Covered catchment experiment at Gårdsjön: changes in runoff chemistry after four years of experimentally reduced acid deposition

Atmospheric deposition was almost entirely excluded from the forested headwater catchment G1 ROOF, by means of a 7000 m² plastic roof that prevents rain and throughfall from reaching the ground. Under the roof an irrigation system was installed to simulate a natural precipitation regime. The intercepted throughfall was substituted by the same amount of clean pre-industrial throughfall. The experiment started in April 1991. During the four years of the treatment, 2960 mm (18 600 m³) of sprinkled solution was applied under the roof, which is about 15 times the mean water storage of the catchment. After four years of treatment major changes in runoff chemistry were observed. The exclusion of all non-marine sulphate (SO₄ ^2–) input to the catchment (i.e. ca 75 % of total SO₄ ^2– input excluded) resulted in significant decline of sulphate in runoff through all four years of treatment. During the fourth year, annual volume weighted SO₄ ^2– concentration was 46 % lower than the two years prior to the treatment. Concentrations of inorganic aluminium Al³⁺ declined 52 % and Mg²⁺ declined 54 %. No change of H⁺ concentration was detected. As the treatment proceeds there seems to be a trend towards less negative acid neutralising capacity in runoff.     

Sulfate Adsorption-Desorption in a Swedish Forest Soil

The Forest Roof project at Lake Gårdsjön located on the Swedish Southwest coast near Gothenburg was established to assess the response of the ecosystem following a major reduction in loading of sulfur acidic deposition. The objective of the study was to describe the adsorption-desorption processes of sulfate in Haplorthod soils at the Gårdsjön site. A sequential batch technique was used to sequentially leach the upper three mineral horizons with forest floor leachate containing different concentrations of sulfate. Results showed that adsorption-desorption processes were highly irreversible and hysterestic in nature. Moreover, the soil solution SO₄ ^2-, pH and DOC were interrelated. These findings suggest that improvements in the prediction capability of models for acidified surface waters, lakes and soils are needed. Therefore, existing sulfate models with fully reversible Langmuir or Freundlich approaches need to account for additional irreversible reactions.     

Nitrex: The timing of response of coniferous forest ecosystems to experimentally-changed nitrogen deposition

In large regions of Europe and eastern North America atmospheric deposition of inorganic nitrogen (N) compounds has greatly increased the natural external supply to forest ecosystems. This leads to N saturation, in which availability of inorganic N is in excess of biological demand and the ecosystem is unable to retain all incoming N. The large-scale experiments of the NITREX project (NITRogen saturation EXperiments) are designed to provide information regarding the patterns and rates of responses of coniferous forest ecosystems to increases in N deposition and the reversibility and recovery of impacted ecosystems following reductions in N deposition.The timing of ecosystem response generally followed a hypothesized cascade of response. In all sites N outputs have responded markedly but to very different degrees within the first three years of treatment. Within this time significant effects on soil processes and on vegetation have only been detected at two sites. This delayed response is explained by the large capacity of the soil system to buffer the increased N supply by microbial immobilization and adsorption. We believe that this concept provides a framework for the evaluation and prediction of the ecosystem response to environmental change.     

Liming and fertilization of acid forest soil: Short-term effects on runoff from small catchments

Increased atmospheric deposition of strong acids and deposition of potentially acidifying compounds (e.g. ammonium) has caused a decline in pH and exchangeable base cations in forest soils in Sweden. In recent years, attention has been paid to liming of forest soil as a method to counteract the effects of acid deposition. Experiments with liming, fertilization and woodash treatment of acid forest soils started in 1984. The aim of this study was to determine the effects of low doses of lime (500 to 1500 kg ha^–1) in combination with N fertilizers on tree growth, nutritional status of trees as well as soil, and runoff chemistry. This paper describes the short term effects of liming and fertilization on runoff from ten small catchments in two regions in south Sweden. The effects of liming were small in both areas. In the catchments fertilized with N (NH₄NO₃), a substantial leakage of various N species appeared in runoff after treatment. The increased N output was dominated by nitrate. The excess leakage of N during 2 yr after fertilization was 25 and 13% as an average of the applied N in the two study areas. The mobile nitrate increased the base cations output via runoff with 10 to 100% during 1 yr after N treatment. The runoff of Al increased with 60 to 1009, the first year in the fertilized catchments. Mobilization of cations was also influenced by ammonium, especially K that was exchanged by ammonium on the surface of the soil particles. The effects of woodash-treatment were small, however, sulfate in the ash leaked out following application and about 100% of the added sulfate was found in runoff during the first year.     

Effects of nitrogen deposition on the acidification of terrestrial and aquatic ecosystems

The concentration of ammonium and nitrate in precipitation has increased during this century. The deposition of N compounds (wet + dry) is reaching 30 to 40 kg ha^?1yr^?1 in many areas in Central Europe and above 20 kg in the southern parts of Scandinavia. In extreme situations throughfall data indicate depositions above 60 kg ha^?1yr^?1 in Central Europe and above 40 kg ha^?1yr^?1 in south Sweden. Very high depositions are observed on slopes at forest edges and adjacent to areas with animal farms and manure spreading. In areas with low N deposition almost all deposited N (>95%) will be absorbed in the tree canopies or in the soil. In areas with high deposition an increased outflow is observed which in some cases reach 10 to 15 kg ha-lyr-1. The increased output is an indication of N saturation of the ecosystem and it leads to acidification effects in soils, soilwater, groundwater and surface waters.     

