Thesis (M. Phil.)--University of Hong Kong, 2010. / Includes bibliographical references (p. 129-132). Also available in print.

published_or_final_version / Architecture / Master / Master of Landscape Architecture

No description available.

No description available.

The enhanced potential vorticity gradients along the polar jet serve as a wave guide for trapped Rossby waves. These tropopause level, synoptic scale Rossby waves are of primary importance for weather development at the surface. In particular, extreme weather events have been linked to the existence of overlying upper level meridionally elongated filaments of stratospheric (high potential vorticity) air which form during the breaking of these waves. Motivated by the desire to understand the conditions under which these waves are formed and to improve their predictability, the current thesis discusses one excitation mechanism---the potential vorticity anomaly associated with a hurricane approaching the extra-tropics. / Attention is directed toward the adiabatic interaction of the cyclone with the polar jet before the two features meet. The hurricane's ability to excite Rossby waves is verified from observations of past interactions and theoretical study using a mechanistic model with idealized settings. The nature of the interaction between the cyclone and the polar jet is found to be sensitive to parameters such as the cyclone's radius and PV anomaly. Three different regimes have been identified. It is also concluded that the same parameters have influence on the skill of the Rossby wave prediction. The more intense the cyclone is, the harder it is to accurately predict the response of the polar jet.

Freshwater ecosystems reflect the condition of their surrounding landscape, and thus are particularly vulnerable to anthropogenic stressors associated with human land-use. One of the most prevalent stressors on stream ecosystems in agricultural regions, such as the Canterbury Plains of New Zealand, is eutrophication, or increased primary productivity. The aim of this thesis was to investigate effects of eutrophication on stream communities, specifically food web structure and ecosystem function. From a food web perspective, eutrophication is a shift in the form and amount of available energy from externally-produced (allochthonous) to internal (autochthonous) basal resources. Such shifts are frequently associated with land-use intensification, due to riparian vegetation removal and increased nutrient inputs, both of which enhance autochthonous production. A field survey across a gradient of eutrophication showed that eutrophic stream food webs are largely autochthonously-based and often contain large numbers of defended primary consumers, which form trophic bottlenecks and prevent energy from reaching higher trophic levels. Consequently, while there is more total energy available, less of that energy is in a usable form for stream food webs. Moreover, I found that eutrophic streams are largely composed of generalist consumers, which shift their diets to refocus on autochthonous resources with increasing productivity. Given that eutrophication causes food web resources to become more homogenous and was a primary driver of food web change, I tested whether reintroducing allochthonous subsidies would alter or reverse the negative effects of eutrophication. To do this I conducted a short-term community assembly experiment and a year-long population biomass accrual study. I found that the simplified, generalist-dominated communities in eutrophic streams did not respond to changes in resource diversity as predicted by food web theories, which are based on more diverse food webs. After restoration of allochthonous subsidies, defended generalist taxa continued to dominate the invertebrate communities. However, while restoring allochthonous subsidies did not mitigate the numerical dominance of defended consumers, the biomass accrual of other, previously excluded desirable taxa, such as mayflies and predatory invertebrates, increased following resource additions. This indicates that more energy reached the top of the food web, suggesting that resource additions alleviated trophic bottlenecks. Overall, my findings have advanced current knowledge about key mechanisms driving food web responses to both anthropogenic stress and to restoration efforts, which can be applied to improve management and restoration of stream ecosystems.

