• Sonuç bulunamadı

Spatial and temporal variations of aridity indices in Iraq


Academic year: 2021

Share "Spatial and temporal variations of aridity indices in Iraq"


Yükleniyor.... (view fulltext now)

Tam metin



Spatial and temporal variations of aridity indices in Iraq

NerminŞarlak1&Omar M. A. Mahmood Agha2,3

Received: 1 February 2016 / Accepted: 8 May 2017 / Published online: 3 June 2017 # Springer-Verlag Wien 2017

Abstract This study investigates the spatial and temporal var-iations of the aridity indices to reveal the desertification vul-nerability of Iraq region. Relying on temperature and precip-itation data taken from 28 meteorological stations for 31 years, the study aims to determine (1) dry land types and their delin-eating boundaries and (2) temporal change in aridity condi-tions in Iraq. Lang’s aridity (Im), De Martonne’s aridity (Am), United Nations Environmental Program (UNEP) aridity (AIu), and Erinç aridity (IE) indices were selected in this study

because of the scarcity of the observed data. The analysis of the spatial variation of aridity indices exhibited that the arid and semi-arid regions cover about 97% of the country’s areas. As for temporal variations, it was observed that the aridity indices tend to decrease (statistically significant or not) for all stations. The cumulative sum charts (CUSUMs) were ap-plied to detect the year on which the climate pattern of aridity indices had changed from one pattern to another. The abrupt change point was detected around year 1997 for the majority of the stations. Thus, the spatial and temporal aridity charac-teristics in Iraq were examined for the two periods 1980–1997 and 1998–2011 (before and after the change-point year) to observe the influence of abrupt change point on aridity

phenomena. The spatial variation after 1997 was observed from semi-arid (dry sub humid) to arid (semi-arid) especially at the stations located in northern Iraq, while hyper-arid and arid climatic conditions were still dominant over southern and central Iraq. Besides, the negative temporal variations of the two periods 1980–1997 and 1998–2011 were obtained for almost every station. As a result, it was emphasized that Iraq region, like other Middle East regions, has become drier after 1997. The observed reduction in precipitation and increase in temperature for this region seem to make the situation worse in future.

Keywords Aridity indices . Climate classification . Potential evapotranspiration . Spatial distribution . Temporal trend

1 Introduction

Aridity is a major, permanent risk for most territories in the world. Water resources and the quality of water are decreasing in many places (Lungu et al.2011). Assessing and monitoring aridity help to confront the threat of land degradation and desertification. Therefore, the necessary step in any venture related to water resources management, land reclamation for agricultural, and the like is to identify dry and semi-dry re-gions on a climatic basis. It is important to realize that there are different definitions and characterizations of aridity indices. The principal of which are Lang (1920), De Martonne (1926), UNESCO (1979), UNEP (1992), and Erinç (1996). These indices have been employed by many researchers to display any spatial variation of water scarcity in any region (e.g., Paltineanu et al. 2007 and Haider and Adnan2014). According to the use of climate factors, these indices can be mostly split into two main groups: one of them considers only

* Nermin Şarlak nsarlak@kmu.edu.tr Omar M. A. Mahmood Agha omar_alomary2@yahoo.com


Department of Civil Engineering, Faculty of Engineering, Karamanoglu Mehmetbey University, 70200 Karaman, Turkey


Department of Civil Engineering, Faculty of Engineering, University of Gaziantep, 27310 Gaziantep, Turkey

3 Department of Dams and Water Resources Engineering, University

of Mosul, Mosul, Iraq DOI 10.1007/s00704-017-2163-0


annual precipitation (P) or/and temperature (T); the other one further includes potential evapotranspiration (PET).

Lang (1920) and De Martonne (1926) proposed aridity index that are based on precipitation and temperature. The international association UNESCO (1979) improved Lang and De Martonne indices by using the potential evapotranspi-ration rates instead of temperature. In 1996, UNESCO aridity index was simplified by United Nations Environment Program (UNEP).

