- Email: [email protected]
Estimation of regional irrigation water requirement and water supply risk in the arid region of Northwestern China 1989–2010
Agricultural Water Management 128 (2013) 55–64
Contents lists available at SciVerse ScienceDirect
Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat
Estimation of regional irrigation water requirement and water supply risk in the arid region of Northwestern China 1989–2010 Yanjun Shen a,∗ , Shuo Li a,b , Yaning Chen c , Yongqing Qi a , Shuowei Zhang d a Key Laboratory for Agricultural Water Resources, Center for Agricultural Resources Research, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Shijiazhuang 050021, China b College of Resources and Environmental Sciences, Hebei Normal University, Shijiazhuang 050016, China c State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China d Department of Geography, The State University of New York-Buffalo, Buffalo, USA
a r t i c l e
i n f o
Article history: Received 9 November 2012 Accepted 21 June 2013 Keywords: Irrigation water requirement Crop water requirement Water resources Penman–Monteith equation Water supply risk Arid region in Northwestern China (ARNWC)
a b s t r a c t Water use in agricultural sector shares more than 90% of the total water withdrawal in the arid region of Northwestern China (hereafter, ARNWC). Irrigation water demand is therefore essential to the water resources allocation to economy and natural ecosystems in the highly water deficit region. In this study, we analyzed the spatial and temporal variations of irrigation water demand as well as crop water requirement by combining the modified Penman–Monteith equation recommended by FAO and GIS technology. Crop and irrigation water requirements for 5 main crops, including wheat, corn, cotton, oilseed and sugar beet, from 1989 to 2010 were calculated and the spatio-temporal variations were analyzed. The results suggested that the demand of irrigation water in the ARNWC showed increasing trend during the past two decades, which mainly caused by fast increase in cotton cultivation areas, because irrigation water requirement for cotton was much larger than the other crops. The changes in cotton growing area significantly affected the spatial pattern of water demand. A total of 44.2 billion m3 water was withdrawn for irrigation in year 2010. Larger amount of water was consumed for crops in Northern Xinjiang and Tarim River Basin than Qilian-Hexi region. Irrigation water requirement reaches its maximum in July and August. It is revealed that the critical period for water supply is during April and May through comparing the monthly irrigation water requirement with water availability, i.e. river discharge. Even though the annual water resources are much larger than the requirement, but for some basins, there is severe physical water shortage during the critical water use period in April and May. The water resource supply is expected to be facing more difficulties in future. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Water is one of the essential resources in arid and semi-arid regions especially for agriculture. The arid region in Northwestern China (ARNWC) is one of the mostly water stressed regions in the world (Shen and Chen, 2010). This region is also an important foodand cotton-producing region in China, but the imbalance between water supply and demand is also very prominent in the region. Agricultural irrigation consumed most water resources in this region, and accounted for 91.8% of the total water consumption (Geng et al., 2006). The shortage of water resources has become a major limiting factor for the socio-economic development in the ARNWC. Even in places with relatively abundant water resources, it is not sufficient to meet the water demand of crop growth all the year round. With increasing acreage of crop growing, unreasonable
∗ Corresponding author. Tel.: +86 311 85825464; fax: +86 311 85815093. E-mail address: [email protected] (Y. Shen). 0378-3774/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.agwat.2013.06.014
crop planting structure, high irrigation quota, and low water use efficiency, the shortage of water resources is becoming increasingly serious. Therefore, quantifying agricultural irrigation water demand in the region, and analyzing the spatial and temporal characteristics of the irrigation water demand, are of importance to help improve the water management toward sustainable use. Currently, there are several methods to calculate crop irrigation water requirement, including ground observation (based on crop planting structure and irrigation model), the crop model method, the remote sensing method, and the Penman–Monteith method recommended by the UN Food and Agriculture Organization (FAO). Ground observation can help calculating accurate water consumption estimation at farmland scale, but can hardly estimate the water requirement in a region. Remote sensing method always combines with ground meteorological observations and crop survey to estimate crop water demand (Ma et al., 2005; Mohamed et al., 2004). However, due to the error originated from the spatial and temporal resolutions of remote sensing data, there are still many technical issues need further, e.g. the interpolation method from
56
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 1. Geomorphologic setting and zoning of the ARNWC. The embedded figure shows the distribution of annual precipitation, precipitation in the study area is generally less than 200 mm/yr.
instantaneous satellite image to daily or seasonal water consumption. The Penman–Monteith method has a simple equation and better physical basis, and been widely used in the estimation of crop water requirement. Liu et al. (2005) calculated water requirements of major crops in North China using the Penman–Monteith method. Xiao et al. (2006) calculated and analyzed the spatiotemporal characteristics of water demand of corn in China. Er-Raki et al. (2007) studied water demand of wheat in Morocco using the Penman–Monteith equation combined with land surface remote sensing data. Du et al. (2006) studied the response of irrigation to cotton production and physiology in the oasis in Northwestern China. Although the above studies looked at the water requirement of unit area crop, they did not consider the impact of rainfall on crop irrigation water demand, and could not quantify the total water demand at a regional scale, especially, in the ARNWC. In this paper, based on Penman–Monteith method and combined ground observation data of crop growing acreage, we calculated and analyzed the temporal and spatial variations of agricultural irrigation water demand in the ARNWC. The objectives of this study are: (1) to give a whole picture of spatial and temporal changes of irrigation water demand and the water availability and (2) to provide quantitative information for spatial irrigation water demand as well as the water demand structure for different crops in the hyper arid region of China. We hope these results can help improve the water resources management in this highly water deficit region. 2. The ARNWC The ARNWC is located in the central part of Eurasian continent, including all of Xinjiang, Hexi Corridor in Gansu and the west of the Helan Mountain of Inner Mongolia; its geographical location is between longitude 73–125◦ E and latitude 35–50◦ N (Luo and Yang, 2003); the total area is approximately 2.02 million square kilometers, accounting for more than one fifth of China’s total terrestrial area. The Kunlun–Altun–Qilian Mountains are the south boundary of the area (Fig. 1), which combined with the Qinghai-Tibet Plateau, blocks the vapor from Indian Ocean. Altai Mountain is the northern boundary, blocking the vapor from the Arctic Ocean. Tianshan Mountains lie in the middle of the region and form the landscapes
of alpine, inland basins, gobi desert and widespread sandy desert. This inland region is controlled by continental climate, with little rainfall throughout the year and strong evaporative potential. The average annual precipitation is only 130 mm, with uneven spatial and temporal distributions. The highest rainfall occurs at the eastern end of the Tianshan Mountains in the Ili River Valley, which could amount up to 800 mm/yr or above; annual precipitation is generally less than 50 mm in the oasis areas, where most agricultural activities exist. However, the potential evaporation in this arid region can be as large as 3200 mm/yr, 8–10 times the annual rainfall (Shen and Chen, 2010; Chen et al., 2012). The hyper arid region is thus one of the most arid regions in the world (Shen and Chen, 2010). Abundant radiation and heat resources in this region make it an ideal place for thermophilous and photophilous crops growth, so it becomes an important high-quality and also high-yield cotton production zone (Xu et al., 2011). Other crops like wheat and corn are also widely cultivated in this region. The arable land is mainly distributed in oases embedded at the edge of the deserts, relying on surface water and groundwater for irrigation. At the end of 2010, the total area of farmland was 4.9 million hectares, effective irrigation reached 4.682 million hectares, and the total crop growing area amounted to 5.55 million ha. Cotton growing area was 1.5 million ha, and its acreage and production accounted for 31% and 42.7% of the country’s total, respectively. According to Bureau of Xinjiang Water Resources (2011), the annual withdrawal for irrigation reached to 48.83 billion m3 , accounting for 91.3% of the total water consumption in Xinjiang in 2010. The irrigated area in northwestern arid regions increased from 2.7 million hectares in 1989 to about 4.4 million hectares (Fig. 2) in 2010, and the fastest increase goes to cotton growing land. In the irrigated farmland, wheat and corn growing acreage was about 2 million hectares, with little change in the total over the past 20 years. Beet and oilseed acreage was approximately 0.4–0.5 million hectares, which slightly declined in recent years. The cotton growing area showed the most fast increase, from only 0.4 million hectares in 1989, mainly distributed in Tarim River Basin, to 1.5 million hectares by 2008. In 2010, the cotton growing area slightly declined to 1.3 million hectares, still accounting for ∼1/3 of the total cropping areas.
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 2. The variations of irrigated crop growing area in ARNWC.
