No description available.

No description available.

No description available.

No description available.

Two locations, Lake Ivanhoe and the Maitland interchange of Interstate 4, were selected to study the impact of bridge runoff on receiving water bodies. The Lake Ivanhoe location includes two similar bridges, one without scuppers and one with scuppers. The Maitland interchange site has several borr ponds which drain to Lake Lucien. Samples were collected from Lake Ivanhoe below the bridges and in the open lake away from the bridges. Also samples were collected from the east pond, west pond, and lake Lucien, at the Maitland interchange. Samples included water, sediments, plants and benthos to detect differences, if any, in heavy metal concentration due to sampling location. Heavy metals tested included: ZN, Cu, Cr, Cd, Pb, As, Fe, and Ni. Dissolved oxygen, temperature profiles and secchi desk transparency were measured in the field. Additional water quality parameters such as pH, turbidity, carbon and phosphorus were evaluated. Results indicated significant differences in specific heavy metal concentrations exist between samples collected beneath a bridge with scuppers compared to samples collected beneath a bridge without scuppers. Also, the drainage ponds appear to contain more heavy metals than the adjacent Lake Lucien at the Maitland interchange.

<p>Disturbed peatlands may undergo a dramatic alteration in ecohydrological conditions, potentially limiting the recolonisation of peat-forming species like <em>Sphagnum</em>. A poor fen was experimentally drained in 1984, both the drained and undrained portion of the peatland burned in 2001, providing an unique opportunity to examine the ecohydrological response to ‘double disturbance’.</p> <p>The undrained site<em> </em>was characterized by a healthy recovery of<em> </em>peatland microform <em>Sphagnum</em> species, low soil water pressure (Ψ), high volumetric soil moisture (θ) content and high and stable water table position. However, the drained site showed no recolonization of <em>Sphagnum</em> with <em>Brome</em> grasses representing the dominant surface cover nine years post-wildfire.</p> <p>While the study period was generally wet and as such Ψ did not exceed thresholds limiting <em>Sphagnum</em> growth (≥ -100 mb) during the study period, a series of ecohydrological influences were found to be operating, limiting <em>Sphagnum</em> recolonisation at the drained site. The physical peat structure following drainage and wildfire has been considerably altered, changing the moisture retention and water storage properties of the peat, largely through substantive increases in bulk density (ρ_b). Moreover, specific yield (<em>Sy</em>) has also decreased the drained peat having become more humified, increasing unstable water table fluctuations. As such, this has lowered the resilience to drought. Only smaller decreases in θ are required to reach Ψ ≥ -100 mb at the drained and impose ecophysiological stress on <em>Sphagnum</em> growth. Dense canopy cover (<em>Betula</em> and <em>Sali</em>x) has limited available radiation at the surface to recolonisation, shading out the surface, further limiting <em>Sphagnum</em> recolonisation.</p> / Master of Science (MSc)

Yes / Event monitoring is an essential application of Smart City platforms. Real-time monitoring of gully and drainage blockage is an important part of flood monitoring applications. Building viable IoT sensors for detecting blockage is a complex task due to the limitations of deploying such sensors in situ. Image classification with deep learning is a potential alternative solution. However, there are no image datasets of gullies and drainages. We were faced with such challenges as part of developing a flood monitoring application in a European Union-funded project. To address these issues, we propose a novel image classification approach based on deep learning with an IoT-enabled camera to monitor gullies and drainages. This approach utilises deep learning to develop an effective image classification model to classify blockage images into different class labels based on the severity. In order to handle the complexity of video-based images, and subsequent poor classification accuracy of the model, we have carried out experiments with the removal of image edges by applying image cropping. The process of cropping in our proposed experimentation is aimed to concentrate only on the regions of interest within images, hence leaving out some proportion of image edges. An image dataset from crowd-sourced publicly accessible images has been curated to train and test the proposed model. For validation, model accuracies were compared considering model with and without image cropping. The cropping-based image classification showed improvement in the classification accuracy. This paper outlines the lessons from our experimentation that have a wider impact on many similar use cases involving IoT-based cameras as part of smart city event monitoring platforms. / European Regional Development Fund Interreg project Smart Cities and Open Data REuse (SCORE).

Road construction through sulfidic materials in Virginia has resulted in localized acid rock drainage (ARD) that threatens water quality, fill stability, integrity of building materials, and vegetation management. The objectives of this study were: i) to develop a state-wide sulfide hazard rating map based on characterization of the geologic formations associated with acid roadcuts, ii) to estimate depth to sulfidic sediments in the Coastal Plain based on landscape relationships, and iii) to evaluate potential acidity testing procedures on diverse materials. Geologic formations associated with acid roadcuts were characterized by potential peroxide acidity (PPA) and S content, and grouped into four categories. Listed in order of increasing severity, these formations included: the Tabb Formation (Coastal Plain), the Lynchburg Group of the Ashe Formation (Blue Ridge), the Chesapeake Group and Lower Tertiary deposits (Coastal Plain), the Millboro shale, Marcellus shale, Chatanooga shale and Needmore Formation (Valley and Ridge), and the Quantico Formation (Piedmont). Evaluation of landscape parameters near Richmond, Virginia, indicated that the likelihood of encountering sulfidic materials within a given depth at a specific location was related to elevation and mapped soil types. Elevation and soil map units were assigned to risk classes to indicate the likelihood of encountering sulfides within a depth of 9 m. Comparison of PPA and S content for 296 diverse samples indicated that S may serve as a screening tool to evaluate materials without carbonates. Comparison of PPA and conventional Acid-Base Accounting (ABA) for 14 diverse samples indicated that PPA and ABA were highly correlated, with PPA yielding 0.60 to 0.95X the amount of acidity as ABA. Potential acidity by Soxhlet extraction and PPA were equivalent for 3 of 4 diverse samples. Average acidity and metal contents of leachate from Soxhlet extractors were correlated with acidity and metals of road drainage. Sulfide hazard analysis should be an essential step in the pre-design phase of highway construction and other earth-disturbing activities. / Ph. D.