Many studies used aridity indices in their calculations for different regions. For example, Turkes (1999) analyzed the spatio-temporal variations of UNEP aridity index series in Turkey for the period 1930–1993. He found that dry sub-humid climatic conditions were dominant over most of central Anatolia. Semi-arid climatic conditions extend over Konya sub-region and the eastern part of Anatolia. He also indicated that some of stations in Aegean region showed a general shift around the 1960s from humid conditions to dry sub-humid climatic conditions. Baltas (2007) used Johansson Continentality, Kerner Oceanity, De Martonne, and Pinna Combinative indices to study the spatial distribution of

climatic indices in northern Greece. He claimed that De Martonne index is more appropriate for that region. In addition, Abdulla (2008) assessed the Thornthwaite moisture index for four stations (Sulaimaniya, Mosul, Baghdad, and Basra) in Iraq’s dry land. He found that the values of moisture deficit index were negatives for all stations. These values were greater in Sulaimaniya and Mosul than in Baghdad and Basra. This result is logical because Sulaimaniya and Mosul are close to mountainous regions, where the temperature as well as potential evaporation is lower than central and southern Iraq. Önder et al. (2009) estimated the potential effect of climate change on aridity and land cover in Turkey for 2070s using the observed data of 1990s. He used the regional climate model (RCM) to make the calculations. Aridity index was deter-mined by calculating the ratio of annual precipitation to refer-ence crop evapotranspiration. The results showed an increase of aridity in the southern and western coastal regions of Turkey for future data.

Additionally, recent research has increasingly used the aridity indices to identify dryness status of the climate at a given region. Deniz et al. (2011) studied the spatial

Fig. 1 Location of the meteorological stations in Iraq


v a r i a t i o n s o f c l i m a t e i n d i c e s i n Tu r k e y u s i n g Continentality index, Oceanity index, De Martonne aridity index, and Pinna Combinative index for two separate pe-riods (1960–1990 and 1991–2006). They found that the semi-dry areas in the second period were greater than in the first period. They did not identify any dry areas for the two periods in Turkey. Moreover, Some’e et al. (2013) investigated spatio-temporal trend of aridity index for the period 1966–2005 in Iran. The results showed positive trends in the annual aridity index series at 45% of the stations. However, only two stations indicated statistically significant positive trends at 95% confidence level. Tabari et al. (2014) examined De Martonne and Pinna indices for Iran for the period 1966–2005. De Martonne index and Pinna index showed that arid and semi-arid regions repre-sented 88 and 96% of Iran’s areas, respectively. They cated that the relationship between the values of two indi-ces was quite strong. Furthermore, Haider and Adnan (2014) investigated the assessment of aridity for the period

1960–2009 at 54 stations located in Pakistan. They dem-onstrated that about 75 to 85% of the overall region of Pakistan was actually arid climate and most part of this region lies in southern Pakistan. They emphasized that the southern area of Pakistan was more prone to dryness than northern part.

Water supply in Iraq is also prone to the impact of climate change and human activities. Nevertheless, climate change in Iraq has not been studied adequately. In this study, the spatial and temporal variations of the aridity indices were investigat-ed to reveal the desertification vulnerability of Iraq region. Calculations were based on temperature and precipitation data taken from 28 meteorological stations for 32 years. The pur-pose of this study is to determine (1) dry land types and their delineating boundaries and (2) temporal change in aridity con-ditions in Iraq. The results of this study are expected to help utilize limited water resources rationally and adjust the cropping pattern in the region to achieve sustainable productivity.

Table 1 Site information, including latitude, longitude, and elevation for each selected station

Number Station names ST no. Latitude (N) Longitude (E) Elevation (m) Missing data (%)

1 Rabiah 40602 36o34 42o3 382 14 2 Tel Afar 40603 36o11′ 42o22′ 273 13 3 Sinjar 40604 36o16′ 41o55′ 465 10 4 Mosul 40608 36o16′ 43o20′ 223 2 5 Salahuddin 40611 36o15′ 44o07′ 1075 17 6 Arbil 40616 35o55′ 43o57′ 420 15 7 Kirkuk 40624 35o17′ 44o42′ 331 5 8 Khanaqin 40637 34o18′ 45o21′ 202 14 9 Al-Kaem 40627 34o20′ 41o08′ 177.5 13 10 Anah 40629 34o19′ 41o48′ 139.5 15 11 Baiji 40631 34o49′ 43o27′ 115.5 1 12 Tikrit 40633 34o18′ 43o42′ 107 20 13 Haditha 40634 34o01′ 42o22′ 108.7 14 14 Al Khalias – 33o47 44o33 52 15 15 Rutba 40642 33o01 40o18 630.8 0 16 Ramadi 40645 33o18 43o22 45.1 11 17 Baghdad 40650 33o06 44o21 31.7 8 18 Kerbela 40656 32o23 44o07 29 0 19 Hella 40657 32o17 44o37 27 0 20 Nukhaib 40658 31o56 42o15 305 19 21 Azizyia – 32o48′ 45o06′ 24 20 22 Al-Hai 40665 32o06′ 46o05′ 17 0 23 Najaf 40670 31o47′ 44o26′ 32 0 24 Diwaniya 40672 31o45′ 45o02′ 20 4 25 Samawa 40674 31o13′ 45o34′ 11.4 3 26 Nasiriya 40676 30o53′ 46o18′ 7.6 0 27 Amara 40680 31o38′ 47o15′ 9.5 5 28 Basra 40689 30o15′ 47o51′ 2.4 0


