COMPARATIVE EFFECTIVENESS OF TILLANDSIA USNEOIDES L. AND PARMOTREMA PRAESOREDIOSUM (NYL.) HALE AS BIO-INDICATORS OF ATMOSPHERIC POLLUTION IN LOUISIANA (U.S.A.) F. B. PYATT1∗ , J. P. GRATTAN2 , D. LACY1 , A. J. PYATT1 and M. R. D. SEAWARD3 1 Department of Life Sciences, Nottingham Trent University, Nottingham, U.K.; 2 Institute of Geography and Earth Sciences, University of Wales, Aberystwyth, U.K.; 3 Department of
Environmental Science, University of Bradford, Bradford, U.K. (∗ author for correspondence, e-mail:
[email protected].
(Received 28 January 1997; accepted in revised form 1 April 1998)
Abstract. Samples of the epiphytic bromeliad Tillandsia usneoides L. (Bromeliaceae) and the lichen Parmotrema praesorediosum (Nyl.) Hale, growing on bald cypress trees in southern Louisiana, were chemically examined by means of X-ray micro-probe analysis to determine their comparative elemental content. The plants were found to effectively bio-accumulate heavy metals and sulfur from the atmosphere; the accumulatory capacities and implications are discussed. Partitioning occurs within the plants of T. usneoides and consequently any analytical procedure should standardise on precisely which parts are to be analysed. The bio-accumulation of certain heavy metals such as manganese, nickel and cadmium increases with age of the T. usneoides. Keywords: bio-accumulation, heavy metals, Louisiana, Parmotrema praesorediosum, partitioning, sulfur dioxide, Tillandsia usneoides, X-ray micro-probe analysis
1. Introduction Both the bromeliad Tillandsia usneoides L. (also known as Spanish moss) and the lichen Parmotrema praesorediosum (Nyl.) Hale are common in the southern states of the U.S.A., especially in swamplands; this investigation focused on the area of the Audubon Park adjacent to New Orleans in the sub-tropical state of Louisiana. T. usneoides there is an important and prolific epiphyte particularly on trees of Taxodium sp. It is pendulous and hangs 1 m or more from the lateral branches of the trees. The foliose lichen P. praesorediosum is horizontally positioned and covers extensive areas of both branches and trunks of the same trees and is also widespread in the Audubon Park area and the immediate vicinity. This lichen is a pan-tropical species but restricted in the U.S.A. to the south-eastern states; according to Hale (1965) it is apparently a weedy species of areas subjected to human disturbance. The bromeliad is absent from the town of New Orleans which is situated some 10 km from the Audubon Park, this is conceivably a response to the diminished humidity in the city and the enhanced values of some of the atmospherically transported pollutants. The lichen is largely absent from the New Water, Air, and Soil Pollution 111: 317–326, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.
318
F. B. PYATT ET AL.
Orleans vicinity although a very limited number of depauperate thalli are evident on the outer margins of the urban area and this too may largely be attributed to atmospheric pollution. Although many lichens and bromeliads have been researched in terms of their bio-accumulatory capacity, these morphologically and physiologically different plants have not been compared in terms of their relative effectiveness as bio-monitors. The site, with its nearby industrial activities, urban areas and power producing plants provided a unique opportunity to investigate this.
2. Experimental Procedure Material of both epiphytes was carefully collected during 1996 from three cypress trees (Taxodium sp.) in Audubon Park some 10 km from the centre of New Orleans, Louisiana. The trees chosen were of approximately the same height and growth form. Samples from each Taxodium tree were collected at a height of ca. 2 m from the ground and from the same south-westerly aspect, air-dried and subsequently transported to the laboratory in clean, sealed polythene bags. In the laboratory all samples were thoroughly washed in de-ionised water, air-dried, pooled according to type to give the following six sub-samples: 1. Apices of Tillandsia shoots to a distance of 2.5 cm from the tips. 2. Portions of Tillandsia shoots at a distance of 2.5–5.0 cm from the apices. 3. Older and more robust material of Tillandsia obtained 15.0–20.0 cm from the apices. 4. Bulked samples which were collected solely from the exterior of the Tillandsia wefts. 5. Bulked samples from within the Tillandsia wefts; these were conceivably partially protected from full exposure to atmospheric pollutants by the exterior parts of the plants. 6. Thalli of Parmotrema praesorediosum from which all the substratum material had been carefully removed without damage to the lichen thalli. Each sub-sample was individually mounted on a 13 mm diameter carbon stub and secured with conductive carbon cement. The samples were provided with a light coating of anti-static spray as described elsewhere (Pyatt and Lacy, 1988) and, when dry, placed in a Cambridge Stereoscan 600 electron microscope. Samples were analysed by electron probe X-ray microanalysis with a Link System 860 series 2 computer using a ZAF 4 program. Areas of 10 000 µm2 on each sample were examined by means of the microscope visual display monitor at a magnification of 500X for 100 sec of live time at 25 kv (electron beam energy). This procedure only analyses material from sodium upwards in the Periodic Table but has proved particularly useful in side-by-side comparisons of similar material as was required in this investigation. Ten replicates of each sub-sample were employed in each case
TILLANDSIA USNEOIDES AND PARMOTREMA PRAESOREDIOSUM AS BIO-INDICATORS
319
to separately determine the elemental/ionic content of the various morphological portions of Tillandsia and Parmotrema.
