CHANGES IN ACID PRECIPITATION-RELATED ... - Springer Link

CHANGES IN ACID PRECIPITATION-RELATED WATER CHEMISTRY OF LAKES FROM SOUTHWESTERN NEW BRUNSWICK, CANADA, 1986–2001 W. PILGRIM1∗ , T.A. CLAIR2 , J. CHOATE1 and R. HUGHES1 1 New Brunswick Department of Environment and Local Government, Sciences and Reporting Branch, P.O. Box 6000, Fredericton, New Brunswick, E3B 5H1 Canada; 2 Environmental Conservation Br., Environment Canada – Atlantic Region, P.O. Box 6227,

Sackville, N.B. E4L 1G6 (∗ author for correspondence, e-mail: [email protected])

Abstract. Between 1986 and 2001, thirty-nine lakes in southwestern New Brunswick in Atlantic Canada were surveyed for acid precipitation-related water quality changes. Most of the study lakes are located on granite bedrock and represent the most acid sensitive lakes in the province. Between 1987 and 1992, hydrogen ion deposition to the lake study area averaged 452 eq ha−1 yr−1 , compared to 338 eq ha−1 yr−1 between 1993 and 2000, a 25% reduction. The lake chemistry data were evaluated by dividing the lakes into four clusters for each survey year based on their acid neutralizing capacity. Twenty percent of the lakes (cluster IV) had an average ANC of 40 µeq L−1 or greater and maintained an average pH of greater than 6 over the duration of the study period. A pH of 6 or greater is considered a healthy benchmark for maintaining biodiversity. The remaining 31 lakes (clusters I to III) had an average ANC of less than 40 µeq L−1 and maintained an average pH of less than 6. Other lake chemistry changes included a general decline in lake sulphate and colour over the duration of the survey period, followed by more recent improvements in calcium ion, pH and ANC, and notably higher but declining aluminum levels in lower ANC and pH lakes. Nitrate accounted for 37% of the acid deposition to the study area, however it was not detectable in the lakes. Although acid deposition has declined and these lakes are beginning to show signs of acid recovery, 80% of the study lakes remain acid sensitive having little buffering capacity with low calcium, pH and ANC. Keywords: acid recovery, acid neutralization capacity, acidification, Bay of Fundy, lake chemistry, New Brunswick, nitrate, sulphate

1. Introduction The southwestern part of New Brunswick is located in a transboundary influenced air shed and acid rain, ozone and mercury are monitored in this area (Cox et al., 1989; Davies, 1993; NESCAUM et al., 1998). The geology of southwestern New Brunswick is mixed containing Ordovician and Devonian granites, Silurian greywacke, slates, sandstones and limestone, as well as Mississippian silicic flows (Potter et al., 1979; Shilts, 1981). The present study, examines acid precipitation related chemistry trends for 39 lakes in southwestern New Brunswick that were periodically sampled over a 15-year period and compares water quality changes to acid deposition to the study area. Environmental Monitoring and Assessment 88: 39–52, 2003. © 2003 Kluwer Academic Publishers. Printed in the Netherlands.

40

W. PILGRIM ET AL.

Over the past two decades, sulphur dioxide emissions in the Canada and United States have decreased by 28%, in the U.S. from 23.5 Mt in 1980 to 17.8 Mt in 1998 (GAO, 2000) and in Canada from 4.6 Mt in 1980 to 2.5 Mt in 1999 (CCME, 1999; Ro et al., 1998). In response to reductions in emissions, acid deposition has decreased and some aquatic ecosystems in eastern North America are showing signs of acid recovery while others are not (Stoddard et al., 1999; Dupont et al., 2002; Clair et al., in press).

