The Science of the Total Environment, 83 (1989) 113-125 113 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
'RED H E R R I N G ' L A K E S AND S T R E A M S I N THE ACID-RAIN L I T E R A T U R E
STEPHEN A. NORTON*, DAVID F. BRAKKE** and ARNE HENRIKSEN
Norwegian Institute for Water Research, P.O. Box 33, Blindern N-0313 Oslo (Norway)
(Received October 20th, 1988; accepted November 29th, 1988)
ABSTRACT
Examples of naturally acidic lakes and streams that occur in areas not receiving acid rain are frequently cited in the literature on acidic precipitation. Both organic acidity and the effect of salt have been postulated as explanations for naturally occurring acidity. The example lakes have been presented as being representative of the processes responsible for regionally acidified lakes in Scandinavia and eastern North America. We have analyzed the published water chemistry data for the examples cited and conclude that the interpretations are incorrect because of faulty or incomplete chemical analysis, and errors in data compilation, summary, and analysis. Further- more, the use of analogy is incorrect in some cases because the examples are not representative of lake types in regions where there are large numbers of recently acidified lakes.
Organic acidity and the sea-salt effect may contribute to the acid-base status of a lake or stream, but there is no evidence that either one is responsible for the regional and recent chronic acidification observed in lakes and streams in areas of North America and Scandinavia receiving acidic precipitation.
INTRODUCTION
I n t h e l a s t s e v e r a l y e a r s n u m e r o u s a r t i c l e s h a v e b e e n w r i t t e n t h a t s y n t h e s i z e a n d d e b a t e s c i e n t i f i c u n d e r s t a n d i n g o f p r o c e s s e s r e s p o n s i b l e for r e c e n t r e g i o n a l a c i d i f i c a t i o n o f l a k e s a n d s t r e a m s ( C o w l i n g , 1982; K r u g a n d F r i n k , 1983, 1984; R i c h t e r , 1983, 1984; J o h n s o n e t al . , 1984; S e i p a n d D i l l o n , 1984; W r i g h t , 1984; H e n r i k s e n , 1984; L e f o h n a n d K l o c k , 1985; K r u g e t al . , 1985; S c h i n d l e r , 1988; R e u s s e t al . , 1988). M a n y s c i e n t i s t s (cf. N a t i o n a l A c a d e m y of S c i e n c e s , 1986; S c h i n d l e r , 1988) h a v e c o n c l u d e d , b a s e d o n t h e r e l a t i o n s h i p b e t w e e n a c i d i c p r e c i p i t a t i o n a n d t h e d i s t r i b u t i o n o f a c i d i c l a k e s , t h a t d e p o s i t i o n o f s u l f u r h a s c a u s e d t h e r e c e n t , r a p i d , r e g i o n a l - s c a l e a c i d i f i c a t i o n o f s t r e a m s a n d l a k e s . C o n v e r s e l y , t h e r e a r e s c i e n t i s t s w h o a r g u e t h a t t h e p r i m a r y c a u s e s o f a c i d i f i c a t i o n a r e n a t u r a l , i n c l u d i n g t h e p r o d u c t i o n o f o r g a n i c a c i d s in w a t e r s h e d so i l s , t h e s e a - s a l t e f fec t (cf. K r u g a n d F r i n k , 1983), a n d l a n d u s e c h a n g e s ( R o s e n q v i s t , 1978a, b).
* Present address: Department of Geological Sciences, University of Maine, Orono, ME 04469, U.S.A. ** Present address: Institute for Watershed Studies, Western Washington University, Bellingham, WA 98225, U.S.A.
These opposing views have stimulated new research that might not have been performed otherwise (e.g., Seip, 1980) and have prompted the re-evalua- tion of much existing data (e.g., Kramer et al., 1986). New large-scale surveys of surface water chemistry have been conducted to characterize the current chemistry of lakes in several sensitive areas (Linthurst et al. (1986) in eastern U.S.A.; Landers et al. (1986) in western U.S.A.; Henriksen et al. (1988a) in Norway; Kortelainen and Mannio (1988) in Finland). Summary analyses of these surveys conclude that excess anthropogenic S deposition is the principal cause of recent regional lake acidification. Acidic lakes are mainly restricted to areas receiving acidic deposition; lakes having low concentrations of base cations in the western U.S.A. or northern Norway receive lower atmospheric loading of anthropogenic S and few lakes are acidic.
