Biomass, morphology and nutrient contents of fine roots in four Norway spruce stands
- First Online:
- Cite this article as:
- Borken, W., Kossmann, G. & Matzner, E. Plant Soil (2007) 292: 79. doi:10.1007/s11104-007-9204-x
Fine root systems may respond to soil chemical conditions, but contrasting results have been obtained from field studies in non-manipulated forests with distinct soil chemical properties. We investigated biomass, necromass, live/dead ratios, morphology and nutrient concentrations of fine roots (<2 mm) in four mature Norway spruce (Picea abies [L.] Karst.) stands of south-east Germany, encompassing variations in soil chemical properties and climate. All stands were established on acidic soils (pH (CaCl2) range 2.8–3.8 in the humus layer), two of the four stands had molar ratios in soil solution below 1 and one of the four stands had received a liming treatment 22 years before the study. Soil cores down to 40 cm mineral soil depth were taken in autumn and separated into four fractions: humus layer, 0–10 cm, 10–20 cm and 20–40 cm. We found no indications of negative effects of N availability on fine root properties despite large variations in inorganic N seepage fluxes (4–34 kg N ha−1 yr−1), suggesting that the variation in N deposition between 17 and 26 kg N ha−1 yr−1 does not affect the fine root system of Norway spruce. Fine root biomass was largest in the humus layer and increased with the amount of organic matter stored in the humus layer, indicating that the vertical pattern of fine roots is largely affected by the thickness of this horizon. Only two stands showed significant differences in fine root biomass of the mineral soil which can be explained by differences in soil chemical conditions. The stand with the lowest total biomass had the lowest Ca/Al ratio of 0.1 in seepage, however, Al, Ca, Mg and K concentrations of fine roots were not different among the stands. The Ca/Al ratio in seepage might be a less reliable stress parameter because another stand also had Ca/Al ratios in seepage far below the critical value of 1.0 without any signs of fine root damages. Large differences in the live/dead ratio were positively correlated with the Mn concentration of live fine roots from the mineral soil. This relationship was attributed to faster decay of dead fine roots because Mn is known as an essential element of lignin degrading enzymes. It is questionable if the live/dead ratio can be used as a vitality parameter of fine roots since both longevity of fine roots and decay of root litter may affect this parameter. Morphological properties were different in the humus layer of one stand that was limed in 1983, indicating that a single lime dose of 3–4 Mg ha−1 has a long-lasting effect on fine root architecture of Norway spruce. Almost no differences were found in morphological properties in the mineral soil among the stands, but vertical patterns were apparently different. Two stands with high base saturation in the subsoil showed a vertical decrease in specific root length and specific root tip density whereas the other two stands showed an opposite pattern or no effect. Our results suggest that proliferation of fine roots increased with decreasing base saturation in the subsoil of Norway spruce stands.
KeywordsCa/Al ratioFine rootsFine root biomassFine root morphologyLimingMn concentrationNitrogen depositionNorway spruce
Distribution and morphology of fine roots are the result of internal factors as the genotype of plant species and several external factors such as temperature, precipitation, soil properties, nutrient availability and competition between plants (Leuschner et al. 2004; Braun et al. 2005; Majdi et al. 2005). Fine root systems are dynamic and may respond within months to changes in soil chemical properties and water availability because of high turnover rates. Lifespan (turnover time) of less than 1 year and more than 10 years have been reported for fine roots <2 mm in diameter from different tree species (e.g., Majdi et al. 2005; Trumbore & Gaudinski, 2003). The turnover time and biomass of fine roots is of relevance for growth of individual trees and competition with other plants because a large fraction of annually produced carbohydrates is required to maintain the fine root system including the transfer of carbohydrates to mycorrhiza. Janssens et al. (2002) estimated that 28–49% of gross primary production (GPP) is transferred to the root system in a Scots pine forest.
