Advertisement

Plant and Soil

, Volume 424, Issue 1–2, pp 503–524 | Cite as

Effects of UV radiation and rainfall reduction on leaf and soil parameters related to C and N cycles of a Mediterranean shrubland before and after a controlled fire

  • L. Díaz-Guerra
  • D. Verdaguer
  • M. Gispert
  • G. Pardini
  • J. Font
  • J. A. González
  • E. Peruzzi
  • G. Masciandaro
  • L. Llorens
Regular Article

Abstract

Background and aims

In the Mediterranean basin, reduction in cloudiness owing to climate change is expected to enhance solar ultraviolet (UV) levels and to decrease rainfall over the coming years, which would be accompanied by more frequent and intense wildfires. The aim of the present study was to investigate the role of solar UV-A and UV-B radiation in C and N pools of a Mediterranean shrubland and whether drier conditions could alter this role before and after a fire.

Methods

Over a three-year field experiment, 18 plots of 9 m2 were subjected to three UV conditions (UV-A + UV-B exclusion, UV-B exclusion or near-ambient UV-A + UV-B exposure) combined with two rainfall regimes (natural or reduced rainfall). Several parameters related to C and N cycles in the soil and in the leaves and litter of two dominant plant species (Arbutus unedo and Phillyrea angustifolia) were measured before and after an experimental fire.

Results

UV-A exposure increased soil moisture throughout the study period, as well as respiration before the fire. The additional presence of UV-B decreased β-glucosidase activity at 5–10 cm depth and soil respiration and pH. UV-B exposure also raised leaf C concentration in P. angustifolia and δ15N values in A. unedo. Reduced rainfall often emphasized the opposite effects of UV-A and UV-B on the studied parameters. After the fire, most of the UV and rainfall effects were lost.

Conclusion

UV-A exposure seems to stimulate soil biological activity and, thus, C and N turn-over, while the effect of UV-B would be the opposite. At least in the short term, the “homogenizing influence” of fire would probably have a stronger effect on the C and N cycles than the expected changes in UV and rainfall levels.

Keywords

Carbon cycle Drought Fire Mediterranean shrublands Nitrogen cycle UV radiation 

Introduction

As a consequence of climate change, cloudiness reduction in the Mediterranean basin will decrease overall precipitation and increase ultraviolet (UV) radiation fluxes, both UV-B (280–315 nm) and UV-A (315–400 nm), reaching terrestrial ecosystems in the near future (IPCC 2013; Sanchez-Lorenzo et al. 2017). Models also predict that Mediterranean ecosystems will be exposed to an increase in fire frequency over the coming years (IPCC 2013), which could trigger changes in plant communities favoring the persistence and expansion of highly resilient communities such as Mediterranean shrublands (Acácio et al. 2009). Shrub ecosystems have been spreading in Spain and other parts of Europe in the last decades (Tárrega et al. 2001; Riera et al. 2007), as a result of the increase in wildfire occurrence together with agricultural abandonment (Díaz-Delgado et al. 2002; Lloret et al. 2002). Mediterranean-type terrestrial communities deserve special attention for its role in fuel potential, plant variety and soil quality. Despite this, current knowledge about the UV effects on the functioning, and in particular on the biogeochemical cycles, of Mediterranean shrublands is limited, with even less information being available about the interactive effects between UV levels and other environmental factors, such as water availability or fire (Zepp et al. 2007; Sardans and Peñuelas 2013).

Specifically, increases in UV-B and UV-A may directly alter C and N cycles of Mediterranean shrublands through the stimulation of photodegradation of plant litter and its phototransformation into soil microorganism-available forms. In arid and semi-arid environments, photodegradation by direct sunlight exposure plays an important role in the breakdown of organic matter, particularly because of a UV-induced decline in the lignin concentration of the soil litter (Day et al. 2007; Henry et al. 2008; Dirks et al. 2010). Enhanced lignin degradation in plant litter leaves the N easily available to microbes (Foereid et al. 2010) facilitating the enzymatic degradation and the microbial access to labile C compounds (Austin and Ballaré 2010; Baker and Allison 2015). However, direct sunlight exposure in the UV range can also be harmful for soil microorganisms (Hughes et al. 2003), somewhat hindering the C and N release by biological decomposition (Zepp et al. 2007).

UV radiation effects on litter decomposition and, thus, on C and N cycles may also be mediated by UV-induced chemical responses in plants which can vary depending on the species (Caldwell et al. 2007; Austin et al. 2016). Exposure to enhanced UV radiation can increase plant production of phenylpropanoid compounds, such as phenols (Searles et al. 2001; Bassman 2004; Julkunen-Tiitto et al. 2005; Li et al. 2010), which are used as UV-absorbing compounds (UACs) and free radical scavengers in leaves (Agati and Tattini 2010). Higher amounts of phenolic compounds in the litter can delay soil organic matter decomposition and mineralization (Castells et al. 2004), and inhibit nitrification due to their harmful effects on soil microorganisms and enzyme activities (Erickson et al. 2000; Castells et al. 2004; Castaldi et al. 2009; Formánek et al. 2014), thus decreasing available soil N. Enhanced UV-B exposure during plant growth may also directly increase (Yue et al. 1998) or decrease (Pancotto et al. 2005), depending on the species (Zepp et al. 1998), leaf N concentration.

The activity of soil enzymes involved in the biological decomposition of organic matter might also be altered by the change in solar UV fluxes (Nannipieri et al. 2002; Caldwell 2005). One of these soil enzymes is β-glucosidase, which controls the C cycle through the breakdown of labile cellulose and other carbohydrate polymers, enhancing nutrient release from organic compounds and thus facilitating microbe metabolism (Sardans et al. 2008a). Despite the importance of this enzyme in the C cycle, present knowledge about how UV radiation affects β-glucosidase activity is limited (Gallo et al. 2006; Choudhary et al. 2013), with even less information available in Mediterranean ecosystems (Baker and Allison 2015). In a field experiment with mung bean cultivars, enhanced UV-B radiation stimulated root accumulation and secretion of phenolic compounds, which depleted microbial biomass of the rhizosphere leading to a reduction of β-glucosidase activity; on the contrary, at the non-rhizosphere soil, reduced root activity resulted in nutrient accumulation, increasing the microbial population and thus β-glucosidase activity (Choudhary et al. 2013). Conversely, several studies performed in dryland ecosystems found that β-glucosidase activity in litter samples was unaffected by changes in UV exposure (Gallo et al. 2006; Baker and Allison 2015). Clearly, more information is needed to disentangle how changes in UV radiation can affect the activity of soil enzymes, and in particular of β-glucosidase.

Carbon and nutrient cycles may also be substantially affected by other components of climate change, such as altered patterns of rainfall, which can interact with UV effects (Erickson et al. 2015). Unlike what happens with UV radiation, the effects of drought on the biogeochemical cycles of Mediterranean ecosystems have been extensively investigated, especially in relation to soil microbial activity and litter decomposition (Incerti et al. 2011; Sardans and Peñuelas 2013). Drier conditions tend to attenuate soil microbial activity, leading to reduced respiration rates (Rey et al. 2002) and enzyme activities (Gallo et al. 2006), along with increases in soil C concentration (Sardans et al. 2008a). In turn, higher C:N ratios would delay the mineralization process and eventually the transformation of organic N into plant-available forms (Bengtsson et al. 2012). With lower plant N uptake, the C:N ratio tends to increase in plants and, consequently, the soil is enriched with hardly mineralizable organic debris. Under water shortage, plants often increase their content of phenolic compounds (Hofmann et al. 2003), which become an additional factor that can further slowdown decomposition rates (Castells et al. 2004). Therefore, soil water availability is related to many variables and processes that may also be affected directly or indirectly by UV radiation. Because of that, rainfall regime might be an important factor modulating UV effects on C and N levels in Mediterranean ecosystems. Indeed, there is evidence that the degree of photodegradation can vary with soil water content (Gallo et al. 2006; Brandt et al. 2007). In addition, metabolic activity of soil microbiota can be strongly limited by both high UV fluxes (Hughes et al. 2003) and low soil moisture (Sherman et al. 2012). Interactive effects between UV and water supply on litter decomposition can also be modulated by plant responses to both factors. At plant level, enhanced UV radiation in combination with low soil moisture conditions have been reported to increase plant production of phenolic compounds being this effect dependent on plant species (Hofmann et al. 2003; Ren et al. 2007). In Mediterranean species, the direction of these effects can vary among specific phenols, despite the total pool of phenols not being changed (Nenadis et al. 2015).

The evolution and dynamics of most Mediterranean-type ecosystems are also linked to wildfires (Lloret et al. 2002; Paula and Pausas 2006), with many species showing post-fire regeneration mechanisms, such as resprouting (Pausas et al. 2004). Plant resprouting capacity is associated to storage of resources in belowground organs to ensure post-disturbance nutrient supply (Verdaguer and Ojeda 2002). In soils of Mediterranean shrublands, decreases in organic C and increases in total N have been reported in the short term after a fire, being dependent on factors such as soil moisture, vegetation type and climatic conditions (Caon et al. 2014). Therefore, effects of UV fluxes and rainfall regime on the biogeochemical cycles of Mediterranean shrublands could be modulated after a fire by changes in soil C and nutrients and the reduction in plant aerial biomass. Moreover, the post-fire regeneration of the vegetation could also be affected by the levels of UV radiation and soil water availability, for instance, through their effects on the capacity of plants to store resources.

In this context, the main objectives of this study were: 1) to assess the role of UV radiation (UV-A and UV-B) on the C and N cycles of a Mediterranean shrubland, before and after a fire, and 2) to elucidate whether this role can be altered by water availability. To achieve these goals, we performed a field experiment where the levels of UV and rainfall reaching the ecosystem were modified. Different parameters related to C and N cycles were measured at soil, litter and plant level before and after an experimental fire. We hypothesized that: (i) UV exposure will affect soil C and N levels through effects on litter decomposition, which would be supported by changes in related variables, such as soil respiration rates and β-glucosidase activity; (ii) UV-induced changes in soil C and N levels will be mediated by alterations in C and N concentrations of plant leaves and litter; (iii) UV effects will be modulated by the amount of rainfall; and (iv) fire-induced changes in soil C and N and/or in plant cover will alter the interactive effects between UV fluxes and rainfall regime on C and N cycles.

