FENG Xu-yu ,PU Jing-xuan ,LlU Hai-jun ,WANG Dan ,LlU Yu-hang ,QlAO Shu-ting,LEl Tao,LlU Rong-hao,

1 College of Water Resources Science and Engineering,Taiyuan University of Technology,Taiyuan 030024,P.R.China

2 College of Water Sciences,Beijing Normal University,Beijing 100875,P.R.China

3 Beijing Key Laboratory of Urban Hydrological Cycle and Sponge City Technology,Beijing Normal University,Beijing 100875,P.R.China

4 Shanxi Province Key Laboratory of Soil Environment and Nutrient Resources,Taiyuan 030031,P.R.China

Abstract Alternate partial root-zone drip fertigation (ADF) is a combination of alternating irrigation and drip fertigation,with the potential to save water and increase nitrogen (N) fertilizer efficiency. A 2-year greenhouse experiment was conducted to evaluate the effect of different fertigation frequencies on the distribution of soil moisture and nutrients and tomato yield under ADF. The treatments included three ADF frequencies with intervals of 3 days (F3),6 days (F6) and 12 days (F12),and conventional drip fertigation as a control (CK),which was fertilized once every 6 days. For the ADF treatments,two drip tapes were placed 10 cm away on each side of the tomato row,and alternate drip irrigation was realized using a manual valve on the distribution tapes. For the CK treatment,a drip tape was located close to the roots of the tomato plants. The total N application rate of all treatments was 180 kg ha-1. The total irrigation amounts applied to the CK treatment were 450.6 and 446.1 mm in 2019 and 2020,respectively;and the irrigation amounts applied to the ADF treatments were 60%of those of the CK treatment. The F3 treatment resulted in water and N being distributed mainly in the 0-40-cm soil layer with less water and N being distributed in the 40-60-cm soil layer. The F6 treatment led to 21.0 and 29.0% higher 2-year average concentration of mineral N in the 0-20 and 20-40-cm soil layer,respectively and a 23.0% lower N concentration in the 40-60-cm soil layer than in the CK treatment. The 2-year average tomato yields of the F3,F6,F12,and CK treatments were 107.5,102.6,87.2,and 98.7 t ha-1,respectively. The tomato yield of F3 was significantly higher (23.3%) than that in the F12 treatment,whereas there was no significant difference between the F3 and F6 treatment. The F6 treatment resulted in yield similar to the CK treatment,indicating that ADF could maintain tomato yield with a 40% saving in water use. Based on the distribution of water and N,and tomato yield,a fertigation frequency of 6 days under ADF should be considered as a water-saving strategy for greenhouse tomato production.

Keywords: alternate partial root-zone irrigation,drip fertigation,soil water,soil mineral content,tomato yield

1.lntroduction

Alternate partial root-zone irrigation (APRI) is a highly efficient water-saving irrigation technology that has been extensively researched and practiced in China and abroad(Selimet al.2012;?olaket al.2018;Abboudet al.2021;Wanget al.2021). APRI alternates the supply of water to different parts of the root zone. Roots in the drying zone can produce a drying signal that restricts stomatal opening,reducing excessive plant transpiration and soil evaporation(Kang and Zhang 2004;Kirdaet al.2007;Liuet al.2008;Forner-Gineret al.2011). APRI may save irrigation water by 30.0-50.0% and improve water use efficiency (WUE) by 26.7-61.0%,without significantly affecting photosynthesis(Leibet al.2005;Shahnazariet al.2007;Duet al.2008;Wanget al.2012,2019). In addition,APRI can cause a degree of water stress in crops,thereby stimulating root growth,enhancing root activity,and markedly increasing the root-to-shoot ratio (Yanget al.2010;Zhanget al.2017;Shuet al.2020). Moreover,APRI can promote nutrient transfer from stems and leaves to fruit,improving fruit quality and maintaining crop yield (Kang and Zhang 2004;Duet al.2007,2008;Wanget al.2019;Shuet al.2020).

