1 Introduction

Intercropping, which involves simultaneous growth of two or more crop species within the same space during a considerable part of their life span, is mainly employed in organic or sustainable farming systems around the world, including tropical, subtropical, and temperate regions (Lithourgidis et al. 2011). Farming industries in many regions are showing increased attention to intercropping systems because of growing concerns over the environmental implications of the use of chemically intensive farming practices (Fujita et al. 1992; Ehrmann and Ritz 2014). Previous research has shown that intercropping, especially cereals with legumes, had several major advantages over monocropping, such as maximized crop growth, higher yield (Cecilio et al. 2011), and better land use efficiency (Zhang and Li 2003). Intercropping practice can also utilize resources more efficiently (Javanmard et al. 2009), increase soil microbial diversity, balance the nutrients in the soils (Hauggaard-Nielsen and Jensen 2005), reduce damages caused by pests and diseases (Hauggaard-Nielsen et al. 2001), and improve C and N dynamics (Oelbermann and Echarte 2011).

Wheat (Triticum aestivum L.) and faba bean (Vicia faba L.) are traditional and abundant cash crops grown in subtropical, arid, and semi-arid regions. Faba bean is well known to be one of the most commonly used grain legumes intercropped with wheat to enhance benefit and reduce environmental impacts. Specifically, when faba bean is intercropped with wheat, the benefits to non-N fixing wheat include increased yield, enhanced land equivalent ratio, improved grain quality, and complementary use of N. Studies to investigate the general performance of intercropping have mainly focused on crop growth, yield and quality of the grain, aboveground nutrient-use efficiency, and availability (Vandermeer 1989; Knudsen et al. 2004). However, the effect of belowground interactions between wheat and faba bean on yield have not been fully studied.

Nitrogen is an indispensable nutrient for crop growth and production. The suitable N management should fulfill crop N requirements, and, at the same time, lower the possibility of N leaching into groundwater. Thereby, the introduction of intercropping systems that allow optimal use of atmospheric and soil N sources to sustain improved yield has been proposed (Ehrmann and Ritz 2014). Atmospheric N fixation and soil inorganic N supply can be complementarily utilized by the companion crops in cereal-legume intercropping. Cereals that have deeper and finer root growth outcompete legumes to uptake and utilize more inorganic soil N within cereal and legume intercrops (Jensen 1996). In wheat/faba bean intercropping system, substituted N input to the soil N pool is attributable to the symbiotic N2 fixation ability of faba bean (Hauggaard-Nielsen et al. 2001). The atmospheric N fixed by faba bean is available to supplement the N demand of wheat. Thus, symbiotic N2 fixation relative to N uptake by faba bean within the intercropping system is especially of relevance in low soil fertility or low input systems.

The dynamics of N transformation processes determine the supply, conservation, and losses of N in the rhizosphere of intercropping plants, and govern crop N uptake, N use efficiency, and yield of crops (Stark 2000). To date, the mechanisms for improving crop production in intercropping system have not been fully examined in relation to the dynamics of N transformation in soils. Therefore, the main aim of this study was to investigate the relationship between intercropping and crop production and develop a better understanding of N transformation within such intercropping systems.

The 15N dilution technique was recognized as the most widely used method to quantify gross N transformation rates in soils (Häbteselassie et al. 2006; Wan et al. 2009; Zhu et al. 2011). The latest developments of numerical 15N tracing technique offer opportunities to simultaneously estimate specific N transformations among several processes (Mary et al. 1998; Müller et al. 2007; Rütting and Müller 2007), and to acquire a deeper understanding of the gross N transformation within cereal-legume intercropping practices. A field experiment and a 15N incubation study were established to explore the underlying mechanisms for enhanced crop productivity in the intercropping system based on the N transformation processes that are affected by the above- and belowground interactions of interspecies.

Therefore, the primary objective of the study was to assess the effect of the N availability improvement on yield along with aboveground biomass of intercrops; secondly, to evaluate the soil gross N transformation, net N nitrification, and mineralization rates in wheat/faba bean intercrops, as compared with their sole crops; and thirdly, to identify the potential environmental benefits by adopting such intercropping practice in arable farming system.

