Past and future chemistry changes in acidified Nova Scotian Atlantic ...

11 downloads 0 Views 428KB Size Report
Nova Scotian Atlantic salmon (Salmo salar) rivers: a dynamic modeling approach. Thomas A. Clair, Ian F. Dennis, Peter G. Amiro, and B.J. Cosby. Abstract: ...
1965

Past and future chemistry changes in acidified Nova Scotian Atlantic salmon (Salmo salar) rivers: a dynamic modeling approach Thomas A. Clair, Ian F. Dennis, Peter G. Amiro, and B.J. Cosby

Abstract: Atlantic salmon (Salmo salar) populations have been extirpated from a number of rivers in Nova Scotia, Canada, as a result of acid rain. We applied the model of acidification of groundwater in catchments (MAGIC) to 35 regional rivers to estimate pre-industrial water chemistry conditions and the potential future changes in water chemistry under three acid deposition scenarios for the region. Our model results indicate that water chemistry in the study streams remained relatively unchanged until the 1950s and reached their maximum effects on pH in the mid-1970s. The main effects of acid deposition have been a decrease in pH and an increase in base cations to surface waters, as the ion-exchange processes in soils release soil cations into surface waters. We forecast future water chemistry in the rivers using three deposition scenarios: no change in sulfate deposition from year 2000 and 10% and 20% sulfate reductions per decade. We show that the more rapid the reduction in acid deposition, the faster the recovery. We also show that although stream water acidity will recover within a few decades, in most streams, base cations will not recover to pre-industrial levels within the next 100 years. Résumé : Plusieurs populations de saumons atlantiques (Salmo salar) ont été éliminées des rivières de NouvelleÉcosse à cause des précipitations acides. Nous avons appliqué le modèle MAGIC (modèle d’acidification des eaux souterraines dans les bassins hydrogéographiques) à 35 rivières de la région pour estimer les conditions chimiques de l’eau avant l’industrialisation et pour prédire les changements futurs de chimie de l’eau dans ces milieux selon trois scénarios de précipitations acides dans la région. Les résultats de la modélisation indiquent que les conditions chimiques de l’eau sont demeurées à peu près inchangées jusqu’aux années 1950 et que les effets les plus importants sur le pH se sont produits au milieu des années 1970. Les effets principaux ont été une baisse du pH et une augmentation des cations basiques dans les eaux de surface, alors que les mécanismes d’échanges ioniques des sols libéraient les cations vers les eaux superficielles. Nous prédisons les caractéristiques chimiques de l’eau dans les rivières selon trois scénarios, soit des précipitations de sulfates inchangées au taux de 2000 et des réductions de 10 % et de 20 % des sulfates par décennie. Plus la réduction des précipitations acides se fera rapidement, plus rapide aussi sera la récupération. Bien que l’acidité de l’eau des rivières puisse se rétablir en quelques décennies, les concentrations de cations basiques ne retrouveront pas leur niveau pré-industriel au cours du prochain siècle. [Traduit par la Rédaction]

Clair et al.

1975

Introduction Acid rain, in the form of sulfuric (H2SO4) and nitric (HNO3) acids, is deposited in catchments. Some of the hydrogen ions accompanying the acid anions are neutralized, first by calcium and magnesium carbonates and then by alumino-silicate minerals. Increases in acid loadings will then cause increases in the calcium and magnesium (base cations) dissolved from soil minerals and leached into runoff waters. If base cation production from weathering of bedrock and soils cannot keep up with acid leaching, then these

will gradually be depleted from catchment ion-exchange sites and replaced by hydrogen and aluminum ions (Kirchner 1992). Large parts of Nova Scotia are underlain by poorly weatherable slates and granites producing soils with naturally low amounts of base cations (Shilts 1981) so that this region is especially vulnerable to acidification effects. Acid deposition effects were first noticed in Nova Scotia in the 1960s. The pH of surface waters used by fish hatcheries in the region was seen to be decreasing, which caused a number of obvious problems to hatchery managers (Farmer et al. 1979). In the mid-1980s, the region received over