Most temperate and boreal forests are nitrogen limited, but high atmospheric input of ammonium and nitrate have increased the availability of nitrogen in many temperate forests of Europe and some regions of north-east America during past decades. An analysis of forest inventories suggest an increased biomass production of European forests which at least is partially caused by elevated atmospheric N deposition (Spiecker 1999). The implications of increasing N availability for fine root systems and their nutrient status are still unclear. An increase of fine root N concentrations was observed in Norway spruce forests along a deposition gradient from northern Europe (low N input) to central Europe (high N input) (Högberg et al. 1998). Application of ammonium sulphate (100 kg N ha−1 yr−1) enhanced the N concentration of fine roots in a 28-year old Norway spruce stand in Sweden (Majdi and Rosengren-Brinck 1994). In the same study, Persson et al. (1995) found no effect on fine root biomass, but the amount of fine root necromass increased and the live/dead ratio of fine roots dropped from 2.2 to 0.9 in the humus layer. By contrast, long-term application of ‘clean rain’ with reduced amounts of nitrate and ammonium enhanced fine root biomass and reduced N concentration of fine roots in a Norway spruce forest in Germany although the average input of about 11 kg N ha−1 y−1 is close to the annual net uptake of Norway spruce (Lamersdorf and Borken 2004). The study suggests that Norway spruce compensated the decrease in N availability by proliferation of the fine root system.
The effect of N deposition on fine root systems may be superimposed by other chemical soil properties. Soil acidity and aluminum toxicity have been considered as potential problems for fine root systems (Cronan and Grigal 1995; Matzner and Murach 1995; Jentschke et al. 2001). In acid mineral soils, high Al concentrations in soil solution may cause damages to root tips through inhibition of cell division and cell elongation (Puhe 2003). High Al concentrations may suppress the uptake of Mg and Ca which can lead to nutrient imbalances, in particular, at high N deposition rates (Joslin et al. 1988; Matzner and Murach 1995; de Wit et al. 2001). Soil acidification may diminish the penetration of fine roots in mineral soil horizon and may alter the turnover and morphology of fine roots. Godbold et al. (2003) reported higher fine root turnover and decreased live/dead ratio, decreased specific root length and specific root tip density of Norway spruce in acidified soils as compared to less acidified soils.
Puhe (2003) proposed damage classes for the root system of Norway spruce and suggested that abnormalities of the typical root structure indicate soil chemical stress as a result of nutrient imbalances, soil acidification or Al toxicity. Damage of root systems may reduce the stress resistance as well as the annual growth increment of Norway spruce. This tree species covers about 1/3 of the German forest area and is therefore ecologically and economically important, but little is known about the long-term effects of atmospheric deposition on the vitality of fine root systems. The atmospheric emissions of sulfur and protons dramatically declined during the past 15 years whereas the emissions of inorganic N remained on a high level (UBA 2006).
We hypothesized that fine root biomass is negatively and fine root N concentration positively correlated with increasing leaching losses of inorganic N. The objectives of this study were to compare (1) live and dead fine root mass, (2) nutrient contents and (3) morphological properties of live fine roots from different soil depths of four mature Norway spruce stands. Fine root properties were related to long-term averages of nutrient fluxes in throughfall and seepage as well as to stand and soil parameters.
Materials and methods
Characteristics of the four study sites
Mean annual precipitation [mm]
Mean annual temperature [°C]
Growing season [d]
Tree age in 2005 [yr]
Stand density [trees ha−1]
Mean increment [m3 ha−1 yr−1]
Thickness of humus layer [cm]
Amount of humus layer [kg m−2]
aAmount of humus layer [kg m−2]
Mean fluxes of inorganic N [kg N ha−1a−1] from 1996–2003
20 cm soil depth
40 cm soil depth
Mean molar Ca/Al ratio in soil solution from 1996–2003
20 cm soil depth
40 cm soil depth
Tree age ranged between 84 and 99 years at the four sites (Table 1). The stand density of 354 trees per hectare is lowest in GOL while the other stands have a similar stand density of 500 to 600 trees per hectare. Because of dramatic growth depressions at AOE and HOE in the extremely dry summer of 2003 the mean increment of the stands is given for the more representative time period from 1997 to 2002.