2 Materials and methods

2.1 Study area and dataset

Iraq is located in southwest Asia and northeast Arabian home-land. Iraq shares borders with six countries: Turkey, Iran, Syria, Jordan, Kuwait, and Saudi Arabia. It covers an area of 435,052 km2, which lies between the latitudes of 29° 5′ and 37° 22′ N and the longitudes of 38° 45′ and 48° 45′ E. Topographically, Iraq is surrounded by mountainous area in the northeast, in which the highest peak is 3611 m above sea level, and by desert area in the southwest covering around 40% of the country’s areas.

The climate in Iraq varies from region to region since cli-mate depends on the geographical location of the region. Rainy season starts in northern and northeastern Iraq in November and ends in May, while in the rest of the regions, it begins from December to March. Average annual precipita-tion is about 216 mm ranging from 100 mm in the south to 1200 mm in the northeast. The monthly average temperature ranges from higher than 48 °C in July and August to below zero in January.

Historical records of precipitation and temperature for 32 years (1980–2011) from 28 climate stations were ac-quired from the Iraqi Meteorological Organization and Seismology (IMOS). The location of the meteorological stations and their geographical coordinates are indicated in Fig.1 and Table 1. Missing data in each station were estimated from nearby stations by utilizing normal ratio method. Standard normal homogeneity (SNH) and Pettit test were applied to detect the inhomogeneity in the time series. Since 3 out of 28 precipitation series were found to be inhomogeneous according to SNH test (Mahmood Agha andŞarlak2016), Al-Kaim, Kerbela, and Hilla sta-tions were excluded in further analyses.

2.2 Aridity indices

Many aridity indices have been proposed to identify the status of arid climate and to find out the classification of climate regarding water availability at a given region.

Lang (1920) defined aridity index as the ratio of annual precipitation to mean annual temperature. De Martonne modified Lang’s formula by adding a constant of 10 in the denominator (De Martonne 1926). However, this in-dex is undefined at mean annual temperature of −10 °C. UNEP proposed using the PET rate instead of mean an-nual temperature to define aridity like UNESCO index (UNEP 1992). The difference between UNESCO and UNEP’s aridity indices is in the methodology used to estimate potential evapotranspiration rate. While Penman formula (Penman 1948) was proposed by UNESCO to calculate PET, Thornthwaite formula (Thornthwaite 1948) was proposed by UNEP. Actually, Thornthwaite formula is based mainly on temperature with an adjust-ment being made to the number of daylight hours. Although this method tends to exaggerate the PET com-paring to the Penman formula, the scarcity of observed data in this region forced us to favor it over Penman’s. Erinç’s aridity index, on the other hand, is the ratio of mean annual precipitation total to annual maximum tem-perature. The class boundaries to delineate the area and formulation of each selected indices are summarized in Table 2.

2.3 Trend analysis

One widely used non-parametric test for the detection of trends in water resources and environmental studies is M a n n - K e n d a l l ( M K ) . T h e Wo r l d M e t e o r o l o g i c a l Organization highly recommended this test to estimate trends

Table 2 The class boundaries to delineate the area and formulation of each selected indices

Index Classification Remarks Reference

Im = P/MAT Im <20: arid; 20 to 40: semi-arid; 40 to 60: humid; >160: very humid

P—the annual precipitation (mm); MAT—the mean annual temperature (°C)

Lang (1920) Am = P/(MAT + 10) Am <5: arid; 5 to 15: semi-arid; 15 to 20: dry

sub-humid; 20 to 30: moist sub humid; 30 to 60: humid; >60: very humid

P—the annual precipitation (mm); MAT—the mean annual temperature (°C)

De Martonne (1926) AIu = P/PET AIu <0.05: hyper arid; 0.05 to 0.2: arid; 0.2 to 0.5:

semi-arid; 0.5 to 0.65: sub humid; >0.65: humid

P—the annual precipitation (mm); PET—the potential evapotranspiration (mm) according to Thornthwaite (1948). PET = 16 K (10 T/I)^m where K is a correction coefficient, I is a heat index, and m is a coefficient depends on I.

