- Email: [email protected]
Lead and zinc in the Wallsend Burn, an urban catchment in Tyneside, UK
The Science of the Total Environment 269 Ž2001. 49᎐63
Lead and zinc in the Wallsend Burn, an urban catchment in Tyneside, UK A. Mellor U Di¨ ision of Geography and En¨ ironmental Management, Lipman Building, Uni¨ ersity of Northumbria, Newcastle upon Tyne, Tyne and Wear, NE25 9DG, UK Received 14 December 1999; accepted 2 September 2000
Abstract This paper examines lead and zinc concentrations in topsoils and stream sediments of public access areas in an urban catchment in Tyneside, UK. It examines the extent and severity of metal contamination, explores spatial patterns in relation to urban and industrial development, and makes inferences about potential metal mobility. Total and acetic-acid extractable lead and zinc concentrations, organic content and pH were determined on 121 topsoil and 22 stream sediment samples using standard laboratory procedures. Using the lowest trigger thresholds for total lead and zinc, almost 75% and 91%, respectively, of topsoil samples were classified as contaminated; proportions were rather lower for acetic acid extractable metals. Similarly, approximately 45% and 95% of stream sediment samples were contaminated with lead and zinc, respectively. The spatial distribution of metal concentrations was characterized by a hotspot pattern, with highest values in central and southern parts of the catchment where there is a long urban and industrial history. The potential mobility of zinc is considerably greater than that of lead in both topsoils and stream sediments, and for both metals is slightly higher in the stream sediments than in the topsoils; both of these differences are statistically significant Ž P- 0.05.. The implications of the findings in this paper for assessment and monitoring of metal contaminated areas are explored. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Heavy metals; Lead; Zinc; Urban catchment; Soils; Stream sediments
1. Introduction The widespread occurrence of heavy metal contamination in urban areas and its potential imU
Tel.: q44-191-227-3758; fax: q44-191-227-4715. E-mail address: [email protected] ŽA. Mellor..
pact on human health are now well documented in the literature Že.g. Thornton, 1991; BEES Environmental Health, 1997; Oliver, 1997; Mielke and Reagan, 1998.. Such contamination has been reported in soils, airborne dusts, road sweepings, road-side vegetation, produce from urban allotments, urban stream water and fluvial and estuarine sediments ŽCulbard et al., 1988; Moir and
0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 8 1 1 - 1
50
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Thornton, 1989; Bubb and Lester, 1994; Kelly et al., 1996; Macklin, 1996.. Uptake of heavy metals by humans, largely through respiratory or digestive pathways, has contributed to elevated concentrations of such metals in blood, bone and tissue samples, and consequently to poor health, in a number of urban populations ŽBEES Environmental Health, 1997; Mielke et al., 1999.. Of particular concern is the strong association between elevated lead concentrations in blood samples from children, as a result of dust inhalation and pica-related ingestion, and impaired mental development Že.g. Ericson and Mishra, 1994; Jin et al., 1997.. Despite the potential health implications, there have been relatively few studies, conducted at an appropriately fine spatial scale, of the severity and spatial distribution of heavy metal pollution in urban areas. Douglas et al. Ž1993. examined the spatial distribution of lead in soils in part of south Manchester using a 70-m sampling grid, whilst Kelly et al. Ž1996. collected four topsoil samples per km2 in a study of heavy metals in Richmond upon Thames and Wolverhampton. In Tyneside, Aspinall et al. Ž1988. collected topsoils from relatively undisturbed grassy areas in public open spaces using a 1-km2 grid. Point data were then aggregated to 1981 census wards in an attempt to assess the number and spatial distribution of people exposed to elevated heavy metal concentrations. Mellor and Bevan Ž1999. examined the spatial distribution of lead in topsoils and stream sediments of public open spaces in the Ouseburn catchment in Newcastle upon Tyne. Here topsoil samples were collected using a 200-m grid, and stream sediments were collected at 500m intervals. In many of these studies, spatial patterns of heavy metal contamination were difficult to explain due to the complex industrial history and continued overprinting of pollution in urban areas. Indeed, it is only in the last 20 years or so, through the introduction of environmental monitoring and legislation, that such pollution has been documented. This study focused on lead and zinc in the topsoils and stream sediments of Wallsend Burn, a largely urban catchment in Tyneside, UK. Its aims were as follows:
䢇
䢇
䢇
to examine the extent and relative severity of lead and zinc contamination within the catchment; to explore the spatial distribution of the two metals and to relate this to former and present patterns of urban and industrial development within the catchment; to make inferences about the potential mobility of lead and zinc within the catchment.
The implications of the findings of this study for assessment and monitoring of soil and stream sediment pollution in urban areas were also considered.
2. Study area The Wallsend Burn is a 6-km-long lowland stream that drains an area of approximately 20 km2 to the east of Newcastle upon Tyne ŽFig. 1.. The central and northern parts of the catchment are mainly suburban with relatively low relief, whilst the southern part is rather steeper and more urban and industrial in character. The solid geology consists of carboniferous coal measures, including a mixture of sandstones, shales and coal bands. Superficial deposits consist largely of glacial till which covers most of the catchment. Dominant soil types are gleys and brown earths of the Brickfield and Hallsworth Series ŽJarvis et al., 1984.. The catchment is traversed by a major road ŽA1056., which runs east᎐west across its centre, linking the city of Newcastle upon Tyne with the coastal towns of Whitley Bay, North Shields and Tynemouth ŽFig. 1.. To the north of this, the land is relatively open with a number of public green spaces and recreational areas, and a small amount of agricultural land. These spaces are punctuated with relatively new housing estates, such as Hadrian Park and Battle Hill, mostly constructed since 1970. Although there are some open spaces in the south of the catchment, such as Bigges Main golf course, Wallsend Dene and Holy Cross, much of this area is characterized by a dense network of older Žmostly pre-1930s. housing, roads and industrial sites.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
51
Fig. 1. Location map.
These differences in density of development are reflected in the character of the stream network. In the northern half of the catchment, the streams are relatively small and discontinuous ŽFig. 1.. This is due to extensive channelization and the construction of underground drainage around and beneath the newer housing estates and roads. In the southern part of the catchment, stream channels are larger and more continuous, with some channelization, particularly in and just above the lower tidal reaches. Here, stream flow is greater, largely due to surface drainage from impermeable urban surfaces and the three combined sewer overflow ŽCSO. discharge points. Industry, which comprises approximately 5% of the catchment area, discharges its effluent directly to sewer; hence industrial contamination should only occur at times of CSO. The main activities include chemical industries, oil rig construction and fitting, waste processing and incineration, scrap metal processing, and former gas and coke works.
Most of these activities are located in the south of the catchment. Historically coal mining has been a significant industry within the catchment, notably at Bigges Main between Walkerville and Wallsend, the Rising Sun to the north of Battle Hill, and at Howdon ŽFig. 1.. Deep mining has now ceased in the catchment, and these sites have since been restored as recreational and amenity areas such as Bigges Main golf course and the Rising Sun Country Park. Other public open spaces include large areas of sports fields at Benton and Holy Cross, Wallsend Dene and Richardson-Dees Park in Wallsend.
3. Methodology Using a combination of Ordnance Survey maps Ž1:25 000 scale., colour air photographs Ž1:10 000 scale. and site visits as appropriate, the catch-
52
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 2. Locations of Ža. topsoil sampling sites and selected industrial activities, and Žb. stream sediment sampling sites and combined sewer overflows.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
ment area was defined. All public access land areas and stream networks were then mapped, together with a number of potential sources of heavy metal contamination, notably sites of former industrial activity such as areas of colliery spoil deposition, former and present waste disposal sites, major road networks and CSOs. Mapped information was then entered onto a geographical information system ŽGIS. using Arc-View software to facilitate spatial representation. Public access areas were systematically sampled by a 200-m grid over the appropriate parts of the catchment, amounting to a total of 121 topsoil Ž0᎐10 cm. samples ŽFig. 2.. Twenty-two stream sediment samples were collected at 200-m intervals along the main stream and its tributaries working upstream from the high tide limit ŽFig. 2.. All samples consisted of an aggregate of five sub-samples collected using a trowel from the corners and centre of a 1-m2 quadrat. The soil and sediment samples were air-dried, ground with a mortar and pestle and passed through a 2-mm sieve. Lead and zinc contents Žboth total and acetic acid extractable ., organic matter content and pH were then determined. Total lead and zinc were extracted by treating 0.5 g of sample with a mixture of 10 ml concentrated nitric and 5 ml hydrochloric acids in a microwave digestor; the extract was then diluted to 100 ml with distilled water. The more mobile and bioavailable lead and zinc were extracted by treating 5 g of sample with 100 ml of 0.5 M acetic acid ŽAspinall et al., 1988.. Although the latter extraction is not widely used, it was selected in this investigation to facilitate comparison with other similar studies in the Tyneside region ŽAspinall et al., 1988; Macklin, 1996; Mellor and Bevan, 1999.. The extracts were then filtered, and lead and zinc concentrations were determined using a PerkinElmer 2380 double beam atomic absorption spectrophotometer. Results were validated using the same analytical procedures on 20 sub-samples from a standard reference soil ŽORM Laboratory of the Government Chemist, reference number GBW07406.; at least 95% of the sub-samples exhibited lead and zinc concentrations within "2 S.E. of the means quoted by the supplier. Organic
53
matter content was determined by loss on ignition and pH was established from 10 g of sample suspended in 25 ml of distilled water using a Jenway 3020 pH probe ŽRowell, 1994..
