The sources of large and unique platinum-metal placers were the zonal alkali-ultrabasic massifs becoming active in the Cretaceous time that gravitated to the periphery of the areal of the North-Asian Superplume ( NAS) influence． Origin of the conditions favorable for the formation of the placer platinum presence in the region is governed by the plume ore-generating magmatism of several generations ( in Proterozoic and Phanerozoic) ． Ultramafite-mafite magmatism was responsible for the origin of the stratified zonal massifs subjected in the Late Paleozoic and Mesozoic to the action of granitogene derivations of the NAS． The latter provided the development of the ore-forming magmagene-fluid-metasomatic processes and favorable conditions for the platinum group elements concentration．

Based on analysis of drilling，logging and field profile data，six sequence boundaries in Permian are identified in Bachu and Tazhong regions of Tarim Basin． All sequence boundaries are of type I sequence boundaries， and are characterized by down cut． According to the six sequence boundaries，the Permian in this area can be divided into 5 third-order sequences，and all the sequences correspond with classic sequence model of Vail． Sequence Psq4 indicates lake transgressive system tract ( TST) and highstand system tract ( HST) ． Sequences Psq1，Psq2，Psq3，Psq5 indicate low stand system tract ( LST) ，transgressive system tract and highstand system tract． LST is deposited by incised channel infilling with features of fluvial facies． TST is deposited by shore-shallow lake and semi-deep lake． HST is deposited by semi-deep lake，shore-shallow lake and delta． In addition，volcanic rocks are present on the top part of HST in sequence Psq3． Incised channel infilling and deltaic deposits were mainly distributed on western slope of Tadong uplift．

GC-GIS system is a geochemical data processing system based on fractal theory． The system realized quantity statistics function by calling Surfer and MapInfo software，and it is compiled with Visual Basic language． This system is designed to integrate the functions both quantity statistics of Surfer and spatial data management of MapInfo． A new algorithm of fractal is added up to GC-GIS． Taking example for Weichang region of Hebei to test the system，the processing results show that the model can match the real distribution of mine well．

Taizishang slope of Qingjiang Reservoir in Huibei is within the range of atomization rainfall． The authors created the finite element model with ANSYS 10. 0 and used strength reduction FEM to calculate thesafety coefficient of the slope on conditions with no rain，and theatomization rainfall lasting for 0. 2 hours and 0. 4 hours． The results show that the atomization rainfall would decrease the stability of the slope，and with rainfalls increasing，the stability would reduce much more．

The coastline changes along Yemen's the Red Sea ( Al-muka，Al-khohah，Al-tiaf，Ras Katib and Al- Urji spits) were studied using a series of landsat images ( MSS，TM and ETM + ，1972，1989，2000，2006) ， coupled with geomorphological，sedimentological and meteorological findings． Comparison of satellite images provided a viable means for establishing long-term coastal changes ( accretion and erosion) as observed in the studied spits ( Al-Urj，Ras Sham，Ras Maemoon，Ras Katib，Al-Mandar，Nukhaylah，Mujamilah，Ras Ashab Abu-Zahr and Mukha) ． The rate of the spit accretion has the greatest value up to 89 pixel /year corresponding to 72 290 m2 /year in Mukha，while the spit erosion shows greatest value up to 131 pixel /year corresponding to 106 404 m2 /year in Mujamilah． The patterns of accretion and erosion along the spits depend on the spit direction， natural processes mainly of wave-induced longshore currents，the sediments supply and depth of sea．

In recent years，the bridge safety monitoring has been paid more attention in engineering field． However， the financial and material resources as well as human resources were costly for the traditional monitoring means． Besides，the traditional means of monitoring were low in accuracy． From an engineering example，based on neural network method and historical data of the bridge monitoring to construct the BP neural network model with dual hidden layer structure，the bridge temperature field and its effect on the behavior of bridge deflection are forecasted． The fact indicates that the predicted biggest error is 3. 06% of the bridge temperature field and the bridge deflection behavior under temperature field affected is 2. 17% by the method of the BP neural network， which fully meet the precision requirements of the construction with practical value．

In order to decrease relative settlement，foundation treatment plays an extremely important role in bridgehead transition section，especially，the situation of building the bridge piles firstly，and then processing piles． On the basis of engineering practice，the authors analyzed the influence of foundation treatment on bridge piles in bridgehead transition section by finite-element method ( FEM) ． This research has positive significance in predicting displacement of bridge pile，directing construction of foundation treatment，and improving quality of engineering and so forth．

The technical characteristics and operational principal of the reverse circulation DTH hammer drilling technology is integrated introduced in this paper． Using simulation technology to assist the drilling tool design， the optimum designed DTH hammer gets stable performance，and the bit has reverse circulation ability，and further more，it has already developed serial products． The reverse circulation DTH hammer drilling technique has been fully applied in field such as the geological core exploration in complex strata，water well drilling，and large diameter rock embedding piles drilling in hard rock，and obviously，it is achieved favorable economic and social benefits．

This paper presents a nonlinear multidimensional scaling model，called kernelized fourth quantification theory，which is an integration of kernel techniques and the fourth quantification theory． The model can deal with the problem of mineral prediction without defining a training area． In mineral target prediction，the pre-defined statistical cells，such as grid cells，can be implicitly transformed using kernel techniques from input space to a high-dimensional feature space，where the nonlinearly separable clusters in the input space are expected to be linearly separable． Then，the transformed cells in the feature space are mapped by the fourth quantification theory onto a low-dimensional scaling space，where the scaled cells can be visually clustered according to their spatial locations． At the same time，those cells，which are far away from the cluster center of the majority of the scaled cells，are recognized as anomaly cells． Finally，whether the anomaly cells can serve as mineral potential target cells can be tested by spatially superimposing the known mineral occurrences onto the anomaly cells． A case study shows that nearly all the known mineral occurrences spatially coincide with the anomaly cells with nearly the smallest scaled coordinates in one-dimensional scaling space． In the case study，the mineral target cells delineated by the new model are similar to those predicted by the well-known WofE model．