Effects of drought on soil phosphorus availability and fluxes in a burned Mediterranean shrubland

Geoderma 191 (2012) 61–69

Contents lists available at SciVerse ScienceDirect

Geoderma journal homepage: www.elsevier.com/locate/geoderma

Effects of drought on soil phosphorus availability and fluxes in a burned Mediterranean shrubland M. Belén Hinojosa a,⁎, Antonio Parra b, David A. Ramírez b, José A. Carreira c, Roberto García-Ruiz c, José M. Moreno b a b c

Centro de Investigación del Fuego, Fundación General de Medio Ambiente de Castilla-La Mancha, Universidad de Castilla-La Mancha, Campus Fábrica de Armas, E-45071 Toledo, Spain Dpto. Ciencias Ambientales, Universidad de Castilla-La Mancha, Campus Fábrica de Armas, E-45071 Toledo, Spain Dpto. Biología Animal, Biología Vegetal y Ecología, Universidad de Jaén, Campus Las Lagunillas, E-23071 Jaén, Spain

a r t i c l e

i n f o

Article history: Received 2 June 2011 Received in revised form 6 November 2011 Accepted 10 January 2012 Available online 2 February 2012 Keywords: P fractions P transformation rates Fire Drought Climate change

a b s t r a c t Little research has been done to study the combined fire-drought impacts on the structure and functioning of Mediterranean fire-prone ecosystems. In this study we have evaluated post-fire patterns of soil phosphorus (P) availability and transformation rates in a Mediterranean shrubland subjected to different drought treatments. Precipitation was manipulated in various treatments to evaluate the combined effects of fire and moisture in the soil. The following treatments were included: long-term historical average precipitation, or historical control (HC); moderate drought (MD, 25% reduction from HC); severe drought (SD, 45% reduction from HC); and environmental control without rainfall manipulation (EC). In late summer, these plots were burned in order to evaluate the joint effects of drought and fire. In order to compare burned and non-burned scenarios, unburned plots were maintained without rainfall manipulation. Post-fire soil samples, collected in the spring, were analyzed for P fractions, P transformation rates (net solubilization and immobilization, gross mineralization) and phosphatase activity. Mid- to high-intensity fire caused a short-term increase of rapid-to-mid turnover inorganic P pools in the soil, as well as a decline of rapid turnover organic P pools (including microbial P) and acid phosphatase activity. Fire also induced an overall rise in the flux among P pools, including gross P mineralization, net microbial immobilization and net geochemical solubilization, with the latter being the most affected. Moreover, under drier than average conditions, this ‘mineralizing’ effect of fire was partially (rapid turn over inorganic P, total P flux among pools and net solubilisation) or completely (net immobilization) offset. Thus, our results indicate that a drier environment will limit the extent of post-fire peak in soil P availability. This effect may act synergistically with increased water stress to further inhibit vegetation recovery after fire, possibility which is particularly relevant for fire-prone Mediterranean shrublands where P limitation is widespread. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Water deficit and fire disturbance regimes are the two main environmental factors controlling the structure and function of Mediterranean-type ecosystems (Di Castri and Mooney, 1973). Water constraint is the prevalent limiting factor affecting primary production, species richness and soil genesis in Mediterranean-type regions (Yaalon, 1997). Fire regimes play key evolutionary and disturbance-recovery roles in shaping plant communities and landscapes in these regions (Rundell, 1998a). Fire also has a strong influence on the biogeochemical cycles of several key elements in Mediterranean ecosystems (DeBano et al, 1998; Moreno and Oechel, ⁎ Corresponding author. Dpto. Ciencias Ambientales, Universidad de Castilla-La Mancha, Campus Fábrica de Armas, E-45071 Toledo, Spain. Tel.: + 34 925268800; fax: + 34 925268840. E-mail address: [email protected] (M.B. Hinojosa). 0016-7061/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.geoderma.2012.01.015

1994; Rundell, 1983). Fire has long been recognized as a major, although poorly conservative, mineralizing agent which strongly impacts nutrient availability in Mediterranean soils (St. John and Rundell, 1976; Stock and Lewis, 1986). In turn, nutrient supply is often the key factor affecting plant species, functional types, and the structure, dynamics and distribution of plant communities in these regions (Carreira and Niell, 1992; Henkin et al., 1998; Specht, 1973). In this sense, phosphorus (P) limitation is especially widespread (Cramer, 2010; Henkin et al., 2006; Mayor and Rodà, 1994), since total P in Mediterranean soils is commonly limited, and even more so the proportion of readily available and labile P forms (Carreira et al., 1997; Sardans et al., 2005). Both soil water and fire disturbance regimes are very sensitive to drivers of climate change (sensu IPCC, 2007). Mediterranean-type ecosystems may be particularly vulnerable to climate change drivers (Moreno et al., 2010; Sala et al., 2000; Schröter et al., 2005). Climate models project reductions in soil water content and increase in air

62

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

temperature for the Mediterranean Basin (Christensen et al., 2007; Schröter et al., 2005). The way in which rainfall pattern changes will alter soil nutrient availability has been well explored in temperate and boreal biomes (e.g., Allison and Treseder, 2008; Emmett et al., 2004; Johnson et al., 2008), and are also being addressed in the Mediterranean context. Studies generally agree that, in Mediterranean systems, under drier conditions, soil labile inorganic P (Pi) and phosphatase activities decline, accompanied by an increase in soil organic P forms that are not directly available to plants (García-Fayos and Bochet, 2009; Sardans and Peñuelas, 2004, 2005, 2007, 2010; Sardans et al., 2006). There is a relatively large body of research analyzing short and long term fire effects on nitrogen (N) and P availability in Mediterranean soils (Carreira et al., 1994, 1997; Debano and Conrad, 1978; Marion et al., 1991; Rundell, 1983). The role of fire on nutrient cycles is exerted through direct (e.g. volatilization, ash deposition, chemical transformation of charred litter and plant materials) and indirect (e.g. post-fire erosion and changed conditions for soil microbial activity) effects. In relation to fire, the effects on P cycle have received less attention than the effects on C or N cycle. In contrast to N, loss of P from plant biomass through volatilization during fire is low, and the resulting ash provides a high P inputs to the soil in readily available inorganic forms (Arianoutsou and Margaris, 1981; Carreira and Niell, 1995). Heat during a fire can also affect the topsoil by converting relatively non-labile forms of P, such as organic P (Po) (e.g., phosphate monoand di-esters) to labile forms of inorganic P (Pi) that are soluble or weakly surface-adsorbed, such as PO43 – (Cade-Menun et al., 2000; Galang et al., 2010; Giovannini et al., 1990). As a result, there is frequently a short-term, post-fire peak in Pi which quickly declines (Certini, 2005). While some studies have shown no change in P availability after burning (e.g., Binkley et al., 1992; Boyer and Miller, 1994), these apparent discrepancies can be attributed to differences in intensity and timing of fire, in pre-fire soil P levels, and/or in the selected P availability index. Other short-term effects, commonly found after fire, are the decrease of microbial biomass and phosphatase activity of surface soils (Boerner and Brinkman, 2003; Eivazi and Bayan, 1996; Saá et al., 1998). However, microbial activity tends to recover shortly thereafter due to increased substrate availability and altered soil temperature and moisture conditions. Recovery of microbial P and phosphatase activity has been reported following the initial post-fire decline (Arévalo et al., 1994; Serrasolsas and Khanna, 1995). According to the findings reported above, P supply in Mediterranean soils may depend on a trade-off between the direct effects of drought conditions which could depress mineralization, and an indirect impact of fire disturbance and its role as a mineralizing agent. However, no previous studies have attempted to quantitatively distinguish between the joint effects of both drought and fire on soil nutrient cycling and availability. In this study, we have assessed how the occurrence of a fire in a Mediterranean shrubland affects the short-term pattern of change in soil P cycling and availability, depending on the particular drought conditions. A better understanding of the role of fire in Mediterranean ecosystems under scenarios of drought is relevant to calibrate mechanistic models of succession and management of burned areas. We hypothesized that increasing drought in post-fire environments will limit the extent of the short-term peaks in soil P availability following fires, primarily because of the effect of drier conditions on: i) decreasing biological and biochemical mineralization, and ii) increasing chemical precipitation of soluble inorganic P forms due to the higher ionic strength of the soil solution in reduced moisture soils. To test this hypothesis, we studied the effects of fire on the patterns of P fractions, P transformation rates (net solubilization and immobilization, gross mineralization) and phosphatase activity in soils from a Mediterranean shrubland under various drought treatments.

