Comparison of drip fertigation and negative pressure fertigation on soil water dynamics and water use efficiency of greenhouse tomato grown in the North China Plain

Agricultural Water Management 184 (2017) 1–8

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Comparison of drip fertigation and negative pressure fertigation on soil water dynamics and water use efficiency of greenhouse tomato grown in the North China Plain Yinkun Li, Lichun Wang, Xuzhang Xue ∗ , Wenzhong Guo ∗ , Fan Xu, Youli Li, Weituo Sun, Fei Chen Beijing Research Center of Intelligent Equipment for Agriculture, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China

a r t i c l e

i n f o

Article history: Received 15 September 2016 Received in revised form 25 December 2016 Accepted 29 December 2016 Keywords: Negative pressure fertigation Drip fertigation Soil water storage Greenhouse tomato Water use efficiency

a b s t r a c t Maintaining a stable soil water supply is the key for solar greenhouse vegetable production across the North China Plain. A three-season field experiment was conducted over 2 years to evaluate two methods of applying Yamazaki tomato nutrient solution (negative pressure and drip fertigation; NF and DF, respectively) for production of greenhouse tomato and water use efficiency (WUE). Soil moisture in the surface (0–20 cm) and entire soil profile (0–100 cm), as well as soil water storage (SWS) and crop evapotranspiration (ET) levels were measured during the growing season. Then, plant growth, fruit yield, and WUE were compared. The variations in soil moisture (0–20 cm) were small for the NF treatment, with ranges of 20.0–25.0% and 22.2–24.3% in the early spring and autumn winter seasons, respectively, which were less than the ranges of 19.7–28.5% and 21.4–26.7% for DF. The average SWS did not significantly differ between DF and NF treatments, while SWS in NF (318.6–339.3 mm) during the growing season showed small fluctuations compared with DF (315.7–342.9 mm). The ET over the whole growing season varied in the range of 224.0–319.9 mm, which was higher during fruit-set and flowering than other growth periods. With its higher irrigation amount, DF had a higher ET level than NF, but there was no significant difference in early spring. The consecutive and stable water supply of NF improved tomato plant height and stem diameter (P < 0.05) and improved fruit yield and WUE by 1.6–8.2% and 9.9–30.5% (P < 0.05), respectively, compared with DF. These results demonstrate that the NF system can save more water (11.3% and 32.0% in the ES and AW seasons, respectively) than DF. As a new mode of integrated water and fertilizer management, NF is appropriate for vegetable production in solar greenhouses. © 2016 Published by Elsevier B.V.

1. Introduction Tomato (Solanum lycopersicum) is one of the main commercial vegetables grown on the North China Plain, and is mainly cultivated in greenhouses during early spring (ES) and autumn–winter (AW) (Yuan et al., 2001; Ren et al., 2010; He et al., 2016). Water and fertilizer supply can significantly affect tomato growth in solar greenhouses and watering must be carried out with high efficiency, because vegetables have a high requirement for water whenever irrigation is the only source (Wang et al., 2009; Liu et al., 2009; Fan

∗ Corresponding authors at: Beijing Research Center of Intelligent Equipment for Agriculture, Beijing Academy of Agriculture and Forestry Sciences, Beijing Nongke Mansion, No. 11 Shuguang Huayuan Middle Road, Haidian District, Beijing 100097, China. E-mail addresses: [email protected] (X. Xue), [email protected] (W. Guo). http://dx.doi.org/10.1016/j.agwat.2016.12.018 0378-3774/© 2016 Published by Elsevier B.V.

et al., 2014; Wang et al., 2015a). However, high rates of water and nitrogen (N) inputs are commonly used in tomato production in China (Chen et al., 2004; Liu et al., 2013). The total irrigation and N rate are about 1000 mm and 1049 kg N ha−1 per growing season, respectively, and have increased the production costs and resulted in large losses of water and N (Zhu et al., 2005; Ren et al., 2010). Many studies have shown that the increasing use of irrigation and chemical fertilizers has caused many environmental problems (Ren et al., 2010; Fan et al., 2014; Du et al., 2014; Cao et al., 2015), such as contaminating groundwater and nitrous oxide emissions. Previous research indicated a direct relationship between large nitrate losses and inefficient fertigation and irrigation management (Zotarelli et al., 2009; Shan et al., 2015). Therefore, water and fertilizer inputs should be carefully managed, especially in greenhouse vegetable production (Zhang et al., 2012; Chen et al., 2013; Wang et al., 2015a). Drip fertigation is one of the best techniques for applying water and fertilizer to vegetables (Simonne et al., 2006; Karlberg

2

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

Fig. 1. Structure diagram of negative pressure fertigation system.