Received 19 March 2004. Accepted 22 October 2004. Published on the NRC Research Press Web site at http://cjfas.nrc.ca on 18 December 2004. J18033 T.A. Clair1 and I.F. Dennis. Environmental Conservation Br, Environment Canada – Atlantic Region, P.O. Box 6227, Sackville, NB E4L 1G6, Canada. P.G. Amiro. Science Branch, Fisheries and Oceans Canada, Bedford Institute of Oceanography, P.O. Box 1006, Dartmouth, NS B2Y 4A2, Canada. B.J. Cosby. Department of Environmental Sciences, University of Virginia, Charlottesville, VA, 22903, USA. 1

Corresponding author (e-mail: [email protected]).

Can. J. Fish. Aquat. Sci. 61: 1965–1975 (2004)

doi: 10.1139/F04-196

© 2004 NRC Canada

1966

20 kg·ha–1·year–1 of sulfate (SO4–2), most of which originated in central Canada and the east coast of the United States (Shaw 1979). This deposition clearly had an effect on the fish populations of the region. Watt (1987) showed that formerly viable Atlantic salmon (Salmo salar) populations were extirpated in a number of river catchments and were greatly reduced in others. He was also able to make the link between the reduction of fish and increases in acidification. Since the early 1980s, acid rain controls in both Canada and the United States caused a significant reduction (45%) in SO4–2 deposition in the region (Beattie et al. 2002). However, these reductions have not yet been reflected in positive changes in water chemistry. Clair et al. (2002) examined trends in water chemistry from over 40 lake sites in Nova Scotia from 1982 to 1997. They found that although surface water SO4–2 concentrations decreased along with the deposition, pH and acid neutralization capacity (ANC) did not change. Moreover, they also noted a marked decrease in base cation concentrations (Ca+2 + Mg+2 + K+ + Na+) in a number of the sites. A similar situation was noted in biological studies, as the reduced SO4–2 changes occurring were not seen to have helped local fish populations (Watt et al. 2000). The 16-year data series available for trend analysis is quite good for these kinds of studies, but the results measured tended to present as many questions asked as answered. These changes in water chemistry and fish populations, or lack thereof, emphasized the need to apply geochemical models to the systems to see what kinds of results should be expected. To help answer these questions, we applied the model of acidification of groundwater in catchments (MAGIC) (Cosby et al. 2001) to the rivers in Nova Scotia, which were previously studied by Watt (1987) and which also used to have endemic Atlantic salmon populations. There were therefore a number of questions that needed to be answered in order to allow us to better manage our responses to the acidification of Nova Scotia’s surface waters. First, it was important to determine what were the initial water chemistry conditions at these sites before acidification. Much of our analysis of possible mitigation measures should be moderated by an understanding of what conditions existed in pre-acidification times. We felt that modeling the geochemistry of the catchments might present us with information on earlier conditions and thus produce more realistic management decisions. Secondly, there was a need to understand where the water chemistry in these ecosystems might be heading under different future acid deposition scenarios. This is most important to managing expectations of stakeholders so that they be aware of the time frames involved in the chemical and ecological recovery of these systems. It is also important that decision makers be made aware of the consequences of the choices they make when they decide on the relative rates of acid emission reductions. To present them with predictions of potential changes, we applied hypothetical acid deposition scenarios to the catchments to determine how water chemistry might change. We used three potential scenarios to provide a range of options: no further decreases in deposition from the year 2000 or 10% or 20% decreases in SO4–2 deposition per decade from the year 2000. We did not address the

Can. J. Fish. Aquat. Sci. Vol. 61, 2004

issue of nitrogen acidification as it currently is not obviously causing problems in this region.

