TRACE METAL FLUXES IN A SPHAGNUM PEATLAND – HUMIC LAKE SYSTEM AS A CONSEQUENCE OF DRAINAGE MAREK KRUK∗ and KATARZYNA PODBIELSKA Department of Applied Ecology, University of Warmia and Mazury, 10-957 Olsztyn, Poland (∗ author for correspondence, e-mail:
[email protected],)
(Received 28 September 2004; accepted 20 July 2005)
Abstract. The study was conducted in a small (7.4 ha) peatland system with a humic lake concentrically surrounded by Sphagnum mat (Caricetum limosae) and Sphagnum bog (Ledo-Sphagnetum) in the Mazurian Lakeland in North-Eastern Poland. The peatland was situated in a forested catchment with a total area of 16.25 ha, and was influenced by surface drainage, artificially designed for forestry purposes. The content of zinc (Zn), cadmium (Cd), lead (Pb), copper (Cu) and hydrogen (H) ions was analysed in waters from precipitation, humic lake, mat, bog and surface outflow. The yearly budget of these ions, expressed as difference between atmospheric inflow and surface outflow, for the whole catchment was presented. The drainage of the peatland system over three years was accompanied by an increase in pH of lake, mat and bog waters. The whole catchment retention of H+ and trace metals studied decreased yearly in absolute values under these conditions. The peatland system began to leach Pb into surface waters and lost its ability to intercept Cd. These effects could be caused by intensity of outflow of water and sulphates and lowering of water storage. However, the potential to retain Zn and Cu ions was preserved. Keywords: Sphagnum peatland, trace metals, hydrogen ion, drainage
1. Introduction The drainage of peatlands for the purpose of changing the composition of tree species for economic reasons, as well as to protect the forests neighbouring the bogs against periodical flooding is a standard technique applied in forest management practices (Heikurainen et al., 1978; Lundin and Bergquist, 1990; Richardson and McCarthy, 1994). However, the effects of such activities on the ecosystem level have not yet been sufficiently documented. The drainage of a peatland increases the amount of water discharge (Seuna, 1974; Mustonen, 1975; Lundin and Bergquist, 1990), which results in a sequence of changes in chemical element fluxes within a given landscape. This alters the peatland biogeochemical properties, peatland catchment and the chemical composition of the discharged water, which may consequently affect the functioning of a water body receiving such water. The studies focused primarily on the loss of phosphorus (Crisp, 1966; Knighton and Stiegler, 1980; Sallantaus and P¨atil¨a, 1985), nitrogen (Dowding, 1981; Grootjans et al., 1985; Kirkham and Wilkins, 1993; Kruk, 1997) and sulphates (Andersson and Eriksson, 1978; Kruk, 1996). Less attention was paid to the trace metal fluxes through peatlands after a drainage system was installed. Water, Air, and Soil Pollution (2005) 168: 213–233
C
Springer 2005
214
M. KRUK AND K. PODBIELSKA
The aim of this study was to determine the effect of the drainage of a Sphagnum peatland – humic lake ombrothrophic system on the content of trace metal ions in the water components of the peatland and the fluxes of these elements in the peatland catchment system. It will be showed as the effect of the changes in the drainage hydrologic conditions, peatland water pH, fluxes of hydrogen ion (H+ ) and leaching of sulphates (SO2− 4 ). The study included the water-soluble forms of zinc (Zn), copper (Cu), lead (Pb) and cadmium (Cd). The literature provides information on the great heavy metal adsorption properties of Sphagnum moss and moss peat connected with enhanced cation exchange capacity of living cell walls, as well as, decomposed organic remains (Clymo, 1983). It is especially confirmed towards zinc and copper (Coupal and Lalancette, 1976; Ringquist and (Oumlau) Oborn, 2002; Ringquist et al., 2002). The increased copper outflow by 25% after peatland drainage was reported by Lundin and Bergquist (1990). High Pb retention indices on Sphagnum dominated peatland were obtained by Hemond (1980) and Kruk (2000). Krosshavn et al. (1992) examined the heavy metal bonding by peat of different pH and concluded that 75–85% of Zn and Cd may be adsorbed by moss peat with neutral pH, whereas the adsorption of Pb and Cu is not determined by pH. The greater mobility of lead ions in hollows compared to hummocks may be related to the effect of water levels in lower parts of a bog (Damman, 1978; Urban et al., 1991). The water level fluctuations caused redox potential and its changes seem to be the major mechanism responsible for mobility or removal of Pb in peat environment (Clymo and Hayward, 1982) and Cd in mineral soils (Ito and Iimura, 1975). Lead and Cadmium form sulphides in anaerobic conditions, which can be oxidized during lowering water levels and produces SO2− 4 and mobile metal ions (Ito and Iimura 1975; Damman, 1978). An experimental study reported 210 Pb mobility in Sphagnum peat cores showed very limited mobility tendency of soluble and particulate Pb (Vile et al., 1999).
2. Study Area The research covered a small 7.4 hectare area of a peatland complex including a small humic lake (0.23 ha, 2.5 m average depth) surrounded by Sphagnum mat (0.52 ha) and Sphagnum bog with a typical lagg zone (Figure 1). The studied area is situated in the Olsztyn Lakeland in the Baltic Lakelands in north-eastern Poland. The geographical co-ordinates are: 20◦ 25 eastern longitude and 53◦ 45 northern latitude. This system and the surrounding forest form a forest-peatland catchment covering an area of 16.25 ha. The dominating flora species of Sphagnum mat representing the Caricetum limosae Br.-Bl. 1921 association are: Sphagnum spp. L., Oxycoccus quadripetalus Gilib., Carex limosa L., Carex rostrata Stokes. and Drosera rotundifolia L.. The bog is covered by vegetation representing the Ledo-Sphagnetum magellanici Sukopp 1959 em. Neuh¨ausl. 1969 association: Sphagnum spp., Eriophorum vaginatum L.,
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
215
Figure 1. Schematic position of Sphagnum peatland – humic lake system. Black patch – humic lake, dark grey patch – Sphagnum mat, grey patch – Sphagnum bog with lagg, dashed thick line – border of catchment area, solid fine line – surface outflow from the peatland.
