimento a lo largo del límite entre la parte baja del talud y la vegetación. En lugares donde el pastoreo ha sido activo, las reservas de alimento estaban compuestas consistentemente de vegetación recolectada de las plantas que crecían dentro del talud. Esta vegetación del talud pareció ser de menor diversidad y las especies de plantas de menor valor nutricional en comparación con la plantas del antepaís. Si esta diferencia es real, comprometería la calidad del forraje para el ramoneo del verano y el almacenamiento de alimento para el invierno. Esta condición, combinada con las condiciones térmicas potencialmente menos favorables del verano e invierno en las zonas altas del talud en comparación con los extremos inferiores del talud, sugieren que el pastoreo podría ser un factor que esté comprometiendo las condiciones de la población y el estatus de las picas. Hay estudios recientes que han observado una extirpación mayor de las poblaciones de picas en las cordilleras de la Gran Cuenca (principalmente en Nevada) en comparación con regiones adyacentes. Debido a que el pastoreo por ganado doméstico se permite en muchos terrenos públicos a lo largo del hábitat de la pica en la Gran Cuenca, pero no se permite (o está mucho más restringido) en el hábitat de la pica en la Sierra Nevada, California, los efectos del pastoreo podrían estar contribuyendo a las diferencias regionales observadas en la viabilidad de las picas. Mountains, and Rocky Mountains, pikas persist across wide elevation bands, exploiting diverse talus landforms and saturating available habitat. By contrast, populations in Great Basin mountain ranges appear to have higher extirpation rates, lower abundances, and higher percentages of unoccupied habitat (Beever et al. 2003, 2008, 2011, Millar and Westfall 2010). Attribution of these differences remains uncertain, and several hypotheses have been proposed to explain the status of pikas in the Great Basin. Multivariate analyses indicate that climate change, expressed through chronic sum mer warming and/or acute winter cold stress, is the dominant factor influencing observed de clines in the Great Basin (Beever et al. 2003, 2010, 2011, Rodhouse et al. 2010, Wilkening et al. 2011); however, island biogeographic consid erations could also explain these patterns. Habitat within Great Basin ranges is scattered and disjunct, and mountain ranges are isolated, greatly restricting dispersal within and especially between ranges. This situation, coupled with the metapopulation dynamics exhibited by pika populations, predicts lower habitat occupancy within ranges and lower recolonization rates across ranges for regions like the Great Basin relative to extensive cor dillera (Smith 1974b, Hafner and Sullivan 1995, Gorrell et al. 2005, Millar and Westfall 2010). Both climatic and island biogeographic hypotheses are supported by paleoecological evidence from the Great Basin, which documents extirpations of pikas from ranges without recolonization during the Holocene and shows that extirpations correlate with low elevations and warm temperatures (Hafner 1993, Grayson 2005). Given questions about the vulnerability of pikas to changing climates, focus on Great Basin populations is important. I offer preliminary ob servations to suggest another factor that might influence the status of pikas in Great Basin ranges: grazing by domestic livestock on vegetation communities adjacent to pika talus bases (forefields). Grazing can remove preferred forage, driving pikas to poor-quality habitat high in talus fields and thereby impacting their population health. A difference in range management policies exists between the primary public agency units that administer pika habitat in this region. Thus, a grazing effect, if present, could be pervasive through much of the Great Basin mountain regions. In Nevada and a small portion of California, a significant extent of pika habitat is administered by the USDA Forest Service (Hum boldt–Toiyabe National Forest [HTNF]) or the Bureau of Land Management. These administrative units maintain active domestic livestock allotments throughout many mountainous re gions, including vegetation communities at elevations up to and including alpine habitat. By contrast, most public agencies administering land in the Sierra Nevada (except a piece managed by the HTNF) currently exclude grazing from most midto high-elevation regions that support pikas (Michele Slaton, Inyo National Forest, personal communication). Under typical conditions, pika territories average <50 m in diameter and pikas preferentially locate haypiles along lower talus borders adjacent to meadows, and grassand shrublands that support choice forage (Smith and Ivins 1984, Smith and Weston 1990). Although pikas are generalized herbivores, they nonetheless select high-value forage species depending on season and availability (Huntly et al. 1986). During warm seasons, pikas forage daily in talus forefields within a few meters of the talus border. By midto late summer, they begin to collect vegetation for winter consumption. This vegetation comes from the same forefields but is often a different species mix than the forage consumed directly. Pikas stockpile this vegetation in large hoards (“haypiles”) under boulders within the talus field, usually one to several per territory (Millar and Zwickel 1972). Haypile vegetation is eaten during winter when fresh forage is unavailable. Haypile plants are selected on the basis of high nutritional and caloric content, as well as capacity for preservation during many months in cold, wet conditions (Dearing 1997). Haypiles are often located close to (<6 m from) the talus/ vegetation interface (Millar and Zwickel 1972). This short distance minimizes the time and risk for pikas traveling from haypile to forefield. Pikas augment their haypiles, especially in late winter or early spring, by running either over snow to collect foliage from conifer trees whose branches drape onto the snowpack or through snow tunnels to collect vegetation along the talus borders. Location of haypiles near the bor ders of talus fields thus appears critical for year-round access to abundant and high-quality forage. In addition to their proximity to forage, lower locations within a talus field appear to have 426 WESTERN NORTH AMERICAN NATURALIST [Volume 71