SEARCH

SEARCH BY CITATION

Keywords:

  • ammonium toxicity;
  • germination;
  • plant diversity;
  • soil column experiments;
  • survival

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Restoration of formerly species-rich wet heaths and matgrass swards has not always been successful. The constraints on this restoration process are not yet fully understood and need further investigation, particularly the accumulation of ammonium in the soil after sod cutting, i.e. the removal of the vegetation and topsoil layer. This accumulation is known from sod cutting experiments in dry heaths, but had not previously been studied in wet heaths and matgrass ecosystems.
  • 2
    In 2000, sods were cut from two degraded Dutch wet heaths. Soil chemistry and germination in the sod-cut plots were measured at irregular intervals between April 2000 and August 2001. To test the influence of ammonium on germination and survival, a glasshouse dose–response experiment was conducted with two endangered wet heath plant species.
  • 3
    In both wet heaths, an accumulation of KCl-extractable ammonium up to 600 µmol kg−1 dry soil was found in the upper 10 cm of the soil within the first year after sod cutting. These high ammonium concentrations lasted for about 10 months. Germination was very low in the sod-cut plots in 2000 and 2001, and few target species were found, although they were present in the vicinity.
  • 4
    The dose–response experiment indicated a significant, negative correlation of both germination and survival with increasing ammonium addition for both plant species. Mean soil ammonium concentrations of the control, 100 and 250 µm ammonium treatments were significantly lower than those of the 500 and 1000 µm ammonium treatments (47, 45, 70, 144 and 252 µmol kg−1 dry soil, respectively).
  • 5
    Maximum concentrations of KCl-extractable ammonium in the field corresponded to water-extractable concentrations that were higher than those found to be limiting germination and growth in the glasshouse experiments. The low germination in the field is likely to have been adversely affected by high concentrations of ammonium as a result of sod cutting.
  • 6
    Synthesis and applications. High ammonium concentrations occur in wet heaths following sod cutting. Low rates of germination of restoration target plant species occur under such conditions. To increase the success of wet heath restoration, the accumulation of ammonium after sod cutting should be prevented by additional measures, such as liming. Because sod cutting is also applied as a restoration measure in the restoration of other ecosystems, such as fens, the effects on increased soil ammonium concentrations need further attention.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Heathlands were once widespread in Western Europe, but their total distribution area and species composition have declined considerably. Therefore, heathlands are regarded as an internationally endangered habitat type of high conservation value (Pitcairn, Fowler & Grace 1991; Gimingham 1992; Webb 1998). In particular, the species composition of wet heath communities (Ericion tetralicis) and related wet matgrass swards (Nardo-Galion saxatilis) have changed drastically (e.g. Sansen & Koedam 1996). In the Netherlands, these species-rich ecosystems used to be common on sandy Pleistocene deposits. The soils were oligotrophic, slightly buffered (within the cation buffer range) and had a relatively low ammonium to nitrate ratio (Schaminée, Van’t Veer & Van Wirdum 1995; Schwertz, Schaminée & Dijk 1996). In many areas almost all of the endangered (target) species, such as Cirsium dissectum, Dactylorhiza maculata, Drosera intermedia, D. rotundifolia, Epipactis palustris, Gentiana pneumonanthe, Lycopodiella inundata, Narthecium ossifragum, Parnassia palustris, Pedicularis sylvatica, Rhynchospora alba, R. fusca and Succisa pratensis, have now disappeared from these ecosystems (Roelofs et al. 1996; Bobbink et al. 1998a). This decline in species diversity has been attributed to increased soil acidification as a consequence of atmospheric deposition and lowering of the groundwater table (Van Breemen et al. 1982; Houdijk et al. 1993; Roelofs et al. 1996; Bobbink, Hornung & Roelofs 1998b; Roem, Klees & Berendse 2002). Furthermore, enhanced soil nitrogen concentrations due to atmospheric deposition have resulted in the dominance of grasses over the characteristic wet heath species (Heil & Diemont 1983; Aerts & Berendse 1988; Pitcairn, Fowler & Grace 1991; Houdijk et al. 1993).

At the end of the 1980s, recognition of the deterioration of wet heaths and matgrass swards led to the introduction of restoration measures in the Netherlands. These measures aimed to restore oligotrophic and weakly buffered soil conditions to enable successful germination and establishment of the endangered wet heath and wet matgrass sward species. Examples of these restoration measures are sod cutting, which is the removal of vegetation and the topsoil layer to remove excess nutrients (Bakker 1989; Snow & Marrs 1997; Mitchell et al. 2000), and restoration of the former hydrology to increase the influence of groundwater (Roelofs et al. 1996).

In a few cases, these management practices have resulted in the restoration of species-rich wet heaths. Soil buffering capacity and pH were effectively restored, enabling endangered plant species to return to these sites or to increase their abundance (Jansen, De Graaf & Roelofs 1996; Roelofs et al. 1996). However, in many more instances these measures have resulted in species-poor dwarf shrub vegetation with few endangered plant species present. Field observations have led to the identification of possible causes. In areas where discharging groundwater had been acidified, as a result of acidification and cation depletion of the surrounding catchment areas, soil buffering capacity could not be restored. Suitable soil conditions could not be created through restoration measures (Roelofs et al. 1996). The removal of seeds from the soil seed bank by sod cutting might have been another constraint, because most seeds are concentrated in the upper layers of the soil (Putwain & Gillham 1990; Pywell, Putwain & Webb 1997). In this paper we consider a third constraint: increased soil ammonium concentrations after sod cutting, possibly up to toxic levels for endangered wet heath species. This phenomenon has been found in dry heaths (De Graaf et al. 1995, 1998b), but has not previously been studied in wet heath areas.