pH, cation exchange capacity (CEC) and base saturation of the humus layer and different mineral soil horizons at the four Norway spruce sites AOE, FLO, GOL and HOE
Mineral soil horizons
CEC (μmolc g−1)
Base saturation (%)
CEC (μmolc g−1)
Base saturation (%)
CEC (μmolc g−1)
Base saturation (%)
CEC (μmolc g−1)
Base saturation (%)
The soil at FLO, a Haplic Podzol, is developed on precambric gneiss while the soil at GOL, a Dystric Cambisol, developed on cambric phyllite and phyllitic shist. FLO has the most acidic top soil with a pH (CaCl2) of about 2.8 in the humus layer, 2.9 (0–4 cm), 3.1 (4–6 cm) and 4.4 (30–45 cm) in the mineral soil. The subsoil at FLO (45–60 cm) has a pH value of about 4.5, low CEC of 40 μmolc g−1 soil and a base saturation of 5% down to 1.3 m depth. At GOL, the relatively high pH (CaCl2) value of 3.8, CEC of 359 μmolc g−1 soil and the base saturation of 68% in the humus layer results from a dolomite (MgCaCO3) plus Kalimagnesia (30% K2O, 10%MgO, 17% S) addition of about 3–4 ha−1 in 1983. This treatment increased the pH (3.5) and in the mineral soil at 0–2 cm, but had little effect on pH at 2–7 cm (3.1), 7–17 cm (3.3) and 17–40 cm (4.0) depth. CEC (26 μmolc g−1 soil) and base saturation (4%) are extremely low in the subsoil (40–70 cm).
Molar ratios of Ca/Al of soil solution ranged between 2.3 and 17.2 in the humus layer and between 0.1 and 1.0 in the mineral soil at 20 cm depth (Table 1). The high Ca/Al ratio of 17.2 in the humus layer and of 1.0 in the mineral soil at GOL may be explained by the liming in 1983. The increase in the Ca/Al ratio from 20–40 cm depth at AOE and HOE corresponds with the increase in base saturation in the subsoil (Table 2). FLO had the lowest Ca/Al ratio of 0.1 in soil solution at 20 and 40 cm depth.
Moder is the dominating humus type in all stands with mean thicknesses of 5.0–8.5 cm, however, the variation in the amount of organic matter is large at individual sites and may considerably differ from the samples taken in this study using a soil corer (see below). Long-term averages of N input by throughfall varied between 17 and 26 kg ha−1 yr−1 (Table 1). Leaching losses of 4 to 9 kg N ha−1 yr−1 were measured at 40 cm soil depth in AOE, FLO and GOL, however, N solute fluxes were exceptional high in the humus layer and mineral soil at HOE.
Sampling and sample preparation
Soil samples were collected from the four sites within four weeks between September and October of 2005 using a corer of 8 cm in diameter and 40 cm in length. The humus layer and the mineral soil at 0–10, 10–20 and 20–40 depth were consecutively taken to avoid compaction of the soil samples. We have randomly chosen four trees on an 1 ha area and collected 3 soil samples 2 m apart to the trees, i.e. in total 12 soil samples per depth and site. All samples were transported in cooling boxes, weighed and afterwards stored at 2°C in a cooling chamber until sample preparation. An additional soil sample was taken at each tree to determine gravimetric water contents of the humus layer and mineral soil depths. These samples were sieved (2 mm) to remove gravels and roots and then dried at 60°C (humus layer) and 105°C (mineral soil) over 48 h.
Roots and gravels were washed out of soil samples with tap water using a sieve of 2 mm mesh size within 4 weeks after sampling in the field. Roots from the humus layers were further cleaned in a supersonic bath for 30 minutes because of attached fine organic matter particles. After cleaning, roots were visually sorted in bio- and necromass using a binocular and by means of root colour and elasticity, degree of cohesion of cortex and primary root structure. Live roots were stored no longer than 2 weeks at 2°C in deionised water for further morphological analysis. Dead fine roots <2 mm were dried at 40°C over 48 h for determination of necromass.
