Home Range Use by Abert Squirrels: A Comparative Analysis

Abstract

-We compared data from an analysis of space use by Abert squirrels (Sciurus aberti) performed in 1991 with data from a previous study performed in 1971 at the same study site (Farentinos, 1979). In both studies, home range estimates based on the 100% minimum convex polygon (MCP) method were positively correlated with sample size. The number of home range sightings used for calculations in the current study was significantly larger for both males and females when compared to used by Farentinos (1979), as were estimates of home range. Results from the current study also contradict Farentinos' earlier findings males have larger home ranges during the breeding season than the nonbreeding season, and males have larger home ranges during the breeding season than females year around. The method used to estimate home range sizes may have influenced the results, and the 100% MCP method may be of limited use for drawing meaningful biological comparisons, unless sample size is held constant. RESUMEN-Comparamos datos de un anilisis de uso de espacio por ardillas Sciurus aberti llevado a cabo en 1991 con datos de un estudio previo llevado a cabo en 1971 en el mismo sitio de estudio (Farentinos, 1979). En ambos estudios, las estimaciones de rango de hogar que se basaron en el metodo de poligono convexo minimo 100% (MCP) estuvieron positivamente correlacionadas con el tamafio de la muestra. El nuimero de avistamientos de rango de hogar usado para los cilculos en el presente estudio fue significativamente mis grande para ambos machos y hembras cuando se compar6 con el usado por Farentinos (1979), asi como fueron las estimaciones de rango de hogar. Los resultados del presente estudio tambien contradicen los resultados anteriores de Farentinos de que los machos tienen mis grandes rangos de hogar durante la epoca de apareamiento que durante el resto del afio, y de que los machos tienen mis grandes rangos de hogar durante la 6poca de apareamiento que las hembras durante todo el afio. El mitodo usado para estimar los rangos de hogar pudo haber influenciado los resultados, y que el metodo 100% MCP es de uso limitado para obtener comparaciones biol6gicas significativas, a menos que el tamafio de la muestra sea mantenido constante. An animal's home range has been defined as that area traversed by the individual in its normal activities of food gathering, mating and caring for young (Burt, 1943:351). However, published home range values are only an estimate of an animal's activity, and are dependent on the methods used to obtain these estimates. Methods used to calculate home ranges (Mohr, 1947; Hayne, 1949; Mohr and Stumpf, 1966; Jennrich and Turner, 1969; Koeppl et al., 1977; Dixon and Chapman, 1980; Schoener, 1981; Anderson, 1982; Bekoff and Mech, 1984) have different underlying models can produce different results for similar data sets (Boulanger and White, 1990). These results depend on type of movement patterns used by animals (Van Winkle, 1975; Boulanger and White, 1990) and, in some instances, by the sample size used to calculate home range estimates. For example, the 100% minimum convex polygon method usually shows large increases in home range size as sample size increases (Boulanger and White, 1990; Bekoff and Mech, 1984). In a previous study of home range sizes of Abert squirrels (Sciurus aberti) in Boulder Co., Colorado from December 1969 to May 1971, Farentinos (1979) reported males had significantly larger home ranges than females especially during the breeding season. However, Farentinos used the 100% minimum convex polygon (MCP) method (Mohr, 1947) for calculating home range sizes. Because Farentinos' estimates were based on sample sizes between 18 and 254 observations, differences he obThis content downloaded from 207.46.13.157 on Thu, 09 Jun 2016 07:29:37 UTC All use subject to http://about.jstor.org/terms 254 The Southwestern Naturalist vol. 45, no. 3 served might have resulted from his methods rather than the squirrels' ranging patterns. The purpose of this study was to re-examine space use by Abert squirrels using two different methods to estimate home range sizes. By using the same study site and field methods, we were able to compare the 100% MCP measurements obtained by Farentinos (1979) and those obtained using larger sample sizes. If 100% MCP home range sizes for Abert squirrels are dependent on sample size, then home range estimates computed from the present study should be larger than those reported by Farentinos. In addition, 95% minimum convex polygons were calculated for the present set of data. Similar comparisons of space use between