Population dynamics of small mammals in relation to production of cones in four types of forests in the northern Sierra Nevada, California.
RESUMEN--Estudiamos ensamblajes de pequenos mamiferos en cuatro tipos de bosques de coniferas (abeto blanco, abeto rojo, abetos mixtos y pino-cedro) en la Sierra Nevada de California durante dos temporadas (2003-2004). Evaluamos la produccion de conos por especies de coniferas dominantes en cada ano. La produccion de conos fue mayor, en general, en el otono de 2003, pero varia por tipo de bosque y entre las especies de coniferas. Paralelamente, el promedio de densidades maximas de ratones del campo (Peromyscus maniculatus) aumento en 2004 (desde 0,7-7,3 individuos/ha hasta 65,7-112,7 individuos/ha). Los numeros de ardillas manto dorado (Spermophilus lateralis) fueron similares en ambos anos; tipico de las especies que invernan, esta especie ocurrio a bajas densidades en mayo (6,6 [+ or -] 0.2), densidades fueron maximas en septiembre (24,5-32,5 individuos/ha), y sus numeros se redujeron en octubre (9,2 [+ or -] 4,8). Las ardillas rayadas orejonas (Tamias quadrimaculatus) alcanzaron mayores densidades en bosques de abeto rojo (48,2 [+ or -] 13,4 individuos/ha) y en bosques de abetos mixtos (36,0 [+ or -] 13,5 individuos/ha) que en los bosques de abeto blanco (7,6 [+ or -] 2,7 individuos/ha), y todas las poblaciones alcanzaron sus numeros maximos en septiembre. La ardilla rayada de Allen (Tamias senex) se mantuvo en densidades menores de T. quadrimaculatus excepto durante el mes de septiembre de 2004 cuando alcanzaron altas densidades (54,6 [+ or -] 26,8 individuos/ha). La sobrevivencia de P. maniculatus fue dependiente de una interaccion entre el tipo de bosque y el mes, con efectos adicionales de sobrevivencia invernal y la produccion promedia de conos en el otono de 2003. La sobrevivencia de S. lateralis vario por mes, mientras que la sobrevivencia de ambas especies de Tamias vario con la interaccion de tipo de bosques y el mes, mas efectos adicionales de la sobrevivencia invernal y la produccion de conos para T. quadrimaculatus. Las ratas cambalacheras patas oscuras (Neotoma fuscipes) estuvieron presentes en las elevaciones bajas y llegaron a mayores densidades en bosques de pinos y cedros. Las ardillas voladora del norte (Glaucomys sabrinus) no fueron capturados comunmente, y se encontraron principalmente en los bosques de abeto rojo.
The Sierra Nevada is a prominent ecoregion in the western United States and possesses a diverse flora and fauna, including threatened and endangered species. Although the Sierra Nevada is well known and accessible, there have been surprisingly few studies on the basic ecology of the region, especially of the mammalian fauna. Small mammals are abundant and diverse throughout the range and constitute important prey for predatory species of concern such as the California spotted owl (Strix occidentalis occidentalis; Smith et al., 1999), northern goshawk (Accipiter gentilis; Promessi et al., 2004), and American marten (Martes americana; Clark et al., 1987). Although many of these species are known to exhibit fluctuations in size of populations and dynamics of assemblage (e.g., King, 1968; Murie and Michener, 1984; Kirkland and Layne, 1989; Zabel and Anthony, 2003), details of these dynamics and factors underlying them remain poorly understood. Informed management in the face of environmental variability relies upon scientific assessment of the range of variation that may be predicted and the factors influencing this variation. Such information also is needed as we move toward a philosophy of adaptive management of forests (Graham and Krueger, 2002; McNeely, 2004).
We have been studying small mammals at a series of replicate grids in the northern Sierra Nevada. In this contribution, we provide basic information on populations of small mammals over 2 years in four major types of forest (white fir, red fir, mixed-fir, and pine-cedar) in the northern Sierra Nevada and relate these to production of cones by conifers as a measure of productivity. Application of advanced quantitative methods allows us to provide baseline data on survival and population densities for these species. Although this time series is not extensive, this is one of the longer studies of its sort in this region and provides some insight to the role of availability of resources in the form of annual variation in production of conifers.
MATERIALS AND METHODS--We studied small mammals in the Plumas National Forest near Quincy, California, at elevations of 1,220-2,100 m in habitats that included white fir, red fir, mixed-fir, and pine-cedar forests. Although there was considerable overlap in composition of species in each type of forest, they can be characterized roughly as follows. White fir forests were composed primarily of densely packed white fir with little to no understory and an abundance of downed wood. Red fir forests were dominated by red fir (Abies magnifica), with western white pine (Pinus monticola), sugar pine (P. lambertiana), and lodgepole pine (P. contorta var. murrayana) as secondary species. Understory consisted primarily of large mats of pinemat manzanita (Arctostaphylos nevadensis) interspersed with snowberry (Symphoricarpos) and saplings of red fir. Mixed-fir forests were a mixed-conifer forest consisting of Douglas fir (Pseudotsuga menziesii), ponderosa pine (Pinus ponderosa), sugar pine, white fir (Abies concolor), and incense cedar (Calocedrus decurrens). Finally, pinecedar forests were on rocky exposed slopes supporting sparse ponderosa and Jeffrey pine (P. jeffreyi), and incense cedar. Understory species also were sparse and consisted primarily of several species of manzanitas and lilacs (Artostaphylos and Ceanothus).