2. Study Area and Methods Thirty-nine, semi-remote, coastal, headwater lakes were surveyed by helicopter in July 1986, 1993, 1995, 1998 and September 2001. The study area is located in the southern part of the province of New Brunswick on the Atlantic coast of Canada (Figure 1). The study area is approximately 2000 km2 in size representing 3% of the area of the province. The lakes are located less than 50 km from the Bay of Fundy coast and range in size from 2 to 80 ha. Surface water samples were taken from the middle of the lake. Samples were transported to the New Brunswick Department of Environment and Local Government laboratory (accredited by the Canadian Association of Environmental Analytical Laboratories) in Fredericton where they were preserved the same day as collected. Samples were analyzed for: pH, sulphate − (SO2− 4 ), nitrate (NO3 ), alkalinity (G-ALK), total organic carbon (TOC), dissolved organic carbon (DOC), chloride (Cl− ), sodium (Na+ ), magnesium (Mg2+ ), calcium (Ca2+ ), colour (Apparent Colour Units), iron (Fe) and aluminum (Al) ion. Major ions were corrected for sea-salt contribution using Cl− as the indicator ion (Watt et al., 1979). Annual acid deposition was extracted from the provincial acid deposition data base using three stations that were located in the lakes study area. Musquash, South Oromocto Lake (SOL) and Pennfield, were part of the provincial-wide acid precipitation network (Tims, 1986). The Musquash station closed in 1989 and data from the two remaining two stations were subsequently used. Acid deposition monitoring in the lake study area started in 1986 with the first full set of annual data being available in 1987. A two-way analysis of variance (ANOVA) and t-tests were first used to determine the significance of general water chemistry changes for all lakes combined. The lake data showed a high degree of variance in water chemistries, and changes in major ion chemistry were more uniform when the lakes were grouped into clusters according to their acid buffering ability.

CHANGES IN WATER CHEMISTRY OF LAKES

41

Figure 1. Location and geology of study area in southwestern New Brunswick in Atlantic Canada.

42

W. PILGRIM ET AL.

3. Results 3.1. ATMOSPHERIC DEPOSITION Between 1987 and 2000, the annual average precipitation measured at the three acid deposition-monitoring sites ranged from 1095 mm to 1621 mm, with an overall average of 1311 mm (Figure 2a). Sea salt corrected (SSC) wet SO2− 4 deposition peaked in 1989 and generally declined for the remainder of the period. Nitrate also peaked in 1989, declined to its lowest level in 1991 and remained relatively constant after that (Figure 2b). Hydrogen ion peaked in 1989, and then decreased substantially until 1992. Since that time, H+ deposition has remained relatively constant, with smaller inter-annual fluctuations. For the first part of the survey period (1986 to 1992), average H+ deposition was 452 eq ha−1 yr−1 , compared to 338 eq ha−1 yr−1 between 1993 and 2000, a 25% reduction (Figure 2c). Calcium ion deposition to the study area showed high variability over periods of 1–3 years, with a slight downward trend (Figure 2d). 3.2. L AKE CHEMISTRY STATISTICS The ANOVA was applied to all of the data collected for the five survey periods and 2+ + showed significant differences for SSC-SO2− 4 , Ca , H , ANC and TOC over the 15-year period (Table I). The ANOVA also showed significant differences for most parameters between lakes sampled. A paired t-test was further used to examine the changes between survey years. Generally, SSC-SO2− 4 of the 39 lakes declined 2+ and ANC, pH and Ca decreased in the early part (1986 to 1993) of the study then increased between 1993 to 2001. Between 1986 and 1993, the paired t-test 2+ and TOC (Table II). showed significant decreases in SSC-SO2− 4 , ANC, SSC-Ca 2− The paired t-tests further indicated that SSC-SO4 was significantly lower (p = 0.03) in 2001 than in 1986. The increase in pH between 1995 and 2001 was significant (p = 0.01). SSC-Ca2+ significantly (p = 0.001) decreased between 1986 and 1998 then increased. 3.3. L AKE CLUSTERS The best interpretation of the data was made by dividing the lakes into clusters according to the level of their ANC. Within the clusters, more consistent changes were observed for pH, ANC, sulphate, calcium ion, colour and aluminum (Figures 3 and 4). Cluster I lakes were those, which had an average of less than zero ANC and also had the lowest pH, base cations and colour, as well as the highest sulphate and aluminum levels. The largest number of lakes, 41% were in cluster II which had an average ANC between 0–20 µeq/L. From the clusters we were able to determine which lakes were more susceptible to chemical changes caused by acidification, which is useful information for modeling and assessment purposes.

