Oecologia DOI 10.1007/s00442-016-3581-x
PHYSIOLOGICAL ECOLOGY – ORIGINAL RESEARCH
Phosphorus resorption by young beech trees and soil phosphatase activity as dependent on phosphorus availability Kerstin Hofmann1 · Christine Heuck1 · Marie Spohn1
Received: 13 November 2015 / Accepted: 1 February 2016 © Springer-Verlag Berlin Heidelberg 2016
Abstract Motivated by decreasing foliar phosphorus (P) concentrations in Fagus sylvatica L. forests, we studied P recycling depending on P fertilization in mesocosms with juvenile trees and soils of two contrasting F. sylvatica L. forests in a greenhouse. We hypothesized that forests with low soil P availability are better adapted to recycle P than forests with high soil P availability. The P resorption efficiency from senesced leaves was significantly higher at the P-poor site (70 %) than at the P-rich site (48 %). P fertilization decreased the resorption efficiency significantly at the P-poor site to 41 %, while it had no effect at the P-rich site. Both acid and alkaline phosphatase activity were higher in the rhizosphere of the P-poor than of the P-rich site by 53 and 27 %, respectively, while the activities did not differ in the bulk soil. Fertilization decreased acid phosphatase activity significantly at the P-poor site in the rhizosphere, but had no effect on the alkaline, i.e., microbial, phosphatase activity at any site. Acid phosphatase activity in the P-poor soil was highest in the rhizosphere, while in the P-rich soil, it was highest in the bulk soil. We conclude that F. sylvatica resorbed P more efficiently from senescent leaves at low soil P availability than at high P availability and that acid phosphatase activity in the rhizosphere but not
Communicated by Hakan Wallander. Electronic supplementary material The online version of this article (doi:10.1007/s00442-016-3581-x) contains supplementary material, which is available to authorized users. * Marie Spohn marie.spohn@uni‑bayreuth.de 1
Department of Soil Ecology, Bayreuth Center of Ecology and Environmental Research (BayCEER), University of Bayreuth, Bayreuth, Germany
in the bulk soil was increased at low P availability. Moreover, we conclude that in the P-rich soil, microbial phosphatases contributed more strongly to total phosphatase activity than plant phosphatases. Keywords Forest nutrition · Nutrient resorption · Rhizosphere · Phosphomonoesterase · Plant–microbe interaction
Introduction Foliar phosphorus (P) contents have been decreasing in a range of temperate forests, and especially in Fagus sylvatica L. forests, during the last decades, indicating a decreasing P availability at these sites (Flückiger and Braun 1998; Duquesnay et al. 2000; Ilg et al. 2009; Braun et al. 2010; Crowley et al. 2012; Talkner et al. 2015). It seems likely that P is more efficiently recycled in forests with low P availability than in forests with high P availability (Aerts and Chapin 2000). However, to what extent F. sylvatica forests can adjust P recycling processes to decreased P availability has not yet been studied. P can be recycled in forests at the tree level by resorption of P from senescent leaves and at the ecosystem level by mineralization of soil organic P. Nutrients resorbed during senescence are directly available for subsequent plant use, reducing a plant’s dependence on nutrient uptake from soil (Aerts 1996; Aerts and Chapin 2000; Reed et al. 2012). In a meta-analysis on plant nutrient resorption, Aerts (1996) found that deciduous trees and shrubs resorbed on average 50 % of the foliar P before leaf abscission. P resorption in deciduous trees and shrubs was not very responsive to P availability as P resorption decreased upon enhanced nutrient supply only in 35 % of
13
Oecologia
