Response of rodents to wildfire and livestock grazing in an Arizona desert grassland.
|Abstract:||Livestock grazing and fire influence the composition of desert grassland communities, including their rodent populations. However, there have been few studies of the interactions between grazing and wildfire in arid grasslands of the southwestern United States. We trapped rodents and measured vegetation on grazed versus ungrazed plots before (2001) and following (2002-2007) an intense 15,000 ha wildfire in southeastern Arizona. The fire reduced grass canopy on grazed plots for 2 y and on ungrazed plots for 3 y. Some rodents in the family Cricetidae (genera Baiomys, Reithrodontomys and Sigmodon) were more abundant on ungrazed plots before the fire. Cricetidae as a whole declined following the fire and did not return to preburn levels until the sixth postfire year (2007). Nine of ten cricetid species contributed to this general pattern. By contrast, the abundant species of Heteromyidae (Chaetodipus hispidus, C. baileyi, Perognathus flavus) increased following the fire, especially on ungrazed plots. These results are consistent with a model predicting that fire-caused reductions in grass cover should favor Heteromyidae over Cricetidae. Fires elsewhere in the Southwest have had little impact on rodent populations, but these were smaller and cooler burns with relatively minor effects on vegetation. Future studies of large wildfires of varying intensities would further elucidate the generality of the model.|
Rodents (Environmental aspects)
Livestock (Environmental aspects)
Grasslands (Environmental aspects)
Animal behavior (Research)
Wildfires (Environmental aspects)
Bock, Carl E.
Jones, Zach F.
Kennedy, Linda J.
Bock, Jane H.
|Publication:||Name: The American Midland Naturalist Publisher: University of Notre Dame, Department of Biological Sciences Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Earth sciences Copyright: COPYRIGHT 2011 University of Notre Dame, Department of Biological Sciences ISSN: 0003-0031|
|Issue:||Date: July, 2011 Source Volume: 166 Source Issue: 1|
|Topic:||Event Code: 310 Science & research Canadian Subject Form: Animal behaviour|
|Product:||Product Code: 3523860 Livestock Feeders NAICS Code: 333111 Farm Machinery and Equipment Manufacturing|
|Geographic:||Geographic Scope: United States Geographic Name: Arizona Geographic Code: 1U8AZ Arizona; 1USA United States|
Fire and livestock grazing are major factors affecting the structure and function of arid grasslands, including those of the southwestern United States (McPherson, 1995; Whitford, 1997; Valone and Sauter, 2005; Yarnell et al., 2007). Rodents are an abundant and ecologically significant component of these ecosystems, responsive to habitat condition but also affecting plant community composition through selective herbivory (Curtin et al., 2000; Roth et al., 2009).
We trapped rodents between Nov. 2000 and Nov. 2001 on grazed versus ungrazed grassland plots paired across fencelines separating cattle ranches from a 32 y old livestock exclosure in southeastern Arizona (Jones et al., 2003). Results showed that rodent assemblages in dense ungrazed grasslands were dominated by species in the family Cricetidae, especially in the genera Baiomys (pygmy mice), Reithrodontomys (harvest mice) and Sigmodon (cotton rats). By contrast, species of pocket mice (family Heteromyidae, genera Perognathus and Chaetodipus) were relatively abundant in the sparser grasslands still being grazed by livestock.
Based on these results and studies from other southwestern grass and shrublands (e.g., Brown and Heske, 1990; Hayward et al., 1997), we developed a model predicting that livestock grazing will favor Heteromyidae over Cricetidae by reducing vegetative ground cover (Jones et al., 2003). Results of subsequent studies have been generally consistent with our model (Valone and Sauter, 2005; Geluso, 2009; Rowe et al., 2010).
In a related component of the model (Jones et al., 2003), we predicted that fire also would favor certain cricetid rodents over heteromyids by temporarily reducing ground cover. However, data suitable to test this aspect of the model remain inconclusive, largely because the fires studied were small and/or relatively cool and, therefore, had little impact on vegetation (Fitzgerald et al., 2001; Valone et al., 2002; Killgore et al., 2009). In Apr. 2002 a large and intense wildfire burned across the study area we had sampled for rodents in 2000-2001, including all the grazed and ungrazed plots. This circumstance presented an unusual opportunity to examine the effects of an unplanned wildfire, while having preburn data available for comparison.