Ledum palustre L., Vaccinium uliginosum ChCl in the ground cover and Pinus silvestris L. and Betula pubescens ChAss. in tree layer. In the bank strip of the peatland, the bog-like species are replaced by more fertile habitat species. For example, willow shrubberies with Salix cinerea ChCl/O/All. have propagated on a large scale and the wet sub-areas are dominated by Alnus glutinosa DCl. The peatland consists of Sphagnum peat with a small degree of decomposition (10%) and low ash content (5%). The average depth of the peat layer was 4.0 m and the maximum reached 4.8 m. The bed consists of a layer of organic gyttja to 4.9 m depth (Dokumentacje torfowisk, 1976). The catchment surrounding the examined peatland covers 8.85 ha and is overgrown by a mature coniferous forest stand dominated by Pinus silvestris on sandy and clay soils. The area of the peatland occupied by Sphagnum bog is supplied exclusively with atmospheric precipitation. In addition, a narrow border zone with mineral soils of the catchment is also supplied with flooding water or run off water, particularly during spring thawing. Approximately a quarter of a year prior to the study, a ditch for draining water from the peatland was deepened and purged from the eastern area of the peatland. The ditch cuts through the peatland about 50 metres from the bank at a depth of about 50 cm. Within the peatland, the water in the ditch is mainly stagnant and starts to flow only outside the border of the peatland complex along the terrain slope (Figure 1). 3. Methods 3.1. G ENERAL
ASSUMPTION
The study was carried out for three years, to understand the tendency of behaviour of peatland ecosystem after drainage. The hydrological measurement and sampling
216
M. KRUK AND K. PODBIELSKA
began about three months after deepening the drainage ditch, when water discharge had been stabilized after drainage manipulation. The determination of trace metal behaviour properties in the examined Sphagnum peatland drainage system includes two methodological approaches. The first is an analysis of the metal content in the system internal waters of: bog, mat and a centric humic lake. This approach focuses on the investigation of the effect of an artificial water outflow on the variability of the concentration of the examined elements in the primary components of the system. The second approach examines the entire biogeochemical system behaviour under the influence of artificial drainage. This system includes water fluxes from the atmospheric deposition to surface outflow forming a catchment area system. In the case of peatlands, their isolation from the catchment surroundings into an inputoutput system is a methodologically difficult task to perform, as a hydrogeological measurement network is necessary (Kruk, 1990). Therefore, as a simplification, a catchment model was used for the biogeochemical characteristics of the examined catchment-peatland complex and included both the peatland (46% of its area) and its forested supply area. In relation to the existence of organic gyttja under the peat layer in the peatland, having a minimum hydraulic conductivity value, a prerequisite was made: the examined catchment system does not have an underground water exchange. It has to be underlined that the catchment approach applied to the evaluation of mineral balances of peatland systems is widely represented in the literature (Crisp, 1966; Verry and Timmons, 1982; Lundin and Bergquist, 1990). 3.2. H YDROLOGICAL
MEASUREMENTS AND SAMPLING
Hydrological examinations of the fluxes of precipitation and water outflow from the peatland-catchment system were used to determine the loads of trace metals and H+ ions. The hydrological measurements were completed on monthly basis and samples for analyses were taken once a month for 3 years from November 1996 to October 1999. The values of the water flow rate for the outflow from the examined peatland were obtained through the measurement of the water velocity and the flow cross-section along a 5-metre segment (of a 40 cm diameter pipe). The measurements were conducted in the pipe with the use of stopwatch and submerged floats. The time of flow of the submerged float along a pipe was measured in three to five repetitions during a day and the averages were calculated. The changes in the volume of water storage were calculated from the measurement of changes in water table in the Sphagnum bog using water mark. Precipitation data in mm month−1 was obtained from a Meteorological Station in Olsztyn located at 2.5 km distance. Yearly and monthly (March–October) evapotranspiration values were calculated from the water budget equation: P = E + R + S
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
217
where S is change in volume of water storage; P is precipitation; E is evapotranspiration; R is surface outflow. Turnover time (t) for peatland and lake area (7. 4 ha) was calculated approximately as relation (Mitsch and Gosselink, 1993): t = V /Q where: V is volume of water in peatland (water capacity for moss peat was estimated at 0.9) and lake; Q is volume of yearly total inflow: precipitation + inflow from watershed estimated as precipitation × 0,27 (regional outflow/precipitation ratio (Dynowska and Pociask-Karteczka, 1999)). Surface water outflow has been sampled from the peatland at a distance of about 20 metres outside from the borders. Peatland water was sampled from Sphagnum bog and Sphagnum mat through squeezing water from the surface spongy layer. Water was collected also from the central area of the humic lake. In order to assess the spatial distribution of pH and the metal content in water of mat and bog, 6 samples were taken from these areas, (e.g. from hummocks and hollows in the bog) in August 1998 and a variability coefficient was calculated as a standard deviation from the mean ratio. Precipitation, wet and dry together, was collected in the Meteorological Station into 2 or 3 polyethylene bottles (15 m between them) equipped with a funnel with 2 polyethylene nets: 1 mm mesh protecting against insects and a smaller net (0.1 mm) suspended at the end of the funnel. The containers were protected against exposure to sunlight by an aluminium foil cover. The metal concentration and pH in rainwater for a given month were calculated as an average from 2 or 3 sites (± std) to overcome the spatial variability of precipitation. 3.3. C HEMICAL
ANALYSIS
The sample was filtered through a Teflon filter with a resolution of 0.45 µm (Machery-Nagel, Germany). Next, the samples were acidified with HNO3 (0,1 N) (Aristar, BDH, UK), and mineralised in UV over 7 h. Concentrations of Zn, Cu, Pb and Cd were determined by inverse voltamperometry (DPASV) with use of a polarographic-voltamperometric apparatus Methrom 646 VA Processor, 675 VA Sample Changer, Switzerland). Standards from Merck, Germany and nitric acid (to stabilise pH of samples) from Fluka, Switzerland were used. pH of collected waters was measured potentiometrically (Orion Analyser 940, USA). S-SO2− 4 was determined by using ion chromatograph Methrom 690 (Switzerland). 3.4. STATISTICAL
METHODS
Coefficient of variability was used to express the spatial distribution of the content of metals in precipitation (mean from 4 data, 3 samples all), and in Sphagnum mat
218
M. KRUK AND K. PODBIELSKA
and Sphagnum bog waters (from 1 data, 6 samples all for both). The coefficients of Pearson’ correlation (R) were calculated and the regression lines were showed for the dependencies of trace metals outflow from hydrological parameters and sulphate outflow. The significance level of correlations was estimated. The statistical analysis was provided using STATISTICA 6.