Several processes may cause the accumulation of ammonium after sod cutting. First, sod cutting usually does not remove all of the soil organic material. The remaining roots, in which retranslocation of nitrogen cannot take place, have a low carbon : nitrogen ratio and are readily decomposed. As a result, mineralization rates might initially be enhanced after sod cutting (Berendse 1990). Secondly, the uptake of ammonium by plants is eliminated through removal of the vegetation by sod cutting. Thirdly, atmospheric deposition of ammonium is high in the Netherlands (Bobbink & Heil 1993; Erisman, Bleeker & Van Jaarsveld 1998). Finally, it has been suggested that sod cutting removes nitrifying bacteria, which are mainly present in the topsoil layer (Troelstra, Wagenaar & De Boer 1990). The conversion of ammonium into nitrate by these micro-organisms might therefore be considerably reduced. Because of these processes, the ammonium input into the soil is increased, whereas the uptake of ammonium and conversion into nitrate are reduced, resulting in the accumulation of ammonium in the topsoil layer.

It has been shown in hydroculture experiments that high ammonium concentrations have negative effects on the germination and survival of several endangered heath species (De Graaf et al. 1998a; Lucassen et al. 2003). However, the response of such species to high ammonium concentrations has never been investigated in soil column or field experiments. Research may provide insights into the constraints on the current restoration measures in wet heaths and matgrass swards.

This paper describes the results of a field study to determine whether ammonium accumulates in the soils of wet heaths after sod cutting. In addition, effects of increased ammonium concentrations on the germination and survival of seeds of wet heath species were examined using field and soil column experiments.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

study areas

The nature reserve ‘Leemputten’, in the Municipality of Ermelo, is situated on the Veluwe in the centre of the Netherlands, about 8 km east of Ermelo (52°17′N, 5°44′E). Sandy deposits overlay layers consisting of loam from the Riss glacial period. Large variation in soil conditions, and consequently in species composition, can be found due to the former human use of these loam layers. Species-rich communities like Ericetum tetralicis subass. orchietosum, Gentiano pneumonanthes–Nardetum and Cirsio dissecti–Molinietum are intermingled with species-poor vegetation in which Calluna vulgaris, Erica tetralix and Molinia caerulea are dominant (nomenclature follows Schaminée, Van’t Veer & Van Wirdum (1995) and Swertz, Schaminée & Dijk (1996) for plant communities, and Van der Meijden (1996) for vascular plants). The current management of this area consists of annual mowing of the grasslands and irregular sod cutting of the heaths.

The heathland area ‘Havelte-Oost’ is owned by the Ministry of Defence and is located approximately 65 km to the north of Leemputten, 8 km north of Meppel (52°48′N, 6°13′E). The topsoil layer consists of sandy deposits with low loam content, although shallow loam layers are also present. The vegetation types found in Leemputten, except for Cirsio dissecti–Molinietum, can be found here too, as well as species-rich, dry Nardo–Galion saxatilis communities (De Graaf et al. 1994). Since 1990, large areas have been irregularly treated by sod cutting or mown and are now dominated by Erica tetralix. By contrast, wetter parts of Havelte-Oost have not been treated by sod cutting, and are now dominated by Molinia caerulea.

Both areas are characterized by low pH (4·2–4·5), slightly buffered soil conditions, low aluminium : calcium ratios and very low soil nitrate concentrations (Table 1). Strikingly, the Erica tetralix-dominated zone has lower base cation and nitrogen concentrations than the other areas. The carbon : nitrogen ratios are between 21 and 30, and are higher in the Erica tetralix-dominated zone compared with the Molinia caerulea-dominated areas.

Table 1.  Soil conditions at the start of the experiment of the untreated humus layers (FH) and upper 5 cm of the mineral soil (M) layers of Leemputten and Havelte-Oost. In Havelte-Oost, there were Erica- and Molinia-dominated zones, whereas the plots in Leemputten were all dominated by Molinia. Water- and 0·2 m KCl-extractable concentrations are indicated by single and double asterisks, respectively. Ion concentrations are presented in µmol kg−1 dry soil
 LeemputtenHavelte-Oost
Molinia-dominatedMolinia-dominatedErica-dominated
FHMFHMFHM
pH-(H2O)   4·21 (0·02)   4·25 (0·02)   4·38 (0·06)   4·41 (0·09)   4·36 (0·04)  4·44 (0·02)
Ca** + Mg** + K*6646 (1118)4050 (1094)8170 (1081)4776 (2471)2762 (255)883 (150)
Al* 272 (54) 173 (22) 285 (20) 231 (32) 242 (8)147 (9)
Al*/Ca**   0·1 (0·02)   0·09 (0·03)   0·07 (0·02)   0·14 (0·04)   0·16 (0·03)  0·47 (0·20)
NH4** 357 (114) 116 (37) 643 (79) 156 (86)  72 (71) 50 (17)
NO3*  41 (26)  32 (18)  41 (31)  33 (33)  19 (19)  0
PO4*  13 (3)   9·8 (2·4)   6·6 (1·9)   4·09 (2·08)   2·4 (0·5)  0·97 (0·29)
Soil moisture (%)  54 (3)  42 (5)  57 (4)  42 (4)  53 (5) 30 (3)
Ntot (%)   0·31 (0·06)   0·21 (0·03)   0·43 (0·09)   0·22 (0·04)   0·27 (0·05)  0·12 (0·01)
Ctot (%)   7·3 (1·2)   5·5 (0·8)   9·1 (1·3)   5·7 (0·8)   7·3 (1·7)  3·5 (0·3)
C : N (%/%)  24·7 (2·1)  27·9 (3·1)  21·7 (1·3)  26·1 (1·7)  26·4 (1·8) 30·2 (2·3)

