This article was published in an Elsevier journal. The attached copy is furnished to the author for non-commercial research and education use, including for instruction at the author’s institution, sharing with colleagues and providing to institution administration. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

ARTICLE IN PRESS Journal of Arid Environments Journal of Arid Environments 71 (2007) 327–332 www.elsevier.com/locate/jaridenv

Short Communication

Distribution of geckos in northern Peru: Long-term effect of strong ENSO events? A. Catenazzi, M.A. Donnelly Department of Biological Sciences, OE167, Florida International University, University Park, Miami, FL 33199, USA Received 14 February 2007; received in revised form 14 April 2007; accepted 4 May 2007 Available online 27 June 2007

Abstract Arid ecosystems of western South America that are affected by ENSO events experience short-term pulses in precipitation that trigger changes in the recruitment and growth of vegetation. We hypothesized that strong ENSO events, such as those of 1982/83 and 1997/98, could influence animal communities over several decades. We surveyed populations of three sympatric species of geckos (Phyllodactylus spp.) at the Illescas Peninsula, northern Peru, in 2004, and compared our results with those collected by Huey [1979. Parapatry and niche complementarity of Peruvian desert geckos (Phyllodactylus): the ambiguous role of competition. Oecologia 38, 249–259] in 1968. We found that the number of mesquite trees was much higher in 2004 than it was in 1968. The frequency of the most arboreal species of gecko, Phyllodactylus reissi, was higher in 2004 than it was in 1968. P. reissi was also more likely to be associated with mesquite trees in 2004 than in 1968. Carbon and nitrogen stable isotopes could not be used to identify the most important source of primary productivity for the diet of geckos, because the values of mesquite trees and Capparis shrubs were similar. However, nitrogen isotopic values suggested that one species of gecko was incorporating marine-derived nutrients in its diet. r 2007 Elsevier Ltd. All rights reserved. Keywords: Arid ecosystems; Community ecology; Lizards; Prosopis; Sechura desert; Stable isotopes

1. Introduction Environmental fluctuations, such as El Nin˜o Southern Oscillation (ENSO) events, can have strong impacts on plant and animal terrestrial communities (Holmgren et al., 2001, 2006b). Many of these strong impacts are caused by short-term or catastrophic events, such as abundant rainfall, fire or flooding. In ecosystems that have been continuously influenced by ENSO events, the frequency of this phenomenon over several decades may also trigger long-term changes in plant cover, plant–animal interactions, and species diversity. In the Sechura desert in northern Peru, the two very strong ENSO events of 1982/83 and 1997/98 were accompanied by unusually high precipitation of up to 400 cm (Huama´n Solis and Garcia Pen˜a, 1985) and a spectacular regeneration of trees (Block and Richter, 2000; Holmgren et al., 2001). Abundant rainfall during 1997/98 even Corresponding author. Present address: Department of Biology, University of South Florida, 4202 E. Fowler Avenue, Tampa,

FL 33620, USA. Tel.: +1 305 396 2626; fax: +1 817 582 9551. E-mail address: [email protected] (A. Catenazzi). 0140-1963/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jaridenv.2007.05.003