STUDY SITE, GRAZING TREATMENTS AND FIRE CONDITIONS
This study took place on the Sonoita Plain, Santa Cruz County, Arizona (31[degrees]39'N, 110[degrees]32'W). Ungrazed plots were on the Appleton-Whittell Research Ranch, a 3160 ha sanctuary of the National Audubon Society from which all livestock were removed in 1968. Grazed plots were on the adjacent Babacomari and Diamond C cattle ranches. Plot elevations ranged from 1400 to 1560 m. Temperatures at the study area vary from a mean daily Jan. minimum of -3.0 C to a mean daily Jun. maximum of 32.6 C. Long term average annual precipitation at the field site is 43 cm, about 60% of which occurs during the Jul.-Aug. summer monsoon (Bock and Bock, 2000).
The Sonoita Valley consists of mesas, slopes and low benches adjacent to largely ephemeral drainages. Warm season perennial bunchgrasses are the dominant vegetation throughout, including species with a Madrean floristic affinity in the genera Aristida, Bouteloua, Eragrostis, Heteropogon and Hilana (Bock and Bock, 1986; McLaughlin et al., 2001).
The northern half of the sanctuary and adjacent grazed lands include scattered mesquite (Prosopis velutina) and a variety of low shrubs such as Baccharis pteronioides and Isocoma tenuisecta, especially on upland mesas and benches. To the south, oak savannas (Quercus arizonica and Q. emoryi) become increasingly abundant. Scattered throughout, but especially on slopes, are a variety of emergent succulents, including Agave palmeri, Yucca elata and Dasylirion wheeleri.
The Diamond C Ranch was grazed by cows and calves moved rotationally for short durations on small (about 50 ha) pastures (R. Jelks, pers. comm.). Stocking densities and rotation times varied substantially among years, but a typical level involved an overall stocking density of about one cow-calf unit/13 ha across the ranch as a whole, and a grazing duration per pasture of less than one week. Grazing on the Babacomari Ranch involved steers and heifers rotated seasonally through relatively large (>500 ha) pastures grazed for about 45 d/y, at a stocking density of about one animal/12 ha across the ranch as a whole (D. Ruppel, pers. comm.).
This study took place between Jun. 2001 and Oct. 2007. On 29 Apr. 2002 a wildfire ignited at the northern edge of the San Rafael Valley south of the Research Ranch, passed through the sanctuary and adjacent cattle ranches on 30 Apr. and eventually consumed about 15,000 ha in Santa Cruz and Cochise counties. There are four reasons why the fire was unusually intense, such that it burned both grazed and ungrazed lands. First, the preceding summer had been wet and productive, resulting in abundant fine fuels. Second, the winter of 2001-2002 had experienced a drought, such that these fuels were very dry going into spring. Third, the fire occurred in Apr., which is before the start of the growing season in this part of Arizona, so that nearly all grass cover consisted of dead growth from the previous year. Finally, winds during the fire were sustained at speeds up to 65 km/h.
We trapped rodents on six cross fence sites around the northern, western and eastern perimeters of the Research Ranch, where a boundary fence separated grazed from adjacent ungrazed plots. Four sites were in mesquite grassland whereas two were in oak savannah. Each plot included two parallel 300 m traplines 50 m apart, beginning 20 m from the boundary fence and running perpendicular to it.
During each trapping event, 120 Sherman livetraps were set out singly at 10 m intervals along the four 300 m traplines (two grazed, two ungrazed, 30 traps each). Traps were set for three consecutive nights, except in 2007 when traps were run for two nights. Traps were opened in the evenings and closed in the mornings after captured animals were identified, marked by flu clipping or with ink and released.
We sampled each plot once between Jun. and early Nov. 2001 (preburn), once between Jun. and Aug. 2002 and twice per year between Jun. and Oct. of 2003 through 2007. The lack of fall trapping in 2002 could have affected results compared to the other years. However, 2002 capture rates of Cricetidae and Heteromyidae were intermediate between those of 2001 and 2003, suggesting that continued trapping in fall 2002 would not have affected our overall conclusions. We did not trap in winter (Dec.-Feb.) because some heteromyids may hibernate during this season in southern Arizona (Hoffmeister, 1986).
We computed the number of different animals (not previously marked) captured/100 trap nights as a measure of relative abundance of each species among plots and years. Total sample effort over the 7 y was 24,460 trap nights, resulting in capture of 1662 rodents of 15 species. Nomenclature follows Wilson and Reeder (2005).