soil sampling and analysis

In spring 2000, permanent plots were randomly laid out in both study areas. Havelte-Oost was divided into two zones on the basis of the dominance of Molinia caerulea or Erica tetralix. These zones will be referred to as Molinia- and Erica-dominated zones, respectively. In these zones sods were cut in May 2000 from four permanent plots of approximately 2 × 3 m. In Leemputten, sods were cut in April 2000 from six Molinia-dominated plots of 2·5 × 2·5 m. Sod cutting was carried out by the local Nature Conservation agencies. In both areas a corresponding number of untreated control plots were established also.

Three soil samples of the upper 10 cm of the soil were collected (auger diameter 2·5 cm) in each of the sod-cut and untreated plots from April 2000 until August 2001. All samples were divided into two and three layers, respectively (Table 2). The mean depth of the soil layer removed by sod cutting was about 7 cm and comprised most of the humus layer (FH). Access restrictions due to Foot and Mouth Disease prevented sampling in March and April 2001.

Table 2.  The soil samples of the sod-cut plots were divided into two layers of 5 cm each, whereas the soil samples of the untreated plots contained an extra soil layer (FH) that had not been treated by sod cutting
Soil layerUntreated plotSod-cut plot
1upper 7 cm (FH)
20–5 cm0–5 cm
36–10 cm6–10 cm

The soil samples were transported to the laboratory in a cool box, stored at 4 °C and processed within 2 days. After homogenizing the three samples of each plot, 15 g of fresh soil was extracted on a rotary shaker (100 r.p.m.) for 1 h with 100 mL 0·2 m KCl or demineralized water. The soil suspensions were centrifuged for 5 min at 4000 r.p.m. Supernatants were filtered through a Whatman GF/C filter and stored at −20 °C until further analysis. Soil pH was determined in the remaining soil solutions. Soil moisture content was measured after drying 15 g of fresh soil at 105 °C for 24 h. Ca, Mg, Al, K, NH4, NO3, PO4 and SO4 concentrations were analysed calorimetrically using a continuous-flow analyser (Skalar 40, Skalar Analytical BV, Breda, the Netherlands).

groundwater sampling

In Havelte-Oost, piezometers with a depth of 1 m were inserted in both the Molinia- and the Erica- dominated zones. In Leemputten, the piezometers were installed at three locations between the sod-cut plots. The level of the groundwater tables was measured on the same dates as the soil samples were collected.

germination and survival experiments

To test the effects of increased ammonium addition on the germination and survival of two endangered wet heath species, 100 PVC cylinders (diameter 15 cm, height 20 cm) were filled with soil material originating from a heathland reserve (pH-H2O = 4·3). The cylinders were put in an open glasshouse in which the temperature and light regime were not controlled and were subjected to outdoor conditions, except rainfall.

Twenty seeds of Cirsium dissectum were placed on the soil surface of 25 cylinders and 20 seeds of Succisa pratensis were placed in a second series of 25 cylinders. Germination of these seeds was investigated in a dose–response experiment of 82 days, starting on 30 June 2000. Five concentrations of (NH4)2SO4, namely 0 (control treatment), 100, 250, 500 and 1000 µmol L−1, were sprayed on the soil columns three times a week. Each treatment had five replicates that were randomly distributed in the glasshouse. The total amount of added solution corresponded to the average yearly rainfall in the Netherlands, namely 800 mm.

On the soil surface of each of the other 50 cylinders two seedlings of either Cirsium dissectum or Succisa pratensis were placed. These seedlings had been grown at room temperature on wetted filter paper and were transferred to the cylinders after 4 weeks. The same five (NH4)2SO4 treatments were applied to these cylinders. Each treatment had five replicates, resulting in a total of 10 seedlings per treatment. Survival was recorded for both species for a period of approximately 10 weeks. In the 0 and 1000 µmol L−1 (NH4)2SO4 treatments of Succisa pratensis fewer seedlings could be used. In the 0 µmol L−1 treatment, one of the five replicates contained only one seedling (resulting in a total of nine seedlings). In the 1000 µmol L−1 treatment two replicates contained one seedling (eight seedlings in total for this treatment).

In the top 10 cm of each soil column, one Rhizon soil moisture sampler was installed. Soil moisture was recorded five times during the experiment. The ammonium concentrations in the soil moisture samples were measured with the continuous-flow analyser. Based on the soil moisture contents, these ammonium concentrations were expressed as µmol kg−1 dry soil, to be able to compare them with the ammonium concentrations in the field.

Germination of species was quantified in the field as well. During two growing seasons, the presence and cover of seedlings were visually estimated in a 1 × 1 m subplot within each sod-cut plot according to the Braun–Blanquet approach (Westhoff & Van der Maarel 1978).

statistical analysis

Statistical tests were carried out using the Statistical Program for the Social Sciences 8·0 (SPSS Inc. 1989–1997). All data were tested for normality using a one-sample Kolmogorov–Smirnov test. The data of soil ammonium concentrations were logarithmically transformed to stabilize variances between groups. If Repeated Measures tests (General Linear Model) detected significant interactions between time and treatment, one-way anova tests were carried out for each sample date separately.