2. Material and methods 2.1. Study area, experimental design and soil sampling The study was conducted in a Cistus-Erica Mediterranean shrubland of central Spain located in the Coto Nacional de Quintos de Mora (Lat. 39° 25′ N; Long 04° 04′ W), in Montes de Toledo mountains. The study site (900–925 m altitude, 20% slope and a northwest aspect) has a typical Mediterranean climate with a mean annual temperature of 14.9 °C and mean annual precipitation of 622 mm (only 7% occurs during the summer) (Parra et al., 2012). Soil is a Dystric Cambisol (FAO soil classification) and the parent rock consists mainly in quartzite. A rainfall manipulative experiment was initiated in April 2009. Rain manipulation treatments were defined based on long-term (1948–2006) precipitation records from a nearby meteorological station (“Los Cortijos” meteorological station; 39°19′ N, 4°04′ W; AEMET, Ministerio de Medio Ambiente, Rural y Marino, Spain). Firstly, the long-term average precipitation for each two-week period during the year was calculated, excluding extreme values of the data series (99th percentile) for each of the time period. Secondly, a biweekly rainfall schedule for each experimental treatment was designed by modifying the historical long-term rainfall pattern as follows (Fig. 1): i) environmental control (EC) without manipulation and no changes in the natural rainfall pattern (540 mm fell during the hydrological year of experimentation); ii) historical control (HC) simulating long- term historical average precipitation (600 mm y − 1 and drought during July and August); iii) moderate drought (MD), 25% decrease in rainfall relative to HC treatment (450 mm y − 1 and drought from May to September; and iv) severe drought (SD), 45% decrease in rainfall relative to HC treatment (325 mm y − 1 with drought from April to October). The simulated rainfall patterns were consistent with projections for the Mediterranean region (Christensen et al., 2007), i.e., lengthened summer drought and rainfall concentrated in winter. The target rainfall for each treatment was reached by excluding rainfall using automatic rainout shelters or watering plots by sprinkler irrigation at the end of each biweekly period; depending if natural rainfall had been higher or lower, respectively, than the average of the long-term rainfall assigned for the treatment on this period. A more complete description of the mechanical system used to manipulate rainfall in this experiment can be read at Parra et al., 2012). This approach was chosen to maximize the natural rainfall input, and minimize rainout shelter effects on the microenvironment of the plots. The four different treatments were applied to replicate 6 × 6 m plots (n = 4) (Fig. 1). At the end of the 2009 dry season (late September) these plots were burned (hereafter denoted by adding “+” to the corresponding rainfall treatment reference) in order to assess the joint effects of drought and fire. Experimental fire was homogeneous across all plots; significant differences in fire intensity were not recorded among rainfall treatments. The mean residence time above 100 °C was 13.76 min, and the mean maximum temperature was 710 °C (Parra et al., 2012). In order to compare burned and non-burned treatments, an additional set of four plots without rainfall manipulation was left unburned (EC-), resulting in a total of five replicate treatments: EC−, EC+, HC+, MD+ and SD+. Detailed information about the experimental design, rainfall manipulations and fire is provided in Parra et al. (2012). Soil samples were collected in May 2010, eight months after burning. We expected that this mid-spring sampling date would provide the greatest soil moisture differences among drought treatments, together with optimum temperature for soil microbial activity. Four composite samples, each one consisting of four soil cores, were collected from the top 5 cm of mineral soil in each experimental plot. Soil samples were transported in an isothermic bag (4 °C) to the

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

MD+

HC+

EC+

EC-

63

SD+ EC- Environmental control, unburned

EC-

MD+

HC+

EC+

SD+

EC+ Environmental control, burned HC+ Historical control, burned

EC+

SD+

HC+

MD+

EC+

EC-

HC+

SD+

MD+

70 m

EC-

MD+ Moderate drought (25% reduction HC), burned SD+ Severe drought (45% reduction HC), burned

100 m Fig. 1. Design layout of the rainfall manipulation and fire experiment at Quintos de Mora (Toledo, Spain).

laboratory, where they were sieved (b2 mm) before further analysis. The collected soil was determined to be sandy loam (74% sand, 20% silt and 6% clay) with 3.7% carbonate content (Laboratorio Agroalimentario Regional de Albacete). 2.2. General physico-chemical soil properties Soil water content was estimated gravimetrically. Soil pH and conductivity were determined in a slurry (1:2.5; soil:H2O) according to McLean (1982). Particle size distribution was determined using a densimeter according to the Bouyoucos method (Gee and Bauder, 1986). Total nitrogen was analyzed according to the Kjeldahl method. Organic matter content was estimated by weight loss after ignition (Nelson and Sommers, 1982) and total C was estimated indirectly following Walkley (1947).

process allows the calculation of inorganic P concentrations in the different fractions. Additionally, aliquots of the bicarbonate and NaOH extracts were subjected to a sulfuric acid-potassium persulfate digestion (D'Elia et al., 1977). The concentration of organic P in these two fractions was estimated by subtracting inorganic P from the total P, determined in the corresponding digested extract. Microbial P (micr-P) was estimated as the difference in bicarbonate-extractable Pi between chloroform-fumigated and nonfumigated pairs of replicated soil samples, using a conversion factor of KP = 0.4 (Brookes et al., 1985; Hedley and Stewart, 1982). Finally, total soil P was determined on separate soil subsamples (0.2 g DW) after nitric-perchloric acid (5:3 v:v) digestion (Sommer and Nelson, 1972). Residual P was calculated as the difference between total P and the sum of all P fractions obtained in the sequential extraction scheme. All data are expressed on dry weight basis.