Fig. 3 shows the spatial distribution and variation of growing area of the five major crops from 1989 to 2010. In the Tarim River Basin, double crops wheat–corn or single crops cotton are the major cropping types; while in Northern Xinjiang and Qilian-Hexi Region, due to heat constraints single crop such as spring wheat or spring
57
corn are mainly grown. The north and south slopes of the Tianshan Mountains is the transition zone, where sugar beet and oilseed are mainly grown. Fig. 3 also reflects changes in the structure of the crops in the past 20 years. The crops grown in 1989 were dominated by wheat and corn. And cotton acreage was mainly distributed in several big irrigation districts such as Shihezi, Aksu, Kashi, etc., and the acreage accounted only for 10.4% of the total growing area, far less than food crops like wheat and corn. Oilseed, sugar beet and other cash crops were mainly planted in the Northern Xinjiang’s Ili, Tacheng and Gansu’s Minle region. However, by the year 2000, with the improvement of irrigation facilities and the increase of cotton procurement price, cotton acreage rapidly increased to 1045 thousand hectares, accounting for 25.6% of the total sowing areas; the Northern Tianshan Mountain and the oases in the edge of the Tarim Basin increased most significantly. Cotton was patchily distributed in Gansu’s Dunhuang and Yumen region. The changes of oilseed and sugar beet were minor. By the end of 2010, due to the water restrictions, the cotton area declined throughout the study region. The main crops became cotton, wheat, and corn; oilseed, sugar beet, and other cash crops scattered in northern part near Altay Mountains and central Hexi Corridor in Gansu (Fig. 3).
Fig. 3. Spatial distribution of crop growing acreage in the ARNWC in different periods (1989–2010). The red, green, and blue color composition indicates the acreage percentage of the three types of cultivation, wheat/corn, cotton, and beet/oilseed. (For interpretation of the references to color in this artwork, the reader is referred to the web version of the article.)
58
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 4. Crop coefficient (Kc ) values for normal/spring crops (a) and winter crops (b).
3. Methods 3.1. Data Daily meteorological observations are collected from China Meteorological Administration (CMA). The main meteorological variables include the maximum and minimum temperatures, wind speed, relative humidity and sunshine duration. Observations on crop growth and phenology are collected from several agricultural meteorological stations. According to the characteristics of the time series and spatial locations, 67 meteorological stations were selected for the calculation and analysis in this study. Monthly river discharge data from the gauges at mountain foot were selected for analyzing the supply and demand situation of each basin. Socio-economic statistical data were collected from the Statistical Yearbook of the provinces, municipalities and prefectures (1989–2010) in the study regions, including arable land at the county level and the various crops sown. Land use data for years 1990 and 2000 were provided by the Data Center of Resources and Environment, Chinese Academy of Sciences. These data were combined with socio-economic data and statistics to create the spatial distributions of crops sowing area in different years. 3.2. Calculation of crop irrigation requirement Crop irrigation requirement is calculated using the following formula: CWR = ETc − Pe
(1)
ETc = Kc · ET0
(2)
IWR =
S · CWR Ic
(3)
where CWR is the crop water requirement of a certain crop; ET0 is the reference evapotranspiration; Kc is the crop coefficient; S is the acreage of the crop; Pe is the effective rainfall, which indicate the rainfall transferred to soil moisture; IWR is the irrigation water requirement of the crop; Ic is the irrigation efficiency, by considering the loss in water diversion to field from river channel. CWR can vary widely and is affected by crop variety, soil property, and aquifer conditions, weather conditions, and other factors (Sakellariou-Makrantonaki, 2006). 3.3. Reference evapotranspiration Reference evapotranspiration (ET0 ) was calculated using the modified Penman–Monteith equation recommended by United Nations Food and Agriculture Organization (FAO) (Allen et al.,
1998). Reference ET was defined and calculated based on the evapotranspiration of well watered short grass with 0.12 m in height, resistance of 70 s/m, and the albedo of 0.23. This is very similar to the open surface, highly consistent, vigorously growing, ground cover without the water shortage (Allen et al., 2005). Reference evapotranspiration is only related to meteorological factors (Shahid, 2011), so it can be used in different regions, and has universal significance for inter-comparison. The reference ET was calculated using the following equation: ET0 =
0.408 (Rn − G) + (900/(T + 273))u2 (es − ea ) + (1 + 0.34u2 )
(4)
where ET0 is the r eference evapotranspiration (mm/d) Rn is the net radiation at the land surface [MJ/(m2 d)]; G is the soil heat flux density [MJ/(m2 d)]; u2 is 2 m height average wind speed in 24 h (m/s); es is the saturation vapor pressure (kPa); ea is actual vapor pressure (kPa); is slope of vapor pressure–temperature curve [kPa/◦ C]; and is psychrometric constant [kPa/◦ C] (Er-Raki et al., 2007). 3.4. Crop coefficient during the growing period Crop coefficient is the ratio of well watered crop evapotranspiration and the reference ET, i.e. ETc /ET0 , commonly written as “Kc ”. Kc value generally can be obtained through field experiments. The Kc values reflect the relative water consumption capacity of crop during different growing stages of different crops. In general, the crop coefficient varies during four developing stages: the initial growth period, the rapid development period, mid-maternity, and mature period (as illustrated in Fig. 4). Kc value is generally very small at the initial growth phase because of low crop cover and most water consumption is from soil evaporation. During the rapid development period, leaf area increases rapidly and causes a rapid increase in the Kc value. The Kc value reached maximum when crops fully developed, and remained stable over a period of time. As the crop matured, the Kc value start to decreasing due to foliage senescence. In the current study, we used the crop coefficients according to Duan et al. (2004). The crop coefficients for some typical subregions are shown in Table 1. 3.5. Effective rainfall The effective rainfall at agricultural land is defined as the portion of rainfall penetrating canopy, infiltrated in soil layer and stored in the crop root zone, which can be used later by crops as transpiration. It does not include the interception of plant canopy, surface runoff and the amount of water percolating to the layer below the root zone. Effective rainfall is usually calculated as follows: Pe = P × ı
(5)
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
59
Table 1 Crop coefficients of the major crops in different growth stages in some sites in ARNWC. Sub-region
Crop
Development stage Initial
Developing
Middle
Sowing (mm/dd)
Harvesting (mm/dd)
Total (days)
Late
Yanqi, Xinjiang
Spring wheat Spring corn Cotton Oilseed
0.17 0.21 0.26 0.31
0.17–1.16 0.21–1.20 0.26–1.20 0.31–1.15
1.16 1.20 1.20 1.15
1.16–0.40 1.20–0.35 1.20–0.70 1.15–0.35
03/07 04/27 04/20 06/14
07/05 09/17 10/16 09/29
121 144 180 108
Zhangye, Gansu
Sugar beet Cotton Spring wheat Spring corn
0.34 0.23 0.23 0.23
0.34–1.21 0.23–1.21 0.23–1.16 0.23–1.20
1.21 1.21 1.16 1.20
0.21–0.70 1.21–0.72 1.16–0.40 1.20–0.35
03/15 04/11 03/13 04/13
09/24 10/11 07/17 09/26
194 184 127 167
where P is the precipitation; ı is an empirically effective utilization coefficient of rainfall, which is adopted as 0.52 in our study region (Xu et al., 2010). 3.6. Irrigation efficiency Irrigation efficiency refers to the ratio of the net amount of water used by crops to the total amount of water withdrawal from river channels or reservoirs. It is mainly composed of the canal water utilization coefficient and the field water utilization coefficient. The canal water utilization coefficient is used to measure the degree of utilization of canal water delivery; field water utilization coefficient is used to measure the degree of utilization of soil water (Wu and Zhou, 2011). Improvements in irrigation infrastructure as well as the development of water-saving technologies have led to an obvious improvement of irrigation efficiency in the ARNWC (Table 2). According to the “China Water Conservancy Yearbook” and the regional water resources bulletin, the irrigation water utilization coefficient has increased from 0.38 to 0.48 during past 20 years (Table 2). In this paper, the irrigation water utilization coefficients (Table 2) were used to calculate irrigation requirement. 3.7. Spatial interpolation In this study, we employed the Thiessen polygon method for spatial interpolation of the crop water requirement calculated based on the ground meteorological observations at 67 stations. Point to the surface interpolation methods included the inverse distance weighting method, the spline function method and the kriging method (Zhu and Zhang, 2005). In many areas these methods are able to better reflect regional differences. In the study region, the three mountains split the region into two large inland depressions, where widespread deserts with arable land and oases scattered along the mountain foot fringes. This unique topography
makes natural and geographical conditions of the arable land in each watershed (such as altitude, weather and soil) similar, so is the crop water demand. We used the simple Thiessen polygon for dimensional interpolation in this study. The distribution of crop growing area is rasterized by combining the statistics and land use maps. The Thiessen polygon division and arable land distribution of the study region are shown in Fig. 5. 4. Results 4.1. Crop water requirement and its trend We calculated the crop water requirement for the five main crops from 1989 to 2010, and detected the trend using the Mann–Kendall test (Table 3). Table 3 shows that sugar beet is the highest water consumption crop with annual CWR of 710 mm, and oilseed is the lowest with CWR of 352 mm. There are some differences among the spatial distributions of regional crop irrigation requirement (Table 3) due to the difference of the geographical conditions of the various sub-regions. Based on the Mann–Kendall test, it was found that the annual water requirement of the five major crops over the past 20 years showed increasing trend, especially of oilseeds and sugar beet. The increase of CWR is not significant for all the crops in Tarim Basin, but significant in Northern Xinjiang (NX) and Hexi Region (QH), mainly due to the relatively low temperature in those two sub-regions. The reference ET in Northern Xinjiang and Qilian-Hexi Region showed significant increasing trend, implies increasing demand of irrigation water for the major crops of per unit area in those regions. 4.2. The spatial distribution of irrigation water requirement Based on the above results of crop water requirement, we calculated and generated the distribution maps of irrigation water requirement in different periods (Fig. 6).