Data from the germination experiment revealed an interaction between time and treatment. Therefore, a one-way anova was carried out on the data at the end of the experiment. Post-hoc Tukey tests were carried out to test for significant differences between treatments at that date. The significance of (linear) correlations between germination and addition of (NH4)2SO4 was tested by calculating Pearson correlation coefficients if data were normally distributed. Spearman's rank correlation coefficients were calculated for data that did not fit normality.

The survival of both plant species during the experimental period was tested using Kaplan–Meier tests. Nonparametric log rank statistics were used to compare differences in survival distributions between the treatments.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

soil ammonium concentrations

In Leemputten, no significant differences in ammonium concentrations between the sod-cut and control plots were found 2 weeks after sod cutting (Fig. 1). In the subsequent months, soil ammonium concentrations in both layers of the sod-cut plots increased strongly and reached maximum values of 572 and 327 µmol kg−1 dry soil at the end of February 2001, for the 0–5 cm and 6–10 cm layers, respectively. At that time, the differences in ammonium concentrations of the 0–5 cm and 6–10 cm layers between sod-cut and control plots were highly significant (P < 0·001). The ammonium concentrations of the untreated vegetation did not show such an increase over time. Significant differences in ammonium concentrations between the sod-cut plots and the untreated plots were found from June 2000 until May 2001.

image

Figure 1. Ammonium concentrations in the upper soil layers of the sod-cut plots and untreated vegetation in Leemputten (n = 6) from April 2000 to August 2001. Vertical bars indicate SE. Open squares = sod-cut 0–5 cm, filled squares = sod-cut 6–10 cm, open circles = untreated 0–5 cm, filled circles = untreated 6–10 cm. Sod cutting took place on 12 April 2000.

Download figure to PowerPoint

In the Molinia-dominated zone of Havelte-Oost the same differences in ammonium concentrations between the soil of sod-cut and control plots were found (Fig. 2). Within 2 months, ammonium concentrations of the 0–5 cm soil layers of the sod-cut plots reached their highest mean value (558 µmol kg−1 dry soil), which was significantly higher than those of the untreated plots (P = 0·018). In November 2000 a sharp decline in ammonium concentrations was recorded, possibly because of dilution by high rainfall. A second high value was found in April 2001, 11 months after sod cutting, although this difference was not significant. The same pattern was observed in the 6–10 cm soil layer, although the increase in ammonium concentration was recorded 2 months later than in the 0–5 cm soil layer. The highest ammonium concentration in this sod-cut soil layer was found in April 2001 (496 µmol kg−1 dry soil), which was the only significant difference compared with the control treatments. From April 2001 onwards, soil ammonium concentrations of the sod-cut and untreated plots were the same.

image

Figure 2. Ammonium concentrations in the sod-cut plots and untreated vegetation of the Molinia-dominated zone of Havelte-Oost (n = 4) from May 2000 to August 2001. Vertical bars indicate SE. Open squares = sod-cut 0–5 cm, filled squares = sod-cut 6–10 cm, open circles = untreated 0–5 cm, filled circles = untreated 6–10 cm. Sod cutting took place on 19 May 2000.

Download figure to PowerPoint

In the Erica-dominated zone of Havelte-Oost all soil ammonium concentrations were much lower compared with the Molinia-dominated zone (Fig. 3). The ammonium concentrations of both sod-cut soil layers were generally higher compared with those of the control treatments in this zone. In the upper soil layers of the sod-cut plots the highest ammonium concentration (109 µmol kg−1 dry soil) was again found within 2 months of sod cutting. It was significantly higher than that of the 0–5 cm layers of the untreated plots (P = 0·01). Significantly higher concentrations were also found in September 2000 and April 2001. The 6–10 cm soil layers showed a delayed increase in ammonium concentration. Here the highest concentration was found in April 2001, almost 1 year after sod cutting. At that time, the difference in ammonium concentrations between the 6–10 cm layers of the sod-cut and control plots was highly significant (P = 0·002), which was also the case for the subsequent sampling dates.

image

Figure 3. Ammonium concentrations in the sod-cut plots and untreated vegetation of the Erica-dominated zone of Havelte-Oost (n = 4) from May 2000 to August 2001. Vertical bars indicate SE. Open squares = sod-cut 0–5 cm, filled squares = sod-cut 6–10 cm, open circles = untreated 0–5 cm, filled circles = untreated 6–10 cm. Sod cutting took place on 19 May 2000.

Download figure to PowerPoint

groundwater table

The groundwater table of Leemputten and Havelte-Oost showed different patterns over time (Fig. 4). The groundwater table in the Leemputten area was more or less stable and ranged from 10 to 2 cm below soil surface during summer and spring, respectively. In Havelte-Oost much larger fluctuations in groundwater tables were found. The Molinia- and Erica-dominated zones showed similar fluctuations. At the beginning of the summers of 2000 and 2001 the groundwater tables were low, about 90 cm below soil surface. By contrast, groundwater levels were approximately 10 cm below soil surface in early autumn and winter.

image

Figure 4. Groundwater tables of Havelte-Oost and Leemputten in centimetres below soil surface from June 2000 to December 2001. Filled and open circles = Molinia- and Erica-dominated zone of Havelte-Oost, respectively; open triangles = Leemputten.