The alternative explanations of recent acidification (Richter, 1983, 1984; Lefohn and Klock, 1985; Krug et al., 1985) have been repeatedly cited. We have reviewed the examples of lakes and streams used and in each case have found serious analytical and/or interpretive errors. In some cases, spurious informa- tion or interpretations on a single lake have been advanced as being represen- tative of a large number of lakes in regionally acidified areas of the Northern Hemisphere, thus casting doubt on the interpretation of large bodies of high quality data from those areas. A number of the examples we have found to be unsound were published in two articles by Richter (1983, 1984) and then re-cited by Krug et al. (1985) and Lefohn and Klock (1985) elsewhere.
This paper emphasizes three points: (1) careful scrutiny of water chemistry data before using any part of it; (2) use of information that represents the general case and population level when possible; and (3) clarification and correction of the interpretations (by Richter, 1983, 1984) that are critical with respect to points originally raised by Krug and Frink (1983).
NATURALLY ACIDIC SURFACE WATERS
We have examined many of the data sets cited by Richter (1984), Krug and Frink (1984), Krug et al. (1985), and Lefohn and Klock (1985) in support of their discussions about acidic and acidified lakes and streams. We discuss the data and their interpretations for individual sites. We illustrate our discussions with calculations based on the original data (Tables 1-3) because of transcrip- tion errors committed by the above authors. Sea-salt corrected values have been calculated by us from the original data.
Lakes
Lake 57.1 Lake 57.1, in western Norway, is located ~ 100 km north of Sognefj orden and
10km from the Norwegian Sea, in an area receiving precipitation with a volume-weighted pH of -~ 4.8 (Wright et al., 1977). Richter (1984) (re-cited by Krug et al., 1985) states: "This (acidification from the sea-salt effect) is well illustrated by Norwegian data in Table 1 which shows that lake water pH was decreased markedly by runoff greatly enriched in salts" (parentheses are ours). The data cited are from 1974 and 1976. Richter maintained that the water
quality in 1976 resulted from a sea-salt effect, using the water quali ty in 1974 as a point of reference. This argument is unsound. Lake 57.1 is a dimictic lake and the flushing time is much less than one year, calculated on a whole lake basis, and even less if calculated using the epilimnetic volume. Consequently, the water quality in 1976 was determined by events in 1976, not in 1974.
Evaluat ion of the water chemistry from the point of view of a sea-sait effect results in the following. If a salty precipitation event occurs, the watershed receives a solution with Na/C1 close to 0.86 (that in sea-salt). In 1974, excess Na (measured total Na corrected for sea-salt) was + 28 ttequiv 1 1. The Na/C1 ratio was 1.13. One would expect that the Na/C1 ratio produced by the mixing of incoming precipitation and groundwater would be intermediate between 0.86 and 1.13 (it is 1.05). However, if there is to be a sea-salt effect, Na ~ (and typically Mg 2+) must be adsorbed by the soil and H ÷ released. The excess Na value must be less than the original value (+ 28 in 1974). However, the value in 1976 was + 62. These two water chemistries cannot be linked to demonstrate a sea-salt effect.
In point of fact, other data available to Richter, but not cited, do suggest a sea-salt effect on the acidity status of Lake 57.1. The pH varies from 6.1 to 4.9 and C1 and pH do co-vary reciprocally. However, the degree to which the effect of salt operates can only be estimated by interpreta t ion of frequent water samples (see, e.g., Wright et al., 1988). The maximum sea-salt effect we have observed for Lake 57.1 is 36 #equiv 1-1, on an episodic basis. This acid contribu- tion can be added to that of SO*(excess anthropogenic SO4) of the lake, which is continuously 30-35 pequiv l- 1. The decline in SO*, ra ther than indicating less acidification, as suggested by Richter and Krug et al., is nearly compensated by a reduction in the sum of Ca* + Mg*. This reduction in base cations is caused by dilution from precipitation.