Materials and methods

Study area and experimental design

A field experiment involving UV radiation and rainfall reduction was conducted from August 2011 to June 2014 in a Mediterranean shrubland at the Gavarres Massif (41° 53′ 57″ N, 2° 54′ 43″ E) near Cassà de la Selva (Girona, NE of the Iberian Peninsula). The study area was situated at about 250 m above sea level. The vegetation was dominated by Arbutus unedo, Erica scoparia and Phillyrea angustifolia whose relative abundances in the study site just before the experiment (spring 2011) were around 12%, 36% and 17%, respectively. Other woody Mediterranean species present were Quercus suber, Pinus pinaster, Calluna vulgaris, Viburnum tinus, Daphne gnidium, Ulex parviflorus and Cistus salviifolius, along with an herbaceous layer composed mainly of Brachypodium retusum and Carex oedipostyla. The soils of the study area were mostly Inceptisols, classified as Typic Haploxerept according to Soil Taxonomy System (Soil Survey Staff 2010), with A, B, C/R horizon development over a Palaeozoic granitic parent material. Climatological variables, such as global solar irradiation, temperature and rainfall, were monitored throughout the study period (Fig. 1) by a meteorological station located at Cassà de la Selva, 3 km away from the study site.
Fig. 1

Monthly averages of daily global solar irradiation (MJ m−2) and temperature (°C), together with accumulated rainfall (mm) for each month, along the study period. Data set was obtained from the meteorological station of Cassà de la Selva (177 m above sea level, 41° 52′ 28″ N, 2° 55′ 37″ E)

In August 2011, eighteen plots (3 × 3 m per plot) were distributed over the study area on a south-facing slope to ensure a high solar exposure. All plots had a similar slope, and their distribution and the relatively high plant cover of the soil (64% on average before the beginning of the experiment) minimized the effects of sporadic runoff and/or leaching. In each of these plots, plastic filters were installed above the vegetation on metallic frames with a 10° slope towards the south and at a height of around 1.5 m at the center of the plot. These filters were made of different materials, which excluded or transmitted solar UV-A and/or UV-B radiation, allowing the establishment of three different UV conditions (see below). At the south-face side of each plot, a 35 cm-wide filter made of the same type of plastic that covered the plot was also placed in order to prevent plant exposure to unfiltered solar radiation. Filters covering the plots also stopped the rainfall, which was collected in a tank (310 L) placed next to each plot, allowing to combine the three UV conditions with two different rainfall regimes (see below). Each one of the six different UV x rainfall conditions was replicated three times, with plots being distributed in three blocks (six plots per block). In each plot, several parameters related to C and N levels were analyzed at soil, litter and plant leaf level. Litter and plant leaves were studied from the two dominant species A. unedo and P. angustifolia (Table 1). In February–March 2013, all the vegetation of the experimental plots was burned in a controlled fire (see below).
Table 1

Sampling months and parameters analyzed from soil, and from litter and plant leaves of Arbutus unedo and Phillyrea angustifolia before and after the experimental fire

Grey colour indicates sampling months, which correspond to the end of the different seasons, except in the case of February 2013, which is a sampling performed just before the fire. EC, electrical conductivity

* In June 2014, we only collected litter from A. unedo, since production of P. angustifolia litter was too low

UV-radiation treatment

As detailed in Nenadis et al. (2015), the three UV conditions applied were (Table 2):
  • UV0 plots (i.e., plots where UV-A and UV-B were excluded): This condition was achieved by means of a 2-mm-thick polycarbonate filter (PC0100UV, PolimerTecnic, Girona, Spain) which allowed the transmission, on average, of only 5% of UV-B (280–315 nm) and 6% of UV-A (315–400 nm) solar radiation.

  • UVA plots (i.e., plots where UV-B was excluded): Plots under this condition did almost not receive UV-B radiation (3% on average), whereas average transmission of UV-A radiation was 52%. To accomplish this, a 0.25-mm-thick polyester filter (Melinex, Ponscosta, Valencia, Spain) was used.

  • UVAB or control plots (i.e., plots exposed to near-ambient UV radiation levels): These plots were aimed to provide similar microclimate conditions (degree of shading and temperature) to those found in UV0 and UVA plots. They were covered by a 3-mm-thick methacrylate filter (MC0100XN, PolimerTecnic, Girona, Spain), which transmitted, on average, 80.5 and 85% of UV-B and UV-A radiation, respectively.

Table 2

Percentage of UV radiation and photosynthetic photon flux density (PPFD) transmitted through the filter in each UV condition in the field (UVAB, UVA and UV0)

 

UV radiation treatment

UVAB plots

UVA plots

UV0 plots

Filter type

Methacrylate

Polyester

Polycarbonate

UV-A radiation

84–86%

49–55%

5–7%

UV-B radiation

79–82%

2–4%

4–6%

GENa

81–83%

3–4%

4–6%

PGb

83–84%

35–46%

4–5%

PPFD

88–94%

82–87%

77–90%

UV radiation fluxes were expressed as unweighted UV-A and UV-B radiation and also using the plant response action spectrum (GEN) and the new plant growth weighting function (PG)

aPlant response action spectrum according to Caldwell (1971)

bPlant growth weighting function according to Flint and Caldwell (2003)

Spectral transmittances of filter materials in the UV and visible bands were assessed and verified periodically in the laboratory using a deuterium/halogen lamp and a CCD spectrometer (Avantes; The Netherlands). Effective in situ reduction in UV radiation and photosynthetic photon flux density (PPFD) under filters were determined using a double monochromator spectroradiometer (SR9910, Irradian Ltd., UK). Since spectral measurements could not be taken continuously, we measured erythemal irradiance to assess the UV doses by means of two UV-S-E-T Kipp & Zonen sensors (The Netherlands): the first one was placed at the experimental site during several days each season; the second one was located at the radiometric station of the Environmental Physics Group (EPG) at the University of Girona (41° 97′ N, 2° 82′ E, 115 m above sea level), 16 km far from the study site, where it was taking measurements continuously. The erythemal UV irradiance data (UVE; Commission International de l’Éclairage, CIE) in combination with the spectral measurements and radiative modelling allowed obtaining continuous series of unweighted UV irradiances. Series of irradiances weighted according to the generalized plant action spectrum (GEN) from Caldwell (1971) and to the new plant growth response spectrum (PG) from Flint and Caldwell (2003) were also obtained (Nenadis et al. 2015).

UV-A doses were estimated from PPFD measurements in combination with the radiative model. PPFD was determined using continuous measurements of a quantum sensor (Li-190SA, Li-cor, USA), located at the EPG station, which was verified against spectroradiometric measurements. Also, PPFD measurements were performed seasonally at different points of the plots and at vegetation canopy level to confirm that filters reduced or transmitted PPFD levels adequately. Filters were periodically cleaned and they were replaced when radiation transmittance characteristics were not optimal or when they were damaged by strong winds.

Rainfall treatment

Half of the plots received 100% of the natural rainfall (NR plots), whereas the other half were watered with 70% of the rainfall throughout the study period, except in winter when they were watered with 90% (reduced rainfall or RR plots). To achieve these two levels of rainfall, the precipitation collected in the tanks placed beside each plot was used to irrigate the plots according to the above rainfall conditions. Percentages of reduced rainfall were established based on the changes in precipitation expected for the Mediterranean basin in the near future as a consequence of climate change (IPCC 2013). Throughout the study period, soil moisture was significantly lower in RR plots compared to NR ones (Fig. 2), which confirms that the treatment was properly applied.
Fig. 2

Mean values of soil moisture (%) throughout the study period under the two experimental rainfall regimes: natural rainfall (NR) and reduced rainfall (RR). Error bars represent the standard error of the mean (n = 9). The significance level was set at p ≤ 0.05

Experimental disturbance (fire)

Vegetation of the plots was completely burned by specialist firefighting personnel in February–March of 2013. Just before the fire, the entire experimental infrastructure was removed, being rebuilt after the fire. The experimental setup was fully functional again by the end of March 2013.

Soil parameters measured in situ

Measurements of soil moisture, temperature and respiration rates were performed in situ at midday, on sunny days, in five points distributed over each plot area. These parameters were measured at the end of each season throughout one year before the fire, and another year after the fire (Table 1). In the post-fire period, monthly measurements of soil moisture were also taken from May 2013 to June 2014 to confirm that the rainfall treatment was properly applied (Fig. 2). Soil moisture was determined as the percentage of volumetric water content by means of a time domain reflectometer (FieldScout TDR 300 Soil Moisture Meter, Spectrum Technologies, Inc., Aurora, USA), with two 20-cm probe rods, providing instantaneous readings.

Soil CO2 fluxes were measured with a portable infrared gas analyzer (IRGA; CIRAS-2, PP-Systems, Amesbury, USA) connected to an SRC-1 soil respiration chamber. Once the closed chamber (10 cm diameter × 15 cm height) was placed on the soil surface, the flux of CO2 was measured by the IRGA for one minute. Carbon dioxide concentration was then calculated and expressed as μmol CO2 m−2 s−1. Data were calibrated according to soil temperature, which was determined just before the respiration measurements using a thermometer with a 10-cm probe rod (HANNA Instruments, Woonsocket, USA).

Soil parameters measured in the laboratory

For each plot, soil was sampled at two depths (A, 0–5 cm; and B, 5–10 cm) at the end of autumn and spring before the fire (December 2011 and June 2012) and after the fire (December 2013 and June 2014) (Table 1). Soil was also sampled just before the experimental fire (February 2013). At each sampling date, and for each depth, samples were collected from five points distributed over the plot area and then mixed and homogenized in order to have one representative sample per plot and depth. In the laboratory, samples were air dried and sieved to 2 mm before the analyses. Soil organic C and total N were analyzed for each one of the five seasons. Soil pH and electrical conductivity were measured for all the samples except for June 2012, while β-glucosidase enzyme activity was analyzed for all the samples except for February 2013 (Table 1).

Organic C was quantified by the dichromate wet oxidation method in presence of concentrated sulphuric acid (Forster 1995). The concentration of total N was determined by means of the Kjeldahl method (Forster 1995). Briefly, 1 g of soil was digested with 98% H2SO4 for 1 h at 175 °C and 1.5 h at 370 °C for organic N mineralization. Ammonium was then distilled with a Kjeldahl Distiller Pro-nitro I (J.P. Selecta, Instrumentación Científica Técnica S.L., La Rioja, Spain).

Soil pH was determined using 1:2.5 soil water ratios and a Crison 20 pH meter, and electrical conductivity with a 1:5 soil water ratios and a Crison micro CM 2200 conductivity meter (Crison Instruments S.A., Barcelona, Spain).

The determination of β-glucosidase activity was conducted using the method of Masciandaro et al. (1994), which is based on the release of ρ-nitrophenol (ρNP) from the 0.05 M 4-nitrophenyl-β-D-glucopyranoside (ρNPG), used as substrate of the enzyme (Hayano and Tubaki 1985). The concentration of ρNP released from 0.5 g of dried soil was determined spectrophotometrically at 398 nm (Tabatabai and Bremner 1969). Thus, β-glucosidase activity was expressed as mg ρNP kg−1 h−1.

Plant and litter cover

Plant and litter cover of each plot was measured by means of the vertical “pin-point” method (Arévalo et al. 2011) just before the start of the treatments (May 2011) and, then, annually throughout the experimental period, in June 2012, 2013 and 2014 (Table 1). In each plot, data were collected from 5 parallel 3-m transects oriented east-west, along which 30 measuring points (one each 10 cm) were taken; hence, in total, 150 data points were considered per plot. For all these points, plant presence or absence was determined, as well as soil cover (which was classified as bare or covered with litter). Then, the percentage of points with vegetation presence, as well as those with soil litter, were calculated in relation to the total number of points sampled per transect, obtaining 5 values of plant and litter cover per plot.

Litter and leaf parameters

Four litter traps were installed in each plot at the beginning of the experiment. Three of these traps were positioned below the three dominant species (A. unedo, P. angustifolia and E. scoparia) whereas the fourth was placed in an area without vegetation. Leaf litter was sampled before (in June and September 2012) and after (in June 2014) the fire (Table 1). For each sampling date, one sample per plot was obtained by joining the leaf litter accumulated in the four traps. After collection, leaf litter of A. unedo and P. angustifolia were separated for subsequent analysis. In June 2014, leaf litter of P. angustifolia was too scarce to be analyzed.