Download figure to PowerPoint

germination in the field

The maximum plant cover by seedlings originating from the seed bank or from surrounding plants was only 38% (Leemputten, August 2001). In Havelte-Oost these values were only 13% and 4% for the Molinia- and Erica-dominated zones, respectively.

The number of species that germinated in the sod-cut plots during two growing seasons was small (Table 3). The largest number of species was found in Leemputten (seven). In both areas less than two target species were found, even though a large number of target species were present in the vicinity of the sod-cut plots. The largest numbers of seedlings were recorded for the more common species such as Erica tetralix, Molinia caerulea, Juncus bulbosus and Phragmites australis.

Table 3.  Mean number of species (minimum − maximum), target species and cover percentages (minimum − maximum) of the sod-cut plots of Leemputten (n = 6) and Havelte-Oost (n = 4 in each zone) in August 2001. The total cover is the percentage of the 1 × 1 m subplots that is covered by vegetation. Furthermore, the frequency of occurrence of the five most common species is indicated as a percentage of total plots
 LeemputtenHavelte-Oost
Molinia-dominatedErica-dominated
Total cover (%) 36 (5–60) 13 (2–25)  4 (1–10)
Number of species
 Total  7 (4–10)  5·5 (5–7)  3·5 (3–4)
 Target  2  1·75  0·5
Molinia caerulea100100100
Erica tetralix100100 50
Drosera intermedia 83 25100
Phragmites australis100 50  0
Juncus bulbosus 50 75  0

germination and survival experiments

The cumulative germination percentage of Cirsium dissectum seeds in the control treatment was 68% at the end of the experiment. This rate was significantly higher than that of the 1000 µm ammonium treatment (Fig. 5). The 100, 250 and 500 µm treatments had intermediate germination percentages and did not differ significantly from the others. However, at the end of the experiment, a highly significant negative correlation between germination of Cirsium dissectum and ammonium concentration was found (Pearson, r = −0·649, P < 0·001).

image

Figure 5. The cumulative germination percentage of Cirsium dissectum in the glasshouse in relation to increasing addition of ammonium (as µmol L−1 (NH4)2SO4). Significant differences between treatments are indicated by different letters.

Download figure to PowerPoint

The germination of Succisa pratensis was in general very low and did not exceed 15% (data not shown). The control treatment had the highest germination (15%), followed by the 250, 100 and 1000 µm treatments (respectively 14, 5 and 5% germination), whereas in the 500 µm NH4 treatment no seeds germinated at all. At the end of the experiment a significant negative correlation was found between germination and ammonium concentration (Spearman's rho = −0·436, P = 0·015).

There were significant differences in survival percentages of Cirsium dissectum and Succisa pratensis seedlings between the treatments (Figs 6 and 7). For Cirsium dissectum, the 0 and 100 µm ammonium treatments had the highest survival percentages, 60 and 50% respectively, compared with the other treatments. However, these differences were only significantly higher than that of the 1000 µm ammonium treatment (P = 0·023 and 0·016 for 0 and 100 µm, respectively).

image

Figure 6. Survival (%) of Cirsium dissectum seedlings in relation to increasing addition of ammonium (as µmol L−1 (NH4)2SO4). Significant differences between treatments are indicated by different letters.

Download figure to PowerPoint

image

Figure 7. Survival (%) of Succisa pratensis seedlings in relation to increasing addition of ammonium (as µmol L−1 (NH4)2SO4). Significant differences between treatments are indicated by different letters.

Download figure to PowerPoint

A similar relationship was found for survival of plants of Succisa pratensis. There was a tendency for survival to decrease with increasing addition of ammonium (Fig. 7). The survival of control plants was significantly higher than those of the 500 and 1000 µm treatments (P = 0·007 and 0·039, respectively), although no other significant differences between treatments were found.

The ammonium concentrations in the soil moisture of the columns, averaged over the total duration of the experiment, differed significantly between the treatments (data not shown). The highest ammonium concentration was found in the 1000 µm treatment (252 µmol kg−1 dry soil), which was significantly higher than that of the 500 µm treatment (144 µmol kg−1 dry soil, P = 0·000). Both treatments had higher ammonium concentrations than those of the control, 100 and 250 µm treatments (47, 45 and 70 µmol kg−1 dry soil, respectively; P = 0·000).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Experimental sod cutting of two degraded Dutch wet heaths resulted in a strong accumulation of ammonium in the upper 10 cm of the soil. The soil ammonium concentrations increased for a period of 10–12 months and reached maximum values of about 570 µmol kg−1 dry soil in both Leemputten and the Molinia-dominated zone of Havelte-Oost. In the untreated vegetation the ammonium concentration did not exceed 150 µmol kg−1 dry soil. Such an accumulation of ammonium has previously been found in dry heaths (De Graaf et al. 1995, 1998b), but has never been measured in wet heathland ecosystems.

The accumulation of ammonium after sod cutting can be attributed to several factors, such as atmospheric deposition of ammonium and the absence of uptake by plants. The temporary decrease of the carbon : nitrogen ratio of the soil organic matter, as a result of high nitrogen contents of remaining roots after sod cutting, will also contribute to increased mineralization rates (Berendse 1990). Another important factor might be the decreased nitrification. Because nitrifying bacteria are mainly present in the topsoil layers (Troelstra, Wagenaar & De Boer 1990), they may be removed by sod cutting. Preliminary results of soil incubation experiments showed no nitrification in soil of sod-cut plots, in contrast to soil of untreated plots (our unpublished data). The decrease in ammonium concentrations of the sod-cut plots, which was found in both areas after about 10 months, might then be explained by the recovery of the populations of nitrifying bacteria. However, future experiments are necessary to elucidate the effect of sod cutting on the nitrification process.