et al., 2007; Qiu et al., 2011). Many studies have reported that higher yield and higher water and fertilizer use efficiencies can be achieved with this technique compared with other irrigation and fertilization methods (Marino et al., 2014; Zhang et al., 2017). Because of the high water and fertilizer efficiency, application of drip fertigation has increased quickly inside solar greenhouses in China. In recent years, the technique of negative pressure irrigation has been widely studied and used in solar greenhouses (Li et al., 2010a; Li et al., 2016). This new irrigation technology supplies water at a negative pressure, and soil water content can be controlled accurately and continuously by adjusting the suction of the negative pressure irrigation device (Zou et al., 2007; Li et al., 2008). This technique allows automatic irrigation according to crop requirements and changing environmental factors (Li et al., 2010b). Li et al. (2010a) showed a negative correlation between water supply tension and soil water content, with gravimetric soil water content maintained at 14.2–42.3% for water supply suction of 1–13 kPa. With the use of negative pressure irrigation, the soil water content can be maintained in an unsaturated state, which reduces water losses from soil evaporation and underground leakage (Li et al., 2008). Compared with conventional irrigation (flooding) methods, greenhouse tomato yield and water use efficiency (WUE) with negative pressure irrigation method increased by 8.6 and 11.0% (P < 0.05), respectively (Li et al., 2008). Studies of negative pressure irrigation have mainly focused on potted or soilless cultivation and on crop water management strategies (Li et al., 2010b; Xu et al., 2014). Integration of water and fertilizer can be achieved with negative pressure irrigation, and this technology has been patented (Li et al., 2015). However, research on integrating water and fertilizer for tomatoes in solar greenhouses with negative pressure irrigation systems is rare. Drip irrigation as the best water-saving technology has been widely used in greenhouses (Cetin and Uygan, 2008; Liu et al., 2013), but few studies have compared drip and negative pressure irrigation. The objectives of this research were to (i) quantify the soil moisture, soil water storage (SWS) and crop evapotranspiration (ET) during the tomato growing season; and (ii) determine the effects of integrative water and fertilizer management practices on tomato growth, fruit yield and WUE. 2. Materials and methods 2.1. Site description The present study was conducted during 2014–2015 at the National Experiment Station for Precision Agriculture (40◦ 10 43 N,

116◦ 26 39 E), Xiaotanshan Beijing, China. The site is warm temperate, with an annual mean air temperature of 11.8 ◦ C, annual sunshine duration of 2684 h and precipitation of 550.3 mm. The experimental greenhouse was 28 m long and 7.5 m wide, with a wall made of brick-concrete, and brackets constructed with welded metal wires. The soil used for this study was collected from the 0–20 cm surface layer of cultivated soil and was classified as silt loam. The main properties of the soil were pH 6.75 (1:5, soil/water), bulk density 1.39 g cm−3 , organic matter 23.3 g kg−1 , total N 1.57 g kg−1 , available phosphorus 96.4 mg kg−1 , available potassium 158.9 mg kg−1 and mass water content at field capacity of 26.3%.

2.2. Experimental design Two different fertigation treatments were compared: negative pressure fertigation (NF) and drip fertigation (DF). The NF system was established based on a negative pressure device, which could apply integrative water and fertilizer management for tomatoes in the greenhouse (Li et al., 2015). The structure diagram of NF system is shown in Fig. 1, and was used in each test plot. The system consisted of five parts connected by water pipes: liquid storage barrel, liquid constant-level barrel, pressure pipe, air bottle and water feeder. This system was sealed and connected with 14-disk water feeders at the terminal. Water feeders were ceramic and had 20 cm diameters. The feeders were pervious to water but not air, and were buried vertically, 25 cm deep in the soil, at equal intervals of 35 cm. The tomato seedlings were planted on both sides of the water feeder at a distance of 10 cm from the feeder. Before the NF system was run, the water feeder was filled with water to eliminate air from the system. The system functions as follows. When soil dries, it absorbs water from the water feeder driven by soil water potential. Then, the water level in the air bottle begins to decline and its pressure reduces, which leads to a direct increase in vacuum of the pressure pipe. The water in the liquid constantlevel barrel then enters the air bottle through the pressure pipe under the action of atmospheric pressure difference. In the liquid constant-level barrel, a float valve keeps water at the same level. When the water level declines, water in the liquid storage barrel automatically enters the liquid constant-level barrel and when the water level reaches its previous level, the filling ceases under control of the float valve. Finally, the irrigation amount was obtained by recording the water-level difference in the liquid storage barrel every day at 8:00 am.

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

3

Table 1 Irrigation and nutrient (N, P2 O5 and K2 O) rates used in the experiment in 2014. Season

Treatment

Growth stage

Irrigation (mm)

N (kg ha−1 )

P2 O5 (kg ha−1 )

K2 O (kg ha−1 )

ES 2014

NF

SD FW FS PK Total SD FW FS PK Total SD FW FS PK Total SD FW FS PK Total

48.8 65.1 117.8 67.0 298.7 56.8 81.6 144.7 53.5 336.6 34.3 57.6 60.6 62.4 214.9 36.4 102.7 79.1 65.3 283.5

61.3 81.7 148.0 84.3 375.3 71.3 102.5 181.8 67.2 422.8 43.0 72.3 76.1 78.3 269.7 45.6 128.9 99.5 82.0 356.0

23.2 30.9 56.0 31.8 141.9 27.0 38.8 68.7 25.4 159.9 16.3 27.3 28.8 29.6 102.0 17.3 48.8 37.6 31.0 134.7

91.3 121.8 220.3 125.3 558.7 106.2 152.6 270.7 100.1 629.6 64.0 107.6 113.3 116.6 401.5 68.0 192.0 148.1 122.1 530.2

DF

AW 2014

NF

DF

NF: negative pressure fertigation. DF: drip fertigation. SD: Seedling stage. FW: Flowering stage. FS: Fruit-set period. PK: Picking period. ES: early-spring season, from mid March to mid July. AW: autumn–winter season, from end July to mid December.