UNEP (1992)

IE= P/Tmax IE<8: hyper arid; 8 to 15: arid; 15 to 23: semi-arid;

23 to 40: dry sub humid; 40 to 55: humid; > 55: very humid

P—the annual precipitation (mm); Tmax—the annual mean maximum temperature (°C)

Erinç (1996)


of environmental data (Yenigun et al.2008). According to this test, the null hypothesis (H0) equals the non-existence of trend

(in the time series), whereas the alternative hypothesis (H1)

equals the existence of trend.


Sen’s slope estimator (SS) involves computing the slopes of all data pairs, Qi. These slopes are then used in order to

estimate the median of these N values of Qi. The slope Qiis

computed as follows

Qi¼ xi−xj

i−j ð1Þ

where xiand xjrepresent the data values during the time period

i and j (i > j), respectively. SS is the value of Qiat the median

of N. If N is odd, the SS is calculated as Qmed ¼ QN þ1 2 ;

other-wise, it is calculated by Qmed ¼ QN 2þ QNþ22

h i

=2. A positive value of Qirefers to upward trends, whereas a negative value

of Qirefers to downward trends in the time series (Sen1968b).

3 Results and discussions

The spatial variability of aridity based on Lang’s (Im), De Martonne’s (Am), UNEP (AIu), and Erinç (IE) aridity indices

was plotted as shown in Fig.2. This figure reveals the effect of aridity index type on the classification of aridity. While Lang’s aridity index classified the region as arid and semi-arid, Erinç and UNEP (De Martonne) classified the same region as hyper-arid (hyper-arid), hyper-arid (semi-hyper-arid), semi-hyper-arid (dry sub-humid), dry sub-humid, and sub-humid (moist sub-humid). The highest value of aridity index was found at only one station (Salahuddin). The elevation of this station (1075 m) is higher than other stations. Receiving relatively abundant

Table 3 The demarcated percentages of areas for the periods 1980– 1997, 1998–2011, and 1980–2011

Period Classification Lang De Martonne UNEP Erinç

1980–2011 Hyper arid 30.6 79.4 Arid 93.3 71.55 55.5 13.6 Semi-arid 6.7 19.65 12.3 5.4 Dry sub-humid 8.5 1.6 1.6 Moist sub-humid 0.3 1980–1997 Hyper arid 14.8 76.25 Arid 90 63 67.7 13.15 Semi-arid 10 23.7 14.5 7.4 Dry sub-humid 10.8 2.9 3.2 Moist sub-humid 2.6 1998–2011 Hyper arid 42.7 82.6 Arid 95.6 74.1 46.1 12.6 Semi-arid 4.4 16.2 11.2 4.7 Dry sub-humid 9.7 0.1 Moist sub-humid