Morphology of fine roots
Morphology of live fine root was investigated using an image evaluation software (Regent Instruments Inc. 2003, WinRhizo 2003b). For this purpose, the roots were scanned in a water bath at 400 dpi resolution. The images were analyzed with the software to identify the morphological parameters of fine roots. Roots with a diameter greater than 2 mm were excluded from the analysis. The measured parameters were fine root length, average diameter, number of forks, number of crossings, and number of root tips and surface area of fine roots. Root length density (m l−1 soil), specific root length (m g−1 root), root tip density (tips l−1 soil), specific root tip density (tips g−1 root), and specific root area (cm−2 g−1 root) were calculated for each sample, except samples without live roots (in total n = 10). After morphological analysis, fine root biomass was determined by drying of roots at 40°C at least over 48 h until weight constancy was reached. Some additional fine roots were dried first at 40°C and afterwards at 105°C in order to test the difference in gravimetric water contents. The difference was less than 0.5%, indicating that fine root biomass was not overestimated using a drying temperature of 40°C.
Because the amount of fine root material was too small in some mineral soil samples for chemical analysis fine roots from the 10–20 and 20–40 cm depth of each soil core were pooled to one sample prior to chemical analysis. After grinding and milling of dried fine roots the C and N concentrations were analysed using an elemental analyzer (CHN-O-Rapid, Heraeus Elementar, Hanau, Germany). Concentrations of Al, Fe, Ca, K, Mg, P, S, Mn and Na were determined after HNO3 pressure digestion at 170°C by inductive coupled plasma atomic emission spectroscopy (gcp Electronics, Penrith, Australia). The element concentrations of roots, in particular K, might have been underestimated because of diffusion losses following storing of live roots in deionised water.
Statistical analysis of the data was carried out using Statistica 6.1 (Statsoft Inc. 2002). The Shapiro–Wilk test was performed to analyze the distribution of all data sets. Because the histograms of most of the data were skewed even after log or root transformation the one-factorial Kruskal–Wallis-ANOVA was performed to test the site effect on fine root biomass, fine root density, live/dead ratios of fine roots, element concentrations and morphological parameters per soil depth or the sum of all depths. Additionally, the global significance of soil depth (vertical gradient) of the same parameters was separately tested for each site. When the Kruskal–Wallis Test was significant at α = 0.05, the Mann–Whitney U-Test was used as a post-hoc test to compare gradually the differences between two spruce stands. According to the sequential Bonferroni Method, the significance level was adjusted by α/k, α/k−1 to α/k−5, where k is the number of pairs (n = 6) for each parameter. The lowest P value of the Mann–Whitney U-Test was compared with the lowest adjusted significance level (α/k) and the second lowest P value was compared with the second lowest significance level (α/k-1). Linear regressions were performed to evaluate the relationships between different parameters.
Biometric properties of fine roots
The total amount of fine root biomass down to 40 cm was similar at AOE (338 g m−2), GOL (329 g m−2) and HOE (302 g m−2). A significant smaller fine root biomass was found at FLO (210 g m−2) which may be partly explained by the small amount of humus layer in our samples from FLO. A previous soil survey revealed different amounts of organic matter in the humus layer (e.g., 3.2 kg m−2 in our samples versus 9.7 kg m−2 in the soil survey at AOE, Table 1). Consequently, the determination of fine root biomass in the humus layers was not representative, considering the large deviations between the soil surveys and the soil cores in this study.
Chemical properties of fine roots
Mean element concentrations, C/N ratios and molar Ca/Al ratios of live fine roots from the four Norway spruce sites at AOE, FLO, GOL and HOE (n = 12). Different letters represent significant differences among the stands for the same depth. Global significant effects of soil depth are indicated by § for each site
Mg [mg g−1]
K [mg g−1]
Mn [mg g−1]
P [mg g−1]
Ca [mg g−1]
Al [mg g−1]
Ca/Al [mg g−1]
49.3 b, §
1.46 a, §
34 a, n.s
0.9 a, n.s
2.1 a, §
0.5 a, §
1.1 b, §
6.3 b, §
0.6 a, §
8.0 a, §
51.2 a, n.s.