4. Results 4.1. HYDROLOGY The variation in precipitation between years during three-year studies was small and ranged from 668 to 675 mm (Table I). Secondly, a significant variability of the precipitation for particular months was observed. The winter months and the summer months, particularly August of 1997 and September 1998 and 1999, were low in precipitation (below 30 mm) (Figure 2). On the other hand, the months with high precipitation were May (about 100 mm) and July (as much as 180 mm) in 1997 and April and June of 1999 (about 110 mm) (Figure 2). The catchment outflow was characterised by considerably lower values in comparison to the precipitation each year and almost each month (Table I, Figure 2). It is worth underlining that contrary to the total yearly precipitation, the values of the surface outflow increased from 102 mm to 144 mm (Table I). The increased monthly values of the outflow occurred in spring months, especially in 1999 and the late autumn months of 1997, and reached a peak of approximately 25 mm (Figure 2). The outflow in January 1997 amounted 0 mm, due to period of strong and long frost which froze water in the ditch. The changes in the volume of water storage and evapotranspiration show that water retention of bog was highly variable in particular months after peatland drainage. The increased amounts of evapotranspiration were noted in summer months, when TABLE I Hydrology of drained peatland-catchment system in Masurian Lakeland, NE Poland, during the three years November 1996 – October 1999. Annotations in mm Change in volume of Outflow/ Mean turnover Year Precipitation Outflow Evapotranspiration water storage Precipitation time (years) I II III ∗
668 675 669
102 129 144
566 539 508
0.0 −7 −17
0.153 0.191 0.215
4.25∗ 1.44∗∗
Calculated for a total peat deposit (4 m depth). Calculated for a surface active layer (1, 3 m depth estimated from the largest difference in changes in water storage (Figure 2)).
∗∗
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
219
Figure 2. Monthly dynamics of precipitation (solid line), evapotranspiration from March till October (dashed lines), changes in volume of water storage (thick grey line) and surface outflow (solid thick line) in the peatland system during 3 years of the study.
the majority of water from precipitation returned in the same season to the atmosphere (Figure 2). The elevated levels of precipitation caused also the increase of water retention in next month, as it happened in August after considerable rains in July 1997, in July and September 1998, and in May and August 1999. The water storage decreased regularly in months when outflow was greater: in February and November 1997 and also in July this year. The similar relation can be observed in April of 1999 year (Figure 2). The changes of water storage in winters are probably caused by alternately surface retention and thawing of snow. The run-off increased in the periods of enhanced precipitation or directly after them, as in June, July and October 1997, and in the spring in 1999 (Figure 2). Consequently, the periods with precipitation deficit, particularly during summer, caused lowering of the outflow or even its ceasing, as happened in September 1997 and September 1999. Similarly, relatively low amounts of precipitation in summer and autumn 1998 resulted in low water discharge in outflow (Figure 2). There is also a levelling effect visible in 1998 and 1999 years, which is caused probably by lesser differences in atmospheric input than in the first year of the study (Figure 2). The turnover time of water exchange in peatland system calculated for the whole profile was quite long (4.25 years), however, it is much shorter (1.44 years) when we take into account only surface active layer appointed from the changes of water storage (Table I). Draining the peatland with a ditch most likely resulted in gradual changes of the peatland system hydrology. The amount of water outflow in the following years of the study consequently increased, but evapotranspiration ratio, as well as water
220
M. KRUK AND K. PODBIELSKA
storage decreased. The relation between the outflow to precipitation, i.e. water outflow index in the first year was only 0.153 while in the third year it reached 0.215, which was a 40% increase (Table I). 4.2. DISTRIBUTIONS
AND DYNAMICS OF CONCENTRATIONS
pH
AND TRACE METAL
The highest values of pH were reported in the precipitation water (on average 5.9) and the lowest in the outflow from the peatland (only 4.1) (Table II). The pH values also showed a spatial variability in the internal waters of the peatland complex. It is characteristic that the pH decrease gradient was from the centre: the humic lake of pH 5.8, through Sphagnum mat (pH 5.1) until the formed peatbog occupied by the Ledo-Sphagnetum magellanici complex, whose near-surface waters had pH 4.5 (Table II). The seasonal variability of pH in the waters of the examined peatland system was clear. The lowest values of pH were reported in the first year in the lake and in the first and the second year in the bog, as well as in the first two years in the mat. On the other hand, the highest pH values were reached in the final year of the study (Figure 3). The increase in the pH of the humic lake occurred consequently from about 5.0 in the first year to about 6.5 at the end of the study. This regular increase is confirmed by the highly and significantly correlated (R = 0.87) trend line (Figure 3). The highest mean monthly