the analyzed studies. However, in interspecies comparisons along pedogenic chronosequences it was found that P resorption efficiency (the percentage of foliar P withdrawn before leaf abscission) increased strongly with decreasing P availability (Richardson et al. 2004; Hayes et al. 2014). This finding is in line with a meta-analysis showing that P resorption efficiency increased with decreasing latitude, suggesting that it is highest in ecosystems with old and therefore P-depleted soils (Yuan and Chen 2009). The mineralization of soil organic P is a key process of P recycling in forest ecosystems. Organic P comprises up to 95 % of the total P in temperate forest soils (Cross and Schlesinger 1995; Yang and Post 2011; Stutter et al. 2012), and has to be mineralized before plants can take it up (Vance et al. 2003). Organic P mineralization is catalyzed by phosphatases. The dominant group of phosphatases are phosphomonoesterases, which are relatively non-specific phosphatases that act on a wide range of low-molecular-weight P compounds with a phosphomonoester bond (Nannipieri et al. 2011). According to their pH optima, phosphomonoesterases can be separated into acid and alkaline phosphomonoesterases. While microbes are capable of producing both acid and alkaline phosphomonoesterases that have a broader pH optimum and are also active at a high soil pH, plants and ectomycorrhiza can only produce acid phosphomonoesterases (Dick et al. 1983; Juma and Tabatabai 1988; Nannipieri et al. 2011). With respect to P acquisition, plant–microbe interactions can be mutualistic as well as competitive. Microorganisms can mineralize more organic P than they need, which is beneficial for plants because it provides them with inorganic P. However, microbes might also compete with plants for P and immobilize a large amount of P in their biomass (Richardson et al. 2009; Spohn and Kuzyakov 2013a; Heuck et al. 2015). Soil phosphatase activity strongly varies spatially in soil. It has been shown in several studies on herbaceous plants that the rhizosphere is a hotspot of phosphatase activity (Tarafdar and Jungk 1987; Rastin et al. 1988; Spohn and Kuzyakov 2013a), which in part is likely due to a stimulation of microbial phosphatase production by root exudates (Spohn et al. 2013b). Concerning trees, it has been found that the rhizosphere of Pinus elliotti was a hotspot of acid phosphatase activity in P-poor, but not in P-rich soil (Fox and Comerford 1992). Since the spatial variability of phosphatases is very high, variability in time is very difficult to detect by traditional, destructive enzyme assays (Harrison and Pearce 1979; Turner et al. 2002). This might be the reason why previous studies report contradicting findings concerning the question as to whether there is a seasonal dynamic of phosphatase activity in temperate soils
13
(Harrison and Pearce 1979; Turner et al. 2002). However, non-destructive, in situ methods allow the measurement of phosphatase activity in the same microsite several times, which might help to differentiate seasonal and spatial variability. The objective of this study was to gain insight into P recycling in the plant–microbe-soil system of two contrasting F. sylvatica L. forests depending on P availability using a non-destructive enzyme assay. We hypothesized that young beech trees of the P-poor site resorb P of senescent leaves more efficiently than young beech trees at the P-rich site as an adaptation to low P availability, and that phosphatase activity is higher in the soil of the P-poor site than in the soil of the P-rich site. Concerning the distribution in time and space, we hypothesized that phosphatase activity is highest in spring after leaf expansion when the stored nutrients have been used, and that it is generally higher in the rhizosphere than in the bulk soil due to combined plant and microbial activities.