Because white fir forests were a major component of the region, they were emphasized in the study. We established three sets of three experimental grids (control, light thin, and heavy thin) in white fir forests of similar condition. These plots were sampled for 2 years (2003, 2004) prior to any thinning treatments to determine baseline conditions; these comprise the data presented in this report. To provide basic information on populations of small mammals inhabiting additional major types of forest, we established three grids in each of three other types (red fir, mixed-fir, and pinecedar) to mimic the experimental structure within white fir forest. Thus, we established a total of 18 trapping grids, with 9 in white fir forests and 3 in each of 3 other forest types. Coppeto et al. (2006) divided our white fir sites into white fir (n 5 5), mixed-conifer (n 5 3), and mixed-fir (n 5 1, in addition to the three sites herein treated as mixed-fir), based on levels of codominance of white and red fir. Whereas Coppeto et al. (2006) were attempting to assess environmental and habitat features most clearly allied with different species of small mammals, observations reported here are designed to integrate with those of other research modules in this project, which emphasize vegetation, songbirds, spotted owls, and behavior of fire; toward this end, we are using types of forest as defined across all modules of this study.
Trapping grids were established in spring 2003. Each site consisted of a nested grid of 100 Sherman live traps (model XLF) set at 10-m intervals in a 10 by 10 array, and centered within a larger grid of Tomahawk live traps (model 201) set at 30-m intervals in a 6 by 6 array. We attempted to place all grids randomly in the habitat by picking a random point for the corner of the grid; however, high density of roads in the Plumas National Forest made this difficult. For a complete description of sites see Coppeto et al. (2006).
Each grid was trapped for 5 consecutive nights each month from May or June through October (depending on snow melt). Traps were baited with rolled oats and black-oil sunflower seeds mixed with peanut butter and set in early evening. Sherman traps were checked early the following morning and closed. Tomahawk traps also were checked early the following morning, but were left open until mid-morning when they were closed. All traps were closed 1200-1600 h PDT to prevent thermal stress to captured animals. In addition, traps were supplied with polyester insulation during cold nights. Captured animals were weighed, sex determined, reproductive condition assessed (females: perforate, imperforate, swollen vagina, pregnant, or lactating; males: scrotal or not), and location of capture was recorded. Uniquely numbered Monel eartags (Model 1005-1; National Band and Tag Co., Newport, Kentucky) were applied to each ear. All fieldwork was conducted under the auspices of an approved animal care and use protocol and met guidelines recommended by the American Society of Mammalogists (Gannon et al., 2007).
To evaluate effects of production of seeds by conifers on abundance of small mammals, we measured availability of cones during autumn 2003 and 2004. Only mature cones were assessed in these surveys, so that our measure estimates availability in the year of measurement. We measured cones on 10 randomly selected individual trees of each species on each grid. For this we selected mature, adult trees with pointed crowns, as tall as or taller than the surrounding canopy (Gordon, 1978), and sufficiently far apart that their crowns did not touch. For grids with <10 individual trees of a given species, additional trees were assessed as close to the grid as possible (<500 m). We counted the same trees in both years and within a 2-week period to prevent confounding temporal factors. Counting was performed by standing at a distance of [greater than or equal to] 1.5 times height of the tree and manually counting cones using binoculars. Two people counted cones independently, but simultaneously, and we used the mean of these counts for all analyses; variation among paired observers was greatest for Douglas fir (50.0 [+ or -] 8.7 cones) owing to the numerous small cones. Other conifers had little inter-observer variability in counts: white fir (12.9 [+ or -] 1.7), red fir (2.3 [+ or -] 1.4), ponderosa pine (2.2 [+ or -] 1.3), Jeffrey pine (1.1 [+ or -] 0.4), western white pine (5.7 [+ or -] 1.4), and sugar pine (0.3 [+ or -] 0.2). For each tree, we recorded height, diameter at breast height (DBH), species, and crown class (Gordon, 1978). Temporal differences in production of cones were determined using repeated-measures analysis of variance (rmANOVA) with year and habitat as treatments, and individually counted trees as the repeated measure.
Estimates of population density and apparent survival ([PHI]) were made using a Cormack-Jolly-Seber model in program MARK (White and Burnham, 1999). Ten monthly trap sessions were used as the sampling structure, and trap sessions were separated by 1 month (with the exception of the 7-month winter period between October 2003 and May 2004). Trapping was temporarily halted on all plots during July 2004, and, therefore, we did not obtain data for some grids. The July anomaly was included in the capture probability models in program MARK, and no significant effect was observed due to missing data. Histories of encounters were generated for each individual and grouped by type of forest for analysis.