4. Results Descriptive statistics for all properties examined are shown in Table 1. With the exception of organic matter content and pH, all properties had mean values in excess of median values and high standard deviations reflecting positively skewed distributions and a high degree of variation. Mean total lead and zinc concentrations in the topsoils were 130 mgrkg and 282 mgrkg, respectively. These concentrations are comparable with those from other investigations conducted on soils of urban areas in the UK. For example, Culbard et al. Ž1988. obtained geometric mean lead concentrations ranging from 80 to 1300 mgrkg from garden soils in a number of urban locations, including Newcastle upon Tyne Ž350 mgrkg.. Mean total lead concentration in the stream sediments Ž103 mgrkg. was slightly Table 1 Descriptive statistics for topsoils Ž n s 121. and stream sediments Ž n s 22. Property
Mean
Median
Total Pb: topsoils Žmgrkg. Total Pb: stream sediments Žmgrkg. Total Zn: topsoils Žmgrkg. Total Zn: stream sediments Žmgrkg. Acetic acid extractable Pb: topsoils Žmgrkg. Acetic acid extractable Pb: stream sediments Žmgrkg. Acetic acid extractable Zn: topsoils Žmgrkg. Acetic acid extractable Zn: stream sediments Žmgrkg. Organic content: topsoils Ž%. Organic content: stream sediments Ž%. Acidity ŽpH.: topsoils Acidity ŽpH.: stream sediments
129 103
115 82
69 67
282 473
253 267
138 571
6
5
7
10
7
11
62
35
66
134
91
119
18 17
17 16
4 7
6.1 7.4
6.0 7.5
S.D.
0.9 0.7
54
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
lower than in the topsoils, whereas that for zinc Ž473 mgrkg. was considerably higher ŽTable 1.. Total lead concentrations are comparable with those of Macklin Ž1996. and Mellor and Bevan Ž1999. who analysed stream sediment samples from the river Ouseburn a few kilometres to the west of the study area Žmean concentration s 68 mgrkg.. Mean acetic acid extractable lead and zinc concentrations in the topsoils were considerably lower than those for the total metals at 6 mgrkg and 62 mgrkg ŽTable 1., and on average constituted 4.7% and 23.4% of the total metals, respectively. These concentrations compare well with those of Mellor and Bevan Ž1999. who obtained a mean lead concentration of 16 mgrkg. Mean acetic acid extractable lead and zinc concentrations in the stream sediments were 11 mgrkg and 134 mgrkg, respectively, both of which were considerably higher than those in the topsoils. On average, this fraction of lead and zinc constituted 9.4% and 29.8% of the total metals, respectively, these proportions being slightly higher than those in the topsoils. Once again, concentrations of acetic acid extractable lead compare well with those in stream sediments of the adjacent river Ouseburn Žmean s 15 mgrkg. ŽMellor and Bevan, 1999.. Mean organic matter contents in the topsoils and stream sediments were 18% and 17%, and their pH values 6.1 and 7.4, respectively. These values are comparable with those in grassland and woodland soils elsewhere in northern England ŽJarvis et al., 1984.. Spatial distributions of total and acetic acid extractable lead and zinc in the topsoils and stream sediments are shown in Figs. 3᎐6. Highest concentrations were found in southern and central parts of the catchment, although the spatial pattern for zinc was somewhat more scattered than that for lead, particularly in the stream sediments. Although organic matter contents and pH values did not display any clear spatial patterns, organic contents were highest in wooded and less disturbed green spaces, and in the smaller tributary streams. Similarly, pH values were lowest in less disturbed wooded areas and highest in areas where agrochemicals had been used Že.g.
adjacent to golf courses.; pH values in the stream sediments were consistently higher than in the topsoils.
5. Discussion 5.1. Extent and se¨ erity of contamination A number of widely recognized systems have been used to define contamination thresholds in soils, although none exists for the designation of contamination in stream sediments. In the system adopted by the UK ŽICRCL, 1987., trigger concentrations are only provided for the total metal and not for more mobile, bioavailable forms. Metals are also separated into group A contaminants which may be hazardous to health Že.g. lead. and group B contaminants which are phytotoxic but not normally hazardous to health Že.g. zinc.. In addition, trigger concentrations for group A contaminants differ according to land use, so that in domestic gardens and allotments, the figure for lead is 500 mgrkg whereas in parks, playing fields and open spaces it is 2000 mgrkg. Trigger concentrations for group B contaminants are not differentiated in this way and are only applicable to soil in which produce for human and animal consumption is grown; the trigger concentration for zinc is 300 mgrkg. Application of the trigger concentration for lead to the topsoils and stream sediments from Wallsend Burn reveals that none of the samples was significantly contaminated. It should be noted, however, that 31% of topsoils and 45% of stream sediments have zinc concentrations in excess of the trigger threshold. In response to recent changes in the Environmental Protection Act Ž1995. the ICRCL Ž1987. system is soon to be replaced by a risk assessment model based on identification of contaminant sources, pathways and receptors, and new Contaminated Land Exposure Assessment ŽCLEA. guidelines. For this approach to be successful, contaminant guidelines must accommodate the issue of bioavailability which will vary between different metallic forms. The guidelines must also be readily interpreted by land remedia-
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 3. Total metal concentrations in topsoils: Ža. lead, Žb. zinc.
55
56
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 4. Total metal concentrations in stream sediments: Ža. lead, Žb. zinc.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 5. Acetic acid extractable metal concentrations in topsoils: Ža. lead, Žb. zinc.
57
58
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 6. Acetic acid extractable metal concentrations in stream sediments: Ža. lead, Žb. zinc.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
tion practitioners, site owners and by Local Authorities whose task it is to establish contaminated land registers. In the Netherlands, a more stringent system based on much lower threshold concentrations is in use ŽDepartment of Soil Protection, 1996.. The Dutch system also specifies action values which, if exceeded, signal the need for clean up operations to be carried out. The lower threshold or ‘optimum’ concentrations for lead and zinc are 85 mgrkg and 140 mgrkg, respectively. Similarly, the action values for these metals are 530 mgrkg and 720 mgrkg. Seventy-five percent of topsoil and 45% of stream sediment samples from Wallsend Burn exceed the lower threshold with respect to total lead, whilst none of the samples exceed the higher action concentration. Similarly, 91% of topsoil and 95% of stream sediment samples from Wallsend Burn exceed the lower threshold with respect to zinc, whilst 2% of topsoil and 14% of stream sediment samples exceed the action concentration. Using a more statistical approach ŽDavies, 1983., Aspinall et al. Ž1988. determined the highest probable concentrations of total and available lead and zinc in uncontaminated soils in Tyneside. For total lead and zinc, these were 80 mgrkg and 345 mgrkg, respectively, and for available lead and zinc they were 14 mgrkg and 11 mgrkg. This study concluded that more than two thirds of the Tyneside soils were contaminated with respect to lead, although the proportion was considerably lower for zinc. The proportion of the population living in contaminated areas, calculated on the basis of census wards, was also found to be high at 84% and 12% for total lead and zinc, respectively. With reference to the samples collected from Wallsend Burn, it can be seen that 76% of topsoils and 55% of stream sediments exceed the threshold concentration for total lead, whilst only 7% of topsoils and 14% of stream sediments are in excess of that for available lead. In relation to total zinc, 26% of topsoils and 41% of stream sediments exceed the threshold concentration, whilst all samples exceed that for available zinc.