2.3. Soil phosphorus fractionation 2.4. Phosphorus transformation rates and phosphatase activity Field-moist soil samples were analyzed for soil P fractions using a variation of the P fractionation scheme of Hedley et al. (1982), to sequentially extract inorganic and organic P held in a variety of bioavailable and poorly-available pools. Briefly, 3.5 g of each soil sample were placed in a 50-ml centrifuge tube containing 30 ml of distilled water and a 5 × 2.5 cm 2 section of an anion-exchange membrane (type 204-U-386 Ionics Inc., Watertown, MA, USA), equivalent to 1 g of dry resin (Cooperband and Logan, 1994), previously saturated with Cl −. (LiCl). After being shaken for 16 h, the membrane was removed and treated with 0.5 M HCl to elute the PO43– ions sorbed onto it. This procedure provides the membrane-extractable inorganic P (Pi) fraction (hereafter, memb-Pi), which represents the most readily bioavailable P and includes soil solution Pi as well as the most labile adsorbed Pi (Sibbesen, 1978). The soil remaining in the tube was recovered by centrifugation (3500 rpm for 10 min.) and extracted with 0.5 M NaHCO3 at pH 8.5 to obtain the rest of the labile P pool. This pool is considered a good indicator of short- to mid-term plant available-P and includes both inorganic (bic-Pi) and organic (bic-Po) forms of P, reversibly sorbed onto soil colloids and surfaces (Bowman and Cole, 1978). Following additional centrifugation steps, the moderately to poorly labile pool was sequentially extracted from the remaining soil with 0.5 M NaOH followed by 1.0 M HCl. The NaOH-extractable fraction includes non-occluded, inorganic P (NaOH–Pi) and organic P (NaOH–Po) forms bound to Fe and Al minerals (Ryden et al., 1977). The HCl-extractable fraction includes both inorganic and organic, Ca-bound forms of P (Williams et al., 1971), although in the current investigation, only the inorganic fraction was quantified (HCl-Pi). All the supernatants obtained during the sequential extraction were filtered, neutralized, and then directly analyzed for phosphate by the ascorbic acid-molybdenum blue method (John, 1970). This

Soil P transformation rates were evaluated using a modification of the radiation–autoclaving–incubation procedure of Zou et al. (1992). Three subsets were obtained by sub-sampling each field-moist composite soil sample. One subset was used as a control and the other two were sterilized, using HgCl2-addition at a rate of 2500 mg kg − 1 of dry soil (Wolf and Skipper, 1994), instead of γ-radiation (Viñegla et al., 2006). To apply the HgCl2-treatment, the corresponding samples were evenly mixed with the appropriate amount of HgCl2. All samples were then incubated for 12 d, adding water when needed to maintain the original soil moisture. Soil respiration rates were measured during incubation, using the aqueous alkali trapping method (Zibilske, 1994) to check for the degree of microbial sterilization achieved by the HgCl2 treatment, and we found that soil respiration was reduced by 92% after this treatment. One of the two HgCl2-treated subsets was then autoclaved (120 °C for 30 min) three times to denature soil enzymes. As a result, three independent treatments were created for analysis: (i) control, (ii) HgCl2-addition, and (iii) HgCl2addition plus autoclaving. Immediately after treatment, all samples of the three soil subsets were examined for membrane-extractable P during a 48 h incubation period. This was achieved by shaking 2.5 g of soil with 30 ml deionized water and a 12.5 cm 2 section of an anion-exchange membrane (type 204-U-386 Ionics Inc., Watertown, MA, USA) in 50 ml centrifuge tubes in a reciprocal shaker (200 strokes min − 1). To ensure complete soil sterilization in the HgCl2 and the HgCl2 plus autoclaving treatments, and to maintain sterile conditions during incubation with membranes, additional HgCl2 (2500 mg kg − 1 of dry soil) was added, along with the deionized water (30 ml), to the corresponding centrifuge tubes. Following incubation, membranes were extracted with 30 ml 0.5 M HCl on a reciprocal shaker for 1 h. Phosphate

64

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

Table 1 General physico-chemical soil properties in burned (EC +) and unburned (EC−) plots under environmental conditions, and in the burned plots subjected to different rainfall treatments (HC+: long-term historical average precipitation; MD+: moderate drought with 25% reduction from HC; SD+: severe drought with 45% reduction from HC). Data represent the mean (standard deviation). Different uppercase or lowercase letters indicate a significant difference (p b 0.05, n = 4) between burned and unburned treatments and among levels of rainfall, respectively. Treatments

Soil moisture %

pH

Conductivity dS m− 1

Total N %

Organic C %

C/N

OM %

Comparison factor: unburned vs burned plots under environmental conditions EC25.1 (1.92) A 5.57 (0.02) A 0.08 (0.02) A EC + 21.2 (2.83) A 6.66 (0.05) B 0.11 (0.02) A

0.21 (0.03) A 0.20 (0.02) A

5.66 (0.73) A 4.06 (0.43) A

27.6 (3.71) A 21.4 (0.39) B

9.77 (1.25) A 7.01 (0.74) A

Comparison factor: levels of rainfall manipulation in burned plots HC + 21.51 (3.64) a 6.58 (0.08) a MD + 12.75 (3.79) b 6.61 (0.09) a SD + 6.27 (2.39) c 6.63 (0.00) a

0.23 (0.02) b 0.19 (0.02) a 0.19 (0.02) a

5.19 (0.50) a 4.29 (0.46) a 4.12 (0.58) a

21.8 (0.15) a 22.1 (0.32) a 22.1 (0.31) a

8.95 (0.86) a 7.41 (0.79) a 7.11 (0.83) a

0.11 (0.02) a 0.16 (0.02) b 0.15 (0.01) b

concentration in the solutions was determined by the ascorbic acid– molybdenum blue method of John (1970). By extracting P with membranes from the control soils, the net balance between solubilization of inorganic P, mineralization of organic P and immobilization of solution P can be determined. Membrane-extractable P from the HgCl2-treated samples is an estimate of the sum of solubilized P and mineralized P (HgCl2-sterilization avoids microbial immobilization of solution P). Phosphorus extracted from the HgCl2-treated plus autoclaved soils is derived only from the solubilization of inorganic P (autoclaving of HgCl2treated soils additionally avoids P mineralization by phosphatase activity), and gives an estimate of the net P solubilization rate. Therefore, the difference in the concentrations of membrane-extractable P between the HgCl2-addition and the HgCl2-addition plus autoclaved treatments is an estimate of the gross P mineralization rate. The difference between the HgCl2-addition and control treatments is an estimate of the P immobilization rate. To check for the direct effect of HgCl2-addition and autoclaving on membrane-extractable P, three subsamples of 2.5 g moist soil from each of the control, HgCl2-addition, and HgCl2-addition plus autoclaving treatments were extracted with anion exchange membranes as above, except that extraction time was only 1 h. If differences were found in membrane-P concentrations before and after HgCl2-addition and HgCl2-addition plus autoclaving, correction factors were applied in the calculation of P transformation rates, to allow for direct treatment effects (Zou et al., 1992). Effects of fire and rainfall manipulation treatments on phosphatase activity were also tested. Both acid and alkaline phosphomonoesterase (PMEase) activities were measured on soil samples following the method developed by Tabatabai (1982), using p-nitrophenyl phosphate (pNPP) as enzyme substrate. 2.5. Statistical analysis A one-way ANOVA was performed to test the effects of fire (comparison of burned and unburned soils), and to test the effect of rainfall manipulation treatments on burned soils. ANOVA requirements of normality (Kolmogorov–Smirnov test) and homogeneity of variance (Bartlett–Box F test) were checked at α = 0.01. For variables that did not meet such requirements, log (x + 1) transformation was sufficient to achieve normality and homocedasticity. Statistical differences were determined at a significance level of α = 0.05, n = 4. Differences among levels of a treatment were tested using the Fisher Least Significant Difference (LSD) test. 3. Results 3.1. General soil properties At sampling time (mid spring), the soil water content was not significantly different between burned and unburned plots under

