J Cave Karst Stud 81 2 — Tectonophysics — Boroujeni B, Ashjari J, Karimi H Geological and hydrological effective factors in the high permeability zone of several dam sites of the Zagros region. Iran J Cave Karst Stud 81 1 :9— J Acta Carsol — Pergamon, London. Geoscience 9 7 — Deere D, Miller R Engineering classification and index properties for intact rock. Technical report no.
Dickson JAD Carbonate identification and genesis as revealed by staining. J Sediment Petrol 36 2 — Unwin Hyman, pp Ghabezloo S, Pouya A Numerical modelling of the effect of weathering on the progressive failure of underground limestone mines. Book Google Scholar. Ghobadi MH The influence of aperture of joints on solubility of carbonaceous rocks. In Proceedings international symposium on engineering geology and the environment, pp — Environ Geol — AAPG Bull — Lashkaripour GR, Rastegarnia A, Ghafoori M Assessment of brittleness and empirical correlations between physical and mechanical parameters of the Asmari limestone in Khersan 2 dam site, in southwest of Iran.
J Afr Earth Sci — Pet Sci — Milanovich PT Karst hydrogeology. Rotated Varimax raw factor loadings for Sinjsko polje and 5c is observed.
The elements indicate pollution by phos- neighboring area on a 5! This grid encompasses Fe 0. Bearing in Sr P0. Hence, there is virtually tent of hydrolysis reactions Retallack ; the ratio in- no possibility of finding Mg in the terra rossa and brown dicates the degree of soil draniage, leaching and the time soils during the long-term process of weathering of the of development in soil profiles. Other weathering ratios carbonate rocks. Maps presented in Fig.
It should neate the polje itself and lowest part of the polje floor also emphasize a marked difference between a wide river from its surroundings. The soils terra rossa-luvisols and valley and carbonate neighborhood. This possibility is much large cations K c, Ba 2c provide charge balance on sur- enhanced if a closer look is taken at the spatial distribu- faces and in interlayer sites of clays, and the smaller ca- tion of Pb-Zn-P association on the F3-factor-score map tions Ca 2c, Na c, and Sr 2c are quantitatively removed Fig.
Comparison with the pedologic map of the re- from percolating waters. On the slopes encircling the gion Fig. The high area. The soil samples from the surround- ample, can be effectively used to differentiate karst polje ing area obviously contain a much higher concentration by geochemical means, further investigations of other of Cu, which bonds together with Sr and Mg in a rela- large poljes in Croatia are to be made to confirm these tionship that is not characteristic for the soils developed observations.
Moreover, it should be noted that Cu and Sr make one set of variables that is inversely re- lated to Mg standing alone. Again, the two geochemically contrasted regions are markedly differentiated as can be Conclusions seen from the F4-factor-score map Fig.
The low fac- tor-score values outline the alluvial Cetina valley, particu- The following conclusions can be drawn from the geo- larly the narrow tract along both riverbanks. This is a chemical characterization of Sinjsko polje by means of belt where Cu and Sr is deposited and concentrated in al- factor analysis. On the 1. Within the polje three factors were estimated.
It shows itself to be somehow extraneous to the karstic background. They are largely concerned with an association between one set possibly derived from the sulphates of a mixed origin. Cu of variables with Zn, Ni, Co, Fe, V, Cr, Ti, and Al op- is very probably anthropogenic and its distribution could erating against Sr and Ca in the other set, which rela- be ascribed to the use of copper vitriol in agricultural ac- tionship must be a mirror of the presence of anhydrite tivity vineyards , while Sr is most likely of natural ori- and gypsum outcrops.
The second factor F2 relates a much simpler geo- anhydrite. O Fig 6a—d 2. Four factors are estimated to be significant for the ex- Factor-score maps of Sinjsko polje and neighboring area — planation of the geochemical structure of data assessed 5! Some of the tors F1, F2, and F3 as in the factor matrix pertaining data used are from the General Geochemical Map of Croatia to the finer grid, although there are some very signifi- project funded by the Ministry of Science and Technology of cant contrasts.
The authors would also like to thank a. The first factor is additionally loaded with Ba, which all of the Institute of Geology staff in Zagreb, who were in the is to say that contribution of this element is somewhat field for sampling during the 2 years of field study in the area, and Dr. Durn from the Faculty of Mining-Geology and Pe- greater from soils not in direct contact with alluvial troleum Engeneering, University of Zagreb for his sugestions sediments of the Cetina river.
