Research site description
A 3-year stationary field experiment was conducted at the Yuanziqu experimental station of the Bayannur Academy of Agricultural and Animal Husbandry Sciences (40° 90′ N, 107° 17′ E), Linhe, Inner Mongolia, China, from 2019 to 2021. The site has a continental monsoon climate typical of the northern mid-temperate zone, with a mean annual temperature from 3.7 to 7.6 °C, and the potential evaporation is 2200–2400 mm15. The total precipitation during the wheat growth period (March–July) was 66 mm, 110 mm and 47 mm in 2019, 2020, and 2021, respectively. Daily air temperature and precipitation during the field trial period are presented in Fig. 1.
Daily maximum temperature, daily minimum temperature and precipitation during the growth period (March–July) of spring wheat from 2019 to 2021 in the field experiment at Linhe, Inner Mongolia, China.
The soil at the experimental site is a silt loam. The major physical and chemical properties of the 0–20 cm soil layer at the experimental site in 2019 were as follows: a bulk density of 1.48 g cm−3, pH 8.3, organic matter content of 15.49 g kg−1, total N concentration of 1.20 g kg−1, nitrate (NO3− N) concentration of 3.98 mg N kg−1, Olsen-P concentration of 32.3 mg P kg−1 and available K concentration of 180.0 mg K kg−1.
Experimental design and field management
A local popular spring wheat (Triticum aestivum L.) cultivar, Yongliang No. 4, was used in the trials. The sowing dates were 20 March, 16 March and 13 March in 2019, 2020 and 2021, respectively; the harvest dates were 15 July, 15 July and 5 July in 2019, 2020 and 2021, respectively. A total of 375 kg ha−1 wheat seeds were sown at a depth of 5 cm.
Five fertilization treatments were set in the three consecutive experimental years, including the control (CK), farmer practice (FP) and three balanced fertilization treatments (BF1, BF2 and BF3), as presented in Table 1. The P and K fertilizers as single superphosphate and K2SO4 were basal dressed, respectively. Both basal- and top-dressing of N fertilizer as urea were applied as shown in the research program. The basal applications occurred during sowing, and the remaining N fertilizer was top-dressed at the tillering stage (Table 1). The experimental plot was 10 m × 7 m with 13 cm row spacing and a buffer zone of 1 m between plots. The plots were laid out in a completely randomized block design with three replications.
Every plot was ridged around its border to ensure the uniformity of irrigation. Flood irrigation from Yellow River water was performed according to the local policy and farmer practices. Irrigation water (30 mm) was applied at the tillering, jointing, heading and grain filling stages of spring wheat in 2019–2021. Disease, weed, and pest control, as well as other management, were performed according to local standard methods.
Sampling and sample analysis
Three 50 cm-long rows of spring wheat plants were selected randomly and pulled from each plot, from which 10 large, middle and small seedlings were picked out during each sampling effort. Then, the roots were cut off from the junction between the root and the stem, two plant parts (leaves and stems) before the heading stage, three plant parts (leaves, stems and spikes) at the heading stage, and four plant parts (leaves, stems, glumes and grains) at the grain filling stages and maturity were separated and pooled. The samples were dried for 30 min at 105 °C and then at 80 °C in an oven (DHG-9070A, China) until they reached a constant weight; the dry weight was then measured.
The N concentrations in leaves, stems, spikes and grains of spring wheat were measured with three replications depending on the crop stage, following the Kjeldahl procedure using an element analyzer (Vario El cube, Elementar, Germany).
Three soil cores containing 0–100 cm of soil were taken from each plot using an auger at the harvest of spring wheat each year. The soil samples of each 20 cm layer were collected separately and sealed immediately in a marked plastic bag. The extracts were immediately measured for nitrate-N concentration as described by Dai et al. (2015) with a continuous flow analyzer (SKALAR SAN++, Netherlands)16. The soil nitrate-N concentration was expressed on the basis of the oven-dried soil.
Grain yield was evaluated at maturity by selecting two 2 m2 (avoiding border rows) randomly and harvested. A fresh weight of ∼ 1 kg of grain from each plot was weighed in the field, and the water content from each plot was oven dried for the calculation. The actual yield was adjusted by a grain water content of 13%17. Grains per spike, 1000-grain weight and spike number were determined at three 50-cm sites sampled randomly from each plot.