Lake 25.1 Lake 25.1 is located --~ 90 km south of Sognefjorden, ~ 40 km from the open
ocean but only 10km from salt water (Wright et al., 1977). Precipi tat ion has an average pH of ~ 4.6. This lake was also cited as an example of acidification due to neutral-salt effects in soils (Richter, 1984). The pat tern of water chemistry (Table 1) for 1974 and 1976, respectively, was: C1, 59 and 166gequiv 1-1, Na/C1, 0.93 and 1.05; Na*, +4 and +31~equiv 1-1. The pat tern is the opposite of what is expected if these two water chemistries were linked by an event due to the sea-salt effect (Norton et al., 1987; Wright et al., 1988). As in the case of Lake 57.1, the contr ibut ion of excess SO4 to the acidity status of this lake is 30-35 gequiv l-1. A sea-salt effect could occur at Lake 25.1, but the data do not support this interpretat ion. The same hydrologic arguments raised in the Lake 57.1 discussion also apply to Lake 25.1.
Lake Pedder Lake Pedder (as well as the Maria lakes discussed below) can be located on
older maps of southwestern Tasmania. These lakes were situated approximate-
119
ly 40 km from the ocean, in an area receiving rainfall exceeding 2 m year- l , with high concentrations of marine Salts. These lakes no longer exist, having been inundated by a hydroelectric power reservoir. Nonetheless, it is instruc- tive to discuss their chemistry (from Buckney and Tyler, 1973a, b) and mis- interpretations by Richter (1984) (re-cited by Krug and Frink, 1984); Krug et al. (1985) introduced a third sample (Table 2). Richter (1984) suggests '~that relatively low pH (for the Tasmanian lakes cited) results from a combination of weak organic acidity and salt effects" (parentheses are ours).
Richter cites the partial chemistry of two samples from Lake Pedder, taken 13 days apart, as evidence for the effect of salt and acidification due to organic acids. The referenced analytical data (Table 2) are suspect. First, HCO3 is reported as 12 pequiv 1-1. There should be virtually zero HCO3 and negative alkalinity at pH = 4.64. This discrepancy may be caused by improper calcula- tion of alkalini ty (Henriksen, 1982). Second, it is difficult to invoke organic acids as an acidifying agent because color only increased from 120 to 150Pt color units between samples, suggesting that the anionic contribution changed no more than ~ 10-15pequiv 1-1. Third, Lefohn and Klock (1985) state that Buckney and Tyler (1973a) concluded ~'th~low pH waters (are due to) natural ly occurring humic acids" (parentheses are ours). No such statement is made by Buckney and Tyler (1973a). Krug et al. (1985) cite data for dissolved organic carbon (DOC) in their table; no such data are given by Buckney and Tyler (1973a or b). Given the measured value of color in Lake Pedder (120-150Pt units), some organic contribution to acidity must occur. Assuming that the analytical data are correct in Buckney and Tyler (1973a), the organic contribu- tion is 25-37 pequiv 1 1 for the first two samples cited in Table 2. It is 109 ttequiv 1 1 for the third sample, a considerable increase even though color increased by only 30 Pt units. This latter value is close to what might be an expected contribution given a color of 150; the first two samples are lower than would be expected based on the color. These discrepancies suggest analytical problems in the original data.
If we accept the data at face value, chemical conditions changed as follows: C1, 260-451pequiv 1 1; Na/C1, 1.05-0.86; Na*, 50 to - 1/tequiv 1 1. These data suggest a nearly complete replacement of water in the ground/surface water complex by salty rainwater. The amount of acidification from the effect of salt cannot be determined because the volumes of mixing waters are not known. Both Na/C1 and Na* would be reduced independently because of dilution by precipitation that contained sea-salt. The maximum calculated sea-salt effect is 51pequiv 1 1. If there had been 1:1 dilution, the sea-salt effect would have been sufficient to drop the pH and alkalinity to their final values. Thus, these data, at first glance, do support a distinct sea-salt effect, as Richter implies, but uncertainties associated with the analytical data remain unresolved.
Krug et al. (1985) cite data from a third sample from Lake Pedder. Similar difficulties are encountered as discussed above. Positive alkalinity is reported for a pH of 4.8; color is reported as 120 Pt units, which certainly contributes to the acidity status. The Na/C1 ratio is 0.70 and the Na* is -32#equiv 1-1.
120
These observations suggest a sea-salt effect of at least 32 #equiv 1 1. However, it is curious, and unexplained, that all analyses for 1969, 1970 and 1971 show Na/C1 less than 0.71 and negative Na*, with the most negative Na* values ( - 72 to -79) associated with pH values only over 5.1. This pattern is the exact opposite of that expected from the sea-salt effect, suggesting that there are significant analytical problems, probably including an overestimate of C1. These problems influence the interpretation of all data for Lake Pedder.
