Environmental Science & Technology
Permanent URI for this communityhttp://hdl.handle.net/1903/2216
Browse
14 results
Search Results
Item Towards an Autonomous Algal Turf Scrubber: Development of an Ecologically-Engineered Technoecosystem(2010) Blersch, David Michael; Kangas, Patrick C.; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The development of an autonomous and internally-controlled technoecological hybrid is explored. The technoecosystem is based on an algal turf scrubber (ATS) system that combines engineered feedback control programming with internal feedback patterns within the ecosystem. An ATS is an engineered, high-turbulent aquatic system to cultivate benthic filamentous algae for the removal of pollutants from an overlying water stream. This research focuses on designing a feedback control system to control the primary production of algae in an ATS through monitoring of the algal turf metabolism and manipulation of the turbulence regime experienced by the algae. The primary production of algae in an ATS, and thus the potential of the waste treatment process, is known to be directly related to the level of turbulence in the flowing water stream resulting from the amplitude and frequency of the wave surge. Experiments are performed to understand the influence of turbulence on the biomass production rate of algae in an ATS. These results show that biomass production is correlated with wave surge amplitude at a constant frequency. Further, the influence of turbulence on the net ecosystem metabolism of an algal turf in an ATS was investigated. Results showed that both net primary production and respiration, measured through the diurnal change of inorganic carbon, follow a subsidy-stress relationship with increasing wave surge frequency, although some of this trend may be explained by the transfer of metabolic gases across the air-water interface. A feedback control algorithm, developed to monitor the net primary production and manipulate a controlling parameter, was found to converge quickly on the state of maximum primary production when the variance of the input data was low, but the convergence rate was slow at only moderate levels of input variance. The elements were assembled into a physical system in which the feedback control algorithm manipulated the turbulence of the flow in an ATS system in response to measured shifts in ecosystem metabolism. Results from this testing show that the system can converge on the maximum algal productivity at the lowest level of turbulence--the most efficient state from an engineering perspective--but in practice the system was often confounded by measurement noise. Investigation into the species composition of the dominant algae showed shifting relative abundance for those units under automated control, suggesting that certain species are more suited for utilizing the technological feedback pathways for manipulating the energy signature of their environment.Item Advanced imaging and data mining technologies for medical and food safety applications(2009) Jiang, Lu; Tao, Yang; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)As one of the most fast-developing research areas, biological imaging and image analysis receive more and more attentions, and have been already widely applied in many scientific fields including medical diagnosis and food safety inspection. To further investigate such a very interesting area, this research is mainly focused on advanced imaging and pattern recognition technologies in both medical and food safety applications, which include 1) noise reduction of ultra-low-dose multi-slice helical CT imaging for early lung cancer screening, and 2) automated discrimination between walnut shell and meat under hyperspectral florescence imaging. In the medical imaging and diagnosis area, because X-ray computed tomography (CT) has been applied to screen large populations for early lung cancer detection during the last decade, more and more attentions have been paid to studying low-dose, even ultra-low-dose X-ray CTs. However, reducing CT radiation exposure inevitably increases the noise level in the sinogram, thereby degrading the quality of reconstructed CT images. Thus, how to reduce the noise levels in the low-dose CT images becomes a meaningful topic. In this research, a nonparametric smoothing method with block based thin plate smoothing splines and the roughness penalty was introduced to restore the ultra-low-dose helical CT raw data, which was acquired under 120 kVp / 10 mAs protocol. The objective thorax image quality evaluation was first conducted to assess the image quality and noise level of proposed method. A web-based subjective evaluation system was also built for the total of 23 radiologists to compare proposed approach with traditional sinogram restoration method. Both objective and subjective evaluation studies showed the effectiveness of proposed thin-plate based nonparametric regression method in sinogram restoration of multi-slice helical ultra-low-dose CT. In food quality inspection area, automated discrimination between walnut shell and meat has become an imperative task in the walnut postharvest processing industry in the U.S. This research developed two hyperspectral fluorescence imaging based approaches, which were capable of differentiating walnut small shell fragments from meat. Firstly, a principal component analysis (PCA) and