We were unable to collect data on fuel loads or moisture content before the fire because it was an unplanned event. As an alternative, in May 2002 (1 mo after the fire) we recorded whether ground vegetation was blackened or not, at 200 points spaced at 1 m intervals along each of the rodent traplines. This gave us a measure of the completeness with which the fire burned standing dead grasses, forbs and shrubs on grazed versus ungrazed areas before the first postfire growing season had begun.
We measured vegetative canopy at the height of the growing season in late summer 2001-2007, on 20 grazed and 20 ungrazed 20 m diameter plots that were part of a long term monitoring project at the field site that began in the mid 1980s (Bock et al., 2007). Four 10 m transects were set out in cardinal compass directions from each plot center. We placed a 20 x 50 cm sampling frame at 2 m intervals along each transect, and visually estimated percent canopy cover for each plant species along with percent bare ground (no plant canopy) inside each frame (n = 20 frames/plot). Frame data were averaged to generate single percent cover values for grasses, forbs, shrubs and bare ground on each plot.
We compared percent of the ground blackened by the fire on grazed versus ungrazed traplines using the Mann-Whitney U test. The following variables met assumptions for parametric statistical analyses (Zar, 2010): (1) captures/100 trap nights for total Cricetidae, (2) captures/100 trap nights for total Heteromyidae, (3) percent grass canopy and (4) percent bare ground (no vegetation canopy). We analyzed the vegetation data using repeated measures ANOVA, with grazing as the treatment variable and year as the repeated measure. We interpreted significant interactions between treatment and year as evidence that the fire changed pre-burn differences between grazed and ungrazed vegetation plots.
Because the paired grazed and ungrazed traplines were in close proximity and matched in terms of habitat, rodent capture data across the fencelines could not be considered statistically independent. Therefore, we analyzed trapping results for combined Cricetidae and combined Heteromyidae in the following two ways. First, we pooled data from grazed and ungrazed traplines into single values for each family on each of the six sites; and then analyzed these using repeated measures ANOVA to test for significant year effects on abundance. Second, we subtracted each ungrazed value from its paired grazed value and analyzed these differences using repeated measures ANOVA to test for differences in grazing effects on rodents across the years. In both cases, we applied post-hoc Fisher's PLSD tests to look for significant pairwise differences between years.
We chose not to analyze trapping data for individual rodent species statistically because results for many species failed to meet assumptions necessary for parametric analyses, and because the large number of tests (15 species) would have increased the probability of Type-I errors to a very high level (Zar, 2010). However, because we found statistically significant differences among treatments and years for the two rodent families as a whole, we explored data for individual species heuristically to determine which species contributed in what particular ways to the family-wide patterns.
All statistical tests were performed in Statview 5.0.1 (SAS Publishing, 1999), with P < 0.05 considered significant. Unless otherwise noted, values shown are means + SE.
Precipitation the year before the fire (2001) was 43 cm; and it varied from 42 to 47 cm in 2004-2007, all at or near the long-term average of 43 cm. However, the 2 y following the fire were relatively dry: 22 cm in 2002 and 27 cm in 2003. Proportions of the ground charred by the Ryan Fire ranged from 99 to 100% on the six ungrazed plots, significantly higher than on the six grazed plots (range: 41 to 95%; Z = 2.89, P = 0.004).
Because grasses comprised 75% of vegetation cover, percent grass canopy and percent unvegetated ground were essentially mirror images of one another (Fig. 1). The fire reduced grass canopy for 2 y on grazed plots and 3 y on ungrazed plots (Fig. 1), resulting in a significant interaction between treatment and year (Grazing effect: [F.sub.1,38] = 37.2, P < 0.001; year effect: [F.sub.39,6] = 46.4, P < 0.001; interaction: [F.sub.1,38,6,6] = 31.3, P < 0.001). Grass cover was relatively high on grazed plots in 2004 compared to other years, perhaps due to a residual effect of reductions in livestock numbers that had been necessitated by the fire. Ungrazed grass cover returned to levels similar to the preburned condition by 2005 (Fig. 1). Results were similar for percent unvegetated ground (Grazing effect: [F.sub.1,38] = 43.1, P < 0.001; year effect: [F.sub.39,6] = 24.0, P < 0.001; interaction: [F.sub.1,38,6,6] = 23.7, P < 0.001).
