Flood modeling with the fuzzy SMRGT method and an example of the Kalecik Basin

Flood has been one of the too difficult and complex hydraulic and hydrological problems for human beings to take precautions among the natural disasters since the past. It is very difficult and almost impossible to detect earthquakes in advance with today's technology. However, it can be said that it is easier to predict floods than earthquakes. In this study, the flow coefficient and flood flow rate of Kalecik basin were determined. There are quite realistic models prepared using various methods for this purpose, With excessive urbanization, the increase in reinforced concrete structures and, accordingly in the roof covers, the replacement of stabilized roads with asphalt roads, the increase in the use of cobblestones and their derivatives, the narrowing of river volumes due to settlements built on the riverbanks increases the amount of precipitation that flows into a basin or region, in other words, the flow coefficient. The increase in the flowing part of the precipitation can be considered as the cause of human-induced floods. In addition to all these, considering the increase of extreme meteoerological events (especially torrential rains) with the effect of global climate change and the temporal and spatial changes of these extreme events, flood risk is increasing for almost every region and basin. Floods are situations where natural events that seriously damage the housing, infrastructure, superstructure, agricultural lands and the environment around them, depending on the flow rate, turn into disasters due to human origin. That is why it is necessary to calculate and analyze the relationship between precipitation-flow uncertainty and the amount of precipitation that passes into the flow very accurately. As a matter of fact, the hydrological calculation of the flow will be of great benefit both for the prevention of possible floods and for the dimensioning of the hydraulic structures to be designed. The rain-flow relationship is evaluated by the flow coefficient. This relationship, which is indispensable in the determination of floods but contains uncertainty, is determined by various statistical, stochastic, deterministic and black box methods in the current literature. Black box or fuzzy models may be preferred instead of using deterministic methods in natural events involving uncertainty. However, black box methods often do not take into account the physical aspect of the event. For this reason, it is useful to use methods that take into account the physical cause-effect relationship of the event. In this study, fuzzy logic was preferred because the flow coefficient used in the calculation of the flood flow rate also reflects the physical cause-effect relationship of the event. Thus, both the uncertainty and the physical aspect of the hydrological event were not ignored. Two points to be considered in fuzzy modeling are to correctly determine fuzzy sets and fuzzy rule base. Hence, SMRGT, which is new and has the opportunity to determine both together, has been preferred. Kriging method and Digital Elevation Model (DEM) were used for the maps of the basin. Meteorological, geomorphological and land use-related features were taken into account to determine the flow coefficient. Although the model has been applied to the Kalecik Basin, it can be easily used for all basins if the limit values of the variables are expanded. Then, the flow coefficient found was used to calculate the flood flow rate and to determine the risk map. Rational Method was also taken into account in flood calculations. The dissertation consists of six main chapters in total. In the introduction, general information about the subject Flood is given. National and international studies related to flow coefficient, global climate change and flood are given in the second part, methodology is given in the third part, models, calculations and maps developed within the scope of the thesis are given in detail in the fourth part. In the fifth chapter, conclusions and recommendations are presented. The reference list is included in the last section.