Author: Wang, Qi; Li, Fengrui; Zhang, Enhe; Li, Guan; Vance, Maureen
Date published: April 1, 2012
Heihe River Basin is the second largest inland river basin in the arid regions of Northwest China, and one of the main areas for grain-production in China. Maize (Zea mays L.) and wheat (Triticum aestivum L) are produced on 85 % of the total cultivated area in the region (Su et al., 2002; 2006). This region is dependent mainly on snow melt-water from mountains and underground water for irrigation. In order to increase the grain yield and to get high economic returns, farmers normally reclaim sandy lands using high fertilizers and irrigation rates. In Northwest China the current average rate of N application and irrigation reaches 250-350 kg N ha-1 yr-1 and 64-94 cm, respectively, in single-cropping systems (Su et al., 2007b). However, a high rate of nitrogen (N) fertilizer and irrigation application has led to an over-use of fresh water resources and an increase in soil nitrate nitrogen (NO3 -N) accumulation and leaching (Ju et al., 2004). Nitrate nitrogen leaching ranges from 15 to 55 % of applied N fertilizer in North China Plain (Zhu and Chen, 2002). This has resulted in groundwater contamination (Zhang et al., 1996). Nitrate-nitrogen pollution of ground and drinking water due to high N fertilizer and irrigation rates has become a serious issue in this arid zone of Heihe River Basin, China (Yang et al., 2008). The investigations showed that nearly 50 % of groundwater has NO3 -N content, exceeding 50 mg L-1, the allowable limit for drinking water. In some areas of Northwest China, the NO3 -N content in groundwater and drinking water reaches 300 mg L-1 (Zhang et al., 1995; 1996). Spring wheat is usually grown from March to July in warm temperatures and high evapotranspiration, especially in June and July, when the average precipitation is only 52.5 mm during the wheat growing season. The adverse effect of this stress can be partially mitigated by adequate water and N supply (Pandey et al., 2001). Heihe River is the only surface water resource for irrigation and the increased demand of this limited water sources presents a constraint in crop production with deteriorating quality as result of increased consumption (Ma et al., 2005; 2009). Zhang et al. (1996) found that fertilizers are applied at double or even triple the rates required for agricultural production. The trend of N application and irrigation is still on the rise, and this causes a reduced N use efficiency and increased risk of NO3 -N leaching into the groundwater (Ma et al., 2005; Chen et al., 2007). The soils are relatively poor in plant nutrients and they have a loose sandy structure on recently reclaimed sandy farmlands. Water shortage has become the major obstacle to the plant production which relies thoroughly on irrigation (Zhang et al., 1996; Kang et al., 2004; Yang et al., 2008). Many factors affect wheat yield and NO3 -N accumulation in soil. These include crop N uptake dynamics, N fertilizer management, rainfall, irrigation, soil texture, and N transformation in the soil. However, N fertilizer and irrigation are two major factors influencing wheat yield and NO3 -N accumulation but these can be controlled by the grower (Ottman and Pope, 2000; Yin et al., 2007). Irrigation effectively increases crop yield although water-use efficiency (WUE) decreases as the irrigation rate increases (Al-Kaisi et al., 2003). N fertilization also increases crop yield when the soil N supply is low (Fredrick et al., 1995a: 1995b; Sexton et al., 1996). N fertilizer applied at rates higher than the optimum requirement for crop production may cause an increase in NO3 -N accumulation below the root zone and pose a risk of NO3 -N leaching (Zhu et al., 2003; Fang et al., 2006). N application rate was found to be the main factor causing NO3 -N leaching. No NO3 -N leaching was found when the N application was below 150 kg ha-1, but NO3 -N leaching increased at a rate of 225 - 300 kg N ha-1 (Campbell et al., 1993; Fan et al., 1998). In general, increased soil water content enhances crop yield response to N fertilization, especially when high N rates are applied (Norwood, 2000). Interest in the relationship between wheat production and water-use is being increased because of the increasing scarcity and cost of irrigation water in this dry region. Profits and risks inherent in irrigation management decisions depend directly on crop-water production function (Pandey et al., 2001; Zhu et al., 2002). Preferred management practices for N fertilization recommend that N fertilizer is applied in split applications or timely according to crop needs. NO3 -N uptake efficiency has been shown to be the greatest with split N applications to wheat (Alcoz et al., 1993; Sowers et al., 1994; Ayoub et al., 1995). Many studies have been carried out to investigate the effects of N application rates on grain yield and overall N balance in the soil (Liu et al., 2001, 2003; Ju et al., 2003). Few efforts have been made to assess yield of spring wheat, WUE and NO3 -N accumulation in soil under different irrigation and N fertilizer application rates, specifically on recently reclaimed sandy farmland. This study can provide insights on how deficit irrigation and N rates can be manipulated for maximization of grain yield of spring wheat and the minimization of NO3 -N leaching below root depth in the soils. The object of this study was to find the optimum N application rate on a recently reclaimed sandy farmland in Heihe River Basin, Northwest China.