Fig. 5. Thiessen polygon division and farmland distribution in the study region, 2000.
60
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Table 2 Development of irrigation efficiency in the study region during past 20 years. Year
1990
1995
2000
2005
2010
Irrigation water efficiency
0.38
0.40
0.42
0.45
0.48
Table 3 Annual average crop water demands in the ARNWC and trends (1989–2010, unit: mm/yr). Region
Wheat
Corn
Cotton
Oilseed crops
Sugar beet
ET0
TRB MK trend
464 0.479
353 −0.085
679 0.085
310 0.197
749 0.31
1126 3.525**
NX MK trend
400 2.453*
495 2.397*
614 2.566*
365 2.679**
649 2.876**
1036 0.902
176 0.649
QH MK trend
519 2.34*
643 2.425*
731 2.285*
376 1.241
730 2.651**
1227 2.933**
163 0.931
Average MK trend
456 1.776
492 1.805
670 1.748
352 2.228*
710 2.228*
1123 2.735**
132 0.874
MK trend indicate the statistic value for Mann–Kendall test. * Significant at 0.05 level. ** Significance at 0.01 level.
Fig. 6. The spatial distribution of irrigation water demand in the ARNWC in 2010 (a) and its change compared to that in 1989 (b).
Precipitation 63 0.423
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
61
Fig. 7. The intra-annual variations of irrigation water demand in different sub-regions in the ARNWC (1989–2010) (a. Tarim Riber Basin; b. Northern Xinjiang; c. Qilian-Hexi Region).
By 2010, the crop water requirement in the study region amounted to 21.2 billion m3 , of which, the Northern Xinjiang, the Tarim River Basin and Qilian-Hexi Region were 9.2, 9.82 and 2.18 billion m3 , respectively. If the averaged irrigation efficiency can reach to 0.48 in this region, the irrigation water requirement for the major five crops amounted to 44.2 billion m3 in the northwestern arid region in 2010 (Fig. 6a). This number increased 12.9 billion m3 comparing with that in 1989. Areas with large irrigation water requirement were mainly in the oases Kuitun, Shihezi, Aksu, etc., where the cotton acreage accounted for more than half of the total acreage. Grain crops occupy the most part in the Qilian-Hexi Region, resulting in a relatively low irrigation requirement. Since 1989, the increase in crop water requirement caused mainly by temperature rising and the changes in crop planting structure, resulted in the overall increasing trend of irrigation water demand in the study region. The places with significant increase in irrigation water demand are the Oases located at both sides of the Tianshan Mountains and Kashgar in Tarim River Basin (Fig. 6b). This distribution pattern was consistent with cotton acreage during the past 20 years. The phenomenon indicated that the change in planting structure, especially the expansion of cotton growing area are the major reason to lead to the substantial increase in irrigation water demand both in time and space. In contrast, the planting structure in the Qilian-Hexi Region and Northern Xinjiang Altay region changed minor, so the increase in crop acreage did not result in a substantial increase of the total water demand of the regional irrigation. There are some places even showed obvious decrease in irrigation water demand, such as Urumqi, Shihezi, and other places with fast urbanization or desertification. Fast urbanization in past 20 years encroached large arable land, together with desertification at the fringe of oases, played important role in decreasing of irrigation land and resulted in decrease in irrigation water demand (Fig. 6b).
4.3. Monthly distribution of irrigation water requirement The intra-annual variation of irrigation water requirement in different sub-regions was calculated and plotted in Fig. 7. Monthly distributions of irrigation water demand in different regions were markedly different. In the Tarim River Basin, there were two peaks occurring in May and August, respectively, reflecting double cropping growing patterns. This is because the cropping systems were mainly one-season cotton and spring wheat and summer corn in this sub-region due to abundant heat resources. In May, winter wheat in the Tarim River Basin gradually entered to the jointing and heading stage, irrigation water demand increased rapidly and reached up to 2 billion m3 /month. Meanwhile, the cotton is at the seedling stage. In June, wheat is harvested, irrigation water demand is reduced, resulting in the first peak seen in this study. After June, water demand in Tarim River Basin was mainly due to
cotton and summer corn. In Tarim River Basin, cotton was in boll division and the flowering stage in August; meanwhile, the summer corn also entered the critical water demand period, and as a result, the irrigation requirement reached the maximum value of about 3.5 billion m3 /month. In Northern Xinjiang and Qilian-Hexi Region, one-season spring crops such as, wheat, corn, cotton, etc. grow, resulting only one peak of water demand in June and July, when irrigation water requirement can be peaked as much as 3.8 billion m3 /month in Northern Xingjiang and 1.3 billion m3 /month in Qilian-Hexi Region (Fig. 7). In Qilian-Hexi Region, the water demand of the local crops per unit area was much greater than that in Xinjiang. However, smaller growing areas determined the total irrigation requirement was less than the other 2 sub-regions. The major crops in Northern Xinjiang spring wheat and spring corn reached their maximum water requirements of around 1.0 billion m3 in June, when is about 1 month earlier than the water demand peak. Although the total water demand in Qilian-Hexi Region was much smaller than those in the Southern and Northern Xinjiang, relatively short and small rivers in Hexi region had the available water resources not as much as needed too, therefore, in this region the water shortage issue is actually more serious and caused widespread ecological degradation as well. 4.4. The inter-annual variation of irrigation requirement The variation of irrigation requirement for different crops in different sub-regions were calculated and illustrated in Fig. 8. The irrigation requirement has largely increased during the past 20 years because the reclamation and increase of crop growing area and the changes in planting structure since 1989. The total irrigation requirement in 2010 exceeded to 44.0 billion m3 , which was about 12.9 billion m3 larger than that in 1989. The irrigation requirement was low in 2003, consistent with the fact that crop growing area at that time was almost the fewest in recent years. Tarim River Basin shared more than 50% of the total irrigation water, whilst Qilian-Hexi Region only shared less than 13%. Irrigation demand in Tarim river basin was slightly greater than that in Northern Xinjiang mainly because of the large portion of cotton growing. If look at the water demand for different crops, grain crops wheat and maize shared more than half before 1997; and from then on, cotton became the largest water consumer. The sugar beet and oilseed only occupied around 10–12% of the total water withdrawal throughout the study period (Fig. 8). 4.5. Analysis of water demand and supply risks in some typical basins Agricultural land in the ARNWC mainly located at oases with irrigation sourced from river water, which mainly originated from
62
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 8. Inter annual variations of irrigation water requirement integrated for different crops and different sub-regions.
snow/glacier melting and summer rainfall in mountain areas. Climate change has large impacts on not only the quantity of river discharge but also river hydrograph in the northwestern arid basins in China. The latter is more important with regard to water supply. In this section, we analyze the water supply/demand situation of
three typical river basins, such as Yarkand, Hei, and Kaidu-Konqi River Basins, at both annual and seasonal bases. We compared the annual river discharge, which is considered as the maximum water resources available to society, with irrigation water requirement and found that at the annual base, the irrigation
Fig. 9. Intra-annual distribution of river discharge (Q) and irrigation water requirement (IWR) in three typical river basins (a, c, e); and the inter-annual variations of IWR and runoff in critical water demand period May (b, d, f). (a and b: Yarkand River; c and d: Hei River; e and f: Kaidu-Konqi River).