Gaussian mixture model (PCA-GMM)-based Bayesian classification method was introduced. PCA was used to extract features, and then the optimal number of components in PCA was selected by a cross-validation technique. The PCA-GMM-based Bayesian classifier was further applied to differentiate the walnut shell and meat according to the class-conditional probability and the prior estimated by the Gaussian mixture model. The experimental results showed the effectiveness of this PCA-GMM approach, and an overall 98.2% recognition rate was achieved. Secondly, Gaussian-kernel based Support Vector Machine (SVM) was presented for the walnut shell and meat discrimination in the hyperspectral florescence imagery. SVM was applied to seek an optimal low to high dimensional mapping such that the nonlinear separable input data in the original input data space became separable on the mapped high dimensional space, and hence fulfilled the classification between walnut shell and meat. An overall recognition rate of 98.7% was achieved by this method. Although the hyperspectral fluorescence imaging is capable of differentiating between walnut shell and meat, one persistent problem is how to deal with huge amount of data acquired by the hyperspectral imaging system, and hence improve the efficiency of application system. To solve this problem, an Independent Component Analysis with k-Nearest Neighbor Classifier (ICA-kNN) approach was presented in this research to reduce the data redundancy while not sacrifice the classification performance too much. An overall 90.6% detection rate was achieved given 10 optimal wavelengths, which constituted only 13% of the total acquired hyperspectral image data. In order to further evaluate the proposed method, the classification results of the ICA-kNN approach were also compared to the kNN classifier method alone. The experimental results showed that the ICA-kNN method with fewer wavelengths had the same performance as the kNN classifier alone using information from all 79 wavelengths. This demonstrated the effectiveness of the proposed ICA-kNN method for the hyperspectral band selection in the walnut shell and meat classification.Item Modeling Approaches for Treatment Wetlands(2009) Carleton, James N.; Montas, Hubert J; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Although treatment wetlands can reduce pollutant loads, reliably predicting their performance remains a challenge because removal processes are often complex, spatially heterogeneous, and incompletely understood. Although initially popular for characterizing wetland performance, plug flow reactor models are problematic because their parameters exhibit correlation with hydraulic loading. One-dimensional advective-dispersive-reactive (ADE) models are also inadequate because longitudinal dispersion in wetlands is often non-Fickian as a result of steep velocity gradients. Models that make use of residence time distributions have shown promise in improving performance characterization, particularly when interdependencies of stream-tube scale velocities and reaction rate coefficients are considered (the "DND" approach). However this approach is limited to steady-state conditions, and to an assumption that transverse mixing is nil. This dissertation investigates three aspects of wetland modeling and is organized in a journal paper format. The first paper describes development of a DND model which accommodates non-steady-state conditions. The model processes flow and inlet concentration time series, and calculates as output effluent concentration time series. A version of the code allows optimization of model parameters by minimization of summed squared deviations between predicted and measured effluent concentrations. In example comparisons, model results compare favorably with measured data. The second paper develops an analytical solution to a two-dimensional advective-dispersive-reactive equation, in which all flux terms are expressed as power functions of the transverse dimension. For uniform inlet concentration this idealized heterogeneity model is similar to a DND model, but with the inclusion of transverse diffusion. An example is used to illustrate the beneficial impact that transverse mixing has on reactor performance. The third paper describes development of a model based upon a stochastic interpretation of the ADE. The solution technique that is employed results in a bicontinuum model that for steady-state conditions becomes a weighted sum of two exponential decline functions. For low and intermediate degrees of mixing, model results nicely match those of the corresponding idealized heterogeneity model, and for high mixing they match results of the corresponding one-dimensional ADE. Comparisons against data suggest the bicontinuum model may represent wetland performance better than the DND model in some but not all cases.Item Hyperspectral Reflectance as an Indicator of Foliar Nutrient Levels in Hybrid Poplar Clone OP-367 Grown on Biosolid Amended Soil(2009) Griffeth, Tommy; Felton, Gary; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Trees of the genus Populus are fast growing trees that require considerable amounts of water and nutrients to meet physiological growth demands. The determination of correlations between hybrid poplar leaf spectral reflectance in the 325-1100 nm range, laboratory foliar analysis of leaf macronutrient and