Samplings of A. unedo and P. angustifolia leaves were always conducted at the end of winter and summer before and after the fire (i.e., March and September 2012, September 2013 and March 2014) (Table 1), and always on sunny days during hours of maximum solar irradiation. Leaves of both species were taken from the top of the canopy of each plant, selecting always south-facing fully-developed leaves exposed to solar radiation. For each plot and sampling date, we collected three leaves from three different plants of P. angustifolia, and four leaves from one or two plants of A. unedo (always from different branches).

Once in the laboratory, litter and leaf samples of both species were dried in an oven at 45 °C for 5 days and grounded using a ball mill (Mixer Mill MM 400, Retsch GmbH, Haan, Germany). From each litter sample, three subsamples of 3–4 mg of powder were encapsulated into tin (Sn) capsules to have replicas of each analysis. In the case of leaves, the different samples were analyzed separately. Analyses of C and N concentrations, as well as of 15N and 13C, were performed at the University of California (UC Davis Stable Isotope Facility, Davis, USA), using an elemental analyzer (PDZ Europa ANCA-GSL, Sercon Ltd., Cheshire, UK) linked to a continuous flow isotope ratio mass spectrometer (IRMS; PDZ Europa 20–20 IRMS, Sercon Ltd., Cheshire, UK). The final delta values were expressed relative to atmospheric nitrogen for δ15N and relative to PDB standard for δ13C, according to the following equation:
$$ \updelta \mathrm{Z}={\left({\mathrm{R}}_{\mathrm{sample}}/{\mathrm{R}}_{\mathrm{standard}}-1\right)}^{\ast}\kern.02em 1000 $$
where Z is the heavy isotope of either N or C, and R is the ratio of heavier to lighter isotope (15N/14N or 13C/12C) for the sample and the standard. The long-term standard deviation was 0.3‰ for 15N and 0.2‰ for 13C.

Statistical analysis

A Principal Component Analysis (PCA) was performed using data for six soil variables (moisture, temperature, respiration, organic C, total N and β-glucosidase activity) determined in four sampling dates (December 2011, June 2012, December 2013 and June 2014). The six variables were previously normalized and mean values were used in the case of those variables measured at two soil depths.

To evaluate the differences between pre- and post-fire data for soil and plant leaf parameters, and their interactive effects with the two treatments, we performed three-way ANOVAs using fire, UV and rainfall treatments as fixed factors. To analyze the effects of the two treatments, pre- and post-fire data were also analyzed separately. Soil parameters determined from composite samples per plot and depth (organic C, total N, C:N ratio, pH, electrical conductivity and β-glucosidase) were analyzed by means of repeated-measures ANOVAs for each depth, with UV and rainfall treatments as factors. Treatment effects on soil moisture, temperature and respiration, as well as on leaf C, N, δ13C, δ15N and C:N ratio, were tested by ANOVA analyses, since data for these parameters were obtained from several soil points or leaves per plot. In the case of soil parameters, sampling date, and UV and rainfall treatments were used as factors, while, for leaf parameters, plant species was also included as a factor. To avoid pseudoreplication (Hurlbert 1984), mean values of each parameter per plot were used for all these statistical tests.

Treatment effects on plant and litter cover, as well as on litter quality variables (C, N, δ13C, δ15N and C:N) for each species, were analyzed within each sampling date by means of two-way ANOVAs (with UV and rainfall treatments as fixed factors). For plant and litter cover, pre-treatment data (May 2011) was also included in the statistical tests as a co-variable.

In the case of significant UV effects, Fisher’s LSD post-hoc pairwise comparisons were applied to determine differences among UV conditions (UVAB, UVA and UV0). When the interaction between factors was significant, treatment effects were assessed within the levels of the other factor. The Kolmogorov–Smirnov test was used to test normality, while the homogeneity of variances was analyzed with the Levene’s test. For all the statistical tests, the significance level considered was p ≤ 0.05. PCA was done with PRIMER 6 software (PRIMER-E Ltd., Plymouth, UK) and other statistical analyses were done using SPSS software (IBM SPSS Statistics, Corporation, Chicago, USA).

Results

PCA on soil parameters

Three “principal components” (PCs) were obtained from the PCA performed with the six soil variables determined in four sampling dates, explaining 84.0% of the variance of the data set (Table 3, Fig. 3). Soil organic C, total N and β-glucosidase activity were the most important parameters related to PC1 (factor loadings >0.50) while soil moisture and temperature showed the highest contribution (positive and negative, respectively) to PC2 (Table 3). Although soil moisture also contributed to PC3, soil respiration was the most relevant variable related to this component (Table 3).
Table 3

Principal component solution on six soil variables (n = 72)

Parameters

PC1

PC2

PC3

Moisture (%)

−0.019

0.603

0.527

Temperature (°C)

−0.056

−0.695

0.014

Respiration (μmol m−2 s−1)

0.200

−0.382

0.802

Organic C (mg g−1)

0.556

0.018

−0.275

Total N (mg g−1)

0.610

0.080

0.034

β-glucosidase (mg ρNP kg−1 h−1)

0.525

−0.020

−0.034

Variance explained (%)

 Absolute

38.2

30.2

15.6

 Cumulative

38.2

68.4

84.0

Factor loadings ≥0.50 in absolute value are marked in bold

Fig. 3

Ordination plot by principal component analysis (PCA) of the studied experimental plots along four sampling dates, representing PC1 vs. PC2 (a) and PC3 vs. PC2 (b), according to soil data of moisture (%), temperature (°C), respiration (μmol m−2 s−1), organic C (mg g−1), total N (mg g−1) and β-glucosidase activity (mg ρNP kg−1 h−1)

PC2 clearly segregated December from June samplings, due to higher soil moisture and lower soil temperature in December than in June months (Fig. 3a, b). No clear separation was observed along PC1, although values obtained in June 2012 tended to be more positive than those obtained in June 2014, indicating higher overall values of organic C, total N and β-glucosidase activity in the first sampling date (Fig. 3a). PC3 separated December 2011 from December 2013 data, mainly as a result of higher respiration in December 2011 associated to slightly higher soil moisture (Fig. 3b). No segregation was observed in response to the treatments.

Differences between pre- and post-fire periods

Significant differences were found in most of the studied parameters between pre- and post-fire periods regardless of the treatments. At the soil level, temperature, electrical conductivity (at the two studied depths) and organic C at depth B were significantly higher in the post-fire period, while respiration and total N at depth A decreased by 9.7% and 23%, respectively, in relation to pre-fire values (Table S1). As a consequence, soil C:N ratio was 19.8% and 17.5% higher at depth A and B, respectively, after the fire. Contrasting differences were obtained for the β-glucosidase activity between the two depths studied, since the activity of this enzyme declined by 21% at depth A whereas it increased by 47% at depth B in the post-fire period compared to pre-fire values. Soil moisture and pH showed similar values before and after the fire.

As expected, there were significant differences in soil cover by litter and plants before (June 2012) and after the fire (June 2013) (Table S2). In June 2014, vegetation cover already showed similar values to those found before the fire, while litter cover was still lower (Table S2). Regarding the chemical properties of the leaf litter of A. unedo, values of δ15N and N concentration were significantly higher after the fire (1.34‰ and 30.7%, respectively, compared to June 2012 and 1.14‰ and 55.1%, respectively, compared to September 2012) (Table S2). Conversely, δ13C and C concentration values did not vary among sampling dates. As a consequence, the C:N ratio of A. unedo leaf litter in June 2014 was 31.1% lower than in September 2012 (Table S2). In the case of P. angustifolia, leaf litter production after the fire was too low to be analyzed.

In the two studied species, leaf C concentration was significantly lower after the fire (by 1.5% in A. unedo and 1.2% in P. angustifolia) (Table S1). Since N concentration of P. angustifolia leaves was a 23% higher after the fire, the leaf C:N ratio of this species was a 19.2% lower in the post-fire period. A. unedo leaves also showed a decrease in δ13C values (by 0.4‰) after the fire, while, for both species, δ15N values were 1‰ higher in the post-fire period, in accordance with the results found for the leaf litter of A. unedo.

Regarding inter-specific differences, despite P. angustifolia had higher leaf C concentration than A. unedo over the study period (F1,120 = 488.4, p < 0.001), differences in leaf N concentration between the two species varied before and after the fire (Table S1). In the pre-fire period, N concentration in P. angustifolia leaves was lower than in A. unedo (F1,48 = 7.0, p = 0.011), but, after the fire, the contrary was found (F1,48 = 46.9, p < 0.001). These differences led to a higher C:N ratio of P. angustifolia leaves in the pre-fire period (F1,48 = 19.5, p < 0.001) followed by a lower ratio after the fire (F1,48 = 31.0, p < 0.001; Table S1). Throughout the study period, δ13C values did not differ between the two species, although P. angustifolia showed lower δ15N values than A. unedo (F1,120 = 105.8, p < 0.001).

Effects of UV radiation and rainfall regime

At soil level

Regardless of the watering regime and along all the experimental period, soil moisture was significantly higher in UVA and UVAB plots than in UV0 ones (Table 4). In addition, at depth A (0–5 cm), pre-fire soils from UVA and UV0 plots showed around 5% higher values of pH than those from UVAB plots.
Table 4

Overall mean ± S.E. for all the studied parameters in the soil (depth A, 0–5 cm; depth B, 5–10 cm) and in the leaves of Arbutus unedo and Phillyrea angustifolia under the three different UV radiation conditions (UVAB, UVA and UV0) and the two rainfall regimes (natural rainfall, NR; reduced rainfall, RR)

 

UV radiation (UV)

Rainfall (R)

Interactions

UVAB

UVA

UV0

p-value

NR

RR

p-value

PRE-fire

Soil

Moisture (%)

10.918 ± 1.185 a

12.024 ± 1.389 a

8.569 ± 1.037 b

<0.001

11.619 ± 1.108

9.388 ± 0.867

0.001

Temperature (°C)

17.850 ± 0.698

17.635 ± 0.671

17.713 ± 0.664

ns

17.928 ± 0.567

17.537 ± 0.537

ns

Respiration (μmol m−2 s−1)

1.815 ± 0.095 b

2.136 ± 0.128 a

1.747 ± 0.105 b

0.038

1.835 ± 0.095

1.964 ± 0.091

ns

UV × R

Depth A

 pH1:2.5

6.008 ± 0.098 b

6.273 ± 0.057 a

6.313 ± 0.058 a

0.027

6.147 ± 0.074

6.249 ± 0.058

ns

 EC1:5 (dS m−1)

0.064 ± 0.011

0.059 ± 0.006

0.066 ± 0.008

ns

0.060 ± 0.008

0.066 ± 0.006

ns

 Organic C (mg g−1)

17.191 ± 0.702

18.683 ± 0.853

18.084 ± 1.080

ns

18.269 ± 0.634

17.703 ± 0.813

ns

 Total N (mg g−1)

1.490 ± 0.080

1.632 ± 0.085

1.564 ± 0.097

ns

1.463 ± 0.072

1.661 ± 0.067

ns

 C:N ratio

12.223 ± 1.051

11.641 ± 0.466

11.928 ± 0.765

ns

13.168 ± 0.783

10.694 ± 0.324

0.010

 β-glucosidase (mg ρNP kg−1 h−1)