3. Results The results of the analyses are presented in Table I. There was no significant difference in the results from the analyses of the individual pooled replicates of each type which subsequently provided the means for the chemical constituents of the various samples being chemically analysed. The results (Table I) list all elements which were present above the detection level of the instrumentation employed in the samples analysed. To determine that the detected accumulation of elements in different samples was significant, and not the result of random variability, an analysis of variance test was conducted on the data for each element. The test employed was the KruskalWallis H test which can be used to determine whether there is a significant difference between three or more samples. Although little used in environmental sciences, it is a useful and powerful alternative to analysis of variance. Significance was determined at the 0.05 confidence level. This test is usefully employed in this type of study as it is a non-parametric test and does not rely on potentially unrealistic assumptions about the distribution of the variable. To apply this test, the data for the overall samples must first be ranked, from lowest to highest. Thus, for example, every value of chloride in the entire data set was assigned a rank. Identical values are given the mean of the ranking they would otherwise have received. The test is described fully and in detail in Ebdon (1990). 3.1. S ODIUM
AND CHLORIDE
These elements were generally well represented in all the Tillandsia samples, the sodium concentrations varying from 23.8% (Tillandsia sub-sample 3, old material), to the relatively low value of 0.7% in the lichen thalli; chloride had a maximum value of 42.3% (Tillandsia sub-sample 2), but was poorly represented in Parmotrema with a value of 3.9%. The concentration of Na and Cl in the Tillandsia apices was significantly different from the concentration measured in the external and internal weft samples. Slightly enhanced concentrations of both sodium (19.4%) and chloride (42.3%) occurred behind the apices, which suggests that partitioning had occurred, but this is not confirmed by the statistical analysis. Tillandsia proved the more effective bioaccumulator of both ions. The sodium and chloride values possibly reflect a response to the relative proximity of the coast but more particularly to the marine influence which extends up the tidal Mississippi River to the study site. An additional source of chloride is from urban and industrial fuel combustion sources. Shacklette and Connor (1973) have also reported enhanced sodium contents of bromeliads collected from coastal areas.
320
TABLE I Elemental content of T. usneoides and P. praesorediosum as a percentage of the total content (weight) of the elements detected Sub-sample 1 Apices Element 18.3 σ 35.2 σ 1.0 σ 7.5 σ 13.6 σ 0.2 σ 5.2 σ 5.1 σ 11.4 σ 0.1 σ 0.2 σ 1.3 σ 0.5 σ 0
0.62 2.57 0.32 1.52 2.33 0.04 0.47 0.25 1.44 0.01 0.07 0.14 0.1
3 Old material (15–20 cm from tips)
4 Bulk samples (External)
5
19.4 σ 42.3 σ 1.1 σ 5.0 σ 12.7 σ 0.2 σ 3.2 σ 2.9 σ 10.9 σ 0.1 σ 0.6 σ 0.5 σ 0.9 σ 0.2 σ
23.8 σ 40.4 σ 0.8 σ 5.1 σ 8.1 σ 0.1 σ 3.5 σ 4.9 σ 2.4 σ 0.2 σ 1.4 σ 0.6 σ 2.6 σ 1.0 σ
5.2 σ 16.6 σ 0.9 σ 10.7 σ 34.0 σ 1.0 σ 7.3 σ 4.9 σ 10.1 σ 0.3 σ 0.4 σ 4.5 σ 3.6 σ 0.2 σ
3.6 σ 15.7 σ 0.8 σ 6.9 σ 22.5 σ 0.7 σ 9.9 σ 3.2 σ 33.9 σ 0.2 σ 0.03 σ 1.8 σ 1.0 σ 0
0.78 2.59 0.31 1.19 2.38 0.03 0.36 0.12 2.02 0.01 0.12 0.14 0.32 0.03
2.5 2.59 0.15 1.18 1.99 0.01 0.22 0.42 1.56 0.05 0.11 0.14 0.09 0.15
1.31 2.58 0.16 1.47 2.4 0.09 0.52 0.37 1.78 0.02 0.12 0.11 0.37 0.07
6 Parmotrema
(Internal) 1.06 2.58 0.15 1.6 2.00 0.08 0.56 0.22 2.12 0.06 0.04 0.07 0.36
0.7 σ 3.9 σ 0.03 σ 8.6 σ 37.5 σ 2.9 σ 8.0 σ 4.7 σ 22.8 σ 0.15 σ 0.1 σ 4.4 σ 0.3 σ 0.3 σ