Urban, agriculture and coal mining land use/cover impacts upon aquatic habitat, toxicity and biodiversity were investigated in Leading Creek, a 388 km2 watershed in southeastern Ohio. Abandoned strip mine land (ASML) and active deep underground mines were examined along with abandoned near-surface underground mine land (AUML). The work focused on assessment of aquatic toxicity, water quality, and biodiversity through investigation of associated ecological responses for both treated and untreated AMD. Relations were examined among land use/cover, chemistry, and various ecological and toxicological endpoints. Sources of data (scale 1:24000) included Landsat5 imaging from 1988 and 1994, and directly digitized extents of underground mining activities dating to the 19th century, with more recently created strip mines. USEPA and Ohio EPA qualitative habitat scoring protocols were used. Land use/cover thresholds were established using ASML=3%, AUML=2% to 10%, Urban=3% to 5%, and Bare Soil=3%. Biodiversity was assessed using qualitative benthic macroinvertebrate taxon richness and abundance, for total and EPT groups, respectively. A better understanding of acid mine drainage (AMD) was demonstrated linking land use/cover, coal bed, sediment, and water column chemistry to aquatic ecotoxicity through examination of the origin and fate of sulfate, magnesium, iron, manganese, and zinc. Key findings in risk assessment of Leading Creek indicated that (1) abandoned near-surface underground mine lands (AUML) were associated with >90% of untreated AMD reaching Leading Creek; (2) degradation to aquatic ecology was primarily associated with water quality degradation due to AMD, not with sediment quality degradation; (3) modest habitat destruction, especially sedimentation effects, were observed for ASML>3%, and urbanization>5% in small subsheds; (4) unique chemical signatures differentiated mining techniques instream; and (5) in situ Corbicula fluminea growth rates were dependent upon drainage area. Sporadic signs of agricultural and urban impacts were indicated from acute toxicity with Ceriodaphnia dubia and chronic in situ toxicity testing with C. fluminea. Both the ecotoxicological tests were shown to be reliable indicators of AMD impact from AUML, on watershed and subwatershed scales. AMD was strongly associated with depressed biodiversity, low pH, and elevated zinc. Ecotoxicity monitoring supported interconnections found between sediment and water chemistry, land use/cover, and biodiversity. / Ph. D.

This dissertation reports the design of a process (patent pending) to remove iron from acid mine drainage (AMD) without the formation of metal hydroxide sludge. The system includes the oxidation of ferrous iron in a packed bed bioreactor, the precipitation of iron within a fluidized bed, the removal of manganese and heavy metals (Cu, Ni, Zn) in a trickling filter at high (>9) pH, with final neutralization in a carbonate bed. The technique avoided the generation of iron oxyhydroxide sludge. In the packed bed bioreactor, maximum substrate oxidation rate (R_,max) was 1500 mg L⁻¹ h⁻¹ at dilution rates of 2 h⁻¹, with oxidation efficiency at 98%. The half-saturation constant (similar to a Ks) was 6 mg L⁻¹. The oxidation rate was affected by dissolved oxygen below 2 mg L⁻¹, with a Monod-type Ko for DO of 0.33 mg L⁻¹. Temperature had a significant effect on oxidation rate, but pH (2.0 to 3.25) and supplemental CO₂ did not affect oxidation rates. Iron hydroxide precipitation was not instantaneous when base was added at a OH/Fe ratio of less than 3. Induction time was found to be a function of pH, sulfate concentration and iron concentration, with a multiple R² of 0.84. Aqueous [Al (III)] and [Mn (II)] did not significantly (α = 0.05) affect induction time over the range of concentrations investigated. When specific loading to the fluidized bed reactor exceeded 0.20 mg Fe m⁻² h⁻¹, dispersed iron particulates formed leading to a turbid effluent. Reactor pH determined the minimum iron concentration in the effluent, with an optimal at pH 3.5. Total iron removals of 98% were achieved in the fluidized bed with effluent [Fe] below 10 mg L⁻¹. Further iron removal occurred within the calcium carbonate bed. Heavy metals were removed both in the fluidized bed reactor as well as in the trickling filter. Oxidation at pH >9 caused manganese to precipitate (96% removal); removals of copper, nickel, and zinc were due primarily to sorption onto oxide surfaces. Removals averaged 97% for copper, 70% for nickel and 94% for zinc. The treatment strategy produced an effluent relatively free of iron (< 3 mg/L), without the formation of iron sludge and may be suitable for AMD seeps, drainage from acidic tailings ponds, active mine effluent, and acidic iron-rich industrial wastewater. / Ph. D.