Methods Study area The southeastern portion of Nova Scotia is a narrow strip of land draining directly into the Scotian Shelf of the Atlantic Ocean (Fig. 1). Streams and rivers draining this area do not exceed 80 km in length from their headwaters to the estuaries. Beginning in the 1980s, water from 40 streams and rivers containing Atlantic Salmon populations were sampled sporadically by staff of Fisheries and Oceans Canada (DFO) for water chemistry. As the water chemistry varied greatly depending on the years sampled, we use model results normalized for the year 2000 to illustrate the relative water chemistry between the sites (Table 1). Angling catch data extending to 1936 were also available from these sites, so they were known to have had salmon from pre-acidification times. Earlier work by DFO (Watt et al. 2000) related the survival capacity of salmon in these streams to specific 1980 pH ranges. They classified rivers with mean pH values greater than 5.4 as not significantly affected by acidification, and those with pH ranges between 5.4 and 5.0 were identified as having depleted stocks. Sites with pH values between 4.7 and 5.0 still had residual populations, whereas populations at sites with mean annual pHs of less than 4.7 were extirpated. When modeling pH, we will use this classification to place changes in water chemistry into an ecological context. The area’s waters are not only low in natural dissolved ions, but many are also high in natural dissolved organic carbon (DOC), which produces natural organic acids. These originate from littoral wetlands along the water courses and are drained into water courses by runoff. At high concentrations, DOC increases surface water acidity, especially in waters low in basic cations. Clair and Ehrman (1998) have shown that DOC concentrations in this region are controlled by climate. We therefore assume that over the longer term, weather patterns and thus DOC inputs to these catchments will not change, so that their long-term effect on acidity will be constant. Model description A number of mathematical models of soil and surface water acidification in response to atmospheric deposition were developed in the early 1980s (e.g., Christophersen and Wright 1981; Cosby et al. 1985). These models were based on process-level acidification information and were built for a variety of purposes ranging from estimating transient water quality responses for individual storm events to estimating chronic acidification of soils and baseflow surface water. The MAGIC model has now been in use for more than 15 years and has been constantly modified and upgraded (Cosby et al. 1985, 2001). It has been applied extensively in North America and Europe to both individual sites and regional networks of sites. The utility of MAGIC for simulating a variety of water and soil acidification responses at the laboratory, plot, hillslope, and catchment scales has been tested using long-term monitoring data and experimental manipulation data. It has been used to reconstruct the history of acidification, to examine current patterns of recovery, and © 2004 NRC Canada

Clair et al.

1967

Fig. 1. Location of Nova Scotia, Canada, and the study rivers relative to the rest of eastern North America. The study rivers are numbered (see Table 1 for descriptions).

to simulate the future trends in stream water acidity in both individual catchment and regional applications at a large number of sites across North America and Europe (e.g., Wright and Cosby 2003; Clair et al. 2003). MAGIC is a lumped-parameter model of intermediate complexity developed to predict the long-term effects of acidic deposition on surface water chemistry. The model simulates soil solution chemistry and surface water chemistry to predict the monthly and annual average concentrations of the major ions in these waters. MAGIC consists of (i) a section in which the concentrations of major ions are assumed to be governed by simultaneous reactions involving sulfate adsorption, cation exchange, dissolution–precipitation–speciation of aluminum and dissolution–speciation of inorganic carbon and (ii) a massbalance section in which the flux of major ions to and from the soil is assumed to be controlled by atmospheric inputs, chemical weathering, and losses to runoff. At the heart of MAGIC is the size of the pool of exchangeable base cations in the soil. As the fluxes to and from this pool change over time owing to changes in atmospheric deposition, the chemical equilibria between soil and soil solution shift to give changes in surface water chemistry. The degree and rate of

change of surface water acidity thus depend on both flux factors and the inherent characteristics of the affected soils. Atmospheric deposition and net uptake–release fluxes for the base cations and strong acid anions are required as inputs to the model. These inputs are generally assumed to be uniform over the catchment. Atmospheric fluxes are calculated from concentrations of the ions in precipitation and the rainfall volume into the catchment. The atmospheric fluxes of the ions must be corrected for dry deposition of gas, particulates, and aerosols and for inputs in cloud/fog water. The volume discharge for the catchment must also be provided to the model. In general, the model is implemented using average hydrologic conditions and meteorological conditions in annual or seasonal simulations, i.e., mean annual or mean monthly deposition, precipitation, and stream discharge are used to drive the model. Values for soil and surface water temperature, partial pressure of carbon dioxide, and organic acid concentrations must also be provided at the appropriate temporal resolution. For complete details of the model, see Cosby et al. (1985, 2001). As implemented for the Nova Scotian streams, the model is a two-compartment representation of a catchment. Atmo© 2004 NRC Canada

1968

Can. J. Fish. Aquat. Sci. Vol. 61, 2004 Table 1. River name, location, and modeled pH, acid neutralization capacity (ANC), dissolved organic carbon (DOC, in mmol·m–3), and calcium concentrations (in µequiv.·L–1) for 2000. River No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Site

Lat.