2 Materials and methods

2.1 Field site and experimental design

A field study of intercropping between wheat and faba bean during the 2014–2015 cropping seasons (November–April) was established at an experimental site of Yunnan Agricultural University in Kunming city, Yunnan province, southwest China (25° 7′ N, 102° 45′ E, 1895 m asl). The experimental site was characterized as a moderately subtropical climate; and annual precipitation was c. 1000 mm, of which approximately 85% fell in the season from May to October. The average annual air temperature over the past 30 years was 15 °C.

The soil was a moderately well-drained sandy loam (sand = 52%; silt = 43%; clay = 5%) that had moderate fertility (Table 1). The field site had been previously used for wheat and faba bean intercropping for 1 year. The field study encompassed three treatments: (1) wheat sole crop with 0.2-m row width, (2) faba bean sole crop with 0.3-m row width and 0.2-m plant distance, and (3) wheat/faba bean intercropping with six rows of wheat (0.2-m spacing between adjacent rows) and two rows of faba bean (with 0.3-m spacing between adjacent rows). The field study was conducted in a random block design with three replicates, each with a plot area of 8 × 6 m. Fertilizers application rates to the wheat crops were nitrogen (N at 225 kg ha−1), phosphate (P at 75 kg ha−1), and potassium (K at 75 kg ha−1). Basal fertilizers applied before sowing were half of the N fertilizer, and all of the P and K fertilizer, with the remaining half of the N fertilizer top-dressed at wheat jointing stage. The faba bean crop was applied the same P and K rate as the wheat crop but with only half of N (112.5 kg ha−1). All of the fertilizers were applied as basal fertilizers to faba bean. The fertilizers used in the experiment were urea, superphosphate, and potassium sulfate. Field preparations, inter-row distances, fertilization, and irrigation corresponded to the common cultivation practice in the region, but without spraying pesticides or fungicides. The cultivars used in this study were Yuxidabaidou for faba bean and Yunmai 42 for wheat. Wheat and faba bean were sown on 6 November 2014 and harvested on 20 April 2015.

Table 1 The physical and chemical properties of the rhizosphere soils of intercropped faba bean/wheat and sole crops
Full size table

2.2 Plant and soil samples collection

Plant and soil samples were collected at harvest. Four types of soil sample collected from the rhizosphere were monocropped wheat (MW), intercropped wheat (IW), monocropped faba bean (MF), and intercropped faba bean (IF). In the sole crops, plant samples were randomly collected, while in the intercrops, wheat or faba bean plants were sampled from the rows adjacent to the other crop. Rhizospheres of 10 plants were carefully excavated in each plot (three replicates and three treatments), by keeping as much of their associated roots as possible (sample taken 20-cm depth around plants). After the soil loosely adhering to the plant roots was shaken off vigorously, the tightly adhering soil (rhizosphere soil) from each of the ten plants was removed with a paint brush and mixed to be a composite sample for each plot. Then, the soil samples were immediately stored in a refrigerator at 4 °C prior to sieving. Soil samples were homogenously sieved through 2-mm pores after root pieces had been removed. Each soil sample was divided into two subsamples. One sub-sample was air-dried for soil physical and chemical property analysis, and another sub-sample was used for the incubation experiment. The soil physical and chemical properties are presented in Table 1. Plant samples were dried in an oven at 65 °C for 72 h. The dry matter, aboveground biomass, and grain yields of wheat and faba bean were measured.

2.3 Laboratorial 15N tracing incubation study

The 15N incubation study was carried out in agreement with Müller et al. (2004, 2007). The incubation study consisted of an ammonium labeled (15NH4NO3, 20% atom% 15N excess) and a nitrate labeled (NH415NO3, 20% atom% 15N excess) treatment in triplicate. Twenty grams of fresh soil (oven-dried basis) was weighed out and transferred into a series of 250-ml conical flasks. A 1-ml portion of 15NH4NO3 or NH415NO3 solution was added to each flask to reach the concentration of 20 mg NH4+–N kg−1 soil and 20 mg NO3−–N kg−1 soil. The incubated soil was maintained at 60% water-holding capacity at 25 °C for a 72-h period. The flasks were aerated by removing the silicone rubber stoppers for 1 h every 2 days. The determination of NH4+–N and NO3−–N concentrations and the respective 15N enrichment was conducted at 0.5, 24, 48, and 72 h for each 15N label treatment in triplicate.