59
5.2. Spatial distribution The difference in spatial pattern between lead and zinc could be explained by the fact that zinc is generally more mobile in soils and sediments than lead ŽMerrington and Alloway, 1994; Lee and Touray, 1998.. It may also be explained by differences in the sources of the two metals, much of the lead having a more localized urban and industrial source, and much of the zinc having a more widespread geological source, notably the carboniferous shales ŽPrade et al., 1991.. Superimposed upon the catchment-wide distribution of lead and zinc are more localized ‘hotspot’ patterns where sites with high metal concentrations are surrounded by sites with much lower concentrations. This spatial pattern is a common feature of many types of geochemical pollution and has been reported for lead contamination in the adjacent Ouseburn catchment ŽMellor and Bevan, 1999.. Elevated lead and zinc concentrations in the southern part of the catchment are likely to reflect the long history of urban and industrial development. Potential sources of these metals, and of lead in particular, include vehicle emissions from the dense network of roads, a pattern that has been widely reported in other urban areas Že.g. Douglas et al., 1993; Sanka et al., 1995.. This influence can be seen most clearly at sites adjacent to the main coast road where relatively high lead concentrations can be found ŽFigs. 1 and 2a and Fig. 4a.. Elevated metal concentrations are also associated with open spaces adjacent to older housing and demolition sites which are prevalent in the south of the catchment. Similar associations have been reported in Manchester ŽDouglas et al., 1993. and in Richmond upon Thames and Wolverhampton ŽKelly et al., 1996. where leaded paint used in older properties was identified as an important source of lead in soils. Another possible source of elevated metal concentrations in the south of the catchment is the metal smelting and finishing, and glass-making factories in the lower Ouseburn ŽMellor and Bevan, 1999.. This area
60
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
lies to the south-west of the catchment in a generally up-wind direction, and although these industries have now ceased, it is likely that airborne pollution from them would have been widespread in the past. Such industrial sources of heavy metals in soils have been reported elsewhere Že.g. Bloeman et al., 1995.. Another possible source of heavy metals in soils of the Wallsend Burn catchment is the carboniferous pyritic shale ŽPrade et al., 1991. which constitutes a significant proportion of the coalbearing geology of this area. Elevated lead and zinc concentrations are found at the Bigges Main Golf course in the south-west of the catchment and at the Rising Sun Country Park in the central area, both of which represent extensive areas of colliery spoil restoration. The association between elevated lead concentrations in topsoils and restored colliery spoil sites was also reported in the Ouseburn catchment, only a few kilometres to the west of the Wallsend Burn catchment ŽMellor and Bevan, 1999.. More local examples of pollution sources include the former waste incinerator, a gas works site and former metal scrap yard at Howdon in the south-east of the catchment. Unlike soils, stream sediments undergo phases of transfer and post-depositional mobilization, interspersed with periods of temporary storage and stability ŽFoster and Charlesworth, 1996.. Such temporal and spatial complexity creates difficulties when attempting to interpret spatial patterns of heavy metals ŽMacklin, 1996.. Highest concentrations of lead and zinc are found in the more urban, and formerly more industrialized, lower reaches of the stream. Here, surface drainage during prolonged or intense rainfall events may carry significant quantities of metals from impermeable surfaces, including roads and car parks. Combined Sewer Overflow, which is common during such events, may also represent another important source of lead and zinc in this lower part of the catchment, as has been reported in the adjacent Ouseburn catchment ŽMellor and Bevan, 1999. and elsewhere ŽDeely and Ferguson, 1994.. 5.3. Potential metal mobility A crude estimate of potential lead and zinc
mobility within topsoils and stream sediments of the Wallsend Burn catchment can be gained through examination of the concentrations of acetic acid extractable metals expressed as a percentage of the total metals. In the topsoils the mean values for lead and zinc are 4.7% and 23.4%, respectively, whilst in the stream sediments they are 9.4% and 29.8%. In both cases, these differences were found to be statistically significant following t-test analysis Žfor topsoils t s 9.11, P- 0.001; for stream sediments t s 8.33, P- 0.001.. It therefore appears that the potential mobility of zinc is considerably greater than that for lead in both topsoils and stream sediments of the Wallsend Burn catchment. The difference in potential mobility between these metals may be accounted for by the fact that zinc is an essential micronutrient in many plants and, as a consequence, is likely to be more soluble and mobile in soil᎐plant ecosystems than lead which plays no known beneficial role in living organisms ŽRoss, 1994.. In contrast, lead from industrial and urban pollution sources is dominated by relatively inert particulate forms which are stable and persistent in soils and sediments ŽMellor and Bevan, 1999.. In addition to differences in potential mobility between lead and zinc, there are differences in the potential mobility of both metals between the topsoils and stream sediments, with slightly higher percentages in the stream sediments. Once again, these differences were found to be statistically significant following t-test analysis Žfor lead t s 13.23, P- 0.001; for zinc t s 5.43, P- 0.001.. The differences in potential mobility may result from differences in both the sources of these metals and the geochemical environments between topsoils and stream sediments ŽMellor and Bevan, 1999.. The influences of organic content and pH on the behaviour and potential mobility of heavy metals, particularly in soils, is well documented Že.g. Nelson and Campbell, 1991; Bryan and Langston, 1992; Hargitai, 1995; Vanstraalen and Bergema, 1995; Sauve et al., 1997; Hooda and Alloway, 1998.. In the topsoils and stream sediments of the Wallsend Burn catchment, these influences were explored using Pearson correlation analysis. However, the only statistically sig-
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
nificant correlations found in the topsoils were those between organic content and total lead Žq0.408, P- 0.001., and organic content and total zinc Žq0.309, P- 0.01.. None of the correlations between stream sediment properties was found to be statistically significant. The absence of significant correlations with organic content in the stream sediments suggests that the type of organic matter could be just as important as the overall content. Hargitai Ž1995., e.g. indicates that although Pb is strongly organophilic, it is associated most strongly with the humic fraction, the presence of which is likely to be limited in stream sediments. Although metal mobility is known to increase with decreasing pH ŽNelson and Campbell, 1991; Vanstraalen and Bergema, 1995., the absence of any correlations with this property suggests that the pH range was too narrow to demonstrate such relationships in the topsoils and stream sediments of Wallsend Burn ŽMellor and Bevan, 1999..
6. Conclusions The extent and severity of lead and zinc contamination in topsoils and stream sediments depends on which of the many systems of trigger thresholds is adopted. Using the most stringent of the systems considered here, more than two thirds of topsoil and stream sediment samples from Wallsend Burn are found to be contaminated. In general, however, zinc contamination appears to be more significant than lead contamination in both topsoils and stream sediments, and there appears to be no clear difference in the severity of contamination between topsoils and stream sediments. Highest concentrations of lead and zinc are found in southern and central parts of the catchment, where urban and industrial development have been greatest, although the spatial pattern for zinc is somewhat more scattered than that for lead. As with all spatial sampling, the density of the sampling network can considerably affect the conclusions drawn ŽMellor and Bevan, 1999.. The spatial complexity of lead and zinc distributions, often in the form of contaminated hot spots,
61
creates particular difficulties when designing sampling strategies to assess and monitor soil and sediment contamination; such strategies must achieve a balance between representativeness and cost. Sampling design may be optimized using a geostatistical approach where spatial variability and interpolation are used to predict and model contaminant concentrations in un-sampled areas Že.g. Kuzel et al., 1994.. The potential mobility of zinc is significantly greater than that of lead in both topsoils and stream sediments of Wallsend Burn. This observation may reflect the fact that zinc is more mobile in soil-plant systems, whilst lead is dominated by relatively inert, particulate forms that are more stable and persistent in soils and sediments. Both metals are found to be significantly more mobile in the stream sediments than in the topsoils, possibly reflecting differences in geochemical environment and metal sources. The complex behaviour of stream sediments over time, in comparison with the relative stability of soils, makes interpretation of spatial patterns difficult. In order to develop a fuller understanding of the behaviour and fate of heavy metals in the environment, it is recommended that investigations be extended beyond that of the total fractions to consider the various different forms of these metals ŽRamos et al., 1994; Floresrodriguez et al., 1994; Macklin, 1996.. It is suggested that there is a need for detailed baseline investigations of the spatial distribution of heavy metals, and other pollutants, in urban areas, perhaps with a view to formulating a database that would be of use to Local Authorities, organizations involved in monitoring and assessment of contamination, and other parties concerned with the planning and development process ŽBullock, 1987; Macklin, 1996.. For example, a more thorough inventory of contaminants at urban ‘brownfield’ sites may facilitate their development in preference to the more precious and often more pressurized rural ‘greenfield’ sites ŽMellor and Bevan, 1999.. It is further suggested that such investigations should take a holistic approach by integrating data collection across the soil and stream systems of catchment areas, rather than focusing on soils alone. Not only would this
62
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
allow development of a more thorough understanding of the behaviour and fate of pollutants in the environment, but it would also parallel the more integrated approach to monitoring and assessment of the environment adopted by the UK Environment Agency, and emphasized in their Local Environment Agency Plans ŽLEAPs.. The catchment, now widely recognized as a fundamental unit of environmental management, is central to these plans. Finally, such an approach could integrate well with the new risk assessment approach to contaminant guidelines ŽCLEA. based on identification and quantification of contaminant source᎐pathway᎐receptor linkages.