Figure 2. Annual precipitation, sulphate, nitrate, hydrogen and calcium ion deposition measured in the lakes study area of southwestern New Brunswick, 1987–2000.


43

44

W. PILGRIM ET AL.

TABLE I Two way analysis of variance for changes in SSC-sulphate, calcium, pH, ANC and TOC, 1986, 1993, 1995, 1998, 2001. ANOVA SO4 Source of Variation Lakes Years Error Total

SS 30753.82 1788.023 11973.82 44515.66

df 38 4 152 194

MS 809.311 447.0057 78.77513

F 10.27369 5.674452

P-value 2.74E-26 0.000277

F crit 1.484082 2.431165

SS 125147.9 6506.603 17114.23 148768.7

df 38 4 152 194

MS 3293.365 1626.651 112.5936

F 29.25002 14.4471

P-value 1.17E-52 5.08E-10

F crit 1.484082 2.431165

SS 2.31E-09 3.4IE-11 3.03E-10 2.65E-09

df 38 4 152 194

MS 6.08E-11 8.53E-12 2.00E-12

F 30.45011 4.271678

P-value 8.58E-54 0.002644

F crit 1.484082 2.431165

SS 169108.2 3481.92 18130.45 190720.6

df 38 4 152 194

MS 4450.217 870.4799 119.2793

F 37.30923 7.297831

P-value 1.28E-59 2.11E-05

F crit 1.484082 2.431165

SS 1039.318 58.3333 414.2496 1511.901

df 38 4 152 194

MS 27.35047 14.58333 2.725326

F 10.03567 5.351038

P-value 8.74E-26 0.000465

F crit 1.484082 2.431165

ANOVA Ca Source of Variation Lakes Years Error Total ANOVA H+ Source of Variation Lakes Years Error Total ANOVA ANC Source of Variation Lakes Year Error Total ANOVA TOC Source of Variation Lakes Years Error Total

45


TABLE II Significance of changes in mean concentrations of sulphate, calcium ion, hydrogen ion, acid neutralizing capacity of lakes surveyed in 1986, 1993 and 2001. Variable

1986

1993

p

1993

2001

P

1986

2001

p

60.71 185.6 38

53.16 168.8 38

0.0001

53.16 168.8 38

53.99 475.6 38

0.78

60.71 185.6 38

53.99 475.6 38

0.03

SSC-Ca+2 Mean 61.93 Variance 644.8 df 38

52.98 529.6 38

0.001

52.98 529.6 38

57.11 967.4 38

0.09

61.93 644.8 38

57.11 967.4 38

0.14

SSC-SO2− 4 Mean Variance df

H+ Mean Variance df

3.28E-06 2.83E-06 0.03 1.6E-11 1.5E-11 38 38

2.83E-06 2.67E-06 0.72 1.5E-11 1.18E-11 38 38

ANC Mean Variance df

29.3 746.3 38

22.4 939.3 38

0.02

22.4 939.3 38

TOC Mean Variance df

6.43 6.29 38

5.56 6.70 38

3.32E-05 5.56 6.70 38

3.28E-06 2.67E-06 0.14 1.6E-11 1.18E-11 38 38

32.1 1257.1 38

0.0001 29.3 746.3 38

32.1 1257.1 38

0.44

5.98 8.39 38

0.42

5.98 8.39 38

0.41

6.43 6.29 38

4. Discussion 4.1. C RITICAL LOADS The long-term goal of The Canada-Wide Acid Rain Strategy for Post-2000 is to meet the environmental threshold of critical loads for acid deposition across Canada. An Integrated Assessment Model (IAM) used in the 1997 Canadian acid rain assessment report predicted that the sulphate critical load for New Brunswick lakes was 167 eq ha−1 yr−1 (8 kg ha−1 yr−1 ) (Jeffries, 1997). A sulphate critical load for lakes is the maximum amount of wet sulphate deposition that will allow 95% of the lakes to maintain a pH of 6.0 or more (Jeffries, 1997). This is the underlying assessment tool by which the effects of acidifying emissions are measured in Canada and the process by which to examine whether or not further