2.4 Soil chemical properties

The soil properties (i.e., pH (soil/water ratio, 1:2.5), soil organic carbon (SOC), and total N) were analyzed in accordance with the handbook of soil agro-chemical analysis procedures (Lu 2000). NH4+–N and NO3−–N concentration was determined by a continuous-flow analyzer (Skalar, Breda, Netherlands) after 2 mol l−1 KCl extraction at the soil/KCl ratio of 1:5. Soil alkali-hydrolyzable N is one of the indicators of soil N-supplying capacity (Roberts et al. 2011). Soil alkali-hydrolyzable N content was quantified by the method of Roberts et al. (2011). Briefly, 5 g of soil was distilled with 2 mol L−1 NaOH for 5 h and then distilled with 10 mol L−1 NaOH for 7 min. Boric acid (40 g L−1) was used to absorb the liberated NH3 using the method of direct steam distillation. Soil alkali-hydrolyzable N content was quantified by conductometric titration.

An automated C/N analyzer coupled with isotope ratio mass spectrometer was employed to determine isotopic composition of the NH4+–N and NO3−–N in the incubated soil (Europa Scientific Integra, Crewe, UK). Respective 15NH4+–N and 15NO3−–N measurements were conducted by MgO and Devarda’s alloy distillation (Feast and Dennis 1996; Lu 2000; Zhang et al. 2012). Briefly, a portion of the extract passed through MgO steam distillation to obtain NH4+–N; the remaining extract in the flask was continuously distilled to acquire NO3−–N after Devarda’s alloy addition. The liberated NH3 was trapped in boric acid solution and was acidified and converted to (NH4)2SO4 using 0.02 mol l−1 H2SO4 solution. The soluble NH4+–N in H2SO4 solution was then dried at 65 °C for analysis of 15N abundance. The recovery of NH4+–N and NO3−–N in a standard solution (1 g NH4+–N l−1 and 1 g NO3−–N l−1) was also measured using the same method. NH4+–N (> 99%) and NO3−–N (> 96%) were recovered in the standard solution.

2.5 15N tracing model

The simultaneously occurring gross N transformations in soil were quantified in accordance with the Ntrace model described by Müller et al. (2007 Fig. 1). A description of Ntrace model is shown in the attached supporting information file. In the model, initial NH4+ and NO3− concentrations were estimated for time zero using back extrapolation of data at t = 0.5 and 24 h according to Müller et al. (2004). Adsorbed NH4+ on the surficial exchange sites (NH4ads) after N addition can be calculated by subtracting determined NH4+ from applied NH4+. The average N transformation rates over the incubation time were calculated according to parameters of kinetic equations, with units represented as milligrams nitrogen per kilogram soil per day.

Fig. 1
figure 1

15N tracing model used for data analysis based on Müller et al. (2007) (NH4+, ammonium; NH4+ads, adsorbed NH4+; Nlab, labile soil organic N; NO3−, nitrate; Nrec, recalcitrant soil organic N. For explanations of N transformations, see Table 3)

Full size image

2.6 Data calculation and statistical analyses

Net mineralization was estimated from the difference in Min N concentration between initial and end of the incubation period, while net nitrification rates were calculated as the change in the NO3−–N concentration divided by the incubation period. The ratio of autotrophic nitrification rate to mineralization rate (M = MNrec + MNlab) was expressed as nitrification capacity (NC), denoting the capability of the soil to convert available NH4+ to NO3−. The ratio of total NO3− conservation rate (NO3− immobilization + dissimilatory nitrate reduction to ammonium (DNRA)) to total nitrification rate was defined as the capacity of the soil to retain NO3− (NR).

