Wetland, which contains about 20 - 30% of global soil carbon pool (Lal, 2008), is one of the world’s most important environmental resources for long-term carbon storage, and plays a vital role in global carbon cycling, especially in mitigating carbon concentration in the atmosphere. However, it is also the ecosystem that has been most seriously abused and suffering from continuous degradation and loss across the world. During the past few centuries, about 50% of the world’s wetland has been lost due to increasing anthropogenic disturbances and global warming (Mitsch & Gosselink, 2007; Gibbs, 2000; Dugan, 1993; Zedler and Kercher, 2005). One typical example is the wetland in Dongting Lake area of China, which was once China’s largest freshwater wetland and now has become the second one. During the past few decades, the Lake has experienced many significant changes causing the rapid degradation, shrinkage and fragmentation of its wetland. Therefore, monitoring the changes of the Lake wetland in spatial distribution and temporal trend and predicting its potential dynamics under climate change and human induced disturbances are becoming increasingly important for linking policy decision-making with regulatory actions and subsequent land-use activities. The overall objective of this project is to monitor the wetland changes in the Lake area and predict its dynamics in the future using proposed land use and land cover (LULC) classification, change detection and modelling approaches. To start with, this study examined the spatiotemporal dynamics of the Lake wetland patterns during the past half century through analyzing remotely sensed images acquired on six time points, including 1978, 1984, 1994, 2001, 2004, 2009, and 2013. A hybrid knowledge-based classification method which combines supervised and expert classification systems was first applied to conduct image classifications with special attention to the classification accuracy of the wetland categories including water, paddy field, reed and marsh categories. After that, a post-classification based change detection technique was carried out to monitor the dynamics of the Lake wetland. The error matrices and Kappa coefficients were than used to assess the classification accuracy. The classification results demonstrated that the proposed hybrid classification approach could discriminate the wetland categories from others with the high accuracy of 96.9%, 93.7%, 82.6%, and 82.4% for water, paddy field, reed, and marsh categories, respectively. The LULC analysis based on the classification showed that wetland area (reed and marsh) in the Lake area has decreased with a dramatic decrease trend after the Three Gorges Dam being fully operated in 2003. To predict future wetland changes and allocate the changes effectively, an integrated model incorporating the logistic, the Markov, and the Conversion of Land use and its Effects (CLUE-S) models has been developed and utilized to 1) produce the LULC probability surface maps; 2) to simulate the LULC change demand in 2013 and 2025 of which the demand for 2013 was then used for validating the results of this integrated model by comparing with the actual LULC maps of the same year; 3) to allocate the simulated changes of 2013 and 2025 based on the obtained LULC probability surface maps and some user-defined rules including land use conversion rules and conversion elasticity. The results from the model validation indicated that the integrated model performed very well with an overall modelling accuracy and Kappa statistic of 80.2% and 74.9%, respectively. The results also suggested that the wetland area is likely to undergo further decrease of another 256.3 km2 by 2025. In summary, this study focused on the development of a unique and integrated approach for the LULC image classification, change detection and prediction of the wetland area – Dongting Lake region in which the landscape was complex and experiencing fast and dramatic changes due to the construction of the TGD. The approach can be easily extended to other wetland associated studies. By providing the information of the long-term wetland dynamics and simulation of its future changes in the Lake area, this research will also enhance our understanding of wetland resources, their dynamics and relationships with human activity induced disturbances and thus promote our ability to make informed use and wise restoration regulations of wetlands.

Paketklassifikation ist eine Kernfunktionalität vieler Netzwerksysteme, wie zum Beispiel Firewalls und SDN-Switches. Für viele dieser Systeme ist Durchsatz von höchster Bedeutung. Weitere wichtige Eigenschaften sind dynamische Aktualisierbarkeit und hohe Regelsatz-Ausdrucksfähigkeit. Die Kombination dieser Eigenschaften macht Paketklassifikation zu einem schwierigen Problem. Diese Arbeit befasst sich mit dem Design von Klassifikationssystemen und -algorithmen, welche mindestens zwei dieser Eigenschaften vereinen. Es werden hybride Systeme sowie Systemspezialisierung verwendet, um effiziente Ansätze zum Paketklassifikationsproblem in drei Bereichen zu erarbeiten: Klassifikationsalgorithmen, Regelsatztransformation und hardwarebasierte Architekturen. Die Beiträge im Bereich der Klassifikationsalgorithmen sind Jit Vector Search (JVS) und das SFL-System. JVS verbessert existierende Techniken durch spezialisierte Suchdatenstrukturen und durch Nutzung von SIMD-Fähigkeiten der CPU, was in fast optimaler Klassifikationsperformanz bei kaum erhöhten Vorberechnungszeiten resultiert. Das hybride SFL-System hingegen kombiniert einen Klassifikationsalgorithmus mit einem Änderungspuffer, um sowohl hohe Klassifikations- als auch Aktualisierungsperformanz zu ermöglichen. Bezüglich Regelsatztransformationen wird die RuleBender-Technik vorgestellt, welche Suchbäume in Regelsätze für Firewalls mit Sprungsemantik kodiert. Somit kann der Durchsatz dieser Systeme unter Beibehaltung komplexer Regelsatzsemantik um eine Größenordnung gesteigert werden. Schließlich wird der MPFC-Ansatz vorgestellt, welcher einen Regelsatz in einen auf einem FPGA implementierbaren Matching-Schaltkreis übersetzt. Die generierten Schaltkreise sind hochoptimiert und kleiner als generische Matching-Schaltkreise. Um dynamische Regelsatzänderungen zu ermöglichen, wird der hybride Consul-Ansatz konzipiert, welcher MPFC-Matcher mit generischen Matching-Schaltkreisen kombiniert. / Packet classification is a core functionality of a wide variety of network systems, such as firewalls and SDN switches. For many of these systems, throughput is of paramount importance. Further important system traits are dynamic updateability and high expressiveness in terms of rule set semantics. The combination of several of these properties turns packet classification into a hard problem. This work focuses on the design of classification systems and algorithms that combine at least two of the abovementioned characteristics. To this end, the concepts of hybrid systems and system specialization are employed to obtain efficient approaches to the packet classification problem in three domains: classification algorithms, rule set transformation, and hardware-centric architectures. The contributions in the domain of classification algorithms are Jit Vector Search (JVS) and the SFL system. JVS improves upon existing techniques through specialized search data structures and by exploiting SIMD capabilities of the underlying CPU, which results in near-optimal classification performance at only slightly increased preprocessing times. In contrast, the SFL system is a hybrid approach that combines a classification algorithm with an update buffer to allow for high classification as well as update performance. With respect to rule set transformation, the RuleBender technique is proposed, which encodes search tree structures into rule sets of firewalls with jump semantics. That way, the throughput of these systems can be improved by an order of magnitude, while maintaining complex matching semantics. Finally, the MPFC approach is proposed, which translates a given rule set into a matching circuit that can be implemented on an FPGA. The generated circuits are highly optimized and significantly smaller than those of generic matchers. To allow for dynamic rule set updates, the hybrid Consul approach is devised, which combines MPFC circuits with a generic matcher.