APRI alternately supplying water to both sides of roots results in the soil water content on the wetted side being higher than on the dry side (Bauerleet al.2008). Given the water potential gradient,moisture in the soil moves from the wetter side to the drier side (Duet al.2007). As irrigation proceeds,the water content difference between both sides gradually decreases (Duet al.2008). It has been found that APRI decreased the vertical infiltration depth of soil moisture by 20-40 cm,and that deep leakage of soil moisture was reduced,compared with conventional irrigation (Pan and Kang 2000;Wanget al.2019).

Research has shown that APRI can promote nutrient uptake by crops. Lehrschet al.(2000) reported that APRI increased soil N uptake by maize by 21.0% compared with furrow irrigation. Wanget al.(2012) also revealed that,compared with conventional drip irrigation (CDI),APRI increased the agronomic N use efficiency (NUE) by 11.0%.Alternate dry and wet conditions on both sides of the root zone is conducive to promoting the growth and uptake capacity of roots (Skinneret al.1998;Liuet al.2020). In addition,the frequent dry-wet cycles are beneficial to soil respiration and promote bacterial growth related to nutrient metabolism,thereby accelerating soil nutrient transformation(Wanget al.2017). Moreover,APRI can reduce the leaching of nitrate and stimulate soil N mineralization,thus increasing nitrate accumulation in the upper soil layer and promoting nutrient uptake (Shahnazariet al.2008;Wanget al.2008b;Topaket al.2016).

Drip fertigation can provide water and fertilizer directly to the crop root zone conveniently and efficiently and reduce irrigation and fertilizer input. The frequency of fertigation affects the amount of water and N applied each time,thereby impacting soil water and nutrient concentrations near the roots (Zotarelliet al.2009). The frequency of fertigation also affects the accumulation and distribution of plant biomass(Hebbaret al.2004;Sensoyet al.2007),along with N uptake and crop yield (Zotarelliet al.2009;Uzen and Cetin 2016).ADF combines the advantages of APRI and drip fertigation,which has the potential to improve both WUE and NUE (Liuet al.2020). It was reported that the root N uptake efficiency of ADF was 2.5% higher than that of CDI (Fuet al.2017). In addition,ADF was found to induce more15N accumulation and promote plant uptake of both fertilizer-15N and soil N than CDI (Liuet al.2020). Under ADF conditions,the alternate supply of water and fertilizer to both sides of the root system make the dynamic changes in soil moisture and nutrients more complex. Thus,determining the distribution and migration of water and N in the crop root zone under ADF will be of significance for improving WUE and fertilizer use efficiency in farmland.

The objective of this experiment was to study the effects of different fertigation frequencies on soil water and nutrient distribution and tomato yield under ADF conditions. It was hoped that the experimental results would provide insights for the effective utilization of water and N fertilizers in tomato production.

2.Materials and methods

2.1.Experimental site

A greenhouse experiment was conducted from May 2019 to October 2020 in Hecun Village (38°04′N,112°89′W),Yangqu County,Taiyuan City,China. This region experiences a semiarid continental monsoon climate,with annual average temperature,precipitation,and evaporation of 6.0°C,459 mm,and 1 546.9 mm,respectively. The experimental soil was sandy loam,with a pH of 8.4,a bulk density of 1.43 g cm-3,field capacity of 0.31 cm3cm-3,1.4%organic C,0.91 g kg-1total N,29.9 mg kg-1Olsen P,and 173 mg kg-1exchangeable K in the soil depth of 0-20 cm.Meteorological data were collected from an automatic weather station installed inside the greenhouse.

The solar greenhouse was 50-m long in the south-north direction and 7.5-m wide in the east-west direction,and had a maximum height of 3.5 m. The total area cultivated in the greenhouse was 268.8 m2,and the crop rows were in an east-west direction. The greenhouse was covered by a transparent polyethene film. A removable membrane section of～50 cm at the bottom of each side of the greenhouse served as a vent. These were removed manually as required by the internal conditions of the greenhouse.