Soil water storage
The high irrigation treatments were 735 mm in 2006 and 895 mm in 2007. The amounts of water supplied to these sandy lands were higher than those applied to old farmlands mainly due to the loose structure and high evaporation rate. Soil water storage, differences between N application rates at a given irrigation rate, and between irrigation treatments at a given N application rate in the 0 - 200 cm profile are given in Fig. 1. Soil water storage fluctuated greatly in response to supplemental irrigation. This increased with supplemental irrigation and afterwards decreased with ET, and showed small fluctuations with rainfall in 2006 and 2007. Soil water storage declined quickly from the heading stage to the flowering stage in this area of high evaporation and high cropwater- requirements. The total rainfalls, during the spring wheat growing seasons, were 43.9 mm (2006) and 61.0 mm (2007). These were much less than the amounts of irrigation supplied during the periods. The ET rate is very high in the study area. The maximum rainfalls of 10.2 mm (2006) and 14.4 mm (2007) (seen as a considerable rainfall in this region) can only moisten the top of the dry soil layer (2.5-5 cm in a flat field). However, the evaporation capacity can reach 10 mm per day in warm periods. So, after only 1-2 days the rain-water retained in the soil evaporates completely. Thus, rainfall has little effect on the soil water storage in this area. After analyzing the differences in soil water storage between different irrigation rates, and between different N application rates, the differences between treatments were not significant. This indicated that irrigation treatments and N application rates have little effect on soil water in the 0-200 cm soil layer. This is mainly due to the fact that the soil water content was higher than the field capacity, and permeated from the measured layer 24 h after irrigation in the well- drained sandy farmland.
Analysis of variance
Irrigation and N treatments had significant effects on grain yield, kernel numbers, straw yield and WUE at P < 0.05. An exception was the effect of irrigation treatment on straw yield in 2007. However, their interactions were not significant, except for the interaction on WUE in 2006 and on grain yield in 2007 (Table 1). The results of Table 1 indicated that, in most cases, N had more effect on grain yield, kernel numbers and WUE than irrigation. The amount of irrigation in 2007 was greater than that in 2006. This suggested that the interaction of irrigation and N treatment on WUE in 2006 and on grain yield in 2007 were significant.
Grain yield response varied by year and is presented separately. Yields were generally greater in 2007 than that in 2006. The average grain yield were 1589 kg ha-1 in 2006 vs. 2803 kg ha-1 in 2007 (Table 2). The greater grain and straw yield in 2007 is mainly explained by a higher amount of irrigation and coincidental rainfall. This rainfall was 16.4 mm and occurred from 14 to 17 June during the filling stage (the crucial stage for forming grain yield), while no rainfall occurred from 21 May to 22 June (flowering stage to the filling stage) in 2006 (Fig. 2). The grain yield increased with the irrigation application. The high irrigation treatment (I1.0-1.0ET) had a significantly higher grain yield than the low irrigation treatment (I0.6-0.6ET). The differences in grain yield between the high and the medium irrigations (I0.8-0.8ET), and between the low and the medium irrigations were not significant (Table 2). Grain yield was decreased by withholding irrigation at various growth stages at all N rates, the reduction rates were 21.7 % (2006) and 19.8 % (2006) at low irrigation, compared with high irrigation. The grain yield response to N was quadratic for all irrigation regimes. The grain yield significantly increased with increasing N, up to 221 kg N ha-1, under 3 irrigation regimes in the 2 years (Fig. 3). The highest grain yield obtained with application of 221 kg N ha-1 treatments were 2089 kg ha-1 (2006) and 3462 kg ha-1 (2007). Treatment N221 (221 kg N ha-1) had significantly higher grain yield response than those in the N0 (0 kg N ha-1), N79 (79 kg N ha-1) and N140 (140 kg N ha-1) treatments. The difference in grain yield between N221 and N300 (300 kg N ha-1) was not significant in the 2 years. Obviously, the amount of supplied water influenced the response to N. The N221 treatment gave grain yield increases of 156.2, 56.4, 17.7 and 8.2 % compared to N0, N79, N140 and N300 at all irrigation rates in 2006, while there were increases of 75.5, 39.0, 21.4 and 6.9 % in 2007.
The differences in kernel numbers between irrigations treatments were not significant in the 2 years (Table 2). Similar to grain yield, kernel numbers significantly increased with increasing N up to 221 kg N ha-1 under the 3 irrigation regimes in 2 years. The highest value of kernel numbers obtained with N221 treatment were 4469 (2006) and 10864 (m-2) (2007).
Like grain yield, straw yield data showed that the individual effects of irrigation and N were greater than their interactive effects. Aboveground, the straw yield was lower in 2006 (1809 kg ha-1) than in 2007 (2533 kg ha-1) (Table 2). Irrigation treatments had significant effects on the straw yield in 2006, but not in 2007. Straw yield decreased with decreasing irrigation rates, the reductions of I0.6 and I0.8 were 15.7 % and 8.5 % compared with I1.0 in 2006, while the reductions were 8.6 % and 3.8 % in 2007. Straw yield significantly increased with N rate increasing. Applications of 221 kg N ha-1 in 2006 and 300 kg N ha-1 in 2007 proved sufficient to reach maximum straw yield.