The second factor of the coarser-grid factor model distinguishes the absence of Mg with regard to the previous one. Elemental composition of the third factor in the fac- References tor model of 5! The fourth factor F4 has not been met as the out- and minerals, — Survey Canada Paper , come of factorizing the data pertaining to a finer grid Ottawa of the Sinjsko Polje valley and can be used as one Colak A, Bogunovic M Pedologic Map of Yugoslavia, which delineates polje from its surrounding area.
Wiley, differences in drainage patterns which clearly delineate New York the polje itself and lowest part of the polje floor from Dillon WR, Goldstein M Multivariate analysis: meth- its surroundings. Unwin Hyman, London Geochemical mapping of carbonate terrains.
Bull Inst Gams I The polje: the problem of definition. In: Herak M, Stringfield ment contamination of soils in a karstic polje — an example VT eds Karst — important karst region of the northern from Sinj polje, Croatia. Geol Croat 48 : 67—86 hemisphere. Basil Blackwell, Lon- dology. To introduce tidal boundary conditions, the canal and sea surface water bodies within modelling transects were simulated explicitly using the high-K approach discussed by Mulligan et al.
In this approach, large hydraulic conductivity values are assigned to surface water regions. Hydraulic conductivity values were manually calibrated using numerous day tidal simulations with tidally varying boundaries. Representative model values for hydraulic conductivity were determined by adjusting parameters, until simulated amplitudes from the model matched with observed amplitudes Dausman and Langevin A simplified zone continuum model Langevin was developed to define the representative hydraulic conductivity distribution within the domain.
Within the gypsum layer, five discrete zones of different permeability were identified based on core data. To the west of the canal, the first gypsum layer and a m buffer of the canal were defined as the most permeable zones. These zones are characterized as having the highest density of variously oriented fractures and karst features.
To the east of the canal, a uniform permeability value was set for the gypsum layer due to insufficient stratigraphic information.
Different sets of hydraulic conductivity values for the five zones were tested by the day tidal simulations until agreement between field and simulated data was achieved in terms of heads and TDS. The agreement was measured by the root mean square error RMSE. The best fit was obtained for the hydraulic conductivity values shown in Table 2. Finally, porosity was kept constant throughout the simulation because the feedback mechanism involving dissolution, porosity changes, flow and concentration field modification cannot be approximated in numerical codes.
Because of the LEA, the initial water composition must also be in chemical equilibrium. Since the reactive-transport model is based on the transient flow field resulting from the SEAWAT tide-influenced simulation, the salt concentration distribution at the beginning of this simulation is the linkage between the second and the third modelling phase in terms of chemical species distribution.
The chemical composition of groundwater at the beginning of the tide-influenced simulation is variable from cell to cell, in proportion to cell salinity. However, to simplify the conversion of salt concentration into chemical species distribution and the related setting of the initial chemical equilibrium, the salinity distribution was approximated by eight discrete zones with uniform salinity values.
To achieve saturation conditions, chemical equilibrium modifies both aqueous species and mineral concentrations. For example, anhydrite can be completely dissolved in favour of gypsum. The activities of the adsorbed species on exchanger sites, resulting from chemical equilibrium, are also entered into PHT3D as initial conditions, and they are involved in ion exchange reactions during the simulation. For boundary conditions, the hydrochemical composition of water entering the model domain during the reactive transport simulation is kept constant during the simulation and given by the initial concentrations at the first inland column, the canal, and the sea.
Results The results from the simulation phases were compared with experimental data collected during monitoring activities. The observation wells defined within the model domain are s3pz, s21pz, s22pz and s40pz Fig. Since the observation wells of the monitoring network are long-screened wells, they are simulated as observation boreholes screened over the model layers.
The associated simulated head value is the average of the calculated hydraulic heads of the screened layers, to approximate the water table location with an average RMSE of about 0. The simulated heads from the equilibrium model are in good agreement with the average water table elevation along the model transect. The simulated salinity distribution from the equilibrium model is also in good agreement with the average salinities along the model transect.
For these reasons, the first phase results are deemed as suitable to represent the initial conditions for the second phase simulation. For the tide-influenced model, the simulated groundwater level fluctuations in observation wells were compared with the monitoring data for the entire simulation period about 1 year.
The signal phases match and the greater amplitude differences, particularly for s3pz, can be explained by local flow conditions such as empty cavities that are not explicitly modelled. The observed TDS concentration distribution is also satisfactorily reproduced by the model. The logs do not refer to the same time, because the chemical survey was carried out from September to December , whereas the simulation period was between February and March However, the comparison refers to periods of the year with similar climatic conditions.
For the inland well s27pz , the simulated transition from freshwater to saltwater is offset by about 5 m. The last validation test performed for the tide-influenced model concerns groundwater velocities. Simulated Darcy velocities registered at the instant of two extremes of a tidal cycle were compared to logs of the groundwater velocity for low and high tide.
