Marine-derived nutrients can increase primary productivity and change species composition of terrestrial plant communities in coastal and riverine ecosystems. We hypothesized that sea nutrient subsidies have a positive effect on nitrogen assimilation and seedling survival of non-nitrogen fixing species, increasing the relative abundance of non-nitrogen fixing species close to seashore. Moreover, we proposed that herbivores can alter the effects of nutrient supplementation by preferentially feeding on high nutrient plants. We studied the effects of nutrient fertilization by seabird guano on tree recruitment and how these effects can be modulated by herbivorous lizards in the coastal dry forests of northwestern Peru. We combined field studies, experiments and stable isotope analysis to study the response of the two most common tree species in these forests, the nitrogen-fixing Prosopis pallida and the non-nitrogen-fixing Capparis scabrida. We did not find differences in herbivore pressure along the sea-inland gradient. We found that the non-nitrogen fixing C. scabrida assimilates marine-derived nitrogen and is more abundant than P. pallida closer to guano-rich soil. We conclude that the input of marine-derived nitrogen through guano deposited by seabirds feeding in the Pacific Ocean affects the two dominant tree species of the coastal dry forests of northern Peru in contrasting ways. The non-nitrogen fixing species, C. scabrida may benefit from sea nutrient subsidies by incorporating guano-derived nitrogen into its foliar tissues, whereas P. pallida, capable of atmospheric fixation, does not.
Citation: Havik G, Catenazzi A, Holmgren M (2014) Seabird Nutrient Subsidies Benefit Non-Nitrogen Fixing Trees and Alter Species Composition in South American Coastal Dry Forests. PLoS ONE 9(1): e86381. doi:10.1371/journal.pone.0086381
Editor: Paul V. A. Fine, University of California, Berkeley, United States of America
Received: October 1, 2013; Accepted: December 9, 2013; Published: January 22, 2014
Copyright: © 2014 Havik et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was partially funded by an Alberta Mennega Stichting grant to GH, a NWO Meervoud grant to MH (836.05.021) and a Swiss National Science Foundation Postdoctoral fellowship to AC (116305). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The composition and dynamics of ecological communities in coastal marine and terrestrial ecosystems are partially driven by nutrient fluxes that link sea and adjacent land –. Some of the most striking examples of this interaction are found in the unproductive arid west coasts of North and South America where sea nutrient subsidies from the productive Pacific Ocean permeate through whole food webs affecting terrestrial plants , – and consumers , , –, in various ways –. Marine nutrients reach land primarily through colonies of seabirds and marine mammals, or through organic debris, carcasses and algae deposited by tidal action , , and are further dispersed through trophic interactions , , ,  and wind .
Seabird guano is a major nutrient path from the surrounding marine environments into terrestrial roosting and nesting sites worldwide . It is well established that seabird guano increases soil nitrogen and phosphorus, although it can also affect soil pH, salinity and moisture . These changes in soil characteristics generally translate into enhanced foliar nutrient levels and plant growth , especially when sufficient rainfall is available , , , . However, at very high guano concentrations, plant cover can be reduced  or completely depleted , which has been attributed to ammonium toxicity  or acidic conditions . Plant communities in seabird colonies are often dominated by cosmopolitan and ruderal species ,  because the combined effects of seabirds include nutrient fertilization, physical disturbance and seed dispersal , . Experimental manipulations of seabird guano have found contrasting responses across tree species, suggesting that differences in plant functional traits could help explain changes in plant composition in seabird-associated areas .
The overall effects of marine subsidies on arid coastal woody plant communities likely depend on how herbivores respond to plant nutrient enrichment and on how they are distributed along the sea-land gradient of nutrient fertilization. In pulsed systems, herbivores can limit plant growth , , but their effect on tree recruitment in arid coastal systems has not been directly assessed. Since herbivores usually respond positively to increasing plant nutrient content , we hypothesize that they can potentially offset the positive effects of sea nutrient subsides on tree seedling establishment.
In this paper, we assess the effects of seabird nutrient input and herbivory on tree seedling establishment in an arid coastal system. We hypothesised that non nitrogen-fixing tree species benefit more strongly from marine-derived nitrogen than nitrogen-fixing species since non-fixers depend on soil nitrogen sources and are not able to uptake atmospheric nitrogen through associations with rhizobial bacteria. We combined stable isotope analysis, field observations and experiments to test the following predictions: (1) marine nutrient fertilization has a positive effect on nitrogen assimilation of non-nitrogen fixing trees; (2) abundance of non-nitrogen fixing trees increases near the sea shore; (3) herbivory by lizards can offset the positive effects of nitrogen fertilization on tree seedling survival.