Long.

pH

Carleton R. Tusket R. Barrington R. Clyde R. Roseway R. Jordan R. Sable R. Westfield R. Medway R., Caledonia Broad R. West R. Mersey R. Medway R., Hwy. 3 Lahave R. Petite R. Gold R., New Ross Mushamush R. Gold R., Hwy. 3 Middle R., Hennigar L. Middle R., Chester Connaught L. outflow East R., Chester Officers Camp L. Timber L. outflow East R. Wood Road Ingram R. Sackville R. Salmon R., Lake Echo Musquodoboit R. Pembroke R. Ship Harbour R. Tangier R. West R. Sheet Harbour East R. Sheet Harbour Salmon R., Port Moser R. Ecum Secum R. Liscomb R. West River, St. Mary’s St. Mary’s R.

43.9 43.9 43.5 43.6 43.7 43.8 43.9 44.4 44.4 44 44.5 44.1 44.2 44.4 44.3 44.7 44.5 44.6 44.6 44.6 44.6 44.6 44.7 44.7 44.6 44.7 44.8 44.7 44.9 45.3 44.8 44.8 44.9 44.9 44.9 45 45 45 45.3 45.2

–65.9 –65.8 –65.5 –65.4 –65.3 –65.2 –65.1 –65.0 –64.9 –64.8 –64.8 –64.8 –64.7 –64.6 –64.5 –64.4 –64.4 –64.4 –64.3 –64.2 –64.2 –64.1 –64.2 –64.1 –64.1 –64.0 –63.8 –63.4 –63.2 –62.9 –62.9 –62.7 –62.5 –62.4 –62.2 –62.3 –62.2 –62.1 –62.1 –62.0

5.7 5.0 4.9 5.0 4.7 5.3 4.9 5.0 5.5 4.8 5.1 5.7 5.5 5.8 5.7 5.6 6.4 5.2 4.9 5.0 4.8 5.2 5.5 6.6 6.2 5.1 5.1 5.0 6.3 5.8 6.0 5.4 5.5 5.7 5.8 5.7 5.8 5.5 5.8 6.0

ANC 38.7 45.2 20.2 13.9 43.7 16.4 14.3 50.1 56.0 19.4 50.4 27.4 66.7 78.0 50.5 90.3 66.5 68.1 43.6 46.0 23.1 46.9 127.2 135.5 115.0 37.3 58.7 13.5 160.8 83.2 43.1 34.0 29.8 28.5 32.2 31.0 55.5 40.4 52.8 82.2

DOC 16.6 42.8 27.4 19.1 59.0 14.4 20.9 49.6 33.0 29.7 44.2 12.0 40.4 34.7 25.7 47.9 15.5 54.4 46.4 46.9 33.4 37.4 77.4 34.2 38.1 35.2 48.9 18.6 52.4 38.1 14.5 22.5 18.9 13.6 13.9 14.5 24.6 23.4 24.0 31.7

Ca+2 61.6 61.5 35.6 32.5 26.0 28.3 24.0 32.4 37.8 24.3 42.7 33.0 45.6 74.9 62.4 58.1 58.0 50.9 24.9 39.6 23.6 49.8 118.7 128.7 107.6 35.4 99.2 61.2 345.6 255.4 50.6 36.9 37.5 32.9 40.9 38.9 64.4 38.1 37.3 51.5

Note: The rivers are listed from west to east, as acid deposition is greatest in the western region and is reduced with distance from New England and central Canada. Lat., latitude; Long., longitude; R., river; Hwy., highway.