Economic nitrogen rates
Predicted maximum grain yield and N rates required for maximum grain yield and profitable N rates for the 3 irrigation regimes were calculated from regression equations (Fig. 3) for each season (Table 3). The most profitable N rates were less than the maximum N rates (required for maximum yield) in the all irrigation regimes. The predicted N rates to achieve maximum grain yield were 44.3% (2006) and 55.0 % greater (2007) than the optimum economic N rates for the I0.6 treatment. For the I0.8 treatment, the N rates were 37.9 % (2006) and 37.0 % greater (2007). For I1.0 treatment, the N rates were 33.8 % (2006) and 31.3 % greater (2007), respectively. The predicted N rates to achieve the maximum yield ranged from 237 to 310 kg ha-1, while the optimum economic N rates ranged from 174 to 226 kg ha-1. This information is useful to farmers farming with limited irrigation in these arid regions.
In both years, crop water use was linearly related to the amount of irrigation (the regression equations between crop water use and the amount of irrigation were not showed in this paper). The low crop water-use, related to high irrigation treatments, was most likely due to high permeation over the root layer and high ET. In 2007, WUE was higher than that observed in 2006 due to the better climatic conditions which were favorable to the formation of grain yield (Table 2). The WUE decreased with increasing irrigation application rates. The low irrigation treatment had higher WUE (3.5 and 5.0 kg ha-1 mm-1 in 2006 and in 2007, respectively) than that of high irrigation treatment (2.7 and 3.8 kg ha-1 mm-1 in 2006 and in 2007, respectively). The medium irrigation treatment had a medium WUE (3.1 and 4.4 kg ha-1 mm-1 in 2006 and in 2007, respectively). In both years, the WUE increased significantly with N application rates up to rate of 221 kg ha-1. The N221 treatment had the highest WUE value (4.1 and 5.3 kg ha-1 mm-1 in 2006 and in 2007, respectively) among the various N treatments. There was a tendency for low irrigation treatments to have the highest WUE over the 2 wheat-growing seasons. This may indicate that deficit irrigation reduces grain yield, but increases WUE.
Nitrate nitrogen accumulation
As the NO3 -N accumulation of N79 was between N0 and N140 in the 3 irrigation regimes over the 2 years, the NO3 -N accumulation of N79 is not discussed here. Before the sowing stage, the differences of NO3 -N accumulation within 200 cm depth, ranging from 57.3 to 62.7 kg ha-1 in 2006 and from 16.3 to 22.4 kg ha-1 in 2007, between treatments were not notable. At low irrigation level, the NO3 -N accumulation of N221 at harvesting (67.4 kg ha-1in 2006 and 26.6 kg ha-1 in 2007) was higher than the average value before sowing (60.1 kg ha-1 in 2006 and 20.0 kg ha-1 in 2007) in the 200 cm depth soil. Similar to the N221, the N300 had a higher NO3 -N accumulation at harvesting (120.1 kg ha-1 in 2006 and 33.6 kg ha-1 in 2007). At medium irrigation level, only N300 had higher NO3 -N accumulation at harvesting stage (139.9 kg ha-1 in 2006 and 27.7 kg ha-1 in 2007) than the average value before sowing. Comparisons of the NO3 -N accumulations before sowing and at harvesting under deficit irrigation (Fig. 4) indicates a significant NO3 -N accumulation increase with N application rates, especially when the application rate is over 221 kg N ha-1. At high irrigation level, the NO3 -N accumulation at harvesting (the average values were 32.9 kg ha-1 in 2006 and 14.6 kg ha-1 in 2007) was lower than that accumulated before sowing within 200 cm over the 2 years. This suggests that more NO3 -N leached to deeper soil layers and more N was absorbed by plants under high irrigation treatment. This confirmed that there would be little NO3 -N leaching to a depth below the 200 cm at a N rate of below 221 kg N ha-1 under deficit irrigations. By comparing the average values over the 4 nitrogen rates, it was found that the medium and the low irrigations had more NO3 -N accumulation than the high irrigation in 200 cm soil depth. The NO3 -N accumulation (the average value was 50.0 kg ha-1) at harvesting in 2006 was higher than that accumulated (the average value was 18.3 kg ha-1) before sowing in 2007 in the 0-200 cm soil profiles. There was no plant growth and no N absorbed, and this indicated that NO3 -N was leached with the winter irrigation water applied on 15 November 2006.
Effects of irrigation on grain yield
In this 2 year study, grain yield significantly increased with irrigation rates. This agreed with the results of Pandey et al. (2001). Conversely, water lack at any growth stage reduced grain yield, kernel numbers and straw yield. In both years, during the growing season, the rainfall was low, and the high temperature and high ET demand had adverse effects on wheat growth, especially from the flowering to the filling stages of wheat. The adverse effects of this stress can be partially mitigated by an adequate supply of water (Pandey et al., 2001). Medium irrigation (I0.8) and low irrigation (I0.6) consistently resulted in lower grain yield (20 % and 8%), kernel numbers (9 % and 5 %) and straw yield (12 % and 6 %) than high irrigation (I1.0) over the 2 years period. These results are similar to Al-kaisi et al. (2003) who observed significant and positive impacts of irrigation levels on corn grain yield response to N rates, and they agreed with Hergert et al. (1993) and O'Neill et al. (2004) that grain yield under high irrigation were significantly greater than that under deficit irrigation. As a result of the present high ET and limited irrigation, crops will not grow reliably under water-stressed conditions in the studied area. This indicates that it is difficult to obtain satisfactory grain yield without irrigation, suggesting that supplemental irrigation is necessary.