Estimated population parameters under the Cormack-Jolly-Seber data type were apparent survival between primary sessions ([PHI]) and probability of capture (p). Measures of abundance (i.e., minimum number known alive) were adjusted using probability of capture to account for the untrappable population. Program MARK is sensitive to small samples; consequently, we did not segregate our data by age as this would greatly reduce samples, especially for juveniles. Factors used to build models of survival and probability of capture included no difference (.), type of forest (white fir, red fir, mixed-fir, and pine-cedar), time (month), year, over-winter survival (November 2003-May 2004), and production of cones in autumn1 2003. Over-winter survival was modeled as a single event representing time from end of seasonal trapping to beginning of trapping the following season. Because program MARK provides monthly estimates of survival, probabilities of over-winter survival represent the cumulative probability of survival across all winter months (ca. November through April-May). Over-winter survival was modeled independently for each habitat type. Abundance values were corrected by the area trapped (1.10 ha for Sherman traps; 2.20 ha for Tomahawk traps) to obtain densities per hectare.
We developed several specific models to identify factors important to survival of focal species. These models used a combination of type of forest, month, production of cones, year, or over-winter survival. Because small mammals may exhibit a lag in their response to availability of food (McShea, 2000), we used data for production of cones only from 2003, as this reflected the food resources that would have been available during late autumn 2003 and early 2004. We applied two models for production of cones: mean production by each species (giving total production of cones) and mean production by all conifers combined, excluding Douglas fir. Because Douglas fir consistently produced large quantities of cones, we chose to eliminate this species from the latter model because the comparably large number of cones produced represented a stable source of food, whereas variability in production of cones by other species may generate years of high and low availability of food. Selection of model was determined using Akaike's Information Criterion adjusted for small samples (AICc) in program MARK.
Goodness-of-fit tests for individual histories of encounters were performed using RELEASE and bootstrap simulation methods in program MARK. Goodness-of-fit tests were used to assess the assumption of homogeneity of capture (test 2) and survival (test 3). If individuals did not exhibit independence in trappability, then estimated sampling variances would be under estimated (i.e., overdispersion). Burnham et al. (1987) suggested that an estimate of overdispersion ([??]) is calculated from the summation of tests 2 and 3 above, with values of [??]. 1.0 indicating violation of the assumption of independence, and [??] 1.3 as evidence of significant overdispersion. AICc for species exhibiting overdispersion should be adjusted using [??] as a correction factor in the computation of AICc (Burnham and Anderson, 2002). Because tests on our data revealed significant overdispersion for P. maniculatus, we used a corrected value ([??] = 1.85) to compensate for our observed overdispersion (Table 2). All other species showed [??] values <1.3, and therefore, were not adjusted. All statistical analyses were performed using SAS (SAS Institute, Inc., 2000).
RESULTS--We captured 2,969 individuals of 10 species of small mammal in 123,840 trapnights (one trapnight is one trap set for 1 night) distributed evenly across all sites during 2 years of trapping. These included North American deermice (Peromyscus maniculatus), dusky-footed woodrats (Neotoma fuscipes), montane and long-tailed voles (Microtus montanus and M. longicaudus, herein treated together), California and golden-mantled ground squirrels (Spermophilus beecheyi, S. lateralis), shadow and long-eared chipmunks (Tamias senex, T. quadrimaculatus), Douglas' squirrel (Tamiasciurus douglasii), and northern flying squirrel (Glaucomys sabrinus).
Not all species were present in the four types of forest (Table 1). All species were captured in mixed-fir forests, although M. montanus was not captured in 2003 and G. sabrinus was not captured in 2004. All species except S. lateralis were captured in white fir forests, but T. douglasii and G. sabrinus were captured there only in 2004. Neotoma fuscipes and S. beecheyi were not captured in red fir forests. Overall, pine-cedar forests had the lowest species richness and abundances, with no capture of T. douglasii, S. lateralis, or G. sabrinus; in 2003, neither chipmunk nor M. montanus was captured at these sites.
Densities of North American deermice varied both annually and monthly within years (Fig. 1). Densities were lower throughout 2003 than in 2004, when this was the most abundant species in all habitats. Densities during 2003 remained <10 individuals/ha on all sites (range, 0.7-7.3). By contrast, in 2004, densities were much greater (maximum densities of 86.0, 112.7, 77.4, and 65.7 individual/ha in red fir, mixed-fir, white fir, and pine-cedar forests, respectively). A single peak was observed in populations of North American deermice (except mixed-fir) during 2004, suggesting a single reproductive episode. The reproductive peak occurred during June in all types of forest, except mixed-fir forest, which peaked in September.
