Effect of gnawing byMicrocavia australis(Rodentia, Caviidae) onGeoffroea decorticans(Leguminosae) plants

Journal of Arid Environments (1999) 41: 79]85 Article No. jare.1998.0470 Available online at http:rrwww.idealibrary.com on

Effect of gnawing by Microcavia australis ( Rodentia, Caviidae) on Geoffroea decorticans ( Leguminosae) plants

Marcelo F. Tognelli*, Carlos E. Borghi† & Claudia M. Campos† * Department of Wildlife, Fish, and Conservation Biology, University of California, Davis, CA 95616, U.S.A. † Unidad de Zoologıa ´ y Ecologıa ´ Animal-IADIZA, CC 507-CP 5500-Mendoza, Argentina (Received 20 November 1997, accepted 3 November 1998) The impact of gnawing by Microcavia australis on the survival of Geoffroea decorticans plants was studied in the Monte Desert of Argentina. Sampled plants ( N s 105) were selected randomly from two different colonies of M. australis. For each plant sampled the following properties were measured: the diameter of the trunk at ground level; the distance to the nearest burrow of M. australis; the proportion of the trunk perimeter that had been gnawed; and the condition of the plants. It was found that gnawing by M. australis has a negative impact on the survival of G. decorticans plants. Damage of the plants is inversely related to the distance from the burrow systems ( rs s y0.283, p - 0.003) and to the proportion of the trunk perimeter gnawed ( rs s y0.554, p - 0.0001). The differential survival of plants with a larger diameter trunk and of plants farther away from M. australis colonies may have long-term effects on the abundance and distribution of G. decorticans. q 1999 Academic Press Keywords: herbivory; Monte Desert; plant survival; Microcavia australis; Geoffroea decorticans; plant]animal interaction

Introduction Knowledge of factors that limit the establishment of long-lived perennial plants in desert communities is necessary for a better understanding of species distributions, abundances and diversity in these communities (McAuliffe, 1986). Mammal community composition can have an important effect on vegetation structure and species diversity (Brown & Heske, 1990; Asquith et al., 1997). Locally, mammalian herbivores may have significant effects on the diversity and abundance of plant species, either by direct consumption (Huntly & Inouye, 1988; Longland, 1991; Danell et al., 1994; Huntly & Reichman, 1994) or as an indirect consequence of physical disturbance due to herbivory (Platt, 1975; Inouye et al., 1987; Huntly & Reichman, 1994). Herbivore foraging may also act as an important regulating mechanism in the establishment of seedlings at an 0140]1963r99r010079 q 07 $30.00r0