micronutrient concentrations, and leaf water potential datasets were analyzed using Full Cross-Validation and Test Set Models via the partial least squares (PLS) method of regression analysis. Based on an evaluation of the slope of the Predicted vs. Measured regression line, the root mean squared error (RMSE), and r-squared, the majority of the models constructed did not adequately model foliar concentrations from spectral data. However, the models for H, N, P, K, Cu and Al had values (slope of the Predicted vs. Measured regression line greater than 0.50 and r-squared values greater than 0.50 in at least one type of model) that warrant future study.Item Comparison of Infiltration Equations and their Field Validation with Rainfall Simulation(2006-12-13) Turner, Ellen Rebecca; Felton, Gary K; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Infiltration is a complex process with many factors contributing to the rate. Different approximate equations for infiltration differ in the parameters they require and predict different infiltration rate curves. Five equations including those of Kostiakov, Horton, Holtan, Philip and Green-Ampt were compared to determine which one most accurately predicted measured infiltration rates from rainfall simulation events at two different locations. Parameters were developed from measured infiltration data and laboratory analyses of soil samples. The Green-Ampt, Holtan and Philip equations with respective root mean squared errors of 0.15, 0.17, and 0.19 cmh-1, provided the first, second and third best estimates of infiltration rates, for observed infiltration data at the University of Maryland's Research and Education Center in Upper Marlboro, Maryland. An atypical infiltration curve was observed for the Poplar Hill site on the Eastern Shore of Maryland for which infiltration rate was constant and equal to rainfall rate.Item IMPROVEMENT IN ESTIMATING POLLUTION TRANSPORT BY DEVELOPING STREAMFLOW COMPONENTS ASSESSMENT IN THE GIS ENVIRONMENT(2006-11-10) Nejadhashemi, Amirpouyan; Shirmohammadi, Adel; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Water by nature is a suitable domain for the transport of contaminants through watersheds. Evaluating the relative amounts of stored or moving water via the different components of the hydrological cycle is required for precise and strict management and planning of water resources. One of the most challenging parts of this process is the separation and quantification of baseflow from the total streamflow hydrograph. The aim of this study was to separate the storm runoff hydrograph into its components, thus being able to infer about sources and hydrological pathways of the storm runoff. The specific objectives of this study were to identify the most accurate and user-friendly streamflow partitioning method, to evaluate the accuracy of each of these methods using separately measured surface and subsurface flow data, and finally to improve available techniques or develop a more precise approach for separation of hydrograph components. In the early stage of this study, forty different streamflow partitioning methods were reviewed and classified into three-component, analytical, empirical, graphical, geochemical and automated methods and five methods were identified as being the most relevant and least input intensive. The performance of these methods were tested against independently measured surface and subsurface flow data obtained on a field scale watershed Boughton's method produced the most consistent and accurate results. However, its accuracy depends upon the proper estimation of the end of surface runoff, and the fraction factor (α). It was demonstrated that incorporating physical and hydrologic characteristics of a watershed can significantly improve the accuracy of hydrograph separation techniques when used jointly with enhanced recession limb analysis, calibration approach, and time-discretization method. Finally, simulation of the model for different scenarios (e.g., soils, land use, etc.) was performed within the geographical information systems for a large scale watershed (Little River Watershed in Georgia). Results showed that the weighted discharge method is better than the weighted average curve number method and the modified Boughton's method because it divides a watershed into small filed scale pixels and treats each pixel separately, thus mimicking the field scale station Z conditions where the method was successfully applied.Item Assessment of nitrogen status and vegetation composition in tidal freshwater marshes using partial least squares regression models of hyperspectral canopy reflectance(2006-04-28) Jenkins, Emily Poynter; Tilley, David R; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Hyperspectral canopy reflectance was used to predict sub-surface water nutrients, vegetation composition, and canopy nutrients, which could lead to more useful means for assessing the status of wetlands. Thirty field quadrats at two tidal freshwater marsh sites on the Nanticoke River (Maryland) were treated with five nitrogen levels. During the 2004-05 growing seasons, hyperspectral canopy reflectance was measured using a spectroradiometer with 1nm resolution across the visible and near - infrared spectrum (350-1075 nm), water samples were collected using lysimeters, species cover was quantified, and biomass was collected and analyzed for canopy