Materials and Methods
We conducted our work in the coastal dry forests of north-western Peru, a region where the rich marine waters of the Peru-Chile current face the arid Sechura desert , . Dry forests with an estimated surface of over 3 million ha expand through northern Peru and southern Ecuador . These dry forests are dominated by Prosopis pallida, a nitrogen-fixing leguminous tree species, associated with other tree and shrub species such as the non-nitrogen fixing Capparis scabrida . Vegetation becomes sparser closer to the coast (Fig. 1). We worked along the coast of Bayovar (5° 47′ S, 81° 04′ W; 8 m. a. s. l.; Fig. 1), near the oil pipeline terminal of PetroPerú and protected from human disturbance. This area is controlled by PetroPerú, and PetroPerú authorized our work in the area. We worked along the rocky seashore, rising about 7 m above the water and separated from the beach by cliffs. Large colonies of blue-footed boobies (Sula nebouxii) create abundant deposits of guano up to 50 cm thick (Fig. 1).
a) Guano accumulates in places where Sula nebouxii roost along shore (photograph G. Havik). b) Capparis scabrida growing in guano-rich soil (photograph A. Catenazzi).
Rainfall is concentrated in the summer months (December-May) and is strongly influenced by El Niño Southern Oscillation (ENSO) events. Rainy ENSO events play a critical role in the regeneration of these forests, facilitating tree seedling establishment , ,  and fast growth  that contributes to plants reaching a critical size at which they are less sensitive to herbivore damage . Mean annual precipitation is ca. 50 mm (1961–1983; ), with high interannual variation (e.g. ranging from 2.8 mm in 1996 to 1639 mm in 1998). Mean annual temperature is 24.0°C, ranging between 21.1°C (August) and 27.9°C (February; ).
The two tree species are Prosopis pallida (Fabaceae) and Capparis scabrida (Capparaceae). Prosopis pallida is able to grow under arid conditions with at least 50 to 250 mm rainfall annually . It can grow in nutrient-poor soils where its roots fix nitrogen in association with rhizobial bacteria , . It is a fast growing tree with roots able to reach deep water sources up to 25 m below the surface . Seed dispersal and germination are facilitated by a variety of animals . Nevertheless, a high percentage of seeds remain dormant soon after falling from the tree . Seedling establishment is greatly enhanced during rainy ENSO events , . Capparis scabrida grows under a wide range of conditions, often in association with P. pallida . Seeds are dispersed by foxes and birds and have high germination rates .
The most important herbivore in this region is Dicrodon guttulatum (Teiidae), a lizard endemic to north-western Peru that feeds preferentially on P. pallida  and strongly limits tree seedling establishment . This is the most abundant herbivore at our site because livestock are not allowed within the PetroPerú concession.
We assessed soil nutrient concentrations along five transects perpendicular to the sea at increasing distances (10, 50, 100, 200 m) from the seashore. At each sampling location, we collected approximately 1 kg of soil from the top 30 cm of soil, and removed the rocks. Concentrations of NO3, NH4, PO4 and K were measured at the Resource Ecology Laboratory, University of Wageningen. Dried samples were extracted using a 0.01M CaCl2 solution previous to spectrophotometric determination of NO3 and NH4 by a Segmented Flow Analyzer (Skalar, Breda, The Netherlands). PO4 concentration was measured spectrophotometrically after digestion with H2O2. Concentrations of total nitrogen (N) and carbon (C) were measured in an Elemental Analyzer (EA 1108 CHN-O, Fisons Instruments) based on the dynamic combustion method. Potassium concentration was assessed using the ammonium acetate pH 7.0 protocol.