Survival of P. maniculatus was best described with a model including a type of forest by month interaction, as well as over-winter survival and mean production of cones in autumn (Table 2). Survival of the North American deermouse was highly variable, fluctuating both within and among types of forest (Fig. 1); it was most consistent in white fir forests, remaining between 0.48 [+ or -] 0.29 and 0.71 [+ or -] 0.13 individuals/ha until autumn 2004 when survival decreased. Monthly survival in the remaining types of forest followed a similar pattern, decreasing throughout 2003. Over-winter survival remained high (0.76 [+ or -] 0.13) in all types of forest. In most of 2004, monthly estimates of survival in mixed-fir, red fir, and pine-cedar forests remained below those of white fir forests; survival in white fir forests dropped in September.
Spermophilus lateralis were captured in red fir and mixed-fir forests, but captures in the latter were rare and were insufficient for population analyses. Within red fir forests, populations of S. lateralis increased following hibernation from a low in May or June, peaked in September, and declined by October (Fig. 2). September peaks reached greater densities in 2003 (32.5 [+ or -] 0.8 individuals/ha) than in 2004 (24.5 [+ or -] 3.2 individuals/ha).
Because S. lateralis was abundant only in red fir forests, we removed type of forest from population analyses, and survival was best explained solely by month (Table 2). Mean monthly survival ranged between 0.43 and 0.65 from August through September of both years (Fig. 2). Mean monthly survival followed similar patterns during both years, increasing from July to August before dropping again in October. Although survival was not estimated for June 2003, June 2004 showed the greatest rate of survival of all months sampled (0.79 [+ or -] 0.20). Over-winter survival was high for S. lateralis (96.6%).
Long-eared (Tamias quadrimaculatus) and shadow (T. senex) chipmunks occurred at our study sites. Although both species occurred in white fir, mixed-fir, red fir, and incidentally in pine-cedar forests, their populations varied by both type of forest and year (Fig. 3). Abundances of T. quadrimaculatus averaged 5 times higher in mixed-fir forests than in white fir or red fir forests (Fig. 3) and were highest in September. T. senex reached greater abundances in red fir forests than any of the other types of forest, especially in 2004, when abundances reached levels 10 times that of mixed-fir and white fir forests. T. senex in white fir forests had lower maximum abundances in 2004 (78% that of 2003); conversely, densities in 2004 were higher in both red fir (2.3 times) and Douglas fir (1.4 times) forests.
[FIGURE 1 OMITTED]
Mean monthly survival for T. quadrimaculatus was similar in both years ([F.sub.1,16] = 0.10, P = 0.76). Mean monthly survival followed a similar annual pattern across all types of forest (Fig. 4). Overwinter survival was high ([PHI] = 0.93 [+ or -] 0.02) and did not differ among types of forest. Additionally, survival was similar across both years of study (white fir, 0.59 [+ or -] 07; red fir, 0.69 [+ or -] 0.07; mixed-fir, 0.72 [+ or -] 0.06; [F.sub.1,16] = 0.10, P = 0.76).
Mean monthly survival in T. senex was more variable across types of forest than was that in T. quadrimaculatus (Fig. 4), although survival was similar overall in the two species ([F.sub.1,34] = 0.05, P = 0.82). Survival of T. senex remained constant during July and August and increased in September 2003. Survival in 2004 was quite heterogeneous across types of forest. No difference was detected in probabily of survival in 2003 and 2004 (white fir, 0.65 [+ or -] 0.06; red fir, 0.75 [+ or -] 0.07; mixed-fir, 0.52 [+ or -] 0.10; [F.sub.1,18] = 0.22, P = 0.65).
Neotoma fuscipes were not captured in red fir forests, occurred at low levels (<0.5 individuals/ ha) in white fir and mixed-fir forests, and reached their greatest densities in pine-cedar forests (Table 1). Microtus was uncommon (<1/ ha) in all types of forest (Table 1). T. douglasii and G. sabrinus were captured only rarely across all types of forest, with abundances ,0.2 individuals/ha.
Productivity of cones varied across years ([F.sub.1,1538] = 132.94, P < 0.001; Fig. 5), but not across habitat ([F.sub.2, 14.1] = 2.63, P = 0.09). The largest cone producer was Douglas fir, producing an average of 156.6 [+ or -] 15.9 and 137.2 [+ or -] 14.5 cones/tree across all types of forest (2003 and 2004, respectively). Production of cones by white fir was higher across all types of forest in 2003 than 2004 (22.4 [+ or -] 2.2 versus 4.7 [+ or -] 1.0 cones/ tree; [t.sub.367] = 10.1, P < 0.001), as was red fir (71.4 [+ or -] 8.7 versus 6.7 [+ or -] 6.0 cones/tree; [t.sub.55] = 5.6, P < 0.001), western white pine (83.5 [+ or -] 11.2 versus 37.1 [+ or -] 5.9 cones/tree; [t.sub.60] = 3.9, P < 0.001), and ponderosa pine (10.2 [+ or -] 1.7 versus 0.2 [+ or -] 0.1 cones/tree; [t.sub.200] = 5.9, P < 0.001).
