Vegetation dynamics and their implications for the management of wetlands in the lowlands of Nepal

The maintenance of species richness is often a priority in the management of nature reserves, where consumptive use of resources is generally prohibited. The purpose of this research was to improve management by understanding the vegetation dynamics in the lowlands of Nepal. The objectives were to determine vegetation associations in relation to environments and human-induced disturbances that affect vegetation dynamics on floodplains, where upstream barrages had altered flooding patterns, and consumptive use of plant resources was influencing natural processes. Floodplain vegetation in relation to physical environments and disturbances were studied along transects, perpendicular to the course of the Mahakali River in the western Terai, Nepal. Forest structural changes were studied for three years in ten plots. A randomized split-block experiment with nine burning and grazing treatments was performed in seasonally flooded grasslands. A semi-structured questionnaire was used to assess people's socio-economic status, natural resource use patterns and conservation attitudes. ^ Elevation, soil organic matter, nitrogen, percentage of sand and grazing intensity were significant in delineating herbaceous vegetation assemblages, whereas elevation and livestock grazing were significant in defining forest type boundaries. On the floodplain islands, highly grazed Dalbergia sissoo-Acacia catechu forests were devoid of understory woody vegetation, but the lightly grazed D. sissoo-mixed forests had a well-developed second canopy layer, comprising woody species other than D. sissoo and A. catechu. In grasslands, species richness and biomass production were highest at intermediate disturbance level represented by the lightly grazed and ungrazed early-burned treatments. Ethnicity, education and resource use patterns were important in influencing conservation attitudes. A succession towards the mixed forests would occur in D. sissoo-dominated floodplain forests, where dams and barrages reduce flooding and associated fluvial processes, and if livestock grazing is stopped, as occasionally suggested by nature conservationists. In seasonally flooded grasslands, early burning with moderate grazing would enhance the species diversity and productivity. There is a need to implement a participatory integrated wetland management plan, to include community development, education and off farm income generation, to assure participatory conservation and management of wetlands in Nepal. ^

Exploring Hidden Coherent Feature Groups and Temporal Semantics for Multimedia Big Data Analysis

Thanks to the advanced technologies and social networks that allow the data to be widely shared among the Internet, there is an explosion of pervasive multimedia data, generating high demands of multimedia services and applications in various areas for people to easily access and manage multimedia data. Towards such demands, multimedia big data analysis has become an emerging hot topic in both industry and academia, which ranges from basic infrastructure, management, search, and mining to security, privacy, and applications. Within the scope of this dissertation, a multimedia big data analysis framework is proposed for semantic information management and retrieval with a focus on rare event detection in videos. The proposed framework is able to explore hidden semantic feature groups in multimedia data and incorporate temporal semantics, especially for video event detection. First, a hierarchical semantic data representation is presented to alleviate the semantic gap issue, and the Hidden Coherent Feature Group (HCFG) analysis method is proposed to capture the correlation between features and separate the original feature set into semantic groups, seamlessly integrating multimedia data in multiple modalities. Next, an Importance Factor based Temporal Multiple Correspondence Analysis (i.e., IF-TMCA) approach is presented for effective event detection. Specifically, the HCFG algorithm is integrated with the Hierarchical Information Gain Analysis (HIGA) method to generate the Importance Factor (IF) for producing the initial detection results. Then, the TMCA algorithm is proposed to efficiently incorporate temporal semantics for re-ranking and improving the final performance. At last, a sampling-based ensemble learning mechanism is applied to further accommodate the imbalanced datasets. In addition to the multimedia semantic representation and class imbalance problems, lack of organization is another critical issue for multimedia big data analysis. In this framework, an affinity propagation-based summarization method is also proposed to transform the unorganized data into a better structure with clean and well-organized information. The whole framework has been thoroughly evaluated across multiple domains, such as soccer goal event detection and disaster information management.

Wildfire Assessment Using FARSITE Fire Modeling: A Case Study in the Chihuahua Desert of Mexico

The Chihuahua desert is one of the most biologically diverse ecosystems in the world, but suffers serious degradation because of changes in fire regimes resulting in large catastrophic fires. My study was conducted in the Sierra La Mojonera (SLM) natural protected area in Mexico. The purpose of this study was to implement the use of FARSITE fire modeling as a fire management tool to develop an integrated fire management plan at SLM. Firebreaks proved to detain 100% of wildfire outbreaks. The rosetophilous scrub experienced the fastest rate of fire spread and lowland creosote bush scrub experienced the slowest rate of fire spread. March experienced the fastest rate of fire spread, while September experienced the slowest rate of fire spread. The results of my study provide a tool for wildfire management through the use geospatial technologies and, in particular, FARSITE fire modeling in SLM and Mexico.