124.639 ± 10.591

140.743 ± 11.514

128.745 ± 9.500

ns

124.091 ± 7.521

138.660 ± 9.338

ns

Depth B

 pH1:2.5

5.760 ± 0.062

5.922 ± 0.104

5.825 ± 0.139

ns

5.911 ± 0.062

5.760 ± 0.104

ns

 EC1:5 (dS m−1)

0.047 ± 0.006

0.050 ± 0.007

0.054 ± 0.008

ns

0.047 ± 0.005

0.054 ± 0.006

ns

 Organic C (mg g−1)

10.247 ± 0.526

10.854 ± 0.768

10.121 ± 0.453

ns

10.514 ± 0.472

10.301 ± 0.501

ns

 Total N (mg g−1)

0.840 ± 0.051

0.902 ± 0.054

0.848 ± 0.048

ns

0.881 ± 0.044

0.846 ± 0.039

ns

 C:N ratio

12.534 ± 0.657

12.233 ± 0.739

12.240 ± 0.459

ns

12.339 ± 0.605

12.333 ± 0.393

ns

 β-glucosidase (mg ρNP kg−1 h−1)

38.504 ± 5.405 b

54.119 ± 3.548 a

45.188 ± 4.065 ab

0.017

47.430 ± 4.375

44.444 ± 3.252

ns

Date × UV × R

Plant leaf

A. unedo

 C (mg g−1)

492.004 ± 2.041

488.030 ± 1.924

487.910 ± 1.861

ns

490.783 ± 1.053

487.847 ± 1.988

ns

Date × R

 N (mg g−1)

12.864 ± 0.622

12.005 ± 0.330

12.700 ± 0.550

ns

12.234 ± 0.347

12.811 ± 0.480

ns

 C:N ratio

39.214 ± 1.860

41.031 ± 1.258

39.251 ± 1.775

ns

40.652 ± 1.125

39.012 ± 1.508

ns

 δ13C (‰)

−27.306 ± 0.234

−27.025 ± 0.190

−27.524 ± 0.291

ns

−27.313 ± 0.233

−27.256 ± 0.163

ns

 δ15N (‰)

−1.713 ± 0.226

−2.315 ± 0.271

−2.230 ± 0.314

ns

−2.131 ± 0.206

−2.042 ± 0.248

ns

P. angustifolia

 C (mg g−1)

512.470 ± 1.429 a

506.282 ± 2.099 b

509.480 ± 1.783 ab

0.011

509.654 ± 1.526

509.167 ± 1.603

ns

 N (mg g−1)

11.823 ± 0.543

11.717 ± 0.460

11.755 ± 0.628

ns

11.255 ± 0.392

12.276 ± 0.453

0.020

 C:N ratio

44.406 ± 2.127

44.036 ± 1.930

44.825 ± 2.580

ns

46.351 ± 1.814

42.494 ± 1.647

0.033

 δ13C (‰)

−27.489 ± 0.126

−27.535 ± 0.387

−27.398 ± 0.269

ns

−27.233 ± 0.111

−27.715 ± 0.289

ns

 δ15N (‰)

−3.965 ± 0.383

−3.762 ± 0.396

−4.069 ± 0.352

ns

−3.665 ± 0.324

−4.199 ± 0.269

ns

POST-fire

Soil

Moisture (%)

9.823 ± 0.606 a

10.915 ± 0.902 a

7.526 ± 0.643 b

<0.001

10.609 ± 0.681

8.234 ± 0.518

<0.001

Temperature (°C)

20.364 ± 0.914

20.535 ± 0.857

20.011 ± 0.799

ns

20.429 ± 0.713

20.180 ± 0.683

ns

Respiration (μmol m−2 s−1)

1.746 ± 0.144

1.774 ± 0.110

1.628 ± 0.097

ns

1.661 ± 0.092

1.769 ± 0.099

ns

UV x R

Depth A

 pH1:2.5

6.081 ± 0.088

6.281 ± 0.077

6.318 ± 0.080

ns

6.205 ± 0.077

6.248 ± 0.063

ns

 EC1:5 (dS m−1)

0.083 ± 0.013

0.082 ± 0.005

0.086 ± 0.009

ns

0.075 ± 0.008

0.092 ± 0.007

ns

 Organic C (mg g−1)

15.640 ± 1.593

17.787 ± 1.166

17.298 ± 1.503

ns

17.103 ± 1.089

16.714 ± 1.250

ns

 Total N (mg g−1)

1.126 ± 0.135

1.246 ± 0.064

1.240 ± 0.115

ns

1.214 ± 0.085

1.194 ± 0.092

ns

 C:N ratio

14.439 ± 1.001

14.338 ± 0.675

14.120 ± 0.597

ns

14.387 ± 0.555

14.211 ± 0.689

ns

 β-glucosidase (mg ρNP kg−1 h−1)

83.390 ± 5.078

110.333 ± 10.253

116.817 ± 11.118

ns

106.023 ± 8.961

101.004 ± 7.308

ns

Depth B

 pH1:2.5

5.916 ± 0.093

5.997 ± 0.094

6.037 ± 0.097

ns

6.004 ± 0.087

5.962 ± 0.066

ns

 EC1:5 (dS m−1)

0.055 ± 0.004

0.060 ± 0.004

0.070 ± 0.008

ns

0.055 ± 0.003

0.069 ± 0.005

ns

 Organic C (mg g−1)

11.347 ± 0.685

13.458 ± 0.597

12.357 ± 0.928

ns

12.969 ± 0.599

11.805 ± 0.646

ns

Date x UV x R

 Total N (mg g−1)

0.828 ± 0.059

0.924 ± 0.063

0.858 ± 0.068

ns

0.919 ± 0.051

0.821 ± 0.049

ns

 C:N ratio

13.916 ± 0.546

15.091 ± 1.002

14.466 ± 0.346

ns

14.276 ± 0.302

14.706 ± 0.738

ns

 β-glucosidase (mg ρNP kg−1 h−1)

59.671 ± 6.508

75.263 ± 8.112

67.442 ± 6.813

ns

68.158 ± 5.554

66.759 ± 6.368

ns

Plant leaf

A. unedo

 C (mg g−1)

482.669 ± 1.639

483.505 ± 1.329

479.354 ± 1.482

ns

483.349 ± 1.395

480.336 ± 1.016

ns

 N (mg g−1)

11.952 ± 0.652

11.380 ± 0.688

11.865 ± 0.772

ns

11.401 ± 0.588

12.064 ± 0.542

ns

 C:N ratio

41.711 ± 2.258

44.333 ± 2.822

42.384 ± 2.791

ns

44.348 ± 2.266

41.271 ± 1.931

ns

 δ13C (‰)

−27.980 ± 0.211

−27.675 ± 0.201

−27.472 ± 0.279

ns

−27.884 ± 0.203

−27.534 ± 0.175

ns

 δ15N (‰)

−0.562 ± 0.310 a

−1.257 ± 0.335 b

−1.622 ± 0.245 b

0.011

−1.217 ± 0.248

−1.077 ± 0.277

ns

UV × R

P. angustifolia

 C (mg g−1)

504.430 ± 1.404

503.235 ± 0.955

501.739 ± 1.307

ns

503.830 ± 1.096

502.440 ± 0.930

ns

 N (mg g−1)

15.354 ± 0.758

14.227 ± 0.628

13.843 ± 0.814

ns

13.944 ± 0.582

15.004 ± 0.616

ns

 C:N ratio

33.802 ± 1.760

36.198 ± 1.715

37.712 ± 2.316

ns

37.231 ± 1.566

34.576 ± 1.601

ns

 δ13C (‰)

−27.401 ± 0.254

−27.153 ± 0.182

−27.587 ± 0.177

ns

−27.337 ± 0.150

−27.423 ± 0.192

ns

 δ15N (‰)

−2.541 ± 0.238

−3.172 ± 0.428

−3.101 ± 0.272

ns

−2.493 ± 0.271

−3.383 ± 0.220

0.027

Pre- and post-fire data and analyses are shown separately. Numbers in bold indicate significant differences among the levels of the factor. In the case of UV radiation, significant differences among the UV conditions are also indicated by different letters. For pre-fire and post-fire, n = 12 in each UV condition and n = 18 in each rainfall regime for all variables, except for soil moisture, temperature and respiration (n = 24 and n = 36, respectively) and for pre-fire values of soil organic C, total N and C:N ratio (n = 18 and n = 27, respectively). The significance level considered was p ≤ 0.05. Only significant two-way or three-way interactions were included in the column “interactions”. EC, electrical conductivity; ns, not significant

As it was expected, the reduction in rainfall decreased the soil moisture of RR plots throughout all the study period, being 20% and 23% lower than in NR plots before and after the fire, respectively (Table 4). Before the fire, the reduction in rainfall also decreased by 19% the soil C:N ratio at depth A, but this effect was lost after the fire.

Throughout the study period, there was an interactive effect between the two treatments on soil respiration (Table 4). Indeed, before the fire, exposure to UV-B reduced soil respiration rates (UVAB < UVA, p = 0.051) under natural rainfall, while exposure to UV-A increased soil respiration rates (UVA > UV0, p = 0.010) under reduced rainfall (Fig. 4). On the other hand, control soils (UVAB plots) always exhibited greater respiration rates under drier conditions (pre-fire: p = 0.001; post-fire: p = 0.018), whereas water supply did not significantly affect soil respiration of UVA and UV0 plots (Fig. 4).
Fig. 4

Soil respiration in plots subjected to three UV radiation conditions (UVAB, UVA and UV0) combined with two rainfall regimes (natural rainfall, NR; reduced rainfall, RR), both before and after the fire. Error bars represent the standard error of the mean (n = 12). Asterisks indicate significant differences between NR and RR plots exposed to the same UV condition, whereas different letters indicate significant differences among UV conditions within each rainfall regime. Only significant differences within the same UV or rainfall condition are highlighted. The significance level was set at p ≤ 0.05

At depth B (5–10 cm), the effects of the two treatments on β-glucosidase activity showed a significant interaction with the sampling date before the fire (Table 4), since UVA plots showed significantly higher β-glucosidase activity than UV0 and UVAB plots under reduced rainfall and in December 2011, but not in June 2012, under natural rainfall (p = 0.010) (Fig. 5a). After the fire, the effects of the two treatments on organic C at depth B also differed between sampling dates, since, under natural rainfall, soils of UVA and UV0 plots showed higher organic C values than UVAB plots in June 2014 (p = 0.047), but not in December 2013 (Fig. 5b).
Fig. 5

Soil β-glucosidase activity (a) and organic C (b) at depth B (5–10 cm) from plots subjected to three UV radiation conditions (UVAB, UVA and UV0) combined with two rainfall regimes (natural rainfall, NR; reduced rainfall, RR) along all the sampling dates both before and after the fire. Error bars represent the standard error of the mean (n = 3). Since the interaction between UV radiation, rainfall and sampling date was significant in the pre-fire period for β-glucosidase activity (p = 0.009) and in the post-fire period for organic C (p = 0.012), we analyzed UV effects within the two levels of rainfall for each of these periods and only significant differences are highlighted. Thus, different letters indicate significant differences among UV conditions within a specific sampling date and rainfall regime. The significance level was set at p ≤ 0.05

Finally, plant and litter cover did not show significant differences as a result of the treatments (data not shown).

