The variation features of the Antarctic sea ice (Ⅱ)

In this paper, on the basis of the Antarctic sea ice data from 1972 to 1989 issued by the America JointIce Center, the distribution features of the Antarctic sea ice is analyzed, the net sea ice area indexes are calculated,and the long-range variation periods of the sea ice area index are analyzed with the maximum entropy spectrum, finally the distribution pattern of the Antarctic sea ice and its variation features are obtained.According to its spatial distribution feature, the Antarctic Sea ice is divided into three large regions. Region Ⅰ(0°～120°E) is a zonal area which includes the Prydz Bay area, and sea ice area extending from the Weddell Sea,Region Ⅱ (120°E～120°W) mainly includes the Ross Sea area, and Region Ⅲ (120°W～0°) mainly the WeddellSea area. Of all the regions, the ice area in Region Ⅲ is the largest, and that in Region Ⅰ is the smallest.In the Antarctic,the seasonal changes of the sea ice are very obvious, during summer, in February, there isleast sea ice in the Southern Ocean, the net sea ice area (not include the area of open water) is about 3 190 000 km2,during winter, in September, there is most sea ice in the Southern Ocean, the area index is about 16 840 000 km2,nearly 5. 3 times of that in February. The seasonal change of sea ice is one month lag of the changes of the air temperature, but almost synchronous with that of SST.Of all the three regions divided above, there are some points both common and uncommon in their sea icechange cycles, the common features are that there exist one and a half years, one year and ten-months secondaryperiods in all three regions, but their main periods are not the same, they are about 5, 11 and 6 a in Regions Ⅰ,Ⅱ,and Ⅲrespectively. The main periods of the sea ice change in Regions Ⅰand Ⅱ are very close because the ice areaextended eastward from the Weddell Sea ice area of Region Ⅲ is one of the main components of the Region Ⅰ. It isalso worth pointing out that in Region Ⅱ, southward extension of the Pacific Ocean, there exist not only a 11-yerasmain period but also a 2-yeras secondary period, which does not exist in the other two regions.

Variations and Influence of Cloud Cover in Heze Region in Recent 52 Years

Based on the data of cloud cover, precipitation, temperature, sunshine hours and relative humidity from nine ground meteorological stations in Heze region in the southwest of Shandong Province from 1961 to 2012, changes of total and low cloud cover and its relationship with climatic factors associated in the southwest of Shandong Province in recent 52 years were analyzed. The results showed that average total cloud cover in- creased by 0.89%/10 a, but average low cloud cover decreased by 1.1%/10 a in Heze region in recent 52 years. The positive correlation between the average total cloud cover and temperature in autumn and winter was obvious, that is, when cloud cover increased by 10%, the average temper- ature increased by 0.48 ~C in autumn and increased by 0.83~（3 in winter. The average low. cloud cover negatively correlated with the average tam- perature in each season, and the negative correlation was very significant in spring. When cloud cover increased by 10%, the average temperature decreased by 1.49 ~C. The positive correlation between the average cloud cover and average precipitation was significant. The annual precipitation increased by 148.1 mm when annual mean total cloud cover increased by 10%. When seasonal mean cloud cover increased by 10%, the precipita- tion increased by 48.4, 107.1,55.4 and 12.2 mm in spring, summer, autumn and winter respectively. The annual average total cloud cover and low cloud cover had significantly positive correlation with 〉~0.1, ~〉1.0, ~〉10 and ~〉25 mm precipitation days respectively. The sunshine hours were seriously influenced by cloud cover, and when cloud cover increased by 10%, the sunshine hours decreased by 54.5 h in spring, 134.2 h in sum- mer, 154.3 h in autumn and 60.6 h in winter. The total cloud cover significantly positively correlated with relative humidity in summer and autumn, and when cloud cover increased by 10%, the relative humidity increased by 3.3% in summer and 4.1% in autumn.

Change Features of GPS/MET Precipitable Water Vapor in Different Precipitation in Lianyungang

Based on the GPS/MET water vapor monitoring data and conventional meteorological data at Lianyungang station from April to July, 2015, the relationship between precipitable water vapor and real precipitation was studied. According to different precipitation, change trends of precipitable water vapor in convective precipitation and steady precipitation were analyzed. Results showed that necessary condition of precipitation generation was high precipitable water vapor value in the air. Precipitable water vapor change presented wave-shape and phased characters. In convection precipitation, precipitable water vapor changed frequently and had larger change amplitude, while its change was slow in steady precipi- tation. The appearing time of the maximum values of rainfall intensity and precipitable water vapor was not necessarily consistent, but it was known that severe rainfall usually began at the high-value stage of precipitable water vapor, and high-value stage of precipitable water vapor often corresponded to higher precipitation probability. In addition, precipitable water vapor showed different characteristics in the above two different precipitation, and these results could provide a reference for precipitation forecast.

PRODUCT IMAGE RETRIEVAL BASED ON CO-FEATURES OF THE OBJECT

In this paper, we propose a product image retrieval method based on the object contour corners, image texture and color. The product image mainly highlights the object and its background is very simple. According to these characteristics, we represent the object using its contour, and detect the corners of contour to reduce the number of pixels. Every corner is described using its approximate curvature based on distance. In addition, the Block Difference of Inverse Probabilities (BDIP) and Block Variation of Local Correlation (BVLC) texture features and color moment are extracted from image's HIS color space. Finally, dynamic time warping method is used to match features with different length. In order to demonstrate the effect of the proposed method, we carry out experiments in Mi-crosoft product image database, and compare it with other feature descriptors. The retrieval precision and recall curves show that our method is feasible.