There has been little work done to investigate the importance of denitrification in forest soils. This has been caused by difficulties associated with measurement of the denitrification process and from the assumption that nitrification, and hence also denitrification, was insignificant in acid environments. Nitrification can, however, occur even in the acid conditions found in coniferous forest soils, and is especially important after clear-felling when levels of nitrate in soil and drainage waters are often observed to increase. A potential exists, therefore, for gaseous losses of N <i>via</i> denitrification from such soils. This thesis describes the establishment of a suitable method for measurement of denitrification using the acetylene block technique. This method was used to monitor denitrification losses of N, both as N<SUB>2</SUB> and N<SUB>2</SUB>O, from a peaty-gley soil at Kershope Forest. The total loss of N from the standing forest through denitrification was estimated to be 3.2 kg ha<SUP>-1</SUP> over the year studied. Of this loss, approximately 80% was produced as H<SUB>2</SUB>O. Gaseous loss of N through denitrification represents approximately the same order of magnitude as the N lost from the site <i>via</i> leaching. An adjacent site, clear-felled four years previously, was also monitored for denitrification losses. Although this site was denitrifying at only a slightly greater rate than the standing forest, examination of past records from the site revealed that an estimated 9-40 kg N ha<SUP>-1</SUP>y<SUP>-1</SUP> had been lost in the two years immediately after felling. To assess the factors which controlled denitrification in the field, sub-samples of the soils used for denitrification measurements were analysed for water content, extractable nitrate, and available carbon. None of these factors, however, were found to correlate clearly with the denitrification rate. Further experiments, using laboratory amendments of soil cores, indicated that nitrate concentrations had the greatest effect on denitrification rates, although both the availability of carbon and the aeration status of the soil also affected the rates measured.

Two soils from the Fernow Experimental Forest near Parsons, West Virginia were characterized and evaluated in terms of their susceptibility to enhanced soil acidification via acidic atmospheric inputs. After initial physical, chemical, and mineralogical characterization, the soils were analyzed for their ability to adsorb and retain sulfate (SO₄²⁻) through a series of batch adsorption and desorption experiments. Reconstructed soil profiles were then subjected to water leaching as the preliminary step to a base release study in which each soil will be placed under simulated acid rain and evaluated for base cation release and subsequent changes in soil chemistry. Experiments with SO₄²⁻ adsorption and desorption divide the soils into two categories: (i) shallow surface horizons with high organic matter and little Fe oxide content which had little ability to adsorb additional SO₄²⁻ and low capacity to retain SO₄²⁻; and (ii) deeper subsurface soils with low organic matter and high Fe oxide content which could adsorb SO₄²⁻ at solution concentrations above 0.5 mmol/L. All soil horizons desorbed SO₄²⁻ and had no ability for additional adsorption at solution concentrations below 0.5 mmol/L, which implies that the site may be saturated with respect to natural conditions. Initial mass (IM) and Langmiur modeling were used to illustrate SO₄²⁻ dynamics and make correlations with soil physical, chemical, and mineralogical properties. Fe oxides and Al-saturated organic matter were shown to be the preferential sites for SO₄²⁻ adsorption but may be already saturated or very near saturation. Preliminary results from the base release study indicate that the two soils are dominated by different chemical processes and hence release ions into soil solution at different rates. Ion release is shown to be a function of both reactions on the exchange complex and the dissolution of mineral components. / Master of Science

<p>Soil from local beech-forest, spruce-forest and an industrial area was taken. Control- and test-microcosms containing 150 ml soil were spiked with cypermethrin 0,4 mg/ml soil. Cypermethrin residues were extracted on day seven and 14.</p><p>Cyclohexane and deionized water was utilized in multiple step extraction processes. Samples were analyzed in a Gas Chromatograph (GC) with electron capture detector (ECD). </p><p>Concentration values for the samples were highest for beech-forest soil and lower for the other two soil-types. Statistical differences in concentrations between control- and test-microcosms for each soil-type on day seven and day 14 were evaluated with Mann Whitney U tests. Significant result was only found in the industrial 14-day group. The small amounts of cypermethrin in the extracts could not only be ascribed to a bacterial degradation process. Used insecticide has a high bonding affinity for particles and is sequestered in soil.</p>

