A ten year survey of water quality in 330 Massachusetts streams was conducted to examine the rate and pattern of recovery from acidification. Meta-analysis was used to combine the results of the 330 non-parametric trend tests into an overall test for trend in a variety of water quality variables including pH, acid neutralizing capacity (ANC), and major inorganic ions. Analysis of trends in the raw data indicates both pH and ANC are increasing. After detrending for variations in stream runoff, we estimate the streams are recovering from acidification at a rate of +0.021 pH units/year and +2.4μeq/L/year, for pH and ANC respectively. These trends appear to be related to declines in sulfate (−1.8μeq/L/year), while base cation trends were mixed. Meta-analysis reveals the trends are not always homogeneous between seasons or between sites.While it is commonly assumed that the low ANC systems are most ‘sensitive’ to changes in acid or base inputs, we found the greatest rates of change in ANC were associated with the high ANC systems. The greatest increases in pH were seen in the low ANC streams as expected. The results also suggest streams respond quickly to changes in precipitation inputs and stream monitoring networks may be valuable as an early detection technique for changes in environmental quality.

In June 1983 a whole-catchment liming experiment was conducted at Tjønnstrond, southernmost Norway, to test the utility of terrestrial liming as a technique to restore fish populations in remote lakes with short water-retention times. Tjønnstrond consists of 2 small ponds of 3.0 and 1.5 ha in area which drain a 25-ha catchment. The area is located at about 650–700 meters above sea-level in sparse and unproductive forests of spruce, pine and birch with abundant peatlands. A dose of 3 ton/ha of powdered limestone were spread by helicopter to the terrestrial area. No limestone was added to the ponds themselves. The ponds were subsequently stocked with brown and brook trout.Liming caused large and immediate changes in surface water chemistry; pH increased from 4.5 to 7.0, Ca increased from 40 to 200μeq/L, ANC increased from −30 to +70μeq/L, and reactive-Al decreased from about 10 to 3μmol/L. During the subsequent 11 years the chemical composition of runoff has decreased gradually back towards the acidic pre-treatment situation. The major trends in concentrations of runoff Ca, ANC, pH, Al and NO₃ in runoff are all well simulated by the acidification model MAGIC. Neither the measured data nor the MAGIC simulations indicate significant changes in any other major ion as a result of liming.The soils at Tjønnstrond in 1992 contained significantly higher amounts of exchangeable Ca relative to those at the untreated reference catchment Storgama. In 1992 about 75% of the added Ca remains in the soil as exchangeable Ca, 15% has been lost in runoff, and 10% is unaccounted for.The whole-catchment liming experiment at Tjønnstrond clearly demonstrates that this liming technique produces a long-term stable and favourable water quality for fish. Brown trout in both ponds in 1994 have good condition factors, which indicate that the fish are not stressed by marginal water quality due to re-acidification. The water quality is still adequate after 11 years and >20 water renewals. Concentrations of H⁺ and inorganic Al have gradually increased and approach levels toxic to trout, but the toxicity of these are offset by the continued elevated Ca concentrations. Reduced sulphate deposition during the last 4 years (1990–94) has also helped to slow and even reverse the rate of reacidification. The experiment at Tjønnstrond demonstrates that for this type of upland, remote terrain typical of large areas of southern Norway, terrestrial liming offers a suitable mitigation technique for treating acidified surface waters with short retention times.