2.2.Experimental treatments

The field experiment was laid out in May 2019 and comprised four treatments,which included three frequencies of ADF,being 3 days (F3),6 days (F6) and 12 days (F12)intervals,and a conventional drip fertigation management with fertigation every 6 days (CK). The irrigation amount applied in the CK treatment was based on the cumulative evaporation measured using a 20-cm diameter pan (11-cm deep) (Liet al.2021). This was placed 20 cm above the crop canopy in the center of the greenhouse (Yuanet al.2001) and measured daily at 08:00 a.m. All ADF treatments were irrigated at the same rate. ADF treatments were irrigated at their specified intervals and the amount applied corresponded to 60% of the cumulative evaporation from a 20-cm diameter pan since the previous irrigation (Houet al.2017). Seasonal variation in temperature and evaporation rate from the pan (Ep) during the experimental periods in 2019 and 2020 is shown in Fig.1. In 2019,irrigation started on May 16 and ended on September 12. In 2020,irrigation began on May 22 and ended on September 9. The F3,F6,F12 and CK treatments were irrigated 40,20,10 and 20 times,respectively,in 2019,and 37,19,10 and 19 times,respectively,in 2020. The total irrigation amounts applied for the CK and each ADF treatment were 450.6 and 279.4 mm in 2019,respectively,whereas,in 2020,they were 446.1 and 267.7 mm,respectively. The treatments were arranged in a completely randomized block design with three replications.Each plot had a total area of 6.72 m2(2.8 m×2.4 m) and comprised four rows. Spaces between row and plants were 60 and 30 cm,respectively.

Fig.1 Daily means of temperature (A),evaporation rate from the pan (Ep,B),and cumulative Ep (B) at the experimental site for the study period in 2019 and 2020.

Uniform seedlings of tomato (LycopersiconesculentumMill.,cv.Hongfuxing) were transplanted on 6 May 2019 and 11 May 2020. All treatments received an initial irrigation(10 mm) to ensure uniform seedling establishment. The treatment applications started 10 days after transplantation.Each treatment was applied using an independent gravity drip irrigation system,which consisted of a bucket (350-1 000 L) installed 120 cm above the ground. For the CK treatment,a drip tape was located close to the roots of the tomato plants. For the ADF treatments,two drip tapes were placed 10 cm away on each side of the tomato row,and alternate drip irrigation was realized using manually operated valves on the distribution tapes. The emitter interval on the drip tapes was 20 cm,and the flow rate of the emitter ranged from 0.69 to 0.98 L h-1.

All treatments received the locally recommended fertilizer doses (180 kg ha-1N;225 kg ha-1potassium sulfate;and 750 kg ha-1superphosphate) (Xinget al.2015).The source of N fertilizer used was urea (46% N);25% of the N fertilizer was applied before the fruit development stage of the first cluster,with the remaining 75% applied over the subsequent stages. Urea was firstly dissolved into water in a bucket and then fertilized through the drip irrigation system. N fertilizer was added to each irrigation.The amount of N used per application was calculated by dividing the fertilizer amount for the corresponding period by the number of irrigations. Potassium sulfate was applied through drip irrigation in five splits,at a ratio of 1:1:2:2:2 at the transplanting,seedling,first fruit enlargement,second fruit enlargement,and third fruit enlargement stages,respectively. The P fertilizer was evenly scattered onto the soil surface as a basal fertilizer before the ground was manually ploughed.

A 60-cm deep underground impermeable plastic film was buried between plots to prevent cross movement of water and nutrients. Plants were pruned to about two leaves above the fourth layer of flowering stage. The crop season ended on September 28 in 2019 and September 29 in 2020.All operations relating to weed and pest control were the same for all the treatments and were according to local management practices.

2.3.Data collection

To compare the effects of the different irrigation frequencies on soil water and mineral N content,soil samples were taken approximately every 30 days during the 2019 and 2020 seasons. Soil cores were obtained from each plot using an auger. Soil samples were obtained at 0,5,10,20,and 30 cm away from the tomato stem on either side(a total of nine soil sampling locations in each plot) and at four depths at each location,being 0-10,10-20,20-40,and 40-60 cm deep. The samples collected were stored at 4°C until analyzed.