[FIGURE 2 OMITTED]
Production of cones in autumn exhibited a significant year by habitat interaction ([F.sub.3,1,538] = 9.30, P < 0.001; Fig. 5). Although production declined in all types of forest, declines were much greater in mixed-fir and especially red fir forests than in white fir or pine-cedar forests. Whereas the latter two types of forest exhibited reductions of ca. 36%, the former declined by 46% (mixed-fir) and fully 72% (red fir) in the second year (Fig. 5).
DISCUSSION--Western coniferous forests are characterized by shortened growing seasons interspersed with cold, wet winters when the majority of precipitation falls as snow. As a result, growth and reproduction of plants and animals is highly dependant on the pattern of snowfall and snow melt (Millar and Innes, 1985; Wensel and Turnblom, 1998). Because these patterns are highly variable among years, the amount of available water may vary dramatically between years, impacting primary productivity of these forests (Peterson et al., 2002). For example, productive years for animals may occur when species of conifers synchronize reproduction during a wet, mild winter (Martell and Macaulay, 1981). Conversely, lean years may occur when conditions are harsh (e.g., drought), and conifers forego production of cones. As a result, species that rely on seeds as a primary source of food, such as North American deermice (Gunther et al., 1983), should experience boom-and-bust cycles that follow periods of high and low production of cones.
[FIGURE 3 OMITTED]
North American deermice are habitat generalists that are able to thrive under a variety of conditions (King, 1968; Kaufman and Kaufman, 1989). Although several studies have attempted to identify habitat features important to P. maniculatus, little conclusive evidence links specific habitat features with its survival or density. Studies on the relationship between P. maniculatus and downed wood have yielded positive relationships (Carter, 1993) or no relationship at all (Bowman et al., 2000). Additionally, Smith and Maguire (2004) detected little evidence supporting a relationship with downed wood or shrub cover, and, at our study sites, there was no clear relationship between a suite of habitat variables and density of P. maniculatus (Coppeto et al., 2006). This may have been a consequence of particularly high population densities, such that North American deermice occurred over much of the landscape.
[FIGURE 4 OMITTED]
Although we monitored assemblages of small mammals for only 2 years, the variation in abundance of P. maniculatus was high. Although not known to display cyclic patterns in annual abundance, populations do fluctuate between years of high and low abundance (Wolff, 1985), and in our study, they increased roughly 10-fold between the 2 years. Abundances of North American deermice peaked in June and September 2004, possibly reflecting bimodal reproduction (Fairbairn, 1977; Millar and Innes, 1985; Wolff, 1985). Production of cones was higher in 2003 than 2004, and early snowmelt in March 2004 shifted the beginning of the snow-free period up to May instead of June, giving P. maniculatus additional time and resources for reproduction. The proportion of P. maniculatus during 2004 was comparable to levels reported by Waters and Zabel (1998) for similar forests just north of our study area, suggesting that our trapping success was representative of the species in this region; consequently, we believe that the low values in 2003 were real and not an artifact. Another metric that indicates that populations of small mammals were unusually low in 2003 is the observation that the local population of California spotted owls failed to defend breeding territories and produce young that year (J. Keane, in litt.). Small mammals comprise most of the diet of this species (P. Shaklee and J. Keane, pers. comm.), and during 2003 only 10% of 70 monitored pairs reproduced. In contrast, 42% of 43 pairs successfully reproduced in 2004.
[FIGURE 5 OMITTED]
Survival of North American deermice varied predominantly by type of forest and month, with some additive effects from over-winter survival and mean production of cones. Because we did not trap during winter months, over-winter survival was modeled as a single probability specific to each type of forest, and, therefore, represents a mean survival across all winter months. Production of cones in this model included the mean number of cones produced by all conifers except Douglas fir, which was excluded because of reliably large numbers of cones every year. Although mean production of cones was given as a single numerical value for each habitat, mean production of cones represents an important quantifying covariate for simulation by the model providing additional information not contained in a habitat-only model. Models incorporating production of cones by individual species of conifer were poorly supported, indicating, as expected, that survival of P. maniculatus was influenced by net production of cones rather than that by any particular species of conifer.
Survival of golden-mantled ground squirrels in this study was dependant only on time of year (month). A similar pattern of survival was observed across both years suggesting this is relatively invariant. Similar rates of survival have been reported for Richardson's ground squirrels (S. richardsonii; Michener and Locklear, 1990). Because of an early snow melt in 2004, fieldwork was initiated in June rather than July; we observed a noticeable increase in survival of S. lateralis for June 2004 compared to July of either year. This also was likely an artifact of the procedure used to estimate survival. Michener and Locklear (1990) determined that 97-100% of S. richardsonii marked in early spring had hibernated at their site. As a result, most individuals at their study site in early spring had been marked the previous year and had not yet dispersed, thereby inflating survival. Similarly, the majority of S. lateralis we trapped in June 2004 were recaptures from the previous year.