At leaf litter level

Treatments did not affect leaf litter C and N concentrations or C:N ratio of any of the two studied species (Table 5). Regarding the isotopic composition of litter, δ13C values of P. angustifolia leaf litter were 0.8‰ higher in UVA plots than in UVAB ones in September 2012 (Table 5). On the other hand, the experimental reduction in rainfall decreased by 1.9‰ the δ15N values of P. angustifolia leaf litter in June 2012 (Table 5). For A. unedo litter, we found a significant interactive effect of the two treatments on δ15N values in September 2012 (Table 5). In this sampling date, but only in RR plots, leaf litter of this species showed 2.6 and 2.0‰ lower δ15N values in UVA and UV0 plots, respectively, than in UVAB ones (p = 0.009; Fig. 6a).
Table 5

Overall mean ± S.E. for all the studied parameters in the leaf litter of Arbutus unedo and Phillyrea angustifolia under the three different UV radiation conditions (UVAB, UVA and UV0) and the two rainfall regimes (natural rainfall, NR; reduced rainfall, RR)

 

UV radiation (UV)

Rainfall (R)

UV × R

UVAB

UVA

UV0

p-value

NR

RR

p-value

p-value

June 2012

A. unedo

 C (mg g−1)

488.756 ± 2.635

484.806 ± 2.373

481.291 ± 1.967

ns

486.173 ± 2.266

483.728 ± 1.915

ns

ns

 N (mg g−1)

6.803 ± 0.594

6.336 ± 0.629

7.610 ± 0.920

ns

6.645 ± 0.734

7.188 ± 0.421

ns

ns

 C:N ratio

74.741 ± 6.715

80.532 ± 8.269

68.240 ± 8.635

ns

79.874 ± 7.855

69.134 ± 3.968

ns

ns

 δ13C (‰)

−28.908 ± 0.243

−29.006 ± 0.194

−29.202 ± 0.153

ns

−28.965 ± 0.204

−29.112 ± 0.105

ns

ns

 δ15N (‰)

−2.747 ± 0.223

−3.022 ± 0.493

−2.950 ± 0.573

ns

−2.691 ± 0.334

−3.121 ± 0.373

ns

ns

P. angustifolia

 C (mg g−1)

507.117 ± 4.025

508.839 ± 4.635

509.320 ± 3.772

ns

510.662 ± 3.298

506.594 ± 3.037

ns

ns

 N (mg g−1)

9.462 ± 1.371

7.444 ± 0.331

6.546 ± 0.438

ns

8.265 ± 1.177

7.552 ± 0.643

ns

ns

 C:N ratio

59.873 ± 8.776

68.814 ± 3.497

79.277 ± 4.465

ns

68.055 ± 7.389

70.419 ± 5.098

ns

ns

 δ13C (‰)

−28.450 ± 0.429

−28.456 ± 0.174

−28.561 ± 0.251

ns

−28.375 ± 0.239

−28.585 ± 0.269

ns

ns

 δ15N (‰)

−3.436 ± 0.865

−3.502 ± 1.097

−3.042 ± 0.671

ns

−2.249 ± 0.627

−4.126 ± 0.536

0.032

ns

September 2012

A. unedo

 C (mg g−1)

480.609 ± 3.866

485.788 ± 2.093

481.992 ± 2.782

ns

484.803 ± 3.014

480.790 ± 1.544

ns

ns

 N (mg g−1)

6.672 ± 0.672

5.051 ± 0.487

5.764 ± 0.982

ns

5.968 ± 0.734

5.690 ± 0.506

ns

ns

 C:N ratio

75.842 ± 7.721

100.156 ± 8.607

97.324 ± 16.560

ns

92.922 ± 12.293

89.293 ± 6.892

ns

ns

 δ13C (‰)

−28.714 ± 0.386

−28.776 ± 0.175

−29.131 ± 0.237

ns

−28.613 ± 0.228

−29.134 ± 0.194

ns

ns

 δ15N (‰)

−2.080 ± 0.315

−3.232 ± 0.518

−2.817 ± 0.590

ns

−2.366 ± 0.343

−3.053 ± 0.456

ns

0.034

P. angustifolia

 C (mg g−1)

505.620 ± 4.364

504.880 ± 4.521

499.444 ± 4.780

ns

503.584 ± 2.983

503.046 ± 4.353

ns

ns

 N (mg g−1)

6.127 ± 0.380

6.207 ± 0.230

6.000 ± 0.228

ns

5.811 ± 0.149

6.411 ± 0.247

ns

ns

 C:N ratio

84.070 ± 5.072

81.912 ± 3.201

83.852 ± 3.278

ns

87.158 ± 2.474

79.398 ± 3.134

ns

ns

 δ13C (‰)

−28.334 ± 0.212 b

−27.525 ± 0.213 a

−27.804 ± 0.150 ab

0.040

−27.917 ± 0.200

−27.858 ± 0.187

ns

ns

 δ15N (‰)

−3.586 ± 0.551

−3.177 ± 0.792

−3.890 ± 0.158

ns

−3.080 ± 0.565

−4.022 ± 0.214

ns

ns

June 2014 *

A. unedo

 C (mg g−1)

480.776 ± 2.942

479.314 ± 4.184

483.099 ± 8.516

ns

483.452 ± 5.434

478.674 ± 3.283

ns

ns

 N (mg g−1)

8.078 ± 1.612

7.786 ± 1.220

11.246 ± 1.932

ns

10.087 ± 1.470

7.985 ± 1.198

ns

ns

 C:N ratio

69.968 ± 11.122

69.572 ± 10.321

48.801 ± 7.253

ns

55.749 ± 7.449

69.811 ± 8.628

ns

ns

 δ13C (‰)

−28.777 ± 0.406

−28.770 ± 0.364

−27.934 ± 0.556

ns

−28.024 ± 0.433

−28.964 ± 0.222

ns

ns

 δ15N (‰)

−0.969 ± 0.354

−1.864 ± 0.471

−1.859 ± 0.471

ns

−1.471 ± 0.354

−1.658 ± 0.392

ns

ns

Data from each sampling date are shown separately. Numbers in bold indicate significant differences among the levels of the factor. In the case of UV radiation, significant differences among the UV conditions are also indicated by different letters. For all sampling dates, n = 6 in each UV condition and n = 9 in each rainfall regime for all variables. The significance level considered was p ≤ 0.05. ns, not significant

*In June 2014, we only collected litter from A. unedo, since production of P. angustifolia litter was too low

Fig. 6

Arbutus unedo δ15N in litter (a) and leaves (b) from plots subjected to three UV radiation conditions (UVAB, UVA and UV0) combined with two rainfall regimes (natural rainfall, NR; reduced rainfall, RR) along all the sampling dates both before and after the fire. Error bars represent the standard error of the mean (n = 3). Since there was a significant interaction between the effects of the two treatments (UV radiation and rainfall) on δ15N values of litter samples collected in September 2012 (p = 0.034) and of leaves from the post-fire period (p = 0.002), we analyzed the UV effects within the two levels of rainfall for these sampling dates and only significant differences are highlighted. Thus, different letters indicate significant differences among UV conditions within a specific sampling date and rainfall regime. The significance level was set at p ≤ 0.05

At plant leaf level

UV and rainfall treatments had different effects on the leaf parameters studied depending on the species. In the case of P. angustifolia, leaves from UVA plots had a 1.2% lower C concentration than those from UVAB plots before the fire (Table 4). Also before the fire, rainfall reduction increased foliar N concentration of this species by 9.0%, reducing, as a consequence, the C:N ratio by 8.3%. After the fire, P. angustifolia leaves grown in plots under reduced rainfall had δ15N values 0.9‰ lower than those from plots receiving natural rainfall (Table 4).

In A. unedo leaves, in the pre-fire period, the effect of the rainfall treatment on the C concentration depended on the sampling date (Table 4). Indeed, in March 2012 (but not in March 2014, despite the same tendency was observed), leaves of this species had a 2% lower C concentration in plots under reduced rainfall than in those receiving natural rainfall (p = 0.009; Fig. 7). After the fire, there was a significant interactive effect between UV and rainfall treatments on the δ15N of A. unedo leaves (Table 4), since, only under drier conditions, leaves from UVA and UV0 plots showed, respectively, 1.8 and 1.3‰ lower δ15N values than control ones (Fig. 6b). Treatments did not affect leaf δ13C values of any of the two species studied.
Fig. 7

C concentration in leaves of Arbutus unedo from plots subjected to two rainfall regimes (natural rainfall, NR; reduced rainfall, RR) along all the sampling dates both before and after the fire. Error bars represent the standard error of the mean (n = 9). Since there was a significant interaction between the sampling date and the rainfall treatment before the fire, we analyzed the rainfall effects within the two sampling dates and only significant differences are highlighted. Asterisk indicates significant differences between NR and RR plots within a specific sampling date. The significance level was set at p ≤ 0.05

Discussion

Differences in soil parameters measured in late autumn and late spring

According to the results of the PCA, soil characteristics were only segregated by the season, since soil moisture and temperature were higher and lower, respectively, in December than in June (Fig. 3a). Other studies in Mediterranean shrublands have also reported seasonal patterns for these soil variables (Gispert et al. 2013). Considering only autumn data, soil respiration exhibited greater values in December 2011 (before the fire) than in December 2013 (i.e. nine months after the fire) probably due to the slightly higher soil moisture observed in autumn 2011 (Fig. 3b). Regarding spring data, soils in June 2012 (before the fire) tended to have, in general, higher values of total N, organic C and β-glucosidase activity than in June 2014 (i.e. one year and three months after the fire) (Fig. 3a), although these effects varied at the two studied depths. Higher values of these parameters before the fire (respiration and moisture in December 2011, and organic C, total N and β-glucosidase activity in June 2012) could be related to the greater litter cover, which might enhance soil water retention and nutrient input, stimulating soil microbial activity (Raich and Tufekcioglu 2000; Talmon et al. 2011). In addition, higher litter cover would be expected to diminish soil UV exposure, avoiding potential harmful effects of UV on soil microorganisms.

Differences between pre- and post-fire periods

There were significant differences before and after the experimental fire in most of the parameters studied, regardless of the UV and rainfall treatments. After the fire, soil respiration was lower, while temperature, electrical conductivity and C:N ratio values were higher, compared to pre-fire values (Table S1). Enhanced soil temperature after the fire would probably be related to the decrease in vegetation and litter cover. It is known that, in semi-arid Mediterranean areas, a reduction in shrub cover can decrease soil moisture, increase solar radiation reaching the soil and thus soil temperature (Sherman et al. 2012), and diminish nutrient inputs through decaying debris, attenuating microbial activity and, as a consequence, soil respiration (Raich and Tufekcioglu 2000; Talmon et al. 2011).