Author's personal copy

ARTICLE IN PRESS 328

A. Catenazzi, M.A. Donnelly / Journal of Arid Environments 71 (2007) 327–332

triggered the formation of a temporary lake that occupied up to 2326 km2 of previously barren desert in March 1998 (http://www.imarpe.gob.pe/argen/nina/premota/premota.html). Over the last 30 years, recruitment and the mean annual growth rate of Prosopis pallida in northern Peru has peaked during the ENSO events of 1972/73, 1982/83 and 1997/98 (Holmgren et al., 2006a, b; Lo´pez et al., 2005; Vilela et al., 2003). ENSO events of magnitude similar to those of 1982/83 and 1997/98 did not occur in previous decades throughout the XX century in northern Peru (Hocquenghem, 1998; Woodman, 1985), and the closest very strong ENSO event has been traced back to 1891 (Hocquenghem and Ortlieb, 1992; Ortlieb, 2000). To investigate whether these ENSO-triggered changes were influencing animal communities in the area, we sampled gecko populations near the Illescas Peninsula (51500 S, 811020 W), and compared our results to data gathered by Huey (1979) in 1967 and 1968. We hypothesized that long-term effects of ENSO would include an increase in the relative frequency of arboreal species. 2. Methods Leaf-toed geckos of the genus Phyllodactylus are widely distributed in western Peru, where they occupy a range of habitats from coastal hyper-arid deserts in the south to dry equatorial forests in the north. The highest species richness for this genus is found at the study site, with four sympatric species living in the Illescas Peninsula (Dixon and Huey, 1970): Phyllodactylus clinatus, Phyllodactylus kofordi, Phyllodactylus microphyllus, and Phyllodactylus reissi. P. clinatus was rare at this location (Huey, 1969) and was not found during our work in Illescas. Cerro Illescas (51550 S, 811030 W) is a small mountain range reaching maximum elevations of approximately 500 m. The Cerro lies at the western border of the Sechura desert, a generally flat and sandy desert. According to data from Huey (1979), P. reissi was the most scansorial species (82.7% of all captured individuals found on rocks or trees), followed by P. microphyllus (56.7%) and P. kofordi (36.5%). We captured geckos using headlamps and actively searching on the ground, shrubs and trees during December 2003 and June 2004, for a total of 10 sampling nights. We sampled the same areas surveyed by Huey in 1968: the slopes and washes of the Cerro Illescas, the flatland desert within a few kilometers from the Cerro, and the gentle slope of the Cerro near the tip of the Illescas Peninsula (Aguja Point). We used the same habitat categories defined by Huey (1979): habitat I included the flatland desert occupying most of the Sechura desert (but covering a small area in Illescas, and not considered here for comparisons), habitat II the flatland desert 2–3 km around the Cerro (from the Cerro to the ocean), habitat III the base and washes of Cerro Illescas and habitat IV the Cerro itself. Each habitat was surveyed along three linear transects separated from each other by more than 2 km. We surveyed one transect per night, by searching for geckos on the ground, under logs and on trees during a 4-h period. We sampled vegetation cover by counting the number of shrubs and trees in 20 randomly placed quadrats (20  20 m) within each one of three habitat types (II–IV). Huey (1979) provided similar data for habitat II, number of plants along a linear transect in habitat III, and a qualitative description for habitat IV. The main shrub and tree species in the area are the mesquite or ‘‘Algarrobo’’ (P. pallida), ‘‘Zapote’’ (Capparis scabrida), ‘‘Bichayo’’ (Capparis avicennifolia) and ‘‘Palo verde’’ (Parkinsonia aculeata). In order to establish a link between geckos (which feed on arthropods) and their primary source of energy and nutrients (plants that support their prey), we sampled two widespread primary producers (P. pallida and C. scabrida), the three gecko species and an intertidal lizard (Microlophus peruvianus) for carbon (d13C) and nitrogen (d15N) stable isotope ratios. Stable isotope ratios are often used to track the movement of carbon and nitrogen through the food web (Fry, 2007). We assumed that there could be a difference in the isotopic signature between Prosopis and Capparis, and that this difference could be used to establish a link between the plant and the diet of the geckos. M. peruvianus was included in this analysis because this species lives and forages in the intertidal (Catenazzi et al., 2005); thus its diet includes foods that are derived from the ocean, rather than from terrestrial plants. Values of d13C and d15N are typically higher for marine organisms and/or for terrestrial plants and consumers of localities with high inputs of marine-derived energy and nutrients than for terrestrial producers and consumers without marine input (Catenazzi and Donnelly, 2007; Michener and Schell, 1994). Samples consisted of five replicates of a collection of 10–15 leaves from 3–5 plants, and tail sections from five different individuals for each species of lizards. All dried samples were ground to a powder and defatted prior to isotope analysis by using a solution of dichloromethane: methanol (9:1) and sonication.

Author's personal copy

ARTICLE IN PRESS A. Catenazzi, M.A. Donnelly / Journal of Arid Environments 71 (2007) 327–332

