IN THE EARLY 1970s significant acid rain damage to fish habitat was discovered in Ontario, Quebec and the northeastern United States. There was an obvious implication that fish habitat in the Maritimes might also be impacted, but research was required to assess this.

DFO's first acid rain research project in the Maritimes was in 1976. This study showed that Halifax County lakes have been acidified, but that most of the acid was from local industry (especially power generation). Then, in 1978, during an investigation of fish kills at Mersey salmon hatchery, it was discovered that the Mersey River water was so acidic that it was toxic to juvenile salmon. This could not be blamed on emissions from Halifax, and so was a strong indication of long range transport of acid pollution (LRTAP). DFO then did a synoptic water chemistry survey (1978-80) of all Atlantic salmon waters in the Maritimes. Along the Atlantic coast of Nova Scotia, an unfortunate combination of hard-rock geology, poor soils and prevailing weather patterns had caused severe acidification of the rivers and lakes. This acid pollution is produced primarily by U.S. industry, and through LRTAP it rains out over the Maritimes. As a result, Nova Scotia is the most heavily impacted province in Canada in terms of the percentage of fish habitat that has been damaged or destroyed by acid rain - and the only region in North America where entire river systems have been acidified by this process.

Since acid rain is a trans-boundary problem, Canada has promoted and participated in both national and international air pollution control agreements. The Eastern Canadian SO2 (Sulphur dioxide) Control Program was established in 1985 whereby the seven eastern provinces agreed to achieve, by 1994, a 50 per cent reduction in annual SO2 emissions. In the United States, the Clean Air Act of 1990 require a 40 per cent of the 1980 level, achieved in two phases. Phase I affecting 110 sources took effect on January 1, 1995, and phase II affecting more than 2,000 sources is due to come into effect on January 1, 2000.

These reductions in emissions will result in less acid deposition, and a substantial degree of recovery was expected in Nova Scotia's acidified rivers over the next 10-15 years, especially in rivers with borderline toxicity (mean annual pH 5.0-5.4).

The Atlantic salmon populations of Nova Scotia's acidified rivers are the only Canadian fishery resource for which strong internationally accepted scientific evidence has been published that directly links the decline of the resource to the long range transport of acid pollution. The fate of the remaining salmon populations, and chemical trends in the waters of these rivers are considered to be key indicators of the success or failure of the interprovincial and international acid emission accords.

Monitoring
In 1982 DFO began a monitoring program to follow the fate of fish populations in a few of Nova Scotia's acidified rivers. The rivers in the monitoring program were chosen to provide a wide range of toxicity conditions, so as to be able to detect the biological and chemical changes that could result from either increasing or decreasing levels of acid precipitation.

Then, in 1989 DFO decided that acid rain was an important national issue, so more resources were made available and a national monitoring program was organized. The national program concentrated on lakes and invertebrates (mostly insects) rather than rivers and fish. The local program was expanded to include lakes and invertebrates. However, the rivers in southwestern Nova Scotia had to be dropped due to a policy decision by the national acid rain committee that only Nova Scotian rivers that contained salmon would be monitored. Acid rain had already destroyed the salmon runs in the southwestern rivers.

Water Chemistry Trends
The water chemistry changes that have occurred in the monitored rivers are rather bizarre: they differ from all computer model predictions (which were the basis of establishing the current North American emission controls) and they are difficult to explain under current theories of acid rain chemistry. The prevailing theory of the long-range transport of acid pollution is that gaseous sulfur and nitrogen pollutants are converted to acid in the atmosphere and the acid falls to earth primarily in precipitation. These strong acids in turn increase the rates of leaching of calcium, magnesium and other metals. Wet deposition of sulfate, measured at Kejimkujik National Park in southwestern Nova Scotia, has declined, and sulfate levels in the river waters show a similar decline over the period 1982-96. Unfortunately, this decline in sulfate concentration has been balanced by an equal increase in organic acids, so the river waters have not changed in pH or toxicity. It is known that organic acids and their salts can act as (supposedly weak) pH buffers, resisting pH change. However, the large organic acid buffering capacity of Nova Scotia's acid river waters was entirely unanticipated. Calcium and magnesium have also declined in the river waters, but to a lesser extent than sulfate. The declines in calcium and magnesium can be accounted for by reduced leaching of these materials from the soils. Similar declines of calcium and magnesium in river runoff that are concomitant with sulfate declines have also been reported from the northeastern United States and from Scandinavia, so the phenomenon appears to be general in acid rain impacted regions.