Effects of nitrogen on grain yield
Paolo and Rinaldi (2008) showed that applied N had a marked influence on grain yield and kernel numbers, but had little influence on kernel weights. The results of our experiment agreed partly with the above finding. In general, the 2-year results showed that 221 kg N ha-1, which is the low N fertilizer rate currently used on farmland in the area, had the highest values of grain yield, kernel numbers and straw yield. An exception was the highest straw yield obtained with 300 kg N ha-1 supply under the 3 irrigation regimes in 2006. Compared to N0, N79, N140 and N300 over the 2 years, The N221 treatment showed grain yield increases of 99.1, 45.1, 20.0 and 7.4 %. This, in some degrees, agreed with Mandal et al. (2005), where crops maintained higher biomass with increased water supply in combination with N optimum, causing an ascribed to overall improvement in plant vigor in term of development of leaves, stems and grains. Li et al. (2001) found that treatment with N rate of 225 kg ha-1 resulted in the highest yield and yield components. Paolo and Rinaldi (2008) reported that the N was the most limiting factor for grain yield, and water stress had only a slight effect under lower N rates. Conversely, with adequate or excessive N rates, water became the more limiting factor. Optimizing inputs of both water and N simultaneously minimize the adverse effects of high temperatures and compensate for loss in crop productivity (Gajri et al., 1993). Irrigation and N treatment had significant impacts on wheat grain yield, kernel numbers and straw yield, but the interaction of ''I× N '' was insignificant in most cases.
The economic nitrogen rates and the maximum nitrogen rates
The most profitable N rate ranged from 174 to 226 kg ha-1. These were less than the maximum N rates required for maximum grain yield, which ranged from 237 to 310 kg ha-1 in the all irrigation regimes. This result agreed with the findings of Pandey et al. (2001) who also concluded that the most profitable N rates and the maximum N rates were markedly higher with high irrigation than with deficit irrigation. These results agreed, to a degree, with Russelle et al. (1981) that the maximum N rate for maximum yield was the same under different irrigation conditions. Grain yield of winter wheat and N utilization efficiency reached the highest at an application rate of 225 kg N ha-1, while the economic yield was a maximum at a rate of 150-225 kg N ha-1 (Li et al., 2001).
Effects of irrigation and nitrogen on WUE
The results of the current experiment showed that the WUE was decreased directly with irrigation rates, but increased significantly with N fertilizer applications, when the N application rate reached 221 kg ha-1 in the both years. The low irrigation had the highest WUE (4.25 kg ha-1 mm-1) among the irrigation treatments, and the 221 kg N ha-1 had the highest value (4.75 kg ha-1 mm-) among the N application treatments over the 2 years. Crop water use increased significantly with the increase in water supply in every N level. This is because of the relative decrease in WUE at higher levels of irrigation compared with WUE at deficit irrigation (Zhang and Oweis, 1999). Several studies have reported WUE values that were higher under water deficit than high irrigation condition, especially when irrigation is applied in the critical stages of plant development (Mandal et al., 2005). However, in contrary to the irrigation influence, the N rate positively influenced the WUE of wheat (Howell, 2001). Hussain and Al-Jaloud (1995) concluded that the application of 150-225 kg N ha-1 for high irrigation and 75-150 kg N ha-1 for deficit irrigation would be sufficient to obtain optimum grain yield and higher WUE of wheat. The WUE of winter wheat was reduced markedly by either a deficiency or an excess of N supply. The results of the current experiment showed that the highest WUE was obtained at a 221 kg ha-1 N application rate combined with low irrigation each season.
Effects of irrigation and nitrogen on nitrate nitrogen accumulation
Zhu and Chen (2002) concluded that the accumulation of NO3 -N, in many farmlands within 200 cm depth of soil, was a result of a long period of excessive use of N fertilizer. In our experiment, the N application at rate of 221 kg N ha-1 or above showed more NO3 -N accumulation at harvesting than before sowing, under irrigation deficit conditions,. These values are comparable with the values of 150-180 kg N ha-1 recommended by Zhu and Wen (1992) in Northwest China.
Considering the high WUE, deficit irrigation may still be acceptable for wheat production although it resulted in low grain yield in Heihe River Basin. Due to limited water resources, many irrigation systems have to adopt deficit irrigation practices to cover greater areas and benefit more farmers. In order to minimize NO3 -N accumulation in soil at harvesting stage, the N application rate should be lower than that of 221 kg N ha-1 for spring wheat production in a recently reclaimed sandy farmland. The optimum N rate is lower than the current average N rate (250-350 kg N ha-1) in these areas.