0.27 2.12 0.015 0.97 2.61 0.09 0.72 0.13 3.18 0.07 0.09 0.09 0.01 0.07
F. B. PYATT ET AL.
Na Cl Mg Al Si P S K Ca Cr Mn Fe Ni Cd
2 2.5–5.0 cm (from tips)
TILLANDSIA USNEOIDES AND PARMOTREMA PRAESOREDIOSUM AS BIO-INDICATORS
321
3.2. M AGNESIUM Enhanced concentrations of this cation were not detected in any of the sub-samples of Tillandsia, and the differences in concentration between different parts of the plant are not significant. The differences between the concentration of Mg in Tillandsia and Parmotrema are significant at the 0.05 confidence level. The lower values from the lichen thalli reflect the limited contribution of the photobiont to the thallus biomass. This element, like sodium and chloride, may be derived from a number of natural (e.g., marine influences) and anthropogenic sources. 3.3. A LUMINIUM
AND SILICON
These cations are derived from a number of sources including airborne soil and often shown comparable trends (Pyatt et al., 1995). The data obtained for Al could be divided into three significant groups by statistical analysis. Group 1 comprised Parmotrema and the apices and interior weft of Tillandsia; group 2, the older material of Tillandsia; and group 3, the external weft of Tillandsia sub-sample 4. A markedly enhanced concentration of Al was only detected in the outer plant body of Tillandsia which appears to act as an efficient filter mechanism for large diameter atmospherically transported particulates. The horizontally positioned lichen thalli were also relatively effective in accumulating this cation, probably due to processes including gravitational sedimentation. Silicon behaved similarly in terms of its bio-accumulation in both species, in that the maximum accumulation in Tillandsia was detected in the external weft of the plants. However, the lichen thalli also appear to be remarkably effective in concentrating Si. 3.4. P HOSPHORUS Whilst the majority of values are low, those in Parmotrema are significantly higher than those obtained from the material of the bromeliad. 3.5. S ULFUR Sulfur dioxide is a common atmospheric pollutant world-wide but, in coastal areas, seawater also represents an important source of sulfur. Statistical analysis identified two populations, sulfur was abundant in the lichen thalli and also in the material of the bromeliad obtained from within the plant wefts (sub-samples 4 and 5), and less abundant in the apices and shoots of Tillandsia (sub-samples 1–3). There was no evidence of translocation of sulfur from the younger (exposed) to the older portions of the plant. Benzing et al. (1992) exposed four species of Tillandsia (excluding T. usneoides) to sulfur dioxide and concluded that epiphytic Tillandsia spp. were more effective accumulators of sulfur than lichens, this was not statistically confirmed in the present study.
322
F. B. PYATT ET AL.
3.6. P OTASSIUM A high value of 5.1% was detected in the apices of T. usneoides (sub-sample 1), no doubt indicating its importance in developmental processes; but results obtained for the various sub-samples are not statistically different. 3.7. C ALCIUM This cation is derived from a number of sources including the erosion of soils. The peak values occurred within the plant weft of T. usneoides (33.8%) and on the horizontally presented lichen thalli (27.8%) and these two sub-samples are not significantly different. The values in the other sub-samples were broadly comparable with the highest values occurring in the bromeliad apices; the values are statistically indistinguishable from each other. There was no clear evidence of translocation of calcium to the older portions of the plants and its source may therefore not be atmospheric. 3.8. C HROMIUM Chromium was detectable in all samples analysed (Table I); the concentration varied between 0.1% to a maximum of 0.3% on the external parts of T. usneoides. There was no statistically verified difference between the concentrations of Cr in the various sub-samples analysed. This heavy metal is often generated by both the urban and industrial environment. In the New Orleans area, the boat repair industry and associated activities may contribute to atmospheric pollution in which this cation can be an important contributor. However, these data cannot, with any confidence, be used to infer deposition pathways for this element. 3.9. M ANGANESE This is another important particulate pollutant in urban/industrial areas; it is trapped particularly on the exterior of the bromeliad where the value was 0.4%. The bioaccumulation increased with the age of the material from 0.2% (terminal portions), 0.6% (2.5–5.0 cm from the apices) to a peak value of 1.4% in old portions of the plants. The concentration within the plant weft (0.03%) was markedly less than from the other samples and this may reflect the filtering of large diameter particulates by the outer regions of the plant body. The lichen showed less dramatic bio-accumulation which may appear surprising considering the longevity and physiology of lichens but Tillandsia possesses the advantage/disadvantage of its growth form which facilitates its capacity as a bio-accumulator. Statistical analysis of these data supports these observations and demonstrates marked bio-accumulation particularly by the older material of Tillandsia usneoides (Table I).