concentrations of zinc and copper ions throughout the three years of the study were reported in the precipitation water falling on the peatland; 34.7 and 3.8 ppb respectively (Table II). A rather high content of these metals was also found in the bog waters, whereas the lowest was found in the humic lake water – only 7.1 ppb Zn and 1.9 ppb Cu (Table II). The mean lead ion concentrations differed strongly between the examined water types and ranged TABLE II Trace metals concentrations and pH in waters of drained peatland with humic lake in Masurian Lakeland, NE Poland, during November 1996 – October 1999. Mean ± standard deviation. Coefficient of variability of the content of metals caused by its spatial distribution: in precipitation (mean from 4 data, 3 samples all), and in Sphagnum mat and Sphagnum bog (from 1 data, 6 samples all for both), in % (in parentheses) Water from
n
pH
Zn (ppb)
Cu (ppb)
Pb (ppb)
Cd (ppb)
Precipitation Humic lake Sphagnum mat Sphagnum bog Outflow
34 5.9 ± 0.6 (2) 34.7 ± 29.7 (35) 3.8 ± 2.4 (30) 0.8 ± 2.6 (17) 0.9 ± 1.5 (22) 27 5.8 ± 0.4 7.1 ± 5.6 1.9 ± 0.7 0.4 ± 0.2 1.0 ± 1.4 28 5.1 ± 0.8 (10) 8.2 ± 4.8 (32) 2.8 ± 1.5 (25) 1.2 ± 0.5 (48) 1.6 ± 2.8 (49) 26 4.5 ± 0.8 (10) 19.3 ± 22.0 (58) 3.2 ± 2.0 (39) 2.2 ± 1.1 (56) 1.3 ± 1.7 (63) 32 4.1 ± 0.4
12.3 ± 9.0
2.1 ± 1.3
5.0 ± 2.3
1.8 ± 2.2
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
221
Figure 3. The course of changes of pH in water of main components of humic lake – Sphagnum peatland complex during 3 years of the study. Regression line indicate trend line of pH changes in humic lake water. Correlation coefficient R = 0.87 (significance level p < 0.01), N = 27.
from 5.0 ppb in the outflow waters to 0.4 ppb in the lake. The cadmium ion concentrations varied widely in particular in the water, which exhibited high values of standard deviation exceeding the mean value. On average, the highest concentrations of cadmium were found in the drainage outflow water (1.8 ppb) and water from the mat while the lowest concentrations were reported for precipitation and lake water (0.9 ppb) (Table II). The deviations resulting from the spatial variability in metal concentration in the precipitation was the highest for zinc and copper with variability coefficient over 30% and the lowest was for lead ions – only 17%. Significantly higher values of the coefficient of variability (even exceeding 50%) were reported for the concentration of the examined metals in the waters of mat and bog (Table II). The course of the changes in the concentration of the examined trace metals in internal waters of the peatland complex over a sequence of months and years was irregular. The greatest concentrations were found in Sphagnum bog waters: for Zn in 1997 up to 100 ppb, for Cu in the same year up to 8 ppb and for Pb in the same year up to 6 ppb. However, the highest concentration of Cd (up to 12 ppb) was reported for Sphagnum mat water also in the first year of the study (Figure 4). Almost consequently, the lowest concentrations of trace metals examined, as well as, its seasonal fluctuations, were found in humic lake waters. In case of Sphagnum
Figure 4. Seasonal changes of the concentration of trace metals in waters of the components of humic lake – peatland complex: humic lake – solid line, Sphagnum mat – dashed line, Sphagnum bog – dotted line.
222 M. KRUK AND K. PODBIELSKA
223
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
mat and bog waters elevated concentrations of metals were noted mostly in summer months (Figure 4). 4.3. RETENTION
AND LOSS OF TRACE METALS
The load of hydrogen ions in the precipitation waters supplying the peatland catchment was 64 g ha−1 y−1 in the first year of the study and was significantly (5–10 times) higher in comparison to the following years. However, the outflow of H+ was high and less varied in all the studied years (90–190 g ha−1 y−1 ). Consequently, the catchment and peatland system clearly was losing hydrogen ions with the outflow. The smallest loss of hydrogen ions was observed in the first year and the highest was observed in the second year of the study (Table III). Similar to the supply of hydrogen ions, the greatest inflows of trace metals from the atmosphere into the catchment were also observed in the first year of the study. In the third year, however, the lowest amount of these elements from the atmosphere reached the catchment and the greatest differences between particular years were observed here for lead and cadmium loads (Table III). Yearly outflow loads were usually less differentiated in the years of study than the inflow loads of the precipitation. The highest amounts of Zn and Cu outflow were in the third year and the highest amounts of Pb and Cd were in the second year (Table III). A greater metal inflow with precipitation waters did not result in an increase in their outflow within the peatland draining water in a given year. In fact, based on the calculations of the catchment balance, a greater supply from the atmosphere resulted in a greater retention of trace metal ions in absolute (g ha−1 y−1 ) and relative (in % TABLE III The budget of H and Zn, Cd, Pb, Cu ions expressed by difference between atmospheric input and surface outflow in drained peatland – catchment system in Masurian Lakeland, N-E Poland, for three years: November 1996 – October 1999. Values are in g ha−1 yr−1 and in % of atmospheric input (in parentheses). Negative values indicate losses Budget component
Year H