Materials and methods
Field experiments were conducted at Linze Inland River Basin Research Station, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences (39°21 N, 100°07 E, 1367 m above Mean Sea Level), during two consecutive spring wheat growing seasons in 2006 and 2007. The station was located at the southern edge of Badain Jaran Desert. The experimental fields were reclaimed in 2000, from the Gobi Desert (by which they were surrounded), and had been planted continuously with irrigated maize for the past 5 years. The natural vegetation before reclaimation included Nitraria sphaerocarpa (Maxim.), Reaumria soongorica (Pall. Maxim.), Suaeda glauca (Bge.) and Sillium mongolicum ( Rgl.). The region has a typical temperate desert climate: dry and hot in summer, cold in winter, ample sunshine, very little precipitation, strong winds, and frequent drifting sands. The annual mean air temperature is about 7.6 °C, with an absolute maximum and absolute minimum of 39.1 °C in July, and -27 °C in January. The normal annual precipitation is 117 mm (1965-2005) and about 60 % of the total precipitation occurs from July to September (Su et al., 2007b) and groundwater level ranges from 4 to 10 m.
The soils all formed from diluvial-alluvial materials and are classified as Calci-Orthic Aridosols, and Calciothids according to Chinese Soil Taxonomy. The dominant texture is loamy sand with a very low nutrient concentration and a loose structure (Su et al., 2006; 2007a). Just before sowing in 2006, the soil was sampled at depths of 0 - 100 cm. The physical and chemical properties of the soil are presented in Tables 4 and 5. The total rainfalls during the spring wheat growing seasons were 43.9 mm in 2006 and 61.0 mm in 2007.
The experiment was conducted with 3 replications in a split-plot on a randomized complete block design, with irrigation rates as the main plots and N rates as the sub-plots. The 3 irrigation rates were 0.6, 0.8, and 1.0 of the estimated evapotranspiration (ETc). This was based on a model simulating the actual process of soil evaporation, plant transpiration and the total ET from irrigated farmlands in the oasis of Heihe River Basin during spring wheat-growing seasons. ETc, ET0 and Kc were calculated using following formulae (Allen et al., 1998; Ji et al., 2004; Zhao et al., 2009).
where, is the crop evapotranspiration (mm d-1), Kc the crop coefficient and ET0 reference evapotranspiration (mm d-1). Kc(Tab) is the value of Kc taken from Table 17 of FAO-56 (Allen et al., 1998), RHmin the mean value for daily minimum relative humidity, the mean plant height, Rn the net radiation (MJ m-2 day-1), G the soil heat flux (MJ m-2 day-1), T the air temperature at 2 m height (C), Ê2 the wind speed at 2 m height (m s-1); the saturation vapor pressure deficit (kPa), the slope vapor pressure curve (kPaC-1) and the psychrometric constant (kPaC-1).
Irrigation and nitrogen treatments
The irrigation treatments were 482, 608, 735 mm in 2006 and 576, 736, 895 mm in 2007, respectively. These treatments represented 0.6 ETc (I0.6), 0.8 ETc (I0.8), and 1.0 ETc (I1.0). The N rates of 0, 79, 140, 221 and 300 kg ha.1 in the form of urea were applied within each irrigation treatment and were denoted by N0, N79, N140, N221, and N300, respectively. The rates and dates of N fertilizer supply during the wheat-growing seasons over the 2 years were represented in Table 6. The I0.6 irrigation treatment and N0 treatment were used as controls for the respective irrigation and N treatments. Between all main plots and all subplots, a 2 m alley and a 1.5 m alley were kept to eliminate the influence of lateral water and N movement. Calibrated siphons were used to deliver the required amount of water from the irrigation channel into the hose. In order to measure the delivered amount of water accurately, a water flow meter was placed at the head of the hose for each plot of 9 m~4.5 m. To contain irrigation water, the plots were surrounded by 25 cm-high banks. The amount of irrigation water was greater in 2007 than in 2006. This was mainly being due to a high net radiation in 2007. The rates and dates of irrigation during the wheat growing-season over the 2 years of this experiment are listed in Table 7. Irrigation was stopped at the dough-growth stage of spring wheat.
Prior to 2005, the experimental fields had been under irrigated maize cultivation for 5 years. After the harvest of maize in 2005, the land was plowed once and harrowed. The plot was laid out in September 2005, and uniform winter irrigations of 100 mm were applied on 20 November (2005) and again on 15 November of 2006 to pulverize the soil. A local wheat cultivar (Longfu-920) was used to seed 15 cm rows at 337.5 kg ha-1 in the dry seedbed. This was followed by a uniform 107 mm post-sowing irrigation to all plots, regardless of treatments, to ensure proper germination and establishment of the seeds. During cultivation 41 kg P ha-1 and 39 kg K ha.1 were applied along with the wheat seeds. Seeding was done on 22 March, 2006 and 21 March, 2007. Urea (46 % N) was used as a N fertilizer and applied manually before irrigation. The crop varieties, fertilizer dozes, irrigation interval and planting densities were based on common practices used by local farmers. Manual weeding was used to control the weeds. The crop was harvested manually on 9 July in both years.