In the Molinia-dominated zone of Havelte-Oost a sharp decline in the ammonium concentration of the sod-cut plots was found in November 2000. Meteorological data for September–November indicated that total precipitation during this period was much higher compared with long-term average values (260 and 222 mm, respectively, KNMI 2001). These high monthly precipitations were also reflected in the height of the groundwater table in Havelte-Oost. It is speculated that soil ammonium concentrations of the sod-cut plots were diluted, resulting in the sharp decrease. In Leemputten, the groundwater table was higher compared with Havelte-Oost and was hardly influenced by rainfall fluctuations. Soil ammonium concentrations were therefore not affected by changes in the groundwater table.

The difference in ammonium concentrations between the Molinia- and Erica-dominated zones of Havelte-Oost is interesting. In the latter, soil ammonium concentrations were much lower in the sod-cut plots as well as in the untreated vegetation. In this zone, sod cutting resulted in an increase in ammonium concentration, but to a much smaller extent. This can be explained by the difference in dominant plant species between these zones. Increasing dominance of Molinia is known to increase the flow of carbon and nutrients into the litter fraction of the soil (Aerts & Berendse 1988; Berendse 1990). The percentage of total carbon in the soil of the Molinia-dominated zone was indeed higher than that of the Erica-dominated zone (4·2 and 3·3%, respectively). In addition, the carbon : nitrogen ratios of the soil in the Molinia-dominated zone were lower than those of the Erica-dominated zone (on average 26 and 31, respectively). Because decomposition increases with decreasing carbon : nitrogen ratio, mineralization rates in the Molinia-dominated zone will be higher compared with the Erica-dominated zone (Berendse 1990; Van Vuuren et al. 1992). As a result, the ammonium accumulation after sod cutting was also higher in the zone dominated by Molinia.

KCl-extractable soil ammonium concentrations as high as 570 µmol kg−1 dry soil, found after sod cutting in both degraded wet heaths, are likely to hamper the restoration of formerly species-rich wet heathlands and matgrass swards (Gigon & Rorison 1972; De Graaf et al. 1998a; Lucassen et al. 2003). To compare the field ammonium concentrations with those of the glasshouse experiments, we used the concentrations of the extractions with demineralized water instead of those of the 0·2 m KCl extractions. The patterns of the water-extractable ammonium concentrations in both wet heaths areas were the same as in Figs 1–3, because a strong positive correlation between the ammonium concentrations of these methods existed (NH4–demi = 0·468 × NH4–KCl + 8·481; R2 = 0·789, P = 0·000, n = 493). It is clear that the highest water-extractable ammonium concentrations reached in the field (namely 275 µmol kg−1 dry soil) were above the 251 µmol kg−1 dry soil of the glasshouse 1000 µm treatment, at which significant negative effects on the germination and survival of Cirsium dissectum and Succisa pratensis were found. Furthermore, Lucassen et al. (2003) showed in hydroculture experiments that Cirsium dissectum could tolerate ammonium concentrations up to 250 µmol L−1 when the soil pH was relatively high (pH 6). However, at pH 4, detrimental effects of ammonium on growth and survival of this species were found at concentrations as low as 100 µmol L−1 (De Graaf et al. 1998a; Lucassen et al. 2003). Because soil pH values in wet heaths and matgrass swards are usually between 4 and 5 (Table 1; Roelofs et al. 1996; De Graaf et al. 1998b), the negative effects of high ammonium concentrations on germination and establishment of the endangered species might be substantial. This is also reflected by the poor germination of these species found in the sod-cut plots in both areas. Low germination and poor establishment of target species after sod cutting were found in the restoration of dry heaths (De Graaf et al. 1998b) and also in Cirsio dissecti–Molinietum grasslands (Jansen & Roelofs 1996). In the sod-cut wet heath plots, target species, except for Erica tetralix, were rarely found. By contrast, Molinia caerulea germinated in all sod-cut plots and had the highest cover percentage after two growing seasons. This fast-growing grass species has a broad tolerance of soil conditions and it can become dominant over the endangered wet heath species because it can respond more efficiently to the current high nutrient availability (Berendse & Aerts 1984; Aerts & Berendse 1988; Sansen & Koedam 1996). The poor soil conditions after sod cutting were further illustrated by small-scale growth experiments with seedlings of Cirsium dissectum (data not shown). One year after the start of these experiments, the survival percentage of the seedlings was only 3% in Leemputten (n = 72) and 8% in Havelte-Oost (n = 120).

The lack of viable seeds of wet heath target species, as a consequence of the removal of seeds by sod cutting or poor seed dispersal possibilities in the fragmented Dutch landscape, has also been proposed as a bottleneck in the restoration of species-rich wet heaths and matgrass swards (Jansen, De Graaf & Roelofs 1996; Britton et al. 2000). Because most seeds of wet heath species are concentrated in the topsoil layers (Putwain & Gillham 1990; Pywell, Webb & Putwain 1995), sod cutting is likely to remove many of these seeds. However, this will have played only a minor role in the degraded parts of the two wet heath study areas. These degraded parts are already dominated by Molinia caerulea, the endangered species having disappeared several years ago. Because seeds of most of the endangered wet heath target species are short-lived (Thompson, Bakker & Bekker 1997), the presence of viable seeds of these species in the upper soil layers will probably be low.