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
63
Table 4 Characteristics of river discharge and irrigation water demand in typical basins (billion m3 /yr). Year
1989–2000 2001–2010 Average
Yarkand River Basin
Hei River Basin
Kaidu-Konqi River Basin
Q
IWR
IWR/Q (%)
Q
IWR
IWR/Q (%)
Q
IWR
IWR/Q (%)
7.52 8.11 7.74
3.54 3.65 3.58
47.1 45 46.3
2.47 2.78 2.6
2.23 1.74 2.02
90.3 62.6 77.7
5.17 6.23 5.16
1.44 1.8 1.57
27.9 28.9 28.2
water demand is less than the annual river discharge in the three basins (Table 4). However, the ratio of irrigation water requirement to the runoff (IWR/R) was smaller than 0.4 in the Kaidu-Konqi River. According to water scarcity index (Falkenmark and Rockst¨rom, 2004), defined as the ratio of water use to water availability (WU/Q), when the use to availability ratio is greater than 0.4 the basin is recognized as severely water stressed basin. The IWR/R ratio for Yarkand and Hei River Basins are 0.46 and 0.78, respectively, indicating severe water stress exist in both river basins. In the period of 1989–2000, the ratio (IWR/R) had been amounted up to 0.90 in the Hei River Basin, and 0.47 in Yarkand River Basin; due to a combined effect of increasing in river discharge and the practice of water saving technologies, e.g. drip irrigation with vinyl-film mulching for cotton cultivation, the ratio of IWR/R in 2000 s was reduced to 0.63 and 0.45, respectively, in those two river basins. These basins are still facing to serious water scarcity, and the massive irrigation caused ecological degradation both in the oases and at the downstreams, such as salt accumulation in agricultural land (Hu et al., 2013) and riparian vegetation degradation (Chen et al., 2012). Fig. 9 illustrates the monthly irrigation demand and water supply in the three basins. Except for the Kaidu-Konqi River Basin, where the river discharge is much larger than the demand for all the months, Yarkand River encounters water deficit in April and May. Hei River encountered even more severe shortage in April through June (Fig. 9a, c, e). There is great seasonal water shortage in these two basins due to the inconsistent seasonal distribution of the runoff and irrigation water requirement. The maximum river discharge always occurs in July and August, and the irrigation water requirement also reached its maximum at the same period, but IWR are much less than the river discharge. Therefore, there are not water shortage problem in summer. However, during April–May, when is the critical period for winter wheat flowering and spring crops seedling such as spring wheat, corn, and cotton, water shortage in this period can largely reduce the production (Xiao et al., 2006). Due to still relatively low temperature at mountain area in April and May, the snow/glacier melting water cannot meet the irrigation requirement. Fig. 9 also shows the long-term comparison of irrigation water demand and water availability in May, the IWR is always larger than discharge in Yarkand River and Hei River from 1989 to 2010. Even though river discharge showed increasing trend in these two rivers, and the IWR in Hei River experienced dramatically decreasing trend. Seasonal water shortage has being existed in this period for these two basins. Therefore, pumping groundwater or constructing water reservoir are essential for bridging the deficit. 5. Conclusions Agricultural water consumption accounted for more than 95% of the total water use in the ARNWC, becoming the main factor affecting water resources allocation into socio-economic sectors and ecosystems in the region. However, due to absence of observation data, it is difficult to accurately estimate the irrigation water requirement in the ARNWC. Based on the Penman–Monteith equation and crop coefficient approach, we estimated the
irrigation water demand with combination of the remote sensing and census growing area data of 5 main corps during the past 22 years. The results suggested that irrigation water requirement in 2010 has reached 44.2 km3 , with a net increase of 12.9 km3 compared with that in 1989. Among the five major crops, cotton was the most water consuming crop in the region; the irrigation water demand for cotton reached to 20.7 km3 in 2010, accounting for 46.8% of the total water consumption in the year. In terms of water irrigation intensity, sugar beet consumed the most to 7100 m3 ha−1 y−1 , followed by cotton, corn, wheat and oilseed crops with 6700 m3 ha−1 y−1 , 4920 m3 ha−1 y−1 , 4560 m3 ha−1 y−1 , and 3520 m3 ha−1 y−1 , respectively. In some river basins, even though the annual water resource can meet the irrigation requirement, but there are huge deficit in some critical water demanding periods, such as in April, May, and even June in some specific river, i.e. Hei River. This kind of seasonal water shortage reinforced the water stress in the study region. On the other hand, the unprecedented enlargement of the crop growing area has resulted in rapid increase in irrigation water demand and already caused ecological degradation in the lower reaches of the rivers (Shen and Chen, 2010), such as the destination lake shrinkage and desiccation of river way in downstreams of Tarim river (Chen et al., 2011) and Hei river (Zhao et al., 2007). It is urgent to call wise water management such as water price regulation, crop planting pattern adjustment, as well as water saving technologies in this hyper arid region.
Acknowledgement This study was supported by the National Key Project on Basic Research (973) “Impacts of climate change on hydrological cycles and water resources security in the northwestern arid region of China (2010CB951003)”.
References Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: Guidelines for computing crop requirements. Irrigation and Drainage Paper No. 56, FAO, Rome, Italy, 300pp. Allen, R.G., Pereira, L.S., Smith, M., Raes, D., Wright, J.L., 2005. FAO-56 dual crop coefficient method for estimating evaporation from soil and application extensions. Journal of Irrigation and Drainage Engineering 131 (1), 2–13. Bureau of Xinjiang Water Resources, 2011. Xinjiang Water Resources Bulletin 2010., pp. 48. Chen, Y., Ye, Z., Shen, Y., 2011. Desiccation of the Tarim river, Xinjiang, China, and mitigation strategies. Quanternary International 244, 264–271. Chen, Y., Yang, Q., Luo, Y., Shen, Y., 2012. Ponder on the issues of water resources in the arid region of northwest China. Arid Land Geography 01, 1–9 (in Chinese with English abstract). Du, T., Kang, S., Zhang, J., Li, F., 2006. Yield and physiological responses of cotton to partial root-zone irrigation in the oasis field of northwest China. Agricultural Water Management 84 (1), 41–52. Duan, A., Sun, J., Liu, Y., Xiao, J., Liu, Q., Qi, X., 2004. Irrigation Quota of Major Crops for Northern China. China Agricultual Sci & Tech Press, Beijing, pp. 197. Er-Raki, S., Chehbouni, A., Guemouria, N., Duchemin, B., Ezzahar, J., 2007. Combining FAO-56 model and ground-based remote sensing to estimate water consumptions of wheat crops in a semi-arid region. Agricultural Water Management 87 (1), 41–54. Falkenmark, M., Rockström, J., 2004. Balancing Water for Human and Nature: An Ecohydrological Approach for Sustainability. Earthscan Publications, London, pp. 247.
64
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Hu, S., Shen, Y., Gan, Y., Xang, J., 2013. Effect of saline water drip irrigation on soil salinity and cotton growth in an oasis field. Ecohydrology, DOI: 10.1002/eco.1336. Geng, Y., Min, Q., Chen, S., 2006. Water Demand of Crop Plantation in Northwest China—A Case Study of the Jinghe Watershed. Journal of Ecology and Rural Environment 04, 30–34 (in Chinese with English abstract). Luo, X., Yang, J., 2003. Researches on the Questions and Countermeasures of Sustainable Utilization of Water Resources in the Northwest Area of China. Areal Research and Development 01, 73–76 (in Chinese with English abstract). Liu, X., Li, Y., Hao, W., 2005. Trend and causes of water requirement of main crops in North China in recent 50 years. Transactions of The Chinese Society of Agricultural Engineering 10, 155–159 (in Chinese with English abstract). Ma, L., Zhan, C., Tang, L., Jiang, X., 2005. Review and prospect on the study progress in crop water requirements. Arid Land Geography 04, 115–121 (in Chinese with English abstract). Mohamed, Y., Bastiaanssen, W., Savenije, H., 2004. Spatial variability of evaporation and moisture storage in the swamps of the upper Nile studied by remote sensing techniques. Journal of hydrology 289 (1), 145–164. Sakellariou-Makrantonaki, M., Vagenas, I., 2006. Mapping crop evapotranspiration and total crop water requirements estimation in Central Greece. European Water 13/14, 3–13. Shahid, S., 2011. Impact of climate change on irrigation water demand of dry season Boro rice in northwest Bangladesh. Climatic change 105 (3), 433–453.
Shen, Y., Chen, Y., 2010. Global perspective on hydrology, water balance, and water resources management in arid basins. Hydrol Processes 24 (2), 129–135. Wu, X., Zhou, H., 2011. Influence factor to water transport index in canal system in Xinjiang Irrigation area. Yellow River 09, 93–97 (in Chinese with English abstract). Xiao, J., Liu, Z., Duan, A., Liu, Z., Li, X., 2006. Studies on effects of irrigation systems on the grain yield constituents and water use efficiency of winter wheat. Journal of Irrigation and Drainage 02, 20–23 (in Chinese with English abstract). Xu, X., Zhou, H., Wang, Z., 2010. Study on effective rainfall use efficiency in arid irrigation district. Water Saving Irrigation 12, 44–46 + 50 (in Chinese with English abstract). Xu, C., Yang, X., Li, Y., Wang, W., 2011. Changes of China agricultural climate resources under the background of climate change III. Spatiotemporal change characteristics of agricultural climate resources in Northwest Arid Area. Chinese Journal of Applied Ecology 03, 763–772 (in Chinese with English abstract). Zhao, W., Chang, X., He, Z., Zhang, Z., 2007. Study on vegetation ecological water requirement in Ejina Oasis. Science in China Series D: Earth Sciences 50 (1), 121–129. Zhu, Q., Zhang, W., 2005. Topography-based spatial daily precipitation interpolation by means of PRISM and Thiessen polygon analysis. Scientia Geographica Sinica 02, 233–238 (in Chinese with English abstract).