TILLANDSIA USNEOIDES AND PARMOTREMA PRAESOREDIOSUM AS BIO-INDICATORS
323
3.10. I RON In common with the other heavy metal particulate pollutants, this essential cation accumulated to a higher concentration on the outside rather than the inside of the T. usneoides wefts. The concentrations detected in the lichen thalli were also high. Statistical analysis of the data obtained from each species confirmed that the concentrations of Fe on the plant exteriors (lichens and bromeliads) were statistically indistinguishable; they had accumulated significantly enhanced concentrations of Fe in comparison with the other sub-samples. 3.11. N ICKEL T. usneoides was far more effective in bio-accumulating nickel than the thalli of P. praesorediosum (Table I). The highest concentration occurred on the outside of the bromeliad (3.6%) with a diminished concentration (1%) within the plant weft. Particulate nickel may be impacting on the surface of the plant and relatively less penetrates through the weft layers. There is some evidence for either translocation of nickel within the plant or, possibly, an age effect; thus the values are apices (0.5%), 2.5–5.0 cm from the apices (0.9%) and old portions (2.6%). 3.12. C ADMIUM This heavy metal was not recorded from the terminal portions of the plant or from inside the plant weft. A high concentration (1%) occurred in the older material. As is the case with both Ni and Mn, the tissue concentration of this element increased with age. The values for the other samples, including P. praesorediosum, were not significantly different.
4. Discussion Several surveys have been carried out on the bio-accumulation of lead in Tillandsia spp. including those of Martinez et al. (1971) and Robinson et al. (1973). Padaki et al. (1992) utilised material collected from a single location in east Texas and found concentrations in T. usneoides to be lower than those revealed in other studies utilising epiphytic plants as environmental monitors. Lead however was below the detection limit of 100 ppm and was not recorded in any sample of Tillandsia or Parmotrema during the current research program; this may conceivably indicate air quality improvement since the introduction of lead-free gasoline but may reflect the nature of the amount of shelter from the impact of atmospherically transported pollutants provided by vegetation-screening in the sampling area. It may further be noted that, in the case of the T. usneoides samples, some elements (Table I) generally tend to increase in concentration with the age of the
324
F. B. PYATT ET AL.
material being analysed (e.g., Na, Cl, Mn, Ni and Cd), others decrease in concentration (e.g., Al, Si, S, Ca and Fe) whilst others (e.g., Mg) show no significant trend. These changes reflect various features e.g., the capacity of young material, with a relatively high biomass of actively developing tissue, to effectively bio-accumulate certain elements from the atmosphere environment. Furthermore, it is apparent that when such organisms are employed as bio-indicators of atmospheric quality, it is of paramount importance that only comparable material is employed for analytical purposes. The technique, utilised in this current investigation, demonstrated the presence of 14 elements, including 5 heavy metals, in the materials examined. T. usneoides represents a natural, in situ version of the ‘Sphagnum bags’ developed by Goodman and Roberts (1971) and its use as a biomonitor has many advantages. The pollutants will reach both the bromeliad and the lichen as a result of processes such as sedimentation from a non-turbulent atmospheric environment, impaction and also accumulation from the pedosphere. Tillandsia usneoides is currently an abundant epiphyte on bald cypress trees in the swamp areas of southern Louisiana. It is absent from trees in the town of New Orleans and it is suggested that this absence is largely resultant from decreased habitat availability together with the reduced atmospheric humidity often characteristic of urban areas. It has been found to be a particularly effective bioaccumulator of atmospheric pollutants, both gaseous and particulate, and this may be attributed to its pendulous growth form with a high surface area to volume ratio. These features result in the plant being in intimate contact with the atmosphere. It was found to be at least as effective an accumulator of atmospherically transported pollutants at the lichen Parmotrema praesorediosum which is also epiphytic on the same species of tree. Whilst no lead (detection limit was 100 ppm) was recorded from samples collected in this protected and somewhat isolated area, it is interesting to note that Martinez et al. (1971) working on lead accumulation by Spanish moss noted ‘There must be a mechanism which concentrates this lead’. In many cases the pollutants had accumulated on the external surface of the dense pendulous wefts of the bromeliads; they may subsequently, to some extent, be mobilised during periods of extensive rainfall and eventually reach the pedosphere. However, in some cases, e.g., sulfur and calcium, the element occurred at a higher concentration within the plant body of the bromeliad; this is not surprising in the case of sulfur which readily, in the form of sulfur dioxide, penetrates such structures and will ultimately accumulate within the tissue rather than exclusively on the surface of the plant cuticle. In some cases, the actively growing apices were found to exhibit enhanced concentrations of elements, for example potassium, and there was some evidence of translocation to older parts (or a direct age effect) of the plant, e.g., manganese, nickel and cadmium. It is also apparent that partitioning of elements occurs within the plant body of the bromeliad; hence in comparisons of pollution utilising this bio-indicator, great care must be taken to ensure the comparability of the age of the material being utilised.