spheric deposition enters the soil compartment, and the equilibrium equations are used to calculate soil water chemistry. The water is then routed to the stream compartment, and the appropriate equilibrium equations are reapplied to calculate runoff chemistry. In this application, we assumed as a first approximation that the soils were podsols with average depths of 1 m. In the model calibration, this is one variable that is sometimes modified, but this starting assumption generally held for the streams with which we dealt. Once initial conditions (initial values of variables in the equilibrium equations) have been established, the equilibrium equations are solved for soil water and surface water concentrations. These concentrations are used to calculate the stream dis-

charge output fluxes of the model for the first time step. The mass-balance equations are (numerically) integrated over the time step, providing new values for the total amounts of base cations and strong acid anions in the system. These in turn are used to calculate new values of the remaining variables, new stream discharge fluxes, and so forth. The output from MAGIC is thus a time trace for all major chemical constituents for the period of time chosen for the integration. Calibration procedure The aggregated nature of the model requires that it be calibrated to observed data from a system before it can be used to examine potential system response. Calibration is achieved © 2004 NRC Canada

Clair et al.

by setting the values of certain parameters within the model that can be directly measured or observed in the system of interest (called “fixed” parameters). The model is then run (using observed atmospheric and hydrologic inputs), and the simulated values of surface water and soil chemical variables (called “criterion” variables) are compared with observed values of these variables. If the observed and simulated values differ, the values of another set of parameters in the model (called “optimized” parameters) are adjusted to improve the fit. After a number of iterations, the simulatedminus-observed values of the criterion variables usually converge to zero (within some specified tolerance). The model is then considered calibrated. If new assumptions (or values) for any of the fixed variables or inputs to the model are subsequently adopted, the model must be recalibrated by readjusting the optimized parameters until the simulated-minusobserved values of the criterion variables again fall within the specified tolerance. Further details on model calibration can be found in Clair et al. (2003). Historical loading The estimation of the historical changes at the sites requires a temporal sequence of historical anthropogenic deposition. Thus, as part of the model calibration process, the model should be constrained by some measure of historical deposition to the site. However, such long-term, continuous historical deposition data do not exist. We used historical emissions data as a surrogate for deposition (Lefohn et al. 1999). The emissions for each year in the historical period were normalised to emissions from a year for which observed deposition data are available (the reference year). Using this scaled sequence of emissions, historical deposition was estimated by multiplying the total deposition estimated for each site in the reference year by the emissions scale factor for any year in the past to obtain deposition for that year. This procedure was applied for the Nova Scotian streams to provide estimates of past deposition for the historical simulations (Fig. 2). From 1982 onwards, measured data were used to provide model input. From the year 2000, we simulate the scenarios described above until no SO4–2 deposition occurs, either in 2050 or in 2100. Application to Nova Scotian streams The MAGIC model was calibrated for 40 of the study streams based on the information collected by DFO (Table 1). For each stream, at least 2 years (but not more than 5 years) of data were averaged to provide calibration inputs for MAGIC. The resultant calibration data periods for the streams ranged from the mid-1980s to the mid-1990s. Soil inputs for the model were based on soil profiles sampled in Kejimkujik National Park located in the middle of the study area, which are typical of the region. Total deposition inputs used for the calibration period at each stream were based on observed wet deposition at Environment Canada’s Canadian Air and Precipitation Monitoring Network (CAPMoN) site at Kejimkujik National Park. This regional wet deposition estimate was adjusted at each site for dry deposition of anthropogenic pollutants and sea salts using mass-balance calculations based on the observed stream

1969 Fig. 2. Relative regional SO4–2 deposition pattern estimated for eastern North America, as well as three possible future scenarios: no change from the year 2000 (䊉), 10% decrease per decade (䊏), and 20% per decade (䉱). The broken line represents the estimated historical deposition.

exports. Historical simulations for each stream were run using estimates of past deposition described above (Fig. 2). Although MAGIC calculates a number of soil and water variables, we focus this paper on the three most important to the survival of Atlantic salmon. The first is pH, which has been the focus of most of the toxicological studies done on acidification and biota in this region and elsewhere (Lacroix et al. 1985). The second is ANC, which is a measure of the water’s ability to neutralize acidity. ANC was estimated as the difference in charge between the sum of the base cations (Ca+2, Mg+2, K+, Na+) and the acid anions (SO4–2, Cl–, NO3–) (Stumm and Morgan 1996). The third parameter on which we focus is the base cation calcium (Ca+2), which is important for a number of osmoregulation and bone maintenance processes in fish (Lacroix et al. 1985), as well as construction of invertebrate shells. It is also the main base cation originating in bedrock found in these waters. The concentrations of the others (K+, non-sea-salt Na+, and Mg+2) are usually less than half that of Ca+2 and behave somewhat similarly. There are no detectable levels of ammonium (NH4+) in these waters, so this ion is not considered in the calculations.