environmental control conditions (no rainfall manipulation) (Table 1). As expected, the drought treatments applied to burned plots strongly decreased soil water content (40% and 70% reductions in the MD+ and SD+ treatments, respectively, compared to the long-term historical average precipitation treatment, HC+). Soil electrical conductivity showed no effect due to fire, but it increased significantly with the drought applied to burned soils (Table 1). Conversely, soil pH was not significantly affected by rainfall reduction, although it was higher in burned plots than in unburned plots, under environmental control conditions. Fire did not significantly change soil organic matter and total N. The decrease in total C was not statistically significant but provoked a significant change in C/N ratio. Rainfall treatments had no significant effect on total C or organic matter content of burned soils. However, soil total N was significantly higher in HC+ than in the SD + treatment. 3.2. Soil phosphorus fractionation Neither the burning nor the rainfall manipulation treatments showed significant effects on soil total phosphorus concentrations (Table 2). However, substantial differences among treatments were found for the various analyzed P fractions (Table 2). Mean concentrations of the most biologically available inorganic P fraction (membPi) ranged from 5 to 15 μg P g − 1, and accounted for just 3.0% and 1.6% of the total P in burned and unburned soils, respectively. Fire significantly raised the concentration of memb-Pi on plots under normal environmental conditions. A moderate rainfall reduction (MD+ treatment) also increased memb-Pi in burned plots, although this effect was not observed under more severe drought conditions (SD + treatment). Labile inorganic P levels (bic-Pi) in burned soils were twice those in unburned soils, accounting for 3.4% and 1.7% of total P, respectively. Rainfall reduction on burned soils caused a significant decrease in the bic-Pi fraction (from 19.4 ± 3.9 μg P g − 1 in the HC+ treatment to 9.0 ± 1.1 μg P g − 1 in the SD + treatment). Labile organic P (bic-Po), which accounted for 5% (on average) of total P, was not significantly affected by burning alone or the joint effect of burning and altered rainfall. The microbial P fraction (micr-P) was also nonresponsive to changes in rainfall on burned soils; however, it was reduced by a factor of 1.5 in burned plots under environmental control conditions in comparison to values of unburned soils. This fraction represented 14.6% of total soil P in EC- plots and 9.7% in EC+ plots. Considerably higher concentrations of inorganic, but especially organic, P were extracted from all soils when using a strong base (sodium hydroxide) instead of a weak base (bicarbonate). The concentration of organic P forms always exceeded the concentration of inorganic P in the NaOH-extractable P pool. Excluding residual-P, the NaOH-Pi plus NaOH-Po fractions made the greatest contribution (33% to 41%) to total P. The NaOH-Pi concentration increased significantly after fire, but it decreased on burned soil as a consequence of severe drought. The NaOH-Po fraction was significantly reduced in both the burn and drought treatments. Effects on the Ca-bound

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

65

Table 2 Phosphorus fractions (μg P g− 1 DW) in the upper (0–5 cm) mineral soil layer of study plots. Burned (EC +) and unburned (EC −) plots under environmental conditions, and burned plot with different rainfall (HC+: long-term historical average precipitation; MD+: moderate drought; SD+: severe drought. Data represent the mean (standard deviation). Different uppercase letters within each column indicate significant differences between burned and unburned soils. Different lowercase letters within each column denote significant differences among rainfall treatments in burned soils (p b 0.05, n = 4). Treatments

Phosphorus fractions

Total P

Rapid turnover Pi memb-Pi

Rapid to mid turnover Po bic-Pi

micr-P

Mid to slow turnover P

bic-Po

NaOH-Po

Comparison factor: unburned vs burned plots under environmental conditions EC − 5.3 (1.5) A 5.8 (1.8) A 49.0 (8.7) B 19.6 (2.9) A 116.6 (1.7) A ∑pool: 11.1 (1.9) A ∑pool: 182.3 (6.2) B EC + 10.1 (1.0) B 11.2 (3.1) B 31.8 (7.0) A 13.4 (4.2) A 95.6 (9.2) B ∑pool: 21.3 (2.0) B ∑pool: 141.7 (3.0) A Comparison factor: levels of rainfall manipulation in burned plots HC + 10.1 (0.9) a 19.4 (3.9) c 30.8 (8.4) a 17.5 (4.0) a ∑pool: 28.3 (3.0) a ∑pool: 147.8 (11.7) R1 + 15.3 (2.9) b 14.2 (2.6) b 23.4 (10.0) a 15.7 (2.0) a ∑pool: 30.5 (2.2) a ∑pool: 138.3 (11.7) R2 + 11.1 (2.9) ab 9.0 (1.1) a 22.3 (8.3) a 15.8 (2.1) a ∑pool: 20.0 (3.7) b ∑pool: 125.0 (12.5)

inorganic P fraction (HCl-Pi) were qualitatively similar, though greater, than those for the NaOH-Pi fraction. For the soil HCl-Pi fraction, burning caused a substantial increase in concentration, from 1.9 ± 0.2 in the unburned treatment to 6.6 ± 2.1 μg g − 1 in burned plots under environmental control conditions. In contrast, rainfall reduction represented by the shift from the HC+ to the SD+ treatments involved a 50% drop in the concentration of soil HCl-Pi fraction. Neither fire nor drought treatments affected the occluded, nonlabile and recalcitrant P fractions (residual-P), which typically constitute approximately 1/3 of total P. In summary, fire essentially doubled the pool of rapid turnover Pi (from 11.1 μg P g − 1 to 21.3 μg P g − 1 in the unburned and burned soils, respectively) (Table 2). The pool of rapid turnover Po, on the other hand, decreased 28% as a consequence of fire. There were no significant differences in the mid and slow turnover P pools between burned and unburned soils. Differential rainfall treatments had no effects on the rapid, mid or slow turnover Po. However, in the severe

107.9 (7.6) b a 99.2 (11.7) ab a 86.9 (7.8) a a

NaOH-Pi

HCl-Pi

residual-P

16.9 (0.8) A

1.9 (0.2) A 121.4 (12.6) A ∑pool: 140.2 (12.6) A 21.8 (2.5) B 6.6 (2.1) B 134.6 (14.1) A ∑pool: 163.0 (12.6) A

34.2 (5.6) b

10.2 (3.2) b 146.0 (38.4) a ∑pool: 194.6 (39.3) a 35.9 (3.3) b 8.7 (1.6) b 126.5 (13.0) a ∑pool: 171.1 (13.6) a 25.0 (4.0) a 5.0 (0.5) a 160.2 (28.5) a ∑pool: 190.2 (24.5) a

333.6 (11.1) A 326.0 (10.2) A

346.4 (26.1) a 339.9 (8.7) a 335.3 (12.9) a

drought treatment, the soil pool of rapid turnover Pi decreased significantly (Table 2). 3.3. Soil phosphorus transformation rates Total P flux through the soil solution was significantly higher in burned (4.6 ± 0.1 μg P g − 1 2 d − 1) than in unburned soil (0.8 ± 0.1 μg P g − 1 2 d − 1) in plots under environmental control conditions (Fig. 2). Drought superimposed on burned soils caused a reduction in the total P flux, from a mean of 6.7 ± 0.6 μg P g − 1 2 d − 1, in the plots subjected to historical mean rainfall amounts, to 3.1 ± 0.4 μg P g − 1 2 d − 1 when this rainfall was reduced by 45%. The increasing effect of fire on total P flux was due mainly to a five-fold increase in net chemical solubilization rates; gross mineralization rates and net microbial immobilization rates were unaffected. These results are consistent with the fact that burning had no significant effect on the net change in the size of the soil organic P pool during incubation,

Total P flux

EC-

Pi

EC+

Pi

HC+

Pi

+0.33 (0.06) A

+1.78 (0.50) B

+3.58 (0.50) a

+2.70 (0.37) a

MD+

Pi

SD+

Pi

+1.39 (0.41) b

Net P solubilisation rate

Psol +0.59 A Psol +1.74 B

Psol +2.60 a Psol +3.12 a Psol +2.90 a

+0.38 (0.14) A +0.12 (0.25) A’ +1.37 (1.08) A +1.42 (0.94) A’