In this system, the height of liquid in the pressure pipe applies a particular pressure of −␳gh to the air bottle, where ␳ is the water density (kg m−3 ), h is the distance from the liquid level of the constant-level barrel to the highest point of the pressure pipe (m) (Fig. 1) and g is the gravitational constant (10 N kg−1 ). The value of h also represents the suction of the water supply, e.g. when h = 1 m, then −␳gh = 10 kPa. Thus, the suction of the system can be controlled through adjusting the height of the pressure pipe, which will change the soil water content directly. According to previous studies (Li et al., 2008; Wang et al., 2015b), we set the suction at 5 kPa in this experiment. Yamazaki tomato nutrient solution formula and 1S Yamazaki nutrient solution concentration as the standard concentration were used in this study (Gao et al., 2006; Li et al., 2011). The standard nutrient solution contained: 354 mg L−1 Ca(NO3 )2 4H2 O (N, 17.1%; Ca, 24.4%), 404 mg L−1 KNO3 (K2 O, 46.3%; N, 13.8%), 77 mg L−1 NH4 H2 PO4 (P2 O5 , 61.7%; N, 12.2%) and 246 mg L−1 MgSO4 7H2 O (Mg, 9.8%; S, 13.0%). The amount of nutrients was calculated according to the nutrient concentrations and irrigation rates, and the

irrigation and fertilizer rates (NPK) used in the tomato growth stage are shown in Table 1. The NF treatment received irrigation every day, and the total amounts in ES and AW seasons were 298.7 and 214.9 mm, respectively. The drip irrigation rates of the DF treatment were relatively high and irrigation was applied 16 and 13 times, with total amounts of 336.6 and 283.5 mm in ES and AW seasons, respectively. Compared with DF treatment, total irrigation under NF treatment was 11.3% and 32.0% lower, respectively, in the ES and AW seasons. The treatments were arranged in a completely randomized block design, with three replicates per treatment. The size of each experimental plot was 7.0 m2 (5.0 m × 1.4 m), and plots were isolated by 0.6 m deep PVC boards. Planting beds were prepared 0.8 m apart were 0.6 m wide at the top. On the bed top, two rows of tomato plants were transplanted. Plants were spaced 0.35 m apart in rows and the rows were 0.5 m apart. Prior to transplanting, the drip irrigation system with 0.4 mm thickness and 15 mm inner diameter was buried in each row. The drip lines had emitters spaced 15 cm apart with a discharge rate of 1.38 L h−1 . Then white polyethylene mulch (1.2 m in width) was applied over the beds. The tomato variety used in the experiment was ‘Xianke 8’, and tomatoes were transplanted on 21 March and 31 July, and harvested on 18 July and 16 December in ES and AW of 2014, respectively. The field experiment was carried out continuously in 2015, and tomato planted only in AW seasons. The tomato yield and WUE with the same treatments in AW of 2015 were used to verify the experiment.

2.3. Measurements Soil moisture was measured at a distance of 15 cm from the drippers in the row (Ismail et al., 2008; Liu et al., 2013). Soil water content was determined gravimetrically in the 0–0.3 m soil profile every 5–7 d over the growing stages to calculate the drip irrigation amount. Tomato plants were irrigated to 90% of field capacity (␪f ) when mean soil volumetric water content in the main root zone (0.3 m) was depleted to 70% of ␪f . Irrigation water amounts (Ir, mm) were determined using Eq. (1) (Wang et al., 2015a; Qiu et al., 2015): Fig. 2. The variations of air temperature and relative humidity inside the greenhouse during the experiment in 2014.

Ir = 1000 × ␳ × (0.9␪f − ␪i ) × 0.3

(1)

4

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

NF DF

Plant height(cm)

150 120 90 60 30 0

NF DF

Stem diameter (mm)

12 9 6 3 0 0

20

40

60

80

0

20

40

Day (ES season)

60

80

100

Day (AW season)

Fig. 3. Dynamics of tomato plant height and stem diameter in ES season and AW season in 2014. Error bars are twice the standard error of the mean (n = 3).

Soil water content(%) 0

15

20

Soil water content(%) 0

25

15

20

25

30

0 ES season Soil Depth (cm)

Ending

Beginning

20

NF DF

40 60 80 100 0 AW season

Soil Depth (cm)

20

NF DF

40 60 80 100

Fig. 4. Seasonal soil water content change of negative pressure irrigation (NF) and drip irrigation (DF) treatment in 2014. Error bars are twice the standard error of the mean (n = 3).

where ␪f is the soil field capacity (cm3 cm−3 ), ␳ is the soil bulk density (g cm−3 ) and ␪i is the mean soil volumetric water content (cm3 cm−3 ). Soil water content was also gravimetrically measured by ovendrying (105 ◦ C for 24 h) the core samples that were taken at depth intervals of 20 cm down the 0–100 cm soil profile in each plot at transplanting, seeding (SD), flowering (FW), fruit-set (FS) and pick-

ing stages (PK). The soil water storage (SWS) was calculated using Eq. (2): SWS = h × ␳ × ␪ × 1000