Lake Maria Lake Maria is a term applied to a series of shallow, interconnected pools
(titled 1, 3, 4, 5, 6 and 7 by Buckney and Tyler, 1973a), all of which emptied into Lake Pedder, 1 km away. Although data for these pools are reported separately, they represent virtually one lake. One sample from each of three lakes are used as evidence of lakes being acidified by both the sea-salt effect and organic acids (Richter, 1984; Krug et al., 1985).
The Na* values (Table 2) for these three water samples taken in 1971, 1969 and 1969, range from - 57 to - 137#equiv 1 1 and all values prior to 1972 are strongly negative; 3 years of continuous pH depression is not what one expects from the sea-salt effect. All Na* values in 1972 are positive. Although Buckney and Tyler (1973a or b) do not mention a change in analytical methods at that time, the simplest explanation for the abrupt change in Na* is a change in the method for determination of C1.
Lastly, even if one accepts the original data as correct, the ion balances for Lake Maria samples 3 and 5 do not allow for any contribution from organic anions. Given the color values measured, there must be some contribution. However, if organic acids are important, as claimed, there must be analytical errors in the original data for the other ions, which influence the interpretation of sea-salt effects. No complete evaluation of contributions to acidity can be made from the data available and the conclusions of Richter (1984) and Krug et al. (1985) are unsupported.
Perched Lake Perched Lake, in southwestern Tasmania, is cited by Krug et al. (1985) as an
example " tha t neutral salts can result in acidification of water". They gave chemical data (mean values for samples from 10 m depth, Table 2) for the lake as evidence. There are several problems with this example. First, the lake is used as an example of natural contributions of acidity, such as what might occur in areas receiving acid rain. However, Perched Lake is a sink-hole with no surface inlet or outlet (King and Tyler, 1981), and thus is not representative of the acidic drainage lakes of northeastern North America or Scandinavia, with which it was compared. Its relation to acidic seepage lakes in the U.S.A. (Eilers et al., 1988a-c) is unknown.
Second, there was a large difference between field pH (4.57) and the laboratory pH (5.04). King and Tyler (1981) indicate that a partial oxygen depletion (up to 80%) occurred on a yearly basis (their Fig. 3). The difference
121
in pH observed could be related to CO2 overpressures in bottom waters (ap- proaching 10-15 atm, a value we have observed in lakes in Norway (Norton and Henriksen, 1983)) and degassing of the samples between the time of field sample collection and measurement and laboratory measurement. The overpressures of CO2 could also have been partially preserved during storage (Norton and Henriksen, 1983), and therefore the air-equilibrated pH may be greater than 5.04, which would be consistent with the positive alkalinity.
Third, King and Tyler (1981) report chemical constituents in mgl ' and ~equiv 1 1%. However, Krug et al. (1984) inadvertantly mis-state the data in their Table II, citing ttequiv 1-1% as #equiv 1 1, leading to serious errors.
Finally, both sea-salt corrected Na and K are virtually 0 (Table 2). A sea-salt effect can only be identified definitively in a single sample (or in this case a mean, n = 14) if the Na* is negative. King and Tyler (1981) do show temporal chemical data for the 10-m depth and there was a particularly salty sample in April of 1977; however, the pH remained constant during the salty event.
Nineteen Australian lakes The chemistry for 19 lakes (from Bayley (1964)) from coastal regions of
continental Australia is cited by Krug and Frink (1984) and Krug et al. (1985). First and foremost, these lakes are hardly typical of all the types of lakes that have been subjected to increased SO4 loading in eastern North America or Scandinavia. The 19 lakes are perched seepage lakes, out of contact with the regional groundwater system, situated in hollows in coastal sand-dune complexes. Seepage lakes (typically not perched) do occur commonly in the upper mid-west U.S.A. and Florida, and some of these are acidic (Eilers et al., 1988a-c). However, the acidic lakes in northeastern U.S.A. are largely drainage systems (Linthurst et al., 1986; Brakke et al., 1988a), as are the acidic lakes in southern Norway (Henriksen et al., 1988a). The perched sand-dune lakes of Australia are not representative of the hydrologic setting of, and thus should not serve as chemical examples of, regionally acidified drainage lakes in northeastern North America or in Scandinavia.
