Reconstructing trace metal deposition by dendrochemical analysis: A comparison among tree species.

 

Shaun A. Watmough & Thomas C. Hutchinson

 

Environmental and Resource Studies Program, Trent University,

Peterborough, Ontario, K9J 7B8

 

Email:   swatmough@trentu.ca   Phone: 705-748-1071

 

Abstract

 

Cadmium concentrations in annual rings of sycamore, lime, beech and ash were compared with annual Cd deposition values recorded between 1975 and 1990 at a small woodland adjacent to a Cu-Cd refinery.  During this period Cd deposition decreased dramatically, from over 40 mg/m2/a in 1975 to about 4 mg/m2/a by 1990. Significant positive correlations were found between Cd deposition and Cd concentrations in tree rings of sycamore (r=0.76; p<0.01), lime (r=0.52; p<0.05) and beech (r=0.59; p<0.05).  Cadmium concentrations in ash were much lower and there was no significant relationship between Cd deposition and Cd concentrations in tree rings of ash (r=-0.09; N.S.). Tree species, which lack pronounced heartwood (diffuse porous wood), appear to be suitable monitors of Cd deposition, although differences in metal concentration between tree species as well as between individual trees should be expected.  With increasing sample numbers a more accurate interpretation of metal deposition at a given site may be obtained.

 

 

 

Introduction

 

Dendrochemistry has emerged in recent years as a valuable tool that allows the reconstruction of historical metal deposition.  Trees are not passive recorders of their external environment, however, and there is considerable debate as to whether dendrochemistry is an appropriate tool for environmental monitoring (1,2).  Certain tree species appear to be more suitable than others, but there is considerable uncertainty as to which species maintain an historical record of trace metal exposure.  The application of dendrochemistry may also be restricted by differences in metal behaviour and mobility in wood, and by external environmental conditions (3).  Such differences are likely the main reason why there has not yet been a general consensus as to which tree species are most suited to environmental monitoring.  In this study, Cd concentrations were measured in annual increments of sycamore (Acer pseudoplatanus), beech (Fagus sylvatica), ash (Fraxinus excelsior) and lime (Tilia europaea) trees growing in a woodland nearby a Cu-Cd refinery in northwest England.  Cadmium concentrations in wood (1975-1990) were compared with annual Cd deposition values to determine the suitability of each tree species to monitoring Cd deposition under the same environmental conditions.

 

 

 

Materials and Methods

 

All trees (4 sycamore, 2 beech, 3 ash and 2 lime) were sampled in August 1999 in a small woodland, approximately 0.5 km north of a Cu-Cd refinery in Prescot, north-west England (OS NGR SJ5465 927). Tree cores were taken at breast height (1.5 m) from the south-facing aspect (facing the refinery) of each tree using a 5-mm stainless steel increment borer.  Samples were immediately sealed in dry, plastic straws and were stored at -10oC prior to analysis.  Each core was divided into 1-yr increments dating to 1975.  Increments were separated using a stainless-steel knife, and samples were placed in pre-washed (10% H2SO4) 50 ml borosilicate glass tubes, dried at 70oC for 24 h, weighed and then dry ashed at 400oC in a muffle furnace for 12 h.  The ash was dissolved in trace-grade HNO3 and digested under reflux at 100oC for 6 h.  Samples were diluted with dd-H2O prior to analysis using a Perkin-Elmer SCIEX ELAN 6000 inductively coupled plasma mass spectrometer (ICP-MS). Precision and accuracy were confirmed by repeated analysis of NIST apple leaf standard reference material (1515). 

 

 

Results and Discussion 

 

Cd deposition

 

Atmospheric deposition was monitored, using 760-cm2 deposit gauges (wet plus dry deposition) located adjacent (within 100 m) to the woodland site between 1975 and 1991 (Figure 1).  Cd deposition was around 40 mg/m2/annum in 1975, but decreased to about 4 mg/m2/annum by the late 1980s.  Deposition of Cd reflects the activity of the metal-processing plant and emission control developments. For example, bag filters were installed in 1970, with further improvements made in the 1980s but with varying success; breakages of the filtering system apparently occurred quite frequently (4). 