Contents lists available at SciVerse ScienceDirect
Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat
Estimation of regional irrigation water requirement and water supply risk in the arid region of Northwestern China 1989–2010 Yanjun Shen a,∗ , Shuo Li a,b , Yaning Chen c , Yongqing Qi a , Shuowei Zhang d a Key Laboratory for Agricultural Water Resources, Center for Agricultural Resources Research, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Shijiazhuang 050021, China b College of Resources and Environmental Sciences, Hebei Normal University, Shijiazhuang 050016, China c State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China d Department of Geography, The State University of New York-Buffalo, Buffalo, USA
a r t i c l e
i n f o
Article history: Received 9 November 2012 Accepted 21 June 2013 Keywords: Irrigation water requirement Crop water requirement Water resources Penman–Monteith equation Water supply risk Arid region in Northwestern China (ARNWC)
a b s t r a c t Water use in agricultural sector shares more than 90% of the total water withdrawal in the arid region of Northwestern China (hereafter, ARNWC). Irrigation water demand is therefore essential to the water resources allocation to economy and natural ecosystems in the highly water deficit region. In this study, we analyzed the spatial and temporal variations of irrigation water demand as well as crop water requirement by combining the modified Penman–Monteith equation recommended by FAO and GIS technology. Crop and irrigation water requirements for 5 main crops, including wheat, corn, cotton, oilseed and sugar beet, from 1989 to 2010 were calculated and the spatio-temporal variations were analyzed. The results suggested that the demand of irrigation water in the ARNWC showed increasing trend during the past two decades, which mainly caused by fast increase in cotton cultivation areas, because irrigation water requirement for cotton was much larger than the other crops. The changes in cotton growing area significantly affected the spatial pattern of water demand. A total of 44.2 billion m3 water was withdrawn for irrigation in year 2010. Larger amount of water was consumed for crops in Northern Xinjiang and Tarim River Basin than Qilian-Hexi region. Irrigation water requirement reaches its maximum in July and August. It is revealed that the critical period for water supply is during April and May through comparing the monthly irrigation water requirement with water availability, i.e. river discharge. Even though the annual water resources are much larger than the requirement, but for some basins, there is severe physical water shortage during the critical water use period in April and May. The water resource supply is expected to be facing more difficulties in future. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Water is one of the essential resources in arid and semi-arid regions especially for agriculture. The arid region in Northwestern China (ARNWC) is one of the mostly water stressed regions in the world (Shen and Chen, 2010). This region is also an important foodand cotton-producing region in China, but the imbalance between water supply and demand is also very prominent in the region. Agricultural irrigation consumed most water resources in this region, and accounted for 91.8% of the total water consumption (Geng et al., 2006). The shortage of water resources has become a major limiting factor for the socio-economic development in the ARNWC. Even in places with relatively abundant water resources, it is not sufficient to meet the water demand of crop growth all the year round. With increasing acreage of crop growing, unreasonable
∗ Corresponding author. Tel.: +86 311 85825464; fax: +86 311 85815093. E-mail address: [email protected] (Y. Shen). 0378-3774/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.agwat.2013.06.014
crop planting structure, high irrigation quota, and low water use efficiency, the shortage of water resources is becoming increasingly serious. Therefore, quantifying agricultural irrigation water demand in the region, and analyzing the spatial and temporal characteristics of the irrigation water demand, are of importance to help improve the water management toward sustainable use. Currently, there are several methods to calculate crop irrigation water requirement, including ground observation (based on crop planting structure and irrigation model), the crop model method, the remote sensing method, and the Penman–Monteith method recommended by the UN Food and Agriculture Organization (FAO). Ground observation can help calculating accurate water consumption estimation at farmland scale, but can hardly estimate the water requirement in a region. Remote sensing method always combines with ground meteorological observations and crop survey to estimate crop water demand (Ma et al., 2005; Mohamed et al., 2004). However, due to the error originated from the spatial and temporal resolutions of remote sensing data, there are still many technical issues need further, e.g. the interpolation method from
56
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 1. Geomorphologic setting and zoning of the ARNWC. The embedded figure shows the distribution of annual precipitation, precipitation in the study area is generally less than 200 mm/yr.
instantaneous satellite image to daily or seasonal water consumption. The Penman–Monteith method has a simple equation and better physical basis, and been widely used in the estimation of crop water requirement. Liu et al. (2005) calculated water requirements of major crops in North China using the Penman–Monteith method. Xiao et al. (2006) calculated and analyzed the spatiotemporal characteristics of water demand of corn in China. Er-Raki et al. (2007) studied water demand of wheat in Morocco using the Penman–Monteith equation combined with land surface remote sensing data. Du et al. (2006) studied the response of irrigation to cotton production and physiology in the oasis in Northwestern China. Although the above studies looked at the water requirement of unit area crop, they did not consider the impact of rainfall on crop irrigation water demand, and could not quantify the total water demand at a regional scale, especially, in the ARNWC. In this paper, based on Penman–Monteith method and combined ground observation data of crop growing acreage, we calculated and analyzed the temporal and spatial variations of agricultural irrigation water demand in the ARNWC. The objectives of this study are: (1) to give a whole picture of spatial and temporal changes of irrigation water demand and the water availability and (2) to provide quantitative information for spatial irrigation water demand as well as the water demand structure for different crops in the hyper arid region of China. We hope these results can help improve the water resources management in this highly water deficit region. 2. The ARNWC The ARNWC is located in the central part of Eurasian continent, including all of Xinjiang, Hexi Corridor in Gansu and the west of the Helan Mountain of Inner Mongolia; its geographical location is between longitude 73–125◦ E and latitude 35–50◦ N (Luo and Yang, 2003); the total area is approximately 2.02 million square kilometers, accounting for more than one fifth of China’s total terrestrial area. The Kunlun–Altun–Qilian Mountains are the south boundary of the area (Fig. 1), which combined with the Qinghai-Tibet Plateau, blocks the vapor from Indian Ocean. Altai Mountain is the northern boundary, blocking the vapor from the Arctic Ocean. Tianshan Mountains lie in the middle of the region and form the landscapes
of alpine, inland basins, gobi desert and widespread sandy desert. This inland region is controlled by continental climate, with little rainfall throughout the year and strong evaporative potential. The average annual precipitation is only 130 mm, with uneven spatial and temporal distributions. The highest rainfall occurs at the eastern end of the Tianshan Mountains in the Ili River Valley, which could amount up to 800 mm/yr or above; annual precipitation is generally less than 50 mm in the oasis areas, where most agricultural activities exist. However, the potential evaporation in this arid region can be as large as 3200 mm/yr, 8–10 times the annual rainfall (Shen and Chen, 2010; Chen et al., 2012). The hyper arid region is thus one of the most arid regions in the world (Shen and Chen, 2010). Abundant radiation and heat resources in this region make it an ideal place for thermophilous and photophilous crops growth, so it becomes an important high-quality and also high-yield cotton production zone (Xu et al., 2011). Other crops like wheat and corn are also widely cultivated in this region. The arable land is mainly distributed in oases embedded at the edge of the deserts, relying on surface water and groundwater for irrigation. At the end of 2010, the total area of farmland was 4.9 million hectares, effective irrigation reached 4.682 million hectares, and the total crop growing area amounted to 5.55 million ha. Cotton growing area was 1.5 million ha, and its acreage and production accounted for 31% and 42.7% of the country’s total, respectively. According to Bureau of Xinjiang Water Resources (2011), the annual withdrawal for irrigation reached to 48.83 billion m3 , accounting for 91.3% of the total water consumption in Xinjiang in 2010. The irrigated area in northwestern arid regions increased from 2.7 million hectares in 1989 to about 4.4 million hectares (Fig. 2) in 2010, and the fastest increase goes to cotton growing land. In the irrigated farmland, wheat and corn growing acreage was about 2 million hectares, with little change in the total over the past 20 years. Beet and oilseed acreage was approximately 0.4–0.5 million hectares, which slightly declined in recent years. The cotton growing area showed the most fast increase, from only 0.4 million hectares in 1989, mainly distributed in Tarim River Basin, to 1.5 million hectares by 2008. In 2010, the cotton growing area slightly declined to 1.3 million hectares, still accounting for ∼1/3 of the total cropping areas.
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 2. The variations of irrigated crop growing area in ARNWC.