Further, Krug and Frink (1984) state (numbers in parentheses keyed to points discussed below):
"For example, a s6ries of 19 (1) Australian lakes with a mean pH of 4.65 and 15.7 milligrams per liter of dissolved organic matter (2) have little or no anion deficit (3). The strong acids (4) apparently came from ion exchange with salts in rain". Examination of our numbered items in the quote is revealing.
(1) Table 1 of Bayley (1964) indicates that there are at most 17 lakes to be considered; his abstract states 16 lakes. The incorrect number of lakes used by Krug et al. (1985) in their calculation of the mean chemistry leads to non-trivial errors.
(2) Bayley (1964) clearly outlines his methods. Organic matter was reported as mgO2 consumed (not DOM as reported by Krug et al., 1985), using the permanganate method appropriate for the time. This technique oxidizes only ~40% of the DOC and the absolute value (in mgO21-1) of the amount of
122
oxidation is nearly the same as the amount of measured DOC (expressed in mg 1 1) (Henriksen, unpublished data).
(3) Bayley (1964) clearly states that C1 and H C Q were measured and SOn was determined by difference, i.e. SO4 was assigned a value equivalent to the remaining anions required to balance measured cations, assuming no other anions were present. By definition, this approach yields no anion deficit, the ion balance is perfect, and therefore organic anions = 0. However, given measurements of organic matter, some organic anions would be expected.
Analytical difficulties are suggested because many lakes with pH < 5 have positive alkalinity, none have negative alkalinity, and the average lake with a pH of 4.74 has HCO3 - + 11pequiv 1 1. This could be caused by CO2 over- pressures, but the data are not sufficient to determine this.
(4) Krug and Frink (1984) do not define what is meant by strong acids. If they are suggesting H ÷ due to excess SO4, the statement is without meaning, as is the proposed mechanism of generation. The reported SO4 concentration (calculated by differenc e ) incorporates all analytical errors into it as well as the entire amount of organic anions and NO3 present in the system. The appeal for some type of sea-salt effect to generate significant S Q by anion desorption is chemically unsound.
If, by strong acid, Krug and Frink (1984) meant H ÷ ion, then appealing to the sea-salt effect also fails. Na is the only cation that has positive sea-salt corrected values, the opposite of the expected if sea salts were displacing H ÷. The very low Ca concentrations (no excess Ca*) imply that there is virtually no contact between any inorganic soil and the lake, further reducing the possibility of the sea-salt effect. We suspect that these lakes are natural ly acidic due to the concentration of dissociated organic acids being greater than the concentrations of sea-salt corrected base cations (Henriksen and Brakke, 1988). Unfortunately, the presence and role of organic acids cannot be determined, based on the results generated by the analytical methods used by Bayley (1964).
For Lake Pedder, Lake Maria and the ~'19" Australian lakes, the apparent concentration of DOC, based on various measures, far exceeds typical values for acidified lakes in the eastern U.S.A. (~ 6 mg 1 1; Linthurst et al., 1986) and Norway (~2mgl -1 ; Henriksen et al., 1988a, b). Thus, apart from other problems discussed above, the cited lakes are not representative of recently acidified lakes in those areas.
Streams in Scotland
Richter (1984) describes data from a group of streams in Scotland (Harriman and Morrison, 1982) where "Stormflows depress pH greatly . . . . " and "Sea salt effects therefore appear to be of primary significance to runoff pH in these catchments", and '~Very likely, weak organic acidity also contributes to low pH of these relatively clear streams, although its effect has not been quantified". Each of these three statements needs scrutiny.
123
the water added to a stream as a result of precipitation or snow melt. If pH decreases as discharge rises, there must be either higher concentrations of acid in the water, derived either from the groundwater/soil system or from the atmosphere, or less base due to dilution, or both.
Second, Richter (1984, Table 1) presented the average chemistry of two moorland streams and three forest streams from Harriman and Morrison (1982). The original data are reproduced in Table 3. Intracomparison of the moor streams and the forest streams reveals significant differences with respect to pH and Ca*. Higher Ca* at constant SO4 should result in higher pH, as is the case in these two stream types (Table 2). The argument for a sea-salt effect is poor. The samples represent a complete year of data and cannot reveal an episode (Norton et al., 1987). The Na:C1 ratio in all five streams is very close to that of sea water (0.86), ranging from 0.81 to 0.89, very similar to the ratio in precipitation (0.88) for the year (Harriman and Morrison, 1982). Thus, Na* is nearly zero, with the most negative value ( - 8 pequiv 1-1) occurring in the stream with the highest pH. The demonstrable maximum effect for sea-salt acidification for the average moorland and forest stream is only - 2 and - 1 #equiv 1 ', respectively. There appears to be little question that there is an effect due to afforestation. More deposition of NaC1 occurs in forested areas as compared with moorland, owing to an increased flux of dry deposition. However, there is no significant difference in a demonstrable sea-salt effect between the two types of sites in spite of the higher Na concentrations in the forest streams.