Sampling and sample analysis
The soil samples were taken at increments of 20 cm within 200 cm depth using an auger with a diameter of 3.6 cm during the wheat-growing seasons, before each sowing, at harvesting and 24 h after irrigation, This was performed 6 times in 2006 and 8 times in 2007. The soil samples were weighed then dried to a constant weight at 105C to determine the gravimetric soil moisture content. Before sowing and at harvesting, three soil cores were sampled from each plot, thoroughly mixed, quickly stored in plastic bags and analyzed immediately to determine the NO3 -N concentration. This was determined by using a Flow Solution IV Analyzer (FSIV, O.I. Analytical, U.S.A.) after extraction (with a 1:5 ratio) (w/w) soil:1 mol L -1 KCl solution (Bao 2000). The amount of NO3 -N (kg N ha-1) reserved in 0-200 cm soil profiles was calculated according to the equation modified by Emteryd (1989):
where, Ti is the thickness of soil layer in cm; BDi the bulk density in g cm-3; [NO3 -]i the soil NO3 -N concentration in mg kg-1, and 0.1 the conversion coefficient. Crop water use was calculated by the water balance equation: ETactual= P + I - Dp - R -ΔS, where ETactual is the actual evapotranspiration, P the precipitation; I the depth of irrigation water applied; Dp the drainage beyond the measured depth; R the runoff and S the variation of soil water storage between sowing and harvesting dates within 200 cm, and all measurements were expressed in mm. As Dp and R were negligible, ETactual was calculated as ETactual = P + I - DS . WUE was calculated as grain yield divided by seasonal ETactual (Pandey et al., 2001, Mandal et al., 2005).
At harvest, plants were sampled, from 60 cm in the middle row of each replication, for plant dry weight and yield components. All the plant samples were cut near the soil surface and the below-ground fractions were left in the field. Dry matter of grain and straw was determined by drying the sub-samples in a convection oven at 65C to a constant weight. Kernels numbers per spike and kernels numbers per m2 were calculated from the head number and 1000-kernel weight. Grain yield was hand-harvested from the 16 m2 plot and adjusted to 14 % moisture weight.
The treatment at which the maximum yield occurred was determined by differentiating the resulting equations with respect to the fertilizer N and the N rate at which the first derivative equaled zero. The most profitable N rate was calculated as the N rate at which the ratio of fertilizer N price to grain price equaled the first derivative of grain yield with respect to fertilizer N (Sticker et al., 1995).
Grain yield, kernel numbers, straw yield and WUE were analyzed using an analysis of variance (ANOVA) appropriate for a randomized complete block split-plot design, with irrigation treatment as the main-plot factor and N rate as the split-plot factor. To determine the significant differences of the means between the irrigation treatments and between the N treatments, a One-Way ANOVA was used. Pairs of mean values were compared by the least significant difference (LSD) at the 5% and 1% level using SPSS software (for Windows, Version 11.0), and Duncans multiple range test was used for comparisons. Regression analysis was performed on the relationships between the crop parameters and the irrigation and N rates. Best fit regression models were calculated.
It can be concluded that in areas with similar conditions to the experimental site, optimization of N and irrigation application can improve wheat grain yield and minimize potential NO3 -N leaching. In Heihe River Basin, deficit irrigation results in reduced grain yield, kernel numbers and straw yield, but increased WUE. In general, the 2-year results showed that, under 3 irrigation regimes, the N application rate of 221 kg ha-1 had the highest values of grain yield, kernel numbers, straw yield and WUE, and this application rate can be used as an alternative to that of 300 kg N ha-1. Differential strategies of N application rates were concluded when considering different sectors of farming. The N application rate of 221 kg N ha-1 could lead to high risk of excess soil NO3 -N accumulation within 200 cm soil profile at harvesting under deficit irrigation over 2 years. This indicates that the optimum N rate is below 221 kg ha-1, which is much lower than that used in many places in the middle reaches of these areas. There is little potential for soil NO3 -N accumulation and leaching. Full irrigation leads to NO3 -N leaching and plant N-uptake, resulting in lower NO3 -N accumulation at harvest than prior to sowing. The most profitable N rates range from 174 to 226 kg ha-1 and these would enhance economic returns to farmers.
This research was supported by The National Natural Science Foundation of China (41161090 and 31060178). We would like to thank our anonymous reviewers for their many helpful comments on earlier versions of this manuscript.