(2)

where h is the soil depth (cm) and ␪ is the soil gravimetric water content (g g−1 ). Evapotranspiration (ET, mm) was calculated using the soil water balance equation (Zhang et al., 2011). Because run-off and deep per-

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

ES season

AW season

NF DF

360

340

340

320

320

300

300 SD

0 3-19

FW

FS

5-19 4-19 Date (month,day)

SD

PK

6-19

7-19

7-31

FW

8-31

FS

10-19 9-19 Date (month,day)

PK

11-19

Soil water storage (mm)

Soil water storage (mm)

360

5

0 12-19

Fig. 5. Soil water storage (0–100 cm) under negative pressure irrigation (NF) and drip irrigation (DF) treatment during the experiment in 2014. Error bars are twice the standard error of the mean (n = 3). SD: Seedling stage. FW: Flowering stage. FS: Fruit-set period. PK: Picking period.

Table 2 Changes of 0–20 cm soil moisture (%) during tomato growth stage in ES and AW seasons in 2014. Growing season

Treat.

Growth stage

Av

Sd

Cv

SD

FW

FS

PK

ES

NF DF

25.0 ± 0.72a 25.7 ± 1.00a

21.3 ± 1.39bc 19.7 ± 1.81b

20.8 ± 1.77c 27.7 ± 1.94a

23.8 ± 1.40ab 27.6 ± 0.63a

22.7 25.2

1.98 3.78

8.72 15.0

AW

NF DF

24.0 ± 1.28a 23.0 ± 0.86b

22.2 ± 0.77a 25.8 ± 2.42ab

24.3 ± 1.54a 21.4 ± 3.98b

23.2 ± 1.36a 26.7 ± 1.52a

23.5 24.2

0.93 2.43

3.98 10.1

Values are given as means ± standard error of means (n = 3). Values followed by different letters within a row are significantly different (P < 0.05). Av (%): the average of soil moisture during tomato growing season. Coefficients of variation (Cv %) and Standard deviation (Sd %) of soil moisture during tomato growing season. SD: Seedling stage. FW: Flowering stage. FS: Fruit-set period. PK: Picking period.

colation can be neglected under drip irrigation (Yuan et al., 2001; Chen et al., 2015), Eq. (3) was as follows: ET = Ir + W

(3)

where W is the change in soil water storage (mm). WUE was calculated with Eq. (4): WUE = Y/ET × 100

(4)

where Y is fruit yield (kg ha−1 ). Plant height and stem diameter were measured every 7 d using a ruler and Vernier caliper before tomato topping, respectively. All mature tomato fruits in each plot were manually harvested and fresh yield was measured at each picking time. 2.4. Statistical analysis Dates were analyzed by one-way ANOVA with the Duncan’s multiple range tests to separate means using SAS 9.1 statistical software. In all cases, differences were deemed to be significant if P < 0.05. 3. Results 3.1. Air temperature and relative humidity The variations of air temperature and relative humidity during the ES season tended to increase, and the mean daily air temperature and relative humidity were 24.6 ◦ C and 64.7% inside the greenhouse (Fig. 2). The summer fallow period had the highest air temperature (>30 ◦ C), with no vegetables grown in the greenhouses at this time. The air temperature during the AW season showed a downward trend from 33.8 ◦ C in August to 8.82 ◦ C in December. The mean daily air temperature and relative humidity inside the greenhouse during the AW season were 20.8 ◦ C and 76.6%, respectively.

3.2. Plant height and stem diameter Plant height and stem diameter increased rapidly from the beginning of the experiment onwards (Fig. 3), and then increased slowly after transplanting for 50 d. Plant height did not significantly differ between NF and DF treatments in the ES and AW seasons. However, stem diameter of the NF treatment was significantly higher than the DF treatment at 40 days later of transplanting, and increased by 16.6–19.0% and 31.0–34.5% in the ES and AW seasons (P < 0.05), respectively.

3.3. Soil moisture The soil water content was higher at the end of the growing season than at the beginning in the 0–40 cm soil layer under DF treatment (Fig. 4), and there were significant differences (P < 0.05) in AW season, indicating water recharge in soil during the tomato growth period. Soil water content in the 0–40 cm soil layer was higher at the end of the growing season with DF than with NF, this was linked to the higher irrigated amount in DF treatment, and was irrigated to 90% of field capacity at the end of the growing season. For NF treatment, in the 0–100 cm soil profile, the soil water content did not significantly differ between the beginning and harvesting, and the range was 23.8–25.0% and 23.0–24.0% in ES and AW seasons, respectively, indicating that the water in the soil profile was not significantly affected by tomato growth. There was a larger variation in soil surface moisture (0–20 cm) in the ES than AW season (Table 2), and this was linked to the higher temperature and stronger growth in the ES than AW season (Fig. 2). Soil moisture (0–20 cm) was relatively stable during tomato growing seasons for NF treatment, and the coefficients of variation (Cv) were 8.72 and 3.98 in ES and AW seasons, respectively, which were less than the corresponding values of 15.0 and 10.1 for the DF treatment.

6

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

Table 3 ET (mm) in different growth stages under NF and DF treatments in ES and AW seasons in 2014. Growing season

Treat.