+1.09 (0.34) a +2.07 (0.35) a’

Porg

Porg +0,04 A

4.58 (0.11) B*

Porg +0.98 a

6.73 (0.62) a*

Porg

5.20 (0.43) a*

+1.46 (0.40) a +1.04 (0.43) ab’

-0.42 ab

+1.61( 0.31) a +0.11 (0.63) b’

0.84 (0.06) A*

-0.25 A

Porg -1.50 b

3.11 (0.40) b*

Gross P mineralization rate

Net P immobilization rate Fig. 2. Net solubilization, gross mineralization, and immobilization rates of P in burned soils under different rainfall treatments and in unburned soils. Circles represent P pools; inner numbers denote the net change in the pool size during incubation (μg P g− 1 soil). Processes are represented by arrows and the corresponding P transformation rate (μg P g− 1 2 d− 1). Uppercase letters for the same process (italicized for solubilization, regular type for mineralization, with an apostrophe for immobilization) indicate significant differences among burn treatments; lowercase letters reflect significant differences among rainfall treatments (p b 0.05, n = 4).

66

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

A B

a

A b

b

A

a b

b

Fig. 3. Acid and alkaline phosphatase activities in soils from unburned and burned plots, and in burned plots subjected to different rainfall treatments (for treatment references see Fig. 1). Different uppercase or lowercase letters indicate a significant difference (p b 0.05, n = 4) between burned and unburned treatments and among levels of rainfall, respectively.

despite a significant, three-fold net increase in the soluble P pool (Fig. 2). Regarding the effects of differential rainfall treatments applied to burned soils, the observed decrease in total P flux upon soil incubation relied on significant and similar reductions in both net chemical solubilization and microbial immobilization rates. Thus, the size of the soluble P pool was unchanged upon soil incubation. The net change in the organic P pool significantly shifted from positive in HC+ soils to negative in MD+ and SD + soils, which is explained by increasing (though not significant) gross mineralization in response to elevated drought intensity. Potential phosphomonoesterase (PME) activity was between 200 and 400 μg pNP g − 1 dw soil h − 1 (Fig. 3). No significant correlation was found between PME activity and net P mineralization rates (data not shown). Burning of plots under environmental control conditions, while having no effect on soil alkaline PME, showed a significant reduction in soil acid PME activity, relative to unburned plots. In addition, the activity of both enzymes in burned soils was significantly reduced as a consequence of drought, with activity showing a significant positive correlation with soil water content (r = 0.798 for acid PME and r = 0.772 for alkaline PME, p b 0.01). 4. Discussion Our study site is representative of the type of nutrient-poor Mediterranean heathlands and shrublands that demonstrate widespread P limitation, due to highly weathered soils developed on old Palaeozoic materials in the core of tectonic plates (Specht and Moll, 1983). The study site, located in the center of the Hercynian Iberian Belt, is characterized by nutrient-poor soils and reduced P availability. Mean total P (333.6 μg P g − 1) falls within the mid-range for Mediterranean soils worldwide (Rundell, 1998b). However, the most readily available P fraction (memb-Pi) represented just 1.6% of the total soil P pool in the environmental control, whereas occluded and recalcitrant forms (residual-P) accounted for more than a 1/3 of total P (Table 2). This proportion rose to more than 75% of the total P if the mid to slow turnover fractions (NaOH, HCl and residual) are pooled together. The very low measured concentration of primary, apatitelike, Ca-P minerals (see HCl-Pi in EC- plots, Table 2), together with the high proportion of P bound to secondary Fe and Al minerals (NaOH-fractions), are indicative of highly weathered P-deficient soils (Cross and Schlesinger, 1995; Walker and Syers, 1976). Moreover, the importance of the microbial P fraction (15% of total P in EC- soils) and a Po/Pi ratio as high as 16.4 in the rapid turnover P pool, suggest that phosphate availability in the tested soils rely to a large degree on P from the organic subcycle. Such features are common in soils and ecological successional stages of fire-prone

shrublands where symptoms of P-limitation have been reported (Carreira et al., 1997; Vitousek and Farrington, 1997). Thus, our study site represents a context in which an effect of drier conditions, even if only marginal, on phosphorus availability will be of particular significance for vegetation recovery after fire, since P shortage might mean an additional constraint to that of water stress. By releasing P previously held in senescent vegetation, and by fostering the build-up of rapid turn-over soil P, natural fire-regimes act as mineralizing agents in P-poor Mediterranean ecosystems (St. John and Rundel, 1976; Stock and Lewis, 1986). Indeed, we noted a short-term doubling of memb-Pi and bic-Pi concentrations after burning plots under environmental control conditions (Table 2). Post-fire peaks in soil P availability may result from: i) total and partial pyromineralization of Po compounds in plant tissues, litter and the O horizon, and subsequent deposition as ash and charred material onto the A horizon (Galang et al., 2010); ii) heating of the A horizon that results in partial oxidation of soil Po (Giardina et al., 2000); and iii) enhanced P mineralization after fire due to increased substrate availability (easily mineralizable Po derived from the former processes (Turrion et al., 2010)), and/or to improved process conditions in the soil. Data from our study indicate that direct pyromineralization and heating effects, as well as chemical speciation of the released phosphate, are the main factors explaining the post-fire peak in P availability. First, all the individual inorganic P fractions (including NaOH-Pi and HCl-Pi) significantly increased after fire exposure (Table 2). Fire also induced a soil pH shift from slightly acidic (5.5) to near neutral (6.6) (Table 1). At pH around 6.5, soils commonly show minimum phosphate fixing capacity (McDowell and Condron, 2001). Second, the organic and microbial fractions, as well as phosphatase activities, remained unchanged (bic-Po, alkaline phosphatase) or diminished (NaOH-Po, micr-P, acid phosphatase) (Table 2, Fig. 3). Finally, geochemical mechanisms (net solubilization) explain a greater proportion of the effect of fire on total P flux and the size of the soluble P pool upon soil incubation, than do biologically-mediated mechanisms (mineralization/immobilization balance) (Fig. 2). In summary, our results support the significant role of fire as a mineralizing agent and the occurrence of a post-fire peak in soil P availability. Even though loss of P in runoff and erosion were not measured in our study, as they were minimized in our experimental setup, they should be consider also as potential factors affecting total P in soil after a fire. The effects of reduced water availability on P cycling and availability have recently been assessed for Mediterranean-type climates. Drought treatments applied to an evergreen forest in NW Spain led to the build up of soil organic P that was not directly available to plants, and a subsequent decline in standing biomass P (Sardans and Peñuelas, 2004). In the same long-term experiment, soil enzyme

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

activity decreased (Sardans and Peñuelas, 2005, 2010), and the inorganic P (Pi) to organic P (Po) ratio in the soil solution declined (Sardans and Peñuelas, 2007). Similar responses to increased drought have been reported in an Erica-dominated shrubland (Sardans et al., 2006). García-Fayos and Bochet (2009) found that labile Pi decreased under dry and warm conditions in semi-arid Mediterranean shrublands. Thus, generally the literature appears to be consistent in its demonstration of decreasing mineralization under water-limited conditions in Mediterranean ecosystems. The results of our study, which explored how fire and soil moisture interactions affect the short-term pattern of soil P cycling and availability, distinguished two main effect groups (Fig. 4): i) fire effects reduced by drought and ii) fire effects enhanced by drought.

