One of the main goals of this research is to evaluate the changes in intensity-duration-frequency (IDF) curves under the effect of climate change in the future to show the necessity of adopting new methods in urban flood management. There are two major drawbacks to generating future IDF curves using atmospheric general circulation models. The precipitation values obtained from the mentioned models, on the one hand, have very little spatial accuracy for use in urban areas, and on the other hand, in terms of temporal resolution, they are much larger than the concentration-time used in the design of urban flood control and conveyance structures. In this research, an RCM model is employed instead of GCM models for raising spatial accuracy, and a new time-downscaling algorithm is introduced to generate short-duration rainfalls from 24-hour rainfalls.
Materials and Methods
In this research, in the city of Rasht, as representative of the Semi-Mediterranean climate, a dynamic method through using the PRECIS regional climate change model was employed for spatial downscaling due to the need for high accuracy in urban areas. The PRECIS model data, like many other climate change models, is presented daily (24 hours) and the synoptic station data are available every 3 hours. Meanwhile, durations shorter than 10 minutes to 180 minutes are needed to produce IDF curves. Therefore, in this research, a special two-step method was used to convert daily rainfall into short-duration rainfall up to 10 minutes resolution. Frequency analysis is used for the historical/future short-duration rainfalls to evaluate the extremes of different return periods.
The PRECIS model’s 24-hour rainfall conversion factor to 3-hour rainfall was first obtained, and then this rainfall was multiplied by the second factor, which is used to convert three-hour rainfall to less-duration rainfall. One of the important issues for introducing appropriate boundary conditions to climate change models is whether to use the entire existing historical period or not. In this research, the IDF curves for the far future (2070-2100) were produced and with the same curves in the recent historical period (2003-2020), after ensuring the climatic jump that occurred in the year 2003, were compared. The results showed that for rainfalls with a short duration of 10 to 60 minutes, the intensity of rainfall increases significantly for all return periods.
one main point that has been emphasized in this research concerning predicting future events is to define a correct boundary condition, which in turn needs to examine the possibility of climate change in the historical period. If there is evidence of jumping in the historical data, it should be divided into two periods before and after the jump year, and the period after the occurrence of a change in the desired parameter, which is precipitation here, is used to produce down-scaling coefficients. The special result obtained in this research is although the intensity of rainfall has increased in all durations due to climate change, rainfall with a duration of less than 1 hour and a return period of up to 5 years will increase significantly. Considering that these intensity of rainfalls with the mentioned return periods are of particular importance in the design standard of urban flood control structures, revision of intensity-duration-frequency (IDF) curves to adopt new methods in urban flood management in all urban areas including the city of Rasht is essential.