Marine-derived Nitrogen in Plants and Soils
We measured nitrogen stable isotope ratios (δ15N) in soils and leaves of P. pallida and C. scabrida to assess the contribution of seabird-derived nitrogen in soil and foliar tissues. Values of δ15N have been used successfully to assess the contribution of seabird-derived nitrogen to coastal and insular food webs , ,  because seabird guano typically has high δ15N values that can be over 30‰ , . When terrestrial plants assimilate nitrogen from guano, their δ15N increases and approaches the δ15N of guano: for example in the Gulf of California, δ15N of the saltbush Atriplex barclayana was 5.26±0.85‰ on an island without seabirds, but 36.36±2.72‰ on a seabird island . In the Sechura desert, δ15N values were −0.50±0.20‰ for P. pallida and 2.82±0.98‰ for C. scabrida in places that lack guano . Thus, we expected foliar tissues of P. pallida and C. scabrida near seabird roosting sites to have δ15N values much higher than those measured in places lacking guano. For both tree species, we collected pooled samples of 5 leaves from 3 trees (at least 1.5 m tall, n = 15 leaves) growing at the same 4 distances from shore where other measurements were taken (10, 50, 100, 200 m) along six sea-inland transects (including the same transects used for sampling soil). We measured soil δ15N in the samples used for measurements of nutrient concentrations. All samples were oven dried at 55°C for 72 hours and pulverized in a sterilized mill prior to isotope analysis. Analysis of N concentration and δ15N was conducted on a Finnigan MAT Delta Plus continuous flow isotope ratio mass spectrometer at the Center for Stable Isotope Biogeochemistry, University of California, Berkeley. We analyzed 3–4 mg of leaves, and 10–75 mg of soil. Values of δ15N are expressed relative to its international standard, atmospheric air. We analyzed 18% of our samples in duplicate. The average standard deviation for replicate samples was 0.12‰.
Spatial Tree Distribution
We evaluated the density of seedlings and trees of P. pallida and C. scabrida along the first 200 m from sea inland. We counted all seedlings and trees found in eight (300 m long × 60 m wide) transects perpendicular to shore. We divided each transect in six (50 m long × 60 m wide) plots.
Seedling Establishment along a Natural Sea-inland Gradient of Soil Nitrogen Concentration
We designed a field experiment to assess the effects of distance from shore and herbivory pressure on early seedling survival of P. pallida and C. scabrida. At four distances from shore along a natural sea-inland gradient (10, 50, 100, 200 m), we planted 20 seedlings (10 of each tree species). Half of the seedlings (five of each species) were protected against herbivores. At each distance location, seedlings were planted at 25 cm distance from each other . These treatments were replicated five times, along the same sea-inland transects where we measured soil nutrients and δ15N as described earlier. In summary, we planted 200 seedlings of each species for a total of 400 seedlings. Seedlings were protected against lizard browsing using plastic bottles (Ø: 10 cm, bottle top open). This method is commonly used in restoration projects in Peru because it is effective against herbivores  and experimental trials indicate that soil moisture content remains unaffected (File S1). Seedlings germinated from seeds collected locally and were transplanted when they were three weeks old in February 2009. We watered seedlings twice a week for the first seven weeks for a total of 270 mm, roughly equivalent to the rainfall during moderate ENSO events such as those in 1992 (235 mm, ) and 2002 (268 mm; ). No water was added during weeks with rainfall events of at least 5 mm. Total rainfall during the experimental time was 15 mm. Seedling survival was monitored regularly during three months.
We compared the soil nutrient concentrations (log-transformed) and the nitrogen stable isotope ratios (δ15N) in soils along the sea-inland nutrient gradient using general linear models with distance as a fixed factor and transect as a random factor. We tested for differences in leaf N concentrations and leaf δ15N for each tree species separately using general linear models including distance from shore as a fixed factor and transect as a random factor. We used generalized linear models to assess the effect of distance from shore on the number of naturally established plants for each species separately defining a Poisson distribution and using the factor distance from shore nested within transects.
To analyse the survival patterns of experimental seedlings, we used Cox proportional-hazards regression models, using the R-library survival . We tested for differences in seedling survival for each tree species separately including distance from shore (log-transformed) and herbivory (protected or not) as covariables and cluster as a random factor to account for the grouping of 20 seedlings per location.
Soil Nutrient Concentration along the Sea-inland Gradient
Soil nutrient concentrations were higher closer to shore (Table 1). Statistical tests indicate a significant effect of distance on [NO3] (F3,12 = 39.24; p<0.001), [NH4] (F3,12 = 6.60; p = 0.007), [PO4] (F3,12 = 7.84; p = 0.004), and on [K] (F3,12 = 18.23; p<0.001; see File S2 for details). Transects had no significant effects in any of the models.