The land equivalent ratio (LER) was applied for the estimation of system productivity. This compared the yield acquisition from intercropping with yields obtained by same sole crops (Chapagain and Riseman 2014). The LER for the two intercrop species in proportional replacement design was calculated according to Mead and Willey (1980):

$$ \mathrm{LER}=\frac{\begin{array}{l}\mathrm{Intercropped}\ \mathrm{wheat}\\ {}\mathrm{yield}\end{array}}{\mathrm{Monocrop}\ \mathrm{wheat}\ \mathrm{yield}}+\frac{\begin{array}{l}\mathrm{Intecropped}\ \mathrm{faba}\ \mathrm{bean}\ \mathrm{yield}\\ {}\mathrm{yield}\end{array}}{\mathrm{Monocrop}\ \mathrm{faba}\ \mathrm{bean}\ \mathrm{yield}} $$

The yield unit of sole and intercropped crops is presented as kilograms per hectare. An LER value greater than 1.0 means that the intercropping enhances crop growth and production while an LER value less than 1.0 shows that the intercropping reduces growth and yield.

Statistical analyses of soil characteristics and N transformation rates of monocropping and intercropping soils were performed using SPSS 17.0 software for Windows. Pearson correlation coefficient was analyzed to describe the relationship among selected variables (where p < 0.05 was considered to indicate statistical significance)

3 Results

3.1 Effects of wheat/faba bean intercropping on the yield and aboveground biomass

Under field conditions, grain yields of intercropped faba bean with wheat were 11.4 and 34.2% higher when compared with the corresponding sole crops, respectively (p < 0.05) (Table 2). There was a stronger and more positive effect of intercropping on wheat than on faba bean. The LER of yield and aboveground biomass were 1.15 and 1.21, implying the intercropping of wheat/faba bean promoted an overall improvement of performance in both crops.

Table 2 The grain yield and biomass (mean ± SD) and LER of the wheat and faba bean from different cultivation methods (kg/ha)
Full size table

3.2 Soil ammonium and nitrate concentrations

The dynamics of NH4+ and NO3− concentrations in all the rhizosphere soils from monocropped and intercropped treatments during the incubation period are presented in Fig. 2. Ammonium concentrations decreased (Fig. 2a), while NO3− concentrations gradually increased in all the soils over the incubation period (Fig. 2b), suggesting a net nitrification appearance in all the soils. For the intercropped wheat/faba bean soil, the NH4+ concentration reduced and NO3− concentration increased more rapidly than in the corresponding monocropped soil. The measured net nitrification rates of the intercropped wheat/faba bean soils were 8.2 and 16.5% higher than those determined in the corresponding monocropped soils. The measured net mineralization rates in the intercropped wheat/faba bean soils were greater than in the corresponding monocropped soils, with a relative increase ranging from 38.5 to 78.4% (Fig. 2b). Thus, both net mineralization and net nitrification rates of intercropping soil showed a significant rise though considerable inter-replicate variability was observed (see Table 3).

Fig. 2
figure 2

Mean (±SD) concentrations of the NH4+ pool (a) and NO3− pool (b) in the monocropped faba bean (MF), monocropped wheat (MW), intercropped faba bean (IF), and intercropped wheat (IW) soils during the incubation period

Full size image
Table 3 Gross N transformation rates and measured net nitrification and mineralization rates (mg N kg−1 day−1) in rhizosphere soils from intercropped and monocropped faba bean and wheat during 3-day incubation at 25 °C and 60% water-holding capacity (mean ± SD)
Full size table

3.3 15N enrichments in soil mineral-N

The 15N abundance of the labeled NH4+ pool declined gradually in all the soils during the incubation (Fig. 3a), indicating NH4+ at natural or low abundance diluted the 15N labeled NH4+ pool, as a result of mineralization of organic matter. The increased 15N enrichment in the NO3− pool over the incubation in the 15NH4+-labeled treatment showed active NH4+ oxidation to NO3− (Fig. 3b). The 15N enrichment of NH4+ in the 15NO3−-labeled treatment for all the soils was initially low then increased slightly during the incubation time (Fig. 3c). The reduction of 15N enrichment in the NO3− pool of 15NO3−-labeled treatments implied that produced NO3− at natural or low abundance entered the NO3− pool (Fig. 3d).