TILLANDSIA USNEOIDES AND PARMOTREMA PRAESOREDIOSUM AS BIO-INDICATORS
325
The ranking of elements varied, in the case of T. usneoides, with the age of the material being examined and this may be illustrated by reference to chromium, manganese, iron, nickel and cadmium. Thus: apices : Fe > Ni > Mn > Cr > Cd 2.5 − 5.0 cm from apices : Ni > Mn > Fe > Cd > Cr 15 − 20 cm apices : Ni > Mn > Cd > Fe > Cr Bulked external samples : Fe > Ni > Mn > Cr > Cd Bulked internal samples : Fe > Ni > Cr > Mn > Cd Parmotrema praesorediosum : Fe > Ni/Cd > Cr > Mn Plants of T. usneoides will eventually become detached from the tree as a consequence of various agencies and will then decompose to release any bio-accumulated nutrients and heavy metals to the pedosphere where they will be available for recycling through the vulnerable swamp ecosystem. Some elements will be leached through the soil/sediment whilst a proportion will be mobilised through the ground water and will ultimately enter the Mississippi and thence the Gulf of Mexico; i.e., elements bio-accumulated by T. usneoides will ultimately be released, diluted and thence transferred to other ecosystems where the processes of bio-accumulation will continue. If the concentration of the bulk elements (Na through Ca in Table I) is assumed equal in Tillandsia and Parmotrema, the effectiveness of both species as accumulators of heavy metals can be inferred from Table I. For one of the heavy metals considered (Ni), Tillandsia is clearly more effective than Parmotrema. For the other heavy metals (Cr, Mn, Fe and Cd) the differences between the two species are generally only small (or not statistically significant). It can be concluded that, in pollution studies employing bio-accumulation, Tillandsia is probably at least as effective as an accumulator as the epiphytic lichen Parmotrema. In such studies, Tillandsia might be more appropriate for use as a consequence of its current greater abundance.
References Benzing, D. H., Arditti, J., Nymon, L. P., Temple, P. J. and Bennett, J. P.: 1992, Environ. Exp. Bot. 32, (1), 25. Ebdon, D.: 1990, Statistics in Geography, Oxford. Basil Blackwell. Goodman, G. T. and Roberts, T. M.: 1971, Nature 231, 287. Hale, M. E.: 1965, ‘A Monograph of Parmelia subgenus Amphigymnia’ Contr. U.S. Nat. Herb. 36, 193
326
F. B. PYATT ET AL.
Martinez, J. D., Nathany, M. and Dharmarajan, V.: 1971, Nature 233, 564. Padaki, P. M., McWilliams, E. L. and James, W. D.: 1992, J. of Radional. and Nuclear Chem. 161, 147. Pyatt, F. B., Grattan, J. P., Lacy, D., Gilbertson, D. D., Brayshay, B. A. and Wadsworth, W. A.: 1995, Scottish Geographical Magazine 111, 106. Pyatt, F. B. and Lacy, D.: 1988, Environ. Intern. 14, 407. Robinson, J. W., Christian, C. M., Martinez, J. D. and Nathany, M.: 1973, Environ. Lett. 4(2), 87. Shacklette, H. T. and Connor, J. J.: 1973, Airborne Chemical elements in Spanish Moss. Statistical studies in field geochemistry. Geological Survey Professional Paper 574-E. United States Government Printing Office, Washington.
