Document Type : Full length article
Authors
1
PhD Candidate in Climatology, Department of Geography, Faculty of Literature and Humanities, Razi University Kermanshah, Iran
2
Associate Professor, Faculty of Geography, University of Tehran, Iran
Abstract
Introduction
Selection of the clear and transparent index for the precipitation using long-term homogeneous data is an important point for the researchers. Several investigations have led to different indices for heavy rainfall. In some cases the specific amount of precipitation was used for heavy rainfall (Rahimzadeh, 2005; Masoodian, 2008; Kamiguchi et al., 2006), e.g. Alijani (2002) has suggested the precipitation more than 30 millimeter. Some investigators have used the percentage of daily precipitation as a heavy rainfall index (Mohammadi and Masoodian, 2010), e.g. Easterling et al. (2003) used the greatest annual 5-days total precipitation amount and the percentage of annual precipitation, due to all 24-h rainfall totals exceeding the 95th percentile of daily amounts. A series of international workshops have introduced a set of indicators to show the effect of climate change on extreme events (Folland et al., 1999; Nicholla and Murray, 1999; Manton et al., 2001). Some researchers used several indicators as an index for heavy rainfall (Seibert et al., 2005; Haylock et al., 2006; Haylock and Nicholis, 2000; Osborn and Hulme, 2002; Simonov et al., 2007; Vaidya and Kulkarni, 2007; Campins et al., 2006; Paddock et al., 2008; Kysely and Picek, 2007; Bukantis et al., 2010; Schmidli et al., 2002). For example, Hänsel and Matcshullat (2009) to study monthly trends of daily heavy precipitation indicators used 22 heavy precipitation indicators (HPI) that may be classified into the four groups “A”, “I”, “F” and “M”. “A” stands for average precipitation indicators like monthly precipitation totals and number of wet days. “I” comprises indicators measuring the precipitation intensity, like the SDPI (Simple Daily Precipitation Index) or the percentage of precipitation above the 95th percentile. The frequency of heavy precipitation events is studied by indicators in class “F”, while category “M” includes indicators of heavy precipitation events magnitude. Zhang et al. (2001) proved that annual and seasonal time series of heavy event frequency are obtained by counting the number of exceedances per year. Characteristics of the intensity of heavy precipitation events are investigated by examining the 90th percentiles of daily precipitation, the annual maximum daily value, and the 20-yr return values. Based on the results, uses of percentile indicators are more common compared with threshold indicators and in some studies both the indicators have been used. It seems that the use of heavy rainfall partly depends on the geographical characteristics of the rainfall region. The natural ecosystems adapt themselves with the annual precipitation and extreme events in every region over time. Thus, the amount of precipitation shows the heavy rainfall in a dry station and in a humid station it can be recognized as normal. This study tries to find a simple method to indicate the heavy rainfall with regard to monthly trends based on daily data.
Methodology
To determine an index for heavy precipitation, data of daily precipitation for 40 stations with synchronized meteorological data are distributed homogeneously throughout the country in periods 1961-2011. The probability (1, 5, 10, 20 and 50%) for the entire period of rainy days was calculated using the Weibul equation. A very high percentage of daily precipitation values were obtained with the test 1 percentage, so the occurrence of five percent of daily precipitation was used as an index. The relationship between the ratio of the total mean annual precipitation (mm) and number of days with precipitation equal to or greater than one millimeter with a numerical coefficient may provide the best indicator for the heavy rainfall. Factor analysis of these two components can be selected from among the eleven factors of rainfall data including a total of 86 percent. Finally, the isohyet map was plotted using the numerical index by GIS so the heavy rainfall could be calculated for each part of Iran.
Results and Discussion
To determine the appropriate numerical factor in Iran, all the stations are classified into seven groups using K means cluster analyses, because of the different geographical characteristics and rainfall patterns. The average total annual rainfall was used to classify the groups. Then, the numerical coefficient was separately calculated for each group.
Conclusion
According to the proposed heavy rainfall index, the isohyet map was plotted. The isoline map of numerical coefficient was calculated for each station or related area in order to estimate heavy rainfall. The average error between the proposed index and the five percent probability of daily precipitation is 0.07. Only in Ardabil, Urumia, Dezful and Chabahar Port the error is more than one millimeter. There is no error in Ahvaz and Isfahan, i.e. the proposed index is equal to the five percent probability of daily precipitation. The comparison between the heavy rainfall isohyet map and the total average annual precipitation and the number of days with precipitation equal to or greater than one millimeter in Iran shows the same distribution.
Keywords