Inflow from I atmosphere II III Outflow I II III Budget I II III
64.3 15.4 6.2 91.5 189.8 158.5 −27.2 (−42) −174.4 (−11 times) −152.4 (−25 times)
Zn
Cu
273.2 247.5 107.3 13.5 13.9 14.8 259.7 (95) 233.6 (94) 92.5 (86)
25.8 21.2 18.2 2.6 2.0 4.2 23.2 (90) 19.2 (91) 14.0 (77)
Pb
Cd
10.3 9.4 2.4 6.7 1.6 0.7 5.4 1.0 6.3 3.9 6.1 0.8 4.9 (48) 8.4 (89) −3.9 (−1.7 times) 2.8 (41) −4.4 (−2.7 times) −0.1 (−16)
224
M. KRUK AND K. PODBIELSKA
of supply) values by the peatland catchment area. On the other hand, with a lower supply in the second and third year, the retention was smaller and even a negative balance was reported for Pb and Cd (Table III). The relative balance was 95% in the first year for Zn and 90% for Cu, while in the third year it was 86% and 77%, respectively (Table III). Contrary, lead retention (48% of the supply) in the first year did not maintain the same level in the following years of the study; the peatland became an effective provider of this metal (net) to the outflowing water. Similarly, a strong cadmium retention (which was 89% in the first year) dropped to 41% in the second year and completely disappeared in the third year (Table III). These lower values of the Pb and Cd budgets in two last years of the study seem to be concordant with elevated hydrogen ion outflows and negative budgets in the same time (Table III). These results of the yearly changes of Pb and Cd budgets can be read as uncertain in the light of long turnover time of water throughflow in the peatland catchment amounted 4.25 years. It is however contradictory with suggestion that movement of water in peatland environment is limited into surface layer called acrotelm, so the transport of ions through peatland can be accelerated and visible in the shorter period, 1.44 year (Table I). To verify this suggestion we examined the reaction of the outflow of these metals on the volume of monthly outflow of water from the peatland. The outflows of Pb and Cd were significantly correlated ( p < 0.01) with outflows of water from the system. Very strong dependence of Pb outflow on water discharge (R = 0.92) should be noticed (Figure 5). The enhanced tendency for loss of lead and cadmium as a reaction on hydrological changes, such as increased drainage was confirmed also in a short period of time. The lowering and even loss of accumulation of Pb and Cd ions during 3-year study could be explained also by interaction with sulphur reactions in the oxidized zone of the acrotelm layer. The enhanced amounts of sulphate outflowing by drainage waters indicate sulphide oxidation and release of soluble metals. Indeed, the correlation between outflow of lead and cadmium ions and sulphate loads in drainage waters was significant ( p < 0.01), however, the strength of correlation was not very high: 0.46 for Pb and 0.50 for Cd. (Figure 6). The influence of the oxidation of peat surface layer on Pb and Cd transport can be expressed as dependence of the decrease of water storage in peatland on this mobility. The relationship is very clear in the case of lead, where lowering of water retention controls the outflow of Pb quite tightly (R = 0.89, p < 0.01), however, this relation for cadmium is not significant (Figure 7).
5. Discussion The reaction of the biogeochemical system of the peatland complex to the hydrotechnical treatment, i.c. the deepening of a drain was presented in this paper. The hydrological data, such as the amount of yearly drained water and the
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
225
Figure 5. The dependence of lead and cadmium outflow on outflow of water from Sphagnum peatland complex. Pb – correlation coefficient R = 0.92 (significance level p < 0.01), Cd – R = 0.46 ( p < 0.01). N = 32.
outflow-to-precipitation ratio, as well as the pH values of the internal waters of the peatland, all indicate the existence of a gradient of changes from the beginning of drainage. We have no measurements of metal transport before peatland drainage, however, we assume that biogeochemical properties of the peatland system before the drainage were the most similar to, or better than, the situation in the first year of the study when water outflow ratio was the lowest. The obtained hydrological results agreed with the principal that in the budget of ombrothrophic peatlands the precipitation only slightly exceeds the
226
M. KRUK AND K. PODBIELSKA
Figure 6. The relationship between lead and cadmium outflow and sulphate outflow from the peatland. Pb – correlation coefficient R = 0.46 (significance level p < 0.01), Cd – R = 0.50 ( p < 0.01). N = 32.
evapotranspiration, which is exhibited by the low value of the outflow-toprecipitation ratio (Baden and Eggelsmann, 1964; Romanov, 1968; Hemond, 1980; Ivanov, 1981). The draining of Sphagnum bog over a period of two or three years increased this ratio from 0.19 to 0.46 (Lundin and Bergquist, 1990), which is an even higher ratio than in the system studied in this work. The amount of water outflow after drainage and the described outflow index are approaching the characteristic values for the catchment areas without peatlands. An average outflow index for Poland according to Dynowska and Pociask-Karteczka (1999) is 0.27.
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
227
Figure 7. The dependence of lead and cadmium outflow from the peatland on the changes in water storage. Pb – relation between lead outflow and decreasing of water retention: R = 0.89 ( p < 0.01), N = 17. Cd – correlation is not significant ( p > 0.05).