Results Model hindcast outputs suggest that at the beginning of the 20th century, under pre-acidification conditions, pH in all rivers of the region, except those with high DOC, was higher than the 5.4 threshold for sustainable salmon populations in Nova Scotia (Fig. 3). The data show four regions where rivers are extremely sensitive to acidification. The largest region is in southwestern Nova Scotia (rivers 2 to 14 in Table 1) and shows the greatest sensitivity and reactivity to acid deposition. This region is also the closest to major continental acid pollution sources (Shaw 1979). There is another region of sensitive rivers somewhat to the west of Halifax (rivers 18 to 23, Fig. 1). The sensitivity of rivers draining onto the Scotian Shelf of Nova Scotia is controlled © 2004 NRC Canada

Fig. 3. Model hindcast pH of study rivers from 1920 to 2000, as well as predictions for future changes with (a) 10% and (b) 20% decrease in SO4–2 deposition. Green denotes pH > 5.4, which allows sustainable populations; yellow denotes rivers with pH between 5.1 and 5.4 with threatened populations; orange denotes pH between 5.1 and 4.7 and endangered populations; and red denotes pH < 4.7, with extirpated populations.

1970 Can. J. Fish. Aquat. Sci. Vol. 61, 2004

© 2004 NRC Canada

Clair et al.

by geology, with slates and granites underlying the sensitive and acidic regions. The model hindcasts (Fig. 3, here we only show from 1920) suggest that significant changes to the river chemistry only began to be felt in the 1950s. The lowest pH values likely occurred in the mid-1970s to early 1980s when acid rain emissions were at their peak. Since the 1980s (which coincidentally is when ecosystem and acid deposition monitoring began), acidity has been gradually decreasing in regional waters, as has been measured in monitoring programs (Clair et al. 2002). The model simulations show some recovery between 1980 and 2000, with all but one of the sites (Roseway River) having returned to pH levels above extirpation levels for Atlantic salmon by the year 2000. We then applied the three future deposition forecast scenarios (no change from 2000 (NC) and –10% and –20% SO4–2 reduction per decade) to all stream catchments to see how quickly pH might recover (or not) in the future. Under the NC scenario, the model predicts that stream pH will not change further from the year 2000 values (figure not shown), because in the year 2000, the systems are still receiving acid deposition amounts equal to the natural ability of the catchments to neutralize them. When we simulate a 10% SO4–2 reduction per decade to the region (see Fig. 2), pH is predicted to rebound to sustainable levels by 2070 in most of the rivers (Fig. 3a) except for Roseway River in the southwest, which is very dilute and high in organic acids. By 2100, the model predicts that all rivers should have returned to acidity levels acceptable for sustainable fish populations. The predictions for the 20% SO4–2 reduction per decade scenario (Fig. 3b) provided significantly faster pH recovery times for most rivers. The MAGIC model predicts an almost immediate recovery of pH to levels suitable for salmon within 20 years in most rivers. The difference in recovery time between the two deposition scenarios is striking and indicates that linearity is expected in the response of the rivers to changes in deposition levels. ANC, estimated as the difference between the base cations and acid anions, should somewhat mirror the pH. The hindcast (Fig. 4) shows ANC in pre-acidification times to be greater than 50 µequiv.·L–1 at 75% of the sites before 1920, the lower values being due to the lower buffering capacity caused by the poor geology. Based on historical emissions data and estimated changes in acid deposition, the model suggests that some ANC reductions began in the 1920s in the most sensitive rivers of the region. It shows a decrease in ANC beginning in the 1950s, reaching its lowest values (at or below 0 µequiv.·L–1) in the 1970–1980 period in a number of the streams. Reductions in acid deposition from 1980 to 2000 began an improvement in ANC to the point that no streams had ANC values less than 0 µequiv.·L–1, as was the case in the 1970s and 1980s. Only a few sites returned to annual levels greater than 50 µequiv.·L–1 by 2000. The ANC forecast for the NC scenario suggests that little further recovery could be expected in this parameter going into the future. The models predict some slight improvement in two or three sites, but overall, most of the sites should have ANC values between 25 and 50 µequiv.·L–1, with a few having values below 25 µequiv.·L–1. Under the 10% SO4–2 reduction per decade scenario (Fig. 4a), approximately two-thirds of the sites will have returned to values of >50 µequiv.·L–1 by