Growth stage

Whole growth season

SD

FW

FS

PK

ES

NF DF

44.3 ± 6.39b 59.1 ± 8.04a

79.2 ± 10.2a 82.8 ± 8.41a

119.4 ± 10.9a 118.0 ± 0.58a

52.2 ± 10.1a 60.1 ± 3.48a

295.1 ± 21.4a 319.9 ± 8.83a

AW

NF DF

38.4 ± 1.32a 42.6 ± 4.48a

65.6 ± 3.13b 92.9 ± 6.26a

62.2 ± 2.22a 74.4 ± 8.74a

57.8 ± 1.52a 60.5 ± 7.08a

224.0 ± 6.83a 270.5 ± 11.9b

Values are given as means ± standard error of means (n = 3). Values followed by different letters within a row are significantly different (P < 0.05). SD: Seedling stage. FW: Flowering stage. FS: Fruit-set period. PK: Picking period.

Table 4 Tomato fruit yield and water use efficiency (WUE) in 2014 and 2015. Item

Treatments

ES season 2014

AW season 2014

AW season 2015

Yield (104 kg ha−1 )

NF DF

9.55 ± 0.10a 9.40 ± 0.54a

6.91 ± 0.02a 6.39 ± 0.45a

7.04 ± 0.18a 6.82 ± 0.17a

WUE (kg hm−2 mm−1 )

NF DF

32.3 ± 0.97a 29.4 ± 0.53b

29.8 ± 0.54a 20.1 ± 0.95b

29.6 ± 1.79a 24.1 ± 0.97b

Values are given as means ± standard error of means (n = 3). Values followed by different letters within a column are significantly different (P < 0.05).

3.4. SWS The SWS was relatively high during the tomato SD stage in ES and AW seasons, but low during FW and FS stages for NF treatment (Fig. 5). This may be attributed to stronger tomato growth and greater consumption of water at this stage. Even so, the SWS in NF (318.6–339.3 mm) over the whole growing season fluctuated less compared with DF treatment (315.7–342.9 mm), since the water from the NF system was applied to the subsurface, and was absorbed initially by tomato roots, so soil drying–wetting was avoided. The DF treatment showed opposite changes of SWS during FS and PK stages compared with the NF treatment. This was linked to the higher amount of drip irrigation for the DF treatment in this period (Table 1), being 198.2 and 144.4 mm in ES and AW seasons, respectively, compared with 184.8 and 123.0 mm for the NF treatment. Mean SWS, acquired by averaging the readings taken at the sampled tomato growth stages over the growing season, were much more uniform with no significant differences between DF (328.9 and 327.0 mm in ES and AW seasons, respectively) and NF treatments (correspondingly 331.4 and 324.3 mm). This demonstrated that SWS was adequate under NF conditions. 3.5. Growing season ET The ET of the FW and FS stages contributed to the greatest proportion (about 60%) of total ET for the whole growing season (Table 3). The PK stage accounted for the third highest level of ET, and the SD stage for the lowest. Total ET was significantly (P < 0.05) higher in the DF than NF treatment in the AW season, with no significant difference in the ES season. This was attributed to the higher irrigation in DF (283.5 mm in the AW season), which increased by 31.9% compared with NF treatment (P < 0.05). 3.6. Yield and WUE The tomato fruit yield of the NF treatment increased compared with the DF treatment, with increases of 1.6% in ES of 2014, 8.2% in EW of 2014 and 3.2% in AW of 2015 (Table 4). Despite the ET in the DF treatment being higher than for the NF treatment, the tomato yield was still rather low and it is likely that some water was lost through soil surface evaporation. There were significant increases in tomato WUE for the NF compared with the DF treatment (P < 0.05), with increases of 9.9–30.5%

in both years (Table 4). This was linked to the subsurface irrigation in NF treatment, and lower water loss by evaporation, which has no plant physiological significance, and also increases in transpiration.

4. Discussion Negative pressure irrigation differs from positive pressure irrigation (e.g. drip irrigation), and controls the pressure of the water source in negative values, providing a continuous supply of water according to crop growth needs (Zou et al., 2007). The results of this study were in agreement with those of Li et al. (2010a) who found that soil water content stabilized and did not change significantly with changes in crop growth under negative pressure irrigation conditions. While soil moisture (0–20 cm) of the NF treatment in the present study was 20.0–25.0% and 22.2–24.3% in ES and AW seasons (Table 2), respectively, this was less than the range of 19.7–28.5% and 21.4–26.7% in the DF treatment. In the NF system, the water feeder at 25 cm deep in soil (Fig. 1) supplied water directly to tomato roots (Li et al., 2011). Soil drying–wetting was avoided and the variation in soil moisture (0–20 cm) was more affected by crop growth than other factors. However, in the 0–100 cm soil profile, the soil water content with the NF treatment did not significantly differ between the beginning of the study and harvest (Fig. 4), indicating that water in the soil profile was not significantly affected by tomato growth. This result differed from the DF treatment, in which soil water content significantly differed (P < 0.05) in the 0–40 cm soil layer between harvesting and the beginning of the season (Fig. 4). Temperature in greenhouses can affect soil moisture, but this only appeared to affect the variation in soil moisture of the 0–20 cm soil layer (Table 2). The average air temperature was higher in the ES (24.6 ◦ C) than the AW season (20.8 ◦ C), and the variation in soil moisture (0–20 cm) was also greater for the ES season. Liu et al. (2013) reported that higher air temperature in solar greenhouses was one reason for higher soil evaporation, and there was a strong relationship between evaporation and air temperature with a correlation coefficient of 0.78 (Wang et al., 2009). In general, low or high SWS can negatively affect plant growth (Qi et al., 2011). In the present study, the average SWS did not significantly differ between DF and NF treatments (Fig. 5). The NF treatment had higher SWS than DF treatment only during the tomato seedling stage in the ES season, and the maximum SWS difference was 22.0 mm (P < 0.05). Compared with NF, the DF treatment with higher irrigation increased SWS during FW and FS stages