1800

A)

1600

*

1400 1200 *

1000 800

*

*

600 400

*

* *

*

200 0 * * se

-200 a hat

osp

h id p

Ac

ion

b

lu t so

Ne

at iliz

ss

e min

at iliz

b

r

Gro

ion

ion

at aliz

t

Ne

o imm

% of change with respect to the unburned treatment

% of change with respect to the unburned treatment

i) Fire effects reduced by drought The pool of rapid turnover Pi (primarily the bic-Pi fraction), which doubled in response to fire on non-manipulated rainfall plots, was significantly reduced under severe drought conditions in comparison to burned soil with historical rainfall (HC). However, this pool was still significantly higher in burned soil under severe drought than in unburned soils. A similar tendency was observed for PiNaOH (a fraction of mid-turnover Pi). Thus, drought partially offset the increase of low- and mid-turnover Pi pools induced by fire. Similar results have been reported elsewhere on unburned soils (García-Fayos and Bochet, 2009; Johnson et al., 2008). While fire effects are dominated by the geochemical controls on P availability, the net balance of soluble P pools under different soil moisture conditions may be defined by both, geochemical and biologically-mediated processes. Therefore, in the drought plus fire treatments, the observed post-incubation decrease in total P flux was due to both significant reductions in net chemical solubilization and net microbial immobilization rates. The reduction in net chemical solubilization could be due to increased ionic strength of the soil solution induced by drought (thus, higher ion concentrations due to less dilution, which favour precipitation of phosphate minerals). This hypothesis is supported by a significant increase on soil electrical conductivity of burned soils due to drought, which suppose to be strongly related to ionic strength of soil solution (Alva et al., 1991; Jallah and Smyth, 1995). On the other hand, the reduction in net microbial immobilization may be in response to a reduction of microbial biomass as a consequence of low soil moisture. This possibility is supported by our study data which show a decrease (though non-significant) in microbial P under drought conditions. However, the reduction of net P immobilization under drought conditions may be affected also by the differential response of various groups of soil microorganisms under water stress (Killham, 1994), in relation to i) their tolerance to

67

water stress, ii) P flux rates associated with osmoregulatory strategy, and/or iii) in the extent to which P flux changes with cell death as a consequence of dehydration. ii) Fire effects enhanced by drought Fire caused a moderate drop of the rapid- to mid-turnover organic P (including microbial P) and phosphatase activity in the soil. All these fire effects were magnified under drier conditions. Our study showed that the moderate decrease of rapid- to mid-turnover Po that occurred after fire was exacerbated by drought. In addition, phosphatase enzyme activity, previously decreased as a consequence of fire, was also significantly reduced as a consequence of drought. Thus, our results agree with previous investigations on unburned soil showing a clear decrease of potential phosphatase activity under drier climatic conditions (Li and Sarah, 2003; Sardans and Peñuelas, 2005). In this sense, phosphatase enzyme activity has shown to be once more a sensitive and earlier indicator of ecosystem P limitation (Aon et al., 2001; Arévalo et al., 1994). 5. Conclusions The assessment of drought impacts on phosphorus biogeochemistry in Mediterranean systems should take into account post-fire environments, since the interactive effects of these two factors may enhance, mask or ameliorate any particular singular effect. Increasing water stress had the effect of limiting the extent of the peak in rapid turnover inorganic P pools caused by fire. Drought also constrained the fire-induced rise of the overall flux between soil P pools. In contrast, drier environmental conditions enhanced the moderate drop of the rapid-to-mid turnover of organic P (including microbial P) and phosphatase activity in the soil caused by fire. Overall, our results indicate that a drier environment will limit the extent of the post-fire peak in soil P availability. In P-deficient ecosystems, this effect may act synergistically with increased water stress to further inhibit vegetation recovery after fire. However longer-term studies are needed to determine if these effects persist over time. Acknowledgments This research was sponsored by the Spanish “Ministerio de Ciencia y Tecnología” (project SECCIA, CGL2006-06914) and European Commission (project FUME, FP7-243888). M. B. Hinojosa was supported by Fundación General de Medio Ambiente de Castilla la Mancha through a contract partially financed by the European Social Fund and A. Parra thanks to a FPI grant funded by Ministerio de Ciencia e Innovación. We sincerely thank the “Quintos de Mora” staff, in particular J. M. Sebastian and C. Rodríguez, for facilitating this experiment; and D.G. de Política Forestal de Castilla-La Mancha for allowing and 800 700

B)

HC+ MD+ SD+

*

600 *

500 400 300 200

* *

*

100

*

0 -100 Pir

* * * r Po

Ps

al P

x

flu

tot

Fig. 4. Changes induced by rainfall manipulation treatments on the short-term effects of fire in process rates (A) and P pools (B) of a Mediterranean shrubland soil. Data are expressed as the relative change (%) with respect to unburned plots without rainfall manipulation. Pir: rapid turnover inorganic P pool; Por: rapid turnover organic P pool; Ps: slow turnover P pool. See Section 2.4 and Fig. 2 for details regarding “total P flux” and the displayed P transformation rates. Asterisks indicate significant differences (p b 0.05, n = 4) for modulation exerted by rainfall treatments on the effect of fire.

