RELATIONSHIPS BETWEEN PRECEDING PACIFIC SEA SURFACE TEMPERATURES AND SUBTROPICAL HIGH INDEXES OF MAIN RAINING SEASONS

With correlation analysis and factor analysis methods, the effects of preceding Pacific SSTs on subtropical high indexes of main raining seasons are discussed. The results of correlation analysis show that the effects of SSTs on five subtropical high indexes differ in seasons and regions. The variation of SSTs mostly affects the area and intensity indexes of the subtropical high, followed by the western ridge index, and the effect on the ridge line index is more remarkable than on the north boundary index. The results of factor analysis reveals that the first common factor of SST of each season reflected mainly the inverse relation of SSTs variation between the central and eastern part of equatorial Pacific and the western Pacific, which correlates better with the subtropical high indexes in the main raining seasons than other common factors of SST. The analysis of interdecadal variation indicated that the variation of SSTs was conducive to the emergence of the La Ni?a event before the end of 1970s, such that in the summer the subtropical high is likely to be weaker and smaller and located eastward and northward. After the 1980s, the opposite characteristics prevailed.

A Large-Scale Study of Failures on Petascale Supercomputers

With the rapid development of supercomputers, the scale and complexity are ever increasing, and the reliability and resilience are faced with larger challenges. There are many important technologies in fault tolerance, such as proacrive failure avoidance technologies based on fault prediction, reactive fault tolerance based on checkpoint, and scheduling technologies to improve reliability. Both qualitative and quantitative descriptions on characteristics of system faults are very critical for these technologies, This study analyzes the source of failures on two typical petascale supercomputers called Sunway BlueLight （based on multi-core CPUs） and Sunway TaihuLight （based on heterogeneous manycore CPUs）. It uncovers some interesting fault characteristics and finds unknown correlation relationship among main components＇ faults. Finally the paper analyzes the failure time of the two supercomputers in various grains of resource and different time spans, and builds a uniform multi-dimensional failure time model for petascale supereomputers.

Copula-Based Monte Carlo Scenarios Generation Method for STOPF Problem

This paper introduces a Monte Carlo scenario generation method based on copula theory for the stochastic optimal power flow （STOPF） problem with wind power. By using copula theory, the scenarios are simulated from multivariable joint distribution but only from their dependency matrix. Hence, the scenarios generated by proposed method can contain flail statistical information of multivariate. Here, the details of simu- lating scenarios for multi-wind-farm are explained with four steps： determine margin of one wind farm, fit the copulas, choose optimal copulas and simulate scenarios by Mote Carlo. Moreover, the producing process of scenarios is demonstrated by two adjacent actual wind farms in China. With the scenarios, the STOPF is con- verted into the same amount deterministic sub OPF models which can be solved by available technology per- fectly. Results using copula theory are compared against results from history samples based on two designs： IEEE 30-bus and IEEE 118-bus systems. The comparison results prove the accuracy of the proposed methodology.

Isotopic Composition of Atmospheric Precipitation in the Cis-Ural Region

Using the Global Network of Isotopes in Precipitation(GNIP)method at three stations in the Cis-Ural region(Iskor,Perm,and Kungur),the monitoring results of the isotope composition of atmospheric precipitation conducted in 2016-2019 are reported.Local lines of meteoric waters were obtained for each station and for the Cis-Ural region as a whole.The latitudinal zoning of changes in isotope parameters from north to south in the Cis-Ural region is noted,which is related to meteorological parameters.The northern regions are characterized by isotopically light values compared to the southern regions.Additionally,the correlation relationship between the isotopic compositions of atmospheric precipitation and meteorological parameters at each station is estimated.Correlation analysis data show that about 50%-90%of variations in the isotopic composition of precipitation within the annual cycle are determined by the air temperature.At the same time,the amount of precipitation shows a relationship between the average strength and the isotopic composition of precipitation for a series of observations at Kungur Station;for a series of observations at Iskor and Perm stations,there is no correlation.Differences in the relationship with meteorological data are explained by different sources of moisture introduction to the station and,accordingly,the formation of the isotopic composition of precipitation.The flow of moisture is controlled by the transfer of air masses in different directions,the formation of cyclones and anticyclones.So,for the northern part of the region during the year is characterized by the predominance of cyclones,whereas for the southern-anticyclones.