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

(Table 1, Fig. 5), indicating that SWS was influenced by irrigation amount. Tomato is characterized by high ET. In this study, the total ET over the whole growing season varied in the range of 224.0–319.9 mm. Liu et al. (2013) reported that the seasonal tomato ET was in the range of 249.1–388.0 mm in solar greenhouses in the North China Plain. In the present study, water consumption was greater during FS and FW than other growth stages, and accounted for 57.1–67.3% of total ET, coinciding with rapid increases in tomato plant height and stem diameter (Fig. 3). The ET was higher for DF than NF treatment, but with a significant difference only in the AW season (Table 2), and was linked to the 31.9% higher irrigation in DF compared with NF treatment (P < 0.05). Compared with DF, NF with higher frequency over a long period can maintain a constant soil moisture and nutrient concentrations in the root zone (Silber et al., 2003; Segal et al., 2006; Farneselli et al., 2015). High fertigation frequency has been found to improve crop performance in tomato (Badr et al., 2010; Hebbar et al., 2004; Liu et al., 2013), melon (Sensoy et al., 2007) and bell peppers (Sezen et al., 2006). In the present study, the tomato stem diameter under the NF treatment significantly increased by 16.6–19.0% and 31.0–34.5% at 40 days later of transplanting in the ES and AW seasons, respectively. Improved tomato growth was also reported by Li et al. (2008) with negative pressure irrigation supplying water stably and continually and enhancing the RuBP carboxylase activity of tomato leaves. Compared with the negative impact on crop growth of intermittent and pulsed supply of water and nutrient in the DF treatment, the consecutive water supply with NF improved the stability of the production system and increased tomato fruit yield (Table 4). Irrigation WUE was strongly influenced by water input and is the most important criterion used to assess a production system (Fan et al., 2014). Excessive use of water decreases WUE and increases the risk of environmental pollution (Du et al., 2014). In NF, water and nutrients were directly and precisely applied to the main root zone, which improved WUE by 9.9–30.5% compared with DF (Table 4). In addition, excessive irrigation enhances soil surface evaporation (Olesen et al., 2000) and would have caused a reduction of WUE in DF.

5. Conclusions Our results demonstrate that NF was able to supply water at the appropriate times, and soil water content in the 0–100 cm depth was much more stable than for DF during the tomato growth season. Mean SWS in the top 100 cm of the soil profile was adequate for tomato growth, with no significant difference between NF and DF treatments. Cumulative ET throughout the whole tomato growing season was higher in the DF than the NF treatment, and was affected significantly by the irrigation amount (P < 0.05). Compared with DF, NF treatment for applying Yamazaki tomato nutrient solution formula increased tomato height, stem diameter and fruit yield, and significantly improved WUE (P < 0.05). NF can save a substantially larger amount of irrigation water than DF, and can be used as a new mode of integrating water and fertilizer management for vegetable production in solar greenhouses.

Acknowledgments This work was supported by the National Natural Science Foundation of China (41501312, 51509005), the Talents Project of Beijing (2015000057592G267) and the National Science & Technology Pillar Program during the 12th Five-year Plan Period (2015BAD22B03). The authors are grateful to all the staff of the Beijing Xiaotangshan Experiment Station.