nutrients. ANOVA was used to determine whether nitrogen affected reflectance, species composition, canopy N and P, and partial least squares regression was used to develop reflectance models predictive of these ecosystem properties. Results indicated that hyperspectral radiometry could be used as a remote sensing tool for quantifying sub-surface water nitrogen, vegetation composition, and canopy nutrients in tidal freshwater marshes.Item Modeling and monitoring pathogen transport through vegetated filter strips(2004-08-05) Roodsari, Gholamreza Moosapour; Shirmohammadi, Adel; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Contamination of natural waters by microorganisms directly affects public health. Field application of manure can potentially result in surface and groundwater contamination. The objective of this study was to observe and quantify the effects of vegetated filter strips (VFS) on surface and subsurface transport of fecal oviform (FC) surrogates for bacterial pathogens released from a surface - applied bovine/swine manure. The study included a field-based lysimeter equipped with multi-sensor moisture probes to monitor real-time water content through the soil profile, and with other proper instrumentation to monitor and quantify the spatial and temporal release rates of pathogenic bacteria. Another component of this study involved development and testing of a computer model to predict the surface and subsurface transport of FC. Results showed that bare plots offered no resistance to surface flow, thus FC were detected in runoff at 600 cm from the ridge of the lysimeter within 10 minutes of the rainfall initiation. Results from vegetated plots showed that vegetation substantially attenuated surface flow of water as compared to bare plots. Unlike the bare soil, the results showed that the vegetated soil surface created a much less uniform transport pattern for FC. Vegetation changed transport patterns and levels of FC concentrations much more significantly than soil texture did. Results showed that E.coli and Salmonella cholerasuis behaved similarly and resulted in similar transport patterns in both bare plots. Results also showed that both organisms demonstrated a two-stage exponential release rate with a fast release rate in the first 10 minutes of the rainfall simulation. A one-dimensional convective-dispersive equation using the continuity equation and the Manning's equation were used in MODCHOI model (a modified version of KORMIL2) to predict the surface transport of FC. To simulate the vertical movement of FC, a one-dimensional kinematic wave model was developed and used. Green and Ampt, Philip, and Schmid infiltration models were also applied to the vertical water flow movement. The models simulated the spatial and the temporal distribution of FC in runoff assuming an exponential release of FC from the manure. Simulation results satisfactorily modeled both flow and FC.Item ELECTROPHORETIC REMOVAL OF FINE PARTICULATES FROM AQUCULTURE EFFLUENT(2004-04-26) Hanna, Kalim; Wheaton, Fredrick; Biological Resources Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)As larger waste particles breakdown into smaller pieces under the mechanical stress of a recirculating system, it becomes increasingly more difficult to remove these particles through standard methods. This current work explores the possibility of using an impressed electric field as a means of water clarification. In this study aquaculture effluent is passed through an imposed electric field, where the fluid column is divided into two fluid streams: one closest to the positive electrode, and the other closest to the negative electrode. The water quality of each fluid stream is analyzed to determine if any difference results due to its exposure to the electric field. While this study did show that there was a statistically significant difference in certain water quality parameters between the two fluid streams, it was clear that the process was not efficient enough to be considered a viable and effective means of water clarification.Item Investigation into the system dynamics of a wetland soil technoecoysystem using redox potential as a metabolic indicator and feedback control parameter(2004-05-12) Blersch, David Michael; Kangas, Patrick C.; Biological Resources EngineeringThe engineering of technoecosystems (technological-ecological hybrids) was investigated, focusing specifically on novel behavior exhibited by an ecosystem when given control over its own energy sources via artificial feedback control circuits. A technoecosystem was constructed based upon wetland soil microcosms using redox potential as an indicator of system metabolism and as the controlled parameter. Two types of experiments were performed to elucidate system dynamics. Results of the carbon addition experiments exhibited an increase in the rate of decline of redox potential over time as a result of the feedback control system, indicating increased metabolism in the microcosms. Results of the carbon/nitrate selection experiments showed oscillatory redox potential over time, trending towards regulation of redox potential within a set range. Simple computational models of ecological limiting factors are developed to explain the results. A classification system for technoecosystems is proposed, and implications into technoecosystem intelligence and energetic autonomy are discussed.