Fig. 3
figure 3

Changes in 15N abundance of the NH4+ and NO3− pools in the monocropped faba bean (MF), monocropped wheat (MW), intercropped faba bean (IF), and intercropped wheat (IW) soils when incubated at 25 °C and 60% WHC with 15NH4+ (a, b) and 15NO3− (c, d) as a tracer. Vertical bars are standard deviations of the mean (n = 6)

Full size image

3.4 Effects of wheat/faba bean intercropping on NH4 + dynamics

Total NH4+ supply rates (MNrec + MNlab + RNH4ads + DNO3) were 3.41, 4.92, 4.63, and 9.22 mg N kg−1 day−1 in the MF, IF, MW, and IW soils, respectively. The NH4+ supply rates in the intercropped faba bean/wheat soils were 1.5 and 2.0 times greater than in the corresponding monocropping soils. Gross mineralization rates (MNrec + MNlab) in the MF, IF, MW, and IW soils were 3.25, 4.63, 4.53, and 8.52 mg N kg−1 day−1, respectively, with the intercropped faba bean/wheat soils being 1.4 and 1.9 times greater than those in the corresponding monocropping soils, respectively (Table 3). The ratios of gross mineralization from labile organic N to total gross mineralization (MNlab/(MNrec + MNlab)) were 0.63, 0.53, 0.99, and 0.56 in the MF, IF, MW, and IW soils, respectively. This indicated that intercropping stimulated the mineralization rates, especially from the recalcitrant organic N pool, resulting in a higher total NH4+ supply in the wheat/faba bean intercropping system.

The total NH4+ immobilization was lower than the gross mineralization for all the treatments, leading to a net mineralization. However, the total NH4+ immobilization (INH4-Nrec + INH4-Nlab) in the IF (1.88 mg N kg−1 day−1) and IW soils (1.40 mg N kg−1 day−1) were 1.5 and 2.6 times greater than those in the MF (1.27 mg N kg−1 day−1) and MW (0.54 mg N kg−1 day−1) soils, respectively. The ratio INH4-Nrec/(INH4-Nrec + INH4-Nlab) of all the soils was above 0.97, suggesting that intercropping enhanced the immobilization of NH4+ in the soil as compared with the corresponding monocropping. Immobilized NH4+ (> 97%) was predominantly transferred to the recalcitrant organic N pool irrespective of crop cultivation methods (Table 3). Gross NH4+ immobilization was positively correlated with the soil organic C concentration (r = 0.64, p < 0.05) and C/N ratio (r = 0.48, p < 0.05).

3.5 Effects of wheat/faba bean intercropping on NO3 − dynamics

Heterotrophic nitrification rates (organic N oxidized to nitrate) in all the soils were negligible. Thus, it can be deduced that NO3− production was mostly associated with autotrophic nitrification (NH4+ oxidation) (Table 3). The total NO3− production rates (ONH4 + ONrec) of MF, IF, MW, and IW soils were 9.57, 11.25, 11.56, and 12.89 mg N kg−1 day−1, respectively. High NO3− production rates were observed in the intercropped faba bean/wheat soils, being 1.2 and 1.1 times greater than those in the corresponding monocropping soils, respectively. The gross nitrification rate showed a significantly positive correlation with alkali-hydrolyzable N concentration in all soils (r = 0.731, p < 0.01). There was also a significant positive relationship between the gross nitrification and the gross mineralization (p < 0.05) (Fig. 4).

Fig. 4
figure 4

Relationship between gross nitrification and N mineralization

Full size image

Immobilization of NO3− and DNRA were two important NO3− consumption pathways, but DNRA rates were significantly greater than the nitrate immobilization rates in all the soils (p < 0.05) (Table 3), accounting for 1–4% of the gross nitrification. The DNRA rates of the IF and IW soils were notably greater than those in the corresponding monocropping soils (p < 0.05). A significant positive relationship between DNRA rates and the NH4+ absorption rates was also found (r = 0.592, p < 0.05). The ratio of nitrification rate to total mineral N immobilization (immobilized NH4+ and NO3−) (N/I) was 7.5, 5.9, 21.3, and 9.2 for the MF, IF, MW, and IW soils, respectively, indicating that intercropping decreased the N/I ratio, particularly in IW soil (p < 0.05).