Studies on selection of habitat by chipmunks have yielded mixed results. Chipmunks appear to be generalist species that show either no clear response to forest clearing (Mahan and Yahner, 1997; Sullivan and Klenner, 2000; Cote and Ferron, 2001), or a limited preference for shrub (Smith and Maguire, 2004) or overstory (Bowers, 1995) cover. Yellow-pine chipmunks (T. amoenus) reached densities 57% higher on sites with high shrub cover (Smith and Maguire, 2004). Similarly, T. quadrimaculatus in our study area reached greater densities in mixed-fir forests, which have higher shrub cover than the other types of forest (Coppeto et al., 2006). However, T. senex reached greater densities in red fir forests, where shrub cover was structurally more simple and canopy cover more open (Coppeto et al., 2006).
It should be noted that this study, as with all mark-recapture studies, measures apparent survival ([PHI]) and cannot distinguish between mortality and dispersal. Therefore, animals that disperse from the site are indistinguishable from those who suffer mortality. The problem of separating dispersal from mortality is well known and cannot be partitioned without the use of additional techniques, such as direct visual observation or radiotelemetry (Gaines and McClenaghan, 1980).
This study documents that small mammals in the northern Sierra Nevada exhibit considerable temporal and spatial variation in abundance and survival. In addition, we report species-specific habitat affinities, as well as effects of weather and production of cones on some species. Federal forest-fire policy has evolved over the past century from dedicated fire suppression to use of adaptive management to achieve pre-settlement conditions (Stevens and Ruth, 2005). As managers move toward a philosophy of managing forests to maintain overall health of forests, further insights to the basic biology of species inhabiting these ecosystems will be required. In addition, managers will need to take a more adaptive approach to forest management as abiotic factors, such as weather, play critical roles in regulation of key prey species, such as North American deermice. Understanding the many factors that affect abundance of small mammals will help in making decisions for the long-term stability of many species of special concern, such as the California spotted owl.
This study was conducted with funding from the University of California, Davis and United States Department of Agriculture Forest Service, Pacific Southwest Research Station and United States Department of Agriculture Forest Service, Pacific Southwest Region to DAK, DHVV, and MBJ. For assistance with field work, we thank G. Palmer, A. Derrick, S. Connelly, R. LeChalk, J. Csakany, J. Goldman, D. DeJesus, A. Goldman, M. Gilbart, C. Morcos, D. Smith, K. Marsee, and R. Innes.
Submitted 29 November 2007. Accepted 13 March 2008. Editor was Michael L. Kennedy.
BOWERS, M. A. 1995. Use of space and habitat by the eastern chipmunk, Tamias striatus. Journal of Mammalogy 76:12-21.
BOWMAN, J. C., D. SLEEP, G. J. FORBES, AND M. EDWARDS. 2000. The association of small mammals with coarse woody debris at log and stand scales. Forest Ecology and Management 129:119-124.
BURNHAM, K. P., AND D. R. ANDERSON. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer, New York.
BURNHAM, K. P., D. R. ANDERSON, G. C. WHITE, C. BROWNIE, AND K. H. POLLOCK. 1987. Design and analysis methods for fish survival experiments based on release-capture. American Fisheries Society, Monograph 5:1-437.
CARTER, D. W. 1993. The importance of seral stage and coarse woody debris to the abundance and distribution of deer mice on Vancouver Island, British Columbia. M.S. thesis, Simon Fraser University, Burnaby, British Columbia, Canada.
CLARK, T. W., E. ANDERSON, C. DOUGLAS, AND M. STRICKLAND. 1987. Martes americana. Mammalian Species 289:1-8.
COPPETO, S. A., D. A. KELT, D. H. VAN VUREN, J. A. WILSON, AND S. BIGELOW. 2006. Habitat associations of small mammals at two spatial scales in the northern Sierra Nevada. Journal of Mammalogy 87: 402-413.
COTE, M., AND J. FERRON. 2001. Short-term use of different residual forest structures by three sciurid species in a clear-cut boreal landscape. Canadian Journal of Zoology 31:1805-1815.
FAIRBAIRN, D. J. 1977. Why breed early? A study of reproductive tactics in Peromyscus. Canadian Journal of Zoology 55:862-871.
GAINES, M. S., AND L. R. MCCLENAGHAN, JR. 1980. Dispersal in small mammals. Annual Review of Ecology and Systematics 11:163-196.
GANNON, W. L., R. S. SIKES, AND THE ANIMAL CARE AND USE COMMITTEE OF THE AMERICAN SOCIETY OF MAMMALOGISTS. 2007. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy 88:809-823.
GORDON, D. T. 1978. White and red fir cone production in northeastern California: report of a 16-year study. Pacific Southwest Forest and Range Experiment Station, Research Note PSW-330: 1-4.