[FIGURE 1 OMITTED]
We captured 15 species of rodents (Table 1). Five species, all in the family Cricetidae, were trapped more frequently in ungrazed than in grazed grasslands in 2001, prior to the burn: Baiomys taylori, Reithrodontomys fulvescens, R. megalotis, Sigmodon fulviventer and S. ochrognathus. Captures of each of these species declined following the 2002 fire and did not increase again until 2006 or 2007 (Table 1). A third species of Sigmodon (S. arizonae) was not captured in 2001 through 2005 but was captured more frequently on ungrazed plots than on grazed plots in 2006 and 2007. We captured three other cricetids (Onychomys torridus, Peromyscus boylii and P. maniculatus) somewhat more frequently on grazed than on ungrazed plots prior to the burn, and these three species all declined on the grazed plots through 2005 or 2006 (Table 1). The only cricetid showing no obvious response to the fire was Neotoma albigula.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Captures of Cricetidae as a whole declined following the fire, and did not return to preburn levels until the sixth post-fire year (Fig. 2a; [F.sub.5,6] = 8.19, P < 0.001). Fire neutralized the negative impacts of grazing on the Cricetidae as whole, an effect that persisted until 2007, the sixth post-fire year (Fig. 2b; [F.sub.5,6] = 10.18, P < 0.001). Grazing effects were not significantly different between 2002 and 2006; but the trends were positive in 2002 and 2003, essentially neutral in 2004 and 2005 and negative in 2006 (Fig. 2b).
The two most abundant species in the family Heteromyidae (Chaetodipus hispidus and Perognathusflavus) were more abundant on grazed plots prior to the burn (Table 1; see also Jones et al., 2003) and also increased following the fire, especially on ungrazed plots (Table 1). Chaetodipus baileyi was scarce prior to the burn and generally was captured more frequently on grazed plots following the fire (Table 1). Two other heteromyids (C. intermedius and Dipodomys merriami) were scarce throughout the study.
Captures of combined Heteromyidae increased through 2004 (the third post-fire year), declined precipitously in 2005 and then increased through 2007 (Fig. 3a; [F.sub.5,6] = 6.43, P < 0.001). Grazing effects on the Heteromyidae were positive prior to the burn and again in the sixth post-fire year (2007) but not in the intervening years (Fig. 3b; [F.sub.5,6] = 3.73, P = 0.007).
Reduced grass canopy following the 2002 Ryan Fire can be attributed to the intensity and completeness of the burn, especially in ungrazed areas, coupled with a 2 y postfire drought (2002-2003; Fig. 1). Rodent responses to the burn were substantially in accord with our earlier model predicting that fire-caused reductions in ground cover would favor species in the family Heteromyidae over species in the family Cricetidae (Jones et al., 2003). Cricetidae as a whole declined following the fire and did not return to pre-burn levels for 6 y (Fig. 2a). By contrast, captures of Heteromyidae increased for 3 y following the fire (2002-2004; Fig. 3a). We cannot explain the temporary drop in heteromyid captures in 2005 because it appeared unrelated to postfire changes in ground cover (Fig. 1).
Cricetidae were more abundant in ungrazed than in grazed grasslands before the fire in 2001 and again in 2007 but not in the intervening years (Fig. 2b). This result can be attributed to the fact that the fire spared more ground cover in grazed than in ungrazed areas, essentially eliminating preburn differences in grass cover for three postfire years (Fig. 1). Heteromyidae showed the opposite pattern, being more common in grazed grasslands in 2001 and again in 2007 but somewhat more common in ungrazed grasslands between 2002 and 2006 (Fig. 3b). In this case, it appears the greater intensity of the fire in ungrazed habitat temporarily rendered these areas more suitable to heteromyids than nearby grazed sites.
Fire and grazing affected individual rodent species in ways that were generally consistent with their known habitat associations elsewhere (Table 1). Those Cricetidae that were more common on ungrazed plots before the fire, and also negatively impacted by it, usually have been found in greatest numbers in relatively tall, dense or structurally complex grasslands: Baiomys taylori (Eshelman and Cameron, 1987), Reithrodontomys spp. (Webster and Jones, 1982; Johnston and Anthony, 2008; Cameron et al., 2009) and Sigmodon spp. (Baker and Shump, 1978a, b; Geluso, 2009). By contrast, Peromyscus maniculatus is a habitat generalist (Hoffmeister, 1986) that was relatively common on all our plots regardless of grazing or fire. However, even this species declined somewhat following the burn (Table 1). Onychomys torridus and P. boylii are species of scrub and brush habitats (McCarty, 1975; Kalcounis Rueppell and Spoon, 2009) that were trapped more frequently in grazed grasslands but that apparently also declined following fire (Table 1).
