CAN ENVIRONMENTAL VARIATION AFFECT SEEDLING SURVIVAL OF PLANTS IN NORTHEASTERN MEXICO ?

The effects of global warming increase the frequency and intensity of many climate events such as rainfall. We evaluated the effects of environmental conditions on early stage seedling survival of the native thorn scrub species Caesalpinia mexicana A. Gray, Celtis pallida Torr., Cordia boissieri A. DC., and Ebenopsis ebano (Berland.) Barneby and J.W. Grimes, during the summer of 2009 and 2010. The experimental design had two factors, two levels of rainfall and three microhabitats of thorn scrub: (i) open interspace, (ii) thorn scrub edge and (iii) under the canopy of dense thorn scrub. In dense thorn scrub, seedling survival was higher for Caesalpinia mexicana and Celtis pallida, and for Cordia boissieri and Ebenopsis ebano seedling survival was higher in dense thorn scrub and thorn scrub edge. The effect of rainfall on seedling survival depended on the year. Rainfall in 2010 and dense thorn scrub increased seedling survival of native species. For survival, the limiting factors of microhabitats appear to change across the years. Besides rainfall events, biological aspects like competition and mycorrhiza effects would need to be considered in models of plant establishment.


INTRODUCTION
The early stages of plant growth such as seed germination and seedling survival are most affected by adverse conditions (Harper 1977;Kitajima and Fenner 2000), because survival is strongly affected by biotic and abiotic conditions (Ibañez and Shupp 2001).Water is one of the major factors limiting plant establishment in many ecosystems (Valiente-Banuet and Ezcurra, 1991;Flores et al., 2004).By ameliorating abiotic and biotic stresses, woody plants facilitate the recruitment of plants (García-Moya and McKell, 1970;Franco and Nobel 1989;Tirado and Pugnaire, 2003;Pugnaire et al., 2004), due to microclimatic conditions, including lower evapotranspiration rates and changes in the light quality (Murchie and Horton, 1998;Valladares and Puignaire, 1999;Maestre et al., 2001), and improved soil physical and chemical properties (Cerdá, 1997;Bochet et al., 1999).Global warming can affect rainfall patterns (Hughes, 2000) and thereby plant survival.The objective of this research was to evaluate the effect of yearly variation of rainfall as a limiting factor for seedling survival between contrasting microhabitats: (i) dense thorn scrub; (ii) thorn scrub edge and (iii) open thorn scrub.The hypothesis of this research predicts that yearly environmental variations due to global warming effects are causing yearly differences in microhabitat suitability for seedling establishment.

Study species
The species chosen for this study are quite common in northeastern Mexico.E. ebano (ebano) is a large tree of up to 10 m tall and has a large dense canopy; C. mexicana (árbol del potro) grows under 5 m in height and has an upright, low-density canopy (Estrada and Marroquín, 1991); Cordia boissieri (anacahuita) is a small tree with a growth of up to 6 m (Estrada et al. 2005) and C. pallida (granjeno) is a broadleaf tree of up to 3 m (Benson and Darrow, 1981).

Study site
The study was conducted in an area of thorn scrub near Linares (Nuevo Leon) in northeastern Mexico (24º 51' N, 99º 34' W), from June to September in 2009 and 2010.The climate is sub-humid, semi-tropical with an annual rainfall ranging from 713.3 to 1058 mm that falls mainly in late summer and early autumn.Mean annual temperature is 22.9°C, with a maximum in summer of 45°C and a minimum of -2°C (Cavazos and Molina, 1992).Total rainfall was very different between the two years (Fig. 1); for 2009 it was 154.8 mm and for 2010 i was 716.8 mm.

Treatment of seeds
Seeds of all species were collected in 2008 from the study area from at least 20 mother plants to allow for genetic variation.The seeds were thoroughly mixed; air dried, stored in a site to cool and dry in bottles with silica gel bags for moisture control and left in paper bags for potential after-ripening for 3 months.The seeds were not tested for viability, and no fungal inhibitors were applied to them.All seeds except Caesalpinia mexicana were scarified prior to germination.Mechanical scarification with sand paper was preferred as it has been shown be an effective treatment to promote germination for seeds with hard integuments in northeastern Mexico (Foroughbakhch, 1989).