q 1999 Academic Press

80

M. F. TOGNELLI ET AL.

early stage in the succession of plant communities (McAuliffe, 1986; Sortk, 1987; Gill & Marks, 1991; Molofsky & Fisher, 1993; Ostfeld & Canham, 1993). The impact of herbivores on the growth rate of plants depends on the type of tissue removed as well as the time of the attack in relation to the development of the plant (Crawley, 1983). Because plant vascular tissues occur immediately beneath the bark (Taiz & Zeiger, 1991), partial or total debarking around the perimeter of the trunk will interrupt the carbohydrate supply link between the leaves and the roots (Crawley, 1983), thereby affecting the survival of the plant. In arid and semi-arid ecosystems, herbivory plays a very important role in shaping the composition, function and structure of plant communities (McAuliffe, 1986; Bucher, 1987). In Argentina, the majority of studies on plant]herbivore interactions have been conducted to assess the effect of domestic livestock on grasslands or savannas (e.g. Leon ´ et al., 1984; Sala et al., 1986; Sala, 1988). In the Monte Desert specifically, few studies have been conducted to assess the interactions between wild herbivores and native vegetation (Kufner & Chambuleyron, 1993; Campos, 1997; Campos & Ojeda, 1997; Borruel et al., 1998). The importance of herbivory by medium-sized mammals in the Monte Desert was assessed by Borruel et al. (1998). The cuis, Microcavia australis (Geoffroy and D’Orbigny), is a diurnal herbivorous caviomorph rodent that is widely distributed in Argentina (Cabrera, 1953), reaching high densities in arid zones like the Monte Desert (Borruel et al., 1998). It lives in colonies and locates its burrow systems under vegetation that show a weeping structure (Tognelli et al., 1995). It typically feeds on dicot leaves (Campos, 1997) and has been observed climbing trees and shrubs up to 4 m high in order to feed (Rood, 1970; Mares et al., 1977). In the northern Monte Desert, Mares et al. (1977) reported observations of M. australis stripping the bark from young brea trees ( Cercidium praecox [Ruiz & Pav. Harms]). In the central Monte, we observed individuals of M. australis climbing and gnawing plants of Geoffroea decorticans (Gill. ex Hook.) (chanar ˜ ) that were growing close to their colonies. Also, we observed that in general trees with a larger trunk diameter were less affected by gnawing than trees with smaller diameter trunks. Geoffroea decorticans is a tree characterized by the green colour of the trunk and branches through which it photosynthesizes. It reproduces by seed or by rhizomes (Marzocca, 1957), resulting in a very clumped spatial arrangement of plants of different age and size. In this paper, we examined the effect of gnawing by cuises on chanar ˜ plants, and tested for four specific predictions: (1) plant condition should be worse on highly gnawed plants; (2) damage to plants by gnawing should be greater near cuis colonies than further away; (3) damage by gnawing should be greater on plants with a smaller trunk diameter than on plants with thicker trunks; and (4) heavy gnawing of plants should affect plant survival. If these predictions hold, we would expect a localized (closer to colonies) and differential (on trees of smaller diameter) impact on G. decorticans plants, and this may have long-term effects on the abundance and distribution of this species in particular, and on the Monte Desert forest communities in general.

Materials and methods

˜ ˜´ Biosphere Reserve, located in the centralThe study was conducted in the Nacunan western part of the Mendoza plain (348029 S, 678589 W), 200 km south-east of Mendoza. The Reserve is in the Monte phytogeographic province and contains approximately 13,000 ha of xerophytic vegetation. The climate is semi-arid with a long-term average annual rainfall of 330 mm, concentrated in the summer months. The

EFFECT OF GNAWING BY M. AUSTRALIS

81

vegetation consists of open forest and shrubland steppes (Roig, 1971), which have been protected from grazing and human disturbance since 1970. Open mesquite forest is the most representative plant community, dominated by Prosopis flexuosa DC var. depressa, and accompanied by Geoffroea decorticans, a shrub layer of Larrea divaricata Cav., L. cuneifolia Cav., Atriplex lampa Gill. ex Moq., Lycium chilense Miers ex Bertero, Junellia aspera (Gill. & Hook.) Moldenke, and a species-rich herbaceous layer (Roig, 1971). Within the reserve M. australis is most abundantly found in the mesquite community (Contreras & Roig, 1978). Sampling was performed in two large colonies of M. australis that have numerous G. decorticans plants in their surroundings area. Fifty and 55 randomly selected plants, from two colonies of M. australis, respectively, were sampled. For each plant the following characteristics were measured: (a) the proportion of the total trunk perimeter that had been gnawed, as an indicator of importance of damage by M. australis; (b) the distance to the nearest active burrow (determined by the presence of fresh prints andror scats) as a factor that can influence the probability of damage to plants; (c) the diameter of the trunk at ground level as another factor that can influence the probability of damage in plants; and (d) the condition of the plants at three different times. Condition was recorded based on the amount of new leaves and sprouts present, and the general condition of the plants. Plants were quantifiedrcategorized as: 1 s dead (no buds, no leaves, dead stem); 2 s bad (no buds, no leaves, alive stem); 3 s good (one to three buds, few leaves); 4 s very good (four or more buds, numerous leaves). Dead plants are easy to identify as their bark becomes brown rather than the characteristic green colour. The condition of individual plants was followed over three different periods: the austral winter of 1993; the austral summer of 1994; and the austral autumn of 1995. The area was visited periodically and no freshly gnawed areas were found during these periods. However, during July 1995 (dry season), a new clump of G. decorticans plants gnawed by M. australis was found, which was approximately 2.5 km from the closest colony in our study but still within the reserve. At that site, the diameter of the trunk, plant condition and proportion of the total trunk perimeter gnawed was measured for 206 plants. Because the data were not normally distributed, they were analysed using nonparametric statistics. Even though the data from the two colonies were independent, they were pooled in the analyses in order to increase the sample sizes in each category. Spearman rank correlations (Siegel & Castellan, 1988) were performed to test our first three predictions. Correlations were made between: (a) the proportion of the trunk perimeter gnawed vs. the diameter of the trunk; (b) the proportion of the trunk perimeter gnawed vs. the distance to the colony; (c) the condition of the plants vs. the diameter of the trunk; and (d) the condition of the plants vs. the distance to the colony. Correlations for the proportion of the trunk perimeter gnawed against diameter and condition, for the plants sampled in July 1995, were also included. Finally, in order to test our fourth prediction a survival analysis was performed for two groups. The effect of the factor proportion of the total trunk perimeter gnawed (categorized as: high s more than 50% of the trunk perimeter gnawed, and low s equal or less than 50% of the trunk perimeter gnawed) on the survival rate of the plants (Cox’s F test) was compared.