Acknowledgements This work was funded by the University of Northumbria at Newcastle from monies allocated by the Higher Education Funding Council as part of the 1992 Research Assessment Exercise. The author would like to thank Miss Joanne Cooke for analysis of soil and stream sediment samples, with support from Miss Joyce Moxon and Mrs Ruth Steinberg, Mr Keith Turner and Miss Caroline Dilks for Research into the catchment history, processing of data and associated information on GIS and production of diagrams, and Mr Gary Haley for production of Fig. 1 Žlocation map.. References Aspinall R, Macklin M, Openshaw S. Heavy metal contamination in soils of Tyneside: a geographically-based assessment of environmental quality in an urban area. In: Hooke JM, editor. Geomorphology in environmental planning. Chichester: Wiley, 1988:87᎐102. BEES Environmental Health. Lead, 1997Žwww.beesinc. orgraboutrheallead.html.. Bloeman ML, Markert B, Lieth H. The distribution of Cd, Cu, Pb and Zn in topsoils of Osnabruck in relation to land-use. Sci Total Environ 1995;166:137᎐148. Bryan GW, Langston WJ. Bioavailability, accumulation and effects of heavy-metals in sediments with special reference to United Kingdom Estuaries ᎏ a review. Environ Pollut 1992;76:89᎐131. Bubb JM, Lester JN. Anthropogenic heavy-metal inputs to
lowland river systems, a case study ᎏ the River Stour, UK, Water. Air Soil Pollut 1994;78:279᎐296. Bullock P. Soil protection-a need for a European programme. In: Barth H, Hermite PL, editors. Scientific Basis for Soil Protection in the European Community. Amsterdam: Elsevier, 1987:581᎐587. Culbard EB, Thornton I, Watt J, Wheatly M, Moorcroft S, Thompson M. Metal contamination in British suburban dusts and soils. J Environ Qual 1988;12:226᎐234. Davies BE. A graphical estimation of the normal lead content of some British soils. Geoderma 1983;29:67᎐75. Deely JM, Ferguson JE. Heavy-metal and organic matter concentrations and distributions in dated sediments of a small estuary adjacent to a small urban area. Sci Total Environ 1994;153:97᎐111. Department of Soil Protection. Intervention values and target values ᎏ soil quality standards. The Hague, Netherlands: Ministry of Housing, Spatial Planning and Environment, 1996. Douglas I, Al-Ali J, Clarke M. Lead contamination in Manchester. Land Contamination Reclamation 1993;1:143᎐148. Ericson JE, Mishra SI. Soil lead concentrations and prevalence of hyperactive behavior among school children in Ottawa, Canada. Environ Ecol 1994;16:247᎐256. Floresrodriguez J, Bussy AL, Thevenot DR. Toxic metals in urban runoff ᎏ physicochemical mobility assessment using speciation schemes. Water Science Technol 1994;29:83᎐93. Foster IDL, Charlesworth SM. Heavy metals in the hydrological cycle ᎏ trends and explanation. Hydrol Process 1996;10:227᎐261. Hargitai L. Some aspects of the mobility and distribution of toxic heavy-metal contaminants in soil profiles and river sediments. Int J Environ Anal Chem 1995;59:317᎐325. Hooda PS, Alloway BJ. Cadmium and lead sorption behaviour of selected English and Indian soils. Geoderma 1998;84:121᎐134. Interdepartmental Committee for the Redevelopment of Contaminated Land. Guidance on the assessment and redevelopment of contaminated land. ICRCL Paper 59r83. 2nd ed London: Department of the Environment, 1987. Jarvis RA, Bendelow VC, Bradley RI et al. Soils and their use in northern England. Soil Survey of England and Wales Bulletin, 10. Harpenden, 1984. Jin A, Teschke K, Copes R. The relationship of lead in soil to lead in blood and implications for standard setting. Sci Total Environ 1997;208:23᎐40. Kelly J, Thornton I, Simpson PR. Urban geochemistry ᎏ a study of the influence of anthropogenic activity on the heavy metal content of soils in traditionally industrial and nonindustrial areas of Britain. Appl Geochem 1996; 11:363᎐370. Kuzel S, Nydl V, Kolar L, Tichy R. Spatial variability of cadmium, pH, organic matter in soil and its dependence on sampling scales, Water. Air Soil Pollut 1994;78:51᎐59. Lee PK, Touray JC. Characteristic of polluted artificial soil
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63 located along a motorway and effects of acidification on the leaching behaviour of heavy metals ŽPb, Zn, Cd.. Water Res 1998;32:3425᎐3435. Macklin M. Fluxes and storage of sediment-associated heavy metals in floodplain systems: assessment and river basin management issues at a time of rapid environmental change. In: Anderson MG, Walling DE, Bates PD, editors. Floodplain processes. Chichester: Wiley, 1996:441᎐460. Mellor A, Bevan JR. Lead in the soils and stream sediments of an urban catchment in Tyneside, UK. Water Air Soil Pollut 1999;112:327᎐348. Merrington G, Alloway BJ. The flux of Cd, Cu, Pb and Zn in mining polluted soils. Water Air Soil Pollut 1994;72: 333᎐344. Mielke HW, Reagan PL. Soil as an important pathway of human lead exposure. Environ Health Perspect 1998; 106:217᎐229. Mielke HW, Gonzales CR, Smith MK, Mielke PW. The urban environment and childrens’ health: soil as an integrator of lead, zinc and cadmium in New Orleans, Louisiana, USA. Environ Res Section A 1999;81:117᎐129. Moir AM, Thornton I. Lead and cadmium in urban allotment and garden soils and vegetables in the United Kingdom. Environ Geochem Health 1989;11:113᎐119. Nelson WO, Campbell PGC. The effects of acidification on the geochemistry of Al, Cd, Pb and Hg in fresh-water
63
environments ᎏ a literature review. Environ Pollut 1991;71:91᎐130. Oliver MA. Soil and human health: a review. Eur J Soil Sci 1997;48:573᎐592. Prade K, Hagelgans V, Schweigger T. Heavy metal accumulation in the plinthic horizon of a ferric gleysol. Zeitschr Pflanzenernahr Bodenk 1991;154:227᎐232. Ramos L, Hernandez LM, Gonzalez MJ. Sequential fractionation of copper, lead, cadmium and zinc in soils from or near Donana National Park. J Environ Qual 1994;23:50᎐57. Ross S. Toxic metals in soil-plant systems. Chichester: Wiley, 1994. Rowell DL. Soil science methods and applications. Harlow: Longman, 1994. Sanka M, Strnad M, Vondra J, Paterson E. Sources of soil and plant contamination in an urban environment and possible assessment methods. Int J Environ Anal Chem 1995;59:327᎐343. Sauve S, McBride M, Hendershot W. Speciation of lead in contaminated soils. Environ Pollut 1997;98:149᎐155. Thornton I. Metal contamination of soils in urban areas. In: Bullock P, Gregory PJ, editors. Soils in the urban environment. Oxford: Blackwell, 1991:47᎐75. Vanstraalen NM, Bergema WF. Ecological risks of increased bioavailability of metals under soil acidification. Pedobiologica 1995;39:1᎐9.