At depth A, a higher soil C:N ratio after the fire might be, at least partially explained, by the observed reduction in soil N concentration, which might be linked to enhanced N losses through volatilization, runoff or leaching (Certini 2005; Hart et al. 2005). Lower soil respiration rates and β-glucosidase activity would also be in agreement with enhanced C:N ratios at depth A, suggesting lower decomposition rates (Geisseler and Horwath 2009; Bengtsson et al. 2012) and a post-fire attenuation of biological activity at the topsoil. Post-fire reductions in β-glucosidase activity have been documented (López-Poma and Bautista 2014), being mostly attributed to enzyme denaturation and temporary soil sterilization (Certini 2005; Knicker 2007). Conversely, at depth B (5–10 cm), the increase in the C:N ratio after the fire seemed to be linked to the enhancement in soil organic C concentration (Table S1). Given that soil surface is more exposed to erosion and nutrient leaching after a fire (Certini 2005), deeper soil layers may become enriched in organic C and nutrients (López-Poma and Bautista 2014), which might also explain the increase in β-glucosidase activity found at this subsurface layer.

After the fire, we found higher δ15N values in A. unedo and P. angustifolia leaves (Table S1), as well as in A. unedo litter (Table S2), probably because of a fire-induced 15N enrichment in soil organic matter (Szpak 2014). Post-fire leaves of P. angustifolia, as well as leaf litter of A. unedo, showed higher values of N concentration, which might be explained by a concentration phenomenon due to the lower shoot:root ratio of these plants. Finally, the lower foliar C concentration observed in both species after the fire would presumably reflect a higher carbon investment into growth (Savé et al. 1993; El Omari et al. 2003; Bussotti 2008).

Effects of UV radiation

Throughout the whole study period, plots exposed to UV radiation (UVA and UVAB) showed higher soil moisture content than those not receiving this type of radiation (Table 4). This effect would be associated to UV-A exposure, since there were no significant differences between UVA and UVAB plots. The increase in soil moisture in response to UV-A is intriguing, but might, at least partially, be related to a UV-A-induced reduction of plant transpiration, since this effect was detected for P. angustifolia plants in a parallel study conducted in the same experiment, although this was only observed under reduced-rainfall conditions (Verdaguer et al., in prep). In addition, before the fire, UV-A exposure increased soil respiration rates under reduced rainfall (Fig. 4), which would be in agreement with the observed rise in β-glucosidase activity at depth B (Fig. 5a), likely as a result of stimulated soil microbial activity. Higher microbial activity might be related with the UV-A-induced enhancement of soil moisture, since soil water content has been positively correlated with microorganism activity in Mediterranean shrublands (Sardans et al. 2008a). Nevertheless, we cannot rule out a UV-A effect also on root respiration, which is largely controlled by solar irradiance through its effects on plant photosynthetic rates (Ferréa et al. 2012) and on the supply of photosynthates to roots (Högberg et al. 2001; Matteucci et al. 2015).

Contrary to the observed UV-A effects, the presence of near-ambient UV-B radiation reduced soil respiration and β-glucosidase activity (at depth B) in the pre-fire period (Table 4), suggesting a negative effect of UV-B radiation on soil biological activity. This negative effect might have been mediated by the observed pH reduction (significant at depth A) in response to UV-B exposure (Table 4), since acidity has been negatively linked to enzyme and microorganism activity, mainly through its effects on the availability of mineral nutrients (Eivazi and Tabatabai 1990; Sardans et al. 2008a). Based on previous papers (Rinnan et al. 2006, 2008), the significant pH reduction found at depth A in plots exposed to UV-B could be associated to plant chemical changes in root exudates in response to UV-B exposure. The fact that UV-B effects on soil pH, respiration and β-glucosidase activity were mostly observed before the fire, i.e. when plant cover was higher, supports the idea that these parameters might be, at least partially, influenced by plant responses to this type of radiation. It has been shown that plants receiving enhanced UV-B radiation increase root accumulation and secretion of phenolics, which can affect negatively microorganism and β-glucosidase activity in the rhizosphere (Erickson et al. 2000; Castells et al. 2004; Castaldi et al. 2009; Choudhary et al. 2013).

Unlike what has been observed for the pre-fire period, we have found only punctual UV effects on the studied soil parameters after the fire. Apart from the effect on soil moisture commented above, plots exposed to UV-B radiation showed lower values of soil organic C concentration at depth B in June 2014, but only under natural rainfall (Fig. 5b). Although not significant, the same tendency was observed for β-glucosidase activity (Fig. 5a). Similarly to the pre-fire results, these effects could be mediated by plant responses to UV-B, such as a UV-B-induced increase in root exudation of phenolic compounds (Choudhary et al. 2013), which would reduce soil microorganism activity (Castaldi et al. 2009).

The studied litter and plant parameters responded differently to the UV treatment depending on the species. Despite the UV treatment did not affect significantly δ13C values of plant leaves in any of the two studied species, in September 2012, leaf litter of P. angustifolia from UVAB plots showed lower δ13C values than those found in UVA plots (Table 5). This might reflect a UV-B-induced reduction in the integrated water use efficiency of these leaves while they were alive, which would be in accordance with what it was found in a parallel study, although differences were only significant under reduced-rainfall conditions (Verdaguer et al., in prep.).

Under reduced rainfall, δ15N values of A. unedo litter and leaves were highest in UVAB plots throughout the study period, although differences were only significant in September 2012 for the leaf litter (Fig. 6a) and in the post-fire period for the leaves (Fig. 6b). Increases in leaf δ15N have been correlated with greater biomass allocation to roots versus shoots, allowing plants to exploit more efficiently soil systems and thus increasing water and N uptake (Llorens et al. 2003). Higher leaf δ15N values could also indicate increased nitrification in the soil and, consequently, higher N losses mostly in the form of nitrates (Pardo et al. 2007; Högberg et al. 2014).

Effects of the rainfall regime

The reduction in rainfall, apart from the expected decrease in soil moisture, also decreased the pre-fire C:N ratio at the topsoil, which would be explained by the tendency of organic C to decrease and of total N to increase in these soils (Table 4). In Mediterranean ecosystems, a wide variety of precipitation effects on soil C:N ratio has been reported, although, often, this ratio increases in soils under drought due to the input of plant material with a higher proportion of structural carbon-related compounds in leaves (i.e. more sclerophyllous leaves) (Bussotti 2008; Sardans et al. 2012; Sardans and Peñuelas 2013). On the contrary, the lower soil C:N ratio we found under drier conditions might be, at least partially, related to the higher N concentration and, thus, the lower C:N ratio observed in P. angustifolia leaves (Table 4). A higher N content in these leaves might indicate a greater accumulation of leaf soluble protein, as it has been reported in wet-temperate ecosystems under moderate drought (Lu et al. 2009).

After the fire, the reduction in water availability led to a decrease in the foliar δ15N values of P. angustifolia, suggesting lower N losses at the soil level (Högberg et al. 2014; Ruiz-Navarro et al. 2016) despite we did not detect significant differences in soil total N in response to the rainfall treatment (Table 4). These results contrast with other studies performed with Mediterranean plant species that have found lower leaf N concentration (Sardans et al. 2008b), and higher (Ogaya and Peñuelas 2008) or similar (Llorens et al. 2003) δ15N values under low rainfall. Differences in the intensity of the drought treatment applied in these experiments might explain, at least partially, these contrasting results.

In A. unedo, drier conditions decreased the leaf C concentration in March 2012 (the same tendency was observed in March 2014, although it was not significant), but not at the end of summer (September 2012 and 2013) (Fig. 7), probably due to the scarce rainfall recorded during summer months (Fig. 1), which would have minimized the differences between the two irrigation levels. Considering that, in a parallel study, we found higher photosynthetic rates in this species under drier conditions (Verdaguer et al., in prep.), the observed reduction in leaf C concentration could reflect an enhanced C investment in growth, which is supported by the results of another study (Llorens et al., in prep.).

Concluding remarks

Exposure to UV-A radiation appeared to favor soil biological activity and C turn-over before the fire, since we found higher soil respiration rates and β-glucosidase activity. This might be a consequence of the observed increase in soil moisture in response to UV-A exposure. Also before the fire, the additional presence of UV-B radiation (i.e. UV-A + UV-B exposure) decreased the rate of soil respiration along with soil pH and β-glucosidase activity in relation to UVA plots, increasing C concentration in P. angustifolia leaves. This would suggest an attenuated soil microorganism activity coupled with lower rates of decomposition and C turn-over, which would lead to a slowdown of the C cycle in response to UV-B radiation. Under reduced rainfall, the presence of UV-B radiation also resulted in greater δ15N values in leaves and litter of A. unedo, suggesting higher N losses in the soil, particularly in the soil compartment from which these plants took the N, which might affect negatively N cycling in the ecosystem.

The reduction in soil moisture due to reduced rainfall was coupled with a decrease in the C:N ratio at the topsoil before the fire, likely related to the higher N concentration and the lower C concentration found in P. angustifolia and A. unedo leaves, respectively. Therefore, our results suggest increased decomposition rate and, consequently, a faster C and N cycling in response to drier conditions. In addition, the lower foliar δ15N values recorded in P. angustifolia plants grown under reduced rainfall points to lower N losses in the soil (at least in the soil compartment from which plants of this species took the N) linked to an ecosystem with a tighter N cycle.

Overall, the experimental reduction in rainfall exerted a greater effect on the studied parameters related to N cycle, while the biogeochemical cycle of C was more sensitive to UV radiation, alone or in combination with water supply. Many of the UV effects found were modulated by the rainfall regime; in particular, UV-induced changes in soil respiration and β-glucosidase activity along with UV responses in A. unedo plants were emphasized by rainfall reduction. Unlike A. unedo, interactive effects of UV radiation and rainfall were not found for P. angustifolia plants. Species-specific responses to changes in UV fluxes and rainfall may induce modifications in the competitive ability of these species, ultimately altering their distribution in the next decades. Taking into account the fundamental role of the vegetation on biogeochemical cycles, these changes might affect the evolution and dynamics of Mediterranean shrublands in the future.

Apart from this, the fact that most UV and water effects were observed only before the fire would indicate a homogenizing influence of this perturbation. Thus, given the predicted increase in fire occurrence over the coming years, this factor might play a more important role modulating C and N cycles of Mediterranean shrublands than the projected changes in UV fluxes and rainfall amount.

Notes

Acknowledgements

This research was supported by the Spanish Government (CGL2010-22283 and CGL2014-55976-R) and the University of Girona (ASING2011/3 and MPCUdG2016). We are grateful to the Gavarres Consortium for allowing us to perform the experiment in Can Vilallonga. We also thank Jordi Compte, Meritxell Bernal and Miquel Jover for their help with the field experiment, and Dr. Alan Jones for his comments to improve the manuscript.