Alcoz MM, Hons FM, Haby VA (1993) Nitrogen fertilization timing effect on wheat production, nitrogen uptake efficiency, and residual soil nitrogen. Agron J 85:1198-1203
Al-Kaisi MM, Yin XH (2003) Effects of nitrogen rate, irrigation rate and plant population on corn yield and water use efficiency. Agron J 95: 1475-1482
Allen RG, Jensen ME, Wright JL, Burman RD (1998). Operational estimates of reference evapotranspiration. Agron J 81: 650-662
Ayoub M, MacKenzie A, Smith DL (1995) Evaluation of N fertilizer rate and timing and wheat cultivars on soil residual nitrates. J Agron Crop Sci 175:87-97
Bao SD (2000) Soil and agricultural chemistry analysis. Edition III, Chinese Agriculture Press, Beijing, p 42-49, 264 - 268 (in Chinese)
Campbell CA, Zentner RP, Selles F, Akinremi OO (1993) Nitrate leaching as influenced by fertilization in the brown soil zone. Can J Soil Sci 73:387-397
Chen JY, Taniguchi M, Liu GQ, Miyaoka K, Onodera SI, Tokunaga T, Fukushima Y (2007) Nitrate pollution of groundwater in the Yellow River delta, China. Hydrogeol J 15: 1605 - 1614
Emteryd O (1989) Chemical and physical analysis of inorganic nutrients in plant, soil, water and air. Stencil No 10 Umea, Sweden, p 156-159
Fan BQ, Hu CF, Ping JL (1998) Effect of irrigation and fertilization on nitrate leaching in loamy fluvo-qauic soil. Plant Nutrition and Fertilizer Science 4(1): 16 - 21 (In Chinese)
Fang QX, Yu Q, Wang EL, Chen YH, Zhang GL, Wang J, Li LH (2006) Soil nitrate accumulation, leaching and crop nitrogen use as influenced by fertilization and irrigation in an intensive wheat- maize double cropping system in the North China Plain. Plant Soil 284: 335-350
Fredrick JR, Camberato JJ (1995a) Water and nitrogen effects on winter wheat in the southeastern central plain: I. Grain yield and kernel traits. Agron J 87: 521-526
Fredrick JR, Camberato JJ (1995b) Water and nitrogen effects on winter wheat in the southeastern central plain: physiological responses. Agron J 87:527-533
Gajri PR, Prihar SS, Arora VK (1993) Interdependence of nitrogen and irrigation effects on growth and input-use efficiencies of wheat. Field Crop Res 31:71-86
Hergert GW, Klocke NL, Petersen JL, Norquist PT, Clarke RT, Wicks GA (1993) Cropping systems for stretching limited irrigation supplies. J Prod Agric 6:520-529.
Howell TA (2001) Enhancing water use efficiency in irrigated agriculture. Agron J 93: 281-289
Hussain G., Al-Jaloud AA (1995) Effect of irrigation and nitrogen on water use efficiency of wheat in Saudi Arabia. Agr Water Manage 27: 143-153
Ji XB, Kang ES, Zhao WZ, Chen RS, Jin BW, Zhang ZH (2004) Simulation of the Evapotranspiration from Irrigational Farmlands in the Oases of Heihe River Basin. Journal of Glaciology and Geocryology. 26(6):713-719 (In Chinese)
Ju XT, Liu XJ, Zhang FS (2003) Accumulation and movement of NO3 -N in soil profile in winter wheat - summer maize rotation system. Acta Pedologica Sinica 40(4):538-546 (In Chinese)
Ju XT, Liu XJ, Zhang FS, Roelcke M (2004) Nitrogen fertilization, soil nitrate accumulation and policy recommendations in several agricultural regions of China. Am Biol Teach 33:278-283
Kang S, Su X, Tong L, Shi P, Yang X, Yukuo, ABE, Du T, Shen Q, Zhang J(2004)The impacts of human activities on the water-land environment of the Shiyang River basin, an arid region in northwest China. Hydrolog Sci J 49 (3):413-427
Li SJ, Zhou DX, Li JM (2001) Effect of different nitrogen application on yield, nitrogen distribution and utilization in winter wheat under soil water stress. Acta Agriculturae Boreali-Sinic 16 (3): 86 -91 (In Chinese)
Liu XJ, Ju XT, Zhang FS (2001) Effect of urea application as basal fertilizer on inorganic nitrogen in soil profile. J China Agric Univ 6(5): 63-68 (In Chinese)
Liu XJ, Ju XT, Zhang FS, Pan JR, Christie P (2003) Nitrogen dynamics and budgets in a winter wheat-maize cropping system in the North China Plain. Field Crop Res 83:111-124 (In Chinese)
Ma JZ, Ding ZY, Wei GX, Zhao H, Huang, TM (2009) Sources of water pollution and evolution of water quality in the Wuwei Basin of Shiyang river, Northwest China. J Environ Manage 90: 1168 - 1177
Ma JZ, Wang XS, Edmunds WM (2005) The characteristics of groundwater resources and their changes under the impacts of human activity in arid North-West China - a case study of the Shiyang river Basin. J Arid Environ 61:277-295
Mandal KG, Hati KM, Misra AK, Bandyopadhyay KK, Mohanty M (2005) Irrigation and Nutrient Effects on Growth and Water-Yield Relationship of Wheat (Triticum aestivum L.) in Central India. J Agron Crop Sci 191: 416 - 425
Norwood CA (2000) Water use and yield of limited-irrigated and dry land corn. Soil Sci Soc Am J 64: 365-370