Lead and zinc in the Wallsend Burn, an urban catchment in Tyneside, UK A. Mellor U Di¨ ision of Geography and En¨ ironmental Management, Lipman Building, Uni¨ ersity of Northumbria, Newcastle upon Tyne, Tyne and Wear, NE25 9DG, UK Received 14 December 1999; accepted 2 September 2000
Abstract This paper examines lead and zinc concentrations in topsoils and stream sediments of public access areas in an urban catchment in Tyneside, UK. It examines the extent and severity of metal contamination, explores spatial patterns in relation to urban and industrial development, and makes inferences about potential metal mobility. Total and acetic-acid extractable lead and zinc concentrations, organic content and pH were determined on 121 topsoil and 22 stream sediment samples using standard laboratory procedures. Using the lowest trigger thresholds for total lead and zinc, almost 75% and 91%, respectively, of topsoil samples were classified as contaminated; proportions were rather lower for acetic acid extractable metals. Similarly, approximately 45% and 95% of stream sediment samples were contaminated with lead and zinc, respectively. The spatial distribution of metal concentrations was characterized by a hotspot pattern, with highest values in central and southern parts of the catchment where there is a long urban and industrial history. The potential mobility of zinc is considerably greater than that of lead in both topsoils and stream sediments, and for both metals is slightly higher in the stream sediments than in the topsoils; both of these differences are statistically significant Ž P- 0.05.. The implications of the findings in this paper for assessment and monitoring of metal contaminated areas are explored. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Heavy metals; Lead; Zinc; Urban catchment; Soils; Stream sediments
1. Introduction The widespread occurrence of heavy metal contamination in urban areas and its potential imU
Tel.: q44-191-227-3758; fax: q44-191-227-4715. E-mail address: [email protected] ŽA. Mellor..
pact on human health are now well documented in the literature Že.g. Thornton, 1991; BEES Environmental Health, 1997; Oliver, 1997; Mielke and Reagan, 1998.. Such contamination has been reported in soils, airborne dusts, road sweepings, road-side vegetation, produce from urban allotments, urban stream water and fluvial and estuarine sediments ŽCulbard et al., 1988; Moir and
0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 8 1 1 - 1
50
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Thornton, 1989; Bubb and Lester, 1994; Kelly et al., 1996; Macklin, 1996.. Uptake of heavy metals by humans, largely through respiratory or digestive pathways, has contributed to elevated concentrations of such metals in blood, bone and tissue samples, and consequently to poor health, in a number of urban populations ŽBEES Environmental Health, 1997; Mielke et al., 1999.. Of particular concern is the strong association between elevated lead concentrations in blood samples from children, as a result of dust inhalation and pica-related ingestion, and impaired mental development Že.g. Ericson and Mishra, 1994; Jin et al., 1997.. Despite the potential health implications, there have been relatively few studies, conducted at an appropriately fine spatial scale, of the severity and spatial distribution of heavy metal pollution in urban areas. Douglas et al. Ž1993. examined the spatial distribution of lead in soils in part of south Manchester using a 70-m sampling grid, whilst Kelly et al. Ž1996. collected four topsoil samples per km2 in a study of heavy metals in Richmond upon Thames and Wolverhampton. In Tyneside, Aspinall et al. Ž1988. collected topsoils from relatively undisturbed grassy areas in public open spaces using a 1-km2 grid. Point data were then aggregated to 1981 census wards in an attempt to assess the number and spatial distribution of people exposed to elevated heavy metal concentrations. Mellor and Bevan Ž1999. examined the spatial distribution of lead in topsoils and stream sediments of public open spaces in the Ouseburn catchment in Newcastle upon Tyne. Here topsoil samples were collected using a 200-m grid, and stream sediments were collected at 500m intervals. In many of these studies, spatial patterns of heavy metal contamination were difficult to explain due to the complex industrial history and continued overprinting of pollution in urban areas. Indeed, it is only in the last 20 years or so, through the introduction of environmental monitoring and legislation, that such pollution has been documented. This study focused on lead and zinc in the topsoils and stream sediments of Wallsend Burn, a largely urban catchment in Tyneside, UK. Its aims were as follows:
䢇
䢇
䢇
to examine the extent and relative severity of lead and zinc contamination within the catchment; to explore the spatial distribution of the two metals and to relate this to former and present patterns of urban and industrial development within the catchment; to make inferences about the potential mobility of lead and zinc within the catchment.
The implications of the findings of this study for assessment and monitoring of soil and stream sediment pollution in urban areas were also considered.
2. Study area The Wallsend Burn is a 6-km-long lowland stream that drains an area of approximately 20 km2 to the east of Newcastle upon Tyne ŽFig. 1.. The central and northern parts of the catchment are mainly suburban with relatively low relief, whilst the southern part is rather steeper and more urban and industrial in character. The solid geology consists of carboniferous coal measures, including a mixture of sandstones, shales and coal bands. Superficial deposits consist largely of glacial till which covers most of the catchment. Dominant soil types are gleys and brown earths of the Brickfield and Hallsworth Series ŽJarvis et al., 1984.. The catchment is traversed by a major road ŽA1056., which runs east᎐west across its centre, linking the city of Newcastle upon Tyne with the coastal towns of Whitley Bay, North Shields and Tynemouth ŽFig. 1.. To the north of this, the land is relatively open with a number of public green spaces and recreational areas, and a small amount of agricultural land. These spaces are punctuated with relatively new housing estates, such as Hadrian Park and Battle Hill, mostly constructed since 1970. Although there are some open spaces in the south of the catchment, such as Bigges Main golf course, Wallsend Dene and Holy Cross, much of this area is characterized by a dense network of older Žmostly pre-1930s. housing, roads and industrial sites.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
51
Fig. 1. Location map.
These differences in density of development are reflected in the character of the stream network. In the northern half of the catchment, the streams are relatively small and discontinuous ŽFig. 1.. This is due to extensive channelization and the construction of underground drainage around and beneath the newer housing estates and roads. In the southern part of the catchment, stream channels are larger and more continuous, with some channelization, particularly in and just above the lower tidal reaches. Here, stream flow is greater, largely due to surface drainage from impermeable urban surfaces and the three combined sewer overflow ŽCSO. discharge points. Industry, which comprises approximately 5% of the catchment area, discharges its effluent directly to sewer; hence industrial contamination should only occur at times of CSO. The main activities include chemical industries, oil rig construction and fitting, waste processing and incineration, scrap metal processing, and former gas and coke works.
Most of these activities are located in the south of the catchment. Historically coal mining has been a significant industry within the catchment, notably at Bigges Main between Walkerville and Wallsend, the Rising Sun to the north of Battle Hill, and at Howdon ŽFig. 1.. Deep mining has now ceased in the catchment, and these sites have since been restored as recreational and amenity areas such as Bigges Main golf course and the Rising Sun Country Park. Other public open spaces include large areas of sports fields at Benton and Holy Cross, Wallsend Dene and Richardson-Dees Park in Wallsend.
3. Methodology Using a combination of Ordnance Survey maps Ž1:25 000 scale., colour air photographs Ž1:10 000 scale. and site visits as appropriate, the catch-
52
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 2. Locations of Ža. topsoil sampling sites and selected industrial activities, and Žb. stream sediment sampling sites and combined sewer overflows.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
ment area was defined. All public access land areas and stream networks were then mapped, together with a number of potential sources of heavy metal contamination, notably sites of former industrial activity such as areas of colliery spoil deposition, former and present waste disposal sites, major road networks and CSOs. Mapped information was then entered onto a geographical information system ŽGIS. using Arc-View software to facilitate spatial representation. Public access areas were systematically sampled by a 200-m grid over the appropriate parts of the catchment, amounting to a total of 121 topsoil Ž0᎐10 cm. samples ŽFig. 2.. Twenty-two stream sediment samples were collected at 200-m intervals along the main stream and its tributaries working upstream from the high tide limit ŽFig. 2.. All samples consisted of an aggregate of five sub-samples collected using a trowel from the corners and centre of a 1-m2 quadrat. The soil and sediment samples were air-dried, ground with a mortar and pestle and passed through a 2-mm sieve. Lead and zinc contents Žboth total and acetic acid extractable ., organic matter content and pH were then determined. Total lead and zinc were extracted by treating 0.5 g of sample with a mixture of 10 ml concentrated nitric and 5 ml hydrochloric acids in a microwave digestor; the extract was then diluted to 100 ml with distilled water. The more mobile and bioavailable lead and zinc were extracted by treating 5 g of sample with 100 ml of 0.5 M acetic acid ŽAspinall et al., 1988.. Although the latter extraction is not widely used, it was selected in this investigation to facilitate comparison with other similar studies in the Tyneside region ŽAspinall et al., 1988; Macklin, 1996; Mellor and Bevan, 1999.. The extracts were then filtered, and lead and zinc concentrations were determined using a PerkinElmer 2380 double beam atomic absorption spectrophotometer. Results were validated using the same analytical procedures on 20 sub-samples from a standard reference soil ŽORM Laboratory of the Government Chemist, reference number GBW07406.; at least 95% of the sub-samples exhibited lead and zinc concentrations within "2 S.E. of the means quoted by the supplier. Organic
53
matter content was determined by loss on ignition and pH was established from 10 g of sample suspended in 25 ml of distilled water using a Jenway 3020 pH probe ŽRowell, 1994..