Soil NH4+-N and NO3--N were extracted by agitation of 5 g soil samples in 50 mL of 2 mol L-1KCl solution. The concentration of NH4+-N and NO3--N in the liquid supernatant was determined by ultraviolet spectrophotometry. The mineral N content was calculated as a sum of NH4+-N content and NO3--N content (Fanet al.2020). The remaining soil samples were used for measuring gravimetric soil water content by placing samples in aluminum trays and weighing them before and after drying at 105°C to a constant dry weight. The soil water and mineral N contents of each layer were calculated as the average of the nine points.

During the fruit-ripening period,all fruits in the inner two rows of crop plants in each plot were harvested to measure their fresh yield. At the end of the planting season,the fresh yield of each harvest was summed to give the total yield.

2.4.Statistical analysis

Golden Software Surfer 13.0 was used to draw a spatial distribution map of soil mineral N. One-way analysis of variance (ANOVA) was used to compare tomato yields among treatments with IBM SPSS Statistics Version 26.0(IBM,Armonk,NY,USA). Duncan’s multiple range test was used to determine the significance of differences between treatment means atP=0.05.

3.Results

3.1.Effect of fertigation frequency on soil moisture changes

The gravimetric soil water content in the 0-20 and 20-40-cm layers generally increased with increasing fertigation frequency among the ADF treatments through the two growing seasons (Fig.2-A,B,D and E). Throughout the entire growth period,the F3 treatment increased the 2-year average soil water content in the 0-20-cm layer by 7.5 and 15.9%,and in the 20-40-cm soil layer by 3.5 and 7.6%,respectively,compared with the F6 and F12 treatments. In the 40-60-cm soil layer,the F3 treatment had the lowest soil water content during the entire growth periods. During the whole growth period,the average soil water content in the F12 treatment in the 40-60-cm soil layer was 10.3 and 4.2%higher than that in the F3 and F6 treatments over the two seasons,respectively. The CK treatment,with 40% more water input,resulted in higher soil water content compared with the ADF treatments in all three soil layers.

Fig.2 Gravimetric soil moisture dynamics in the 0-20 cm (A and D),20-40 cm (B and E),and 40-60-cm (C and F) soil layers under different fertigation frequencies in 2019 and 2020,respectively. CK,conventional drip fertigation with an interval of 6 days;F3,F6,and F12,alternate partial root-zone drip fertigation (ADF) frequencies with intervals of 3,6 and 12 days,respectively.Vertical bars represent SE (n=3).

With the same irrigation frequency,the variation in soil water content in the CK and F6 treatments were similar(Fig.2). During the entire growth period,the 2-year average soil water content in the CK treatment was 16.0,14.2,and 13.2% higher than in the F6 treatment in the 0-20,20-40,and 40-60-cm soil layers,respectively. The 2-year average soil water content of the CK treatment in the whole soil profile(0-60 cm) was 14.4% higher than that in the F6 treatment.

3.2.Effect of fertigation frequency on changes in mineral nitrogen dynamics

The mineral N content in the 0-20 and 20-40-cm soil layers increased with the increase in fertigation frequency (Fig.3-A,B,D and E). During the entire growth period,the F3 treatment increased the 2-year average mineral N content in the 0-20-cm soil layer by 10.5 and 21.3% and by 37.1 and 57.7%in the 20-40-cm soil layer compared with the F6 and F12 treatments,respectively. In the 40-60-cm soil layer,the F3 treatment generally had the lowest mineral N content during both growth periods. In the 40-60 cm soil layer,the 2-year average mineral N content of the F12 treatment was 26.6 and 14.8% higher than that of the F3 and F6 treatments,respectively.

Fig.3 Soil mineral nitrogen concentration dynamics in the 0-20 cm (A and D),20-40 cm (B and E),and 40-60-cm (C and F) soil layers under different fertigation frequencies in 2019 and 2020,respectively. CK,conventional drip fertigation with an interval of 6 days;F3,F6,and F12,alternate partial root-zone drip fertigation (ADF) frequencies with intervals of 3,6 and 12 days,respectively.Vertical bars represent SE (n=3).