68

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69

helping us to carry out the experimental burning. We also thank Laboratorio Agroalimentario Regional de Albacete (Dpto. de Medios de Producción) the analysis of some general physico-chemical soil properties, in particular to Ángel M. Salido Castellanos and Juan F. Fernández Gómez. References Allison, S.D., Treseder, K.K., 2008. Warming and drying suppress microbial activity and carbon cycling in boreal forest soils. Global Change Biology 14, 2898–2909. Alva, A.K., Sumner, M.E., Miller, W.P., 1991. Relationship between ionic strength and electrical conductivity for soil solutions. Soil Science 152, 239–242. Aon, M.A., Cabello, M.N., Sarena, D.E., Colaneri, A.C., Franco, M.G., Burgos, J.L., Cortassa, S., 2001. Spatio-temporal patterns of soil microbial and enzymatic activities in an agricultural soil. Applied Soil Ecology 18, 239–254. Arévalo, J.R., Carreira, J.A., Niell, F.X., 1994. Kinetic parameters of phosphatase activity along a xeric dolomitic soil chronosequence related to wildfire. Geomicrobiology Journal 11, 235–245. Arianoutsou, M., Margaris, N.S., 1981. Fire-induced nutrient losses in a phryganic (east Mediterranean) ecosystem. International Journal of Biometeorology 25, 341–347. Binkley, D., Richter, D., David, M.B., Caldwell, B., 1992. Soil chemistry in a loblolly pine forest with interval burning. Ecological Applications 2, 157–164. Boerner, R.E.J., Brinkman, J.A., 2003. Fire frequency and soil enzyme activity in southern Ohio oak–hickory forests. Applied Soil Ecology 23, 137–146. Bowman, R.A., Cole, C.V., 1978. Transformations of organic phosphorus substances in soil as evaluated by sodium bicarbonate extraction. Soil Science 125, 49–54. Boyer, W.D., Miller, J.H., 1994. Effects of burning and brush treatments on nutrient and soil physical properties in young longleaf pine stands. Forest Ecology and Management 70, 311–318. Brookes, P.C., Kragt, J.F., Powlson, D.S., Jenkinson, D.S., 1985. Chloroform fumigation and the release of soil nitrogen: the effects of fumigation time and temperature. Soil Biology and Biochemistry 17, 831–835. Cade-Menun, B.J., Berch, S.M., Preston, C.M., Lavkulich, L.M., 2000. Phosphorus forms and related soil chemistry of Podzolic soils on northern Vancouver Island. 1. A comparison of two forest types. Canadian Journal of Forest Research 30, 1714–1725. Carreira, J.A., Niell, F.X., 1992. Plant nutrient changes in a semi-arid Mediterranean shrubland after fire. Journal of Vegetation Science 3, 457–466. Carreira, J.A., Niell, F.X., 1995. Mobilization of nutrients by fire in a semiarid grosescrubland ecosystem of southern Spain. Arid Soil Research and Rehabilitation 9, 73–89. Carreira, J.A., Niell, F.X., Lajtha, K., 1994. Soil nitrogen availability and nitrification in Mediterranean shrublands of varying fire history and successional stage. Biogeochemistry 26, 189–209. Carreira, J.A., Lajtha, K., Niell, F.X., 1997. Phosphorus transformations along a soil/ vegetation series of fire prone, dolomitic, semi-arid shrublands of southern Spain. Biogeochemistry 39, 87–120. Certini, G., 2005. Effects of fire on properties of forest soils: a review. Oecologia 143, 1–10. Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., Jones, R., Kolli, R.K., Kwon, W.-T., Laprise, R., Magaña Rueda, V., Mearns, L., Menéndez, C.G., Räisänen, J., Rinke, A., Sarr, A., Whetton, P., 2007. Regional climate projections. In: Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller, H.L. (Eds.), Climate change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, pp. 848–940. Cooperband, L.R., Logan, T.J., 1994. Measuring in situ changes in labile soil P with anion exchange resin-impregnated membranes. Soil Science Society of America Journal 58, 105–114. Cramer, M., 2010. Phosphate as a limiting resource: introduction. Plant and Soil 334, 1–10. Cross, A., Schlesinger, W., 1995. A literature review and evaluation of the Hedley fractionation: applications to biogeochemical cycle of P in natural ecosystems. Geoderma 64, 197–214. D'Elia, C.F., Steudler, P.A., Corwin, N., 1977. Determination of total nitrogen in aqueous samples using persulfate digestion. Limnology and Oceanography 22, 2–6. Debano, L.F., Conrad, C.E., 1978. The effects of fire on nutrients in a chaparral ecosystem. Ecology 59, 489–497. DeBano, L.F., Neary, D.G., Ffolliott, P.F., 1998. Fire's Effects on Ecosystems. Effects on Ecosystems. John Wiley & Sons, New York. 333 pp. Di Castri, F., Mooney, H.A., 1973. Mediterranean-type ecosystems: origin and structure. Ecological Studies. Springer, Berlin. 405 pp. Eivazi, F., Bayan, M.R., 1996. Effects of long-term prescribed burning on the activity of select soil enzymes in an oak–hickory forest. Canadian Journal of Forest Research 26, 1799–1804. Emmett, B.A., Beier, C., Estiarte, M., Tietema, A., Kristensen, H.L., Williams, D., Peñuelas, J., Schmidt, I., Sowerby, A., 2004. The response of soil processes to climate change: results from manipulation studies of shrublands across an environmental gradient. Ecosystems 7, 625–637. Galang, M.A., Markewitz, D., Morris, L.A., 2010. Soil phosphorus transformations under forest burning and laboratory heat treatments. Geoderma 155, 401–408. García-Fayos, P., Bochet, E., 2009. Evidence of antagonistic interaction between climate change and erosion on plant species richness and soil properties in semiarid Mediterranean ecosystems. Global Change Biology 15, 306–318.

Gee, G.W., Bauder, J.W., 1986. Particle size analysis. Part 1. In: Klute, A. (Ed.), Methods of Soil Analysis. Am. Soc. Agron., Madison, Wisconsin, USA, pp. 383–409. Giardina, C.P., Sanford Jr., R.L., Dockersmith, I.C., 2000. Changes in soil P and N during slash-and-burn clearing of a dry tropical forest. Soil Science Society of America Journal 64, 399–405. Giovannini, G., Lucchesi, S., Giachetti, M., 1990. Effects of heating on chemical parameters related to soil fertility and plant growth. Soil Science 149, 344–350. Hedley, M.J., Stewart, J.W.B., 1982. Method to measure microbial phosphate in soils. Soil Biology and Biochemistry 14, 377–385. Hedley, M.J., Stewart, J.W.B., Chauhan, B.S., 1982. Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Science Society of America Journal 46, 970–976. Henkin, Z., Seligman, N.G., Kafkafi, U., Noy-Meir, I., 1998. Effective growing days: a simple predictive model of response of herbaceous plant growth in a Mediterranean ecosystem to variation in rainfall and phosphorus availability. Journal of Ecology 86, 137–148. Henkin, Z., Sternberg, M., Seligman, N.G., Noy-Meir, I., 2006. Species richness in relation to phosphorus and competition in a Mediterranean dwarf-shrub community. Agriculture Ecosystems and Environment 113, 277–283. IPCC, 2007. Climate Change 2007: synthesis report. In: Pachauri, R.K., Reisinger, A. (Eds.), Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland. 104 pp. Jallah, J.K., Smyth, T.J., 1995. Predicting soil solution chemical attributes from more easily measured soil and soil solution parameters. Communications in Soil Science and Plant Analysis 26, 3315–3333. John, M.K., 1970. Colorimetric determination of phosphorus in soil and plant materials with ascorbic acid. Soil Science 109, 214–220. Johnson, D.W., Todd, D.E., Hanson, P.J., 2008. Effects of throughfall manipulation on soil nutrient status: results of 12 years of sustained wet and dry treatments. Global Change Biology 14, 1661–1675. Killham, K., 1994. Soil Ecology. Cambridge University Press, Cambridge. Li, X., Sarah, P., 2003. Enzyme activities along a climate change transect in the Judean Desert. Catena 53, 349–363. Marion, G.M., Moreno, J.M., Oechel, W.C., 1991. Fire severity, ash deposition, and clipping effects on soil nutrients in chaparral. Soil Science Society of America Journal 55, 235–240. Mayor, X., Rodà, F., 1994. Effects of irrigation and fertilization on stem diameter growth in a Mediterranean holm oak forest. Forest Ecology and Management 68, 119–126. McDowell, R., Condron, L., 2001. Influence of soil constituents on soil phosphorus sorption and desorption. Communications in Soil Science and Plant Analysis 32, 2531–2547. McLean, E.O., 1982. Soil pH and lime requirement. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis, Part 2, Chemical and Microbial Properties. Soil Sci. Soc. Am., Madison, WI, pp. 199–224. Moreno, J.M., Oechel, W.C., 1994. The role of fire in Mediterranean-type ecosystems. Ecological Studies. Springer-Verlag, New York. Moreno, J.M., Zavala, G., Martín, M., Millan, A., 2010. Forest fire risk in Spain under future climate change. In: Settele, J., Penev, L., Georgiev, T., Grabaum, R., Grobelnik, V., Hammen, V., Klotz, S., Kotarac, M., Kuhn, I. (Eds.), Atlas of Biodiversity Risk. Pensoft, Sofia & Moscow, pp. 72–73. Nelson, D.W., Sommers, L.E., 1982. Total carbon, organic carbon and organic matter. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis, Part 2. American Society of Agronomy, Madison, pp. 539–579. Parra, A., Ramírez, D.A., Resco, V., Velasco, Á., Moreno, J.M., 2012. Modifying rainfall patterns in a Mediterranean shrubland: system design, plant responses and experimental burning. International Journal of Biometeorology 56, 1033–1043. Rundell, P.W., 1983. Impact of fire on nutrient cycles in Mediterranean-type ecosystems with reference to chaparral. In: Kruger, F.J., Mitchell, D.T., Jarvis, J.V.M. (Eds.), Mediterranean-type Ecosystems: The Role of Nutrients. Springer-Verlag, Berlin, pp. 192–207. Rundell, P.W., 1998a. Landscape disturbance in Mediterranean-type ecosystems: an overview. In: Rundell, P.W., Montenegro, G., Jaksic, F.M. (Eds.), Landscape Degradation and Biodiversity in Mediterranean-type Ecosystems. : Ecological Studies, vol. 136. Springer-Verlag, Heidelberg, pp. 3–22. Rundell, P.W., 1998b. Vegetation, nutrition and climate (examples of integration), 3: leaf structure and nutrition in Mediterranean-climate sclerophylls. In: Specht, R.L. (Ed.), Mediterranean-type ecosystems. Kluwer, Dordrecht, pp. 157–167. Ryden, J.C., McLaughlin, J.R., Syers, J.K., 1977. Mechanisms of phosphate sorption of soils and hydrous ferric oxide gel. Journal of Soil Science 28, 72–92. Saá, A., Trasar-Cepeda, M.C., Carballas, T., 1998. Soil P status and phosphomonoesterase activity of recently burnt and unburnt soil following laboratory incubation. Soil Biology and Biochemistry 30, 419–428. Sala, O.E., Chapin III, F.S., Armesto, J.J., Berlow, E., Bloomfield, J., Dirzo, R., HuberSanwald, E., Huenneke, L.F., Jackson, R.B., Kinzing, A., Leemans, R., Lodge, D.M., Mooney, H.A., Oesterheld, M., Poff, N.L., Sykes, M.T., Walker, B.H., Walker, M., Wall, D.H., 2000. Global biodiversity scenarios for the year 2100. Science 287, 1770–1774. Sardans, J., Peñuelas, J., 2004. Increasing drought decreases phosphorus availability in an evergreen Mediterranean forest. Plant and Soil 267, 367–377. Sardans, J., Peñuelas, J., 2005. Drought decreases soil enzyme activity in a Mediterranean holm oak forest. Soil Biology and Biochemistry 37, 455–461. Sardans, J., Peñuelas, J., 2007. Drought changes phosphorus and potassium accumulation patterns in an evergreen Mediterranean forest. Functional Ecology 21, 191–201. Sardans, J., Peñuelas, J., 2010. Soil enzyme activity in a Mediterranean forest after six years of drought. Soil Science Society of America Journal 74, 838–851.

