Spatial and Temporal Variation of Rainfall in Ninava Governorate

Rainfall was studied in four locations in Ninava Governorate which are Mosul, Sinjar and Talafar and Rabea for the period (1974 – 2002). The Mean Monthly, Seasonally and Annual values of Rainfall, Standard Deviation, Coefficient of Variation and Time Series for rainfall was studied during there periods in all locations. Simple and Multiple correlations were found between mean monthly values of rainfall and other Metrological elements (Temperature, Spatial and Temporal Variation of Rainfall in Ninava Governorate. 99 Relative humidity, Atmospheric pressure, Sunshine, Evaporation, Gloudness). Frequency distribution of rainfall data were also calculated. The results indicate a simple negative trends in the Time Series of the mean monthly values of rainfall in all locations, The maximum mean annual value of rainfall were found in sinjar location which reach (389) mm. The maximum frequency of rainfall in Oct., Nov., Apr., May were in the range of (0-30) mm in all stations while Dec., Jan., Feb., March show the maximum frequency in the range of (30-60) mm. Introduction Rainfall analysis is very important in different domains such as: Agricultural planning, Water resources planning, Runoff and Stream Flow Predictions soil Conservation studies, environmental studies, and human life activities[1]. The amount, intensity and areal distribution of rainfall are essential factors in many hydrologic studies[2]. Rainfall varies geographically, temporally and seasonally[3,4]. Bonifacis (2000)[5], He Study the variation of rainfall in philippines during the period (1930 – 1996), He found the 5-year and 10-year moving average and they use same statistical models to evaluate rainfall variability. Bleeg (2003)[6] study the variation of precipitation with time and space in Dohuk governorate and made a correlations between Rainfall and other meteorological elements. Dawood (2004)[7] study the time series of the monthly values of rainfall in different Meteorological stations presents in the north part of Iraq Grimald.(2005)[8] test the multivariate linear Parametric models applied to daily rainfall series, these simple models allow to generate synthetic series preserving both the time correlation (autocorrelation) and the space correlation (cross correlation). In Andes Wouter, (2006) [9] Study the spatial and temporal rainfall variability in mountainous areas using 14 rain gauges located in western mountain range of the Ecuadorian Andes. Odure (2006)[10] Take the data of the mean annual rainfall for the period (1961-1998) using data from 30 stations and he subjected the time series to a power spectrum using the max Entropy spectral Analysis technique in Ghana. Momani (2009)[11] use (Box –Jenkins) model to forecasting the monthly rainfall for the upcoming 10 years in Amman to help decision makers establish priorities in term of water demand management. Abaje.(2010) [12] Analyze the Rainfall trends in Nigeria using the rainfall data during the period (1974-2008). They found that the trend in the rainfall series is decreasing on the annual basis.


Introduction
Rainfall analysis is very important in different domains such as: Agricultural planning, Water resources planning, Runoff and Stream Flow Predictions soil Conservation studies, environmental studies, and human life activities [1].
The amount, intensity and areal distribution of rainfall are essential factors in many hydrologic studies [2]. Rainfall varies geographically, temporally and seasonally [3,4]. Bonifacis (2000) [5], He Study the variation of rainfall in philippines during the period (1930 -1996), He found the 5-year and 10-year moving average and they use same statistical models to evaluate rainfall variability.
Bleeg (2003) [6] study the variation of precipitation with time and space in Dohuk governorate and made a correlations between Rainfall and other meteorological elements.
Dawood (2004) [7] study the time series of the monthly values of rainfall in different Meteorological stations presents in the north part of Iraq Grimald.(2005) [8] test the multivariate linear Parametric models applied to daily rainfall series, these simple models allow to generate synthetic series preserving both the time correlation (autocorrelation) and the space correlation (cross correlation).
In Andes Wouter, (2006) [9] Study the spatial and temporal rainfall variability in mountainous areas using 14 rain gauges located in western mountain range of the Ecuadorian Andes.
Odure (2006) [10] Take the data of the mean annual rainfall for the period (1961-1998) using data from 30 stations and he subjected the time series to a power spectrum using the max Entropy spectral Analysis technique in Ghana. Momani (2009) [11] use (Box -Jenkins) model to forecasting the monthly rainfall for the upcoming 10 years in Amman to help decision makers establish priorities in term of water demand management.
Abaje.(2010) [12] Analyze the Rainfall trends in Nigeria using the rainfall data during the period . They found that the trend in the rainfall series is decreasing on the annual basis.
The objective of this research is to study the variation of Rainfall with space and time in Ninava Governorate using four Meteorological stations (Mosul, Sinjar, Talafar and Rabea). Simple and Multiple correlations were found between mean monthly values of rainfall and other Metrological elements.

Methodology
Ninava Governorate which is located in the northern part of Iraq contain four main meteorological stations (Mosul, Sinjar, Talafar, Rabea). fig(1) show the locations of these stations. The latitude, longitude, Altitude and years of observations for these meteorological stations were presented in table (1). In our research we study 1) Standard deviation and coefficient variation of the monthly values of rainfall were found. 2) Mean monthly rainfall, mean seasonal rainfall and mean annual rainfall. in these four stations 3) Time series of the monthly rainfall values were found in all stations.. 4) Simple and multiple regressions equations were found between mean monthly values of rainfall and other Meteorological elements. 5) Frequency distribution of rainfall data were also found in all stations.

1-Study of standard deviations, coefficient of variation and time series of monthly values of rainfall in all stations.
Standard deviation means a measure of the dispersion of a set of data from the mean. A low standard deviation indicates that the points tend to be very close to the means. Standard deviation can be calculated by the following equation: The coefficient of variation represents the percentage ratio of standard deviation to the mean, and it's a useful to show the variability of rainfall data.CV can be calculated by the following equation: CV =(SD / ) * 100 In this research we obtain the values of SD and CV by Excel program.         Fig(17) Shows the frequency distribution of rainfall data for all stations on the rainy months.
All the ranges were appears in January but the maximum frequency were in the range (30 -60)mm,(31 -49)% for all stations. In February the maximum value were in the ranges (30 -60) and (60 -90) mm, its value between (14, 66)% for all stations.
In April the maximum frequency were in the range (0 -30)mm and its value were between (48 -62)% for all stations. In may the maximum frequency were in the range (o -30)mm was value between (73 -87)%.

5) A highly negative correlation shown between mean monthly values
of rainfall with each of (Temperature, Evaporation, and sunshine hours for all stations, while it gave a highly positive correlations with each of (Relative Humidity, presser, and cloudiness).