O'Neill PM, Shanahan JF, Schepers JS, Caldwell B (2004)Agronomic responses of corn hybrids from different areas to deficit and adequate level of water and nitrogen. Agron J 96:1660-1667
Ottman MJ, Pope NV (2000) Nitrogen fertilizer movement in the soil as influenced by nitrogen rate and timing in irrigated wheat. Soil Sci Soc Am J 64:1883-1892
Pandey RK, Maranville JW, Admou A (2001) Tropical wheat response to irrigation and nitrogen in a Sahelian environment. I. Grain yield, yield components and water use efficiency. Eur J Agron 15: 93-105
Paolo ED, Rinaldi M (2008) Yield response of corn to irrigation and nitrogen fertilization in a Mediterranean environment. Field Crop Res 105: 202-210
Russelle MP, Dierbert EJ, Hauck RD, Stevanovic M, Olsen RA (1981) Effects of water and nitrogen management on yield and 15N-deplete fertilizer use efficiency of irrigated corn. Soil Sci Soc Am J 45: 553-558
Sexton BT, Moncrief JF, Rosen CJ, Gupta SC, Cheng HH (1996) Optimizing nitrogen and irrigation inputs for corn based on nitrate leaching and yield on a coarse-textured soil. J Environ Qual 25:982-992
Sowers KE, Pan WL, Miller BC, Smith JL (1994) Nitrogen use efficiency of split nitrogen applications in soft white winter wheat. Agron J 86:942-948
Sticker JA, Buchholz DD, Hanson RG, Wollenhaupt NC, Mcvay KA (1995) Tillage and rotation effects on corn yield response to fertilizer nitrogen on Aqualf soils. Agron J 87: 409-415
Su PX, Du MW, Zhao AF, Zhang XJ (2002) Study on water requirement law of some crops and different planting mode in oasis. Agricultural Research in the Arid Areas 20 (2): 79 - 85 (In Chinese)
Su YZ, Wang F, Suo DR, Zhang ZH, Du MW (2006) Long-term effect of fertilizer and manure application on soil-carbon sequestration and soil fertility under the wheat-wheat-maize cropping system in Northwest China. Nutr Cycl Agroecosys 75: 285-295
Su YZ, Wang F, Zhang ZH, Du MW (2007a) Soil properties and characteristics of soil aggregate in marginal farmlands of oasis in the middle of Hexi Corridor Region, Northwest China. Agricultural Sciences in China 6(6):706 - 714
Su YZ, Zhang ZH, Yang R (2007b) Amount of irrigation and nitrogen application for maize grown on sandy farmland in the marginal oasis in the middle of Heihe River Basin. Acta Agronomica Sinica 33(12):2007 - 2015
Yang R, Su YZ, 2008. Groundwater nitrate pollution in the oasis agricultural area in the middle reaches of Heihe River, Northwest China. Journal of glaciology and geocryology 30(6): 983- 990 (In Chinese)
Yin F, Fu B, Mao R (2007) Effects of nitrogen fertilizer application rates on nitrate nitrogen distribution in saline soil in the Hai River Basin, China. J Soil Sediment 7 (3):136-142
Zhang H, Oweis T (1999) Water-yield relations and optimum irrigation scheduling of wheat in the Mediterranean region. Agr Water Manage 38: 195-211
Zhang WL, Tian ZX, Zhang N, Li XQ (1995) Investigation of nitrate pollution in ground water due to nitrogen fertilization in agriculture in Northern China. Plant nutrition and fertilizer sciences 1(2): 80 - 87 (In Chinese)
Zhang WL, Tian ZX, Zhang N, Li XQ (1996) Nitrate pollution of groundwater in Northern China. Agr Ecosyst Environ 59:223-231
Zhao WZ, Liu B, Zhang ZH (2009) Water requirements of maize in the middle Heihe River Basin, China. Agr Water Manage 97:215-223
Zhu JG, Liu G, Han Y, Zhang YL, Xing GX (2003) Nitrate distribution and denitrification in the saturated zone of paddy field under rice / wheat rotation. Chemosphere 50: 725-732
Zhu ZL, Chen DL (2002) Nitrogen fertilizer use in China - Contributions to food production, impacts on the environment and best management strategies. Nutr Cycl Agroecosys 63:117-127
Zhu ZL, Wen QX (1992) Nitrogen in soil of China. Jiangsu Science and Technology Publishing House, Jiangsu, p 213-228 (In Chinese)
Qi Wang1, 2, Fengrui Li3, Enhe Zhang4*,Guan Li5, Maureen Vance6
1College of Grassland Science, Gansu Agricultural University, Lanzhou, Gansu Province 730070, China
2Cryosphere Research Station on the Qinghai-Tibet Plateau and State Key Laboratory of Cryospheric Sciences, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou, Gansu Province 73000, China
3Linze Inland River Basin Research Station, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou, Gansu Province 73000, China
4Agronomy College, Gansu Agricultural University, Lanzhou, Gansu Province 730070, China
5College of Information Science and Technology, Gansu Agricultural University, Lanzhou, 730070, China
6The Adult Reading Assistance Scheme, Christchurch, New Zealand
* Corresponding author: email@example.com