Acid mist and ozone effects on the leaf chemistry of two western conifer species

Seedlings of Jeffrey pine (Pinus jeffreyi) and giant sequoia (Sequoiadendron gigantea) were more susceptible to leaf chemical changes following exposure to acid mist (pH 3.4-2.0) or acid mist/ozone combinations, than to ozone alone (0.1-0.2 microl/litre), when plants were exposed to alternating doses of these pollutants for 6-9 weeks. Under acid mist treatment, leaves exhibited higher levels of nitrogen and sulfur, two elements applied in acid mist. In addition, levels of foliar sodium, and, in the case of giant sequia, potassium, as well, increased under acid mist treatment. Iron and manganese were also mobilized, resulting in significant increases in these elements in pine, and decreases in manganese in giant sequoia foliage. The acid treatment also reduced chlorophyll b concentrations in pine, and, to a less significant extent, in giant sequoia. Calcium, magnesium, barium and strontium were differentially accumulated in giant sequoia compared to Jeffrey pine. Under acid mist treatment, all of these elements (except strontium) declined in concentration in giant sequoia, with calcium showing the most significant trend. The more extensive changes in leaf chemistry induced by acid mist are consistent with earlier observations of significant changes in spectral reflectance of these seedlings after 3 weeks of fumigation. Limited foliage samples collected from these two species in 1985 and 1986 in Sequoia/Kings Canyon National Parks in the southern Sierra Nevada do not in themselves indicate any clearcut or severe effects of ozone alone on leaf chemistry of these species, but a mild influence of nitrate-laden acid deposition, possibly in combination with ozone, is consistent with the rise in nitrogen and lignin levels in Jeffrey pine on sites observed to have moderate visible injury symptoms. No firm conclusions about effects of pollutants on leaf chemistry in these field sites is possible without further study.     

Concentrations of two organic contaminants in precipitation, soils and plants in the Essex region of Southern Ontario

Recent research has indicated that the atmosphere is an important pathway by which pollutants enter terrestrial and aquatic ecosystems. We report here concentrations of PCBs and octachlorostyrene (OCS) in precipitation, soils and plants in Essex County, Ontario. The average PCB concentration in urban precipitation (23 ng litre(-1)) was lower than that previously reported for urban areas in the Great Lakes basin. Differences between sites and with varying wind directions were not significant. OCS concentrations in precipitation averaged 1.6 ng litre(-1). Concentrations of PCBs in soils were 2-3 orders of magnitude greater than in precipitation. Concentrations of these pollutants in city soils and plant roots were consistently higher than those from suburban and rural sites. Ratios of urban to suburban concentrations in soils and precipitation were approximately 5:1 for PCBs. However, concentrations of OCS were similar in urban and suburban samples of precipitation, soils and plant tissues. These comparisons suggest an urban source for PCBs, but not OCS. Concentrations of all contaminants in plant leaves, unlike those in precipitation, roots and soils, were relatively similar in urban and suburban areas. That similarity suggests that direct foliar uptake is not an important pathway for pollutant uptake in plants.     

Chemistry of soil solutions under different kinds of vegetation in the vicinity of a thermal power station

The influence of atmospheric deposition on the chemical characteristics of soil solutions in a small catchment area in NW Spain was studied. The soils, developed from slates, were sampled from seven sites supporting different forms of vegetation (deciduous and pine forest and heath). Soil solutions were extracted, by the column displacement method, from soil samples collected monthly from March 1992 until November 1993. The solutions were acidic with a low content of basic cations. The most common ions in all horizons were Cl(-) and Na(+), due to marine influence. In the surface horizons (0-10 cm), relatively high concentrations of SO(2-)4 (150-380 micromol litre(-1)) and Zn (approximately 2 micromol litre(-1)) were obtained, with good correlation between the two ions. These results, along with the prevalence of inorganic forms of Al (50-90% of total Al), were related to the effects of acidic deposition in the catchment area. The more rapid breakdown of litter in the soils under deciduous forest explains the greater ionic concentrations obtained in these solutions.     