Because of the increasing data sizes and demands for low latency in modern data analysis, the traditional data warehousing technologies are greatly pushed beyond their limits. Several stream data warehouse (SDW) systems, which are warehouses that ingest append-only data feeds and support frequent refresh cycles, have been proposed including different methods to improve the responsiveness of the systems. Materialized views are critical in large-scale data warehouses due to their ability to speed up queries. Thus an SDW maintains layers of materialized views. Materialized view maintenance in SDW systems introduces new challenges. However, some of the existing SDW systems do not address the maintenance of views while others employ view maintenance techniques that are not efficient. This thesis presents ViewDF, a flexible framework for incremental maintenance of materialized views in SDW systems that generalizes existing techniques and enables new optimizations for views defined with operators that are common in stream analytics. We give a special view definition (ViewDF) to enhance the traditional way of creating views in SQL by being able to reference any partition of any table. We describe a prototype system based on this idea, which allows users to write ViewDFs directly and can automatically translate a broad class of queries into ViewDFs. Several optimizations are proposed and experiments show that our proposed system can improve view maintenance time by a factor of two or more in practical settings.

AdaBoost has been considered one of the best classifiers ever developed, but two important problems have not yet been addressed. The first is the dependency on the ``weak" learner, and the second is the failure to maintain the performance of learners with small error rates (i.e. ``strong" learners). To solve the first problem, this work proposes using a different learner in each iteration - known as AdaBoost Dynamic (AD) - thereby ensuring that the performance of the algorithm is almost equal to that of the best ``weak" learner executed with AdaBoost.M1. The work then further modifies the procedure to vary the learner in each iteration, in order to locate the learner with the smallest error rate in its training data. This is done using the same weight calculation as in the original AdaBoost; this version is known as AdaBoost Dynamic with Exponential Loss (AB-EL). The results were poor, because AdaBoost does not perform well with strong learners, so, in this sense, the work confirmed previous works' results. To determine how to improve the performance, the weight calculation is modified to use the sigmoid function with algorithm output being the derivative of the same sigmoid function, rather than the logistic regression weight calculation originally used by AdaBoost; this version is known as AdaBoost Dynamic with Logistic Loss (AB-DL). This work presents the convergence proof that binomial weight calculation works, and that this approach improves the results for the strong learner, both theoretically and empirically. AB-DL also has some disadvantages, like the search for the ``best" classifier and that this search reduces the diversity among the classifiers. In order to attack these issues, another algorithm is proposed that combines AD ``weak" leaner execution policy with a small modification of AB-DL's weight calculation, called AdaBoost Dynamic with Added Cost (AD-AC). AD-AC also has a theoretical upper bound error, and the algorithm offers a small accuracy improvement when compared with AB-DL, and traditional AdaBoost approaches. Lastly, this work also adapts AD-AC's weight calculation approach to deal with data stream problem, where classifiers must deal with very large data sets (in the order of millions of instances), and limited memory availability.

The removal of indigenous forest and associated fragmentation of habitats has probably had significant impacts on the diversity of stream communities in New Zealand. In this study I investigated the effects of forest fragmentation on stream invertebrate communities on Banks Peninsula. Six catchments were investigated, three with continuous indigenous forest in the riparian zone and three with fragmented indigenous riparian forest. An extensive benthic survey was conducted at three sites in each river catchment, one downstream on the mainstem of the river and two sites in different headwater tributaries. Adult sampling, consisting of malaise and sticky trapping, was also conducted at a sub-set of sites. Taxonomic richness of both the benthic and adult communities was significantly higher in continuous forest than in fragmented forest, and the composition of benthic communities also differed between continuous and forest fragments. Furthermore, benthic invertebrate densities were higher in fragments than continuous forest sites. The fragments in the headwaters were more likely to support forest specialist taxa (e.g. the stonefly Zelandobius wardi, and the caddisfly Costachorema peninsulae), than the downstream fragments. My results indicate that forest fragmentation has resulted in marked changes in benthic communities on Banks Peninsula, and that location of the fragment within the catchment also is important in influencing the diversity and composition of benthic communities. The maintenance of indigenous forest in the headwaters of streams may be essential for the persistence of endemic and forest specialist taxa on Banks Peninsula.

Thesis (M.S.)--Oregon State University, 2008. / Printout. Includes bibliographical references (leaves 52-54). Also available on the World Wide Web.