 

Cadmium deposition levels in the study region are far greater than background deposition values for urban areas in the UK (~1 mg/m2/a) (4).  These high levels of Cd deposition have resulted in severe contamination of surface soils in the Prescot area.  Adjacent to the refinery, Cd concentrations in soil were >150 mg/kg, although concentrations in soil declined

exponentially with increasing distance from the refinery.  Nevertheless, results of 2 extensive studies (n=99) in 1991 and 1993 indicated that surface soil at the woodland site was heavily contaminated with Cd.  Mean Cd concentrations in surface soil were 12.2 mg/kg (S.D. 9.3) in 1991 and 9.6 mg/kg (S.D. 11.3) in 1993, which are much higher than typical background concentrations of Cd in soil of 0.01-2.00 mg/kg reported by Alloway (5).

 

Cd in tree-rings

 

The decline in Cd deposition was reflected in the Cd concentrations of tree rings of sycamore, beech and lime (Figure 2).  Significant positive correlations were found between Cd deposition and Cd concentrations in tree rings of sycamore (r=0.76; p<0.01), lime (r=0.52; p<0.05) and beech (r=0.59; p<0.05).  There was no significant relationship between Cd deposition and Cd concentrations in ash (r=-0.09; N.S.).   Unlike sycamore lime and beech, ash wood is ring-porous, which typically have pronounced heartwood. In some cases, heartwood tissue in ring-porous wood has been shown to have a very different chemistry from surrounding sapwood (see Watmough and Hutchinson – this proceedings). Ring porous species therefore, appear to be unsuitable for environmental monitoring. Even though Cd concentrations in annual rings of sycamore, lime and beech were positively related to changes in Cd deposition, deposition only explained between 27 and 58% of the variation in wood Cd.  Furthermore, there was considerable variation in the mean concentration of Cd in wood of sycamore (1077 ppb), lime (263 ppb) and beech (333 ppb).  The mechanisms by which trees accumulate trace metals have yet to be completely described, although uptake from soil, foliage and/or bark are possible (6). Metal accumulation in wood will likely differ between tree species, leading to differences in absolute concentration.  When assessing the relative suitability of different tree species to metal monitoring, comparisons of metal patterns over time may be more illustrative than simply comparing metal concentrations.

 

There was much more year-to-year variation in Cd concentrations in lime and beech (only 2 trees sampled), than in sycamore (4 trees), and Cd concentrations in lime and beech were less strongly correlated to deposition.  As mentioned earlier, trees are not passive monitors of their environment, and the uptake and immobilization of metals is unique to individual trees of the same species and depends on local attributes of the soil, hydrology and canopy structure as well as physical characteristics of the individual tree. An individual core from a single tree only gives information on the Cd concentration in a small fraction of the ring produced in a given year, and is unlikely to reflect annual metal deposition at a site.  As more trees are sampled, however, differences between trees likely become less important and a more accurate interpretation of metal deposition at a given site may be obtained.  

 

Moreover, dendrochemistry is probably unsuitable for monitoring changes in Cd deposition remote from point sources where metal deposition values are low.  In this study, Cd deposition between 1975 and 1990 greatly exceeded typical deposition values reported for urban areas (~1 mg/m2), and decreased by more than an order of magnitude during the study period.  Despite this dramatic change in metal deposition, Cd concentrations in wood of sycamore, lime and beech only decreased by about 50% during the same period.  At deposition values typical of less polluted areas, Cd concentrations in wood are likely to be more affected by differences in soil chemistry (particularly soil pH) than by changes in deposition (7).

 

 

 

References

 

1.                  Guyette, R.P.; Henderson, G.S.; Cutter, B.E. (1992).  For. Sci. 38: 727-737.

2.                  Lepp, N.W. (1975).  Environ. Pollut. 9: 49-61.

3.                  Watmough, S.A. (1997).  Environ. Rev.  5: 181-201.

4.                  Dickinson, N.M., Watmough, S.A.; Turner, A.P. (1996). Environ. Rev.  4: 8-24.

5.                  Alloway, B.J. (1990) Heavy Metals in Soils, Blackie, London.

6.                  Watmough, S.A. (1999).  Environ. Pollut. 106: 391-403.

7.                  Watmough, S.A. (2000).  Wat. Air. Soil Pollut.  (submitted March 2000).

 

Figure 1.           Cd deposition at the woodland site (1975-1990)

 

 

 

Figure 2.           Cd concentrations in annual rings of sycamore, lime, beech and ash.