Soil from local beech-forest, spruce-forest and an industrial area was taken. Control- and test-microcosms containing 150 ml soil were spiked with cypermethrin 0,4 mg/ml soil. Cypermethrin residues were extracted on day seven and 14. Cyclohexane and deionized water was utilized in multiple step extraction processes. Samples were analyzed in a Gas Chromatograph (GC) with electron capture detector (ECD). Concentration values for the samples were highest for beech-forest soil and lower for the other two soil-types. Statistical differences in concentrations between control- and test-microcosms for each soil-type on day seven and day 14 were evaluated with Mann Whitney U tests. Significant result was only found in the industrial 14-day group. The small amounts of cypermethrin in the extracts could not only be ascribed to a bacterial degradation process. Used insecticide has a high bonding affinity for particles and is sequestered in soil.

During the Fukushima Daiichi power plant incident sea water was used in an attempt to cool reactor Unit 3. Since sea water contains an excessive amount of chloride, 36Cl has likely been formed and spread in the environment. Because of the long residence time and the presumed high mobility in water there is an increased interest to learn more about the biogeochemical cycle of chlorine from a radiation risk assessment perspective. Chlorine occurs in inorganic form as chloride (Clin) or bound to organic matter as organic chlorine (Clorg) and is commonly found in the environment due to both anthropogenic and natural processes. Though there are still uncertainties regarding all of the components of the chlorine cycle in soil, the chlorination of organic matter has been exemplified by research. The reverse process, Clorg mineralizing into Clin, has however not been thoroughly investigated. For this study the objective was to observe at what rate Clorg mineralizes into Clin, this by using 36Cl as a tracer in forest soil. 36Cl was added to the soil and 36Clorg was formed over a period of approximately 100 days. After chlorination the samples were incubated in different conditions and the amount of 36Clorg was observed over a period of time (180 days). The result showed no evident dechlorination during the experiment period which indicates that Clorg can be stable in the organic horizon in forest soil.

In forest ecosystems, substantial part of carbon enters soil in the form of plant litter. The decomposition of litter and soil organic matter represents an important process affecting nutrient cycling and carbon balance in soils. Fungi are considered the primary decomposers in terrestrial ecosystems due to the production of wide range of extracellular enzymes that allow them to attack the lignocellulose matrix in litter. Even if fungi represent key players in organic matter decomposition, the information about the structure and diversity of their communities is still limited and the roles of individual fungal taxa in forest soils remain unclear. This Ph.D. thesis focused on the characterization of fungal communities in forest soils and their potential to decompose plant litter. The method for in-depth analysis of complex microbial communities from environmental samples was established and used. In addition, single eukaryotic functional gene was analysed in soil for the first time at a depth that allowed reliable estimation of diversity. It was demonstrated that microbial community composition differs among horizons of forest soil profile. Despite similar diversity, significant differences in microbial community composition were observed between the DNA and RNA. Several microbial groups highly...

Dead plant biomass is a key pool of carbon in terrestrial ecosystems. Its decomposition in soil environments is thus an essential process of the carbon cycle. Fungi are considered to be the primary decomposers in soil ecosystems because of their physiological adaptations and enzymatic apparatus composed from highly effective oxidative and hydrolytic enzymes. Many recent works show that in addition to fungi, bacteria may also play a significant role in lignocellulose decomposition and among bacteria, the members of the phylum Actinobacteria are often regarded to significantly contribute to cellulose and lignocellulose decomposition. This thesis is focused on the evaluation of the role that fungi and Actinobacteria play in dead plant biomass degradation. First, it explored mechanisms involved in degradation, in particular the enzymatic breakdown of major lignocellulose components as cellulose, hemicelluloses and lignin. Enzymatic apparatus of the saprotrophic fungus Fomes fomentarius was explored both in vitro as well as in vivo. Several Actinobacteria were isolated from soil and comparative experiments, investigating production of hydrolytic enzymes, were carried out to track the transformation of polysaccharides and lignin by these strains. To explain the roles of lignocellulose decomposers in...

