پديد آورندگان :
حسين زاده، سيدرضا نويسنده دانشيار گروه جغرافيا، دانشگاه فردوسي مشهد , , خانه باد، محمد نويسنده دانشگاه فردوسي مشهد Khanehbad, M , خسروي، عذرا نويسنده كارشناس ارشد ژيومورفولوژي، دانشگاه فردوسي مشهد ,
چكيده لاتين :
Introduction
with a linear pattern Kalat city is been developed along of main channel stream of Kalat River where affected by large flooding and river bed changes in the process of urban development have increased the severity of the floods. The Flood which occurred in September 2001 is an example of the devastating flood that caused significant damaged the rice farmlands ,infrastructures and houses of the city.
Accurate estimate of the maximum flood level and its recurrence intervals play an important role in the urban flood planning and is considered as the main phase in planning and urban flood risk management.
Conventional methods for flood frequency analysis are based on the recorded data of gage stations while these records do not have enough length for occurrence intervals . In other words, most estimates of statistical methods are conducted by short-term data for long periods. For example, a 30-year statistical period is used for flood estimate with a 500-year return period. This is a major mistake in flood discharge estimations by using statistical methods.
During the last three decades in moden geomorphology approach, slack water deposits(SWD) and paleo stage indicators (PSI) provided natural data for flood intensity, level and discharge estimations. Paleoflood hydrology is the knowledge of past large floods based on study flood data before they are recorded by hydrometric stations.
Paleofloods are past or ancient floods that occurred without being recorded by either direct hydrological measurement during the course of operation or observation and/or documentation by non-hydrologists. Type recordings are historical floods. The recordings of paleofloods are natural, not human, and this distinguishes them from instrumental/systematic measurments. In contrast to direct human observation, the natural recordings of paleofloods occur via a variety of indices that are interpreted by the experienced paleoflood hydrologists (Baker, 2008).
Paleoflood term was first prominently used in the 1970ʹs (Costa, 1974, Patton and Baker, 1977), but the term and concept of paleoflood hydrology officially introduced in 1982 by Kuchel and Baker (Jehadi Toroghi and Hosseinzadeh, 2013: 137).
Material and Methods
In this research we have used a different set of data included flood data in the Klat gage station which is located in the main stream channel before Kalat City, satellite imageries, areal photos, 1:50000 scale topographic maps and 100000 scale geological maps and ASTER Digital Elevation Model (DEM).
Stages of this research can be stated as follows:
1- Statistical analysis of maximum yearly flood discharges for Kalat gauge station to predict of river’s probable maximum flood in different return periods with variety suggested methods.
2- Detail study of the large-scale aerial photographs and satellite imagery to analysing of Kalat urbanisation impacts on stream channel pattern, geomorphological mapping of river bed and determining potential sites for paleoflood studies.
3- Field works to identify and select paleoflood evidences (slack water deposits and flood signs) and fully describing stratigraphy and survey the channel geometry and forms.
4- Hydraulic calculations and estimates of discharge.
5- Compare the obtained results by paleoflood methods with the data available from the Kalat gage station.
Results and Discussion
To prevent of flood damages, we need to estimate the probability of occurrence and magnitude of major floods and use appropriate methods and specific facilities to control the impact of flood hazard (Mahdavi, 2009: 262). In order to estimate the return period of floods, frequency analysis should be used. Frequency analysis methods are complex events using probability distributions to assess the likelihood of occurrence of a phenomenon over the time. These analysis are used to obtain the return period of measured events and estimation of its magnitude which is outside of recorded events amplitudes and used for project designing. The first step for frequency analysis is primary recognition of proper distributions for sampled data. For this purpose, peak discharge of floods in Kalat gage station were fitted with different return periods by using different statistical distributions. Based on Kolmogorov–Smirnov test, the generalised extreme value distribution method of maximum likelihood is the best method of distribution. Based on this distribution, floods with 1000-years return period were calculated 65268.9 m3/s, which is a very large amount for Kalat River as well as according to other distributions, this statistical distribution is very different and unrealistic. Thus, the comparison of twenty one statistical methods by distributing various statistical have been done to show that the test statistic K-S, in four ways: Method of Probable Weighted Moments of Gumbel type one, method of maximum likelihood of the log-normal 2-parameter, Method of Probable Weighted Moments in generalised distribution, and Method of Probable Weighted Moments in Pearson type III distribution are almost similar. The maximum measured flood with 4 mentioned methods, in return periods above 100 years and even in return period above 10 years upwards show a large differences between each other. Because of the short duration of the study station data, the measured floods for return periods above 100 years are not accurate enough. So according to the above results and current methods, the exact determination of a function is not free from error.
To study and analysing of paleofloods, primarily sediment deposition sites should have favourable geomorphic conditions for the accumulation of sediments of slack water deposits, among which we can mention the following conditions: Downstream of rock outcrops, rock alcove, shelters and caves along the shallow bedrock channel sides, the confluence of the main tributaries, broadening the channels (Benito et al., 2013, 461) and other regions of the vortex flow disruption.
According to the mentioned criteria in the selection of suitable sites of slack water deposits, after field studies and surveys, three sites were selected. After detection of sites for calculation of maximum flood discharge and water level, based on the top of slack water sediments related to paleofloods and evidences of recent flood marks, we assessed and modelled flood discharges and hydrologic behaviour of the river. For this purpose, first we mapped the channel cross sections of the river path, and then calculated the flood discharge using the modified Manningʹs formula. According to our calculations, the maximum flood discharge is 540 M3/S for site 1, 554 M3/S for site 2, and 624 M3/S regarding to flood marks. the estimated discharges in both sites are similarity because no any tributary channels enter into the selected reach for this study and it resulted the accuracy of measurements.
Conclusion
in hydrologic projects, engineers usually use simple statistical data derived from available short records or other published materials for their studies. researchers with an statistical approach apply many statistical methods for peak flood discharge estimation, we have selected 21 methods in this study and results were compared in return periods of 2 to 1000 years. The results show that statistical methods are not very acceptable for frequency analysis and long-term return period calculations where short-term hydrometric records organize the basic date for analyzing process. Therefore, the hydro-statistic methods for flood management projects, as well as the determination of stream channel dimensions throughout urban area must be revised. Based on paleoflood analysis, a large part of the city is under flood risk, mostly at the city entrance and Darband area, which must be given enough attention in the future planning. due to the lack of long-term hydrometric statistics and measuring tools, paleoflood investigations are the best method to obtain the real data for long periods of time .
