Significant resources are spent on counteracting the effects of acidification, mainly by liming. Due to lower S and N deposition in Europe and North America, authorities are changing directives and strategies for remediation and reducing liming. However, as the acid-base buffer capacity differs in different water bodies, the desirable reduction of the lime dose is variable. In this study, a geochemical model is used to predict pH and inorganic monomeric Al (Ali) when liming is reduced and finally terminated in the 3000 Swedish lakes currently treated with lime. To estimate Ca and Mg concentrations not affected by liming for use in the model, the Ca/Mg ratio in nearby unlimed reference lakes was used. For the modelling of pH and inorganic Al the Visual MINTEQ program including the Stockholm Humic Model recently calibrated for Swedish fresh water was used. The predictions were validated with modelling results from six monitored lakes, in which liming had been terminated. The use of geochemical modelling appeared to be a promising tool for the calculation of accurate lime requirements in acid waters. For simulations in which liming was completely terminated, the pH value decreased by, on average, 1 pH unit to pH 5.7, whereas Ali increased by 17 mu g L (1) to 32 mu g L (1). If liming was reduced by half, the pH would drop only 0.3 pH units and Ali would increase by 2 mu g L (1). Lakes in the south-western part of Sweden were predicted to reach a lower pH and higher Ali, which would be expected due to their greater historical S deposition. The results indicate that liming can be terminated in certain areas and in other areas be reduced without increases in the lake acidity.

While chronic acidification of water bodies has been steadily decreasing, episodic acidification continues to affect stream biology by temporarily decreasing pH and mobilizing aluminum. These events are becoming more common as climate change renders more frequent and intense storms and flooding. Throughout Scandinavia, the effects of acidification have been mitigated by liming since the 1980s, but remediation efforts can now be reduced. While transient acidity may reduce fish populations, also other species in streams are affected. In this in-stream study, two macro-invertebrates (Gammarus pulex and Baetis rhodani), both known as salmonid prey organisms, were exposed to snowmelt in six humic brooks with a natural gradient of pH and inorganic monomeric Al (Al-i). We hypothesize that acid toxicity thresholds can be defined using lethal (mortality) and sublethal (changes in body elemental content) metrics. Periodic observations were made of mortality and whole body concentrations of base cations (BC: Ca, Mg, Na and K) and metals (Al, Fe, Zn and Mn). Mortality increased dramatically at pH < 6.0 and Al-i > 15 mu g/L for G. pulex and at pH < 5.7 and Al-i > 20 mu g/L for B. rhodani. No accumulation of Al was found. The invertebrate body Na concentration decreased when pH dropped, suggesting that osmoregulation in both species was affected. In contrast to general BC pattern, Ca concentration in G. pulex and Mg concentration in B. rhodani increased when pH decreased. Although Ali strongly correlates to pH, the Al composition of soil and bedrock also influences Al availability, potentially contributing to toxic Al; episodes. The estimated values calculated in this study can be used to improve water quality criteria and as thresholds to adjust doses of lime compared to old recommendations in ongoing liming programs. Such adjustments may be critical since both Ali and pH levels have to be balanced to mitigate damage to recovering stream ecosystems.

Although the acid load has decreased throughout Scandinavia, acidic soils still mobilise aluminium (Al) thatis harmful to brown trout. We hypothesise that there are thresholds for Al toxicity and that the toxicity can betraced from the water content to gill accumulation and the consequential physiological effects. During snowmelt,yearlings were exposed to a gradient of pH and inorganic monomeric Al (Ali) in humic streams to studythe toxic effects and mortality. Gill Al and physiological blood analyses [haemoglobin (Hb), plasma chloride(P-Cl) and glucose (Glu)] were measured. As the water quality deteriorated, Al accumulated on the gills; Hband Glu increased; P-Cl decreased, and mortality occurred. Moribund fish had significantly increased gill Aland Hb, suggesting that respiratory disturbances contributed to mortality. Decreased P-Cl and plasmaavailability indicated an ion regulatory disturbance and possibly circulatory collapse. Ali should be lessthan 20 μg/L, and pH higher than 5.0, to sustain healthy brown trout populations. These thresholds can beused to fine-tune lime dose, as both Ali and pH levels have to be balanced to prevent harm in the recoveringaquatic biota. Although Al is tightly linked to pH, local variation in Al availability in soil and bedrock affectsthe Al release and subsequent toxic Ali episodes in some catchment areas.

Aluminium (Al) has been recognised as a main toxic factor alongside pH in acidified water ecosystems. The toxic effect of Al has been attributed to inorganic Al (Al_i), though there are few in situ studies in ambient humic waters which are the focus of this thesis.