4. Results Descriptive statistics for all properties examined are shown in Table 1. With the exception of organic matter content and pH, all properties had mean values in excess of median values and high standard deviations reflecting positively skewed distributions and a high degree of variation. Mean total lead and zinc concentrations in the topsoils were 130 mgrkg and 282 mgrkg, respectively. These concentrations are comparable with those from other investigations conducted on soils of urban areas in the UK. For example, Culbard et al. Ž1988. obtained geometric mean lead concentrations ranging from 80 to 1300 mgrkg from garden soils in a number of urban locations, including Newcastle upon Tyne Ž350 mgrkg.. Mean total lead concentration in the stream sediments Ž103 mgrkg. was slightly Table 1 Descriptive statistics for topsoils Ž n s 121. and stream sediments Ž n s 22. Property
Mean
Median
Total Pb: topsoils Žmgrkg. Total Pb: stream sediments Žmgrkg. Total Zn: topsoils Žmgrkg. Total Zn: stream sediments Žmgrkg. Acetic acid extractable Pb: topsoils Žmgrkg. Acetic acid extractable Pb: stream sediments Žmgrkg. Acetic acid extractable Zn: topsoils Žmgrkg. Acetic acid extractable Zn: stream sediments Žmgrkg. Organic content: topsoils Ž%. Organic content: stream sediments Ž%. Acidity ŽpH.: topsoils Acidity ŽpH.: stream sediments
129 103
115 82
69 67
282 473
253 267
138 571
6
5
7
10
7
11
62
35
66
134
91
119
18 17
17 16
4 7
6.1 7.4
6.0 7.5
S.D.
0.9 0.7
54
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
lower than in the topsoils, whereas that for zinc Ž473 mgrkg. was considerably higher ŽTable 1.. Total lead concentrations are comparable with those of Macklin Ž1996. and Mellor and Bevan Ž1999. who analysed stream sediment samples from the river Ouseburn a few kilometres to the west of the study area Žmean concentration s 68 mgrkg.. Mean acetic acid extractable lead and zinc concentrations in the topsoils were considerably lower than those for the total metals at 6 mgrkg and 62 mgrkg ŽTable 1., and on average constituted 4.7% and 23.4% of the total metals, respectively. These concentrations compare well with those of Mellor and Bevan Ž1999. who obtained a mean lead concentration of 16 mgrkg. Mean acetic acid extractable lead and zinc concentrations in the stream sediments were 11 mgrkg and 134 mgrkg, respectively, both of which were considerably higher than those in the topsoils. On average, this fraction of lead and zinc constituted 9.4% and 29.8% of the total metals, respectively, these proportions being slightly higher than those in the topsoils. Once again, concentrations of acetic acid extractable lead compare well with those in stream sediments of the adjacent river Ouseburn Žmean s 15 mgrkg. ŽMellor and Bevan, 1999.. Mean organic matter contents in the topsoils and stream sediments were 18% and 17%, and their pH values 6.1 and 7.4, respectively. These values are comparable with those in grassland and woodland soils elsewhere in northern England ŽJarvis et al., 1984.. Spatial distributions of total and acetic acid extractable lead and zinc in the topsoils and stream sediments are shown in Figs. 3᎐6. Highest concentrations were found in southern and central parts of the catchment, although the spatial pattern for zinc was somewhat more scattered than that for lead, particularly in the stream sediments. Although organic matter contents and pH values did not display any clear spatial patterns, organic contents were highest in wooded and less disturbed green spaces, and in the smaller tributary streams. Similarly, pH values were lowest in less disturbed wooded areas and highest in areas where agrochemicals had been used Že.g.
adjacent to golf courses.; pH values in the stream sediments were consistently higher than in the topsoils.
5. Discussion 5.1. Extent and se¨ erity of contamination A number of widely recognized systems have been used to define contamination thresholds in soils, although none exists for the designation of contamination in stream sediments. In the system adopted by the UK ŽICRCL, 1987., trigger concentrations are only provided for the total metal and not for more mobile, bioavailable forms. Metals are also separated into group A contaminants which may be hazardous to health Že.g. lead. and group B contaminants which are phytotoxic but not normally hazardous to health Že.g. zinc.. In addition, trigger concentrations for group A contaminants differ according to land use, so that in domestic gardens and allotments, the figure for lead is 500 mgrkg whereas in parks, playing fields and open spaces it is 2000 mgrkg. Trigger concentrations for group B contaminants are not differentiated in this way and are only applicable to soil in which produce for human and animal consumption is grown; the trigger concentration for zinc is 300 mgrkg. Application of the trigger concentration for lead to the topsoils and stream sediments from Wallsend Burn reveals that none of the samples was significantly contaminated. It should be noted, however, that 31% of topsoils and 45% of stream sediments have zinc concentrations in excess of the trigger threshold. In response to recent changes in the Environmental Protection Act Ž1995. the ICRCL Ž1987. system is soon to be replaced by a risk assessment model based on identification of contaminant sources, pathways and receptors, and new Contaminated Land Exposure Assessment ŽCLEA. guidelines. For this approach to be successful, contaminant guidelines must accommodate the issue of bioavailability which will vary between different metallic forms. The guidelines must also be readily interpreted by land remedia-
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 3. Total metal concentrations in topsoils: Ža. lead, Žb. zinc.
55
56
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 4. Total metal concentrations in stream sediments: Ža. lead, Žb. zinc.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 5. Acetic acid extractable metal concentrations in topsoils: Ža. lead, Žb. zinc.
57
58
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
Fig. 6. Acetic acid extractable metal concentrations in stream sediments: Ža. lead, Žb. zinc.
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
tion practitioners, site owners and by Local Authorities whose task it is to establish contaminated land registers. In the Netherlands, a more stringent system based on much lower threshold concentrations is in use ŽDepartment of Soil Protection, 1996.. The Dutch system also specifies action values which, if exceeded, signal the need for clean up operations to be carried out. The lower threshold or ‘optimum’ concentrations for lead and zinc are 85 mgrkg and 140 mgrkg, respectively. Similarly, the action values for these metals are 530 mgrkg and 720 mgrkg. Seventy-five percent of topsoil and 45% of stream sediment samples from Wallsend Burn exceed the lower threshold with respect to total lead, whilst none of the samples exceed the higher action concentration. Similarly, 91% of topsoil and 95% of stream sediment samples from Wallsend Burn exceed the lower threshold with respect to zinc, whilst 2% of topsoil and 14% of stream sediment samples exceed the action concentration. Using a more statistical approach ŽDavies, 1983., Aspinall et al. Ž1988. determined the highest probable concentrations of total and available lead and zinc in uncontaminated soils in Tyneside. For total lead and zinc, these were 80 mgrkg and 345 mgrkg, respectively, and for available lead and zinc they were 14 mgrkg and 11 mgrkg. This study concluded that more than two thirds of the Tyneside soils were contaminated with respect to lead, although the proportion was considerably lower for zinc. The proportion of the population living in contaminated areas, calculated on the basis of census wards, was also found to be high at 84% and 12% for total lead and zinc, respectively. With reference to the samples collected from Wallsend Burn, it can be seen that 76% of topsoils and 55% of stream sediments exceed the threshold concentration for total lead, whilst only 7% of topsoils and 14% of stream sediments are in excess of that for available lead. In relation to total zinc, 26% of topsoils and 41% of stream sediments exceed the threshold concentration, whilst all samples exceed that for available zinc.
