مجله زمین شناسی مهندسی
سال دوم شماره 2 (پاییز و زمستان 1386)

One of the quality management methods of groundwater resources is assessment of pollution potential, where with regard to that the boundary of groundwater resources quality is being determined, and proper management for land efficiency in touch with aquifer could be applied. In present investigation, with regard to permeability of recharge zone in dominant part of the Baghmalek plain and use of considerable amount of agricultural fertilizers, pollution potential of groundwater has been assessed. In order to evaluate the pollution potential of groundwater in the area, AVI method, GOD and DRASTIC models have been taken into account. The collected results has indicated that in spite of sedimentary surfacial coarse grain distribution in relatively vast part of the area, the pollution potential is low in major parts of the area under investigation. The main influential factor for low pollution potential in tremendous part of the area is the fine and relatively thickness of vadose zone.

Flood management and flood prediction are taken into account for a long time. There are many methods to estimate the magnitude of the flood. One of the most recent methods is using Artificial Neural Networks. In this paper, a Neural-Wavelet Network (NWN) using wavelet theory and Artificial Neural Networks (ANN) for Halil river basin in SE of Iran is reported. Furthermore, a new rainfall-runoff pattern is reported. This pattern ncludes the classification of data in the harmonic data groups and the use of Neural-Wavelet Network.The introduced pattern is analyzed by the NWN model and results are prepared with Artificial Neural Networks Back propagation and Radial Base Function (RBF) model results. Halil basin was selected to used for our model. The calculations of the R-square and root mean square error (RMSE) can control the accuracy of our computations. The calculations showed that the accuracy of the results of the Artificial Neural Wavelet Network is more desirable than ANN and RBF. It is showed that the classification of data in harmonies data and using of new pattern increase the efficiency of the models operations as well.

One of the main reasons of destructions while happening earthquake and applying intense force is the brittle deformation of outer crust of the earth. Creep in the coastal sediments may happen by the virtue of existing cracks and fissures, unnatural movement and/or loss of the earth tolerance. This phenomenon in sandy areas may happen by increasing the pressure of gang ionic water and is called liquefaction. Studying the laboratory sample shows that the capability of liquefaction of sandy deposits can be decrease by flattening the land. One of the ways to improve the soil to increase the tolerance while applying dynamic loads is dynamic compaction. The design and usage of dynamic compaction usually depend on experiment. Dynamic compaction is an active design which needs a revision and method reformation in progress according to the reaction of the earth against the hit. In 12 parts of ship manufacturing and repairing factories bed in the Iran ship building and offshore industries complex co. located in 40km away in the west of Bandar Abbas, Dynamic compaction has been used to prevent liquefaction phenomenon. While performing of water in pressing step are happened that the following approaches are suggested: 1- To empty the water of sinkholes by a pump, 2- To perform dynamic compaction in two steps with regarding the adequate time between the steps of applying hits. and 3-Sludge taking in the surrounding areas and using large grain mixture and finally if needed to do deep drainage in a large area.