Third, the ion balances for the five streams have slight excesses of cations over anions (a maximum of 11/~equiv 1-1). This suggests that organic anions can only contribute minimally to the acidity status of the streams (Harriman and Morrison, 1982).
It is clear from the data that SO*plus NO*are ,-- 114 and 139pequiv 1-' for the moorland and forest streams, respectively. These values exceed the sum of Ca* + Mg* + Na* + K* by 10 and 55#equiv 1 1, respectively, producing the slight acidification of the moorland streams (loss of alkalinity and slightly depressed pH) and considerable acidification (including A1 mobilization) of the forest streams. The forest apparently increases the capture efficiency of sulfate over moorland sites, Eis suggested by Harriman and Morrison (1982) and Norton et al. (1988), leading to the lower pH.
SUMMARY
Examples of frequently cited lakes purportedly acidified to pH < 5 by organic acidity and the sea-salt effect (Richter, 1984; Krug et al., 1985; Krug and Frink, 1984; Lefohn and Klock, 1985) have been shown to be based on incorrect interpretations and faulty or questionable analytical data. The cases are neither relevant to the processes causing regional acidification of lakes in eastern North America, and Sweden and Norway, nor representative of the majority of lakes. These incorrectly used examples illustrate the importance of
124
First, the pH of a s t ream may go down or up, depending on the chemis t ry of careful sc ru t iny of ana ly t ica l methods and a c lear under s t and ing of hydro logic sett ings, wa te r chemis t ry and l imnology before da ta are used. Organic acids and the sea-salt effect con t r ibu te to the sensi t ivi ty and acidi ty of lakes (Gorham et al., 1986; Brakke et al., 1987; Wr igh t et al., 1988), but are not responsible for the recent regional acidif icat ion of lakes in eas tern Nor th Amer ica and sou the rn Scandinavia . These na tu ra l con t r ibu t ions to acidi ty must be careful ly considered and evalua ted in all cases. Richter (1984), Krug et al. (1985) and others, in a t tempts to demons t ra te tha t na tu ra l sources have resul ted in the reg ional acidif icat ion of lakes, have used poor and non-represen- ta t ive examples. Na tu ra l sources of acidi ty can be impor tant , but complete and high qual i ty water chemis t ry data are required in order to assess the contri- butions.
ACKNOWLEDGMENTS
This paper was wr i t ten while Brakke and Nor ton were on sabbat ical leave at the Norweg ian Ins t i tu te for Wate r Research from Weste rn Wash ing ton Univers i ty and the Univers i ty of Maine, respectively. We also apprec ia te the g ran t from N T N F to Brakke. Helpful comments were received from Joe Eilers and Gene Likens. Mare l la Bunc ick made m a n y improvements to the text. We were suppor ted in par t by U.S. EPA Coopera t ive Agreement CR812653 with Wes te rn Wash ing ton Univers i ty . The paper has not been subjected to EPA review and no official endorsement from the Agency should be inferred. This paper is a con t r ibu t ion from the Na t iona l Surface Wate r Survey of the EPA Aquat ics Effects Research Program.