46

W. PILGRIM ET AL.

emission reductions are required. Average SSC-SO42− deposition to the study area was 337 eq ha−1 yr−1 , which is 50% above the critical load of 167 eq ha−1 yr−1 . This would suggest that the majority of the lakes had a pH of less than 6.0, which was the case. Acid deposition was higher in the first part of the study period and peaked in 1989, at 525 eq ha−1 yr−1 for SO42− and 316 eq ha−1 yr−1 for NO− 3 (Figure 2b). Both nitrate and sulphate play a role in the acidification-recovery process and need to be simultaneously considered when evaluating critical loads. Total acid deposition to the study area in 2000 was 462 eq ha−1 yr−1 or 13.2% less than the average between 1987 and 2000 of 532 eq ha−1 yr−1 (Figure 2c). In 2000, SSC-SO42− deposition to the study area was 285 eq ha−1 yr−1 and nitrate was 177 eq ha−1 yr−1 , 9% less than the 14-year average of 194 eq ha−1 yr−1 . The Integrated Assessment Model projected that at a deposition rate of 250 eq ha−1 (12 kg ha−1 yr−1 ) of SSC-SO42− , 3050 lakes in New Brunswick and Nova Scotia combined would remain chemically and biologically damaged (Jeffries et al., 1999, 2000). Thirty-one (clusters I to III) of the 39 lakes surveyed had an average pH of less than 6.0 and low ANC levels (Figure 3). In spite of the exceedences of the critical load, there were improvements in the levels of lake sulphate, and more recently in ANC and base cations. Other acid-related changes included overall declines in lake colour and base cations and recent decreases in the level of aluminum ion (Figure 4). The Integrated Assessment Model further predicted that 11 to 25% of lakes in eastern Canada would remain chemically altered after the 2010 sulphur dioxide reductions are made as required in the Canada-U.S. Air Quality Agreement. 4.2. ACID RECOVERY Acid recovery is the opposite of acidification and in this study the term refers to either a decrease in acidifying ions, or an increase in base cations, either change may lead to an improvement in pH or ANC. Between 1986 and 2001, lakes with an average ANC of less than 40 µeq L−1 (clusters I to III) showed a loss of 31% from 18.7 to 13 µeq L−1 of ANC, whereas lakes with greater than 40 µeq L−1 (cluster IV) showed a net increase of 14% from 62.2 to 72.4 µeq L−1 per year. Although these are small changes, the point is that the trends are moving in opposite directions. The different responses between the clusters of lakes illustrate the importance when doing regional trends to compare lakes of similar ANC. It is also essential to compare the same window of time and the longer the period the better. The lakes with the least amount of ANC were located on granite bedrock. The pH of lakes not located on granite bedrock (cluster IV) consistently remained greater than 6.0, implying that an ANC of 40 µeq L−1 was sufficient to buffer the level of acid loading measured during the study period. Lakes in clusters I–III showed improvements in pH after 1995 (Figure 3) suggesting that ANC was not sufficient to buffer acid loading occurring prior to 1995, but were able to buffer subsequent loads. It could also be that the increase in pH post-1995 represented a delayed

Figure 3. Changes in acid precipitation related chemistry in lake clusters in southwestern New Brunswick lakes, 1986 and 2001.