These hydrological parameters are determined by technical properties of canals cutting the peatland system. It can be assumed, that peatland before drainage could be characterized as almost hydrologically closed system, and all hydrologic and biogeochemical processes, which were observed in the period of study, appeared or increased after drainage initiation. The complex of peatland and humic lake constitute of almost closed water reservoir with quite long theoretical turnover time, 4.25 years in average (Table I). There is, however, the question what is the relation between this theoretically calculated time with real removal time in situation in which there is almost no mixing between lower layers and surface active layers. In conclusion, turnover time calculations based on the dimensions of peatland would be overestimated (Mitsch and Gosselink, 1993). It is very likely that exchange of water in the system do not occur in the total 4 m depth of peat deposit, but only in approximately 130 cm surface active layer, appointed by the highest difference between changes in water storage
228
M. KRUK AND K. PODBIELSKA
(Figure 2). The turnover time schould also be differentiated spatially. The water exchanged probably faster in the area near the ditch than in the peripheral zones. Evapotranspiration from the lake-peatland system is a process depending on the evaporation properties of peatland waters and the transpiration capacity of the vegetation. These factors influence the temporal and spatial (zoned) heights of evaporation and the differences in water level in particular parts of the peatland (Hooper and Morris, 1982). In the studied system, these changes in water retention between the lake, mat and bog, especially during the beginning and the final stage of snow retention, are most likely to determine the quite irregular seasonal courses of hydrological parameters measured. To clear the picture of the distribution of precipitation water in an ombrotrophic peatland system, the evapotranspiration dynamic and changes in water storage were taken into account (Figure 2). The reaction for an elevated or extraordinary low precipitation is visible in the same month or with a time lag. For example, high precipitation in July 1997 caused the rise of evapotranpiration and outflow in the same month and increase of water storage in the next month. The precipitation deficit in August 1997 gave decreasing outflow and water storage in consecutive months (Figure 2). There is an important issue to be discussed. Did the observed hydrological changes, as increasing of flow and changes in volume of water storage implied by drainage, have an effect on the pH changes in waters within the lake-peatland system? Did they affect the consequent yearly increase in the pH of lake waters? Presumably, the open waters of the humic lake reacted to the drainage faster than the waters bound in the peat and Sphagnum moss structures of mat and bog. The latter, however, increased pH in the third year of the study, which could indicate to the infiltration of lake waters into surrounding peatland zones, as the result of more intensive drainage in this year (Figure 3). The changes of pH value in humic lake and peatland waters seem to be concordant with an increasing negative budget of H+ ions in the whole system (Table III). Probably, it is connected with a shift in the peatland water troughflow. An additional factor which could have had an effect is a relatively small, especially when compared to the first year of the study, H+ ion supply with precipitation waters. It might be caused by an influence of different air masses. However, it should be remembered that changes in the pH of the precipitation are a factor affecting the environmental acidity of Sphagnum bog most likely to a smaller degree than the cation exchange through the moss cell walls Sphagnum ssp. (Clymo and Hayward, 1982; Urban et al., 1985) or the activity of organic acids originating from decomposition (Gorham et al., 1984; McKnight et al., 1985). Concentrations and the trace metal loads in the atmospheric deposition in Europe are quite variable and are generally determined by the study area location in relation to the industrialised regions. The Mazurian Lakeland area, including the surroundings of Olsztyn, is one of the least polluted regions in Poland. Therefore, the concentrations and the loads of the metals are significantly lower in this northern-eastern region than in southern Poland. These differences are particularly
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
229
visible in the case of lead precipitation; 3–20% of southern Poland deposition of this metal reaches the Mazurian ecosystems (Zimka and Stachurski, 1996; SzarekL ukaszewska, 1999). In the case of zinc, copper and cadmium, these differences are usually 2–4 times higher in southern Poland than in the NE regions, based on the references cited. The obtained metal concentrations for Zn, Cu and Cd are most similar to the results originating from Sweden and Finland (Berkvist and Folkeson, 1992; Helmisaari, 1995). However, the Pb deposition values for NE Poland only reach 2–15% of those for Scandinavia. A similar study of trace metals retention in the peatbogs in Sweden and Norway (calculated with the stratigraphic method) were carried out by Jensen (1997). The peatland-water environment cleanliness is also confirmed by the low concentration of trace metals in the humic lake and the waters of bog and mat. These values fall into the low range, which was established for the cleanest rivers in Poland and also all of Europe (Kabata-Pendias and Pendias, 1993). The results of the study show that ecosystem fluxes of dissolved Pb and Cd submit the most visible changes caused by drainage of bog-humic lake system. These changes consist in the destruction of the mechanism of metal retention in the environment of the drained bog. In this study, the possible transitions between dissolved and particulate forms of metals (Stevenson, 1994) were not investigated. It is worth to indicate, that the method of rain water sampling used in the study can overestimate the content of soluble Pb in relation to Pb in alkyl- compounds due to its photolytic decomposition during transport and in laboratory. However, this process is also very fast in the natural environment (Rhue et al., 1992). The main process that intercepts metal cations by soil organic matter seems to be its binding by negatively charged cell walls or to decomposed organic particles (Clymo, 1983; Friedland, 1990). However, in case of a peatland environment, cation exchange capacity is accompanied widely by forming sulphides in stable anaerobic conditions. In water saturated peat, mobile sulphate ions are reduced into sulphides, which potential for immobilization of metal cations by sulphur should not be neglected (Damman, 1978). The author states also, that in case of water level fall and improvement of aerobic conditions, the oxygenation of sulphides moves Pb into soluble and mobile form. A similar conclusion was found in an experiment on Cd behaviour conducted by Ito and Iimura (1975). In the context of above suggestions, the phenomenon of Pb and Cd leaching from peatland complex, showed in presented study, can be more comprehensive. On the basis of the three years of study the following scenario of losing accumulative properties for Pb and Cd of drained peatland can be proposed. The drainage by ditch accelerated water movement and this phenomenon is clearly visible during the study (Table I). As the result, fluctuations of water storage and outflow (Figure 2) furnished more unstable redox and pH conditions, primarily in Sphagnum bog surface layer acrotelm. It seems to be symptomatic, that in two first years a clear zonation of pH distribution between humic lake, mat and bog occurred, however, in third year, the mentioned gradient disappeared and considerable fluctuations of the pH value were
230
M. KRUK AND K. PODBIELSKA
noted (Figure 3). Simultaneously, year by year, the precipitation supply of H+ ions, as well as Pb and Cd, were decreasing (Table III). However, decreased supply of these metals had no any influence on its leaching from the peatland. We think that increased drainage and water retention fluctuations did not inhibit mobility of metal ions caused by sulphide oxidation and SO2− 4 leaching. The significant dependence of Pb and Cd flows on the water outflow indicates that conditions in peat layer established by drainage caused fast geochemical reactions (Figure 5). The outflow of lead began in the second year of the study, probably because of the shift caused by turnover time in the surface layer (1.44 year). More, sulphate release and intensity of outflow could cause synergic effect on Pb and Cd mobility (Figures 5 and 6), even at the limited atmospheric supply. There is also very clear correlation between lowering of water storage and Pb outflow (Figure 7). Cadmium, however, exhibits more inhibited mode of mobilization; the relations with water and SO2− 4 were weak and dependence on water retention changes did not occur. When we look at the seasonal changes of metal contents in mat and bog waters, there is hard to find any regularity, except may be for cadmium, which concentration in mat waters decreased year by year and even disappeared in bog waters in third year (Figure 4). Do we observe an effect of drainage with the shift in turnover time? When we assume, that drainage affect water movement in direction from central humic lake into ditch, then mentioned above consequent lowering of metal concentrations in lake waters could be caused also by slow water seepage through mat and bog zone into outflow. This phenomenon can explain the increase of Zn, Cu and Pb concentration in bog waters in third year, when content of these metals in humic lake were the lowest (Figure 4).