Table 4 Trend test results for the

period 1980–2011 Stations Lang De Martonne UNEP Erinç

MK SS (mm/oC) MK SS (mm/oC) MK SS (mm/ mm) MK SS (mm/oC) Mosul −1.22* −0.29 −2.51 −0.19 −2.44 −0.005 −2.55 −0.21 Arbil −2.22 −0.29 −2.25 −0.2 −1.8* −0.005 −2.32 −0.24 Salahuddin −1.93* −0.29 −1.77* −0.19 −1.83* −0.005 −1.96 −0.27 Rabiah −2.68 −0.31 −2.38 −0.19 −1.63* −0.006 −2.45 −0.19 Tel Affar −3 -0.29 −2.87 −0.18 −1.73* −0.005 −2.71 −0.21 Sinjar −2.58 −0.29 −2.55 −0.2 −3.29 −0.006 −2.58 −0.25 Kirkuk −1.67* −0.23 −1.53* −0.16 −1.56* −0.004 −1.63* −0.17 Baiji −1.51* −0.11 −1.51* −0.07 −2.09 −0.002 −1.28* −0.08 Tikrit −3.13 −0.15 −3 −0.1 −3 −0.002 −3.06 −0.12 Khanaqin −2.55 −0.32 −2.48 −0.21 −2.28 −0.005 −4.04 −0.23 Haditha −1.77* −0.16 −1.8* −0.11 −2.18 −0.003 −1.7* −0.11 Anah −2.16 −0.09 −2.12 −0.06 −2.64 −0.002 −2.06 −0.06 Ramadi −1.67* −0.09 −1.57* −0.06 −2.16 −0.002 −1.54* −0.06 Baghdad −1.99 −0.07 −1.83* −0.05 −2.58 −0.001 −1.7* −0.05 Al Khalias −4.14 −0.17 −4.07 −0.11 −2.41 −0.003 −4.1 −0.12 Rutba −1.86* −0.07 −1.8* −0.05 −2.29 −0.001 −1.7* −0.05 Azizyia −2.09 −0.06 −2.06 −0.04 −2.29 −0.001 −2.16 −0.05 Al-Hai −2.51 −0.1 −2.51 −0.07 −3.06 −0.001 −2.45 −0.07 Najaf −2.38 −0.1 −2.35 −0.07 −3.19 −0.002 −2.32 −0.08 Diwaniya −1.8* −0.06 −1.77* −0.04 −1.83* −0.001 −1.67* −0.05 Nukhaib −1.56* −0.06 −1.56* −0.05 −1.94* −0.001 −1.56* −0.05 Samawa −0.1* −0.02 0.001* −0.01 −0.44* −0.0002 −0.14* −0.01 Amara −0.83* −0.04 −0.66* −0.02 −1.67* −0.001 −0.57* −0.03 Nasiriya −1.77* −0.05 −1.61* −0.03 −2.48 −0.001 −1.7* −0.04 Basra −1.74* −0.08 −1.7* −0.05 −2.77 −0.001 −1.8* −0.06


precipitation associated with low temperature may be attribut-ed to classifying this area as humid.

The percentage of classified areas are calculated and listed in Table3. Table3shows that the total percentage of hyper-arid, hyper-arid, and semi-arid areas for the whole period ranges from 91.2 to 100. Therefore, Iraq can be considered as dry lands excluding for few areas in the north. Dry lands, classi-fied into hyper-arid, arid, and semi-arid areas, are generally prone to low rainfall and high evapotranspiration. Since dry lands are limited by soil moisture, agriculture in these regions depends on irrigation rather than rainfall. Consequently, the limited water resources in this region should be used cautiously.

Mann-Kendall (MK) test and Sen’s slope (SS) estimator were applied to exhibit the temporal tendency of the selected aridity indices. The results of MK test, shown in Table4, indicate that trends of Lang’s (Im), De Martonne (Am), UNEP (AIu), and Erinç (IE) indices tend to decrease

(statisti-cally significant or not) for all stations. According to the Am (IE) values, 12 (13) out of 25 stations were detected to have

statistically significant decreasing trend. The Salahuddin sta-tion had statistically significant decreasing trend for IEunlike

Am series. Similar to the Am series, 12 out of 25 stations of the Im series showed a significant decreasing tendency but at different station. In other words, statistically significant

decreasing trends in the Am (Im) series were found at Mosul (Baghdad) station. Also, according to the AIu values, 16 out of 25 stations were found to have statistically significant de-creasing trend. Although the number of stations with the sta-tistically significant decreasing tendency differed in each arid-ity index type, each aridarid-ity index had downward tendency.

The magnitude of the decreasing trend in Im, Am, AIu, and IEseries were found by using SS estimator technique (see

Table4). It should be emphasized that the detected trend in stations, which are located in the semi-arid area of Iraq, de-creased more rapidly than other stations. The maximum values of significant decreasing trend slope for the Im and Am series were obtained at Khanaqin station, while the highest slope of significant decreasing trends for AIu series was found at Sinjar station. Among the investigated aridity indices, Im index showed a maximum decreasing trend of aridity indices with a slope of−0.32 to −0.02 (mm/mm)/year, while AIu had a min-imum decreasing trend of aridity indices for all stations with a decrease of −0.006 to −0.0002 (mm/mm)/year. Actually, the maximum decreasing slope in total precipitation data was also detected in Khanaqin station (Mahmood Agha and Şarlak 2016). This emphasizes that the most sensible parameters to aridity in this region is precipitation.