Among the Heteromyidae, Chaetodipus baileyi is a species of deserts and desert scrub (Paulson, 1988a) that was scarce in our study area before the burn (Table 1). Captures rose in subsequent years, especially in grazed areas. The other two heteromyids that were common in our study area, Chaetodipus hispidus and Perognathusflavus, are widely distributed in various grasslands but especially in those that are dry and relatively sparse (Paulson, 1988b; Best and Skupski, 1994). Both species increased in ungrazed grasslands following the Ryan Fire (Table 1).
All but one of the nine Cricetidae in our study area declined following the Ryan Fire, on either grazed or ungrazed plots or both, whereas the three common Heteromyidae all increased. These results support the legitimacy of considering the families as a whole in tests of our model predicting how fire would affect rodent populations by altering habitat structure (Jones et al., 2003).
A limitation to our study is that it involved only one burn in one area. Fires elsewhere in southwestern grass/shrublands have had relatively little impact on rodent populations (Fitzgerald et al., 2001; Valone et al., 2002; Killgore et al., 2009). However, these studies involved prescribed burns that were relatively small, or cool and incomplete, or both. By contrast, the wildfire we studied was of a size and intensity likely typical of prehistoric conditions (Bahre, 1991); and it had significant impacts on both vegetation and rodent populations. In many ways the Ryan Fire was 'the perfect storm,' because of the long term absence of livestock from ungrazed areas, because of fuel and atmospheric conditions at the time of the burn and because of the subsequent drought. Fnture studies of large wildfires of varying intensities would further elucidate the generality of our model.
Arid grasslands outside North America lack Heteromyidae; but assemblages of rodents in the families Cricetidae (South America) and Muridae (Africa and Australia) include species that have responded negatively or positively to fire and grazing, depending upon their particular habitat requirements (e.g., Tabeni and Ojeda, 2003; Letnic and Dickman, 2005; Yarnell et al., 2007). In all such grasslands, including those of the southwestern United States (Whitford, 1997), landscape mosaics of areas that are grazed or periodically burned; and especially those long protected from both, are likely to maintain the highest regional diversity of rodents.
Acknowledgments.--This study was supported by the Arizona Department of Game and Fish, by the Ecology Program of the National Science Foundation, by the National Audubon Society and by the Research Ranch Foundation. R.Jelks, B. Brophy and D. Ruppel kindly permitted access to the Diamond C and Babacomari ranches. We thank the following individuals for field assistance: K. Anderson, B. Audsley, K. Bishop, T. Crook, D. Goodheim, B. Loomis, L.Jones, A. Marshall, L. Schevets, C. Venot and C. Wonka, as well as students in the mammalogy classes at Colorado College in 2005 and 2006 and at Eastern New Mexico University in 2007. Two anonymous reviewers provided constructive comments on a previous draft of the manuscript.
BAHRE, C.J. 1991. A legacy of change: historic impact on vegetation in the Arizona borderlands. University of Arizona Press, Tucson, Arizona. 231 p.
BAKER, R. H. AND K. A. SHUMP, JR. 1978a. Sigmodonfulviventer. Mammalian Species no. 94 (1-4). American Society of Mammalogists, New York.
--AND--. 1978b. Sigmodon ochrognathus. Mammalian Species no. 97 (1-2). American Society of Mammalogists, New York.
BEST, T. L. AND M. P. SKUPSKI. 1994. Perognathusflavus. Mammalian Species no. 417 (1-10). American Society of Mammalogists, New York.
BOOK, C. E. ANDJ. H. BOCK. 2000. The view from Bald Hill: thirty years in an Arizona grassland. University of California Press, Berkeley, California. 197 p.
--,--, L. KENNEDY AND Z. F. JONES. 2007. Spread of non-native grasses into grazed versus ungrazed desert grassland. J. Arid Env., 71:229-235.
BOCK, J. H. AND C. E. BOCK 1986. Habitat relationships of some native perennial grasses in southeastern Arizona. Desert Plants, 8:3-14.
BROWN, J. H. AND E. J. HESKE. 1990. Control of a desert-grassland transition by a keystone rodent guild. Science, 250:1705-1707.
CAMERON, G. N., J. M. WILLIAM AND B. L. KRUCHEK. 2009. Seasonal dynamics of small mammals and vegetation in a gulf cordgrass community. Southwest. Nat., 54:237-247.
CURTIN, C. G., D. A. KELT, T. C. FREY ANDJ. H. BROWN. 2000. On the role of small mammals in mediating climatically driven vegetation change. Ecol. Lett., 3:309-317.