6. Conclusion The characteristic feature of the balance of trace metal flow through the examined peatland complex is a consequent decrease from year to year in the load of these elements in the precipitation and a lack of this tendency or its reverse direction in their outflow from the peatland. It is worth mentioning that from the drained system of peatland and catchment, the largest amounts of Zn and Cu are lost in the third year of the study, although it maintains the properties of the accumulation system for these elements. On the other hand, although the Pb losses are similar in all three years and the losses of Cd vary greatly, the greatest relative losses of these metals occurred in the third year. The amounts of lead and cadmium ions flowing from the peatland complex are correlated with water discharge in outflow and leaching of sulphates from the peat. The outflow of Pb is also dependent from lowering of water storage in the peatland. Continuing high Zn and Cu retention and a clear tendency towards an increasing loss of Pb and Cd ions constitutes a biogeochemical result of the Sphagnum peatland ditching. This effect can play an important role for the
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
231
transport of these toxic metals in the catchment surface waters on the landscape scale. Acknowledgments I would like to express my gratitude for prof. A. Stachurski and prof. J. Zimka from Institute of Ecology PAS for their help in the analysis of trace metals. I am also grateful for three anonymous reviewers, which remarks and comments visible improved a quality of publication. References Andersson, U. M. and Eriksson, E.: 1978, Hydrochemical investigations in three representative basins in Sweden IDH/IHP, Swedish National Science Research Council, Report 50, 71 pp. Baden, W. and Eggelsmann, R.: 1964, ‘Wasserkreislauf eines nordwestdeutschen Hooch-moores’, Schrift. Kurat. Kult. Hamburg 12, 1–156. Berkvist, B. and Folkeson, L.: 1992, ‘Soil acidification and element fluxes of Fagus sylvatica forest as influenced by simulated nitrogen deposition’, Water Air Soil Pollut. 65, 111–133. Clymo, R. S.: 1983, ‘Peat’, in A.J.P. Gore (ed.), Mires: Swamp, Bog, Fen, and Moor, Ecosystems of the World 4A, Elsevier, Amsterdam, pp. 159–224. Clymo, R. S. and Hayward, P. M.: 1982, ‘The ecology of Sphagnum’, in A. J. E. Smith (ed), Bryophyte Ecology, Chapman and Hall, London, pp. 229–289. Coupal, B. and Lalancette, J. M.: 1976, ‘The treatment of wastewaters with peat moss’, Water Res. 10, 1071–1076. Crisp, D. T.: 1976, ‘Input and output of minerals for an area of Pennine Moorland. The importance of precipitation, drainage, peat erosion and animals’, J. Appl. Ecol. 3, 327–348. Damman, A. W. H.: 1978, ‘Distribution and movement of elements in ombrotrophic peat bogs’, Oikos 30, 480–495. Dokumentacje torfowisk [Peatland specifications]: 1976, Centralne Biuro Studi´ow i Projekt´ow Wodno-Melioracyjnych, Warszawa, msc. Dowding, P.: 1981, ‘The effects of artificial drainage on nitrogen processes’, in F. E. Clark and T. Rosswall (eds), Terriestrial Nitrogen Cycles, Ecol. Bull. 33, Stockholm, pp. 615–625. Dynowska, I. and Pociask-Karteczka, J.: 1999, ‘Obieg wody [Water cycle]’, in L. Starkel (ed), ´ Geografia Polski. Srodowisko Przyrodnicze [Geography of Poland. Natural Environment], Wydawnictwo Naukowe PWN, Warszawa, pp. 343–373. Friedland, A. J.: 1990, ‘Heavy metal movement in soils and ecosystems’, in Shaw, J. (ed), Evolutionary Aspects of Heavy Metal Tolerance in Plants, CRC Press, Boca Raton, pp. 7–17. Gorham, E., Bayley, S. E. and Schindler, D. W.: 1984, ‘Ecological effects of acid deposition upon peatlands: A neglected field in “acid rain” research’, Can. J. Fish Aquatic Sci. 41, 1256–1268. Grootjans, A. P., Schipper, P. C. and Van der Windt, M. J.: 1985, ‘Influence of drainage on nitrogen mineralization and vegetation response in wet meadows’, Acta Oecol. Pl. 6, 406–417. Heikurainen, L., Kentt¨amies, K. and Laine, J.: 1978, ‘The environmental effects of forest drainage’, Suo 29, 49–58. Helmisaari, H.-S.: 1995, ‘Nutrient cycling in Pinus sylvestris stands in eastern Finland’, Plant and Soil 327–336. Hemond, H. F.: 1980, ‘Biogeochemistry of Thoreau’s Bog, Concord, Mass.’, Ecol. Monogr. 50, 507– 526.