The cumulative sum chart (CUSUM) curve test was also applied in this study to detect the change point on time series

Table 5 The computed percentage variations of the aridity index for the periods 1980–1997 and 1998–2011 Station name Aridity indices

Lang Variation De Martonne Variation UNEP Variation Erinç Variation

80–97 98–11 % 80–97 98–11 % 80–97 98–11 % 80–97 98–11 % Mosul 20.7 14.6 −29.5 13.8 9.9 −28.1 0.27 0.33 −33.8 13.0 15.1 −28.7 Arbil 23.7 16.5 −30.6 15.9 11.2 −29.6 0.31 0.37 −33.5 15.7 18.3 −30.5 Salahuddin 36.3 29.7 −18.1 23.1 19.3 −16.6 0.57 0.63 −18.4 26.5 29.4 −19.6 Rabiah 21.8 14.7 −32.3 14.0 9.7 −30.8 0.33 0.38 −27.3 13.0 15.1 −30.4 Tel Affar 17.7 12.7 −28.5 11.8 8.6 −27 0.23 0.27 −32.1 11.8 13.8 −27.4 Sinjar 19.7 14.1 −28.3 13.1 9.6 −26.7 0.25 0.31 −33.8 13.8 16.2 −28 Kirkuk 18.2 12.8 −29.6 12.5 8.9 −28.5 0.21 0.25 −34.8 12.2 14.2 −28.5 Baiji 9.8 7.8 −20.8 6.7 5.4 −19.6 0.12 0.14 −28.2 6.6 7.4 −20 Tikrit 9.0 6.3 −30 6.2 4.4 −29.1 0.10 0.12 −34.5 5.9 7.0 −29.7 Khanaqin 14.4 10.2 −28.6 10.0 7.2 −27.4 0.12 0.14 −28.2 9.3 10.9 −27.7 Haditha 6.9 4.5 −34.3 4.7 3.1 −33.2 0.08 0.10 −36.5 4.2 5.1 −32.3 Anah 7.3 6.0 −18.7 4.9 4.1 −17.5 0.10 0.12 −26.6 4.9 5.4 −17.5 Ramadi 5.4 3.8 −28.8 3.7 2.7 −27.8 0.07 0.08 −35.1 3.5 4.1 −27.8 Baghdad 5.4 4.0 −26.3 3.7 2.8 −25.2 0.06 0.08 −34.8 3.5 4.0 −25.5 Al Khalias 8.9 5.5 −38 6.1 3.8 −37.2 0.11 0.14 −42.1 5.4 6.6 −38 Rutba 6.6 4.3 −34.7 4.3 2.9 −33.2 0.09 0.11 −37.9 4.0 4.8 −33.3 Azizyia 5.1 4.0 −22.3 3.6 2.8 −21.7 0.06 0.07 −26.3 3.5 3.9 −22.5 Al-Hai 5.7 4.2 −26.2 4.1 3.0 −25.2 0.05 0.06 −38.4 3.9 4.5 −25.9 Najaf 4.7 3.1 −34 3.3 2.2 −33 0.04 0.06 −45.6 3.0 3.7 −33.6 Diwaniya 4.5 3.6 −18.4 3.1 2.6 −17.3 0.05 0.06 −27.1 3.1 3.4 −18 Nukhaib 4.0 2.6 −34.1 2.8 1.9 −33.2 0.04 0.05 −44.2 2.6 3.1 −34.1 Samawa 3.6 3.9 9.9 2.5 2.8 10.6 0.04 0.04 1.8 2.9 2.8 8.7 Amara 6.5 6.9 6.6 4.6 5.0 8.1 0.07 0.07 −9 5.2 5.0 7.3 Nasiriya 5.0 4.5 −10.6 3.6 3.2 −9.2 0.05 0.06 −27.6 3.7 3.9 −10.2 Basra 5.6 4.6 −19.1 4.1 3.3 −18.1 0.05 0.06 −35 4.0 4.5 −19.3


of aridity indices. In this test, the change point was estimated from the position of the summit of the curve (Talaee et al.