Experimental design
The effects of rainfall limitation and microhabitat for seedling survival were evaluated with two factor ANOVAs.Rainfall exposure was: (1) rainfall of summer 2009 and (2) 2010 (Fig. 1).Microhabitats used were: (i) dense thorn scrub, under cover of woody vegetation; (ii) thorn scrub edge and (iii) open thorn scrub, without woody vegetation, in a mosaic of herbaceous and bare soil.These three microhabitats differed in important characteristics such as soil characteristics, moisture on the soil surface and light intensity (Table 1).Seeds were germinated on germination trays with the soil of the three microhabitats collected from the study site.At 7 weeks the seedlings were transplanted into each microhabitat (June 15, 2009 and 2010), into holes of 15 cm diameter x 10 cm deep; to avoid stress the transplanted seedlings were supplied with water in situ.Transplanted seedlings size (cm) and number of leaves for 2009 and 2010 were: C. mexicana 6.5 ± 0.8; 6.8 ± 1.2, 6.9 ± 0.9; 7.1 ± 0.8, (mean ± S.D.).C. pallida 4.1 ±.0.3; 4.9 ± 0.6, 4.3 ± 0.7; 4.7 ± 1.1.C. boissieri 4.2 ± 0.4; 4.9 ± 0.3, 4.0 ± 0.6; 5.1 ± 0.8 and E. ebano 3.9 ±.0.5; 3.6 ± 0.8, 3.7 ±.0.5; 3.5 ± 0.6.For insect protection the seedlings were sprayed with a dilution of synthetic pyrethrin insecticide once per week.All environments were within an area of 20 ha, on a deep vertisol.Seedlings in all microhabitats were fenced to avoid predation by small mammals.
There were a total of 2400 seedlings, 600 per species (10 seedlings x 3 microhabitats x 2 rainfall treatment x 10 replicate sites).Seedling survival was monitored each week.Percentage of seedling survival was calculated as: percentage of seedling survival (number of live seedlings/total seedlings) * 100.Differences in seedling survival were detected using a two-way ANOVA (α= 0.05).Tukey's T-test was used to verify statistically significant differences between the mean values of the examined factors.Prior to analysis of seedling establishment, data were arcsine square-root transformed to obtain a normal distribution (Sokal and Rholf, 1995).Ninety-five percent confidence intervals were used in graphics to highlight differences for each variable.

RESULTS
For all species survival analysis differed between the years of exposure to rainfall (d.f.= 1, P< 0.001), and according to Tukey's T-test (Table 2) all species showed better seedling survival in 2010.Overall 11 and 37% of the seedlings were alive at the end of the summer of 2009 and 2010.During 2009 the mortality rate was greatest through the summer due to high temperatures and lack of rainfall (Fig. 1), while in 2010 rainfall effects increased seedling survival (Table 2; Fig. 2).
For C. mexicana, 98 seedlings were alive: 55 in dense thorn scrub (37 in 2010 and 18 in 2009) For each species seedling survival was significantly (d.f.= 2, P< 0.001) associated with environment (d.f.= 2, P< 0.001).All species showed better seedling survival in dense thorn scrub (Table 2, Fig. 2), while in edge thorn scrub Cordia boissieri and Ebenopsis ebano also showed higher seedling survivor.The interaction rainfall exposure x microhabitats was significant (d.f.=2, P< 0.001); dense thorn scrub (Table 2) for all species, and also in edge thorn scrub for Cordia boissieri and Ebenopsis ebano, (Table 2).The significant twoway interaction demonstrated that the rainfall of 2010 tended to increase survival on edge and dense thorn scrub, while it had virtually no affect on open interspace.

Ebenopsis ebano
6.6 a 5.4 a 2.9 b 2.0 bc 1.3 c 0.6 c are prone to drought because their rainfall amount critically depends on a few rainfall events (Sun et al., 2006).Moisture due to rainfall in the summer of 2010 in dense and edge thorn scrub had positive effects on seedling survival.
In this study, seedling survivor was limited by water availability in 2009, with high mortality of the species studied here probably caused by the low moisture in the surface of the soil, suggesting an intense environmental selection.Differences found here between species to response stress showed different patterns to ecological stress gradients (Gitlin et al., 2006;Bernal et al., 2011) The nurse plant in dense thorn scrub appears to exert a crucial positive influence on seedling survival in dry years (Jurado et al., 2006;García and Jurado 2003), or the capacity for resource-sharing through common mycorrhizal networks (Chiariello et al., 1982).However, seedling survival was greater in the wet year 2010 (Table 2; Fig. 2), with dense and edge thorn scrub appearing to enhance the effects of rainfall for 2010.
For thorn scrub species climatic changes could affect seedling abundance (Jurado et al., 2011) due to changes in rainfall (Van der Waal et al., 2009;Kusnierczyk and Ettl., 2002) and temperature (Hughes, 2000;Thomas et al., 2004).In addition to microhabitats, differences in rainfall between years could be the main factor limiting seedling survival; rainfall changes may also affect seedling recruitment by direct effect on survival.For the thorn scrub ecosystem, changing rainfall patterns caused by climate change can affect the models for establishment of plants.

Fig. 1 .
Fig. 1.Mean, maximum, and minimum temperatures; rainfall, during summer (June to September) 2009 and 2010 on microhabitat of thorn scrub in Northeastern México Data provided by the Mexican National Water Commission weather station (CNA).

Table 1 .
Environmental differences among microhabitats.Soil characteristics are from analyses of the surface soil samples from a representative pedon of each microhabitat type.Litter samples are from 30×30 cm samples, n=10.Light intensity was measured with a lux meter (LX-1010B) for 5 days at noon in June 2009.