1.51 a, n.s.
34 a, n.s
0.8 a, n.s
2.2 a, §
0.1 b, §
1.3 a, n.s.
7.1 a, §
0.9 a, §
6.0 a, §
50.7 a, §
1.48 a, n.s.
34 a, n.s
0.8 a, §
2.1 a, n.s.
0.2 b, §
1.0 b, n.s.
5.4 b, §
0.9 a, §
4.8 a, §
49.9 b, n.s.
1.68 a, §
30 a, n.s
0.7 a, §
2.0 a, §
0.6 a, §
1.2 ab, n.s.
5.7 b, §
0.5 a, §
8.5 a, §
The chemical analysis of live fine roots revealed no significant differences in the C/N ratio, N, Mg, and P content among the stands (Table 3). Hence, the large variation of inorganic N fluxes in soil solution (Table 1) had no effect on N concentrations of live fine roots. The site had a global significant effect (P < 0.05) on K concentrations of fine roots in the mineral soil, being largest at FLO and GOL. In the humus layer, K concentrations of fine roots were almost identical in all four stands.
The Ca/Al ratios of fine roots were not different among sites, but were significantly different (P < 0.05) between the humus layer and the mineral soil at all sites (Table 3). The highest Ca/Al ratios of 4.8–8.5 were found in the humus layer and lowest ratios in the mineral soil at 20–40 cm depth (0.5–1.1) mainly due to increasing Al concentrations and to a lesser extent by decreasing Ca concentrations. Neither liming (GOL) nor elevated contents of exchangeable Ca cations in the subsoil had an effect on the Ca/Al of fine roots. Moreover, the Ca/Al ratio of fine roots did not mirror the Ca/Al ratio of seepage in the humus layer or mineral soil (Table 1).
Morphological properties of fine roots
Root length density, specific root length, root tip density, specific root tip density and specific root area of fine roots from different soil depths in four Norway spruce stands. Different letters represent significant differences among the stands for the same depth. Global significant effects of soil depth are indicated by § for each site and each morphological parameter
Root length density (m l−1 soil)
Specific root length (m g−1 root)
Root tip density (tips l−1 soil)
Specific root tip density (tips g−1 root)
Specific root area (cm2 g−1 root)
17.5 bc, §
12.6 a, §
6822 a, §
4821 a, §
177 b, §
35.5 a, §
12.9 a, §
11424 a, §
4766 a, §
211 a, n.s.
11.6 c, §
9.0 b, n.s.
3963 b, §
3176 b, n.s.
143 c, n.s.
20.3 ab, §
10.8 a, §
7964 a, §
4518 a, §
167 b, n.s.
Root tip density of 3963 tips l−1 soil was smallest in the humus layer at GOL and significantly different to AOE (6822 tips l−1 soil), FLO (11424 tips l−1 soil) and HOE (7964 tips l−1 soil) (Table 4). Similarly, GOL had the lowest root tip density at the 0–10 cm depth, being significantly different to AOE and HOE. The lowest root tip density of 38 tips l−1 soil was found at the 20–40 cm depth in FLO. All stands showed a strong vertical decrease in root tip density, but the gradients varied among the sites.
Specific root tip density was significantly smaller in the humus layer at GOL as compared with the other sites. No differences were found in the mineral soil depths among the sites. In agreement with specific root length, a clear vertical decrease in specific root tip density from 4821 and 4518 to 1803 and 2143 tips g−1 roots, respectively, was only observed at AOE and HOE.
Specific root area was significantly different in the humus layer among all stands, except AOE and HOE. The highest values were found in FLO, considering all soil depths. Even the lowest specific root area of 201 cm2 g−1 root at 0–10 cm depth was higher than the maximum values of the other sites. Specific root area vertically decreased at AOE (P = 0.01) whereas soil depth had no effect at FLO, GOL and HOE. However, specific root area at FLO increased in the mineral soil from 201 to 327 cm2 g−1 root in the mineral soil.