The relationship between hydrogen and sulphate ions in precipitation-A numerical analysis of rain and snowfall chemistry

Acidic rain has been identified as potentially harmful to the aquatic and terrestrial components of the ecosystem. Sulphate measured in rain and snow has been used as a surrogate indicator of acidic deposition. If sulphur dioxide controls are the means to limit acidic deposition, then the association between sulphate and hydrogen ion concentrations in precipitation is an important factor in establishing such limits. Selected data on rain and snowfall chemistry from the National Atmospheric Deposition Program (NADP), the Electric Power Research Institute's SURE, the utility industries' UAPSP, and the Department of Energy's MAP3S were reviewed. Numerical analyses were performed to assess the relationship between hydrogen and sulphate ion concentrations. The strength of the association between hydrogen and sulphate ions varied from site to site. In the Midwestern and Eastern regions, the Pearson correlation coefficient was over 0.50 while in the Central and Upper Midwestern parts of the United States, the correlation coefficients were less than 0.25. Regardless of the strength of the association between hydrogen and sulphate ions, all but one of the NADP/NTN sites used in our analysis exhibited at least 30% of the anions (sulphate, nitrate, and chloride) associated with cations other than hydrogen. For sites where the strength of the association was weak, between 65% and 98% of the anions appeared to be associated with cations other than hydrogen. Because a large percentage of the anions (i.e. sulphate, nitrate, and chloride) appear to be associated with cations other than hydrogen even at those sites where the association between hydrogen and sulphate ions was strong, the complex chemistry controlling the acidity in precipitation may make it difficult to predict the impact of a reduction in sulphate concentration.     

The impact of aluminium on green algae isolated from two hydrochemically different headwater streams, Bavaria, Germany

Two strains of green algae (Scenedesmus sp. and Chlorella sp.) have been isolated from two headwater streams differing in stream chemistry (Eger river: acidic, low alkalinity; Püttlach river: slightly alkaline, high alkalinity). In this study the growth response of these indigenous algae to increased concentrations of aluminium (Al) is investigated. A semi-continuous culture technique was used for Al-toxicity studies. Algal response was determined by calculating growth rates from turbidity and cell counts. Those Al-species which are well known to be toxic were estimated by equilibrium calculations using the WATEQ computer program (Truesdall & Jones, 1973). The pH-value of the culture media was usually pH=5, except for one of the test series. Tested concentrations ranged from c=4 to 220 micromol litre(-1). The isolated strains of green algae were highly sensitive to increased Al-concentrations. The strain isolated from the Püttlach river was more sensitive than the Eger river algae. A total growth inhibition occurs at Al-concentrations of c=4 micromol litre(-1) if the whole Al was added at once. If Al was added gradually into the growth media the response of the algae was delayed. This is due to Al-enrichment in cells. In our long term toxicity studies, growth inhibition occurs even at nearly neutral pH-conditions (pH=6.5) although Al toxicity is expected at pH-values less than pH=5.5. This new result confirms the need of long-term studies in continuous cultures under simulated natural conditions. This might be the only way to achieve valid conclusions about the fate and the toxicity of environmental pollutants.     

Pollution in the upland environment

As part of a study on the effects of pollutants on forests, a long-term monitoring programme has been operating at two sites (602 and 275 m above sea level (a.s.l.)) in Glentress Forest, south-east Scotland, since June 1987. At these sites, equipment has been recording 20-min data for weather parameters, and cloud and rain event frequency, duration and intensity. The chemical composition of bulk cloud and rainwater has been measured. Cloudwater detection and collection has been made using passive 'Harp wire' gauges (with a cross-sectional area of 0.047 m2) strung with polypropylene filament and having a typical collection efficiency of 29% when compared with independent measurements of windspeed and liquid water content. During 1988, the annual rainfall at the upper site was 1213 mm, occurring over 1776 h. The equivalent cloudwater deposition to a forest with a drag coefficient of 0.06 was estimated to be 375 mm over 1936 h. A 'typical cloud event' lasted 4.5 h and would deposit to forests at a rate of 0.2 mm h(-1). There were significantly higher loadings of suspended particulate material (> 0.2 microm) in cloudwater (mean 18.42, max. 94.5 mg litre(-1)) compared with rainwater (mean 2.6, max. 25.6 mg litre(-1)). There were similar differences in ion concentrations, e.g. for H+ in cloudwater (mean 163, max. 1259 microm) and in rainwater (mean 33, max. 262 microM). It is concluded that cloudwater deposition represents a major pathway for pollutant transfer to the upland environment, especially where the surface vegetation is efficient at capturing cloudwater, i.e. forests. The consequences of this increased pollutant loading to forests and water catchments are yet to be assessed.     