ESHELMAN, B. D. AND G. N. CAMERON. 1987. Baiomys taylori. Mammalian Species no. 285 (1-7). American Society of Mammalogists, New York.
FITZGERALD, C. S., P. R. KRAUSMAN AND M. L. MORRISON. 2001. Short-term impacts of prescribed fire on a rodent community in desert grasslands. Southwest. Nat., 46:332-337.
GELUSO, K. 2009. Records of the yellow-nosed cotton rat (Sigmodon ochrognathus) in southwestern New Mexico. West. N. Am. Nat., 69:548-550.
HAYWARD, B., E. J. HESKE AND C. W. PAINTER. 1997. Effects of livestock grazing on small mammals at a desert cienega. J. Wildl. Manage., 61:123-129.
HOFFMEISTER, D. F. 1986. Mammals of Arizona. University of Arizona Press, Tucson. 602 p.
JOHNSTON, A. N. AND R. G. ANTHONY. 2008. Small-mammal microhabitat associations and response to grazing in Oregon. J. Wildl. Manage., 72:1736-1746.
JONES, Z. F., C. E. BOCK AND J. H. BOCK, 2003. Rodent communities in a grazed and ungrazed Arizona grassland, and a model of habitat relationships among rodents in southwestern grass/ shrublands. Am. Midl. Nat., 149:384-394.
KALCOUNIS-RUEPPELL, M. C. AND T. R. SPOON. 2009. Peromyscus boylii (Rodentia: Cricetidae). Mammalian Species no. 838 (1-14). American Society of Mammalogists, New York.
KILLGORE, A., E. JACKSON AND W. G. WHITFORD. 2009. Fire in Chihuahuan desert grassland: short-term effects on vegetation, small mammal populations, and faunal pedoturbation. J. Arid Env., 73:1029-1034.
LETNIC, M. AND C. R. DICKMAN. 2005. The response of small mammals to patches regenerating after fire and rainfall in the Simpson Desert, central Australia. Austral. Ecol., 30:24-39.
McCARTY, R. 1975. Onychomys torridus. Mammalian Species no. 59 (1-5). American Society of Mammalogists, New York.
McLAUGHLIN, S. P., E. L. GEIGER AND J. E. BOWERS. 2001. Flora of the Appleton-Whittell Research Ranch, northeastern Santa Cruz County, Arizona. J. Ariz.-Nev. Acad. Sci., 33:113-131.
MCPHERSON, G. R. 1995. The role of fire in the desert grasslands, p. 130-151. In: M. P. McCLaran and T. R. Van Devender (eds.). The desert grassland. University of Arizona Press, Tucson, Arizona.
PAULSON, D. D. 1988a. Chaetodipus baileyi. Mammalian Species no. 297 (1-7). American Society of Mammatogists, New York.
--. 1988b. Chaetodipus hispidus. Mammalian Species no. 320 (1-4). American Society of Mammalogists, New York.
ROWE, R. J., J. A. FINARELLI AND E. A. RICKART. 2010. Range dynamics of small mammals along an elevational gradient over an 80-year interval. Glob. Change Biol., 16:2930-2943.
ROTH, G. A., W. G. WHITFORD AND Y. STEINBERGER. 2009. Small mammal herbivory: feedbacks that help maintain desertified ecosystems. J. Arid Env., 73:62-65.
SAS PUBLISHING. 1999. Statview 5.0.1. SAS Institute, Inc., Cary, North Carolina. 528 p.
TABENI, S. AND R. A. OJEDA. 2003. Assessing mammal responses to perturbations in temperate aridlands of Argentina. J. Arid Env., 55:715-726.
VALONE, T. J., S. E. NOra)ELL AND S. K. MORGAN ERNEST. 2002. Effects of fire and grazing on an arid grassland ecosystem. Southwest. Nat., 47:557-565.
--AND P. SAUTER. 2005. Effects of long-term cattle exclosure on vegetation and rodents at a desertified arid grassland site. J. Arid Env., 61:161-170.
WEBSTER, D. ANn J. K. JONES, JR. 1982. Reithrodontomys megalotis. Mammalian Species no. 167 (1-5). American Society of Mammalogists, New York.
WHITVORD, W. G. 1997. Desertification and animal biodiversity in the desert grasslands of North America. J. Arid Env., 37:709-720.
WILSON, D. E. AND D. M. REEDER. 2005. Mammal species of the world: a taxonomic and geographic reference, 3rd ed. Johns Hopkins University Press, Baltimore, Maryland. 2142 p.