The measurements were taken by Fidelibus et al. At distances from the canal similar to those of the boreholes, the comparison shows that the order of magnitude and the behaviour during low and high tide are similar.
There are slight differences in the shape of the velocity - depth graph that can be attributed to local flow peculiarities not present in the model. The assumption of the chemical composition of water being constant in the canal during the simulation produces a plume effect below the canal itself.
Chemical species from the canal spread into the aquifer throughout the highest permeable zone maintaining the same concentration they have in the canal water. This is clearly shown for calcium in Fig.
The concentration distributions over time show the evolution of the plume with concentrations exceeding the surrounding ones. The plume then results from the combination of the assumed permeability field and the advective flow field influenced by the canal, which has a calcium concentration higher than the nearby groundwater. This is demonstrated by the corresponding non-reactive simulation whereby transported species can be considered as non-reactive tracers since the same effect was found in the final results, even if at a slightly lower degree.
To explain the final calcium concentration distribution, geochemical processes must be considered. Additional calcium below the canal can be ascribed to calcite dissolution with the maximum rate in the canal surroundings, which was confirmed by a similar excess in the final carbon concentration, which is representative of bicarbonates.
Results were obtained for solid phases i. As mentioned, calcite dissolution Fig. Additionally, a chemically reactive mixture is more likely to occur near this area than elsewhere. This is in agreement with the results of Rezaei et al. Model results for the variation of gypsum concentration Fig. Usually the concentration of sulphate in groundwater circulating in a coastal aquifer is relatively high, but the concentration of calcium is too low to produce gypsum precipitation unless concurrent geochemical processes occur.
Their laboratory experiments and a multicomponent reactive transport model where local equilibrium is assumed prove that gypsum precipitates during the first stages of seawater intrusion, causing a decrease in sulphate concentration. In the case of the Lesina Marina aquifer, cation exchange does not play such a crucial role because calcium concentrations at exchanger sites Fig. However, additional calcium comes from calcite dissolution and makes gypsum precipitate in the zone close to the canal during canal water intrusion.
Moving away from the canal towards land, the gypsum precipitation decreases the amount of sulphate transported by the intrusion front. The entire concentration range for calcite from The entire concentration range for gypsum, of Calcium and sodium distributions at exchanger sites Fig.
Generally calcium displaces sodium in the sediment exchange complex towards the freshwater flow domain. However, calcium excess causes partial sodium substitution also below the canal, although to a lower degree. A non-reactive run confirms that cation exchange processes play a marginal role. This is a coherent result since calcium and sodium concentrations at exchanger sites in the reactive simulation are one to two orders of magnitude lower than the concentrations in groundwater.
The upper limit of the variation range for calcium and sodium concentrations at exchanger sites is one order of magnitude lower than the respective minimum concentration in the groundwater see calcium in Figs 9 and The black circle indicates the observation points in the freshwater flow domain where the model slightly overestimates salinity.
This overestimation leads to a corresponding overestimation of chemical species concentrations leading to greater calcium and sulphate concentrations than those observed. The simulated concentrations for these points are overestimated for calcium and underestimated for sulphate.
Agreement was measured for the RMSE with values of The equivalent comparison between the data simulated by the non-reactive model and the observed data Fig. Therefore, the reactive processes mainly dissolution in the transition zone make the concentrations for both calcium and sulphate increase.
The agreement worsens for calcium, which was overestimated in the conservative simulation, and it improves for sulphate, which was underestimated in the conservative simulation. Still, with the exception of circled points in Fig. Subsea karst may develop in limestones carrying drainage from adjacent land and may also include features inherited from erosion during past A.
Waltham and P. The major engineering hazard is the downward washing of soil into old and stable rock voids to create failures. Karst morphology Karst has an infinitely variable and complex three-dimensional suite of fissures and voids cut into the surface and rock mass of the limestone Fig.
Surface, rockhead and underground landforms are integrated within karst systems, but fall into five broad groups of features Lowe and Waltham, : Surface micro-features - karren runnels, mostly A. Karst types Surface macro-features combine to make the distinctive landscapes of karst. Caves in karst Caves form in any soluble rock where there is an adequate through flow of water. Viral classification, - University of Anbar lecture-Viral Classification Classification of malocclusion Classification of malocclusion.
Classification of Forests in Zambiazmb-nfms. Immunohistochemical classification of breast. Immunohistochemical classification of breast tumours. Chapter 5 : Soil Classification Ch:5 Classification of. Ch:5 Classification of Soil Classification 5. Plant Classification Segment 1. Plant Classification. Classification of Stars - ps. Classification of Electrocardiographic P-wave Morphology
0コメント