We hypothesized that N availability in soil solution would have an impact on biomass, morphology and N concentration of fine roots in the four Norway spruce stands. Despite the large variation in inorganic N fluxes among the sites neither biomass and density nor morphological parameters of fine roots were related to N fluxes in throughfall or seepage. Furthermore, N concentrations of fine roots in the range between 1.3 and 1.7% (d.w. basis) were not different among the stands at same depths. A typical vertical gradient with highest N concentration in the humus layer was found at AOE and HOE whereas FLO and GOL had the maximum concentration in the mineral soil at the 0–10 cm depth. A vertical gradient in fine root N concentration reflects the N availability in the soil profile which is usually highest in the humus layer.
The N concentrations of fine roots at our sites are within the range reported for fine roots in other Norway spruce stands (1.1–2.1% d.w.) along a N deposition gradient from North to Central Europe (Högberg et al. 1998). Fine roots in the Swedish stand with throughfall input of only 1 kg N ha−1 yr−1 had the lowest N concentration of 1.1%, however, fine root N concentrations of the other three stands in Denmark, France and Germany (1.7–2.1%) did not follow the deposition gradient of 8–27 kg N ha−1 yr−1. A similar study in North American forests revealed a positive correlation between fine root N concentration and NO3- availability (Hendricks et al. 2000). The analysis included data from a N fertilization experiment at the Harvard Forest where fine root N concentration strongly increased from 1.4 to 2.3% in a Red pine stand following addition of N fertilizer. Fine root biomass appeared to decrease here in the humus layer of the high N fertilization plot (150 kg N ha−1 yr−1), but showed no response in the mineral soil (Magill et al. 2004). N-fertilization (ammonium-sulphate) of a Norway spruce stand in Sweden caused not only a strong decline in fine root biomass, but also a decrease in specific root length (Clemensson-Lindell and Asp 1995). By contrast, a general positive effect of N fertilization on fine root biomass, root production and N turnover of fine roots was observed in a N limited Norway spruce stand in northern Sweden (Majdi and Andersson 2005). According to Matzner and Murach (1995) fine root biomass of Norway spruce saplings decline above a threshold value of 2 mg N l−1 in soil solution. Inorganic N concentration in soil solution was mostly above this value in our four stands. Nevertheless, our results suggest that the variation in N deposition between 17 and 26 kg ha−1 yr−1 has no effect on fine root biomass and fine root N concentration and that high N deposition does not harm the vitality of fine root systems as indicated by the live/dead ratios in the stands at HOE. However, the fine root system may be affected in forests with low background deposition and limited N availability (Majdi and Andersson 2005). Experimental reductions in throughfall N input suggest an increase in fine root growth and biomass in coniferous forests (Boxman et al. 1998; Lamersdorf and Borken 2004).
The ratio of live to dead fine roots was almost identical in the humus layer of all four stands, but differed largely in the mineral soils. The vertical decrease in the live/dead ratio at FLO and GOL may be attributed to an accumulation of root necromass as a result of increased fine root production in the mineral soil. Low live/dead ratios may indicate stress conditions such as soil acidification, nutrient imbalances or water shortage (Puhe 2003). Godbold et al. (2003) explained the vertical decline in live/dead ratios of two spruce stands by low soil pH (H2O) (3.6 and 3.8 at 0–10 cm) since two other spruce stands with higher pH (H2O) (4.2 and 4.7 at 0–10 cm) showed almost no vertical differences in the live/dead ratio. Our results do not support these findings because variation in soil pH (CaCl2) was not related to live/dead ratio considering all mineral soil horizons. By contrast, the molar Ca/Al ratio in soil solution that has been identified as a stress indicator was higher in AOE and GOL (both 1.0 at 20 cm depth) than in FLO and HOE (0.1 and 0.2 at 20 cm depth). Physiological stress may be expected when the Ca/Al ratio in soil solution falls below 1.0 (Cronan and Grigal 1995). These authors also estimated a 50% risk of Al toxicity at a molar Ca/Al ratio of 0.2 in fine roots, a value that was not reached at our sites.