Materials and methods Study sites Soils and juvenile trees were collected at two sites with contrasting soil P availability in Germany in February 2014. The site Bad Brueckenau (BBR) that has a high soil P availability is situated in the Rhoen Mountains (50°21.38′N, 9°55.71′E) at 825 m above sea level. The mean annual rainfall is 1031 mm and the mean annual temperature is 5.8 °C. The parent material is tertiary volcanic rock. The soil is a Dystric Skeletic Cambisol (FAO) derived from Basalt, and the prevailing tree species is European beech (Fagus sylvatica L.). The site Unterluess (LUE) that has a low soil P availability is located in the Lueneburg Heath (52°50.32′N, 10°16.06′E) at 115 m above sea level. The mean annual rainfall amounts to 730 mm and the mean annual temperature is 8 °C. The soil is a Hyperdystric Folic Cambisol which developed from sandy Pleistocene sediments, and is vegetated by European beech (F. sylvatica). Samples of A and B horizons were collected with a shovel after removal of the organic layers at both sites, and 2-yearold trees were collected using a fork. Experimental setup The young beech trees and soil samples were transferred to the University of Bayreuth, where the samples of the A and B horizon of each site were sieved (
PHYSIOLOGICAL ECOLOGY – ORIGINAL RESEARCH
Phosphorus resorption by young beech trees and soil phosphatase activity as dependent on phosphorus availability Kerstin Hofmann1 · Christine Heuck1 · Marie Spohn1
Received: 13 November 2015 / Accepted: 1 February 2016 © Springer-Verlag Berlin Heidelberg 2016
Abstract Motivated by decreasing foliar phosphorus (P) concentrations in Fagus sylvatica L. forests, we studied P recycling depending on P fertilization in mesocosms with juvenile trees and soils of two contrasting F. sylvatica L. forests in a greenhouse. We hypothesized that forests with low soil P availability are better adapted to recycle P than forests with high soil P availability. The P resorption efficiency from senesced leaves was significantly higher at the P-poor site (70 %) than at the P-rich site (48 %). P fertilization decreased the resorption efficiency significantly at the P-poor site to 41 %, while it had no effect at the P-rich site. Both acid and alkaline phosphatase activity were higher in the rhizosphere of the P-poor than of the P-rich site by 53 and 27 %, respectively, while the activities did not differ in the bulk soil. Fertilization decreased acid phosphatase activity significantly at the P-poor site in the rhizosphere, but had no effect on the alkaline, i.e., microbial, phosphatase activity at any site. Acid phosphatase activity in the P-poor soil was highest in the rhizosphere, while in the P-rich soil, it was highest in the bulk soil. We conclude that F. sylvatica resorbed P more efficiently from senescent leaves at low soil P availability than at high P availability and that acid phosphatase activity in the rhizosphere but not
Communicated by Hakan Wallander. Electronic supplementary material The online version of this article (doi:10.1007/s00442-016-3581-x) contains supplementary material, which is available to authorized users. * Marie Spohn marie.spohn@uni‑bayreuth.de 1
Department of Soil Ecology, Bayreuth Center of Ecology and Environmental Research (BayCEER), University of Bayreuth, Bayreuth, Germany
in the bulk soil was increased at low P availability. Moreover, we conclude that in the P-rich soil, microbial phosphatases contributed more strongly to total phosphatase activity than plant phosphatases. Keywords Forest nutrition · Nutrient resorption · Rhizosphere · Phosphomonoesterase · Plant–microbe interaction
Introduction Foliar phosphorus (P) contents have been decreasing in a range of temperate forests, and especially in Fagus sylvatica L. forests, during the last decades, indicating a decreasing P availability at these sites (Flückiger and Braun 1998; Duquesnay et al. 2000; Ilg et al. 2009; Braun et al. 2010; Crowley et al. 2012; Talkner et al. 2015). It seems likely that P is more efficiently recycled in forests with low P availability than in forests with high P availability (Aerts and Chapin 2000). However, to what extent F. sylvatica forests can adjust P recycling processes to decreased P availability has not yet been studied. P can be recycled in forests at the tree level by resorption of P from senescent leaves and at the ecosystem level by mineralization of soil organic P. Nutrients resorbed during senescence are directly available for subsequent plant use, reducing a plant’s dependence on nutrient uptake from soil (Aerts 1996; Aerts and Chapin 2000; Reed et al. 2012). In a meta-analysis on plant nutrient resorption, Aerts (1996) found that deciduous trees and shrubs resorbed on average 50 % of the foliar P before leaf abscission. P resorption in deciduous trees and shrubs was not very responsive to P availability as P resorption decreased upon enhanced nutrient supply only in 35 % of
13
Oecologia