Fig. 3 shows the spatial distribution and variation of growing area of the five major crops from 1989 to 2010. In the Tarim River Basin, double crops wheat–corn or single crops cotton are the major cropping types; while in Northern Xinjiang and Qilian-Hexi Region, due to heat constraints single crop such as spring wheat or spring
57
corn are mainly grown. The north and south slopes of the Tianshan Mountains is the transition zone, where sugar beet and oilseed are mainly grown. Fig. 3 also reflects changes in the structure of the crops in the past 20 years. The crops grown in 1989 were dominated by wheat and corn. And cotton acreage was mainly distributed in several big irrigation districts such as Shihezi, Aksu, Kashi, etc., and the acreage accounted only for 10.4% of the total growing area, far less than food crops like wheat and corn. Oilseed, sugar beet and other cash crops were mainly planted in the Northern Xinjiang’s Ili, Tacheng and Gansu’s Minle region. However, by the year 2000, with the improvement of irrigation facilities and the increase of cotton procurement price, cotton acreage rapidly increased to 1045 thousand hectares, accounting for 25.6% of the total sowing areas; the Northern Tianshan Mountain and the oases in the edge of the Tarim Basin increased most significantly. Cotton was patchily distributed in Gansu’s Dunhuang and Yumen region. The changes of oilseed and sugar beet were minor. By the end of 2010, due to the water restrictions, the cotton area declined throughout the study region. The main crops became cotton, wheat, and corn; oilseed, sugar beet, and other cash crops scattered in northern part near Altay Mountains and central Hexi Corridor in Gansu (Fig. 3).
Fig. 3. Spatial distribution of crop growing acreage in the ARNWC in different periods (1989–2010). The red, green, and blue color composition indicates the acreage percentage of the three types of cultivation, wheat/corn, cotton, and beet/oilseed. (For interpretation of the references to color in this artwork, the reader is referred to the web version of the article.)
58
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 4. Crop coefficient (Kc ) values for normal/spring crops (a) and winter crops (b).
3. Methods 3.1. Data Daily meteorological observations are collected from China Meteorological Administration (CMA). The main meteorological variables include the maximum and minimum temperatures, wind speed, relative humidity and sunshine duration. Observations on crop growth and phenology are collected from several agricultural meteorological stations. According to the characteristics of the time series and spatial locations, 67 meteorological stations were selected for the calculation and analysis in this study. Monthly river discharge data from the gauges at mountain foot were selected for analyzing the supply and demand situation of each basin. Socio-economic statistical data were collected from the Statistical Yearbook of the provinces, municipalities and prefectures (1989–2010) in the study regions, including arable land at the county level and the various crops sown. Land use data for years 1990 and 2000 were provided by the Data Center of Resources and Environment, Chinese Academy of Sciences. These data were combined with socio-economic data and statistics to create the spatial distributions of crops sowing area in different years. 3.2. Calculation of crop irrigation requirement Crop irrigation requirement is calculated using the following formula: CWR = ETc − Pe
(1)
ETc = Kc · ET0
(2)
IWR =
S · CWR Ic
(3)
where CWR is the crop water requirement of a certain crop; ET0 is the reference evapotranspiration; Kc is the crop coefficient; S is the acreage of the crop; Pe is the effective rainfall, which indicate the rainfall transferred to soil moisture; IWR is the irrigation water requirement of the crop; Ic is the irrigation efficiency, by considering the loss in water diversion to field from river channel. CWR can vary widely and is affected by crop variety, soil property, and aquifer conditions, weather conditions, and other factors (Sakellariou-Makrantonaki, 2006). 3.3. Reference evapotranspiration Reference evapotranspiration (ET0 ) was calculated using the modified Penman–Monteith equation recommended by United Nations Food and Agriculture Organization (FAO) (Allen et al.,
1998). Reference ET was defined and calculated based on the evapotranspiration of well watered short grass with 0.12 m in height, resistance of 70 s/m, and the albedo of 0.23. This is very similar to the open surface, highly consistent, vigorously growing, ground cover without the water shortage (Allen et al., 2005). Reference evapotranspiration is only related to meteorological factors (Shahid, 2011), so it can be used in different regions, and has universal significance for inter-comparison. The reference ET was calculated using the following equation: ET0 =
0.408 (Rn − G) + (900/(T + 273))u2 (es − ea ) + (1 + 0.34u2 )
(4)
where ET0 is the r eference evapotranspiration (mm/d) Rn is the net radiation at the land surface [MJ/(m2 d)]; G is the soil heat flux density [MJ/(m2 d)]; u2 is 2 m height average wind speed in 24 h (m/s); es is the saturation vapor pressure (kPa); ea is actual vapor pressure (kPa); is slope of vapor pressure–temperature curve [kPa/◦ C]; and is psychrometric constant [kPa/◦ C] (Er-Raki et al., 2007). 3.4. Crop coefficient during the growing period Crop coefficient is the ratio of well watered crop evapotranspiration and the reference ET, i.e. ETc /ET0 , commonly written as “Kc ”. Kc value generally can be obtained through field experiments. The Kc values reflect the relative water consumption capacity of crop during different growing stages of different crops. In general, the crop coefficient varies during four developing stages: the initial growth period, the rapid development period, mid-maternity, and mature period (as illustrated in Fig. 4). Kc value is generally very small at the initial growth phase because of low crop cover and most water consumption is from soil evaporation. During the rapid development period, leaf area increases rapidly and causes a rapid increase in the Kc value. The Kc value reached maximum when crops fully developed, and remained stable over a period of time. As the crop matured, the Kc value start to decreasing due to foliage senescence. In the current study, we used the crop coefficients according to Duan et al. (2004). The crop coefficients for some typical subregions are shown in Table 1. 3.5. Effective rainfall The effective rainfall at agricultural land is defined as the portion of rainfall penetrating canopy, infiltrated in soil layer and stored in the crop root zone, which can be used later by crops as transpiration. It does not include the interception of plant canopy, surface runoff and the amount of water percolating to the layer below the root zone. Effective rainfall is usually calculated as follows: Pe = P × ı
(5)
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
59
Table 1 Crop coefficients of the major crops in different growth stages in some sites in ARNWC. Sub-region
Crop
Development stage Initial
Developing
Middle
Sowing (mm/dd)
Harvesting (mm/dd)
Total (days)
Late
Yanqi, Xinjiang
Spring wheat Spring corn Cotton Oilseed
0.17 0.21 0.26 0.31
0.17–1.16 0.21–1.20 0.26–1.20 0.31–1.15
1.16 1.20 1.20 1.15
1.16–0.40 1.20–0.35 1.20–0.70 1.15–0.35
03/07 04/27 04/20 06/14
07/05 09/17 10/16 09/29
121 144 180 108
Zhangye, Gansu
Sugar beet Cotton Spring wheat Spring corn
0.34 0.23 0.23 0.23
0.34–1.21 0.23–1.21 0.23–1.16 0.23–1.20
1.21 1.21 1.16 1.20
0.21–0.70 1.21–0.72 1.16–0.40 1.20–0.35
03/15 04/11 03/13 04/13
09/24 10/11 07/17 09/26
194 184 127 167
where P is the precipitation; ı is an empirically effective utilization coefficient of rainfall, which is adopted as 0.52 in our study region (Xu et al., 2010). 3.6. Irrigation efficiency Irrigation efficiency refers to the ratio of the net amount of water used by crops to the total amount of water withdrawal from river channels or reservoirs. It is mainly composed of the canal water utilization coefficient and the field water utilization coefficient. The canal water utilization coefficient is used to measure the degree of utilization of canal water delivery; field water utilization coefficient is used to measure the degree of utilization of soil water (Wu and Zhou, 2011). Improvements in irrigation infrastructure as well as the development of water-saving technologies have led to an obvious improvement of irrigation efficiency in the ARNWC (Table 2). According to the “China Water Conservancy Yearbook” and the regional water resources bulletin, the irrigation water utilization coefficient has increased from 0.38 to 0.48 during past 20 years (Table 2). In this paper, the irrigation water utilization coefficients (Table 2) were used to calculate irrigation requirement. 3.7. Spatial interpolation In this study, we employed the Thiessen polygon method for spatial interpolation of the crop water requirement calculated based on the ground meteorological observations at 67 stations. Point to the surface interpolation methods included the inverse distance weighting method, the spline function method and the kriging method (Zhu and Zhang, 2005). In many areas these methods are able to better reflect regional differences. In the study region, the three mountains split the region into two large inland depressions, where widespread deserts with arable land and oases scattered along the mountain foot fringes. This unique topography
makes natural and geographical conditions of the arable land in each watershed (such as altitude, weather and soil) similar, so is the crop water demand. We used the simple Thiessen polygon for dimensional interpolation in this study. The distribution of crop growing area is rasterized by combining the statistics and land use maps. The Thiessen polygon division and arable land distribution of the study region are shown in Fig. 5. 4. Results 4.1. Crop water requirement and its trend We calculated the crop water requirement for the five main crops from 1989 to 2010, and detected the trend using the Mann–Kendall test (Table 3). Table 3 shows that sugar beet is the highest water consumption crop with annual CWR of 710 mm, and oilseed is the lowest with CWR of 352 mm. There are some differences among the spatial distributions of regional crop irrigation requirement (Table 3) due to the difference of the geographical conditions of the various sub-regions. Based on the Mann–Kendall test, it was found that the annual water requirement of the five major crops over the past 20 years showed increasing trend, especially of oilseeds and sugar beet. The increase of CWR is not significant for all the crops in Tarim Basin, but significant in Northern Xinjiang (NX) and Hexi Region (QH), mainly due to the relatively low temperature in those two sub-regions. The reference ET in Northern Xinjiang and Qilian-Hexi Region showed significant increasing trend, implies increasing demand of irrigation water for the major crops of per unit area in those regions. 4.2. The spatial distribution of irrigation water requirement Based on the above results of crop water requirement, we calculated and generated the distribution maps of irrigation water requirement in different periods (Fig. 6).