7

References Badr, M.A., Abou Hussein, S.D., EI-Tohamy, W.A., Gruda, N., 2010. Nutrient uptake and yield of tomato under various methods of fertilizer application and levels of fertigation in arid lands. Gesunde Pflanz. 62, 11–19. Cao, J., Lee, J.H., Six, J., Yan, Y., Zhang, F.S., Fan, M.S., 2015. Changes in potential denitrification-derived N2 O emissions following conversion of grain to greenhouse vegetable cropping systems. Agric. Ecosyst. Environ. 68, 94–100. Cetin, O., Uygan, D., 2008. The effect of drip line spacing, irrigation regimes and planting geometries of tomato on yield, irrigation water use efficiency and net return. Agric. Water Manage. 95, 949–958. Chen, Q., Zhang, X.S., Zhang, H.Y., Christie, P., Li, X.L., Horlacher, D., Liebig, H.P., 2004. Evaluation of current fertilizer practice and soil fertility in vegetable production in the Beijing region. Nutr. Cycl. Agroecosyst. 69, 51–58. Chen, J.L., Kang, S.Z., Du, T.S., Qiu, R.J., Guo, P., Chen, R.Q., 2013. Quantitative response of greenhouse tomato yield and quality to water deficit at different growth stages. Agric. Water Manage. 129, 152–162. Chen, R.Q., Kang, S.Z., Hao, X.M., Li, F.S., Du, T.S., Qiu, R.J., Chen, J.L., 2015. Variations in tomato yield and quality in relation to soil properties and evapotranspiration under greenhouse condition. Sci. Hortic. 197, 318–328. Du, T.S., Kang, S.Z., Zhang, X.Y., Zhang, J.H., 2014. China’s food security is threatened by the unsustainable use of water resources in North and Northwest China. Food Energy Secur. 3, 7–18. Fan, Z.B., Lin, S., Zhang, X.M., Jiang, Z.M., Yang, K.C., Jian, D.D., Chen, Y.Z., Li, J.L., Chen, Q., Wang, J.G., 2014. Conventional flooding irrigation causes an overuse of nitrogen fertilizer and low nitrogen use efficiency in intensively used solar greenhouse vegetable production. Agric. Water Manage. 144, 11–19. Farneselli, M., Benincasa, P., Tosti, G., Simonne, E., Guiducci, M., Tei, F., 2015. High fertigation frequency improves nitrogen uptake and crop performance in processing tomato grown with high nitrogen and water supply. Agric. Water Manage. 154, 52–58. Gao, Y.M., Li, J.S., Cao, Y.E., 2006. Study on the fertigation of tomato soil culture by drip irrigation in greenhouse. Acta Agric. Boreali-occidentalis Sin. 15, 121–126 (in Chinese). He, X.Q., Qiao, Y.H., Liu, Y.X., Dendler, L., Yin, C., Martin, F., 2016. Environmental impact assessment of organic and conventional tomato production in urban greenhouses of Beijing city, China. J. Clean. Prod. 134, 251–258. Hebbar, S.S., Ramachandrappa, B.K., Nanjappa, H.V., Prabhakar, M., 2004. Studies on NPK drip fertigation in field grown tomato (Lycopersicon esculentum Mill.). Eur. J. Agron. 21, 117–127. Ismail, S.M., Ozawa, K., Khondaker, N.A., 2008. Influence of single and multiple water application timings on yield and water use efficiency in tomato (var. First power). Agric. Water Manage. 95, 116–122. Karlberg, L., Rockström, J., Annandale, J.G., Steyn, J.M., 2007. Low-cost drip irrigation—a suitable technology for southern Africa: an example with tomatoes using saline irrigation water. Agric. Water Manage. 89, 59–70. Li, S., Xue, X.Z., Guo, W.S., Li, X., Chen, F., 2008. Effects of negative pressure irrigation on the growth, yield and quality of tomato in greenhouses. Trans. CSAE 24, 225–229 (in Chinese). Li, S., Xue, X.Z., Guo, W.S., Li, X., Chen, F., 2010a. Effects of water supply tension on photosynthetic characteristics and root activity of greenhouse cucumber. Chin. J. Appl. Ecol. 21, 67–73 (in Chinese). Li, S., Xue, X.Z., Guo, W.S., Li, X., Chen, F., 2010b. Effects of water supply tension on yield and water use efficiency of greenhouse cucumber. Sci. Agric. Sinic. 43, 337–345 (in Chinese). Li, S., Xue, X.Z., Qi, F., Zhou, C.J., Guo, W.S., Chen, F., 2011. Effects of different nutrient solution contents on yield and quality of greenhouse potted cucumber. Plant Nutr. Fertil. Sci. 17, 1409–1416 (in Chinese). Li, Y.K., Guo, W.Z., Wang, L.C., Xu, F., Li, Y.L., Sun, W.T., Chen, X.L., 2015. A kind of negative pressure irrigation and fertilization system. China Patent, No., 201520049881.6 (in Chinese). Li, X., Xie, Y.G., Wang, G.D., Xue, X.Z., 2016. Growth and physiological characteristics of eggplant under negative hydraulic head irrigation in greenhouse. Nat. Sci. Educ. 44, 163–169 (in Chinese). Liu, H., Duan, A.W., Sun, J.S., Liang, Y.Y., 2009. Effects of soil moisture regime on greenhouse tomato yield and its formation under drip irrigation. Chin. J. Appl. Ecol. 11, 2699–2704 (in Chinese). Liu, H., Duan, A.W., Li, F.S., Sun, J.S., Wang, Y.C., Sun, C.T., 2013. Drip irrigation scheduling for tomato grown in solar greenhouse based on pan evaporation in North China Plain. J. Integr. Agric. 12, 520–531. Marino, S., Aria, M., Basso, B., Leone, A., Alvino, A., 2014. Use of soil and vegetation spectroradiometry to investigate crop water use efficiency of a drip irrigated tomato. Eur. J. Agron. 59, 67–77. Olesen, J.E., Mortensen, J.V., Jorgensen, L.N., Andersen, M.N., 2000. Irrigation strategy, nitrogen application and fungicide control in winter wheat on a sandy soil. I. Yield, yield components and nitrogen uptake. J. Agric. Sci. 134, 1–11. Qi, Z.M., Helmers, M.J., Kaleita, A.L., 2011. Soil water dynamics under various agricultural land covers on a subsurface drained field in north-central Iowa, USA. Agric. Water Manage. 98, 665–674. Qiu, R.J., Kang, S.Z., Li, F.S., Du, T.S., Tong, L., Wang, F., Chen, R.Q., Liu, J.Q., Li, S.N., 2011. Energy partitioning and evapotranspiration of hot pepper grown in greenhouse with furrow and drip irrigation methods. Sci. Hortic. 129, 790–797. Qiu, R.J., Du, T.S., Kang, S.Z., Chen, R.Q., Wu, L.S., 2015. Assessing the SIMDualKc model for estimating evapotranspiration of hot pepper grown in a solar greenhouse in Northwest China. Agric. Syst. 138, 1–9.