YARELL, R. W., D. M. SCOTT, C. T. CHIMOMBA AND D.J. METCALFE. 2007. Untangling the roles of fire, grazing and rainfall on small mammal communities in grassland ecosystems. Oecologia, 154:387-402.
ZAR, J. H. 2010. Biostatistical analysis, 5th ed. Pearson Prentice-Hall, Upper Saddle River, New Jersey. 944 p.
SUBMITTED 22 JULY 2010
ACCEPTED 10 JANUARY 2011
CARL E. BOCK (1)
Ecology and Evolutionary Biology, University of Colorado, Boulder 80309
ZACH F. JONES
Department of Biology, Eastern New Mexico University, Portales 88130
LINDA J. KENNEDY
Appleton-Whittell Research Ranch, National Audubon Society, HC1 Box 44, Elgin, Arizona 85611
JANE H. BOCK
Ecology and Evolutionary Biology, University of Colorado, Boulder 80309
(1) Corresponding author: Telephone: (970) 59-0343; e-mail: firstname.lastname@example.org.
TABLE 1.--Mean (standard error) captures-100 trap nights of fifteen rodent species on six grazed and six ungrazed plots before (2001) and after (2002-2007) a wildfire that burned an Arizona desert grassland in Apr. 2002 Year Family and species Treatment 2001 2002 Heterornyidae Chaetodipus baileyi Grazed 0.09 (0.06) 0.56 (0.46) Ungrazed 0 0.09 (0.09) Chaetodipus hispidus Grazed 2.52 (0.75) 0.93 (0.54) Ungrazed 1.59 (0.57) 2.23 (0.60) Chaetodipus intermedius Grazed 0 0.19 (0.19) Ungrazed 0 0.09 (0.09) Dipodomys merriami Grazed 0.09 (0.09) 0 Ungrazed 0 0 Perognathus flavus Grazed 1.66 (0.88) 1.95 (1.13) Ungrazed 0.65 (0.34) 1.86 (0.61) Cricetidae Baiomys taylori Grazed 0.37 (0.37) 0.56 (0.56) Ungrazed 1.17 (0.46) 0.28 (0.28) Neotoma albigula Grazed 0.09 (0.06) 0.09 (0.09) Ungrazed 0.19 (0.12) 0.19 (0.19) Onychomys torridus Grazed 0.37 (0.37) 0 Ungrazed 0.19 (0.12) 0.09 (0.09) Peromyscus boylii Grazed 0.28 (0.28) 0.09 (0.09) Ungrazed 0.09 (0.09) 0 Peromyscus maniculatus * Grazed 4.20 (2.63) 2.05 (0.99) Ungrazed 2.15 (1.12) 2.61 (1.76) Reithrodontomys fulvescens Grazed 0.75 (0.45) 0.09 (0.09) Ungrazed 2.89 (1.12) 0.09 (0.09) Reithrodontomys megalotis Grazed 2.20 (0.93) 2.71 (1.28) Ungrazed 3.05 (1.30) 1.03 (0.57) Sigmodon arizonae Grazed 0 0 Ungrazed 0 0 Sigmodon fulviventer Grazed 0.28 (0.28) 0.09 (0.09) Ungrazed 2.61 (1.18) 0 Sigmodon ochrognathus Grazed 0 0 Ungrazed 0.65 (0.42) 0 Year Family and species Treatment 2003 2004 Heterornyidae Chaetodipus baileyi Grazed 0.79 (0.49) 1.44 (0.88) Ungrazed 0.28 (0.18) 0.79 (0.31) Chaetodipus hispidus Grazed 2.32 (0.68) 1.53 (0.63) Ungrazed 5.00 (0.43) 2.73 (0.57) Chaetodipus intermedius Grazed 0 0 Ungrazed 0 0 Dipodomys merriami Grazed 0 0 Ungrazed 0.05 (0.05) 0.19 (0.19) Perognathus flavus Grazed 1.20 (0.53) 4.21 (1.27) Ungrazed 1.34 (0.66) 5.05 (2.07) Cricetidae Baiomys taylori Grazed 0 0 Ungrazed 0.05 (0.05) 0.19 (0.19) Neotoma albigula Grazed 0.14 (0.09) 0 Ungrazed 0.09 (0.09) 0.23 (0.23) Onychomys torridus Grazed 0.05 (0.05) 0 Ungrazed 