males and females, during different times of the year, were made using these estimates of home range size. The 95% MCP method is usually a better predictor of an animal's daily activities because it is less sensitive to sample size and discounts infrequent and unusually long forays (Bekoff and Mech, 1984; Harris et al., 1990). MATERIALS AND METHODS--This study was conducted in a 72 ha stand of Ponderosa pine (Pinus ponderosa) forest located in Boulder, Colorado, commonly called Enchanted Mesa (elevation 1,940 m). A grid system of x and y coordinates was installed on the study area with markers placed approximately every 30 m. This grid system was originally installed by Farentinos in 1969 and recreated for this study from his original notes. Acquisition of home range data followed the procedure outlined in Farentinos (1979). Between May 1989 and October 1991, 13 adult (7 male and 6 female) squirrels were observed year round from sunrise until one-half hour before sunset, or when all squirrels had entered their nests for the evening. Days were normally divided into two observation periods; one-half the number of daylight hours in duration. Dawn to dusk observations were attempted at least one day per week. For both studies, home ranges were estimated only for those squirrels observed a minimum of 10 days for both the breeding and non-breeding season. Farentinos (1979) defined the breeding season as April and May; however, between 1989 and 1991 Abert squirrels mated on Enchanted Mesa between 15 February and 5 June of each year. For these years the breeding season was defined as 1 February to 15 June, and the non-breeding season was defined as 16 June to 31 January of each year. Areas of both 95% and 100% polygons were calculated using the computer package CALHOME (Kie, et al, 1994). We calculated 100% MCP home ranges to compare with Farentinos' (1979) data because he presented only these data, whereas we calculated 95% MCP home range estimates to make predictions on space use patterns. Two-tailed nonparametric Mann-Whitney U-tests or Wilcoxon paired-sample tests were used to test differences between studies, sexes, and seasons, and Spearman's rank correlations were used to test for independence between sample sizes and home range sizes (Zar, 1984). Data are expressed in ha as mean + standard deviation (SD) except where noted. To further test whether differences detected between space use by males and females in Farentinos' (1979) data were due to differences in sample size, a model of home range size as a function of sample size was constructed from data on males using linear regression. Home range values for females were estimated using the number of sightings and compared with the original observed values using a Mann-Whitney Utest. RESULTS--Seventeen (nine female and eight male) adult Abert squirrels inhabited Enchanted Mesa during this study (from May 1989 to October 1991). A total of 4,139 sightings was obtained on 13 (seven female and six male) squirrels met the 10 day minimum sampling criteria during 2,097 h of data collection. The mean number of home range sightings was not significantly different between males (277.3 ? 165.6) and females (301.7 ? 181; U = 58.0, P > 0.05). Using the 100% MCP method, home range size and number of home range sightings were positively correlated for both Farentinos' (1979) data set (rs = 0.733, n = 16, P < 0.002) and the present data set (rs = 0.64, n = 13, P < 0.05). The average sample size used to calculate home range estimates was significantly larger in 1991 (290.5 ? 166.2) than in 1971 (90.9 ? 67.5, U = 182, P < 0.001; Table 1), and therefore overall home range estimates were significantly larger in 1991 (17.82 ? 4.10) than in 1971 (13.73 + 7.80, U = 183, P < 0.001; Table 1). Farentinos (1979) reported males had significantly larger home ranges during the breeding season than during the non-breeding season (Table 1). He also reported during the breeding season, males had significantly larger ranges than females. However, the average number of sightings used by Farentinos (1979) to calculate home range estimates for males during the breeding season (95 + 40.9) This content downloaded from 207.46.13.157 on Thu, 09 Jun 2016 07:29:37 UTC All use subject to http://about.jstor.org/terms September 2000 Halloran and Bekoff-Home range use by Abert squirrels 255 TABLE 1-Number of sightings used in calculations and size of resulting (100%) minimum convex polygon home ranges for adult Abert squirrels (Sciurus aberti) inhabiting Enchanted Mesa during two studies ended in 1971 (Farentinos, 1979) and 1991. Number of sightings Home range size Sex n Mean Range Mean Home range size