Results In the austral winter of 1993 (dry season) 76% of the plants sampled had been freshly chewed upon, but no dead plants were recorded. In 1994 and 1995 none of the plants had freshly gnawed areas. Therefore, the condition of the plants was the only datum

M. F. TOGNELLI ET AL.

82

Table 1. Number of plants in 1993, number of plants in 1995, mean diameter of the trunks (cm) mean distance to the colony (cm) and mean proportion of the total trunk perimeter gnawed, for each condition for the two studied colonies

Condition Bad Good

Dead

Very good

Colony 1 Number of plants in 1993 Number of plants in 1995 Mean trunk diameter (cm) Mean distance to colony (cm) Mean damage

0 7 0.76 320.14 0.82

22 28 1.69 513.11 0.39

24 12 2.35 551.58 0.31

9 8 5.80 253.90 0.03

Colony 2 Number of plants in 1993 Number of plants in 1995 Mean trunk diameter (cm) Mean distance to colony (cm) Mean damage

0 1 0.70 650.00 1.00

6 8 2.16 653.25 0.65

8 12 2.82 859.58 0.42

36 29 3.25 1040.00 0.20

measured in the second and third year. Furthermore, eight plants died by the final year of the study (Table 1). The best condition of plants was associated with a lower proportion of their total trunk perimeter being gnawed ( N s 105, rs s y0.554, p - 0.0001). A significant negative correlation between the total trunk perimeter that had been gnawed and the distance of the plants from the colonies ( N s 105, rs s y0.283, p - 0.003) was also found. The proportion of the total trunk perimeter gnawed was negatively associated with the diameter of the trunk ( N s 105, rs s y0.348, p - 0.0003). The correlations for the variable condition of the plants for only the last sampled period (1995) are given, although the previous periods were also significant. A similar pattern was found for the trees sampled in an extra colony in July 1995. In this colony, a significant negative correlation between the diameter and the proportion of the trunk gnawed ( rs s y0.51, p - 0.0001, N s 206), and between the proportion of the trunk gnawed and plant condition ( rs s y0.31, p - 0.0001, N s 206) was found. Finally, for the factor proportion of the trunk perimeter gnawed a significant difference between the rate of survival of plants with a high proportion of their trunks gnawed and the plants with a low proportion of their trunks gnawed (Cox’s F test; F s 8.97, p s 0.0001) was found.

Discussion Our data show that gnawing by M. australis has an impact on G. decorticans plants. More specifically, our results suggest that: (1) plant condition is worse in highly gnawed plants; (2) damage to G. decorticans plants by gnawing is inversely related to their distance from a M. australis colony; (3) damage is greater on plants with smaller diameter trunks than on plants with trunks of larger diameter; and (4) survival of heavily gnawed plants is lower than on less gnawed plants. Plants located closer to M. australis colonies and having a smaller diameter are highly gnawed upon, with a lower survival rate than others. Many studies support the