2014). CUSUM curve for aridity series at Salahuddin and Nasiriya station showed that the curve reached the maximum


value in 1999. The change point year at Mosul, Tel Affar, Sinjar, and Azizyia stations was found around 1996, while the change point year at Baghdad and Al-Khalias stations

was detected at 1994. In addition, the change point year at Arbil, Rabiah, Tikrit, Baiji, Khanaqin, Haditha, Anah, Ramadi, Rutba, Najaf, Al-Hal, and Basra stations was found


at 1997. Since most of the stations’ change point year was detected at 1997, the spatial and temporal variabilities of the aridity indices were investigated for two series defined before and after 1997 (1980–1997 and 1998–2011) to exhibit the effect of change point year on the classification of aridity.

The demarcated percentages of areas were computed during 1980–1997 as shown in Table3. It was observed that about 63 to 89.4% of areas were lying between hyper-arid to arid, 7.4 to 23.7% areas were considered as semi-arid, 0 to 10.8% areas were dry sub-humid, and 0 to 2.6% areas were moist sub-humid, according to selected aridity indices. However, the percent of areas for the period 1998–2011 ranging from 74.1 to 95.6% lied between hyper-arid to arid with no existence of moist sub-humid area. Consequently, we should note that there is an increase in the arid area of about 7.2% of the total area after 1998.

The computed percentage variations of the aridity in-dex for the periods 1980–1997 and 1998–2011 are shown in Table5. The negative variations were detected between the 1980–1997 and 1998–2011 periods except for two stations—Samawa and Awara. The negative variation can be explained by both the rainfall reduction and tem-perature increase in the second part of the time (Mahmood Agha andŞarlak2016).

In general, there is an obvious extension in arid region toward to northern Iraq and this stretch varies from one index to another. Figures 3 and 4 show clearly the climate zone classifications’ discrepancy for the two periods. In fact, the classification of many stations located in northern Iraq changed from semi-arid (dry sub-humid) to arid (semi-arid) after 1997.

4 Conclusions

This study aims to describe dry land types and their delineat-ing boundaries and the analysis of trend for aridity indices in the study region. For this purpose, four aridity indices and trend tests were applied to the observed climate data from 25 stations located in Iraq for 32 years. As a result, around 87.2 to 100% of the country’s areas were classified as arid and semi-arid according to semi-aridity indices, whereas around 0 to 12.8% of Iraq’s land surface was classified as sub-humid area. These results are similar to research findings conducted in other re-gions of the Middle East such as Iran, Saudi Arabia, and Bahrain. For example, Tabari et al. (2014) found that 88 to 96% of Iran regions were categorized as arid and semi-arid regions. Subyani et al. (2010) classified Saudi Arabia regions as desert and semi-arid. Elagib and Addin Abdu (1997) showed that Bahrain could be classified as arid or hyper-arid area. As for temporal variations, it was observed that the arid-ity indices tend to decrease (statistically significant or not) for all stations in this region.

The change point year was detected in 1997 in most sta-tions. These results are compatible with the IPCC report (IPCC 2007), which emphasized that ocean temperature changed markedly at the end of the 1990s of the last century. We detected an increase in the area of arid and semi-arid on an average of 7.2% of the total area after 1997. This result is in agreement with findings of another study in Turkey. Deniz et al. (2011) indicated that the semi-dry areas in the period from 1991 to 2006 were greater than the period from 1960 to1990.

Finally, we hope that the results of the present study will prove helpful in the future planning, evaluating, and managing of water resources in order to control land degradation and desertification in vulnerable areas by taking necessary precautions.


Abdulla HJ (2008) Evaluation of moisture deficit index in dry land in Iraq. Middle-East J Sci Res 3(3):116–119

Baltas E (2007) Spatial distribution of climatic indices in northern Greece. Meteorol Appl 14:69–78

De Martonne E (1926) Aerisme, et índices d’aridite. Comptesrendus de l’Academie des Sciences 182:1395–1398

Deniz A, Toros H, Incecik S (2011) Spatial variations of climate indices in Turkey. Int J Climatol 31:394–403

Elagib A, Addin Abdu S (1997) Climate variability and aridity in Bahrain. J Arid Environ 36:405–419

Erinç S (1996) Klimatoloji ve Metodları. Alfa Basic-Yayım-Dağıtım, Istanbul

Haider S, Adnan S (2014) Classification and assessment of aridity over Pakistan provinces (1960-2009). Int J Environ 3(4):24–35 IPCC (2007) Climate change 2007: climate change impacts, adaptation

and vulnerability. Working Group II contribution to the Intergovernmental Panel on Climate Change Fourth Assessment Report. Summary for policymakers, 23