Nutrient Concentrations and δ15N of Soil and Tree Leaves
Leaves of P. pallida had higher N concentrations (3.41% ±0.31) than leaves of C. scabrida (1.50% ±0.30; Table 1). The two species differed in their leaf N concentration pattern along the sea shore-inland gradient. Whereas N concentrations of P. pallida did not change among sea-distance classes (F3,11 = 2.94; p = 0.08), leaves of C. scabrida contained more N in plants growing near shore (F3,12 = 15.75; p<0.001).
The δ15N of soil decreased with distance from shore (Fig. 2; F3,10 = 8.35; p = 0.004). Values of leaf δ15N for C. scabrida were more enriched than δ15N values for P. pallida (T- test = −14.71; p<0.001). Also the sea shore-inland pattern for leaf δ15N differed between the two species (Fig. 2). Whereas the δ15N for P. pallida did not vary among distance classes (Fig. 2; F3,10 = 0.67; p = 0.59), the δ15N for C. scabrida decreased with distance from shore (F3,12 = 3.54; p = 0.048). At a distance of 10 m from shore, soil was mostly composed of guano, and δ15N values averaged 22.24±2.75‰ for guano-rich soil, 19.94±1.66‰ for leaves of C. scabrida, and 0.87±0.82‰ for leaves of P. pallida.
Spatial Distribution of Naturally Established Trees
Overall, the density of C. scabrida in our study plots was higher than the density of P. pallida (Fig. 3). We found differences in tree abundance among sea-distance classes for both P. pallida (Wald χ2 = 41.37; p = 0.028) and C. scabrida (Wald χ2 = 121.79; p<0.001). However the distribution pattern was different. Whereas the abundance of C. scabrida was higher close to shore, the abundance of P. pallida was lower. Within the first 50 m from shore, the density of C. scabrida was higher than the density of P. pallida (T-test; p = 0.01; Fig. 3).
Survival of Planted Seedlings
The survival rate of P. pallida seedlings was lower than the survival of C. scabrida seedlings; Fig. 4). Protection from lizards had no significant effect on the survival of the two species (Robust-test; p = 0.085 and p = 0.106 for P. pallida and C. scabrida respectively). Since the effects of herbivores were not significant, we modelled seedling survival as a function of species and distance from shore including a species *distance interaction. We found that only the interaction was significant (p<0.05) indicating that the species responded differently to the sea distance gradient which is also reflected in the models for each species separately. Distance from shore had a positive significant effect on the survival rate of C. scabrida (Robust-test; p = 0.0004; Fig. 4) but no effect on the survival of P. pallida seedlings (Robust-test; p = 0.826; Fig. 4).
Our study did not find a significant effect of herbivory on tree seedling survival in a tropical dry forest subsidized by seabird guano. Woody plant species are common in tropical ecosystems receiving large amounts of seabird guano , , . Previous studies have documented high densities of lizards in systems subsidized by marine-derived resources , , . Therefore, we anticipated a positive effect of seabird-derived subsidies on the strength of top-down interactions at our study site. Our negative results confirm that it is difficult to predict the effect of seabird-derived resources on species interactions, or to generalize their effects on food web composition and functioning .
Our focal lizard species Dicrodon guttulatum is a widespread lizard in the Sechura desert  and has been identified as a major limiting factor for the regeneration of inland dry forests in northern Peru , . At our coastal site experimental protection from lizard herbivory did not affect survival of the two tree species. Therefore, top-down control by herbivorous lizards seems to be weaker in coastal dry forests than it is in inland dry forests. This difference might be explained by lower lizard density in coastal sites (unpublished data). Dicrodon guttulatum feeds primarily on P. pallida , and the abundance of P. pallida can be much higher at inland sites than it is at our coastal site, presumably supporting larger populations of lizards.
Although herbivory did not play a role in seedling survival, we observed differences in the distribution of the two tree species at our coastal site. Nutrient and stable isotopic analyses suggest alternative explanations for the observed pattern of increased dominance of the non-fixing species close to shore. The observed patterns of δ15N and N concentrations in foliar tissue (Fig. 2) suggest a positive response to increased soil N concentrations near shore for the non-fixing C. scabrida but not for the fixing P. pallida. This result is in line with our hypothesis that sea-nutrient fertilization has a positive effect on N assimilation of the non-N fixer. The higher concentrations of NH4 and NO3 close to shore indicate that more soil N is available to plants near shore. Leaf tissues of C. scabrida trees growing near shore had elevated δ15N values, suggesting assimilation of marine-derived N because δ15N in guano-rich soil had similarly high values (Fig. 2). Higher N concentrations in C. scabrida tissue near shore support this result. In this coastal ecosystem, C. scabrida is the most abundant plant species suggesting that it can be an important vector of marine-derived nutrients to the terrestrial trophic food web through diverse types of consumers feeding on its leaves or fruits (e.g. foxes, goats, birds and insects).