Response of two cultivars of Triticum aestivum L. to simulated acid rain

The present experiment was aimed at assessing the impact of simulated acidic precipitation (SAR) on growth, biomass accumulation and yield of two cultivars of wheat (Triticum aestivum L.), Malviya 206 and 234, varying in cuticular thickness and leaf area. Wheat cultivars were exposed to simulated rain acidified to pH 5.6 (control), 5.0, 4.5, 4.0 and 3.0 from 30 days of age, twice a week for five weeks. The plants received ambient precipitation of unknown acidity, as well as the acid rain treatments. Growth parameters such as shoot height, root length, and leaf area were reduced significantly in treated plants at different growth stages. Above and below-ground biomass also decreased significantly in the plants treated with acidic precipitation. Relative to control, the number of grains per plant and yield per m(2) declined significantly at all SAR treatments. The hypothesis that the variety with thinner cuticle and greater leaf area would be more susceptible to acidic precipitation was not supported by the present study.     

Spatial patterns concentrations in upland Wales in relation to catchment forest cover and forest age

Data on nitrate nitrogen were collected weekly during 1984 from 136 sites on streams in upland Wales. Mean nitrate concentrations in summer (0.02 to 1.5 mg litre(-1)) were significantly lower (P<0.001) than in winter (0.02 to 1.26 mg litre(-1)), particularly at sites with mature conifers (>30 years old). Mean concentrations increased significantly with the average age of conifers on each catchment (P<0.001), and with increasing areal cover by trees over 30 years old (P<0.001). Nitrate concentrations increased significantly with stream total hardness (P<0.001), possibly reflecting nitrogen mineralisation in soils of higher base status. Concentrations also increased with stream chloride (P<0.001), which is predominantly atmospherically derived, implying that increased nitrate occurred where general atmospheric inputs of solutes were increased. After accounting for variation in hardness, residual nitrate concentrations still increased with the average age of the conifers (P<0.001), and with catchment cover by mature trees (P<0.001). We infer that some additional nitrate under older conifers is thus independent of catchment sources associated with increasing hardness. Two possibilities are increased inputs and decreased retention of nitrogen within the ecosystem of maturing conifer forest. Residual nitrate after accounting for variations in chloride also increased significantly with conifer age (P<0.01) and cover (P<0.01), a pattern implying that some sources of nitrate may also be independent of increased sea-salt deposition. We allude to the possibility that additional nitrogen deposition adds to nitrogen throughputs from maturing forests, and we discuss the potential ecological role of additional nitrogen in runoff.     

Wetfall Deposition and Precipitation Chemistry for a Central Appalachian Forest

Abstract

Although extensive research on acidic deposition has been directed toward spruce-fir forests, less research has been done on the impacts of air pollution on eastern montane hardwood forests. The purpose of this study was to describe precipitation chemistry for several Appalachian hardwood forest sites at or near the Fernow Experimental Forest (FEF) to assess the potential for problems associated with acidic deposition. Emphasis was placed on seasonal patterns of ionic concentrations (H+, Ca++, NH4+; NO3-, and SO4=) and spatial variability of ionic concentrations and deposition among sites. Seasonal patterns of most ions showed highest concentrations during the summer months and deposition of H+ was especially pronounced during this time. Deposition of all ions was generally greater (related to greater precipitation) at three montane forested sample sites compared to a nonforested riverbottom site. Precipitation chemistry at FEF was similar to other sites throughout the eastern United States and contrasted sharply with mid-western and western sites. Eastern sites, including means for FEF sites, were uniformly 3-4 times higher in H+ and SO4= concentration than the mid-western and western sites. Precipitation at FEF was chronically acidic, more so during the growing season, and highest at higher elevations where environmental stresses can be most severe. Furthermore, there were occasional large discrepancies between the low-elevation site and the higher-elevation forested sites for precipitation chemistry and acidic deposition. These results suggest that synoptic-scale (network) data may greatly underestimate the pollutant conditions to which highelevation forest trees are exposed, since network data rarely take elevation into account and typically are based on annual ionic concentration and deposition means that may be considerably lower than those of the growing season.     