47

48

W. PILGRIM ET AL.

response to the large reduction in acid deposition in 1989–1991, as much of the atmospheric acidifying ions are stored in the soils. There are a number of factors that affect changes in the acid or base ion balance in a lake. Some of these include: atmospheric deposition of either acid or base ions, precipitation amount and runoff, weathering rates, geology, soil type, wetlands and marine influence. Between 1980 and 1995, data from 205 lakes and streams in North America and Europe were examined for acid related trends (Stoddard et al., 1999). As with this study, they found that SO2− 4 decreased at most sites measured, concurrent with measured decreases in atmospheric deposition. This would be expected in the New Brunswick lakes as the study area was influenced by similar amounts of sulphate deposition. Although SSC-SO42− declined between 1986 and 2001 in the cluster I to III lakes, it increased near the end of the study period, as did deposition, which suggests a short response time between deposition and run-off rather than a delayed one. This quick response would be particularly true for lakes with little soil and buffering in their surrounding watersheds. Stoddard et al. also showed decreases in base cations in most parts of North America and found little change in ANC. In comparison, in the New Brunswick lakes, ANC decreased in the early part of the study 1986 to 1993, then improved between 1993 and 2001. Other studies closer to New Brunswick, found that pH and ANC declined in lakes sampled biannually between 1989 and 1995 in Nova Scotia and Newfoundland, and these decreases were matched by decreases in SO2− 4 and base cations (Clair et al., in press). In a recent regional analysis of water quality trends in eastern Canada and the northeast States between 1989 and 1999, it was shown that significant improvements for ANC and pH occurred for lakes in southern Québec and southwestern New Brunswick whereas Nova Scotia study lakes and Maine study lakes showed decreasing trends (Dupont et al., 2002). Further surveys are being carried out through this regional lakes monitoring network formed by states and provinces. Although ANC has shown recent improvements in the SW New Brunswick lakes, the majority of the lakes remain acid sensitive having low ANC. One explanation given in the literature which may help explain the observed lower ANC in the New Brunswick study lakes is that after prolonged acid deposition, base cation depletion can occur in soils where Ca2+ can be used up faster than it is made available through weathering. This results in less calcium ion leaching to streams which leads to lower ANC (USGS, 1999). Nitrate deposition is another factor that can influence acid recovery. Between 1987 and 2001, nitrate deposition to the study area averaged 194 eq ha−1 yr−1 , which contributed approximately 37% of the hydrogen ion loading to the soils in comparison to sulphate, which averaged 337 eq ha−1 yr−1 over the 14-year period. However, nitrate was not detectable in the lakes. This suggests nitrate has not reached saturation levels as seen in some other areas of North America.


49

Calcium ion deposition can be an important source of buffering to watersheds but in this study it did not correlate with changes in ANC. Over the study period, Ca2+ deposition ranged between 73 and 104 eq ha−1 yr−1 (Figure 2d). Organic acids can influence acid recovery in lakes and are important in areas of eastern Canada, (Clair et al., 1992, 1995). In the New Brunswick lakes, organic acid levels did not explain the measured changes in ANC and showed no consistent trend in the lake clusters (Figure 4). TOC levels in 1993 were significantly less than in 1986, which would have reduced natural acidity influence on ANC. However, ANC was lower in the earlier survey period. Organic acid levels showed a significant increase between 1993 and 1998 and decreased again in 2001 (Table II). Marine influence on coastal lakes can affect ANC through weathering rates and also ion exchange processes. Marine sodium can exchange with soil calcium, releasing the Ca2+ to the lake waters and also there can be a retention of marine SO2− 4 in the soils (Thompson, 1982; ICP Report, 2000). The study lakes are all within 50 kilometers of the Bay of Fundy coast and annual precipitation to the study lakes is greater than 1000 mm.