EFFECT OF GNAWING BY M. AUSTRALIS

83

finding that herbivore impact on vegetation is inversely related to the distance from burrows or cover (Longland, 1991; Swihart, 1991; Swihart & Picone, 1991; English & Bowers, 1994), creating local gradients and mosaic patches of vegetation (Huntly, 1991). In the shrublands of central Chile, Fuentes et al. (1983) found that degus ( Octodon degus [Molina]) and European hares ( Oryctolagus cuniculus L.) significantly impact the survival of tree and shrub seedlings. They also found that the effect of the native degus is limited to places close to plant cover or rock piles, where it is protected from predators, where the effect of the introduced rabbit is much wider, presumably because in the absence of native predators it is able to exploit open areas. Habitat utilization, diet and some behavioural patterns of M. australis appear very similar to the Chilean degus (Fulk, 1976; Meserve et al., 1984). The majority of M. australis activities are concentrated close to covered areas, feeding on plants it finds nearby (Rood, 1970). Also, in eastern Argentina, Cassini (1991) found that Cavia aperea Erxleben seldom forages far from covered areas and then only for very short periods of time, lessening the risk of predation. Therefore, predation risk may affect foraging behaviour, and this in turn may explain why G. decorticans plant closer to M. australis burrows are more heavily gnawed than those further away. In the Sonoran Desert, mammalian herbivores are an important factor limiting the establishment of Cercidium microphyllum (Torr.) Rose & I. M. Johnston in bajada habitats (McAuliffe, 1986). However, McAuliffe (1986) reported a different pattern in the distribution of seedlings affected by herbivores than the one we found in our study. He found that Cercidium seedlings growing beneath canopies of other plants suffered less herbivore-induced mortality than seedlings in open areas. He suggests that the association of Cercidium with canopies of other plants provide physical concealment from herbivores. Moreover, many of the plant species with which Cercidium is associated are highly unpalatable to mammalian herbivores, so this may further help in deterring herbivores from foraging in the immediate area (McAuliffe, 1986). We speculate that this difference in the pattern of distribution of plants affected by herbivory may be largely due to different habitat use by different herbivores. In the Sonoran Desert, the most important herbivores were cottontail rabbits ( Sylvilagus audobonii Baird) and jackrabbits ( Lepus alleni Mearns and L. californicus Gray). Nevertheless, Longland (1991) found that L. californicus prefer to forage near plant cover in order to reduce prediction risk. Although Longland’s study contradicts our speculation, he suggests that jackrabbit foraging behaviour is flexible, depending on predation pressure, so it might be that prediction pressure is not very high in the Sonoran Desert. We also found that plants with trunks of smaller diameter had greater probability of being gnawed. While further studies are needed to determine why this is so, it may be that thicker trunks could not be gnawed upon due to anatomical constraints of cuises such as maximum mouth opening size, or incisors not sufficiently proodonts. Alternatively, thicker trunks may have higher fibre content, or some kind of resin or secondary compound that repels rodents (Choo et al., 1981; Crawley, 1983). Another observation of the study is that gnawing by M. australis occurred only in the dry season of 1993 (in the two studied colonies), and also in the other colony found and measured in July 1995. We did not find freshly gnawed areas during wet seasons. Although studies on woodchucks have demonstrated that gnawing by this species is associated with scent marking (Ouellet & Ferron, 1988; Swihart, 1991), we believe that gnawing by M. australis is due to low availability of green vegetation during dry periods, as suggested for herbivory on Larrea spp. for M. australis in the Monte Desert (Borruel et al., 1998). Therefore, trees with trunks of small diameter that survive the season of attack (dry winter) may escape predation ( sensu Janzen, 1970) and grow, passing a threshold size whereby they may no longer be susceptible to gnawing by M. australis, i.e reaching size refugium ( sensu Myster & McCarthy, 1989). This study shows that under natural conditions M. australis in the Monte Desert of

84

M. F. TOGNELLI ET AL.

Argentina may influence the condition and survival of G. decorticans plants near their burrow systems. Such a localized and differential effect on the survival of the plants can cause long-term effects in the distribution and abundance of G. decorticans plants, and the distribution and structure of habitat heterogeneity. We are very grateful to D. Kelt, R. Garcia Gonzalez, D. Gomez Garcia, D. Van Vuren, L. Marone, S. Giannoni, J. Gonnet and J. Priotto for their comments on a draft of the manuscript, and J. Boshoven for helping us with the English version. R. Ojeda and an anonymous reviewer greatly improved the manuscript with their comments.