4 Discussion

4.1 Effects of wheat/faba bean intercropping on mineral N supply

The gross net N mineralization rates and the NH4+ supply rates were greater in the two intercropped soils than in the corresponding monocropped soils (p < 0.05, Fig. 5), resulting in higher NH4+ availability. This is in agreement with previous studies that gross N mineralization rates were higher in the legume-cereal intercropping in comparison to their respective monocropping (Regehr et al. 2015). Therefore, intercropping apparently stimulated the availability of N for crop uptake, which was indicated by the significant correlation between both gross and net N mineralization and gross NH4+ supply rates with the crop yield and aboveground biomass, respectively (p < 0.01, Table 4). Similar to our study, a few recent 15N tracing experiments have also found a positive relationship between the soil mineral N availability (via mineralization) and the crop N uptake as well as yield (Zhang et al. 2012; Wang et al. 2015). Previous studies demonstrated that intercropping enhanced total soil organic C, labile organic C that included particulate organic matter, and the carbohydrate concentration due to diverse root exudates and rhizodeposition in an interspecies interaction system (Zhang et al. 2010), suggesting that intercropping influenced soil N mineralization. This is probably the result of alternation of soil organic matter composition and associated microbial communities and activities.

Fig. 5
figure 5

The gross, net N mineralization rates, and the NH4+ immobilization rates in rhizosphere soils of intercropping faba bean/wheat and corresponding monocrops over 3-day incubation at 25 °C and 60% water-holding capacity. Error bars represent standard deviations (n = 3). Different lowercase letters indicate significant differences at p < 0.05 level among treatments for a certain N transformation rate

Full size image
Table 4 The relationship between different N transformation rates, grain yield and aboveground biomass (kg/ha)
Full size table

N mineralization was dominated by labile organic N mineralization, accounting for about 62.5 and 99.4% of the total gross N mineralization in the faba bean or wheat sole-cropping soil, respectively. Intercropping enhanced the N mineralization from the recalcitrant N (MNrec) in the soil. The ratio of MNrec to the total gross N mineralization increased by 1.1 and 75.7 times in the IF and IW soils compared with the corresponding MF and MW soils, respectively, indicating that intercropping activated the recalcitrant N mineralization. This could be due to a priming effect of stimulated microbial activity that enhanced the mineralization of recalcitrant soil organic matter (SOM), particularly in intercropping soil that contained more organic C relative to monocropping soil (Zhang et al. 2010, 2012).

On average, both the gross and net nitrification rates were increased by 17 and 10% in the intercropped faba bean/wheat soils compared with the respective monocropping soils (Fig. 6), denoting the capacity of NO3− supplying in the intercropping system was higher than that in the monocropping system. Intercropping stimulated the N mineralization and autotrophic nitrification processes. Thereby, higher inorganic N supply due to increased N mineralization and nitrification could contribute to the higher crop available N in these intercropping systems compared with the monocropping system and it likely to be a main mechanism explaining the yield increase.

Fig. 6
figure 6

The gross, net nitrification rates, and the gross NO3− consumption rates in rhizosphere soils of intercropping faba bean/wheat and corresponding monocrops during 3-day incubation at 25 °C and 60% water-holding capacity. Error bars represent standard deviations (n = 3). Different lowercase letters indicate significant differences at p < 0.05 level among treatments for a certain N transformation rate

Full size image

Main driving factors influencing soil N mineralization include soil C and N content as well as the C/N ratio (Barrett and Burke 2000; Wang et al. 2001; Mack and D’ Antonio 2003). However, the relationship between gross mineralization rates and soil properties was insignificant in this study (e.g., soil organic C and N concentrations, C/N ratio, and pH, data not shown). The crop species (e.g., faba bean and wheat) and interspecies interaction may affect the N mineralization dynamics as the quantity and decomposition rate of soil organic matter are associated with plant species and root exudates (Zhang et al. 2011; Sotta et al. 2008). The underlying mechanisms of relatively faster gross rates of organic N mineralization in intercropping soils than monocropping soils need to have further investigation.