5. Conclusions For first part of the survey period (1987 to 1992), average, atmospheric H+ deposition was 452 eq ha−1 yr−1 , compared to 338 eq ha−1 yr−1 between 1993 and 2000, a 25% reduction. In response to this reduction, acid sensitive lakes in southwestern New Brunswick appear to be showing recent signs of acid recovery. However, the response of lakes to reductions in acid deposition was dependent on ANC. Lakes having greater than 40 µeq/L of ANC were able to buffer existing acid loads and maintained a pH of greater than 6.0, whereas lower alkalinity lakes, although showing declines in sulphate and recent increases in pH and ANC, could not maintain a pH of 6.0 and showed a net loss in base cations. Although acid deposition has declined and these lakes are showing signs of acid recovery, 80% of the study lakes remain acid sensitive having little buffering capacity with low calcium, pH and ANC. The study area contains some of the most acid sensitive lakes in the Province and would be likely comparable to coastal lakes elsewhere in the region with similar geologies and water chemistries.

Acknowledgements The authors would like to express their thanks to Jane Spavold-Tims of the Department of Environment and Local Government for her pioneering work on acid deposition and lake acidification in New Brunswick; Sean Fortune of the Department of Environment and Local Government for emissions and deposition data; Tom Brydges, Faculty of Environmental Studies, York University; Hague Vaughn

Figure 4. Changes in colour, aluminum and organic acid levels in the study lakes in southwestern New Brunswick, 1986 to 2001.

50 W. PILGRIM ET AL.


51

of Environment Canada and Paul Arp of the University of New Brunswick for reviewing the manuscript; Bernie Connors, Wade MacNutt, and Jeff Stymiest of the Department of Environment and Local Government for their GIS help with the lake plotting; New Brunswick Power Corporation which supports the operation of the provincial acid rain network; and the Ecological Monitoring and Assessment Coordinating Office of Environment Canada for financial, logistical support in obtaining data and reporting of this assessment in their special issue of this journal.

References CCME (Canadian Council for Ministers of the Environment): 1999, ‘Canadian Acid emissions’, CCME Secretariat, 326 Broadway, Suite 400, Winnipeg, Manitoba, R3C OS5, Canada. Clair, T. A., Pollock, T. L., Collins, P. and Kramer, J. R.:1992, ‘How brown waters are influenced by acidification: the HUMEX lake case study’, Environ. Int. 18, 589–596. Clair, T. A., Dillon, P. J., Ion, J., Jeffries, D. S., Papineau, M. and Vet, R.J.: 1995, ‘Regional Precipitation and Surface Water Chemistry Trends in Southeastern Canada (1983–1991)’, Can. J. Fish. Aquat. Sci. 52, 197–212. Clair, T. A., Ehrman, J. M., Ouellet, A. J., Brun, G., Lockerbie, D., and Ro, C.-U: Changes in freshwater acidification trends in Canada’s Atlantic Provinces: 1983–1997. In press: Water, Air, soil Pollut. Cox, R. M., Spavold-Tims, J. and Hughes, R. N.: 1989, ‘Acid Fog and Ozone, Their Possible Role in Birch Deterioration around the Bay of Fundy, Canada’, Water, Air, Soil Pollut. 48, 263–276. Davies C.: 1993, ‘Wet Deposition Monitoring and Modeling in New Brunswick, An Area Dominated by Wet Deposition due to Long-Range Transport’, Atmos. Environ. 27a, 1035–1049. Dupont, J., Clair, T. A., Couture, S., Estabrook, R., Godfrey, P. J, Jeffries, D. S., Kahl. S., Khan, H., Nelson, S., Peckeenham, J., Pilgrim, W., Stacey, P., Taylor, D. and van Arsdale A.: 2002, ‘Warning Network: Changes in Water Quality in New England and Eastern Canada’. Water Quality Working Group, New England Governors and Eastern Canadian Premiers, P.O. Box 2044 Halifax, NS, B3J 2Z1, e-mail: [email protected] GAO (General Accounting Office): 2000, ‘Acid Rain Emissions and Effects’, United States General Accounting Office, Washington, D.C. 20548, 33 pp. ICP (International Cooperative Program): 2000, ‘The 12-Year Report: Acidification of Surface Water in Europe and North America; Trends, Biological Recovery and Heavy Metals’, Program Center, Norwegian Institute for Water Research, Oslo, Norway, 32 pp. Jeffries, D. S.: 1997, ‘Canadian Acid Rain Assessment. Volume 3, Aquatic Effects’, Environment Canada, National Water Research Institute, Box 5050 Burlington, Ontario, Canada. Jeffries, D. S., Lam, D. C. L., Moran, M. D., Wong, I.: 1999, ‘The Effect of SO2 Emission Controls on Critical Load Exceedances for Lakes in Southeastern Canada’, Water Sci. Technol. 39, 165–171. Jeffries, D. S., Lam, D. C. L., Wong, I. and Moran, M. D.: 2000, ‘Assessment of Changes in Lake pH in Southeastern Canada Arising from Present Levels and Expected Reductions in Acidic Deposition’, Can. J. Fish. Aquat. Sci. 57 (2), 40–49. NESCAUM (Northeast States for Coordinated Air Use Management): 1998, ‘The Northeast States and Eastern Canadian Provinces Mercury Study: a Framework for Action’, NESCAUM, 129 Portland Street, Boston, MA. Oliver, B. G., Thurmond, E. M. and Malcolm, R.:1983, ‘The contribution of humic substances to the acidity of coloured natural waters’, Geochim. Cosmochim. Acta 47, 2031–2035. Potter, R. R., Hamilton, J. B. and Davies, J. L.: 1979, ‘Geological map of New Brunswick’, Map number N.R. – 1, Second edition., Fredericton, N.B., Canada.