GRAHAM, A. C., AND L. E. KRUEGER. 2002. Research in adaptive management: working relations and the research process. United States Department of Agriculture Forest Service, Pacific Northwest Research Station Research Paper 538:1-55.
GUNTHER, P. M., B. S. HORN, AND G. D. BABB. 1983. Small mammal populations and food selection in relation to timber harvest practices in the western Cascade Mountains. Northwest Science 57:32-44.
JONES, C. G., R. S. OSTFELD, M. P. RICHARD, E. M. SCHAUBER, AND J. O. WOLFF. 1998. Chain reactions linking acorns to gypsy moth outbreaks and Lyme disease risk. Science 279:1023-1026.
KAUFMAN, D. W., AND G. A. KAUFMAN. 1989. Population biology. Pages 233-270 in Advances in the study of Peromyscus (G. L. Kirkland, Jr. and J. N. Layne, editors).Texas Tech University Press, Lubbock.
KING, J. A., EDITOR. 1968. Biology of Peromyscus. American Society of Mammalogists, Stillwater, Oklahoma.
KIRKLAND, G. L., JR., AND J. N. LAYNE, EDITORS. 1989. Advances in the study of Peromyscus (Rodentia). Texas Tech University Press, Lubbock.
MAHAN, C. G., AND R. H. YAHNER. 1997. Lack of population response by eastern chipmunks (Tamias striatus) to forest fragmentation. American Midland Naturalist 140:382-386.
MARTELL, A. M., AND A. L. MACAULAY. 1981. Food habits of deer mice (Peromyscus maniculatus) in northern Ontario. Canadian Field-Naturalist 95: 319-324.
MCNEELY, J. A. 2004. Nature vs. nurture: managing relationships between forests, agroforestry and wild biodiversity. Agroforestry Systems 61:155-165.
MCSHEA, W. J. 2000. The influence of acorn crops on annual variation in rodent and bird populations. Ecology 81:228-238.
MICHENER, G. R., AND L. LOCKLEAR. 1990. Differential costs of reproductive effort for male and female Richardson's ground squirrels. Ecology 71:855-868.
MILLAR, J. S., AND D. G. L. INNES. 1985. Breeding by Peromyscus maniculatus over an elevational gradient. Canadian Journal of Zoology 63:124-129.
MURIE, J. O., AND G. R. MICHENER., EDITORS. 1984. The biology of ground-dwelling squirrels. University of Nebraska Press, Lincoln.
PETERSON, D. W., D. L. PETERSON, AND G. J. ETTL. 2002. Growth responses of subalpine fir to climatic variability in the Pacific Northwest. Canadian Journal of Forest Research 32:1503-1517.
PROMESSI, R. L., J. O. MATSON, AND M. FLORES. 2004. Diets of nesting northern goshawks in the Warner Mountains, California. Western North American Naturalist 64:359-363.
SAS INSTITUTE, INC. 2000. Statistical analysis system (SAS) user's guide: statistics, version 8. SAS Institute, Inc., Cary, North Carolina.
SCHOENHERR, A. A. 1992. A natural history of California. University of California Press, Berkeley.
SMITH, R. B., P. M. ZACHARIA, R. J. GUTIERREZ, AND W. S. LAHAYE. 1999. The relationship between spotted owl diet and reproductive success in the San Bernardino Mountains, California. Wilson Bulletin 111: 22-29.
SMITH, T. G., AND C. C. MAGUIRE. 2004. Small-mammal relationships with down wood and antelope bitterbrush in ponderosa pine forests of central Oregon. Forest Science 50:711-728.
STEVENS, S. L., AND L. W. RUTH. 2005. Federal forest-fire policy in the United States. Ecological Applications 15:532-542.
SULLIVAN, T. P., AND W. KLENNER. 2000. Response of northwestern chipmunks (Tamias amoenus) to variable habitat structure in young lodgepole pine forest. Canadian Journal of Zoology 78:283-293.
WATERS, J. R., AND C. J. ZABEL. 1998. Abundances of small mammals in fir forests in northeastern California. Journal of Mammalogy 79:1244-1253.
WENSEL, L. C., AND E. C. TURNBLOM. 1998. Adjustment of estimated tree growth rates in northern California conifers for changes in precipitation levels. Canadian Journal of Forest Research 28:1241-1248.
WHITE, G. C., AND K. P. BURNHAM. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study, Supplement 46:120-138.
WOLFF, J. O. 1985. Comparative population ecology of Peromyscus leucopus and Peromyscus maniculatus. Canadian Journal of Zoology 63:1548-1555.
ZABEL, C. J., AND R. G. ANTHONY, EDITORS. 2003. Mammal community dynamics: management and conservation in the coniferous forests of western North America. Cambridge University Press, Oxford, United Kingdom.