The remarkably high live/dead ratios in the mineral soils at AOE and HOE might have other reasons than slow fine root turnover. The positive correlation between live/dead ratio and Mn concentrations of live fine roots point to enhanced decay rates of dead fine roots in AOE and HOE. Manganese peroxidase is an enzyme that catalyses the degradation of lignin which is often an initial step in the decomposition of woody debris. This initial decomposition phase is of particular importance for the quantification of dead fine root mass because only almost intact root fragments were ascertained by the sample preparation. The positive relationship between initial Mn concentrations and mass loss of needle litter (Berg et al. 1996) might also be applicable for mass loss of fine roots. In another study, the addition of birnessite, a Mn oxide, increased the carbon loss of beech litter by 10% during an incubation experiment over 500 days (Miltner and Zech 1998). However, not only the content but also the availability of Mn in the soil may affect the decomposition of dead fine roots. Genenger et al. (2003) reported decreasing Mn concentrations in fine roots following addition of wood ash to the soil, indicating decreasing Mn availability with increasing pH. Moreover, microorganisms may positively influence the Mn availability due to the release of siderophores (Parker et al. 2004).
To our knowledge there is no study available that has investigated the specific role of Mn in the degradation of root necromass. Because all other chemical properties of fine roots were not different among the stands, Mn might promote at least the decay of root necromass in the mineral soil. It remains unclear, however, why there was no effect of Mn concentration on the live/dead ratio in the humus layer. Possibly, not the Mn concentration of root litter per se is relevant for the decay, but the availability of Mn cations in soil solution. Another explanation for the differences in the live/dead ratio might the slightly higher soil temperatures of approximately 1.5–2.0°C at AOE and HOE. It would explain a higher microbial activity, and thus, a faster decay of fine root detritus at these sites. However, similar to the Mn concentration, differences in soil temperature had no effect on the live/dead ratios in the humus layer.
Fine root densities in the humus layer were not different among the sites, suggesting that the differences in fine root biomass resulted from the enormous spatial variation in the amount of the humus layer. The comparison between the soil survey and the soil cores from this study (Table 1) indicates a large potential of misjudgement in the determination fine root biomass in humus layers. An extrapolation of fine root biomass using the site specific amount of humus layer from soil surveys might provide better estimates than without correction. In mineral soils with high rock and gravel contents or shallow soils fine root density may increase as a result of less space for the root system to proliferate in the soil. Fine root biomass might be overestimated in such soils when rock and gravel fractions as well as the soil depth are not considered in the analysis.
According to the vertical pattern in fine root density, the humus layer is the preferential horizon of fine root growth. The availability of nutrients is mostly highest in the humus layer due to mineralization of soil organic matter and the inputs of nutrients by throughfall. Competition between Norway spruce and other plants, mainly grass species, and microorganisms requires a strong presence of fine roots in the humus layer. Hence, not only the thickness or amount of the humus layer, but also the availability of nutrients may affect the vertical pattern in fine roots in the soil profile as it has been shown by fertilisation experiments (Clemensson-Lindell and Persson, 1995). The vertical gradient of fine root density indicate favourable soil conditions in the mineral soil at AOE as compared to the other sites. Fine roots proliferate the mineral soil more intensively although AOE was only significantly different to FLO and GOL at the 0–10 cm depth and to FLO at 20–40 cm depth. Contrary, fine root density and fine root biomass point to less favourable conditions in the mineral soil at FLO. The influence of chemical soil properties on vertical pattern of fine root density or biomass has been controversial discussed in the past decades. A study by Jentschke et al. (2001) suggests that soil acidity alters the pattern of fine roots in Norway spruce stands by increasing biomass in the humus layer and decreasing biomass in the mineral soil. However, lime addition has a similar, at least transitional effect on the vertical pattern of fine roots because of increasing nutrient availability in the humus layer (Murach and Schünemann 1985). The lime addition at GOL in 1983 had obviously no long-lasting effect on the vertical pattern of fine root density and biomass despite the elevated pH (CaCl2) in the humus layer.