Paralelamente à aceitação dos reservatórios de carbono florestais para mitigar o impacto do dióxido de carbono emitido à atmosfera, gera-se uma demanda de técnicas que permitam mensurar e monitorar o carbono das florestas nos projetos de desenvolvimento limpo. Os erros que provém da amostragem, são muito maiores que os associados ao processo de análise de amostras, assim, é importante desenvolver planos de amostragem adequados que permitam alcançar a precisão desejada e sem viés na informação. Desta maneira, o objetivo do estudo foi conhecer o comportamento do estoque de carbono nos solos, na fazenda Três Lagoas, localizada no Município de Angatuba, São Paulo, Brasil, área com diferente cobertura e uso de solo: Eucalipto, Pastagem e Floresta Nativa, e conhecer como o sistema de amostragem influencia as estimativas assim como o esforço amostral necessário para a determinação do estoque de carbono. Coletaram-se 406 unidades amostrais distribuídas nas profundidades 0-10 cm e 10-20 cm e por meio da estatística clássica determinou-se que o estoque de carbono é de 39,6 MgC.ha-1. O menor esforço amostral necessário foi determinado na amostragem sistemática estratificada, com uma distância de amostragem na grade de 979 m. A geoestatística foi uma ferramenta que permitiu conhecer o alcance prático da variável (650 m). Pelo valor da diferença do critério de Akaike, os modelos: exponencial, Matérn com kappa 1, kappa 2 e esférico foram iguais. Difere entre eles o alcance prático que cada modelo determinou. O modelo esférico apresentou o maior alcance prático e o menor esforço amostral comparado aos outros modelos. Conclui-se que existem diferenças nos sistemas de amostragem empregados, os quais dependem das variáveis de estudo, da profundidade e do uso do solo. Tanto a estatística clássica quanto a geoestatística mostraram ser ferramentas úteis na predição do estoque de carbono em solos. / As the forestry carbon pools are accepted to mitigate the impact of carbon dioxide, a technical demand appears looking to measure and monitor the forest carbon inside clean development projects. The errors coming from sampling are bigger than those associated to the sample analysis process; therefore, its important to develop proper sampling plans that allow reaching accuracy without information bias. This way, the study developed inside Tres Lagoas Farm, in Angatuba, Sao Paulo, Brazil, aims to know the soil carbon stock behavior inside an area with different soil usage and occupation: eucalyptus, pasture and native forest, to know how the sampling system influences the estimative and the sampling efforts needed to carbon stock estimation. Using classical statistics was found that the carbon stock is 39,6 MgC.ha-1 coming from 406 sample units in a 0-10 cm and 10-20 cm depth. The less sampling effort needed was established by a stratified systematic sampling, with a sampling distance of 979m. Geostatistics was an important tool that allows knowing the variable practical range, (650 m). The models: exponential, Matérn with kappa 1, kappa 2 and spherical were the same by the Akaike criterion difference value. Among them, the difference is the practical range that each model showed. The spherical model showed the higher practical range and the less sampling effort compared to the other models used in this study. Concluding, there are differences among the used sampling systems, because they depend on the study, depth and soil usage variables. Classical statistics and geostatistics seemed to be useful tools on the prediction of soil carbon stock.

Lead has a more than three-millennia-long pollution history in Europe. Metal production, burning of coal and use of leaded petrol resulted in a significant pollution of the atmosphere. As a consequence of atmospheric fallout, the Swedish boreal forest is strongly contaminated by airborne lead pollution. High levels of lead in the soil and soil pore water are of concern because the soil fauna, plants and aquatic biota may respond negatively to this toxic element. The fate of the accumulated pool of pollution in the soil is not well known. In this thesis, I determine four variables of importance for increasing our knowledge about the past, present and future lead levels in the boreal forest soil: 1) the pre-pollution atmospheric deposition rate of lead; 2) the upward flux of lead from deeper soil layers to near-surface horizons as a result of plant uptake; 3) the mean residence time of lead in the mor layer (the organic horizon at the surface of forest soils); and 4) the vertical and lateral transport of pollution lead within the mineral soil and to streams. Lead concentration measurements and stable lead isotope analyses were used for distinguishing pollution lead from natural lead in cores from ombrotrophic bogs, forest soil profiles, forest mosses, soil-water and stream-water samples. The results clearly stress that the boreal forest ecosystem is totally dominated by pollution lead. This is proved by low 206Pb/207Pb ratios (mainly between 1.14 and 1.20) in the mor layer, forest plants and stream water, while the local geogenic lead of the mineral soil (C-horizon) has high ratios (&gt; 1.30). The dominance of pollution in the mor layer is caused by high deposition rates of airborne lead pollution, minute transport rates of lead from the mineral soil by forest plants (about 0.02 mg lead m-2 year-1) and a long mean residence time of the deposited lead (~250 years for mature forest). In the pristine pre-pollution environment, lead was a rare element due to low atmospheric deposition rates (0.001 to 0.01 mg m-2 year-1). It is estimated that the present lead inventory in the mor layer is up to 100 times higher than in the pristine environment where ≤ 8 mg m-2 was present in the mor. The levels in this biologically important horizon will decrease at a very slow rate and it will take centuries for the deeper part of the mor layer to fully respond to decreasing atmospheric inputs. In a hypothetic scenario with a ceased atmospheric lead deposition, the pool of pollution lead will ultimately be redistributed to deeper water-saturated soil layers from where a lateral transport to surface waters occurs. In the studied catchment, the export of pollution lead from the soil to the stream is estimated to peak slightly about one thousand years from now.