59
5.2. Spatial distribution The difference in spatial pattern between lead and zinc could be explained by the fact that zinc is generally more mobile in soils and sediments than lead ŽMerrington and Alloway, 1994; Lee and Touray, 1998.. It may also be explained by differences in the sources of the two metals, much of the lead having a more localized urban and industrial source, and much of the zinc having a more widespread geological source, notably the carboniferous shales ŽPrade et al., 1991.. Superimposed upon the catchment-wide distribution of lead and zinc are more localized ‘hotspot’ patterns where sites with high metal concentrations are surrounded by sites with much lower concentrations. This spatial pattern is a common feature of many types of geochemical pollution and has been reported for lead contamination in the adjacent Ouseburn catchment ŽMellor and Bevan, 1999.. Elevated lead and zinc concentrations in the southern part of the catchment are likely to reflect the long history of urban and industrial development. Potential sources of these metals, and of lead in particular, include vehicle emissions from the dense network of roads, a pattern that has been widely reported in other urban areas Že.g. Douglas et al., 1993; Sanka et al., 1995.. This influence can be seen most clearly at sites adjacent to the main coast road where relatively high lead concentrations can be found ŽFigs. 1 and 2a and Fig. 4a.. Elevated metal concentrations are also associated with open spaces adjacent to older housing and demolition sites which are prevalent in the south of the catchment. Similar associations have been reported in Manchester ŽDouglas et al., 1993. and in Richmond upon Thames and Wolverhampton ŽKelly et al., 1996. where leaded paint used in older properties was identified as an important source of lead in soils. Another possible source of elevated metal concentrations in the south of the catchment is the metal smelting and finishing, and glass-making factories in the lower Ouseburn ŽMellor and Bevan, 1999.. This area
60
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
lies to the south-west of the catchment in a generally up-wind direction, and although these industries have now ceased, it is likely that airborne pollution from them would have been widespread in the past. Such industrial sources of heavy metals in soils have been reported elsewhere Že.g. Bloeman et al., 1995.. Another possible source of heavy metals in soils of the Wallsend Burn catchment is the carboniferous pyritic shale ŽPrade et al., 1991. which constitutes a significant proportion of the coalbearing geology of this area. Elevated lead and zinc concentrations are found at the Bigges Main Golf course in the south-west of the catchment and at the Rising Sun Country Park in the central area, both of which represent extensive areas of colliery spoil restoration. The association between elevated lead concentrations in topsoils and restored colliery spoil sites was also reported in the Ouseburn catchment, only a few kilometres to the west of the Wallsend Burn catchment ŽMellor and Bevan, 1999.. More local examples of pollution sources include the former waste incinerator, a gas works site and former metal scrap yard at Howdon in the south-east of the catchment. Unlike soils, stream sediments undergo phases of transfer and post-depositional mobilization, interspersed with periods of temporary storage and stability ŽFoster and Charlesworth, 1996.. Such temporal and spatial complexity creates difficulties when attempting to interpret spatial patterns of heavy metals ŽMacklin, 1996.. Highest concentrations of lead and zinc are found in the more urban, and formerly more industrialized, lower reaches of the stream. Here, surface drainage during prolonged or intense rainfall events may carry significant quantities of metals from impermeable surfaces, including roads and car parks. Combined Sewer Overflow, which is common during such events, may also represent another important source of lead and zinc in this lower part of the catchment, as has been reported in the adjacent Ouseburn catchment ŽMellor and Bevan, 1999. and elsewhere ŽDeely and Ferguson, 1994.. 5.3. Potential metal mobility A crude estimate of potential lead and zinc
mobility within topsoils and stream sediments of the Wallsend Burn catchment can be gained through examination of the concentrations of acetic acid extractable metals expressed as a percentage of the total metals. In the topsoils the mean values for lead and zinc are 4.7% and 23.4%, respectively, whilst in the stream sediments they are 9.4% and 29.8%. In both cases, these differences were found to be statistically significant following t-test analysis Žfor topsoils t s 9.11, P- 0.001; for stream sediments t s 8.33, P- 0.001.. It therefore appears that the potential mobility of zinc is considerably greater than that for lead in both topsoils and stream sediments of the Wallsend Burn catchment. The difference in potential mobility between these metals may be accounted for by the fact that zinc is an essential micronutrient in many plants and, as a consequence, is likely to be more soluble and mobile in soil᎐plant ecosystems than lead which plays no known beneficial role in living organisms ŽRoss, 1994.. In contrast, lead from industrial and urban pollution sources is dominated by relatively inert particulate forms which are stable and persistent in soils and sediments ŽMellor and Bevan, 1999.. In addition to differences in potential mobility between lead and zinc, there are differences in the potential mobility of both metals between the topsoils and stream sediments, with slightly higher percentages in the stream sediments. Once again, these differences were found to be statistically significant following t-test analysis Žfor lead t s 13.23, P- 0.001; for zinc t s 5.43, P- 0.001.. The differences in potential mobility may result from differences in both the sources of these metals and the geochemical environments between topsoils and stream sediments ŽMellor and Bevan, 1999.. The influences of organic content and pH on the behaviour and potential mobility of heavy metals, particularly in soils, is well documented Že.g. Nelson and Campbell, 1991; Bryan and Langston, 1992; Hargitai, 1995; Vanstraalen and Bergema, 1995; Sauve et al., 1997; Hooda and Alloway, 1998.. In the topsoils and stream sediments of the Wallsend Burn catchment, these influences were explored using Pearson correlation analysis. However, the only statistically sig-
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
nificant correlations found in the topsoils were those between organic content and total lead Žq0.408, P- 0.001., and organic content and total zinc Žq0.309, P- 0.01.. None of the correlations between stream sediment properties was found to be statistically significant. The absence of significant correlations with organic content in the stream sediments suggests that the type of organic matter could be just as important as the overall content. Hargitai Ž1995., e.g. indicates that although Pb is strongly organophilic, it is associated most strongly with the humic fraction, the presence of which is likely to be limited in stream sediments. Although metal mobility is known to increase with decreasing pH ŽNelson and Campbell, 1991; Vanstraalen and Bergema, 1995., the absence of any correlations with this property suggests that the pH range was too narrow to demonstrate such relationships in the topsoils and stream sediments of Wallsend Burn ŽMellor and Bevan, 1999..
6. Conclusions The extent and severity of lead and zinc contamination in topsoils and stream sediments depends on which of the many systems of trigger thresholds is adopted. Using the most stringent of the systems considered here, more than two thirds of topsoil and stream sediment samples from Wallsend Burn are found to be contaminated. In general, however, zinc contamination appears to be more significant than lead contamination in both topsoils and stream sediments, and there appears to be no clear difference in the severity of contamination between topsoils and stream sediments. Highest concentrations of lead and zinc are found in southern and central parts of the catchment, where urban and industrial development have been greatest, although the spatial pattern for zinc is somewhat more scattered than that for lead. As with all spatial sampling, the density of the sampling network can considerably affect the conclusions drawn ŽMellor and Bevan, 1999.. The spatial complexity of lead and zinc distributions, often in the form of contaminated hot spots,
61
creates particular difficulties when designing sampling strategies to assess and monitor soil and sediment contamination; such strategies must achieve a balance between representativeness and cost. Sampling design may be optimized using a geostatistical approach where spatial variability and interpolation are used to predict and model contaminant concentrations in un-sampled areas Že.g. Kuzel et al., 1994.. The potential mobility of zinc is significantly greater than that of lead in both topsoils and stream sediments of Wallsend Burn. This observation may reflect the fact that zinc is more mobile in soil-plant systems, whilst lead is dominated by relatively inert, particulate forms that are more stable and persistent in soils and sediments. Both metals are found to be significantly more mobile in the stream sediments than in the topsoils, possibly reflecting differences in geochemical environment and metal sources. The complex behaviour of stream sediments over time, in comparison with the relative stability of soils, makes interpretation of spatial patterns difficult. In order to develop a fuller understanding of the behaviour and fate of heavy metals in the environment, it is recommended that investigations be extended beyond that of the total fractions to consider the various different forms of these metals ŽRamos et al., 1994; Floresrodriguez et al., 1994; Macklin, 1996.. It is suggested that there is a need for detailed baseline investigations of the spatial distribution of heavy metals, and other pollutants, in urban areas, perhaps with a view to formulating a database that would be of use to Local Authorities, organizations involved in monitoring and assessment of contamination, and other parties concerned with the planning and development process ŽBullock, 1987; Macklin, 1996.. For example, a more thorough inventory of contaminants at urban ‘brownfield’ sites may facilitate their development in preference to the more precious and often more pressurized rural ‘greenfield’ sites ŽMellor and Bevan, 1999.. It is further suggested that such investigations should take a holistic approach by integrating data collection across the soil and stream systems of catchment areas, rather than focusing on soils alone. Not only would this
62
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63
allow development of a more thorough understanding of the behaviour and fate of pollutants in the environment, but it would also parallel the more integrated approach to monitoring and assessment of the environment adopted by the UK Environment Agency, and emphasized in their Local Environment Agency Plans ŽLEAPs.. The catchment, now widely recognized as a fundamental unit of environmental management, is central to these plans. Finally, such an approach could integrate well with the new risk assessment approach to contaminant guidelines ŽCLEA. based on identification and quantification of contaminant source᎐pathway᎐receptor linkages.