8

Y. Li et al. / Agricultural Water Management 184 (2017) 1–8

Ren, T., Christie, P., Wang, J.G., Chen, Q., Zhang, F.S., 2010. Root zone soil nitrogen management to maintain high tomato yields and minimum nitrogen losses to the environment. Sci. Hortic. 125, 25–33. Segal, E., Ben-Gal, A., Shani, U., 2006. Root water uptake efficiency under ultra-high irrigation frequency. Plant Soil 282, 333–341. Sensoy, S., Ertek, A., Gedik, I., Kucukyumuk, C., 2007. Irrigation frequency and amount affect yield and quality of field-grown melon (Cucumis melo L.). Agric. Water Manage. 88, 269–274. Sezen, S.M., Yazar, A., Eker, S., 2006. Effect of drip irrigation regimes on yield and quality of field grown bell pepper. Agric. Water Manage. 81, 115–131. Shan, L.N., He, Y.F., Chen, J., Huang, Q., Lian, X., Wang, H.C., Liu, Y.L., 2015. Nitrogen surface runoff losses from a Chinese cabbage field under different nitrogen treatments in the Taihu Lake Basin, China. Agric. Water Manage. 159, 255–263. Silber, A., Xu, G., Wallach, R., 2003. High irrigation frequency: the effect on plant growth and on uptake of water and nutrients. Acta Hortic. 627, 89–96. Simonne, E.H., Studstill, D.W., Hochmuth, R.C., 2006. Understanding water movement in mulched beds on sandy soils: an approach to ecologically sound fertigation in vegetable production. Acta Hortic. 700, 173–178. Wang, Z.Y., Liu, Z.X., Zhang, Z.X., Zhang, Z.K., Liu, X.B., 2009. Subsurface drip irrigation scheduling for cucumber (Cucumis sativus L.) grown in solar greenhouse based on 20 cm standard pan evaporation in Northeast China. Sci. Hortic. 123, 51–57. Wang, C.X., Gu, F., Chen, J.L., Yang, H., Jiang, J.J., Du, T.S., Zhang, J.H., 2015a. Assessing the response of yield and comprehensive fruit quality of tomato grown in greenhouse to deficit irrigation and nitrogen application strategies. Agric. Water Manage. 161, 9–19. Wang, R.S., Kang, Y.H., Wan, S.Q., 2015b. Effects of different drip irrigation regimes on saline–sodic soil nutrients and cotton yield in an arid region of Northwest China. Agric. Water Manage. 153, 1–8.

Xu, G.P., Wang, P., Xue, X.Z., Zhang, F., Chen, F., 2014. Experiment on water use efficiency and yield of different plant type of potted maize under negative pressure water control. Trans. CSAE 30, 148–156 (in Chinese). Yuan, B.Z., Kang, Y.H., Nishiyama, S., 2001. Drip irrigation scheduling for tomatoes in unheated greenhouse. Irrig. Sci. 20, 149–154. Zhang, H.X., Chi, D.C., Wang, Q., Fang, J., Fang, X.Y., 2011. Yield and quality response of cucumber to irrigation and nitrogen fertilization under subsurface drip irrigation in solar greenhouse. Agric. Sci. China 10, 921–930. Zhang, L.D., Gao, L.H., Zhang, L.X., Wang, S.Z., Sui, X.L., Zhang, Z.X., 2012. Alternate furrow irrigation and nitrogen level effects on migration of water and nitrate-nitrogen in soil and root growth of cucumber in solar-greenhouse. Sci. Hortic. 138, 43–49. Zhang, H.M., Xiong, Y.W., Huang, G.H., Xu, X., Huang, Q.Z., 2017. Effects of water stress on processing tomatoes yield, quality and water use efficiency with plastic mulched drip irrigation in sandy soil of the Hetao Irrigation District. Agric. Water Manage. 179, 205–214. Zhu, J.H., Li, X.L., Christie, P., Li, J.L., 2005. Environmental implications of low nitrogen use efficiency in excessively fertilized hot pepper (Capsicum frutescens L.) cropping systems. Agric. Ecosyst. Environ. 111, 70–80. Zotarelli, L., Dukes, M.D., Scholberg, J.M.S., Munoz-Carpena, R., Icerman, J., 2009. Tomato nitrogen accumulation and fertilizer use efficiency on a sandy soil: as affected by nitrogen rate and irrigation scheduling. Agric. Water Manage. 96, 1247–1258. Zou, C.W., Xue, X.Z., Zhang, R.D., Geng, W., Li, S., Chen, F., 2007. Principle and equipment of negative pressure irrigation. Trans. CSAE 23, 17–22 (in Chinese).