Paralelamente à aceitação dos reservatórios de carbono florestais para mitigar o impacto do dióxido de carbono emitido à atmosfera, gera-se uma demanda de técnicas que permitam mensurar e monitorar o carbono das florestas nos projetos de desenvolvimento limpo. Os erros que provém da amostragem, são muito maiores que os associados ao processo de análise de amostras, assim, é importante desenvolver planos de amostragem adequados que permitam alcançar a precisão desejada e sem viés na informação. Desta maneira, o objetivo do estudo foi conhecer o comportamento do estoque de carbono nos solos, na fazenda Três Lagoas, localizada no Município de Angatuba, São Paulo, Brasil, área com diferente cobertura e uso de solo: Eucalipto, Pastagem e Floresta Nativa, e conhecer como o sistema de amostragem influencia as estimativas assim como o esforço amostral necessário para a determinação do estoque de carbono. Coletaram-se 406 unidades amostrais distribuídas nas profundidades 0-10 cm e 10-20 cm e por meio da estatística clássica determinou-se que o estoque de carbono é de 39,6 MgC.ha-1. O menor esforço amostral necessário foi determinado na amostragem sistemática estratificada, com uma distância de amostragem na grade de 979 m. A geoestatística foi uma ferramenta que permitiu conhecer o alcance prático da variável (650 m). Pelo valor da diferença do critério de Akaike, os modelos: exponencial, Matérn com kappa 1, kappa 2 e esférico foram iguais. Difere entre eles o alcance prático que cada modelo determinou. O modelo esférico apresentou o maior alcance prático e o menor esforço amostral comparado aos outros modelos. Conclui-se que existem diferenças nos sistemas de amostragem empregados, os quais dependem das variáveis de estudo, da profundidade e do uso do solo. Tanto a estatística clássica quanto a geoestatística mostraram ser ferramentas úteis na predição do estoque de carbono em solos. / As the forestry carbon pools are accepted to mitigate the impact of carbon dioxide, a technical demand appears looking to measure and monitor the forest carbon inside clean development projects. The errors coming from sampling are bigger than those associated to the sample analysis process; therefore, its important to develop proper sampling plans that allow reaching accuracy without information bias. This way, the study developed inside Tres Lagoas Farm, in Angatuba, Sao Paulo, Brazil, aims to know the soil carbon stock behavior inside an area with different soil usage and occupation: eucalyptus, pasture and native forest, to know how the sampling system influences the estimative and the sampling efforts needed to carbon stock estimation. Using classical statistics was found that the carbon stock is 39,6 MgC.ha-1 coming from 406 sample units in a 0-10 cm and 10-20 cm depth. The less sampling effort needed was established by a stratified systematic sampling, with a sampling distance of 979m. Geostatistics was an important tool that allows knowing the variable practical range, (650 m). The models: exponential, Matérn with kappa 1, kappa 2 and spherical were the same by the Akaike criterion difference value. Among them, the difference is the practical range that each model showed. The spherical model showed the higher practical range and the less sampling effort compared to the other models used in this study. Concluding, there are differences among the used sampling systems, because they depend on the study, depth and soil usage variables. Classical statistics and geostatistics seemed to be useful tools on the prediction of soil carbon stock.