NorthWest Research Associates

Fang Liu (Associate Professor, School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China)

Clustering is the problem of finding a set of groups of similar objects within a data set while keeping dissimilar objects separated in different groups or the group of noise. It is a common technique for statistical data analysis, and used in many fields, including machine learning, pattern recognition, data mining, image analysis, information retrieval, and bioinformatics. Density based clustering techniques are attractive because it can find arbitrary shaped clusters along with noisy outliers. In this talk, the density based clustering methods will be discussed, and the result of applying them in different area (such as text extraction from image, postmortem estimation from eye image) is presented as well.

Chelle Gentemann

Several recent studies have concluded that coupled climate models should utilize a diurnally varying SST to examine the details of the boundary layer response and ensuing air-sea interactions. The global distribution of diurnal warming is clearly linked to wind speed and will therefore respond to the climatic distributions and seasonal or anomalous changes in wind speed, as shown by the response to ENSO wind speed anomalies. The Subtropical High regions in each ocean basin, and the Tropical Indian and Western Pacific Oceans have the largest averages of diurnal warming. The intra-day variability of surface warming has been related to the stability of the boundary layer and atmospheric convection. Since the tropical convection is an important driver of global atmospheric circulation, this example of ocean-atmospheric feedback underscores how diurnal warming of the ocean surface may influence larger scale weather patterns and climate.

Results from several satellites show significant diurnal warming present over large regions. Several models (both empirical and physical) of diurnal variability have been developed, but show little agreement with each other. Comparisons of data and models will be used to discuss the global spatial/temporal distribution of diurnal warming and how accurately we actually understand it.