The δ15N of P. pallida was always close to zero, suggesting that atmospheric N was the source. Moreover, the N concentrations in foliar tissue varied little between P. pallida plants growing in N-rich soils near shore and plants growing further inland where N concentrations were lower. These results are particularly interesting compared to previous studies. Elevated soil N concentrations could inhibit N fixation , for example Prosopis juliflora had lower N-fixation rates and higher δ15N under saline conditions . Our data suggest that P. pallida assimilates atmospheric-fixed N regardless of soil N concentration in the soil and distance from shore. An alternative hypothesis, differences in fractionation during N assimilation, is unlikely to explain these patterns because our aggregated samples pooled leaves from different parts of the plants and from several individuals at each distance class, and because we collected replicated samples along a linear gradient of guano concentration. Therefore, given the lack of correspondence between the spatial patterns of δ15N in soil and stability of δ15N in leaves of P. pallida, the most parsimonious explanation is that P. pallida does not assimilate N from the soil even in places where NH4 and NO3 concentrations are high.
The density of naturally established trees of both species suggests a response to the sea-inland nutrient patterns consistent with results from the stable isotope analysis. Near shore, the density of C. scabrida was higher than the density of P. pallida. Density of P. pallida was low near shore, increased between 50 and 100 m, and remained relatively constant and similar to C. scabrida further away. These data suggest that C. scabrida growing near shore benefits more from high nutrient concentrations than P. pallida. High nutrient concentration may affect the competitive ability of the two tree species by promoting growth or survival of the non-fixing species. Our seedling transplant experiment did not detect an effect of distance from shore on survival. Therefore, high nutrient concentrations are likely to enhance growth rate of the non-fixer without affecting the growth rate of the fixer, as suggested by patterns of increased tissue N concentrations in C. scabrida near shore, and no change in tissue N concentration in P. pallida. The positive effects of marine nutrient subsidies on C. scabrida were restricted to areas very close to the sea guano deposits. Remarkably localized effects have also been found in other seabird studies worldwide , as well as in the transfer of marine-derived nitrogen accumulated and transported by salmon to riverine forests of North America , .
An alternative hypothesis to explain increased dominance of the non-fixer near shore is that the fixer is suppressed by the guano nitrogen due to toxicity. This hypothesis relies on the assumption that P. pallida are getting their nitrogen from fixation rather than from the guano-rich soil near shore, as opposed to C. scabrida incorporating nitrogen from guano-rich soil. This assumption is supported by constant δ15N values in the fixer and decreasing δ15N in the non-fixer with distance from shore. Moreover, in a small laboratory experiment on the germination of P. pallida (n = 50 seeds), we found that seeds germinated better in soils coming from ≥50 m from shore than in guano-rich soils collected near shore . However, our field experiment results show that seedling establishment close to shore is difficult for both tree species. The planted seedlings of C. scabrida did not grow well and ultimately died near shore, whereas the survival of P. pallida was very low throughout the 200 m sea-inland gradient (Fig. 4).
The toxic effect of N for both species’ seedlings could be explained either by the acidifying effect of ammonia on the soil , by N toxicity, or by a combination of both . The effects of P on nitrogen and non-nitrogen fixers are complex. Nitrogen fixing trees require relatively high supplies of P for the process of fixation . In natural populations, the rates of N fixation increase with increasing soil P availability . However, under non-limiting N conditions, experimental evidence shows that N acquisition strategies change with P supply, and fixers rely less on N fixation .
It has been suggested that the toxicity effects of guano may be more common in hyperarid regions . We speculate that nutrient leaching during rainy years could drop nutrient concentrations below toxic levels. This would explain the apparent discrepancy between the natural distribution patterns and the outcome of the field experiment in our study system. Indeed the benefits of marine nutrient fertilization on arid plant communities can be particularly strong during rainy years, like those associated to ENSO, when water availability is not limiting , . It has been suggested that this positive effect can build up organic matter that in turn improves soil water retention and, as a consequence, long-term guano accumulation can have a positive effect on plant productivity even in dry years . However, the relationship between guano accumulation and soil moisture is not always positive .