The aim was to estimate Al_i toxicity and thus also Al_i concentrations in Swedish humic streams. Subsequently it is necessary to analyse Al_i correctly, which was studied by modelling and method intercalibrations. The hypothesis was that the effect of Al_i could be followed via physiological effects and Al accumulation, as well as by mortality. Toxicity was studied by in stream exposures of brown trout (Salmo trutta L.) and two salmonid prey organisms (Gammarus pulex and Baetis rhodani) during spring flood.

The modelling of the Al_i fraction was performed using monitoring data covering all of Sweden with satisfactory results. The essential variables for Al_i modelling were determined; Al, DOC, pH and F, while Fe, Ca and Mg had less effect. The automated analytical procedure for Al_i (with cation exchange followed by complexation with pyrocatechol violet) was modified and validated and showed to be the preferred method for laboratory analyses.

To avoid detrimental effects for brown trout Al_i should be <20 µg/L and pH >5.0; mortality was high when the Al_i was above 50 µg/L. The invertebrates were more sensitive, as mortalities occurred at pH <6.0 and Ali >15 µg/L for G. pulex, and at pH <5.7 and Al_i >20 µg/L for B. rhodani. It is prudent to use a wide view and let the most sensitive species set the tolerance limits; a pH above 5.7-6.0 and Al_i below 15-20 µg/L allows the stream ecosystems to thrive.

Today, as waters are recovering from acidification, the aim of mitigating liming is to carefully adjust dosage to avoid suboptimal water quality. The thresholds found in this thesis can be used to efficiently but carefully decrease liming, as both Al_i and pH levels have to be balanced to sustain the recovering aquatic biota.

Aluminium solubility and toxicity increase with acidification. There is no standardized analytical method for the determination of inorganic monomeric Al (Al(im)), which is the form that causes toxicity to fish. Separation by cation exchange is commonly combined with other analytical methods, such as complexation with pyrochatechol violet (PCV) or 8-hydroxyquinoline (HQ) and total quantification using graphite furnace or inductively-coupled plasma emission. Data from 14 laboratories were obtained for a dilution series of Al(im) samples; the results of the Al(im) analysis were statistically evaluated. The Al(im) levels were altered through pH, which was controlled by the addition of calcium hydroxide. Confounding parameters such as total organic carbon (TOC) or fluoride (F) were controlled. The total determination and HQ methods yielded significantly higher Al(im) concentrations than the PCV method. Pretreatment by passage through a 0.45 microm filter and pH-adjustment of the ion exchange column had no apparent effect on the Al(im) yield. However, ultra filtration (<10 kDa) caused a significant reduction in the Al(im) concentration using the HQ method. The ultra filtrated Al(im) fraction was similar to the PCV results in the interlaboratory comparison. Retention of colloidal bound Al in the cation exchange column may result in overestimation of Al(im) when the total and HQ methods are used. Estimated Al(im) concentrations derived from two equilibrium models were similar to PCV-derived Al(im) concentrations, as well as the HQ method using ultra filtrated water. The fact that the PCV method does not detect colloidal Al, neither before nor after ion exchange, makes this a preferred technique for Al(im) analysis. Because of the variability in the reported Al(im) concentrations that can arise when different analytical procedures are used, the adoption of a single, reliable technique will facilitate inter-study comparisons and provide consistency in the detection of trends in environmental monitoring programs.

As acidification changes from chronic to more episodic in nature, study focus shifts from lakes, where the basin moderates acidity, to streams where acid episodes occur. To assess the effects of acid stress, several macroinvertebrate indices have been used in Scandinavia and elsewhere, based on the presence or absence of species. Benthic diatoms are strongly linked to water quality with small spatial influence, and display a broad spectrum of species. An acidity index for streams based on benthic diatoms could refine assessment resolution, which is required as the acidification declines. During the autumns of 2004 and 2005, 99 benthic diatom samples were collected from 75 humic streams in Sweden that were monitored by monthly water sampling. The relationships between benthic diatoms and measured environmental variables were explored using canonical correspondence analysis. Acidity was the principal factor influencing the diatoms, with the highest eigenvalues noted for pH, alkalinity, and inorganic aluminium. Several weighted averaging (WA) models were developed to infer stream pH, both using 50 streams from 2004 as a calibration dataset cross-validated with 49 streams from 2005 as an independent dataset, and by modeling all 99 streams. The predictive power of the WA models was good. The model with all streams showed a high correlation for mean pH (r2 = 0.85) and minimum pH (r2 = 0.74) values. An acidity index for diatoms, ACID, was proposed. The correlations to mean and minimum pH values were strong, at r2 = 0.85 and 0.77, respectively. Our results confirmed a strong link between benthic diatoms and water quality, particularly acidity. By employing different approaches we demonstrated that diatoms are excellent acidity indicators. The proposed ACID index can be used to assess the acidity state of a stream. ACID is more convenient to use than a WA model for pH and has equal precision.