Along with the decline in calcium and magnesium, the water chemistry shows an equivalent increase in sodium and potassium. Overall then, the total ion content and acid levels in the river waters were the same in 1996 as they were in 1982. The decline in the sulfate has been balanced by an increase in organic anions. The declines in calcium and magnesium have been balanced by the increases in sodium and potassium. A plausible scenario for Nova Scotian rivers that will explain all of the major river water chemistry changes is:

1. Sulfate in the river water has declined due to reduced deposition of acid sulfate in the precipitation.
2. The alkalinity being generated by the reduction in acid sulfate is being absorbed by the buffer capacity of the organic acids.
3. The deposition of acid precipitation over more than 40 years has exceeded the ability of the drainage system to produce calcium and magnesium by the weathering process, and so the soils have become leached (by cation exchange for sodium and potassium which are abundant in sea spray and marine aerosols).
4. Now that acid sulfate deposition is declining, the process has reversed and the soils are accumulating calcium and magnesium (hence the decline in river water concentrations of these substances), and releasing an equivalent amount of sodium and potassium.

The water chemistry changes may actually be more complex than the four points given above. This is merely the simplest scenario that I can devise that adequately explains all of the major river water and precipitation chemical time trends.

For the naturally occurring organic acids to be able to resist a rise in pH this effectively requires a change in their chemical nature. It is urgent that this and other possible scenarios be tested by experiment, and that more realistic computer models be constructed to relate emission controls to water chemistry and toxicity. It is also important to continue the river monitoring program to detect future surprises and to test new computer models. If we do not do the necessary experiments and build new models, then we will have no guide to anticipate the future; and the experiments will be performed anyway, but on us.

When an experiment was conducted in Norway to create a reversal of acidification by roofing over a small (860 m2) acidified drainage area and supplying acid-free artificial precipitation for four years (1984-87), the changes that occurred were very similar to those observed in Nova Scotia's rivers. There were declines in nitrate, sulfate and calcium levels in the runoff, but the calcium decline was not enough to balance the acid sulfate and nitrate declines. This result was interpreted by the experimenters as a recovery, though they noted that there had been no change in the pH of the runoff, because there was an equivalent increase in organic acids.

Implications of the Water Chemistry Trends
The interprovincial and international air quality agreements are based on the hypothesis that reducing sulfate emission/deposition will result in a reduction of acid toxicity in the impacted rivers and lakes. Our results are still marginally compatible with the notion that this acid rain paradigm is operational in Nova Scotia's Southern Upland rivers, but that a recovery will be delayed because of the pH buffering being provided by the organic acids and their salts.

The interprovincial and international agreements only came into effect in 1994 and 1995, and the Canada-USA Accord does not compel major emission reduction until after the year 2000. As this scenario proceeds over the next 5-10 years, the exchange sites for bivalent base cations within the drainage basins of the rivers will eventually approach saturation, and the organic acids will become less effective as pH buffers as they approach their charge density limit. Given, also, that much larger reductions in acid deposition are to be expected after the year 2000, the result should be higher pH levels in the acidified Nova Scotian rivers.

Conclusions
The fate of Nova Scotia's Atlantic salmon resource played a significant role in the negotiation of international controls on acidic emissions, so it is important to monitor these stocks, and the chemistry of their rivers, to assess the chemical and biological responses to the scheduled decline in sulfur dioxide emissions. Our study has shown that, when toxicity levels do drop, those rivers with remnant salmon runs will recover on their own. The probable time frame is 20-50 years.

Recovery of barren rivers will require the order of a century, unless they are restocked from hatcheries. Hatchery restocking can hasten the recovery to 10-20 years, but it carries its own risk of further reducing the genetic diversity in the rivers. There is an urgent requirement for laboratory and field experiments to test the limits of the capacity of Nova Scotia rivers to produce stronger organic acids. This should be followed by experiments to test more realistic theories of acid rain chemistry in Nova Scotia's rivers. As soon as a reliable chemical theory is accepted, a regional computer model must be developed for forecasting toxicity levels in the acidified Atlantic salmon rivers. River monitoring will provide the chemical and biological data needed to test the reliability of the new model.

This article is based on a recent (1999) paper published in the scientific journal "Water, Air, and Soil Pollution", of which Dr. Watt was the senior author. The paper presented detailed analysis of the results from the Federal Department of Fisheries and Oceans' (DFO) acid rain monitoring program in Nova Scotia's acidified Atlantic salmon rivers from 1982 to 1996. The program was discontinued after 1996 due to downsizing in response to budget cuts at DFO.