References Asquith, N.M., Wright, S.J. & Clauss, M.J. (1997). Does mammal community composition control recruitment in neotropical forests? Evidence from Panama. Ecology, 78: 941]946. Borruel, N., Campos, C., Giannoni, S. & Borghi, C. (1998). Effect of herbivorous rodents (cavies and tuco-tucos) on a shrub community in the Monte Desert, Argentina. Journal of Arid Environments, 39: 33]37. Brown, J.H. & Heske, E.J. (1990). Temporal changes in a Chihuahuan Desert rodent community. Oikos, 59: 290]302. Bucher, E.H. (1987). Herbivory in arid and semiarid regions of Argentina. Revista Chilena de Historia Natural, 60: 265]273. Cabrera, A. (1953). Los Roedores Argentinos de la Familia Caviidae. Facultad de Agronomıa ´ y Veterinaria, Universidad de Buenos Aires, Publicacion ´ 6: 1]93. Campos, C.M. (1997). Utilizacion medianos y ´ de los recursos alimentarios por mamıferos ´ pequenos ˜ del desierto del Monte. Doctoral dissertation, Universidad Nacional de Cordoba, ´ Argentina. 183 pp. Campos, C. & Ojeda, R.A. (1997). Dispersal and germination of Prosopis flexuosa (Fabaceae) seeds by desert mammals in Argentina. Journal of Arid Environments, 35: 717]714. Cassini, M.H. (1991). Foraging under predation risk in the wild guinea pig Cavia aperea. Oikos, 62: 20]24. Choo, G., Waterman, P., McKey, D. & Gartlan, S. (1981). A simple enzyme assay for dry matter digestibility and its value in studying food selection by generalist herbivores. Oecologia, 49: 170]178. Contreras, J.R. & Roig, V.G. (1978). Observaciones sobre la organizacion ´ social, la ecologıa ´ y la ˜ ˜ estructura de los habitaculos de Microcavia australis en Nacunan, Provincia de Mendoza. ´ Ecosur, 5: 191]199. Crawley, M.J. (1983). Herbivory. The dynamics of animal]plant interactions. London: Blackwell Scientific Publications. 437 pp. Danell, K., Bergstrom, ¨ R. & Edenius, L. (1994). Effects of large mammalian browsers on architecture, biomass, and nutrients of woody plants. Journal of Mammalogy, 75: 833]844. English, E.L. & Bowers, M.A. (1994). Vegetational gradients and proximity to woodchuck ( Marmota monax ) burrows in an old field. Journal of Mammalogy, 75: 775]780. Fuentes, E.R., Jaksic, F.M. & Simonetti, J. (1983). European rabbits versus native rodents in Central Chile: effects on shrub seedlings. Oecologia, 58: 411]414. Fulk, G.W. (1976). Notes on the activity, reproduction and social behaviour of Octodon degus. Journal of Mammalogy, 57: 495]505. Gill, D.S. & Marks, P.L. (1991). Tree and shrub colonization of old fields in central New York. Ecological Monographs, 61: 183]205. Huntly, N. (1991). Herbivores and the dynamics of communities and ecosystems. Annual Review of Ecology and Systematics, 22: 477]503. Huntly, N.J. & Inouye, R.S. (1988). Pocket gophers in ecosystems: pattern and mechanisms. Bioscience, 38: 786]793. Huntly, N.J. & Reichman, O.J. (1994). Effects of subterranean mammalian herbivores on vegetation. Journal of Mammalogy, 75: 852]859. Inouye, R.S., Huntly, N.J., Tilman, D. & Tester, J.R. (1987). Pocket gophers ( Geomys bursarius ), vegetation, and soil nitrogen along a successional sere in east central Minnesota. Oecologia, 72: 178]184.