Calculation methods
To clarify the effect of nitrate residue in the soil under balanced fertilization, the amount of soil nitrate-N (AN, kg N ha−1) in each layer was expressed as:
$${text{AN}} = left( {{text{Ti}};*;{text{Di}};*;{text{Ci}}} right)/10$$
where Ti is the soil layer thickness (cm), Di is the soil bulk density (g cm−3), Ci is the soil nitrate concentration (mg N kg−1) of the corresponding layer, and 10 is the conversion coefficient16. The AN of 0–20, 20–40, 40–60, 60–80 and 80–100 cm soil layers were recorded and measured, respectively.
Nitrogen accumulation in the vegetative organs and their distribution into the grain were investigated. Based on the dry weight and corresponding measured N concentration in the different organs, apparent N translocation (TA, kg ha−1) and apparent N translocation efficiency (TR, %) were calculated as proposed by Cox et al.18 as follows:
$$begin{aligned} {text{TA}} & {text{ = H}}_{{text{N}}} – {text{M}}_{{text{N}}} {text{TR}} & = {text{TA/H}}_{{text{N}}} *100 end{aligned}$$
where HN is the N assimilation in leaves or stems prior to anthesis (kg ha−1), MN is the N assimilation in leaves or stems at maturity (kg ha−1).
Two parameters of nitrogen use efficiency of spring wheat, nitrogen fertilizer partial productivity (PFPN, kg/kg) and agronomic nitrogen efficiency (NAEN, kg kg−1) were determined using the following formulas:
$$begin{aligned} {text{PFP}}_{{text{N}}} & = {text{ Y}}_{N; , fertilizer} /{text{N}}_{rate} {text{NAE}}_{{text{N}}} & = , left( {{text{Y}}_{{N;fertilizer , {-}}} {text{Y}}_{blank} } right)/{text{N}}_{rate} end{aligned}$$
where YN fertilizer is the grain yield of the plot with dressed N fertilizer (kg ha−1), Yblank is the grain yield of the plot without dressed N fertilizer (kg ha−1), and Nrate is the N rate of the dressed fertilizer plot (kg ha−1). Three measurements for each treatment was recorded and calculated.
Two key indicators were chosen to evaluate the risk of N losses as described by Li et al. (2020)7, including N surplus (kg N per hectare per year, Nsur) and N input (kg N per hectare per year, Ninput). The N surplus was used to evaluate the risk of N losses and the N input to guide farmers’ fertilization practices directly. The detailed calculation is as follows:
$$begin{aligned} {text{N}}_{{{text{sur}}}} & = {text{ N}}_{{{text{fer}}}} + {text{ N}}_{{{text{dep}}}} + {text{ N}}_{{{text{fix}}}} – {text{ N}}_{{{text{har}}}} {text{N}}_{{{text{input}}}} & = {text{ N}}_{{{text{har}}}} + {text{ N}}_{{{text{sur}}}} + {text{ soil N change in stock }};;;( approx 0{text{ in long run}}) end{aligned}$$
where Nfer, Ndep and Nfix represent N from fertilizer, atmospheric deposition and biological fixed N, respectively. Seed N was negligible as it was present in a very small amount compared to the fertilization input19. The total N deposition of spring wheat field and biological N fixation were adopted according to Li et al. (2020). Nhar refers to the harvested N in spring wheat.
Economic analysis
The inputs into local spring wheat production included chemical fertilizer, irrigated water, agricultural chemicals, seed, mechanical effort and labor costs, while income was obtained from the grain and wheat straw. The net income was determined from the difference between the total output and total input. The irrigated water, agricultural chemicals, seed, mechanical effort and labor costs were the same for the different treatments. The prices of the input and output materials were determined according to the average local market prices, and fluctuations were not considered among years.
Statistical analyses
The results were analyzed using SPSS software (version 19.0; SPSS Inc., Chicago, IL, USA). Analysis of variance (ANOVA) and the least significant difference (LSD) test were used, and a P value of 0.05 was considered significant.
Source: Ecology - nature.com