Chemical climatology of high elevation spruce-fir forests in the southern Appalachian mountains

The physical and chemical climatology of high elevation (> 1500 m) spruce-fir forests in the southern Appalachian mountains was studied by establishing a weather and atmospheric chemical observatory at Mt Mitchell State Park in North Carolina (35 degrees 44' 05" N, 82 degrees 17' 15"W). Data collected during the summer and autumn (May-October) of 1986, 1987, and 1988 are reported. All measurements were made on or near a 16.5 m walk-up tower extending 10 m above the forest canopy on Mt Gibbes (2006 m msl), which is located approximately 2 km SW of Mt Mitchell. The tower was equipped with standard meteorological instrumentation, a passive cloud water collector, and gas pollutant sensors for O3, SO2, NOx. The tower and nearby forest canopy were immersed in clouds 25 to 40% of the time. Non-precipitating clouds were very acidic (pH 2.5-4.5). Precipitating clouds were less acidic (pH 3.5-5.5). The dominant wind directions were WNW and ESE. Clouds from the most common wind direction (WNW) were more acidic (mean pH 3.5) than those from the next most common wind direction (ESE, mean pH 5.5). Cloud water acidity was related to the concentration of SO4(2-), and NO3- ions. Mean concentration of H+, NH4+, SO4(2-), and NO3- ions in the cloud water varied from 330-340, 150-200, 190-200 and 120-140 micromol litre(-1) respectively. The average and range of O3 were 50 (25-100) ppbv (109) in 1986, 51 (26-102) ppbv in 1987, and 66 (30-140) during the 1988 field seasons, respectively. The daily maximum, 1-h average, and 24-h average concentrations were all greatest during June through mid-August, suggesting a correlation with the seasonal temperature and solar intensity. Throughfall collectors near the tower were used to obtain a useful estimate of deposition to the forest canopy. Between 50-60% of the total deposition of SO4(2-) was due to cloud impact.     

Changes in sulphate retention, soil chemistry and drainage water quality along an upland soil transect

Soils sampled along an altitudinal transect in an upland area of North East Scotland have been used to investigate downslope changes in the capacity of soils to retain sulphate. Simulated laboratory experiments involving the leaching of reconstituted cores with 'rainfall' containing low (1.85 mg litre(-1) and high (51.90 mg litre(-1) concentrations of sulphate indicate that soils developed on upper slopes have a limited capacity to adsorb sulphate, whereas soils on lower slopes have a much greater sulphate adsorption capacity. Soil drainage water, produced from 'sensitive' upper slope soils may therefore be significantly modified by physico-chemical reactions in lower slopes before reaching watercourses.     

Effects of atmospheric ammonia and ammonium sulphate on Douglas fir (Pseudotsuga menziesii)

Three-year-old Douglas firs (Pseudotsuga menziesii) were fumigated with 180 microg m(-3) NH3 or clean (charcoal-filtered) air. During these fumigations the plants received 15 mm artificial rain weekly, supplemented with 20, 500 or 2500 micromol litre(-1) (NH4)2SO4. Exposure to NH3 and NH4+ for 14 weeks resulted in a change of the nutrient status of the needles. The most remarkable effect was the increase in the N/K ratio, due to both uptake of N and leaching of K. The action of NH3 was stronger than that of NH4+. Both NH3 and (NH4)2SO4 affected the epicuticular wax layer and decreased mycorhiza frequency. Following fumigation and artificial rain treatments, needles were incubated for 8 h in a medium containing 0, 50, 250, 500 and 2500 micromol litre(-1) (NH4)2SO4. Almost no exchange of Ca, Mg and K for NH4+ was found. Therefore this ion exchange probably explains only a minor part of the changes in nutrient status of the whole trees.     

Photochemical oxidant pollution and vegetation: effects of mixtures of gases, fog and particles

Photochemical 'smog' contains mixtures of gases (e.g. ozone, nitrogen dioxide), and dry particles (e.g. nitrates). Intermittent fog in the same geographical area can be acidic with high concentrations of nitric acid. Results from recent field studies in the Los Angeles Basin have emphasized the relative toxicity of these components of photochemical air pollution. Studies have focused on gaseous+fog or gaseous+dry particulate effects on conifers, gaseous+fog effects on crops, and the effects of trace pollutants produced during generation of ozone on crops. Data from these studies indicate that direct alterations in growth and physiological responses were observed only with gaseous pollutants (primarily ozone), or repeated applications of highly acidic fogs (pH < 2.7). Direct particle dry deposition effects are unclear. Few interactions have been found between gaseous pollutants and acidic fog. Charcoal-filtered open-top chambers are highly effective in removing pollutants in the following order: fog (100%) > peroxyacetyl nitrate > ozone > nitrogen dioxide > sulfur dioxide > nitrate ion > ammonium ion > sulfate ion. However, nitric oxide concentrations are higher in charcoal-filtered chambers than in ambient air. The studies point out the importance of considering other components of photochemical pollution in addition to ozone, especially when investigating subtle, long-term effects on vegetation.     

