Microbial biomass phosphorus and C/N/P stoichiometry in forest floor and A horizons as affected by tree species

Highlights

• Forest floor microbial biomass C and P contents were higher under beech than under spruce.

• Tree species did not affect microbial biomass C/P ratios.

• Unlike the microbial biomass C/P ratios, SOC/total P ratios varied widely at all sites.

• In L horizons, microbial biomass P was closely related to total P.

• In F and H horizons, microbial biomass P was closely related to microbial biomass C and N.

Abstract

Forest floor horizons contain significant total P stocks, but information on the contribution of microbial biomass P (P_MB) and on the controlling factors of this pool is limited. Slightly modified fumigation extraction procedures were used to investigate the stoichiometric relationships of P_MB to microbial biomass C (C_MB) and microbial biomass N (N_MB) in the forest floor (L, F, H, and A horizons) at five sites, differing in P availability to trees, under adjacent spruce (Picea abies) and beech (Fagus sylvatica) stands. C_MB, N_MB, and P_MB contents were higher in forest floors under beech than under spruce. Mean stocks of P_MB and total P were roughly 27 and 100 kg ha⁻¹ in the forest floor, respectively, but did not differ between the tree species, due to an increased organic matter accumulation in the forest floor under spruce. This reveals the importance of forest floor horizons and microbial biomass turnover for P nutrition of trees in acidic soils with the humus form moder. C/P_MB ratios declined from roughly 26 in L to 13 in F and H horizons, followed by an increase to roughly 17 in A horizons. The range of C/P_MB ratios was small at all sites in relation to the wide SOC/total P ratios of the litter used as microbial substrate, indicating a relatively strict homeostatic regulation of the forest floor microbial, mainly fungal biomass stoichiometry.

Introduction

Forest floors are important P reservoirs in forest ecosystems. In the presence of moder-type humus, trees have been shown to take up considerably more phosphate from the forest floor than from the underlying mineral soil (Brandtberg et al., 2004, Jonard et al., 2009, Jonard et al., 2010). When P-adsorbing mineral phases are lacking, the net release of plant available phosphate in forest floor horizons is determined by microbial mineralization and immobilization processes (Oberson and Joner, 2005, Rosling et al., 2016, Šantrůčková et al., 2004). Even under near steady state conditions, P is dynamically exchanged between microbial biomass and soil solution (Achat et al., 2010a, Achat et al., 2010c, Oehl et al., 2001). Consequently, the turnover of microbial biomass P (P_MB) replenishes soil solution P and contributes to P nutrition of forest trees (Achat et al., 2012).

The P_MB pool in temperate mineral forest soil horizons has been shown to be more closely correlated to soil organic C (SOC) or total N than to total P contents (Achat et al., 2012, Joergensen et al., 1995a, Joergensen et al., 1995b, Khan and Joergensen, 2012). Achat et al. (2012) and Heuck et al. (2015) concluded that microbial P uptake is largely dependent on C availability, which determines the size of microbial biomass C (C_MB). In the forest floor, where C availability to microorganisms is considerably higher than in mineral soils, as indicated by higher C_MB/SOC ratios (Joergensen and Scheu, 1999a), P contents might influence P_MB contents more strongly. However, the few results available from studies comparing P_MB contents in forest floor horizons differing in total P contents are inconsistent. Saggar et al. (1998) found higher P_MB contents in forest floors of P fertilized Pinus radiata stands in New Zealand than in non-fertilized stands. Clarholm (1993) observed the opposite in Swedish forest floors of P fertilized and non-fertilized Picea abies stands.

C/P_MB and also N/P_MB ratios vary over a considerably wider range than C/N_MB ratios in temperate forest ecosystems (Chávez-Vergara et al., 2016, Joergensen et al., 1995a, Khan and Joergensen, 2012) as well as on a global scale (Cleveland and Liptzin, 2007, Hartman and Richardson, 2013, Xu et al., 2013). According to a modeling study by Manzoni et al. (2010), the high variability of C/P_MB ratios is one of the main reasons for the limited power of litter C/P ratios in predicting net P release rates during litter decomposition. C/P_MB ratios decrease under high P availability, combined with low C availability and high N availability in cell culture (Anderson and Domsch, 1980, Kouno et al., 1999, Lukito et al., 1998) and incubation studies (Salamanca et al., 2006). However, there is only limited knowledge on the impact of N and especially P availability on microbial C/N/P stoichiometry in temperate forest soil ecosystems (Joergensen et al., 1995a, Joergensen et al., 1995b). This is especially true for forest floor horizons, where C, N, and P availability strongly depends on litter stoichiometry and decomposition stage (Mooshammer et al., 2014). Relationships between the elemental composition of forest floor horizons and microbial biomass have rarely been examined, due to methodological constraints (Joergensen and Scheu, 1999b, Saggar et al., 1998). For this reason, forest floor horizons have often been removed (Yang et al., 2010, Zhao et al., 2009) or have not been analyzed (Bing et al., 2016, Yang and Zhu, 2015) in studies on P_MB in forest ecosystems.

Microbial biomass contents and decomposition processes in forest floor horizons can be significantly affected by tree species. Different litter qualities of coniferous and deciduous species in respect to lignin and polyphenol contents, elemental composition, pH as well as physical characteristics (Berg and McClaugherty, 2014, Binkley and Giardina, 1998, Hobbie et al., 2007, Perry et al., 1987, Reich et al., 2005) have been reported to affect decomposition processes. C_MB/SOC ratios have been found to be lower under coniferous than deciduous tree species (Bauhus et al., 1998, Scheu and Parkinson, 1995, Zhong and Makeschin, 2004). Although P contents of beech leaf and spruce needle litter might be similar (Hagen-Thorn et al., 2004, Trum et al., 2011), lower C and N availability to microorganisms in spruce compared with beech forest floors may also be associated with a lower microbial P demand, resulting in lower P_MB contents. In contrast, the higher SOC stocks in spruce forest floors (Berger and Berger, 2012, Cremer et al., 2016, Vesterdal et al., 2013) may compensate for this and result in similar forest floor P_MB stocks.

In the present study, we investigated L, F, and H horizons as well as the upper A horizon of five adjacent beech (Fagus sylvatica L.) and spruce (Picea abies L.) stands at five sites, varying strongly in parent material and P availability. After conducting a methodological pre-study to test the effect of the type of extractant and of the soil/extractant ratio on P_MB estimates (Bergkemper et al., 2016, Brookes et al., 1982, Khan and Joergensen, 2012), we adapted the fumigation extraction method to measure P_MB in forest floor horizons. Our central objectives were to investigate the following hypotheses: (I) C_MB and P_MB contents are higher in forest floors under beech than under spruce, but there is no tree species effect on C/P_MB ratios. (II) Stocks of P_MB in the forest floor do not differ between tree species, due to an increased forest floor accumulation under spruce. (IIIa) P_MB contents are affected by total P in forest floor horizons and (IIIb) by SOC contents in A horizons. (IV) C/P_MB ratios decrease with depth from L to A horizons as the substrate is increasingly degraded.