329

Stable C and N isotope analysis were conducted on a Finnigan MAT Delta Plus continuous flow isotope ratio mass spectrometer. We analyzed 3–4 mg of plant samples, and 2 mg of animal samples. Values of d13C and d15N are expressed relative to their respective international standards, Pee Dee belemnite and atmospheric air. We analysed 18% of our samples in duplicate. The average standard deviation for replicate samples was 0.02% for d13C and 0.12% for d15N. 3. Results and discussion We captured 64 geckos in habitats II, III and IV and two geckos during a rapid survey in habitat I. In comparison with data from 1968, P. reissi increased its frequency relative to other species for habitats II–IV (Fig. 1) supporting our hypothesis regarding a change in arboreal species over the time interval. P. reissi was also more likely to be associated with mesquite trees in 2004 than in 1968: 59% were found on mesquite trees and 27% close to mesquite trees in 2004, whereas Huey (1979) only found about 22% of the individuals on Prosopis. We did not observe this species in rocky outcrops in 2004, in contrast to Huey (1979) finding 24% of the individuals in this microhabitat. In all three habitats, mesquite trees were the dominant species in 2004 (Table 1). The vast increase in the number of mesquite trees accounted for an overall increase in the number of shrubs and trees between 1968 and 2004 (from 182 to 695 plants/ha in habitat II). In habitat II, plant composition differed between 1968 and 2004 (w2 ¼ 337.88, po0.01) because mesquite trees were more than four times more abundant in 2004 than they were in 1968, counting mesquite trees taller than 2 m. Most mesquite trees (414 of 479 in habitat II) were saplings that probably benefited from the heavy rains of 1997/98. By pooling all mesquite trees and saplings, this species was 36 times more abundant in 2004 than it was in 1968. The sharp increase in the number of mesquite trees could account for the increase in the relative abundance of P. reissi, because Dixon and Huey (1970) reported that this species was more likely to be associated with mesquite trees than P. microphyllus and P. kofordi. We counted several trees and saplings of P. aculeata in washes at the foot of the Cerro, a species that was not found by Huey (1979) in 1968. Both species of Capparis were relatively less abundant in habitat III in 2004 than they were in 1968, and the number of plants decreased from 144 to 93 plants/ha for C. avicennifolia and from 22 to 3 plants/ha for C. scabrida. The success in P. pallida tree establishment following strong ENSO events contrasts with the decrease in the relative abundance of Capparis species. Holmgren et al. (2006a, b) suggested that the environmental conditions during ENSO events in northern Peru provided good opportunities for tree establishment and growth. Soil texture (loose sandy soils) with low resistance for root growth may favor mesquite trees that develop deep

microphyllus

kofordi

reissi

Relative frequency (%)

100% 80% 60% 40% 20%

4) (2

IV

96 (1 IV

00

8)

) 00 4

) (2 III

III

(1

96 8

) 00 4 (2

II

II

(1

96 8

)

0%

Habitat and year Fig. 1. Frequency (% total) of species of Phyllodactylus in three habitats in Illescas in 1968 and 2004. Habitats: II (flatland adjacent to Cerro Illescas), III (base and washes of Cerro Illescas), IV (Cerro Illescas). Data for 1968 are from Huey (1979).

Author's personal copy

ARTICLE IN PRESS A. Catenazzi, M.A. Donnelly / Journal of Arid Environments 71 (2007) 327–332

330

Table 1 Relative abundance (% total) of shrub and tree species in three habitats in Illescas in 1968 and 2004 Habitat

Area (m2)

no. of plants

C. avicennifolia

C. scabrida

P. pallida

1968 II III

7360 NA

134 131

79.1 34.3

11.9 48.9

9.0 16.8

0.0 0.0

2004 II III IV

8000 8000 8000

557 139 37

13.2 14.4 2.7

0.4 1.4 24.3

86.0 43.2 62.2

0.4 41.0 10.8

P. aculeata

See text for description of habitats. Full names of plant species, from left to right: Capparis avicennifolia, Capparis scabrida, Prosopis pallida, Parkinsonia aculeata. Data for 1968 are from Huey (1979).

15 Microlophus peruvianus

δ15N(‰)

P. microphyllus

10

P. kofordi

P. reissi

5 Capparis

0 -26

Prosopis

-24

-22

-20

-18

-16

-14

δ C (‰) 13

Fig. 2. Carbon (d13C) and nitrogen (d15N) stable isotope ratios for two widespread primary producers (C. scabrida and P. pallida), the geckos Phyllodactylus kofordi, P. microphyllus and P. reissi from habitats II and III, and the intertidal lizard M. peruvianus. Samples are five replicates of 10–15 leaves from three to five plants for Capparis and Prosopis, and tail sections from five different individuals for each species of lizard.

roots and take advantage of deep phreatic layers over Capparis shrubs with less developed roots. Mesquite is readily dispersed and scarified by goats in northern Peru (Holmgren et al., 2006a, b), and scarification has been shown to promote germination in several species of Prosopis (Vilela and Ravetta, 2001). P. pallida could also respond to increased water availability by allocating resources mainly to growth, as shown by Vilela et al. (2003) for another tree-type mesquite species (P. alba). Anthropogenic disturbance is unlikely to have affected mesquite growth in the region since 1968, although goats could have enhanced seed dispersal and germination. Illescas was (in 1968) and still is a relatively remote area with a small human population mainly dedicated to fishing and seafood harvesting that most likely has no significant effect on terrestrial vegetation. The largest human settlement in the region, an oil pipeline terminal near the tip of the peninsula that was not present in 1968, maintained much of the study area outside of human influence other than that directly involved with the pipeline facilities (access to this area is restricted). Stable isotope ratios of carbon and nitrogen were similar for Capparis and Prosopis (Fig. 2); therefore, these analyses could not differentiate between these two plants as primary sources of productivity for geckos. P. reissi and P. kofordi occupied similar trophic niches according to isotopic ratios, whereas P. microphyllus had more enriched d13C and d15N values that were closer to the values of M. peruvianus than they were to