Lang R (1920) Verwitterung und Bodenbildung als Einführung in die Bodenkunde. Schweizerbart Science Publishers, Stuttgart Lungu M, Panaitescu L, Niţă S (2011) Aridity, climatic risk phenomenon

in Dobrudja. Present Environ Sustain Dev 5(1):179–190

Mahmood Agha OM,Şarlak N (2016) Spatial and temporal patterns of climate variables in Iraq. Arab J Geosci 9(4):1–11

Önder D, Aydin M, Berberoğlu S, Önder S, Yano T (2009) The use of aridity index to assess implications of climatic change for land cover in Turkey. Turk J Agric For 33(3):305–314

Paltineanu C, Tanasescu N, Chitu E, Mihailescu IF (2007) Relationships between the De Martonne aridity index and water requirements of some representative crops: a case study from Romania. International Agrophysics 21(1):81

Penman H (1948) Natural evaporation from open water, bare soil and grass. Proc R Soc A 193:120–146

Sen PK (1968b) On a class of aligned rank order tests in two-way layouts. AMS 39:1115–1124

Some’e BS, Ezani A, Tabari H (2013) Spatiotemporal trends of aridity index in arid and semi-arid regions of Iran. Theor Appl Climatol 111:149–160


Subyani A, Al-Modayan A, Al-Ahmadi F (2010) Topographic, seasonal and aridity influences on rainfall variability in western Saudi Arabia. J Environ Hydrol 18:1–11

Tabari H, Talaee PH, Nadoushani SSM, Willems P, Marchetto A (2014) A survey of temperature and precipitation based aridity indices in Iran. Quat Int 345:158–166

Talaee PH, Some’e BS, Ardakani SS (2014) Time trend and change point of reference evapotranspiration over Iran. Theor Appl Climatol 116(3–4):639–647

Thornthwaite CW (1948) An approach towards rational classification of climate. Geogr Rev 38:55–64

Turkes M (1999) Vulnerability of Turkey to desertification with respect to precipitation and aridity conditions. Turk J Eng Environ Sci 23:363–380 UNEP (1992) World atlas of desertification. Edward Arnold, London UNESCO (1979) Map of the world distribution of arid regions.

Explanatory note, Man and Biosphere (MAB)

Yenigun K, Gumus V, Bulut H (2008) Trends in streamflow of the Euphrates basin, Turkey. Proc ICE—Water Manag 161(4):189–198


Fig. 1 Location of the meteorological stations in Iraq
Table 1 Site information, including latitude, longitude, and elevation for each selected station
Fig. 2 The spatial variability of aridity based on Lang ’s (Im), De Martonne’s (Am), UNEP (AIu), and Erinç (I E ) aridity indices (1980 –2011)
Table 4 Trend test results for the


Benzer Belgeler

-Rıfat Usta 80.Yaş gü­ nü nedeniyle düzenlenen saygı gecesinde konu­ şurken bir ara oradaki Kültür Bakanı Fikri Sağ- lar’a takıldı ve dedi ki:

Ankara Atatürk Lisesi’ni bitirdikten sonra, 1961 yılında İstanbul Devlet Güzel Sanatlar Akademisi, Resim Bölümü &#34;Zeki Faik ¡zer Atelyesi’nde öğrenime

Anterior fontanel yerle~imli dermoid kistle- rin nadir gorOlmesi, ozellikle beyazlarda ve ileri ya~larda literatDrde rastlayamaml~ olmamlz nedeniyle 16 ya~mda tespit edilen

In another study conducted with 229 women, 83% of women approved of abortion under the circumstance that the baby will be disabled or when the health of mother is in danger,

In our study, it was shown that the difference between ossiculoplasty techniques such as PORP, autologous grafts and bone cement, which can be used in the presence of intact

Analitik olarak sonlu elemanlar yöntemiyle bulunmuş olan serbest titreşim frekanslarıyla, piklerin seçilmesi yöntemi kullanılarak (her üç durum için) saptanan serbest titreşim

In according to perform this study heating and cooling process has been stimulated in different situation (shading and insulation) with a software (design

The purpose of this study was to determine the correlation of mean platelet volume (MPV) and neutrophil/lymphocyte ratio (NLR) with disease activation and clinical para- meters