52

W. PILGRIM ET AL.

Ro, C. U., Vet, R. J. and Ord, D.: 1998, ‘Canadian Air and Precipitation Monitoring Network (CAPMoN) Annual Summary Reports (1983–1997)’. Environment Canada, Atmospheric Environment Service, Downsview, Ontario, Canada. Shilts, W.: 1981, ‘Sensitivity of Bedrock to Acid Precipitation: Modification by Glacial Processes’, Technical Paper 81–14, Geological Survey of Canada, Booth Street, Ottawa, Ontario, Canada. Stoddard, J. L., Jeffries, D. S., Lükewille, A., Clair, T. A., Dillon, P. J., Driscoll, C. T., Forsius, M., Johannessen, M., Kahl, J. S., Kellogg, J. H., Kemp, A., Mannio, J., Monteith, D., Murdoch, P. S., Patrick, S, Rebsdorf, A., Skjelkvåle, B. L., Stainton, M. P., Traaen, T. S., van Dam, H., Webster, K. E., Wieting, J and Wilander, A.: 1999, ‘Regional Trends in Aquatic Recovery from Acidification in North America and Europe 1980–95’, Nature 401, 575–578. Stumm, W. and Morgan, J.J.: 1996, ‘Aquatic Chemistry: chemical equilibria and rates in natural waters’, Third Edition. John Wiley and Sons. NY, 1022 pp. Thompson, M. E.: 1982, ‘Exchange of marine sodium for calcium during chemical weathering in the Isle aux Morts River basin, Newfoundland’, Geochim. Cosmochim. Acta 46, 361–365. Tims, J. S.: 1986, ‘A helicopter assisted survey of 95 lakes in North Central and Southwest New Brunswick’, paper presented to the Annual Workshop of the Atlantic Region LRTAP Monitoring and Effects Working Group, October 7, 1986, Bedford Nova Scotia. USGS (United States Geological Survey): 1999, ‘Soil-Calcium Depletion linked to Acid Rain and Forest Growth in the Eastern United States’, Report Number WRIR 98-4267, U.S. Geological Survey, Branch Information Services, Box 25286, Federal Center, Denver, Colorado 80225-0286, 12 pp. Watt, W. D., Scott, D. and Ray, S.: 1979, ‘Acidification and other chemical changes in Halifax County lakes after 21 years’, Limnol. Oceanogr. 24, 1154–1161.