Mean element concentrations of green needles from Norway spruce in four study sites
Site (reference year)
P [mg g−1]
Ca [mg g−1]
Mg [mg g−1]
K [mg g−1]
Mn [mg g−1]
Fine root morphology was mostly significantly different in the humus layer at GOL, indicating that the lime application had a long-lasting effect on morphological properties. Root length density (11.6 m l−1 soil), specific root length (9.0 m g−1 root), root tip density (3963 tips l−1 soil), specific root tip density (3176 tips g−1 root) and specific root area (143 cm2 g−1 root) were smaller than in the other three stands. In contrast to other studies (Puhe 2003; Godbold et al. 2003), this cannot be attributed to chemical stress conditions. It is more likely that increasing root length and higher number of root tips indicate low nutrient availability in order to compensate nutrient deficiencies. In agreement with our results, Püttsepp et al. (2006) reported lower specific root length of 6–10 m g−1 root in the top 5 cm of the soil following application of wood ash in a Norway spruce stand. Small specific root length may indicate a slow rate of root proliferation in the soil and a disadvantage in the acquisition of nutrients or water relative to the cost of resources used for the maintenance and construction of roots (Eissenstat 1991). However, in humus layers with favourable nutrient conditions small specific root length might be no disadvantage for the tree.
With a few exceptions there were no significant differences in the morphology of fine roots in the mineral soil among the stands. Specific root length decreased with soil depth at AOE and HOE whereas an opposite pattern or no significant gradient was found in the soil at FLO and GOL, respectively. The vertical gradients are conform with the vertical gradients in the Ca/Al ratio of soil solution, suggesting that specific root length is possibly associated with the gradient in base saturation. In addition to soil chemical conditions, specific root length also may be affected by the pore system and density of the mineral soil. High porosity and low bulk density allow a faster proliferation, and thus, better access to nutrients and water. According to Puhe (2003) long roots predominate the explorative growth of the root system. Similarly, specific root tip density continuously decreased with depth at AOE and HOE while an increase or no change was observed at FLO and GOL, respectively. The stand at FLO contradict earlier observations that specific root length and specific root tip density decrease with increasing soil acidification or low base saturation in the subsoil.
In conclusion, N concentration, biomass and morphological properties of fine roots of Norway spruce were not affected by differences in solute inorganic N fluxes. Our results suggest that the variation of throughfall N input in the range of 17–26 kg N ha−1 yr−1 as reported for many Central European forests does not harm fine root systems of Norway spruce. The variation in biomass, density and morphology of fine roots was rather small considering the differences in soil properties and climate among the stands. An exception was a limed stand that showed distinct morphological properties of fine roots in the humus layer, indicating a long-term effect of lime addition on fine root morphology of Norway spruce. Our results suggest that specific root length and specific root tip density do not always decrease under adverse soil chemical conditions. Moreover, it remains unclear if the root system of Norway spruce in the top soil may profit from high base saturation in the subsoil because Al, Mg and Ca concentration of fine roots were not affected in these stands. The Mn concentration of soils may play an important role in terms of both root and needle litter decomposition.
We thank Rita Süss, Andreas Puhr, Sabine Horvarth, Kristin Strobel, and Regina Gerlinger for their assistance in the preparation of roots. We like to thank Stephan Raspe and Christoph Schulz from the Bayerische Landesanstalt für Wald und Forstwirtschaft and Wendelin Weis from the Technical University of Munich for providing element fluxes in throughfall and seepage of the four study sites. We are thankful to the members of the Central Analytical Laboratory of the Bayreuth Center of Ecology and Environmental Research (BayCEER) for chemical analysis of root samples. Financial support came from the Bavarian State Ministry of Agriculture and Forestry.