M.B. Hinojosa et al. / Geoderma 191 (2012) 61–69 Sardans, J., Rodà, F., Peñuelas, J., 2005. Effects of water and a nutrient pulse supply on Rosmarinus officinalis growth, nutrient content and flowering in the field. Environmental and Experimental Botany 53, 1–11. Sardans, J., Peñuelas, J., Estiarte, M., 2006. Warming and drought alter soil phosphatase activity and soil P availability in a Mediterranean shrubland. Plant and Soil 289, 227–238. Schröter, D., Cramer, W., Leemans, R., Prentice, I.C., Araújo, M.B., Arnell, N.W., Bondeau, A., Bugmann, H., Carter, T.R., Gracia, C.A., De La Vega-Leinert, A.C., Erhard, M., Ewert, F., Glendining, M., House, J.I., Kankaanpää, S., Klein, R.J.T., Lavorel, S., Lindner, M., Metzger, M.J., Meyer, J., Mitchell, T.D., Reginster, I., Rounsevell, M., Sabaté, S., Sitch, S., Smith, B., Smith, J., Smith, P., Sykes, M.T., Thonicke, K., Thuiller, W., Tuck, G., Zaehle, S., Zierl, B., 2005. Ecosystem service supply and vulnerability to global change in Europe. Science 310, 1333–1337. Serrasolsas, I., Khanna, P.K., 1995. Changes in heated and autoclaved forest soils of S.E. Australia. II. Phosphorus and phosphatase activity. Biogeochemistry 29, 25–41. Sibbesen, E., 1978. An investigation of the anion-exchange resin method for soil phosphate extraction. Plant and Soil 50, 305–321. Sommer, L.E., Nelson, D.W., 1972. Determination of total phosphorus in soils: a rapid perchloric acid digestion procedure. Proceedings. Soil Science Society of America 53, 32–37. Specht, R.L., 1973. Structure and functional response of ecosystems in the Mediterranean climate of Australia. In: Chapman, D., Castri, F., Mooney, H.A. (Eds.), Mediterraneantype Ecosystems: Orign and Structure. Chapman and Hall, London, pp. 113–120. Specht, R.L., Moll, E.J., 1983. Mediterranean-type heathlands and sclerophyllous shrublands of the world: an overview. In: Kruger, F.J., Mitchell, D.T., Jarvis, J.U.M. (Eds.), Mediterranean-type Ecosystems: The Role of Nutrients. Springer-Verlag, Berlin, pp. 41–65. St. John, T.V., Rundell, P.W., 1976. The role of fire as a mineralizing agent in a Sierran coniferous forest. Oecologia 25, 35–45. Stock, W.D., Lewis, O.A.M., 1986. Soil nitrogen and the role of fire as a mineralizing agent in a South African coastal fynbos ecosystem. Journal of Ecology 74, 317–328.

69

Tabatabai, M.A., 1982. Soil enzymes. In: Page, A.L., Millar, E.M., Keeney, D.R. (Eds.), Methods of Soil Analysis. American Society of Agronomy, Soil Science Society of America, Madison, pp. 501–538. Turrion, M.-B., Lafuente, F., Aroca, M.-J., López, O., Mulas, R., Ruipérez, C., 2010. Characterization of soil phosphorus in a fire-affected forest Cambisol by chemical extractions and 31P-NMR spectroscopy analysis. The Science of the Total Environment 408, 3342–3348. Viñegla, B., García-Ruíz, R., Liétor, J., Ochoa, V., Carreira, J.A., 2006. Soil phosphorus availability and transformation rates in relictic pinsapo fir forest from southern Spain. Biogeochemistry 78, 151–172. Vitousek, P.M., Farrington, H., 1997. Nutrient limitation and soil development: experimental test of a biogeochemical theory. Biogeochemistry 37, 63–75. Walker, T.W., Syers, J.K., 1976. The fate of phosphorus during pedogenesis. Geoderma 15, 1–19. Walkley, A., 1947. A critical examination of a rapid method for determination of organic carbon in soils — effect of variations in digestion conditions and of inorganic soil constituents. Soil Science 63, 251–257. Williams, J.D.H., Syers, S.K., Harris, R.F., Armstrong, D.E., 1971. Fractionation of inorganic phosphate in calcareous lake sediments. Proceedings. Soil Science Society of America 35, 250–255. Wolf, D.C., Skipper, H.D., 1994. Soil sterilization. In: Weaver, R.W. (Ed.), Methods of Soil Analysis. Part 2. Microbiological and Biochemical Properties. Soil Sci. Soc. Am., Madison, pp. 41–49. Yaalon, D.H., 1997. Soils in the Mediterranean region: what makes them different? Catena 28, 157–169. Zibilske, L.M., 1994. Carbon mineralization. In: Weaver, R.W. (Ed.), Methods of Soil Analysis, Part 2, Microbiological and Biochemical Properties. Soil Sci. Soc. Am., Madison, WI, pp. 835–863. Zou, X., Binkley, D., Doxtader, K.G., 1992. A new method for estimating gross phosphorus mineralization and immobilization rates in soils. Plant and Soil 147, 243–250.