1971

the year 2050. By the year 2100, when all deposition would have ceased, ANC should have returned to pre-acidification levels. The 20% SO4–2 reduction per decade scenario (Fig. 4b) again shows that an accelerated emission control scenario would have a much more rapid effect on ANC. The model predicts that within 10 years of a 20% per decade sulfate reduction, almost all of the sites would have returned to ANC values greater than 50 µequiv.·L–1, and that within 50 years, all sites would have their values recovered to pre-acidification levels. These predictions, as with the pH values, emphasize the linearity in catchment response to changes in deposition and emphasize that the faster further emission controls are in place, the faster the ecosystems will recover. Base cations in waters draining catchments are important for aquatic biota and also provide a good indication of the weathering and ion exchange processes occurring in the basins. There was a wide range of Ca+2 values in the rivers sampled (Table 1). Modeling showed changes in absolute values at most of the rivers sampled, though patterns were not easily obvious. To better understand the patterns, we plotted the yearly Ca+2 data as the percent difference from pre-industrial conditions (1850 to 1920) (Fig. 5). This allowed us to show how changes in acid deposition would change base cation concentrations relative to the unpolluted state. The hindcasts of absolute values showed changes from pre-acidification beginning at most of the sites in the 1950s. Ca+2 values then increased by 5% to 15% at all sites, whereas those in rivers of the far western and Halifax regions increased by more than 15% in the 1960s and 1970s. As acid deposition decreased from the 1980s, the model predicts a return to pre-industrial concentrations for a few years at all sites. However, continued reductions in deposition cause a continued decrease in dissolved Ca+2 everywhere. Under the 10% SO4–2 reduction per decade scenario, Ca+2 values decrease between 5% and 15% into the foreseeable future, whereas at some of the most sensitive sites, the decrease will be greater than 15%. The greatest decreases will begin at three sites in the Halifax area from 2020 and should be occurring at some western rivers beginning in the 2070s. The 20% SO4–2 reduction per decade scenario shows a more rapid arrival of lower Ca+2 concentrations in the two most sensitive regions. It also shows the return to preacidification levels at three catchments in the late part of the century. Maintenance of SO4–2 deposition at year 2000 levels will not change the relative export of Ca+2 from the catchments in the future, as they will remain with ±5% of preacidification. However, the constant acidification pressure and thus the constant reduction of soil base cations will continue degrading the soils.

Discussion These results show a very important feature of acidification effects on these highly sensitive catchments derived from parent materials poor in buffering capacity. Acid precipitation has the effect of titrating catchment soils. Acid protons exchange with base cations and release them to waters leaving the watershed, so the greater the acid precipitated, the greater the loss of base cations. As the titration is reduced through the decrease in acid deposition, the ion exchange leading to cation loss is also reduced, but from a lower soil cation pool. Thus, runoff waters will contain fewer cations. © 2004 NRC Canada

Fig. 4. Model hindcast acid neutralization capacity (ANC) of study rivers from 1920 to 2000, as well as predictions for future changes with (a) 10% and (b) 20% decrease in SO4–2 deposition. Dark blue denotes mean annual ANC >50 µequiv.·L–1; light blue denotes ANC between 25 and 50 µequiv.·L–1; yellow denotes values between 10 and 25 µequiv.·L–1; orange denotes values between 0 and 10 µequiv.·L–1; and red denotes ANC values