Salinity is another environmental factor that could change along the sea-inland gradient. Earlier work indicates that both P. pallida  and C. scabrida  can tolerate high soil salinity, but future manipulative experiments may contribute to further separate the role of nutrients and salinity.
We conclude that the input of marine-derived N through guano deposited by seabirds feeding in the Pacific Ocean affects the two dominant tree species of the coastal dry forests of northern Peru in contrasting ways. The non-N fixing species, C. scabrida may benefit from sea nutrient subsidies by incorporating marine-derived N into its foliar tissues, whereas P. pallida, capable of atmospheric N fixation, does not.
Assessment of microclimate changes introduced by the herbivore exclusions.
Results of general linear models using distance to shore as fixed factor and transect as a random factor.
We thank Petroperú, Universidad de Piura and R. Rodríguez for logistic support. L. Urbina and N. Atarama assisted in the field and in the laboratory. E. Guerra shared her expertise on N fixation. A.M. van den Driessche conducted the soil nutrient analyses. P. Jansen supervised the survival analysis. E. N. Broadbent, R. J. Naiman and E. Chaneton commented on a previous version of this manuscript.
Conceived and designed the experiments: GH AC MH. Performed the experiments: GH AC. Analyzed the data: GH AC MH. Contributed reagents/materials/analysis tools: GH AC MH. Wrote the paper: GH AC MH.
- 1. Koepcke HW, Koepcke M (1952) Sobre el proceso de transformación de la materia orgánica en las playas arenosas marinas del Perú. Publ Mus Hist Nat Javier Prado Ser A 8: 1–24.
- 2. Mellbrand K, Lavery PS, Hyndes G, Hambäck PA (2011) Linking land and sea: different pathways for marine subsidies. Ecosystems 14: 732–744.
- 3. Piovia-Scott J, Spiller DA, Schoener TW (2011) Effects of experimental seaweed deposition on lizard and ant predation in an island food web. Science 331: 461–463.
- 4. Polis GA, Anderson WB, Holt RD (1997) Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annu Rev Ecol Syst 28: 289–316.
- 5. Spiller DA, Piovia-Scott J, Wright AN, Yang LH, Takimoto G, et al. (2010) Marine subsidies have multiple effects on coastal food webs. Ecology 91: 1424–1434.
- 6. Anderson WB, Polis GA (1999) Nutrient fluxes from water to land: seabirds affect plant nutrient status on Gulf of California islands. Oecologia 118: 324–332.
- 7. Polis GA, Hurd SD, Jackson TD, Sánchez-Piñero F (1997) El Niño effects on the dynamics and control of an island ecosystem in the Gulf of California. Ecology 78: 1884–1897.
- 8. Wait DA, Aubrey DP, Anderson WB (2005) Seabird guano influences on desert islands: Soil chemistry and herbaceous species richness and productivity. J Arid Environ 60: 681–695.
- 9. Catenazzi A, Donnelly MA (2007) The Ulva connection: marine algae subsidize terrestrial predators in coastal Peru. Oikos 116: 75–86.
- 10. Polis GA, Hurd SD, Jackson TD, Sánchez-Piñero F (1998) Multifactor population limitation: variable spatial and temporal control of spiders on Gulf of California islands. Ecology 79: 490–502.
- 11. Stapp P, Polis GA, Piñero FS (1999) Stable isotopes reveal strong marine and El Niño effects on island food webs. Nature 401: 467–469.
- 12. Anderson WB, Wait DA, Stapp P (2008) Resources form another place and time: responses to pulses in a spacially subsidized system. Ecology 89: 660–670.
- 13. Fariña JM, Sepulveda M, Reyna MV, Wallem KP, Ossa-Zazzali PG (2008) Geographical variation in the use of intertidal rocky shores by the lizard Microlophus atacamensis in relation to changes in terrestrial productivity along the Atacama Desert coast. J Anim Ecol 77: 458–468.
- 14. Polis GA (1998) Stability is woven by complex webs. Nature 395: 744–745.
- 15. Catenazzi A, Donnelly MA (2007) Role of supratidal invertebrates in the decomposition of beach-cast green algae (Ulva sp.). Mar Ecol Prog Ser 349: 33–42.