It is also possible that poor seed dispersal of the target species may have contributed to the low germination and establishment in the field plots. Although populations of the target species are still present in the less degraded areas within 10–50 m of the sod-cut plots, seed dispersal from these plants to the sod-cut plots might be low (Bakker & Berendse 1999). Analysis of the soil seed banks of these two areas will elucidate this matter. However, even if seed dispersal is abundant, our results show that successful germination and establishment of the target species will be hampered by the inappropriate soil conditions after sod cutting.

In conclusion, our results indicate that sod cutting alone may not be an effective measure to restore formerly species-rich wet heaths and matgrass swards. The accumulation of ammonium after sod cutting hampers the germination and establishment of target species. Therefore, additional measures are necessary to prevent or reduce this build-up of ammonium and to overcome a possible scarcity of seeds. Experiments are in progress to investigate if the application of lime after sod-cutting might be an effective additional restoration measure.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was financially supported by the Netherlands Organization for Scientific Research and was conducted within the Stimulation Programme Biodiversity (project nr. 014.22.062). The Ministry of Defence and the Municipality of Ermelo kindly gave permission to work in the areas Havelte-Oost and Leemputten, respectively. We would like to thank T. G. Rouwenhorst, P. J. M. van der Ven, S. A. Robat and R. A. M. Welschen for their help with the chemical analyses.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aerts, R. & Berendse, F. (1988) The effect of increased nutrient availability on vegetation dynamics in wet heathlands. Vegetatio, 76, 6369.
  • Bakker, J.P. (1989) Nature Management by Grazing and Cutting. Kluwer Academic Publishers, Dordrecht, the Netherlands.
  • Bakker, J.P. & Berendse, F. (1999) Constraints in the restoration of ecological diversity in grassland and heathland communities. Trends in Ecology and Evolution, 14, 6368.
  • Berendse, F. (1990) Organic-matter accumulation and nitrogen mineralization during secondary succession in heathland ecosystems. Journal of Ecology, 78, 413427.
  • Berendse, F. & Aerts, R. (1984) Competition between Erica tetralix L. & Molinia caerulea (L.) Moench. as affected by the availability of nutrients. Acta Oecologia/Oecologia Plantarum, 5, 314.
  • Bobbink, R., De Graaf, M.C.C., Verheggen, G.M. & Roelofs, J.G.M. (1998a) Heeft Het Heischrale Milieu Nog Toekomst? Effectgerichte Maatregelen En Behoud Biodiversiteit in Nederland (eds R.Bobbink, J.G.M.Roelofs & H.B.M.Tomassen), pp. 131159. University of Nijmegen, Nijmegen, the Netherlands.
  • Bobbink, R. & Heil, G.W. (1993) Atmospheric deposition of sulphur and nitrogen in heathland ecosystems. Heathlands: Patterns and Processes in a Changing Environment (eds R.Aerts & G.W.Heil), pp. 2550. Kluwer Academic Publishers, Dordrecht, the Netherlands.
  • Bobbink, R., Hornung, M. & Roelofs, J.G.M. (1998b) The effects of air-borne nitrogen pollutants on species diversity in natural and seminatural European vegetation. Journal of Ecology, 86, 717738.
  • Britton, A.J., Marrs, R.H., Carey, P.D. & Pakeman, R.J. (2000) Comparison of techniques to increase Calluna vulgaris cover on heathland invaded by grasses in Breckland, South East England. Biological Conservation, 95, 227232.
  • De Graaf, M.C.C., Bobbink, R., Roelofs, J.G.M. & Verbeek, P.J.M. (1998a) Differential effects of ammonium and nitrate on 3 heathland species. Plant Ecology, 135, 185196.
  • De Graaf, M.C.C., Verbeek, P.J.M., Bobbink, R. & Roelofs, J.G.M. (1998b) Restoration of species-rich dry heaths: the importance of appropriate soil conditions. Acta Botanica Neerlandica, 47, 89111.
  • De Graaf, M.C.C., Verbeek, P.J.M., Cals, M.J.R. & Roelofs, J.G.M. (1994) Effectgerichte Maatregelen Tegen Verzuring En Eutrofiëring Van Matig Mineraalrijke Heide En Schraallanden. University of Nijmegen, Nijmegen, the Netherlands.
  • De Graaf, M.C.C., Verbeek, P.J.M., Verheggen, G.M., Bobbink, R. & Roelofs, J.G.M. (1995) Experimental liming of acidified heathlands and grasslands: effects on soil chemistry and vegetation. Ecosystem Manipulation Experiments: Scientific Approaches, Experimental Design and Relevant Results (eds A.Jenkins, R.C.Ferrier & C.Kirby), pp. 129137. European Commission, Luxembourg.
  • Erisman, J.W., Bleeker, A. & Van Jaarsveld, A. (1998) Atmospheric deposition of ammonia to semi-natural vegetation in the Netherlands – Methods for mapping and evaluation. Atmospheric Environment, 32, 481489.