Fig. 5. Thiessen polygon division and farmland distribution in the study region, 2000.
60
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Table 2 Development of irrigation efficiency in the study region during past 20 years. Year
1990
1995
2000
2005
2010
Irrigation water efficiency
0.38
0.40
0.42
0.45
0.48
Table 3 Annual average crop water demands in the ARNWC and trends (1989–2010, unit: mm/yr). Region
Wheat
Corn
Cotton
Oilseed crops
Sugar beet
ET0
TRB MK trend
464 0.479
353 −0.085
679 0.085
310 0.197
749 0.31
1126 3.525**
NX MK trend
400 2.453*
495 2.397*
614 2.566*
365 2.679**
649 2.876**
1036 0.902
176 0.649
QH MK trend
519 2.34*
643 2.425*
731 2.285*
376 1.241
730 2.651**
1227 2.933**
163 0.931
Average MK trend
456 1.776
492 1.805
670 1.748
352 2.228*
710 2.228*
1123 2.735**
132 0.874
MK trend indicate the statistic value for Mann–Kendall test. * Significant at 0.05 level. ** Significance at 0.01 level.
Fig. 6. The spatial distribution of irrigation water demand in the ARNWC in 2010 (a) and its change compared to that in 1989 (b).
Precipitation 63 0.423
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
61
Fig. 7. The intra-annual variations of irrigation water demand in different sub-regions in the ARNWC (1989–2010) (a. Tarim Riber Basin; b. Northern Xinjiang; c. Qilian-Hexi Region).
By 2010, the crop water requirement in the study region amounted to 21.2 billion m3 , of which, the Northern Xinjiang, the Tarim River Basin and Qilian-Hexi Region were 9.2, 9.82 and 2.18 billion m3 , respectively. If the averaged irrigation efficiency can reach to 0.48 in this region, the irrigation water requirement for the major five crops amounted to 44.2 billion m3 in the northwestern arid region in 2010 (Fig. 6a). This number increased 12.9 billion m3 comparing with that in 1989. Areas with large irrigation water requirement were mainly in the oases Kuitun, Shihezi, Aksu, etc., where the cotton acreage accounted for more than half of the total acreage. Grain crops occupy the most part in the Qilian-Hexi Region, resulting in a relatively low irrigation requirement. Since 1989, the increase in crop water requirement caused mainly by temperature rising and the changes in crop planting structure, resulted in the overall increasing trend of irrigation water demand in the study region. The places with significant increase in irrigation water demand are the Oases located at both sides of the Tianshan Mountains and Kashgar in Tarim River Basin (Fig. 6b). This distribution pattern was consistent with cotton acreage during the past 20 years. The phenomenon indicated that the change in planting structure, especially the expansion of cotton growing area are the major reason to lead to the substantial increase in irrigation water demand both in time and space. In contrast, the planting structure in the Qilian-Hexi Region and Northern Xinjiang Altay region changed minor, so the increase in crop acreage did not result in a substantial increase of the total water demand of the regional irrigation. There are some places even showed obvious decrease in irrigation water demand, such as Urumqi, Shihezi, and other places with fast urbanization or desertification. Fast urbanization in past 20 years encroached large arable land, together with desertification at the fringe of oases, played important role in decreasing of irrigation land and resulted in decrease in irrigation water demand (Fig. 6b).
4.3. Monthly distribution of irrigation water requirement The intra-annual variation of irrigation water requirement in different sub-regions was calculated and plotted in Fig. 7. Monthly distributions of irrigation water demand in different regions were markedly different. In the Tarim River Basin, there were two peaks occurring in May and August, respectively, reflecting double cropping growing patterns. This is because the cropping systems were mainly one-season cotton and spring wheat and summer corn in this sub-region due to abundant heat resources. In May, winter wheat in the Tarim River Basin gradually entered to the jointing and heading stage, irrigation water demand increased rapidly and reached up to 2 billion m3 /month. Meanwhile, the cotton is at the seedling stage. In June, wheat is harvested, irrigation water demand is reduced, resulting in the first peak seen in this study. After June, water demand in Tarim River Basin was mainly due to
cotton and summer corn. In Tarim River Basin, cotton was in boll division and the flowering stage in August; meanwhile, the summer corn also entered the critical water demand period, and as a result, the irrigation requirement reached the maximum value of about 3.5 billion m3 /month. In Northern Xinjiang and Qilian-Hexi Region, one-season spring crops such as, wheat, corn, cotton, etc. grow, resulting only one peak of water demand in June and July, when irrigation water requirement can be peaked as much as 3.8 billion m3 /month in Northern Xingjiang and 1.3 billion m3 /month in Qilian-Hexi Region (Fig. 7). In Qilian-Hexi Region, the water demand of the local crops per unit area was much greater than that in Xinjiang. However, smaller growing areas determined the total irrigation requirement was less than the other 2 sub-regions. The major crops in Northern Xinjiang spring wheat and spring corn reached their maximum water requirements of around 1.0 billion m3 in June, when is about 1 month earlier than the water demand peak. Although the total water demand in Qilian-Hexi Region was much smaller than those in the Southern and Northern Xinjiang, relatively short and small rivers in Hexi region had the available water resources not as much as needed too, therefore, in this region the water shortage issue is actually more serious and caused widespread ecological degradation as well. 4.4. The inter-annual variation of irrigation requirement The variation of irrigation requirement for different crops in different sub-regions were calculated and illustrated in Fig. 8. The irrigation requirement has largely increased during the past 20 years because the reclamation and increase of crop growing area and the changes in planting structure since 1989. The total irrigation requirement in 2010 exceeded to 44.0 billion m3 , which was about 12.9 billion m3 larger than that in 1989. The irrigation requirement was low in 2003, consistent with the fact that crop growing area at that time was almost the fewest in recent years. Tarim River Basin shared more than 50% of the total irrigation water, whilst Qilian-Hexi Region only shared less than 13%. Irrigation demand in Tarim river basin was slightly greater than that in Northern Xinjiang mainly because of the large portion of cotton growing. If look at the water demand for different crops, grain crops wheat and maize shared more than half before 1997; and from then on, cotton became the largest water consumer. The sugar beet and oilseed only occupied around 10–12% of the total water withdrawal throughout the study period (Fig. 8). 4.5. Analysis of water demand and supply risks in some typical basins Agricultural land in the ARNWC mainly located at oases with irrigation sourced from river water, which mainly originated from
62
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Fig. 8. Inter annual variations of irrigation water requirement integrated for different crops and different sub-regions.
snow/glacier melting and summer rainfall in mountain areas. Climate change has large impacts on not only the quantity of river discharge but also river hydrograph in the northwestern arid basins in China. The latter is more important with regard to water supply. In this section, we analyze the water supply/demand situation of
three typical river basins, such as Yarkand, Hei, and Kaidu-Konqi River Basins, at both annual and seasonal bases. We compared the annual river discharge, which is considered as the maximum water resources available to society, with irrigation water requirement and found that at the annual base, the irrigation
Fig. 9. Intra-annual distribution of river discharge (Q) and irrigation water requirement (IWR) in three typical river basins (a, c, e); and the inter-annual variations of IWR and runoff in critical water demand period May (b, d, f). (a and b: Yarkand River; c and d: Hei River; e and f: Kaidu-Konqi River).