- 16. Polis GA, Hurd SD (1996) Linking marine and terrestrial food webs: allochthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. Am Nat 147: 396–423.
- 17. Catenazzi A, Donnelly MA (2008) Sea lions (Otaria flavescens) as hosts of the common vampire bat (Desmodus rotundus). Mar Ecol Prog Ser 360: 285–289.
- 18. Rose MD, Polis GA (1998) The distribution and abundance of coyotes: The effects of allochthonous food subsidies from the sea. Ecology 79: 998–1007.
- 19. Ellis JC (2005) Marine birds on land: a review of plant biomass, species richness and community composition in seabird colonies. Plant Ecol 181: 227–241.
- 20. Schmidt S, Dennison WC, Moss GJ, Stewart GR (2004) Nitrogen ecophysiology of Heron Island, a subtropical coral cay of the Great Barrier Reef, Australia. Funct Plant Biol 31: 517–528.
- 21. Sánchez-Piñero F, Polis GA (2000) Bottom-up dynamics of alloctonous input: direct and indirect effects of seabirds on islands. Ecology 81: 3117–3132.
- 22. Murphy RC (1925) Bird islands of Peru. New York and London: GP Putnam’s Sons.
- 23. Nesbit J, C. (1856) On agricultural chemistry, and the nature and properties of Peruvian guano. London and New York: Longman and Co. 128 p.
- 24. Baumberger T, Affre L, Torre F, Vidal E, Dumas P-J, et al. (2012) Plant community changes as ecological indicator of seabird colonies’s impacts on Mediterranean Islands. Ecol Indic 15: 76–84.
- 25. Mulder CPH, Anderson WB, Towns DR, Bellingham PJ (2011) Seabird islands: Ecology, invasion, and restoration. New York: Oxford University Press.
- 26. Young HS, McCauley DJ, Dirzo R (2011) Differential responses to guano fertilization among tropical tree species with varying functional traits. Am J Bot 98: 207–214.
- 27. Holmgren M, López BC, Gutiérrez JR, Squeo FA (2006) Herbivory and plant growth rate determine the success of El Niño Southern Oscillation-driven tree establishment in semiarid South America. Global Change Biol 12: 2263–2271.
- 28. Illius AW, O’Connor TG (2000) Resource heterogeneity and ungulate population dynamics. Oikos 89: 283–294.
- 29. Hopcraft JGC, Olff H, Sinclair ARE (2009) Herbivores, resources and risks: alternating regulation along primary environmental gradients in savannas. Trends Ecol Evol 25: 119–128.
- 30. Mohtadi M, Hebbeln D, Marchant M (2005) Upwelling and productivity along the Peru-Chile Current derived from faunal and isotopic compositions of planktic foraminifera in surface sediments. Mar Geol 216: 107–126.
- 31. INRENA (1998) Mapa de Bosques Secos del Departamento de Piura. Memoria Descriptiva. Lima: Ministerio De Agricultura, Instituto Nacional de Recursos Naturales, Proyecto Algarrobo y Gobierno de los Países Bajos.
- 32. Pasiecznik NM, Felker P, Harris PJC, Harsh LN, Cruz G, et al.. (2001) The Prosopis juliflora-Prosopis pallida complex: A monograph. Coventry, UK: HDRA.
- 33. Catenazzi A, Donnelly MA (2007) Distribution of geckos in northern Peru: long-term effect of strong ENSO events? J Arid Environ 71: 327–332.
- 34. Sitters J, Holmgren M, Stoorvogel JJ, López BC (2012) Rainfall-tuned management facilitates dry forest recovery. Restor Ecol 20: 33–42.
- 35. Squeo FA, Holmgren M, Miagros J, Albán L, Reyes J, et al. (2007) Tree establishment along an ENSO experimental gradient in the Atacama desert. J Veg Sci 18: 195–202.
- 36. Scheffer M, Van Nes EH, Holmgren M, Hughes T (2008) Pulse-driven loss of top-down control: the critical-rate hypothesis. Ecosystems 11: 226–237.
- 37. Bernex de Falen N, Reves B (1988) Atlas regional de Piura. Lima: Centro de Investigación y Promoción del Campesinado (CIPCA) y Pontificia Universidad Católica del Perú.