With the same fertilization frequency,the F6 treatment generally had a higher mineral N content in the 0-20 and 20-40-cm soil layers than the CK treatment. Throughout the growth period,the 2-year average mineral N content in the F6 group in the 0-20 and 20-40-cm soil layers were 21.0 and 29.0% higher than in the CK treatment,respectively. In addition,ADF treatments generally resulted in lower mineral N content compared with the CK treatment in the 40-60-cm soil layer. Throughout the growth period,the 2-year average mineral N content of the 40-60-cm soil layer in the F6 treatment was 23.0% less than in the CK treatment,indicating that ADF decreased the downward movement of mineral N to the lower soil layer.

3.3.Effect of fertigation frequency on distribution of mineral nitrogen in the soil profile

At tomato harvest,the distribution of mineral N in the soil profile showed that,in a horizontal direction,the highest mineral N content of each treatment mainly occurred within 0-15 cm of the dripper,showing different degrees of surface aggregation (Fig.4). In 2019 and 2020,the mineral N content in each ADF treatment gradually decreased with the increase in soil depth,and was mainly concentrated in the 0-40-cm soil layer. In this soil layer,soil mineral N content was greater under high frequency treatments (Fig.4-A-C and E-G). In the 0-40-cm soil layer,the mineral N content in the F3 treatment (7.3-26.1 mg kg-1in 2019 and 8.7-39.0 mg kg-1in 2020) was higher than in the F6 and F12 treatments.Below the 40-cm soil layer,the mineral N content increased with decrease in fertigation frequency.

The mineral N content in the F6 treatment in the 0-40-cm soil layer was higher than in the CK treatment over both experimental years. In the 40-60-cm soil layer,the mineral N content of the two treatments was opposite to that of the 0-40-cm soil layer;that is,the mineral N content in the CK treatment was higher than in the F6 treatment.

3.4.Effect of fertigation frequency on tomato yield

In ADF treatments,the tomato yield increased with increasing fertigation frequency (Fig.5). The 2-year average yield in the F3 treatment was 4.7 and 23.3% higher than in the F6 and F12 treatments,respectively. The difference between the F3 treatment and the F12 treatment was statistically significant,and the difference between the F3 treatment and the F6 treatment was not statistically significant. With the same fertigation frequency and fertilizer amount,the yield of tomatoes in the F6 treatment was similar to that in the CK treatment.

Fig.5 Effects of different fertigation frequencies on tomato yield in 2019 (A) and 2020 (B). CK,conventional drip fertigation with an interval of 6 days;F3,F6,and F12,alternate partial root-zone drip fertigation (ADF) frequencies with intervals of 3,6 and 12 days,respectively. Treatments without a letter in common are significantly different at P＜0.05. Vertical bars represent SE (n=3).

4.Discussion

4.1.Effect of fertigation frequency on soil moisture dynamics

Irrigation frequency is one of the main factors affecting soil moisture distribution throughout the soil profile (Wanget al.2008a). High-frequency irrigation usually results in an increased water content of the upper soil layer (El-Hendawyet al.2008). This present study showed that,throughout the entire growth period,irrigation with the F3 treatment increased the 2-year average soil water content in the 0-40-cm soil layer by 5.4 and 11.5% compared with the F6 and F12 treatments,respectively (Fig.2-A,B,D and E). This was because with high frequency irrigation a small amount of water was applied multiple times and thus the upper soil layer was more likely to be fully wetted,which reduced water migration (Waddellet al.2000;Caoet al.2003;Guoet al.2018). It was found elsewhere that under drip irrigation,more than 95% of tomato roots were distributed in the 0-40 cm soil layer (Hammami and Daghari 2007;Hammami and Zayani 2009),which could improve water availability to crops when using high frequency irrigation. A higher water content of the upper soil layer is conducive to crop growth and irrigation WUE (Wanget al.2006). Low frequency irrigation can result in a higher deep soil water content compared with high-frequency irrigation.The current study showed that,in the 40-60-cm soil layer,the water content in the F12 treatment was 10.3 and 4.2%higher than that of the F3 and F6 treatments,respectively(Fig.2-C and F). This was because of the large amount of water applied in a single irrigation at low irrigation frequency,which allowed water to easily move toward deep soil. It has been reported that a lower irrigation frequency resulted in a higher infiltration depth and,once the water had penetrated to lower soil depths,it could not be depleted by plant roots(El-Hendawyet al.2008;Shiet al.2018).