Author's personal copy

ARTICLE IN PRESS A. Catenazzi, M.A. Donnelly / Journal of Arid Environments 71 (2007) 327–332

331

congeneric geckos. These enriched values suggest that P. microphyllus was incorporating marine-derived sources of energy and nutrients in its diet, similarly to what has been observed in P. angustidigitus from southern Peru (Catenazzi and Donnelly, 2007). This hypothesis is consistent with the observed distribution of this gecko, because P. microphyllus was most frequent in habitat II (flatland coastal desert around Cerro Illescas). Our results suggest that P. reissi has benefited from the increase in plant cover in Illescas. This species is the largest gecko in the area (up to 75 mm in snout–vent length vs. 58 mm for P. microphyllus), and is also the most arboreal (Dixon and Huey, 1970; Huey, 1979). P. reissi was especially abundant on mesquite trees growing near shore, a few kilometers away from the Cerro. The relative abundance of this gecko in a habitat that was not occupied by trees in 1968 may be the result of the input of energy and nutrients from the sea. Observations from elsewhere in northern Peru also indicate that P. reissi is common wherever mesquite trees grow close to the ocean (P. Venegas, personal communication). In habitat II, P. microphyllus was exclusively terrestrial in 2004. Competition or intraguild predation may account for a shift from arboreal to terrestrial microhabitats for P. microphyllus in 2004. Intraguild predation is supported by observations of saurophagy among Phyllodactylus species (Huey, 1969); during fieldwork in June 2004 one of us obtained an individual of P. microphyllus from the stomach of a female of P. reissi (Catenazzi, pers. obs.). Sampling biases for plants and geckos could account for some differences between 1968 and 2004. These include differences in sampling individual trees and shrubs, location of transects and identification of the main habitat types. However, given the open structure of this arid landscape, sampling bias could not explain the sharp increase in the number of plants, and the dominance of mesquite trees in 2004 as compared to 1968. Concerning geckos, the use of kerosene lantern in 1968 (R. Huey, pers. com.) instead of headlamps in 2004 could have introduced a bias in detecting different species of geckos, especially concerning the position of geckos on ground or on trees. Finally, we acknowledge that there are several limitations in drawing conclusions based on two samples in time, but we argue that, given the simplicity of the system and the strength of ENSO events in the region, our data indicate how different gecko species respond to changes in vegetation cover following strong ENSO events. In summary, we present data suggesting that bottom–up productivity, triggered by ENSO events, may play an important role in structuring geckos’ populations in northern Peru. This finding complements the discussion by Huey (1979) on the limited role of competition as a causal force for parapatry and niche complementarity. Changes in vegetation and the physical environment seem to have favored one species of the assemblage, although one could not rule out the effects of competition and changes in habitat availability. Because of its location on the Peruvian coast, the Sechura desert has a long history of recurrent ENSO events, and of sudden changes in vegetation cover associated with extreme rainfall events (Hocquenghem, 1998). These changes are likely to influence the structure of animal assemblages throughout the desert. Acknowledgments We are grateful to Compan˜ia Min˜era Regional Grau Bayovar for logistic support in Bayovar and to Petroperu´ for permit to visit the Illescas Peninsula. We thank J. Carrillo for field support, and M. Holmgren, R. Huey, J. Watling and two anonymous reviewers for comments on the manuscript. This is contribution number 124 to the program in Tropical Biology at Florida International University.

References Block, M., Richter, M., 2000. Impacts of heavy rainfalls in El Nin˜o 1997/98 on the vegetation of Sechura Desert in Northern Peru: a preliminary report. Phytocoenologia 30, 491–517. Catenazzi, A., Donnelly, M.A., 2007. The Ulva connection: marine green algae subsidize terrestrial predators in coastal Peru. Oikos 116, 75–86. Catenazzi, A., Carrillo, J., Donnelly, M.A., 2005. Seasonal and geographic eurythermy in a coastal Peruvian lizard. Copeia 2005, 713–723. Dixon, J.R., Huey, R.B., 1970. Systematics of the lizards of the gekkonid genus Phyllodactylus of mainland South America. Los Angeles County Museum Contributions in Science 192, 1–78. Fry, B., 2007. Stable Isotope Ecology. Springer, New York.