4.2 Effects of wheat/faba bean intercropping on mineral N conservation

Nitrogen losses are enhanced if mineral N supply is not matched by N consumption (i.e., immobilization and plant N uptake). Intercropping soils have a strong capacity to preserve inorganic N through N immobilization in soil (Regehr et al. 2015; Vervaet et al. 2004), and this was confirmed by our study. The NO3− conservation capacity (gross NO3− immobilization rate + DNRA rate, CNO3), the NH4+ conservation capacity (gross NH4+ immobilization + ANH4, CNH4) and the total inorganic N conservation capacity (CNO3 + CNH4, CN) were significantly greater in the intercropped faba bean/wheat soils as compared with the respective monocropping soils (p < 0.05, Fig. 7). The values of NR (the ratio of total NO3− conservation capacity to total nitrification rate) for intercropped faba bean/wheat soils were 1.5 and 3.7 times higher than those in the monocropping soils, respectively, suggesting that the intercropping soils had higher NO3− retention capacity.

Fig. 7
figure 7

The NO3− conservation capacity (CNO3), the NH4+ conservation capacity (CNH4) and the total inorganic N conservation capacity (CN) in rhizosphere soils of intercropping faba bean/wheat and corresponding monocrops during 3-day incubation at 25 °C and 60% water-holding capacity. Error bars represent standard deviations (n = 3). Different lowercase letters indicate significant differences at p < 0.05 level among treatments for a certain N transformation rate

Full size image

Intercropping can significantly modify rhizosphere processes (Zhang et al. 2004; Ehrmann and Ritz 2014). Rhizodeposition of intercropped crops transfers and reutilizes organic carbon and other nutrients among plants and microbial communities in the soil so that more N may become available in the rhizosphere of intercropped species. Thus, it facilitates crop N uptake and enhances productivity (Ehrmann and Ritz 2014). Most of the abovementioned indicators of inorganic N conservation capacity in this study were significantly correlated with the grain yield and aboveground biomass (p < 0.05, Table 4), implying that intercropping enhanced the capacity of conserving mineral N in the rhizosphere soils. On the one hand, fixed atmospheric N from faba bean root nodules was rapidly mineralized to ammonium-N and then oxidized to nitrate that was transferred by microorganisms or via roots mycorrhizal connections to rhizosphere of wheat to increase N availability. On the other hand, part of the N released from faba bean was incorporated into soil organic matter (organic N pool). Hence, inorganic N was more effectively retained in the intercropping soils through transforming to organic N (immobilization) and NH4+ pools (through DNRA process); NH4+ and NO3− originated from mineralization and nitrification can be effectively preserved to prevent denitrification, leaching or runoff in the intercropping system (Ehrmann and Ritz 2014; Regehr et al. 2015; Wang et al. 2015). Therefore, enhancing inorganic N conservation capacity is an important strategy in intercropped soils to maintain higher soil fertility and yield.

The changes of gross mineral N immobilization under different cultivation methods may be attributable to the changes of the concentration and composition of soil organic C as well as the soil C/N ratio caused by intercropping. This is in line with some previous studies. These studies reported that soil microbes demanded more inorganic N and restrained N mineralization when they decomposed organic matter with high C/N ratio. Moreover, SOC, soil fertility, and fungal communities had a positive relationship (Sollins et al. 1984; Janssen 1996). Our results showed that the NH4+ immobilization rates were significantly related to the soil organic C and C/N ratio (p < 0.05), inferring that soil organic C content and its composition could be critical in influencing inorganic N immobilization (Burger and Jackson 2003; Häbteselassie et al. 2006). In intercropping systems, SOC inputs are more spatially heterogeneous, and more SOC can move, deposit, and cycle in the soil than in monocropped soil (Regehr et al. 2015). Consequently, the significant increase of rhizospheric organic C content and C/N ratio in the intercropped faba bean/wheat soils required more inorganic N supply for the substrates decomposition, which can also explain the increased inorganic N immobilization rate in the intercropping system (Johnson 1992; Stark and Hart 1997; Burger and Jackson 2003). In addition, the greater C availability in the intercropping system apparently accommodated more abundant, diverse, and active microbial communities with greater N demand (Ehrmann and Ritz 2014; Regehr et al. 2015), thus promoting ammonium N immobilization and recycling of NO3− in the soil.