REFERENCES
Bayly, I.A.E., 1964. Aust. J. Mar. Freshwater Res., 15: 56-72. Brakke, D.F. and Henriksen, A. and Norton, S.A., 1987. Nature, 329: 432-434. Brakke, D.F., Landers, D.H. and Eilers. J.M., 1988a. Environ. Sci. Technol., 22: 155-163. Buckney, R.T. and Tyler, P.A., 1973a. Aust. J. Mar. Freshwater Res., 24: 267-273. Buckney, R.T. and Tyler, P.A., 1973b. Int. Rev. Gesamten Hydrobiol., 58: 61-78. Cowling, E.B., 1982. Environ. Sci. Technol., 16: ll0A 123A. Eilers, J.M., Brakke, D.F., Landers, D.H. and Kellar, P.E., 1988a. Verh. Int. Verein. Limnol., 23:
144-151. Eilers, J.M., Brakke, D.F. and Landers, D.H., 1988b. Environ. Sci. Technol., 22:164 172. Eilers, J.M., Landers, D.H. and Brakke, D.F., 1988c. Environ. Sci. Technol. 22: 172-177. Gorham, E., Underwood, J.K., Martin, F.B. and Ogden, J.G., III., 1986. Nature, 324: 451-453. Harriman, R. and Morrison, B.R.S., 1982. Hydrobiologia. 88: 251-263. Henriksen, A., 1982. Vatten, 38: 83-85. Henriksen, A., 1984. Environ. Sci Technol., 18: 631~32. Henriksen, A. and Brakke, 1988. D.F. Environ. Sci. Technol., 22: 8-14. Henriksen, A., Brakke, D.F. and Norton, S.A., 1988a. Environ. Sci. Technol., 22: 1103-1105. Henriksen, A., Lien, L., Traaen, T.S., Sevalrud, I.S. and Brakke, D.F., 1988b. Ambio, 17: 25~266. Johnson, N.M., Likens, G.E., Feller, M.C. and Driscoll, C.T., 1984. Science, 225: 1424-1425.
125
King, R.D. and Tyler, P.A., 1981. J. Mar. Freshwater Res., 32: 501-515. Kortelainen, P. and Mannio, J., 1989. Water Air Soil Pollut. (in review). Kortelainen, P., Mannio, J. and Makinen, I., 1987. Aqua Fenn., 16:221 229. Kramer, J.R., Andren, A.W., Smith, R.A., Johnson, A.H., Alexander, R.B. and Oehlert, G., 1986. In:
Acid Deposition, Long Term Trends. National Academy Press, Washington, DC, pp. 231-299. Krug, E.C. and Frink, C.R., 1983. Science, 221: 520-525. Krug, E.C. and Frink, C.R., 1984; Science, 225: 1427-1429. Krug, E.C., Isaacson, P.J. and Frink, C.R., 1985. J. Air Pollut. Control Assoc., 35: 109-114. Landers, D.H., Brakke, D.E., Overton, W.S., Kellar, P.E., Silverstein, M.E., Schonbrod, R.D.,
Crowe, R.E., Linthurst, R.A., Omernick, J.M., Teague, S.A. and Meier, E.P., 1987. U.S. EPA 600/3-86/056a.
Lefohn, A.S. and Klock, G.O., 1985. J. Air Pollut. Control Assoc., 35: 632~37. Linthurst, R.A., Landers, D.H., Eilers, J.M., Brakke, D.F., Overton, W.S., Meier, E.P. and Crowe,
R.E., 1986. U.S. EPA 600/4.86/007a. National Academy of Sciences. 1986. Acid Deposition. Long Term Trends. National Academy
Press, Washington, DC. Norton, S.A. and Henriksen, A., 1983. Vatten, 346-354. Norton, S.A., Henriksen, A., Wright, R.F. and Brakke, D.F., 1987. Proceedings, Geochemical
Monitoring in Representative Basins, United Nations, New York, pp. 148-150. Norton, S.A., Kahl, J.S., Brakke, D.F., Brewer, G.F., Haines, T.A. and Nodvin, S.C., 1988. Sci. Total
Environ., 72: 183-196. Reuss, J.O., Cosby, B.J. and Wright, R.F., 1987. Nature, 329:27 32. Richter, D.D., 1983. Environ. Sci. Technol., 17: 568-570. Richter, D.D., 1984. Environ. Sci. Technol., 18: 632~37. Rosenqvist, I. Th., 1978a. Sci. Total Environ., 10: 39-49. Rosenqvist, I. Th., 1978b. Sci. Total Environ., 10:271 272. Schindler, D.W., 1988. Science, 239: 149-157. Seip, H.M., 1980. In: D. Drablos and A. Tollan (Eds), Ecological Impact of Acid Precipitation. SNSF
Project, Oslo, pp. 358-366. Seip, H.M. and Dillon, P.J., 1984. Science, 225: 1425-1426. Wright, R.F., et al., 1977. IR 33/77, SNSF Project. Oslo. Wright, R.F., 1984. Science, 225: 1426-1427. Wright, R.F., Norton, S.A., Brakke, D.F. and Frogner, T., 1988. Nature, 334: 422~424.