Author's personal copy

ARTICLE IN PRESS 332

A. Catenazzi, M.A. Donnelly / Journal of Arid Environments 71 (2007) 327–332

Hocquenghem, A.-M., 1998. Para vencer la muerte: Piura y Tumbes: raı´ ces en el bosque seco y en la selva alta, horizontes en el Pacı´ fico y en la Amazonia. Travaux de l’Institut Franc- ais d’E´tudes Andines, 109. CNRS-PICS (Paris, France) and IFEA (Lima, Peru´). Hocquenghem, A.M., Ortlieb, L., 1992. Eventos El Nin˜o y lluvias anormales en la cosa del Peru´. Siglos XVI–XIX. Bulletin de l’Institut Franc- ais d’E´tudes Andines 21, 197–278. Holmgren, M., Scheffer, M., Ezcurra, E., Gutierrez, J.R., Mohren, G.M.J., 2001. El Nin˜o effects on the dynamics of terrestrial ecosystems. Trends in Ecology and Evolution 16, 89–94. Holmgren, M., Lo´pez, B.C., Gutie´rrez, J.R., Squeo, F.A., 2006a. Herbivory and plant growth rate determine the success of El Nin˜o southern oscillation-driven tree establishment in semiarid South America. Global Change Biology 12, 1–9. Holmgren, M., Stapp, P., Dickman, C.R., Gracia, C., Graham, S., Gutie´rrez, J.R., Hice, C., Jaksic, F., Kelt, D.A., Letnic, M., Lima, M., Lo´pez, B.C., Meserve, P.L., Milstead, W.B., Polis, G.A., Previtali, M.A., Richter, M., Sabate´, S., Squeo, F.A., 2006b. Extreme climatic events shape arid and semiarid ecosystems. Frontiers in Ecology and the Environment 4, 87–95. Huama´n Solis, F., Garcia Pen˜a, A., 1985. Condiciones meteorolo´gicas en el Peru´ durante el feno´meno ‘‘El Nin˜o’’ 1982–1983. In: Proceedings of the Seminario Regional Ciencia, Tecnologı´ a y Agresio´n Ambiental, El Feno´meno El Nin˜o. Concytec, Lima, pp. 333–353. Huey, R.B., 1969. Ecological relations of sympatric Phyllodactylus in the Sechura Desert of Peru. Master Thesis, the University of Texas at Austin, 74pp. Huey, R.B., 1979. Parapatry and niche complementarity of Peruvian desert geckos (Phyllodactylus): the ambiguous role of competition. Oecologia 38, 249–259. Lo´pez, B.C., Sabate´, S., Gracia, C.A., Rodriguez, R., 2005. Wood anatomy, description of annual rings, and responses to ENSO events of Prosopis pallida H.B.K., a wide-spread woody plant of arid and semi-arid lands of Latin America. Journal of Arid Environments 61, 541–554. Michener, R.H., Schell, D.M., 1994. Stable isotope ratios as tracers in marine aquatic food webs. In: Lajtha, K., Michener, R.H. (Eds.), Stable Isotopes in Ecology and Environmental Science. Blackwell, pp. 138–157. Ortlieb, L., 2000. The documented historical record of El Nin˜o events in Peru: an update of the Quinn record (sixteenth through nineteenth centuries). In: Dı´ az, H.F., Markgraf, V. (Eds.), El Nin˜o and the Southern Oscillation: Multiscale Variability and Global and Regional Impacts. Cambridge University Press, Cambridge, pp. 207–295. Vilela, A.E., Ravetta, D.A., 2001. The effect of seed scarification and soil-media on germination, growth, storage, and survival of seedlings of five species of Prosopis L. (Mimosaceae). Journal of Arid Environments 48, 171–184. Vilela, A.E., Rennella, M.J., Ravetta, D.A., 2003. Responses of tree-type and shrub-type Prosopis (Mimosaceae) taxa to water and nitrogen availabilities. Forest Ecology and Management 186, 327–337. Woodman, R.F., 1985. Recurrencia del feno´meno El Nin˜o con intensidad comparable a la del an˜o 1982-1983. In: Proceedings of the Seminario Regional Ciencia, Tecnologı´ a y Agresio´n Ambiental, El Feno´meno El Nin˜o. Concytec, Lima, pp. 301–332.