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
63
Table 4 Characteristics of river discharge and irrigation water demand in typical basins (billion m3 /yr). Year
1989–2000 2001–2010 Average
Yarkand River Basin
Hei River Basin
Kaidu-Konqi River Basin
Q
IWR
IWR/Q (%)
Q
IWR
IWR/Q (%)
Q
IWR
IWR/Q (%)
7.52 8.11 7.74
3.54 3.65 3.58
47.1 45 46.3
2.47 2.78 2.6
2.23 1.74 2.02
90.3 62.6 77.7
5.17 6.23 5.16
1.44 1.8 1.57
27.9 28.9 28.2
water demand is less than the annual river discharge in the three basins (Table 4). However, the ratio of irrigation water requirement to the runoff (IWR/R) was smaller than 0.4 in the Kaidu-Konqi River. According to water scarcity index (Falkenmark and Rockst¨rom, 2004), defined as the ratio of water use to water availability (WU/Q), when the use to availability ratio is greater than 0.4 the basin is recognized as severely water stressed basin. The IWR/R ratio for Yarkand and Hei River Basins are 0.46 and 0.78, respectively, indicating severe water stress exist in both river basins. In the period of 1989–2000, the ratio (IWR/R) had been amounted up to 0.90 in the Hei River Basin, and 0.47 in Yarkand River Basin; due to a combined effect of increasing in river discharge and the practice of water saving technologies, e.g. drip irrigation with vinyl-film mulching for cotton cultivation, the ratio of IWR/R in 2000 s was reduced to 0.63 and 0.45, respectively, in those two river basins. These basins are still facing to serious water scarcity, and the massive irrigation caused ecological degradation both in the oases and at the downstreams, such as salt accumulation in agricultural land (Hu et al., 2013) and riparian vegetation degradation (Chen et al., 2012). Fig. 9 illustrates the monthly irrigation demand and water supply in the three basins. Except for the Kaidu-Konqi River Basin, where the river discharge is much larger than the demand for all the months, Yarkand River encounters water deficit in April and May. Hei River encountered even more severe shortage in April through June (Fig. 9a, c, e). There is great seasonal water shortage in these two basins due to the inconsistent seasonal distribution of the runoff and irrigation water requirement. The maximum river discharge always occurs in July and August, and the irrigation water requirement also reached its maximum at the same period, but IWR are much less than the river discharge. Therefore, there are not water shortage problem in summer. However, during April–May, when is the critical period for winter wheat flowering and spring crops seedling such as spring wheat, corn, and cotton, water shortage in this period can largely reduce the production (Xiao et al., 2006). Due to still relatively low temperature at mountain area in April and May, the snow/glacier melting water cannot meet the irrigation requirement. Fig. 9 also shows the long-term comparison of irrigation water demand and water availability in May, the IWR is always larger than discharge in Yarkand River and Hei River from 1989 to 2010. Even though river discharge showed increasing trend in these two rivers, and the IWR in Hei River experienced dramatically decreasing trend. Seasonal water shortage has being existed in this period for these two basins. Therefore, pumping groundwater or constructing water reservoir are essential for bridging the deficit. 5. Conclusions Agricultural water consumption accounted for more than 95% of the total water use in the ARNWC, becoming the main factor affecting water resources allocation into socio-economic sectors and ecosystems in the region. However, due to absence of observation data, it is difficult to accurately estimate the irrigation water requirement in the ARNWC. Based on the Penman–Monteith equation and crop coefficient approach, we estimated the
irrigation water demand with combination of the remote sensing and census growing area data of 5 main corps during the past 22 years. The results suggested that irrigation water requirement in 2010 has reached 44.2 km3 , with a net increase of 12.9 km3 compared with that in 1989. Among the five major crops, cotton was the most water consuming crop in the region; the irrigation water demand for cotton reached to 20.7 km3 in 2010, accounting for 46.8% of the total water consumption in the year. In terms of water irrigation intensity, sugar beet consumed the most to 7100 m3 ha−1 y−1 , followed by cotton, corn, wheat and oilseed crops with 6700 m3 ha−1 y−1 , 4920 m3 ha−1 y−1 , 4560 m3 ha−1 y−1 , and 3520 m3 ha−1 y−1 , respectively. In some river basins, even though the annual water resource can meet the irrigation requirement, but there are huge deficit in some critical water demanding periods, such as in April, May, and even June in some specific river, i.e. Hei River. This kind of seasonal water shortage reinforced the water stress in the study region. On the other hand, the unprecedented enlargement of the crop growing area has resulted in rapid increase in irrigation water demand and already caused ecological degradation in the lower reaches of the rivers (Shen and Chen, 2010), such as the destination lake shrinkage and desiccation of river way in downstreams of Tarim river (Chen et al., 2011) and Hei river (Zhao et al., 2007). It is urgent to call wise water management such as water price regulation, crop planting pattern adjustment, as well as water saving technologies in this hyper arid region.
Acknowledgement This study was supported by the National Key Project on Basic Research (973) “Impacts of climate change on hydrological cycles and water resources security in the northwestern arid region of China (2010CB951003)”.
References Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: Guidelines for computing crop requirements. Irrigation and Drainage Paper No. 56, FAO, Rome, Italy, 300pp. Allen, R.G., Pereira, L.S., Smith, M., Raes, D., Wright, J.L., 2005. FAO-56 dual crop coefficient method for estimating evaporation from soil and application extensions. Journal of Irrigation and Drainage Engineering 131 (1), 2–13. Bureau of Xinjiang Water Resources, 2011. Xinjiang Water Resources Bulletin 2010., pp. 48. Chen, Y., Ye, Z., Shen, Y., 2011. Desiccation of the Tarim river, Xinjiang, China, and mitigation strategies. Quanternary International 244, 264–271. Chen, Y., Yang, Q., Luo, Y., Shen, Y., 2012. Ponder on the issues of water resources in the arid region of northwest China. Arid Land Geography 01, 1–9 (in Chinese with English abstract). Du, T., Kang, S., Zhang, J., Li, F., 2006. Yield and physiological responses of cotton to partial root-zone irrigation in the oasis field of northwest China. Agricultural Water Management 84 (1), 41–52. Duan, A., Sun, J., Liu, Y., Xiao, J., Liu, Q., Qi, X., 2004. Irrigation Quota of Major Crops for Northern China. China Agricultual Sci & Tech Press, Beijing, pp. 197. Er-Raki, S., Chehbouni, A., Guemouria, N., Duchemin, B., Ezzahar, J., 2007. Combining FAO-56 model and ground-based remote sensing to estimate water consumptions of wheat crops in a semi-arid region. Agricultural Water Management 87 (1), 41–54. Falkenmark, M., Rockström, J., 2004. Balancing Water for Human and Nature: An Ecohydrological Approach for Sustainability. Earthscan Publications, London, pp. 247.
64
Y. Shen et al. / Agricultural Water Management 128 (2013) 55–64
Hu, S., Shen, Y., Gan, Y., Xang, J., 2013. Effect of saline water drip irrigation on soil salinity and cotton growth in an oasis field. Ecohydrology, DOI: 10.1002/eco.1336. Geng, Y., Min, Q., Chen, S., 2006. Water Demand of Crop Plantation in Northwest China—A Case Study of the Jinghe Watershed. Journal of Ecology and Rural Environment 04, 30–34 (in Chinese with English abstract). Luo, X., Yang, J., 2003. Researches on the Questions and Countermeasures of Sustainable Utilization of Water Resources in the Northwest Area of China. Areal Research and Development 01, 73–76 (in Chinese with English abstract). Liu, X., Li, Y., Hao, W., 2005. Trend and causes of water requirement of main crops in North China in recent 50 years. Transactions of The Chinese Society of Agricultural Engineering 10, 155–159 (in Chinese with English abstract). Ma, L., Zhan, C., Tang, L., Jiang, X., 2005. Review and prospect on the study progress in crop water requirements. Arid Land Geography 04, 115–121 (in Chinese with English abstract). Mohamed, Y., Bastiaanssen, W., Savenije, H., 2004. Spatial variability of evaporation and moisture storage in the swamps of the upper Nile studied by remote sensing techniques. Journal of hydrology 289 (1), 145–164. Sakellariou-Makrantonaki, M., Vagenas, I., 2006. Mapping crop evapotranspiration and total crop water requirements estimation in Central Greece. European Water 13/14, 3–13. Shahid, S., 2011. Impact of climate change on irrigation water demand of dry season Boro rice in northwest Bangladesh. Climatic change 105 (3), 433–453.
Shen, Y., Chen, Y., 2010. Global perspective on hydrology, water balance, and water resources management in arid basins. Hydrol Processes 24 (2), 129–135. Wu, X., Zhou, H., 2011. Influence factor to water transport index in canal system in Xinjiang Irrigation area. Yellow River 09, 93–97 (in Chinese with English abstract). Xiao, J., Liu, Z., Duan, A., Liu, Z., Li, X., 2006. Studies on effects of irrigation systems on the grain yield constituents and water use efficiency of winter wheat. Journal of Irrigation and Drainage 02, 20–23 (in Chinese with English abstract). Xu, X., Zhou, H., Wang, Z., 2010. Study on effective rainfall use efficiency in arid irrigation district. Water Saving Irrigation 12, 44–46 + 50 (in Chinese with English abstract). Xu, C., Yang, X., Li, Y., Wang, W., 2011. Changes of China agricultural climate resources under the background of climate change III. Spatiotemporal change characteristics of agricultural climate resources in Northwest Arid Area. Chinese Journal of Applied Ecology 03, 763–772 (in Chinese with English abstract). Zhao, W., Chang, X., He, Z., Zhang, Z., 2007. Study on vegetation ecological water requirement in Ejina Oasis. Science in China Series D: Earth Sciences 50 (1), 121–129. Zhu, Q., Zhang, W., 2005. Topography-based spatial daily precipitation interpolation by means of PRISM and Thiessen polygon analysis. Scientia Geographica Sinica 02, 233–238 (in Chinese with English abstract).