Sulphite and sulphate concentrations in weathering products of sandy limestone and in deposition samples

The sulphite and sulphate concentrations in weathering products of limestone and in wet and total deposition samples were measured by the modified West-Gaeke method and by ion chromatography. The sulphite content in the weathering crust and in the runoff rainwater of two historical buildings in Belgium, was always much lower than the sulphate content. The maximum sulphite to sulphate ratio was 2.5%. The sulphite concentration in wet and total deposition samples was in the low or sub ppm range except during two misty periods when the sulphite concentration was 6.8 and 21 mg litre(-1). In those samples a high sulphate concentration was also found (nl. 57 and 137 mg litre(-1), respectively).     

Reproductive status, blood chemistry, gill histology and growth of perch (Perca fluviatilis) in three acidic lakes

Perch (Perca fluviatilis L.) were sampled soon after spawning in three small acidic lakes (pH 4.3-6.1, Al(lab) 5-106 microg litre(-1), Ca2+ 0.01-0.08 mmol litre(-1)) and in one circumneutral lake (pH 5.9-6.4, Al(lab) 4-12 microg litre(-1), Ca2+ 0.06-0.07 mmol litre(-1)) in southern Finland. Due to the delayed spawning of perch in the acidic lakes, sampling in those lakes was performed later than in the reference lake. In spite of that, the gonadosomatic index (GSI) of males in all the acidic lakes was significantly greater than in the reference lake. Of the two lakes with similar low water pH, the effects on reproduction were more prominent in the lake with higher water Al content. The plasma Ca2+ concentrations of females in the acidic lakes were significantly smaller than in the females of the reference lake. The low female:male plasma Ca2+ ratio (1.0-1.32) depicted delay of spawning. Stress in perch in acidic water was also seen in elevated blood haematocrit values, especially in females. On the other hand, a low plasma Cl- level, a common response to acidic water in salmonids, was not detected in perch in the most acidic lakes. The amount of Al accumulated in the gill epithelium was highest in the most acidified lake with high Al concentration, but was also pronounced in a lake with low pH and low Al concentration.     

Seasonal changes in the tissue-metal (Cd, Zn and Pb) concentrations in two ecophysiologically dissimilar earthworm species: pollution-monitoring implications

During two consecutive growing seasons, the same potted individuals of European aspen (Populus tremula), grown from root cuttings of one clone, were fumigated with either ambient air or ozone concentrations of 0 (control), 0.05 or 0.1 microlitre litre(-1). Structure and biomass of the annually formed branches were analysed after excision at the end of each season. Only at 0.1 microlitre litre(-1) was branch weight reduced, and crooked axes occurred in each season. During the second season, branch length and leaf sizes were strongly reduced, while many leaves displayed yellowish deficiency symptoms and lowered cation concentrations. Such leaves contrasted to those showing characteristic O3-bronzing. Although foliage density was enhanced due to reduced branch length, the area of attached foliage was limited by the small leaf sizes, necrotic leaves and premature leaf loss. During mid-summer of the second fumigation period, photosynthetic capacity, carboxylation efficiency and water-use efficiency (WUE) declined in (attached) yellowish and bronze leaves at 0.1 microlitre litre(-1), whereas green leaves at 0.05 microlitre litre(-1) displayed accelerated senescence in late summer while maintaining WUE. It is concluded that the differences in branch growth between the two growing seasons were caused in part by internal changes in those plant organs (root and basal stem), which had experienced both fumigation periods.     

Abundance and population structure of perch (Perca fluviatilis L.) in some acidic Norwegian lakes

Abundance (catch per unit effort, n=3752) and population structure of perch Perca fluviatilis were studied in 30 acidic Norwegian lakes with relation to pH (4.3-5.9), calcium (0.41-2.44 mg litre(-1)), labile aluminium (24-255 microg litre(-1)) and total organic carbon (TOC, 1.7-13.8 mg C litre(-1)). Standard series of bottom gill-nets were used to sample fish populations. A multiple regression analysis showed that abundance was significantly correlated to TOC (P<0.05) and, to a smaller extent, to the Ca in lakes with concentrations between 0.41 and 1.70 mg litre(-1) (P=0.07). The lakes which were inhabited by pike were excluded from the analysis. Recruitment failure seems to be the main cause of reductions in perch numbers in the lakes studied. However, high mortality among adult individuals was also evident, and an episode of fish kill was observed in one of the lakes.     