Sulphur Balances and Dynamics in Three Forested Catchments in Sweden

The deposition of sulphyr (S) has decreased considerably during the last decade and is now at levels below those recorded in the middle of the 1950s in Sweden. However, the sulphate concentrations in surface waters have not decreased to the extent that could be expected from the decreased S deposition. Catchment studies have shown a net S output, indicating a soil-derived S source besides the atmospheric S input. This study was performed in three small, forested catchments in southern and central Sweden, that are part of the national, integrated monitoring program for forests. The aim was to assess S levels and fluxes in and between the different compartments of the ecosystems in order to trace the possible sources of excess S in runoff. The sulphate concentrations in soil water indicate that the excess S originates from mineralization and oxidation of organically bound S in humus and peat, as well as from desorption of sulphate from Fe and Al oxohydroxides in the B-horizon. The latter process will quickly respond to changes in S deposition, while the former probably is independent of S deposition in a short time perspective. This will have implications for the modelling of recovery from soil and surface water acidification.     

Modeling acidification and recover in the roofed catchment at Lake Gårdsjön

The dynamic, biogeochemical model SAFE was applied to a roofed subcatchment G1 at Gårdsjön, Sweden. The roof was installed in 1991, and deposition of anthropogenic S and N reduced by ca. 90%. Initiated from pre-industrial steady-state conditions, SAFE predicts present levels of biologically relevant chemical properties (pH, inorganic Al and base cations). SAFE overestimates the short-term effects of the manipulation on runoff pH, while the modeled decline in inorganic Al and and base cations are comparable to observations. Temporal variability and too few years of measured data make model to data comparison difficult. Sulfate desorption, which is not included in SAFE, may introduce a time lag between modeled and measured data. Reductions of S and N inputs by 90% will lead to a recovery in pH, low A1 but extremely low base cations concentrations due to replenishment of exchange sites.     

Catalytic effect of various metal ions on the methylation of mercury in the presence of humic substances

Methylation of Hg²⁺ (Hg(NO₃)₂) in the presence of fulvic acid (FA) and various metal ions has been studied. The concentrations of Hg²⁺ and FA ranged from 5 to 20 mg L^?1 and 171 to 285 mg L^?1 DOC, respectively. The pH range was 3 to 6.5. FA was isolated from an acid brown-water lake by XAD-8 polymeric adsorbent. Methylmercury production in the dark during 2 to 4 days incubation at 30 °C increased with increasing concentrations of Hg²⁺ ion and FA as well as with additions of metal ions (5 to 10 × 10^?5 mole L^?1 The observed catalytic activity of metal ions followed the order Fe³⁺ (Fe²⁺) > Cu²⁺ ≈ Mn²⁺, > Al³⁺. The production of methylmercury had a pH-optimum around 4 to 4.5 at the conditions tested.     

From broadleaves to spruce – the borealization of southern Sweden

Mixed or broadleaf forests were once common in many regions of Northern Europe, whereas today conifers often dominate. The aim of the study was to investigate the spatial and temporal patterns and processes which underlie this transition to Picea abies dominance in southern Sweden. We use recently developed paleoecological methods to determine long-term changes in the relative abundance of tree species, and digitalized National Forest Inventory (NFI) databases to assess more recent changes in the spatial coverage of Picea throughout the region. The novel combination of the two databases reveals that Picea became a widespread and abundant species in many parts of southern Sweden around 1000 years ago. After a brief decline in abundance starting around 500 years ago, NFI data indicate a rapid increase in the volume of Picea between 1920 and 1950, prior to the large-scale introduction of Picea-dominated plantation forests. The available evidence suggests that abandonment of forest grazing and slash-and-burn cultivation, as well as selection cutting benefited the natural establishment and growth of Picea during the first half of the twentieth century. Hence, prior to the impacts of industrialized forestry that began in the 1950s, other processes were already favouring increased Picea abundance.     

Are ion exchange processes important in controlling the cation chemistry of soil- and runoff waters ?

Soil- and stream water elemental concentrations from a subcatchment in the Lake Gårdsjön area have been used to evaluate the importance of ion exchange processes on the transport of cations to aquatic ecosystems. The importance of cation exchange processes in the upper organic and upper B soil horizons was demonstrated using lysimeter water data from a recharge area and soil water flow simulated with SOIL model during winter rain events with high sea-salt concentration. The importance of the hydrological conditions, such as water flow and water pathway, silicate weathering and the ion exchange of Al with H⁺ on the streambed materials in controlling cation concentrations in soil and stream waters are also discussed. With the SAFE model, the contribution of cations from ion exchange by depletion of base cations from the exchange matrixes compared to from weathering was also assessed. SAFE calculations indicate that the release rate of base cation by ion exchange to runoff water has decreased since 1945 and is very low, approx. 0.1 keq/ha per year, at present time as a result of soil acidification due to S and N inputs.