232
M. KRUK AND K. PODBIELSKA
Hooper, F. F. and Morris, L. S.: 1982, ‘Mat-water phosphorus exchange in an acid bog lake’, Ecology 63, 1411–1421. Ito, H. and Iimura, K.: 1975, ‘Absorption by cadmium by rice plants in response to change of oxidationreduction conditions in soils’, J. Sci. Soil a. Manure 46, 82–89. Ivanov, K. E.: 1981, Water Movement in Mirelands, Academic Press, London, 276 pp. Jensen, A.: 1997, ‘Historical deposition rates of Cd, Cu, Pb and Zn in Norway and Sweden estimated by 210 Pb dating and measurement of trace elements in cores of peat bogs’, Water Air Soil Pollut. 95, 205–210. ´ Kabata-Pendias, A. and Pendias, H.: 1993, Biogeochemia Pierwiastk´ow Sladowych [Biogeochemistry of Trace Metals], Wydawnictwo Naukowe PWN, 364 pp. Kirkham, F. M. and Wilkins, R. J.: 1993, ‘Seasonal fluctuations in mineral nitrogen content of an undrained wetland soil following differing rates of fertiliser nitrogen application’, Agric. Ecosyst. Environ. 43, 11–29. Knighton, M. D. and Stiegler, J. H.: 1980, ‘Phosphorus release following clearcutting of black spruce fen and a black spruce bog’, in Proceeding of Sixth International Peat Congress, August 1980, Duluth, Minnesota, USA, pp. 577–583. Krosshavn, M., Steinnes, E. and Varskog, P.: 1993, ‘Binding of Cd, Cu, Pb and Zn in soil organic matter with different vegetational background’, Water Air Soil Pollut. 71, 185–193. Kruk, M.: 1990, ‘The processing of elements by mires in agricultural landscape: Mass balances based on sub-surface hydrology’, Ekol. Pol. 38, 73–117. Kruk, M.: 1996, ‘Biogeochemical consequences of watershed inflow of elements into fens in agricultural landscape’, Pol. Ecol. Stud. 22, 105–127. Kruk, M.: 1997, ‘Effects of draining on nitrogen flow through mires in agricultural landscape’, Ekol. Pol. 45, 441–460. Kruk, M.: 2000, ‘Biogeochemical functioning of hydrologically modified peatlands and its effect in eutrophication of freshwaters’, Pol. J. Ecol. 48, 103–161. Lundin, L. and Bergquist, B.: 1990, ‘Effects on water chemistry after drainage of a bog for forestry’, Hydrobiologia 196, 167–181. McKnight, D., Thurmann, E., Wershaw, R. and Hemond, H.: 1985, Biogeochemistry of aquatic humic substances in Thoreau’s Bog, Concord, Mass.’, Ecology 66, 1339–1352. Mustonen, S.: 1975, ‘The influence of draining for forestry on hydrology of peatlands’ in E. Hansen (ed), Inadvertent Effects of Man on the Hydrological Cycle, Nordic IHD Report 8, Oslo, pp. 68–69. Richardson, C. J. and McCarthy, E. J.: 1994, ‘Effect of land development and forest management on hydrologic response in southeastern coastal wetlands’, Wetlands 14, 56–71. Rhue, R. D., Mansell, R. S., Ou, L. T., Cox. R., Tang, S. R. and Ouyang Y.: 1992, ‘The fate and behavior of lead alkyls in the environment: A Review’, Crit. Rev. Environ. Control. 22, 169–193. ¨ Ringquist, L. and Oborn, I.: 2002, ‘Copper and zinc adsorption onto poorly humified Sphagnum and Carex peat’, Water Research 36, 2233–2242. ¨ Ringquist, L., Holmgren, A. and Oborn, I.: 2002, ‘Poorly humified peat as an adsorbent for metals in wastewater’, Water Research 36, 2394–2404. Romanov, V. V.: 1968, Hydrophysics of bogs, Clearinghouse for Federal Scientific and Technical Information, Springfield, Virginia, 299 pp. Sallantaus, T. and P¨atil¨a, A.: 1985, ‘Nutrient transport in runoff water from Finnish peatland drainage and peat mining areas’, in Proceedings of the Peat and the Environment’85, International Peat Society Symposium, Jonk¨oping, Sweden, September 1985, pp. 87–99. Seuna, P.: 1974, ‘Influence of forest draining on the hydrology of an open bog in Finland’, in Proceedings of the International Symposium on Forest Drainage, 1974, Jyv¨askyl¨a-Oulu, Finland, pp. 385–393. Stevenson, F. J.: 1994, Humus Chemistry. Genesis, Composition, Reactions, John Wiley & Sons, Inc., New York, 496 pp.
TRACE METAL FLUXES IN A SPHAGNUM PEATLAND
233
Szarek-L ukaszewska, G.: 1999, ‘Input of chemical elements to the forest ecosystem on the Carpathian Foothills (S. Poland)’, Pol. J. Ecol. 47, 191–213. Urban, N. R., Eisenreich, S. J. and Gorham, E.: 1985, ‘Proton cycling in bogs: Geographic variation in north-eastern North America’, in T. C. Hutchinson and K. Meema (eds.), Proceedings NATO Advanced Research Workshop on the Effects of Acid Deposition on Forest, Wetland, and Agricultural Ecosystems, Toronto, May 13–17, 1985, Springer-Verlag, New York, pp. 577–598. Urban, N. R., Eisenreich, S. J., Grigal, D. F. and Schurr, K. T.: 1990, ‘Mobility and diagenesis of Pb and Pb-210 in peat’, Geochim. Cosmochim. Acta 54, 3329–3346. Verry, E. S. and Timmons, D. R.: 1982, ‘Waterborn nutrient flow through an upland-peatland watershed in Minnesota’, Ecology 63, 1456–1467. Vile, A. M., Wieder, R. K. and Novak, M.: 1999, ‘Mobility of Pb in Sphagnum-derived peat’, Biogeochemistry 45, 35–52. Zimka, J. R. and Stachurski, A.: 1996, ‘Forest decline in Karkonosze (Poland). Part II. An analysis of acidity and chemistry of atmospheric precipitation, throughfall and forest stream waters’, Ekol. Pol. 44, 153–177.