With the same irrigation frequency,the water content in the soil profile of the F6 treatment was less than that in the CK treatment (Fig.2). Throughout the two growing seasons,the average soil water content in the 0-20,20-40,and 40-60-cm soil layers in the CK treatment were 16.0,14.2,and 13.2% higher than that in the F6 treatment,respectively. This was because the amount of irrigation in the F6 treatment was 40% lower than that in the CK treatment. Previous studies have shown that APRI had a lower irrigation volume than conventional irrigation and water was less prone to deep percolation (Liet al.2007;Baridehet al.2018). Additionally,there is a water potential gradient between the wet and dry areas under APRI,which increases the radial penetration rate and reduces the downward movement of water (Ebrahimianet al.2013;Kumaret al.2014). Moreover,APRI results in more developed roots,which enables crops to absorb a large amount of water before it moves to deeper soil (Wanget al.2019).

4.2.Effect of fertigation frequency on soil mineral nitrogen

Different fertigation frequencies,with different times and amounts of fertigation,affected the distribution of N in the soil. At tomato harvest,the maximum mineral N content in the horizontal direction in each treatment was mainly concentrated within 0-15 cm of the dripper,and showed different degrees of surface aggregation (Fig.4). This is possibly due to the soil N around the emitter being gradually replaced by the fertilizer solution as the fertigation proceed,so the mineral N concentration tended to decrease as the distance from the emitter increased (Liet al.2009). This is consistent with other research showing that mineral N was mainly distributed in the soil profile 15 cm from the emitter(Haynes 1990;Coteet al.2003).

Generally,a higher fertigation frequency results in a higher mineral N content in the upper soil. Farneselliet al.(2015) found that,under the same N application conditions,root zone nitrate levels were higher with high-frequency irrigation than for low-frequency irrigation. The current study showed that,in the 0-20-cm soil layer,the 2-year average mineral N content in the F3 treatment was 10.5 and 21.3% more than in the F6 and F12 treatments,respectively(Fig.3-A and D),whereas,in the 20-40-cm soil layer,the values were 37.1 and 57.7%,respectively (Fig.3-B and E).This might be because multiple small amounts of fertigation provided more opportunities to replenish N and reduce the risk of leaching loss (Silberet al.2003;Badret al.2010).In the 40-60-cm soil layer,the 2-year average mineral N content of the F12 treatment was 26.6 and 14.8% higher than in the F3 and F6 treatments,respectively (Fig.3-C and F). Previous research showed that a low fertigation frequency led to higher N leaching in soil (Ramoset al.2012)and increased residual nitrate in the deep soil layer (Xianget al.2018). This might be because of the high one-time fertilization with low fertigation frequency,which might result in N not being absorbed by crops and,thus,the downward migration of N from the upper soil layers,thereby increasing the mineral N content in the deeper soil layers.

Under the same N fertilizer supply,the F6 treatment had a 21.0 and 29.0% higher mineral N content in the 0-20-and 20-40-cm soil layers respectively than the CK treatment(Fig.3-A,B,D and E). This is consistent with other research showing that APRI can improve soil surface N accumulation compared with CDI (Wanget al.2010;Liuet al.2020). The average mineral N across both years in the F6 treatment was 23.0% less than in the CK treatment in the 40-60-cm soil layer (Fig.3-C and F). This was because the F6 treatment had 40% less water applied,which is beneficial for reducing the migration of N to deeper soil layers (Xianget al.2018).In addition,APRI alternates wetting and drying on either side of the root zone,resulting in lateral transport and reducing the downward movement of nutrients (Baridehet al.2018).