JAMES A. WILSON, DOUGLAS A. KELT, * DIRK H. VAN VUREN, AND MICHAEL L. JOHNSON
Department of Wildlife, Fish, and Conservation Biology, University of California, 1 Shields Avenue, Davis, CA 95616
(JAW, DAK, DHVV)
John Muir Institute of the Environment, University of California, 1 Shields Avenue, Davis, CA 95616 (MBJ)
Current address of JAW: Department of Biology, University of Nebraska, 6001 Dodge Street, Omaha, NE 681820040
* Correspondent: email@example.com
TABLE 1--Mean minimum number known alive (SE in parentheses) in white fir (n = 9), red fir (n = 3), mixed-fir (n = 3), and pine-cedar (n = 3) forest in the northern Sierra Nevada, California, 2003 and 2004. Incidental species were captured on a single occasion and not recaptured. Type of Neotoma Peromyscus Tamias Tamias Year forest fuscipes maniculatus quadrimaculatus senex 2003 White fir 0.33 1.89 2.19 2.06 (0.16) (0.30) (0.84) (0.63) Red fir 3.83 6.83 6.83 (0.52) (1.58) (1.58) Mixed-fir 0.33 3.08 3.17 3.17 (0.19) (0.62) (0.83) (0.83) Pine-cedar 1.33 2.92 (0.60) (0.55) 2004 White fir 0.04 19.32 2.53 2.34 (0.04) (1.88) (0.57) (0.74) Red fir 34.60 12.13 17.60 (4.29) (2.13) (2.68) Mixed-fir 0.94 34.13 9.31 3.38 (0.37) (4.27) (1.32) (0.78) Pine-cedar 1.33 12.06 0.06 0.06 (0.52) (1.60) (0.06) (0.06) Type of Spermophilus Spermophilus Year forest Microtus beecheyi lateralis 2003 White fir 0.03 0.06 (0.03) (0.05) Red fir 0.75 7.92 (0.39) (2.60) Mixed-fir 1.83 0.33 (0.59) (0.14) Pine-cedar 1.17 (0.36) 2004 White fir 0.02 0.08 (0.02) (0.04) Red fir 0.27 10.33 (0.11) (2.56) Mixed-fir 0.31 0.50 0.06 (0.15) (0.20) (0.06) Pine-cedar 0.67 0.11 (0.33) (0.08) Type of Tamiasciurus Glaucomys Year forest douglasii sabrinus Incidental 2003 White fir Lepus americanus, Sorex, Mustela frenata, Mephitis mephitis Red fir 0.10 0.17 Lepus americanus, Sorex (0.10) (0.11) Mixed-fir 0.10 0.08 Lepus americanus, Sorex, (0.10) (0.08) Spilogale gracilis Pine-cedar Sorex 2004 White fir 0.09 0.02 Lepus americanus, Sorex (0.04) (0.01) Red fir 0.20 0.07 Sorex, Mustela frenata (0.19) (0.06) Mixed-fir 0.06 Sorex (0.06) Pine-cedar Sorex TABLE 2--Results of analyses using program MARK for four species of rodents in the northern Sierra Nevada, California. All species were analyzed individually using a Cormack-Jolly-Seber model in Program MARK. Best-fit models are shown for each species. [PHI] = probability of survival between primary trap sessions, p = capture probability, Akaike's corrected information coefficient (AICc), adjusted for over-dispersion, and weight of model relative to other less-fit models is given. Data for other species were too few for analysis with program MARK. Species Model Peromyscus maniculatus [PHI] (habitat x t + overwinter + mean cones) p (habitat x t) Spermophilus lateralis [PHI] (t) p (t) Tamias quadrimaculatus [PHI] (habitat x t + overwinter + mean cones) p (habitat x t) Tamias senex [PHI] (habitat x t) p (habitat x t) [PHI] (habitat x t + overwinter) p (habitat x t) Model AICc Weight [??] [PHI] (habitat x t + overwinter + 1,740.6 0.99 1.85 mean cones) p (habitat x t) 358.2 0.96 1.14 [PHI] (t) p (t) 923.5 1.00 1.22 [PHI] (habitat x t + overwinter + mean cones) 683.2 0.60 1.23 p (habitat x t) 684.1 0.39 [PHI] (habitat x t) p (habitat x t) [PHI] (habitat x t + overwinter) p (habitat x t)
|Printer friendly Cite/link Email Feedback|
|Author:||Wilson, James A.; Kelt, Douglas A.; Van Vuren, Dirk H.; Johnson, Michael L.|
|Date:||Sep 1, 2008|
|Previous Article:||Relation between species assemblages of fishes and water quality in salt ponds and sloughs in South San Francisco Bay.|
|Next Article:||Survival and abundance of three species of mice in relation to density of shrubs and prescribed fire in understory of an oak woodland in California.|
|A plan for the struggling Sierra Nevada.|
|Impacts of rainforest logging on non-volant small mammal assemblages in Borneo.|
|Bats' eyes adapted for both daylight and ultraviolet vision.|