Supplementary material

11104_2017_3485_MOESM1_ESM.pdf (136 kb)
ESM 1 (PDF 135 kb)
11104_2017_3485_MOESM2_ESM.pdf (126 kb)
ESM 2 (PDF 125 kb)

References

  1. Acácio V, Holmgren M, Rego F et al (2009) Are drought and wildfires turning Mediterranean cork oak forests into persistent shrublands? Agrofor Syst 76:389–400.  https://doi.org/10.1007/s10457-008-9165-y CrossRefGoogle Scholar
  2. Agati G, Tattini M (2010) Multiple functional roles of flavonoids in photoprotection. New Phytol 186:786–793.  https://doi.org/10.1111/j.1469-8137.2010.03269.x CrossRefPubMedGoogle Scholar
  3. Arévalo JR, de Nascimento L, Fernández-Lugo S et al (2011) Grazing effects on species composition in different vegetation types (La Palma, Canary Islands). Acta Oecol 37:230–238.  https://doi.org/10.1016/j.actao.2011.02.006 CrossRefGoogle Scholar
  4. Austin AT, Ballaré CL (2010) Dual role of lignin in plant litter decomposition in terrestrial ecosystems. Proc Natl Acad Sci 107:4618–4622.  https://doi.org/10.1073/pnas.0909396107 CrossRefPubMedPubMedCentralGoogle Scholar
  5. Austin AT, Méndez MS, Ballaré CL (2016) Photodegradation alleviates the lignin bottleneck for carbon turnover in terrestrial ecosystems. Proc Natl Acad Sci 113:4392–4397.  https://doi.org/10.1073/pnas.1516157113 CrossRefPubMedPubMedCentralGoogle Scholar
  6. Baker NR, Allison SD (2015) Ultraviolet photodegradation facilitates microbial litter decomposition in a Mediterranean climate. Ecology 96:1994–2003.  https://doi.org/10.1890/14-1482.1 CrossRefPubMedGoogle Scholar
  7. Bassman JH (2004) Ecosystem consequences of enhanced solar ultraviolet radiation: secondary plant metabolites as mediators of multiple trophic interactions in terrestrial plant communities. Photochem Photobiol 79:382–398CrossRefPubMedGoogle Scholar
  8. Bengtsson J, Janion C, Chown SL, Leinaas HP (2012) Litter decomposition in fynbos vegetation, South Africa. Soil Biol Biochem 47:100–105.  https://doi.org/10.1016/j.soilbio.2011.11.023 CrossRefGoogle Scholar
  9. Brandt LA, King JY, Milchunas DG (2007) Effects of ultraviolet radiation on litter decomposition depend on precipitation and litter chemistry in a shortgrass steppe ecosystem. Glob Chang Biol 13:2193–2205.  https://doi.org/10.1111/j.1365-2486.2007.01428.x CrossRefGoogle Scholar
  10. Bussotti F (2008) Functional leaf traits, plant communities and acclimation processes in relation to oxidative stress in trees: a critical overview. Glob Chang Biol 14:2727–2739.  https://doi.org/10.1111/j.1365-2486.2008.01677.x Google Scholar
  11. Caldwell MM (1971) Solar UV irradiation and the growth and development of higher plants. In: Geise AC (ed) Photophysiology, IV. Academic Press, New York, pp 131–177CrossRefGoogle Scholar
  12. Caldwell BA (2005) Enzyme activities as a component of soil biodiversity: a review. Pedobiologia (Jena) 49:637–644.  https://doi.org/10.1016/j.pedobi.2005.06.003 CrossRefGoogle Scholar
  13. Caldwell MM, Bornman JF, Ballaré CL et al (2007) Terrestrial ecosystems, increased solar ultraviolet radiation, and interactions with other climate change factors. Photochem Photobiol Sci 6:252–266.  https://doi.org/10.1039/b700019g CrossRefPubMedGoogle Scholar
  14. Caon L, Vallejo VR, Ritsema CJ, Geissen V (2014) Effects of wildfire on soil nutrients in Mediterranean ecosystems. Earth Sci Rev 139:47–58.  https://doi.org/10.1016/j.earscirev.2014.09.001 CrossRefGoogle Scholar
  15. Castaldi S, Carfora A, Fiorentino A et al (2009) Inhibition of net nitrification activity in a Mediterranean woodland: possible role of chemicals produced by Arbutus unedo. Plant Soil 315:273–283.  https://doi.org/10.1007/s11104-008-9750-x
  16. Castells E, Peñuelas J, Valentine DW (2004) Are phenolic compounds released from the Mediterranean shrub Cistus albidus responsible for changes in N cycling in siliceous and calcareous soils? New Phytol 162:187–195.  https://doi.org/10.1111/j.1469-8137.2004.01021.x
  17. Certini G (2005) Effects of fire on properties of forest soils: a review. Oecologia 143:1–10.  https://doi.org/10.1007/s00442-004-1788-8 CrossRefPubMedGoogle Scholar
  18. Choudhary KK, Pandey D, Agrawal SB (2013) Deterioration of rhizospheric soil health due to elevated ultraviolet-B. Arch Agron Soil Sci 59:1419–1437.  https://doi.org/10.1080/03650340.2012.713473 CrossRefGoogle Scholar
  19. Day TA, Zhang ET, Ruhland CT (2007) Exposure to solar UV-B radiation accelerates mass and lignin loss of Larrea tridentata litter in the Sonoran Desert. Plant Ecol 193:185–194.  https://doi.org/10.1007/s11258-006-9257-6
  20. Díaz-Delgado R, Lloret F, Pons X, Terradas J (2002) Satellite evidence of decreasing resilience in Mediterranean plant communities after recurrent wildfires. Ecology 83:2293–2303.  https://doi.org/10.1890/0012-9658(2002)083[2293:SEODRI]2.0.CO;2 CrossRefGoogle Scholar
  21. Dirks I, Navon Y, Kanas D et al (2010) Atmospheric water vapor as driver of litter decomposition in Mediterranean shrubland and grassland during rainless seasons. Glob Chang Biol 16:2799–2812.  https://doi.org/10.1111/j.1365-2486.2010.02172.x CrossRefGoogle Scholar
  22. Eivazi F, Tabatabai MA (1990) Factors affecting glucosidase and galactosidase activities in soils. Soil Biol Biochem 22:891–897.  https://doi.org/10.1016/0038-0717(90)90126-K CrossRefGoogle Scholar
  23. El Omari B, Fleck I, Aranda X et al (2003) Total antioxidant activity in Quercus ilex resprouts after fire. Plant Physiol Biochem 41:41–47Google Scholar
  24. Erickson AJ, Ramsewak RS, Smucker AJ, Nair MG (2000) Nitrification inhibitors from the roots of Leucaena leucocephala. J Agric Food Chem 48:6174–6177.  https://doi.org/10.1021/jf991382z
  25. Erickson DJ III, Sulzberger B, Zepp RG, Austin AT (2015) Effects of stratospheric ozone depletion, solar UV radiation and climate change on biogeochemical cycling: interactions and feedbacks. Photochem Photobiol Sci 14:127–148.  https://doi.org/10.1039/c4pp90036g CrossRefPubMedGoogle Scholar
  26. Ferréa C, Zenone T, Comolli R, Seufert G (2012) Estimating heterotrophic and autotrophic soil respiration in a semi-natural forest of Lombardy, Italy. Pedobiologia (Jena) 55:285–294.  https://doi.org/10.1016/j.pedobi.2012.05.001 CrossRefGoogle Scholar
  27. Flint SD, Caldwell MM (2003) Field testing of UV biological spectral weighting functions for higher plants. Physiol Plant 117:145–153.  https://doi.org/10.1034/j.1399-3054.2003.1170118.x CrossRefGoogle Scholar
  28. Foereid B, Bellarby J, Meier-Augenstein W, Kemp H (2010) Does light exposure make plant litter more degradable? Plant Soil 333:275–285.  https://doi.org/10.1007/s11104-010-0342-1 CrossRefGoogle Scholar
  29. Formánek P, Rejšek K, Vranová V (2014) Effect of elevated CO2, O3, and UV radiation on soils. Sci World J 2014:1–8.  https://doi.org/10.1155/2014/730149 CrossRefGoogle Scholar
  30. Forster J (1995) Soil physical analysis. In: Alef K, Nannipieri P (eds) Methods in Soil Microbiology and Biochemistry. Academic Press, London, pp 79–87 105–121Google Scholar
  31. Gallo ME, Sinsabaugh RL, Cabaniss SE (2006) The role of ultraviolet radiation in litter decomposition in arid ecosystems. Appl Soil Ecol 34:82–91.  https://doi.org/10.1016/j.apsoil.2005.12.006 CrossRefGoogle Scholar
  32. Geisseler D, Horwath WR (2009) Relationship between carbon and nitrogen availability and extracellular enzyme activities in soil. Pedobiologia (Jena) 53:87–98.  https://doi.org/10.1016/j.pedobi.2009.06.002 CrossRefGoogle Scholar
  33. Gispert M, Emran M, Pardini G et al (2013) The impact of land management and abandonment on soil enzymatic activity, glomalin content and aggregate stability. Geoderma 202–203:51–61.  https://doi.org/10.1016/j.geoderma.2013.03.012 CrossRefGoogle Scholar
  34. Hart SC, DeLuca TH, Newman GS et al (2005) Post-fire vegetative dynamics as drivers of microbial community structure and function in forest soils. For Ecol Manag 220:166–184.  https://doi.org/10.1016/j.foreco.2005.08.012 CrossRefGoogle Scholar
  35. Hayano K, Tubaki K (1985) Origin and properties of β-glucosidase activity of tomato-field soil. Soil Biol Biochem 17:553–557.  https://doi.org/10.1016/0038-0717(85)90024-0 CrossRefGoogle Scholar
  36. Henry HAL, Brizgys K, Field CB (2008) Litter decomposition in a California annual grassland: interactions between photodegradation and litter layer thickness. Ecosystems 11:545–554.  https://doi.org/10.1007/s10021-008-9141-4 CrossRefGoogle Scholar
  37. Hofmann RW, Campbell BD, Bloor SJ et al (2003) Responses to UV-B radiation in Trifolium repens L. – physiological links to plant productivity and water availability. Plant Cell Environ 26:603–612.  https://doi.org/10.1046/j.1365-3040.2003.00996.x
  38. Högberg P, Nordgren A, Buchmann N et al (2001) Large-scale forest girdling shows that current photosynthesis drives soil respiration. Nature 411:789–792.  https://doi.org/10.1038/35081058 CrossRefPubMedGoogle Scholar
  39. Högberg P, Johannisson C, Högberg MN (2014) Is the high 15N natural abundance of trees in N-loaded forests caused by an internal ecosystem N isotope redistribution or a change in the ecosystem N isotope mass balance? Biogeochemistry 117:351–358.  https://doi.org/10.1007/s10533-013-9873-x CrossRefGoogle Scholar
  40. Hughes KA, Lawley B, Newsham KK (2003) Solar UV-B radiation inhibits the growth of Antarctic terrestrial fungi. Appl Environ Microbiol 69:1488–1491.  https://doi.org/10.1128/AEM.69.3.1488–1491.2003 CrossRefPubMedPubMedCentralGoogle Scholar
  41. Hurlbert SH (1984) Pseudoreplication and the design of ecological field experiments. Ecol Monogr 54:187–211.  https://doi.org/10.2307/1942661 CrossRefGoogle Scholar
  42. Incerti G, Bonanomi G, Giannino F et al (2011) Litter decomposition in Mediterranean ecosystems: modelling the controlling role of climatic conditions and litter quality. Appl Soil Ecol 49:148–157.  https://doi.org/10.1016/j.apsoil.2011.06.004 CrossRefGoogle Scholar
  43. IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, and New YorkGoogle Scholar