4.3.Effect of fertigation frequency on tomato yield

Previous studies have shown that high-frequency fertigation can increase the total root and fine root biomass of plants,promote nutrient uptake,and improve yield (Hebbaret al.2004;Bhatet al.2007;Behera and Panda 2009;Singhet al.2011). The present results also showed that tomato yield increased with increasing fertigation frequency (Fig.5),which was 23.3% higher in the F3 treatment than in the F12 treatment. These findings supported those of Kumaret al.(2016),who reported that the yield of corns with 3 days of fertigation was significantly higher than that for 14 days of fertigation. In the current study,this was because the soil moisture in the F3 treatment remained at a level conducive to nutrient absorption and improved crop growth and yield(Fig.2-A,B,D and E) (Xianget al.2018). In addition,F3 treatment increased the mineral N content in the 0-40 cm soil layer (Fig.3-A,B,D and E),indicating that this treatment resulted in suitable nutrient conditions for this crop. Moreover,a frequent supply of nutrients can increase the internal nutrient flow in crops,thereby enhancing the transport of dissolved nutrients and improving nutrient absorption (Silberet al.2003).

The current results showed that there was no significant difference in yield between the F3 and F6 treatments (Fig.5).These results are consistent with those of other authors(Badr and Abou El-Yazied 2007;Kumaret al.2016),who reported that 3-and 7-day fertigation schedules did not increase crop yield. Although the water and nutrient contents in the F3 treatment were higher than in the F6 treatment,there was no difference in yield. This may be because of the increase in crop root penetration,which can offset the small and transient changes in nutrient concentration(Farneselliet al.2015). In addition,frequent fertigation practices are difficult to manage (Simonneet al.2006;Farneselliet al.2015). Therefore,a 6-day frequency of ADF is recommended.

Previous studies have shown that APRI can reduce irrigation water amount and improve WUE compared with traditional irrigation,without significantly reducing yield (Kanget al.2000;Fuet al.2017). Houet al.(2017) showed that APRI can save～34.9% of the irrigation water amount and maintain tomato yields,relative to conventional irrigation.Wanget al.(2019) also reported that APRI could save 38.8% of irrigation water,and produce a similar tomato yield compared with conventional irrigation. The current study revealed that the F6 treatment,with the same fertigation frequency and 40% less water application,resulted in a similar tomato yield compared with the CK treatment (Fig.5).This was probably because the F6 treatment resulted in higher levels of mineral N in the 0-20-and 20-40-cm soil layer (Fig.3-A,B,D and E). It has been reported that tomato mainly absorbs nutrients from the 0-40-cm soil layer (Wanget al.2019). Additionally,APRI enhances root growth and activity,and promotes N uptake compared to traditional drip irrigation (Liuet al.2020). It also stimulates soil organic N mineralization,which could help to increase nutrient uptake by the crop (Baridehet al.2018). In addition,APRI promotes N transfer from stems and leaves to tomato fruit (Wanget al.2019;Liuet al.2020),indicating that ADF treatment saves 40% of irrigation water and maintains yield compared with conventional approaches.

5.Conclusion

This study describes the soil water and N distribution under different ADF frequencies. The study showed that higher fertigation frequency (3-day intervals) resulted in a higher water and mineral N content in the 0-40-cm soil layer,which reduced the downward movement of mineral N,promoted the absorption of N by the crop,and improved the tomato yield. The fruit yield in the F3 treatment was the highest,being significantly higher than in the F12 treatment,although there was no significant difference between the F3 and F6 treatments. Compared with conventional drip fertigation,the F6 treatment resulted in a higher mineral N content in the 0-40-cm soil layer,and a lower mineral N content in the 40-60-cm soil layer,and maintained a high tomato yield. In summary,ADF fertigation once every 6 days had the greatest potential for saving irrigation water and reducing the downward movement of N,while maintaining the tomato yield. However,this study only evaluated the effects of different fertigation frequencies on soil moisture,N distribution,and yield under ADF;thus,a suitable N level for tomato production needs further study.

Acknowledgements

The study was supported by the National Natural Science Foundation of China (51809189) and the Shanxi Province Key Laboratory of Soil Environment and Nutrient Resources,China (2019002).

Declaration of competing interest

The authors declare that they have no conflict of interest.