Acknowledgements This work was funded by the University of Northumbria at Newcastle from monies allocated by the Higher Education Funding Council as part of the 1992 Research Assessment Exercise. The author would like to thank Miss Joanne Cooke for analysis of soil and stream sediment samples, with support from Miss Joyce Moxon and Mrs Ruth Steinberg, Mr Keith Turner and Miss Caroline Dilks for Research into the catchment history, processing of data and associated information on GIS and production of diagrams, and Mr Gary Haley for production of Fig. 1 Žlocation map.. References Aspinall R, Macklin M, Openshaw S. Heavy metal contamination in soils of Tyneside: a geographically-based assessment of environmental quality in an urban area. In: Hooke JM, editor. Geomorphology in environmental planning. Chichester: Wiley, 1988:87᎐102. BEES Environmental Health. Lead, 1997Žwww.beesinc. orgraboutrheallead.html.. Bloeman ML, Markert B, Lieth H. The distribution of Cd, Cu, Pb and Zn in topsoils of Osnabruck in relation to land-use. Sci Total Environ 1995;166:137᎐148. Bryan GW, Langston WJ. Bioavailability, accumulation and effects of heavy-metals in sediments with special reference to United Kingdom Estuaries ᎏ a review. Environ Pollut 1992;76:89᎐131. Bubb JM, Lester JN. Anthropogenic heavy-metal inputs to
lowland river systems, a case study ᎏ the River Stour, UK, Water. Air Soil Pollut 1994;78:279᎐296. Bullock P. Soil protection-a need for a European programme. In: Barth H, Hermite PL, editors. Scientific Basis for Soil Protection in the European Community. Amsterdam: Elsevier, 1987:581᎐587. Culbard EB, Thornton I, Watt J, Wheatly M, Moorcroft S, Thompson M. Metal contamination in British suburban dusts and soils. J Environ Qual 1988;12:226᎐234. Davies BE. A graphical estimation of the normal lead content of some British soils. Geoderma 1983;29:67᎐75. Deely JM, Ferguson JE. Heavy-metal and organic matter concentrations and distributions in dated sediments of a small estuary adjacent to a small urban area. Sci Total Environ 1994;153:97᎐111. Department of Soil Protection. Intervention values and target values ᎏ soil quality standards. The Hague, Netherlands: Ministry of Housing, Spatial Planning and Environment, 1996. Douglas I, Al-Ali J, Clarke M. Lead contamination in Manchester. Land Contamination Reclamation 1993;1:143᎐148. Ericson JE, Mishra SI. Soil lead concentrations and prevalence of hyperactive behavior among school children in Ottawa, Canada. Environ Ecol 1994;16:247᎐256. Floresrodriguez J, Bussy AL, Thevenot DR. Toxic metals in urban runoff ᎏ physicochemical mobility assessment using speciation schemes. Water Science Technol 1994;29:83᎐93. Foster IDL, Charlesworth SM. Heavy metals in the hydrological cycle ᎏ trends and explanation. Hydrol Process 1996;10:227᎐261. Hargitai L. Some aspects of the mobility and distribution of toxic heavy-metal contaminants in soil profiles and river sediments. Int J Environ Anal Chem 1995;59:317᎐325. Hooda PS, Alloway BJ. Cadmium and lead sorption behaviour of selected English and Indian soils. Geoderma 1998;84:121᎐134. Interdepartmental Committee for the Redevelopment of Contaminated Land. Guidance on the assessment and redevelopment of contaminated land. ICRCL Paper 59r83. 2nd ed London: Department of the Environment, 1987. Jarvis RA, Bendelow VC, Bradley RI et al. Soils and their use in northern England. Soil Survey of England and Wales Bulletin, 10. Harpenden, 1984. Jin A, Teschke K, Copes R. The relationship of lead in soil to lead in blood and implications for standard setting. Sci Total Environ 1997;208:23᎐40. Kelly J, Thornton I, Simpson PR. Urban geochemistry ᎏ a study of the influence of anthropogenic activity on the heavy metal content of soils in traditionally industrial and nonindustrial areas of Britain. Appl Geochem 1996; 11:363᎐370. Kuzel S, Nydl V, Kolar L, Tichy R. Spatial variability of cadmium, pH, organic matter in soil and its dependence on sampling scales, Water. Air Soil Pollut 1994;78:51᎐59. Lee PK, Touray JC. Characteristic of polluted artificial soil
A. Mellor r The Science of the Total En¨ ironment 269 (2001) 49᎐63 located along a motorway and effects of acidification on the leaching behaviour of heavy metals ŽPb, Zn, Cd.. Water Res 1998;32:3425᎐3435. Macklin M. Fluxes and storage of sediment-associated heavy metals in floodplain systems: assessment and river basin management issues at a time of rapid environmental change. In: Anderson MG, Walling DE, Bates PD, editors. Floodplain processes. Chichester: Wiley, 1996:441᎐460. Mellor A, Bevan JR. Lead in the soils and stream sediments of an urban catchment in Tyneside, UK. Water Air Soil Pollut 1999;112:327᎐348. Merrington G, Alloway BJ. The flux of Cd, Cu, Pb and Zn in mining polluted soils. Water Air Soil Pollut 1994;72: 333᎐344. Mielke HW, Reagan PL. Soil as an important pathway of human lead exposure. Environ Health Perspect 1998; 106:217᎐229. Mielke HW, Gonzales CR, Smith MK, Mielke PW. The urban environment and childrens’ health: soil as an integrator of lead, zinc and cadmium in New Orleans, Louisiana, USA. Environ Res Section A 1999;81:117᎐129. Moir AM, Thornton I. Lead and cadmium in urban allotment and garden soils and vegetables in the United Kingdom. Environ Geochem Health 1989;11:113᎐119. Nelson WO, Campbell PGC. The effects of acidification on the geochemistry of Al, Cd, Pb and Hg in fresh-water
63
environments ᎏ a literature review. Environ Pollut 1991;71:91᎐130. Oliver MA. Soil and human health: a review. Eur J Soil Sci 1997;48:573᎐592. Prade K, Hagelgans V, Schweigger T. Heavy metal accumulation in the plinthic horizon of a ferric gleysol. Zeitschr Pflanzenernahr Bodenk 1991;154:227᎐232. Ramos L, Hernandez LM, Gonzalez MJ. Sequential fractionation of copper, lead, cadmium and zinc in soils from or near Donana National Park. J Environ Qual 1994;23:50᎐57. Ross S. Toxic metals in soil-plant systems. Chichester: Wiley, 1994. Rowell DL. Soil science methods and applications. Harlow: Longman, 1994. Sanka M, Strnad M, Vondra J, Paterson E. Sources of soil and plant contamination in an urban environment and possible assessment methods. Int J Environ Anal Chem 1995;59:327᎐343. Sauve S, McBride M, Hendershot W. Speciation of lead in contaminated soils. Environ Pollut 1997;98:149᎐155. Thornton I. Metal contamination of soils in urban areas. In: Bullock P, Gregory PJ, editors. Soils in the urban environment. Oxford: Blackwell, 1991:47᎐75. Vanstraalen NM, Bergema WF. Ecological risks of increased bioavailability of metals under soil acidification. Pedobiologica 1995;39:1᎐9.