Our results highlighted that competent N retention mechanisms were functional to reduce N losses in the intercropping system through effective assimilation by soil microbes. But further research is needed to understand the underlying mechanism of microbial activities for the shift of gross N transformation rates between monocropping and intercropping system.

The rates of immobilization of NO3− into organic N were negligible in intercropped faba bean/wheat soil, which was consistent with the results of a previous study (Zhang et al. 2013). Relative to NO3− immobilization, however, the DNRA process was more vital for conserving NO3− in the intercropping system, comprising 1.6%, 2.4%, 1.0%, and 3.7% of the total produced NO3− rate in the MF, IF, MW, and IW soils, respectively. The significant positive correlation between DNRA and NH4+ absorption may further increase NO3− retention and reduce the potential nitrogen losses through NO3− leaching to groundwater or denitrification in the intercropping soils.

4.3 Environmental significance of wheat/faba bean intercropping

Intercropping significantly decreased the nitrification capacity (the ratio of autotrophic nitrification rate to mineralization rate, NC) by 29.1 and 75.2% for intercropped faba bean/wheat soils compared with the corresponding monocropping soils, respectively (p < 0.05), indicating intercropping could decrease the risk of NO3− leaching, runoff, and denitrification potential. Microbial nitrification and immobilization of NH4+ are two NH4+ consumption pathways; and the ratio of microbial nitrification to NH4+ immobilization rate (N/Ia) can be applied to evaluate the comparative importance of the two consumption processes (Hoyle et al. 2006; Lan et al. 2014). Intercropping decreased the value of N/Ia comparing with the respective sole cropping (Fig. 8), showing that NH4+ would be efficiently conserved in the intercropping system by decreasing the nitrification and its consequent risk of N losses. It has been reported that ammonia volatilization rate had a positive correlation with soil pH (Duan and Xiao 2000). Therefore, NH3 volatilization was not a main N loss pathway under this neutral pH soil condition (pH = 6.5–6.8) for both intercropping and monocropping soils. The adverse effect of NO3− leaching from monocropping systems can possibly be increased, because of the augmentation of nitrification capacity or restriction of NO3− immobilization. The result of this study was in accordance with a former study (Pappa et al. 2011). Accordingly, the environmental degradation caused by NO3− leaching and runoff from soil can be alleviated by converting monocropping to intercropping.

Fig. 8
figure 8

The ratio of gross nitrification to gross ammonium immobilization rate (N/Ia) in rhizosphere soils of intercropping faba bean/wheat and corresponding monocrops during 3-day incubation at 25 °C and 60% water-holding capacity. Error bars represent standard deviations (n = 3). Different lowercase letters indicate significant differences at p < 0.05 level among treatments

Full size image

The interaction of individual N transformation processes controlled soil nitrogen production and preservation. The intercropping system has been demonstrated to be a more suitable land use option in comparison with monocropping systems (Pappa et al. 2012; Regehr et al. 2015), since it is conducive to reduce nitrification and NO3−–N leaching via N immobilization.

5 Conclusions

Our findings demonstrated that wheat/faba bean intercropping was able to significantly increase the productivity of both companion crops under subtropical condition. This intercropping practice could significantly affect N transformation traits that accounted for the enhanced supply and conservation of inorganic N. The combination of greater gross N mineralization, gross nitrification, and inorganic N immobilization promoted a rapid N turnover, leading to improved N availability and N retention capacity in the intercropping system. Therefore, intercropping can promote a higher inorganic N supply and conservation capacity in soil and lower the risk of NO3− leaching to groundwater. These findings illustrate that intercropping improved the N cycling through enhanced mineralization and immobilization rates to promote crops productivity while at the same time N was conserved in the system. In conclusion, wheat/faba bean intercropping can serve as a more promising and efficient system in comparison to their respective monocropping.