RESUMO
According to Milankovitch theory, the lower summer insolation at high latitudes about 115,000 years ago allowed winter snow to persist throughout summer, leading to ice-sheet build-up and glaciation. But attempts to simulate the last glaciation using global atmospheric models have failed to produce this outcome when forced by insolation changes only. These results point towards the importance of feedback effects-for example, through changes in vegetation or the ocean circulation-for the amplification of solar forcing. Here we present a fully coupled ocean-atmosphere model of the last glaciation that produces a build-up of perennial snow cover at known locations of ice sheets during this period. We show that ocean feedbacks lead to a cooling of the high northern latitudes, along with an increase in atmospheric moisture transport from the Equator to the poles. These changes agree with available geological data and, together, they lead to an increased delivery of snow to high northern latitudes. The mechanism we present explains the onset of glaciation-which would be amplified by changes in vegetation-in response to weak orbital forcing.
RESUMO
Chromosome, a knowledge-based analysis system has been designed for the classification of human chromosomes. Its aim is to perform an optimal classification by driving a tool box containing the procedures of image processing, pattern recognition and classification. This paper presents the general architecture of Chromosome, based on a multiagent system generator. The image processing tool box is described from the met aphasic enhancement to the fine classification. Emphasis is then put on the knowledge base intended for the chromosome recognition. The global classification process is also presented, showing how Chromosome proceeds to classify a given chromosome. Finally, we discuss further extensions of the system for the karyotype building.