0 0.23 (0.13) Peromyscus boylii Grazed 0 0 Ungrazed 0 0 Peromyscus maniculatus * Grazed 1.62 (0.83) 1.21 (0.61) Ungrazed 1.25 (0.65) 0.93 (0.47) Reithrodontomys fulvescens Grazed 0 0 Ungrazed 0 0.05 (0.05) Reithrodontomys megalotis Grazed 0.56 (0.45) 0.56 (0.36) Ungrazed 0.09 (0.06) 0.33 (0.15) Sigmodon arizonae Grazed 0 0 Ungrazed 0 0 Sigmodon fulviventer Grazed 0 0 Ungrazed 0 0.14 (0.14) Sigmodon ochrognathus Grazed 0 0 Ungrazed 0.05 (0.05) 0.05 (0.05) Year Family and species Treatment 2005 2006 Heterornyidae Chaetodipus baileyi Grazed 0.23 (0.23) 1.16 (0.78) Ungrazed 0.33 (0.27) 0.42 (0.20) Chaetodipus hispidus Grazed 0.46 (0.28) 1.30 (0.26) Ungrazed 0.60 (0.26) 2.73 (0.99) Chaetodipus intermedius Grazed 0 0 Ungrazed 0 0 Dipodomys merriami Grazed 0 0.19 (0.19) Ungrazed 0 0 Perognathus flavus Grazed 0.51 (0.35) 0.51 (0.22) Ungrazed 0.74 (0.47) 0.70 (0.38) Cricetidae Baiomys taylori Grazed 0.05 (0.05) 0 Ungrazed 0.14 (0.10) 0.14 (0.14) Neotoma albigula Grazed 0 0 Ungrazed 0.19 (0.12) 0.14 (0.10) Onychomys torridus Grazed 0.14 (0.10) 0.05 (0.05) Ungrazed 0.14 (0.06) 0.28 (0.18) Peromyscus boylii Grazed 0.09 (0.06) 0.09 (0.09) Ungrazed 0 0.05 (0.05) Peromyscus maniculatus * Grazed 0.83 (0.36) 2.50 (1.23) Ungrazed 0.56 (0.36) 1.25 (0.63) Reithrodontomys fulvescens Grazed 0 0 Ungrazed 0 0 Reithrodontomys megalotis Grazed 0 0.05 (0.05) Ungrazed 0.14 (0.14) 0.19 (0.09) Sigmodon arizonae Grazed 0 0.09 (0.09) Ungrazed 0 0.74 (0.27) Sigmodon fulviventer Grazed 0 0.28 (0.14) Ungrazed 0 1.07 (0.73) Sigmodon ochrognathus Grazed 0.05 (0.05) 0 Ungrazed 0.05 (0.05) 0.51 (0.24) Year Family and species Treatment 2007 Heterornyidae Chaetodipus baileyi Grazed 2.50 (1.55) Ungrazed 1.02 (0.56) Chaetodipus hispidus Grazed 1.76 (0.64) Ungrazed 1.20 (0.47) Chaetodipus intermedius Grazed 0 Ungrazed 0 Dipodomys merriami Grazed 0.28 (0.14) Ungrazed 0.28 (0.14) Perognathus flavus Grazed 1.11 (0.58) Ungrazed 0.56 (0.28) Cricetidae Baiomys taylori Grazed 0.47 (0.16) Ungrazed 0.65 (0.26) Neotoma albigula Grazed 0.23 (0.11) Ungrazed 0.14 (0.10) Onychomys torridus Grazed 0.42 (0.14) Ungrazed 0.28 (0.14) Peromyscus boylii Grazed 0.23 (0.23) Ungrazed 0.14 (0.10) Peromyscus maniculatus * Grazed 2.31 (0.52) Ungrazed 1.25 (0.68) Reithrodontomys fulvescens Grazed 0.19 (0.09) Ungrazed 0.60 (0.24) Reithrodontomys megalotis Grazed 0.37 (0.14) Ungrazed 0.83 (0.45) Sigmodon arizonae Grazed 0.60 (0.32) Ungrazed 2.73 (0.80) Sigmodon fulviventer Grazed 1.95 (0.81) Ungrazed 5.83 (1.19) Sigmodon ochrognathus Grazed 0.05 (0.05) Ungrazed 0.14 (0.06) * Captures of deer mice (Peromyscus maniealatus) likely also included some white-footed mice (P. leucopus), as these two species are nearly impossible to distinguish in the field (Hoffmeister, 1986)
|Gale Copyright:||Copyright 2011 Gale, Cengage Learning. All rights reserved.|