Frequency analysis is one of the most useful methods of statistics science for estimating the discharge values in which the probability of occurrence of a particular discharge with a certain return period is investigated. The major problem in this regard is the lack of statistics or short-term statistics, which is used to solve this problem by the regionalization technique in flood frequency analysis. Several methods for regionalization are used, most of which are index flood method (IFM).Scaling are used to examine the IFM method. In this research, the effect of trend on scaling was studied. 96 hydrometric stations were selected and stratified stations with no trend were separated by the Mann-Kendall test.Then scaling was done on peak discharge in two data group. The scaling method was performed using the probability weight moment and area of the basin on peak discharge. The results showed that the station with trend has a multiple scaling and station without trend has simple scaling. In other words, the scaling at the station with trend has changed from simple to multiple due to the existence of the trend. This article demonstrates the effect of the trend on the spatial scaling parameter. Therefore, it can be concluded that in the Caspian basin, it is not possible to use the IFM method to estimate the discharge in ungaged areas. This is, of course, the case for stations with trend. For stations without trend, the IFM method can be used. Key words : Flood Frequency Analysis, Spatial Scaling Method, Trend, Probability Weighted Moment, Peak Discharge, Basin Area.