- 38. Shearer G, Kohl DH, Virginia RA, Bryan BA, Skeeters JL, et al. (1983) Estimates of N2-fixation from variation in the natural abundance of 15N in Sonoran desert ecosystems. Oecologia 56: 356–373.
- 39. Perevolotsky A (1991) Goats or scapegoats - the overgrazing controversy in Piura, Peru. Small Ruminant Res 6: 199–215.
- 40. Rodríguez-Rodríguez EF, Bussmann RW, Arroyo-Alfaro SJ, López-Medina SE, Briceño-Rosario J (2007) Capparis scabrida (Capparaceae) a species from Perú and Ecuador in urgent need of conservation plans. Arnaldoa 14: 269–282.
- 41. Van Leeuwen J, Catenazzi A, Holmgren M (2011) Spatial, ontogenetic and sexual effects on the diet of a Teiid lizard in arid South America. J Herpetol 45: 472–477.
- 42. Mituzani H, Wada E (1988) Nitrogen and carbon isotope ratios in seabird rookeries and their ecological implications. Ecology 69: 340–349.
- 43. Young HS, McCauley DJ, Dunbar RB, Dirzo R (2010) Plants cause ecosystem nutrient depletion via interruption of bird-derived spatial subsidies. Proc Natl Acad Sci U S A 107: 2072–2077.
- 44. Wada E, Kadonaga T, Matsuo S, Shibata R, Torii T (1975) 15N abundance in nitrogen of naturally occurring substances and global assessment of denitrification from isotopic viewpoint. Geochem J 9: 139–148.
- 45. Therneau T (2012) A Package for survival analysis in S. R package version. 2: 36–14.
- 46. Burger AE (2005) Dispersal and germination of seeds of Pisonia grandis, an Indo-Pacific tropical tree associated with insular seabird colonies. J Trop Ecol 21: 263–271.
- 47. Barrett K, Anderson WB, Wait DA, Grismer LL, Polis GA, et al. (2005) Marine subsidies alter the diet and abundance of insular and coastal lizard populations. Oikos 109: 145–153.
- 48. Kolb GS, Young HS, Anderson WB (2011) Effects of seabirds on island consumers. In: Mulder CPH, Anderson WB, Towns DR, Bellingham PJ, editors. Seabird islands: Ecology, invasion, and restoration. New York: Oxford University Press. 212–241.
- 49. Kurdali F, Al-Shamma’a M (2009) Natural abundances of 15N and 13C in leaves of some N2-fixing and non-N2 fixing trees and shrubs in Syria. Isotopes Environ Health Stud 45: 198–207.
- 50. Vitousek PM, Howarth RW (1991) Nitrogen limitation on land and in the sea: how can it occur? Biogeochemistry 13: 87–115.
- 51. Caut S, Angulo E, Pisanu B, Ruffino L, Faulquier L, et al. (2012) Seabird modulations of isotopic nitrogen on islands. PLoS One 7: e39125.
- 52. Helfield JM, Naiman RJ (2001) Effects of salmon-derived nitrogen on riparian forest growth and implications for stream productivity. Ecology 82: 2403–2409.
- 53. Hilderbrand GV, Hanley TA, Robbins CT, Schwartz CC (1999) Role of brown bears (Ursus arctos) in the flow of marine nitrogen into a terrestrial ecosystem. Oecologia 121: 546–550.
- 54. Havik G (2010) The role of seabird nitrogen input and lizard herbivores on the recruitment of dominant tree species in coastal dry forests of north Peru. Master thesis. Wageningen, The Netherhlands: Wageningen University.
- 55. Britto DT, Kronzucker HJ (2002) NH4+ toxicity in higher plants: a critical review. J Plant Physiol 159: 567–584.
- 56. Vitousek PM, Cassman K, Cleveland C, Crews T, Field CB, et al. (2002) Towards an ecological understanding of biological nitrogen fixation. Biogeochemistry 57: 1–45.
- 57. Isaac ME, Harmand JM, Lesueur D, Lelon J (2011) Tree age and soil phosphorus conditions influence N-fixation rates and soil N dynamics in natural populations of Acacia senegal. Forest Ecol Manage 261: 582–588.
- 58. Isaac ME, Harmand JM, Drevon JJ (2011) Growth and nitrogen acquisition strategies of Acacia senegal seedlings under exponential phosphorus additions. J Plant Physiol 167: 776–781.