EFFECT OF GNAWING BY M. AUSTRALIS

85

Janzen, D.H. (1970). Herbivores and the number of tree species in tropical forests. American Naturalist, 104: 501]528. Kufner, M.B. & Chambuleyron, M.B. (1993). Distribucion ´ de las colonias de Lagostomus maximus en relacion y su impacto en el medio desertico del monte, ´ a variables del habitat ´ ´ Argentina. Donana, Acta Vertebrata, 20: 81]87. ´ Leon, ´ R.J.C., Rusch, G.M. & Oesterheld, M. (1984). Pastizales pampeanos: impacto agropecuario. Phytocoenologia, 12: 201]218. Longland, W.S. (1991). Risk of predation and food consumption by black-tailed jackrabbits. Journal of Range Management, 44: 447]450. Mares, M.A., Enders, F.A., Kingsolver, J.M., Neff, J.M. & Simpson, B.B. (1977). Prosopis as a niche component. In: Simpson, B.B. (Ed.), Mesquite its Biology in Two Desert Ecosystems, pp. 123]149. Pennsylvania: Dowden, Hutchinson & Ross. 250 pp. Marzocca, A. (1957). Manual de Malezas. Buenos Aires, Argentina: Editorial Coni. 530 pp. McAuliffe, J.R. (1986). Herbivore-limited establishment of a Sonoran desert tree, Cericidium microphyllum. Ecology, 67: 276]280. Meserve, P.L., Martin, R.E. & Rodriguez, J. (1984). Comparative ecology of the caviomorph rodent Octodon degus in two Chilean mediterranean-type communities. Revista Chilena de Historia Natural, 57: 79]89. Molofsky, J. & Fisher, B.L. (1993). Habitat and predation effects on seedling survival and growth shade-tolerant tropical trees. Ecology, 74: 261]265. Myster, R.W. & McCarthy, B.C. (1989). Effects of herbivory and competition on survival of Carya tomentosa (Juglandaceae) seedlings. Oikos, 56: 145]148. Ostfeld, R.S. & Canham, C.D. (1993). Effects of meadow vole population density on tree seedling survival in old fields. Ecology, 74: 1792]1801. Ouellet, J.P. & Ferron, J. (1988). Scent-marking behavior by woodchucks ( Marmota monax ). Journal of Mammalogy, 69: 365]368. Platt, W.J. (1975). The colonization and formation of equilibrium plant species associations on badger disturbances in a tall-grass prairie. Ecological Monographs, 45: 285]305. ˜ ˜´ Deserta, 1: 25]232. Roig, F.A. (1971). Flora y vetetacion ´ de la Reserva Forestal de Nacunan. Rood, J.P. (1970). Ecology and social behavior of the desert cavy ( Microcavia australis ). American Midland Naturalist, 83: 415]454. Sala, O.E. (1988). The effect of herbivory on vegetation structure. In: Werger, M.J.A., van der Aart, P.J.M., During, H.J. & Verhoeven, J.T.A. (Eds), Plant Form and Vegetation Structure, pp. 317]330. The Netherlands: The Hague. 356 pp. Sala, O.E., Oesterheld, M., Leon, R.J.C. & Soriano, A. (1986). Grazing effects upon plant community structure in subhumid grasslands of Argentina. Vegetatio, 67: 27]32. Siegel, S. & Castellan, Jr., N.J. (1988). Nonparametric Statistics for the Behavioral Sciences (2nd Edn). New York: McGraw-Hill. 399 pp. Sork, V.L. (1987). Effects of predation and light seedling establishment in Gustavia superva. Ecology, 68: 1341]1350. Swihart, R.K. (1991). Influence of Marmota monax on vegetation fields. Journal of Mammalogy, 72: 791]795. Swihart, R.K. & Picone, P.M. (1991). Effects of woodchuck activity on woody plants near burrows. Journal of Mammalogy, 72: 607]611. Taiz, L. & Zeiger, E. (1991). Plant Physiology. California: The BenjaminrCummings Publishing Co. 559 pp. Tognelli, M.F., Campos, C.M., Ojeda, R.A. & Roig, V.G. (1995). Is Microcavia australis (Rodentia: Caviidae) associated with a particular plant structure in the Monte desert of Argentina? Mammalia, 3: 327]333.