  • Gigon, A. & Rorison, I.H. (1972) The response of some ecologically distinct plant species to nitrate- and to ammonium-nitrogen. Journal of Ecology, 60, 93102.
  • Gimingham, C.H. (1992) The Lowland Heathland Management Handbook, 1st edn. English Nature, Peterborough, UK.
  • Heil, G.W. & Diemont, W.H. (1983) Raised nutrient levels change heathland into grassland. Vegetatio, 53, 113120.
  • Houdijk, A.L.F.M., Verbeek, P.J.M., Van Dijk, H.F.G. & Roelofs, J.G.M. (1993) Distribution and decline of endangered herbaceous heathland species in relation to the chemical composition of the soil. Plant and Soil, 148, 137143.
  • Jansen, A.J.M., De Graaf, M.C.C. & Roelofs, J.G.M. (1996) The restoration of species-rich heathland communities in the Netherlands. Vegetatio, 126, 7388.
  • Jansen, A.J.M. & Roelofs, J.G.M. (1996) Restoration of Cirsio–Molinietum wet meadows by sod cutting. Ecological Engineering, 7, 279298.
  • KNMI (2001) Data and Products. Klimatologische Gegevens. Archief Maand- En Jaaroverzichten 2000/2001. Royal Dutch Meteorological Institute (KNMI). http://www.knmi.nl/product/.
  • Lucassen, E.C.H.E., Bobbink, R., Smolders, A.J.P., Van der Ven, P.J.M., Van der Gaag, M., Lamers, L.P.M. & Roelofs, J.G.M. (2003) Interactive effects of low pH and high ammonium levels responsible for the decline of Cirsium dissectum (L.) Hill. Plant Ecology, 165, 4552.
  • Mitchell, R.J., Auld, M.H.D., Hughes, J.M. & Marrs, R.H. (2000) Estimates of nutrient removal during heathland restoration on successional sites in Dorset, southern England. Biological Conservation, 95, 233246.
  • Pitcairn, C.E.R., Fowler, D. & Grace, J. (1991) Changes in Species Composition of Semi-Natural Vegetation Associated with the Increase in Atmospheric Inputs of Nitrogen. NERC Institute of Terrestrial Ecology, Penicuik, UK.
  • Putwain, P.D. & Gillham, D.A. (1990) The significance of the dormant viable seed bank in the restoration of heathlands. Biological Conservation, 51, 116.
  • Pywell, R.F., Putwain, P.D. & Webb, N.R. (1997) The decline of heathland seed populations following the conversion to agriculture. Journal of Applied Ecology, 34, 949960.
  • Pywell, R.F., Webb, N.R. & Putwain, P.D. (1995) A comparison of techniques for restoring heathland on abandoned farmland. Journal of Applied Ecology, 32, 400411.
  • Roelofs, J.G.M., Bobbink, R., Brouwer, E. & De Graaf, M.C.C. (1996) Restoration ecology of aquatic and terrestrial vegetation on non-calcareous sandy soils in The Netherlands. Acta Botanica Neerlandica, 45, 517541.
  • Roem, W.J., Klees, H. & Berendse, F. (2002) Effects of nutrient addition and acidification on plant species diversity and seed germination in heathland. Journal of Applied Ecology, 39, 937948.
  • Sansen, U. & Koedam, N. (1996) Use of sod cutting for restoration of wet heathlands: revegetation and establishment of typical species in relation to soil conditions. Journal of Vegetation Science, 7, 483486.
  • Schaminée, J.H.J., Van’t Veer, R. & Van Wirdum, G. (1995) Oxycocco-Sphagnetea. De Vegetatie Van Nederland, Deel 2. Plantengemeenschappen Van Wateren, Moerassen En Natte Heiden (eds J.H.J.Schaminée, E.J.Weeda & V.Westhoff), pp. 287304. Opulus Press, Leiden, the Netherlands.
  • Schwertz, C.A., Schaminée, J.H.J. & Dijk, E. (1996) Nardetea. De Vegetatie Van Nederland, Deel 3. Plantengemeenschappen Van Graslanden, Zomen En Droge Heiden (eds J.H.J.Schaminée, A.H.F.Stortelder & E.J. Weeda), pp. 263286. Opulus Press, Leiden, the Netherlands.
  • Snow, C.S.R. & Marrs, R.H. (1997) Restoration of Calluna heathland on a bracken Pteridium infested site in North West England. Biological Conservation, 81, 3542.
  • Thompson, K., Bakker, J.P. & Bekker, R.M. (1997) Soil Seed Banks of NW Europe: Methodology, Density and Longevity. Cambridge University Press, Cambridge, UK.
  • Troelstra, S.R., Wagenaar, R. & De Boer, W. (1990) Nitrification in Dutch heathland soils. 1. General soil characteristics and nitrification in undisturbed soil cores. Plant and Soil, 127, 179192.
  • Van Breemen, N., Burrough, P.A., Velthorst, E.J., Van Dobben, H.F., De Wit, T. & Ridder, T.B. (1982) Soil acidification from atmospheric ammonium sulphate in forest canopy throughfall. Nature, 299, 548550.
  • Van der Meijden, R. (1996) Heukels’ Flora Van Nederland, 22nd edn. Wolters-Noordhoff, Groningen, the Netherlands.
  • Van Vuuren, M.M.I., Aerts, R., Berendse, F. & De Visser, W. (1992) Nitrogen mineralization in heathland ecosystems dominated by different plant species. Biogeochemistry, 16, 151166.
  • Webb, N.R. (1998) The traditional management of European heathlands. Journal of Applied Ecology, 35, 987990.
  • Westhoff, V. & Van der Maarel, E. (1978) The Braun-Blanquet Approach. Classification of Plant Communities (ed. R.H.Whittaker), pp. 287378. Dr W. Junk bv Publishers, The Hague, the Netherlands.