the analyzed studies. However, in interspecies comparisons along pedogenic chronosequences it was found that P resorption efficiency (the percentage of foliar P withdrawn before leaf abscission) increased strongly with decreasing P availability (Richardson et al. 2004; Hayes et al. 2014). This finding is in line with a meta-analysis showing that P resorption efficiency increased with decreasing latitude, suggesting that it is highest in ecosystems with old and therefore P-depleted soils (Yuan and Chen 2009). The mineralization of soil organic P is a key process of P recycling in forest ecosystems. Organic P comprises up to 95 % of the total P in temperate forest soils (Cross and Schlesinger 1995; Yang and Post 2011; Stutter et al. 2012), and has to be mineralized before plants can take it up (Vance et al. 2003). Organic P mineralization is catalyzed by phosphatases. The dominant group of phosphatases are phosphomonoesterases, which are relatively non-specific phosphatases that act on a wide range of low-molecular-weight P compounds with a phosphomonoester bond (Nannipieri et al. 2011). According to their pH optima, phosphomonoesterases can be separated into acid and alkaline phosphomonoesterases. While microbes are capable of producing both acid and alkaline phosphomonoesterases that have a broader pH optimum and are also active at a high soil pH, plants and ectomycorrhiza can only produce acid phosphomonoesterases (Dick et al. 1983; Juma and Tabatabai 1988; Nannipieri et al. 2011). With respect to P acquisition, plant–microbe interactions can be mutualistic as well as competitive. Microorganisms can mineralize more organic P than they need, which is beneficial for plants because it provides them with inorganic P. However, microbes might also compete with plants for P and immobilize a large amount of P in their biomass (Richardson et al. 2009; Spohn and Kuzyakov 2013a; Heuck et al. 2015). Soil phosphatase activity strongly varies spatially in soil. It has been shown in several studies on herbaceous plants that the rhizosphere is a hotspot of phosphatase activity (Tarafdar and Jungk 1987; Rastin et al. 1988; Spohn and Kuzyakov 2013a), which in part is likely due to a stimulation of microbial phosphatase production by root exudates (Spohn et al. 2013b). Concerning trees, it has been found that the rhizosphere of Pinus elliotti was a hotspot of acid phosphatase activity in P-poor, but not in P-rich soil (Fox and Comerford 1992). Since the spatial variability of phosphatases is very high, variability in time is very difficult to detect by traditional, destructive enzyme assays (Harrison and Pearce 1979; Turner et al. 2002). This might be the reason why previous studies report contradicting findings concerning the question as to whether there is a seasonal dynamic of phosphatase activity in temperate soils
13
(Harrison and Pearce 1979; Turner et al. 2002). However, non-destructive, in situ methods allow the measurement of phosphatase activity in the same microsite several times, which might help to differentiate seasonal and spatial variability. The objective of this study was to gain insight into P recycling in the plant–microbe-soil system of two contrasting F. sylvatica L. forests depending on P availability using a non-destructive enzyme assay. We hypothesized that young beech trees of the P-poor site resorb P of senescent leaves more efficiently than young beech trees at the P-rich site as an adaptation to low P availability, and that phosphatase activity is higher in the soil of the P-poor site than in the soil of the P-rich site. Concerning the distribution in time and space, we hypothesized that phosphatase activity is highest in spring after leaf expansion when the stored nutrients have been used, and that it is generally higher in the rhizosphere than in the bulk soil due to combined plant and microbial activities.
Materials and methods Study sites Soils and juvenile trees were collected at two sites with contrasting soil P availability in Germany in February 2014. The site Bad Brueckenau (BBR) that has a high soil P availability is situated in the Rhoen Mountains (50°21.38′N, 9°55.71′E) at 825 m above sea level. The mean annual rainfall is 1031 mm and the mean annual temperature is 5.8 °C. The parent material is tertiary volcanic rock. The soil is a Dystric Skeletic Cambisol (FAO) derived from Basalt, and the prevailing tree species is European beech (Fagus sylvatica L.). The site Unterluess (LUE) that has a low soil P availability is located in the Lueneburg Heath (52°50.32′N, 10°16.06′E) at 115 m above sea level. The mean annual rainfall amounts to 730 mm and the mean annual temperature is 8 °C. The soil is a Hyperdystric Folic Cambisol which developed from sandy Pleistocene sediments, and is vegetated by European beech (F. sylvatica). Samples of A and B horizons were collected with a shovel after removal of the organic layers at both sites, and 2-yearold trees were collected using a fork. Experimental setup The young beech trees and soil samples were transferred to the University of Bayreuth, where the samples of the A and B horizon of each site were sieved (