Response of Quercus pubescens leaves exposed to geothermal pollutant input in southern Tuscany (Italy)

The paper reports a case of evident and widespread leaf damage on trees in southern Tuscany (Central Italy) attributed to the input of pollutants produced in a geothermal area. The main potentially phytotoxic substances are boron and hydrogen sulphide. Trees affected are conifers as well as both evergreen and deciduous broadleaves. In the present study the possible impact of geothermal pollutants on Quercus pubescens leaves has been considered. Leaf samples coming from three sampling locations (S1 inside the geothermal area; S2 on the margins; S3 outside) and three consecutive dates (June, July and August) were analyzed for the following parameters: sulphur and boron concentration; leaf area; leaf mass per area; chlorophyll fluorescence (Fv/Fm); chlorophyll a, chlorophyll b and carotenoid concentrations. Anatomical and ultrastructural observations were also performed. In all sampling location sulphur and boron concentrations are greater than the background values recorded in southern Tuscany in a previous survey. The sulphur concentration in leaves was higher in S1 than S2 and S3, but did not increase throughout the survey period. Boron reached the greatest concentrations in S2 and showed a continuous increase over the study period. Leaves subjected to a higher load of pollutants were smaller in size (in terms of leaf area), but were more sclerophyllous. Damaged chloroplasts and reduced Fv/Fm values were observed at S1 and S2, but chlorophyll concentration values were higher at S1. Such an apparent anomaly can possibly be explained by the onset of compensation and recovery mechanisms. Foliar injuries appeared to be related to boron concentration.     

Sensitivity of modelled sulphate and nitrate aerosol to cloud,pH and ammonia emissions

A Lagrangian dispersion model has been used to predict daily sulphate aerosol in 2006 at five UK rural measurement sites and hourly nitrate aerosol in April 2003 at Harwell (UK). The sensitivity of aqueous phase sulphate production to the meteorological input has been investigated. Large differences were found between cloud fraction and cloud liquid water output from the regional and mesoscale Met Office Unified Model. The impact on the sulphate aerosol was relatively small, with the mesoscale meteorology giving better results.Sulphate aerosol production in the aqueous phase was found to be very sensitive to modelled cloud pH. As the cloud becomes acidic sulphate production is greatly limited, conversely if the cloud is basic large amounts of sulphate aerosol are produced. A fixed model pH of 5.8 was found to produce better results than allowing the model to calculate pH which resulted in large over-predictions of measured sulphate aerosol in some episodes.Nitrate aerosol was not sensitive to cloud variables or pH, but showed a slight increase with 30% more ammonia emissions, and a slight decrease with 30% less ammonia.Sulphate production in model runs with fixed pH was not sensitive to changing ammonia emissions, however the sulphate production with modelled pH was very sensitive to plus or minus 30% ammonia. This work suggests that good modelling of ammonia is essential to correct estimation of aqueous phase sulphate aerosol if cloud pH is modelled. It is concluded that modelling to estimate the effect of reduced ammonia emission scenarios on future ambient aerosol levels should also take into account the neutralising effect of ammonia in cloud and hence the effect on aqueous phase production of sulphate.     

Chemical and biological effects of acid, aluminium and lime additions to a Welsh Hill-stream

Chemical and biological responses to simultaneous additions of acid, aluminium and lime were investigated in contiguous 250m-reaches of a chronically acidic stream in Wales. Treatments were applied for 24 h, and from the upstream end were as follows: zone A-untreated, pH 5.0, 0.37 mg litre(-1) filterable Al; zone B-acidified to pH 4.5, 0.40 mg Al litre(-1) (47% of Al attributed to release from the stream bed due to acid additions); zone C-acidified to pH 4.5 and Al dosed to 0.67 mg litre(-1); zone D-dosed with limestone slurry, resulting in pH 7.2, 0.13 mg Al litre(-1). In all reaches, the chemistry of the interstitial water at depths of 0.15 and 0.3 m never fell below pH 5.5, with corresponding decreases in Al and increases in base cation concentrations. Brown trout, Salmo trutta, and crayfish, Austropotamobius pallipes, held in the stream showed decreases in plasma [Na(+)] and haemolymph [Na(+)], respectively, in all acidic zones (A, B, C): these responses were mitigated by liming (zone D). Thus both chronic and simulated episodic levels of pH and dissolved Al were sub-lethally toxic to test species of aquatic fauna. This experiment also demonstrates a stream bed source/sink of Al, and the availability of a possible refuge from acidic surface waters within the substratum.