  44. Julkunen-Tiitto R, Häggman H, Aphalo PJ et al (2005) Growth and defense in deciduous trees and shrubs under UV-B. Environ Pollut 137:404–414.  https://doi.org/10.1016/j.envpol.2005.01.050 CrossRefPubMedGoogle Scholar
  45. Knicker H (2007) How does fire affect the nature and stability of soil organic nitrogen and carbon? A review. Biogeochemistry 85:91–118.  https://doi.org/10.1007/s10533-007-9104-4 CrossRefGoogle Scholar
  46. Li F-R, Peng S-L, Chen B-M, Hou Y-P (2010) A meta-analysis of the responses of woody and herbaceous plants to elevated ultraviolet-B radiation. Acta Oecol 36:1–9.  https://doi.org/10.1016/j.actao.2009.09.002 CrossRefGoogle Scholar
  47. Llorens L, Peñuelas J, Estiarte M (2003) Ecophysiological responses of two Mediterranean shrubs, Erica multiflora and Globularia alypum, to experimentally drier and warmer conditions. Physiol Plant 119:231–243.  https://doi.org/10.1034/j.1399-3054.2003.00174.x
  48. Lloret F, Calvo E, Pons X, Díaz-Delgado R (2002) Wildfires and landscape patterns in the eastern Iberian peninsula. Landsc Ecol 17:745–759.  https://doi.org/10.1023/A:1022966930861 CrossRefGoogle Scholar
  49. López-Poma R, Bautista S (2014) Plant regeneration functional groups modulate the response to fire of soil enzyme activities in a Mediterranean shrubland. Soil Biol Biochem 79:5–13.  https://doi.org/10.1016/j.soilbio.2014.08.016 CrossRefGoogle Scholar
  50. Lu Y, Duan B, Zhang X et al (2009) Intraspecific variation in drought response of Populus cathayana grown under ambient and enhanced UV-B radiation. Ann For Sci 66:613–624.  https://doi.org/10.1051/forest/2009049
  51. Masciandaro G, Ceccanti B, García C (1994) Anaerobic digestion of straw and piggery wastewaters II. Optimization of the process. Agrochimica 38:195–203Google Scholar
  52. Matteucci M, Gruening C, Ballarin IG et al (2015) Components, drivers and temporal dynamics of ecosystem respiration in a Mediterranean pine forest. Soil Biol Biochem 88:224–235.  https://doi.org/10.1016/j.soilbio.2015.05.017 CrossRefGoogle Scholar
  53. Nannipieri P, Kandeler E, Ruggiero P (2002) Enzyme activities and microbiological and biochemical processes in soil. In: Burns RG, Dick RP (eds) Enzymes in the environment: activity, ecology and applications. Marcel Dekker, New York, pp 1–33Google Scholar
  54. Nenadis N, Llorens L, Koufogianni A et al (2015) Interactive effects of UV radiation and reduced precipitation on the seasonal leaf phenolic content/composition and the antioxidant activity of naturally growing Arbutus unedo plants. J Photochem Photobiol B Biol 153:435–444.  https://doi.org/10.1016/j.jphotobiol.2015.10.016
  55. Ogaya R, Peñuelas J (2008) Changes in leaf δ13C and δ15N for three Mediterranean tree species in relation to soil water availability. Acta Oecol 34:331–338.  https://doi.org/10.1016/j.actao.2008.06.005 CrossRefGoogle Scholar
  56. Pancotto VA, Sala OE, Robson TM et al (2005) Direct and indirect effects of solar ultraviolet-B radiation on long-term decomposition. Glob Chang Biol 11:1982–1989.  https://doi.org/10.1111/j.1365-2486.2005.01027.x Google Scholar
  57. Pardo LH, Hemond HF, Montoya JP, Pett-Ridge J (2007) Natural abundance 15N in soil and litter across a nitrate-output gradient in New Hampshire. For Ecol Manag 251:217–230.  https://doi.org/10.1016/j.foreco.2007.06.047 CrossRefGoogle Scholar
  58. Paula S, Pausas JG (2006) Leaf traits and resprouting ability in the Mediterranean basin. Funct Ecol 20:941–947.  https://doi.org/10.1111/j.1365-2435.2006.01185.x CrossRefGoogle Scholar
  59. Pausas JG, Bradstock RA, Keith DA et al (2004) Plant functional traits in relation to fire in crown-fire ecosystems. Ecology 85:1085–1100CrossRefGoogle Scholar
  60. Raich JW, Tufekcioglu A (2000) Vegetation and soil respiration: correlations and controls. Biogeochemistry 48:71–90CrossRefGoogle Scholar
  61. Ren J, Dai W, Xuan Z et al (2007) The effect of drought and enhanced UV-B radiation on the growth and physiological traits of two contrasting poplar species. For Ecol Manag 239:112–119.  https://doi.org/10.1016/j.foreco.2006.11.014 CrossRefGoogle Scholar
  62. Rey A, Pegoraro E, Tedeschi V et al (2002) Annual variation in soil respiration and its components in a coppice oak forest in Central Italy. Glob Chang Biol 8:851–866.  https://doi.org/10.1046/j.1365-2486.2002.00521.x CrossRefGoogle Scholar
  63. Riera P, Peñuelas J, Farreras V, Estiarte M (2007) Valuation of climate-change effects on Mediterranean shrublands. Ecol Appl 17:91–100.  https://doi.org/10.1890/1051-0761(2007)017[0091:VOCEOM]2.0.CO;2 CrossRefPubMedGoogle Scholar
  64. Rinnan R, Gehrke C, Michelsen A (2006) Two mire species respond differently to enhanced ultraviolet-B radiation: effects on biomass allocation and root exudation. New Phytol 169:809–818.  https://doi.org/10.1111/j.1469-8137.2005.01577.x CrossRefPubMedGoogle Scholar
  65. Rinnan R, Nerg AM, Ahtoniemi P et al (2008) Plant-mediated effects of elevated ultraviolet-B radiation on peat microbial communities of a subarctic mire. Glob Chang Biol 14:925–937.  https://doi.org/10.1111/j.1365-2486.2008.01544.x CrossRefGoogle Scholar
  66. Ruiz-Navarro A, Barberá GG, Albaladejo J, Querejeta JI (2016) Plant δ15N reflects the high landscape-scale heterogeneity of soil fertility and vegetation productivity in a Mediterranean semiarid ecosystem. New Phytol 212:1030–1043.  https://doi.org/10.1111/nph.14091 CrossRefPubMedGoogle Scholar
  67. Sanchez-Lorenzo A, Enriquez-Alonso A, Calbó J et al (2017) Fewer clouds in the Mediterranean: consistency of observations and climate simulations. Sci Rep 7:41475.  https://doi.org/10.1038/srep41475 CrossRefPubMedPubMedCentralGoogle Scholar
  68. Sardans J, Peñuelas J (2013) Plant-soil interactions in Mediterranean forest and shrublands: impacts of climatic change. Plant Soil 365:1–33.  https://doi.org/10.1007/s11104-013-1591-6 CrossRefGoogle Scholar
  69. Sardans J, Peñuelas J, Estiarte M (2008a) Changes in soil enzymes related to C and N cycle and in soil C and N content under prolonged warming and drought in a Mediterranean shrubland. Appl Soil Ecol 39:223–235.  https://doi.org/10.1016/j.apsoil.2007.12.011 CrossRefGoogle Scholar
  70. Sardans J, Peñuelas J, Estiarte M, Prieto P (2008b) Warming and drought alter C and N concentration, allocation and accumulation in a Mediterranean shrubland. Glob Chang Biol 14:2304–2316.  https://doi.org/10.1111/j.1365-2486.2008.01656.x CrossRefGoogle Scholar
  71. Sardans J, Rivas-Ubach A, Peñuelas J (2012) The C:N:P stoichiometry of organisms and ecosystems in a changing world: a review and perspectives. Perspect Plant Ecol Evol Syst 14:33–47.  https://doi.org/10.1016/j.ppees.2011.08.002 CrossRefGoogle Scholar
  72. Savé R, Alegre L, Pery M, Terradas J (1993) Ecophysiology of after-fire resprouts of Arbutus unedo L. Orsis 8:107–119Google Scholar
  73. Searles PS, Flint SD, Caldwell MM (2001) A meta-analysis of plant field studies simulating stratospheric ozone depletion. Oecologia 127:1–10.  https://doi.org/10.1007/s004420000592 CrossRefPubMedGoogle Scholar
  74. Sherman C, Sternberg M, Steinberger Y (2012) Effects of climate change on soil respiration and carbon processing in Mediterranean and semi-arid regions: an experimental approach. Eur J Soil Biol 52:48–58.  https://doi.org/10.1016/j.ejsobi.2012.06.001 CrossRefGoogle Scholar
  75. Soil Survey Staff (2010) Keys to soil taxonomy. Eleventh, USDA-NRCS, Washington, DCGoogle Scholar
  76. Szpak P (2014) Complexities of nitrogen isotope biogeochemistry in plant-soil systems: implications for the study of ancient agricultural and animal management practices. Front Plant Sci 5:1–19.  https://doi.org/10.3389/fpls.2014.00288 CrossRefGoogle Scholar
  77. Tabatabai MA, Bremner JM (1969) Use of p-nitrophenyl phosphate for assay of soil phosphatase activity. Soil Biol Biochem 1:301–307.  https://doi.org/10.1016/0038-0717(69)90012-1 CrossRefGoogle Scholar
  78. Talmon Y, Sternberg M, Grünzweig JM (2011) Impact of rainfall manipulations and biotic controls on soil respiration in Mediterranean and desert ecosystems along an aridity gradient. Glob Chang Biol 17:1108–1118.  https://doi.org/10.1111/j.1365-2486.2010.02285.x CrossRefGoogle Scholar
  79. Tárrega R, Luis-Calabuig E, Valbuena L (2001) Eleven years of recovery dynamic after experimental burning and cutting in two Cistus communities. Acta Oecol 22:277–283.  https://doi.org/10.1016/S1146-609X(01)01125-0
  80. Verdaguer D, Ojeda F (2002) Root starch storage and allocation patterns in seeder and resprouter seedlings of two cape Erica (Ericaceae) species. Am J Bot 89:1189–1196.  https://doi.org/10.3732/ajb.89.8.1189
  81. Yue M, Li Y, Wang X (1998) Effects of enhanced ultraviolet-B radiation on plant nutrients and decomposition of spring wheat under field conditions. Environ Exp Bot 40:187–196.  https://doi.org/10.1016/S0098-8472(98)00036-7 CrossRefGoogle Scholar
  82. Zepp RG, Callaghan TV, Erickson DJ (1998) Effects of enhanced solar ultraviolet radiation on biogeochemical cycles. J Photochem Photobiol B Biol 46:69–82.  https://doi.org/10.1016/S1011-1344(98)00186-9 CrossRefGoogle Scholar
  83. Zepp RG, Erickson DJ III, Paul ND, Sulzberger B (2007) Interactive effects of solar UV radiation and climate change on biogeochemical cycling. Photochem Photobiol Sci 6:286–300.  https://doi.org/10.1039/b700021a CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2017

Authors and Affiliations

  1. 1.Department of Environmental Sciences, Faculty of SciencesUniversity of Girona, Campus MontiliviGironaSpain
  2. 2.Department of Chemical Engineering, Agriculture and Food Technology, Polytechnic SchoolUniversity of Girona, Campus MontiliviGironaSpain
  3. 3.Faculty of Sciences and TechnologyUniversity of Vic – Central University of CataloniaVicSpain
  4. 4.Department of Physics, Polytechnic SchoolUniversity of Girona, Campus MontiliviGironaSpain
  5. 5.Institute of Ecosystem Studies, CNRPisaItaly

Personalised recommendations