Abstract
Intercropping, the cultivation of two or more crop species on the same area of land may improve yield, forage quality or soil health. The objective of the present study was to study the yield, growth rate, interspecific competition in terms of dry matter and N yield, quality and financial outcome of the intercrops of faba bean (Vicia faba L. var. equina) with barley (Hordeum vulgare L.) in different spatial arrangement (i.e. 1:1, 2:2, 2:1 alternate rows, and mixed in the same row). Growth rate of faba bean was lower in the intercrops than in the monocrop, due to the strong competitive ability of barley. Faba bean-barley intercrop (FB:B) 2:1 produced the highest biomass. The land equivalent ratio (LER), relative crowding coefficient (K), actual yield loss (AYL) and system productivity index (SPI) values were greater for the FB:B 2:1 indicating an advantage of intercropping in terms of dry matter and N yield. The partial K, aggressivity (Α) and partial AYL values indicated barley as the dominant species in the intercrops for dry matter and N yield. The highest monetary advantage index (MAI) and intercropping advantage (IA) values was also recorded for the FB:B 2:1 intercrop indicates that this intercrop had the highest economic advantage. Based on biomass production and the competition indices for dry matter and N yield, the intercropping of FB:B 2:1 intercrop was more advantageous than faba bean or barley monocrops and the other intercropping treatments that were tested.
Keywords: competition, financial yield, growth rate, indices, protein content.
Introduction
In recent years there are several environmental challenges that are attributed to modern agriculture such as reduce soil, water, and air quality. Modern farmers especially in developed countries use chemically intensive practices to maintain soil productivity, together with the other management practices that decrease soil organic matter and increase soil erosion, acidification, and salinization (Dumanski et al., 1986). In addition, sustainable management of plant nutrients such as nitrogen (N) is challenging due to the increase in cost of N fertilizers coupled with N losses through ammonia volatilization, nitrous oxide emissions which reduce the air quality and nitrate leaching which contribute to ground water contamination (Fageria and Baligar, 2005). Therefore, it is important to develop sustainable and ecologically sound nutrient management practices that can be applied to large farms. For most ecosystems one strategy that can address many of these concerns is the inclusion of grain legumes with cereal crops a practice that is known as intercropping (Anil et al., 1998; Lithourgidis et al., 2011a; Brooker et al., 2015; Bedoussac et al., 2015).
There many studies that showed that intercropping of two or more crops cannot lead necessarily to higher forage biomass yield compared with the monocrops (Mead and Willey, 1980; Midya et al., 2005). However, higher LER values have been reported in many studies but not for all intercropping systems (Lithourgidis et al., 2011a; Dordas and Lithourgidis, 2011) indicating that intercropping can lead to higher yield compared with monocrops.
The use of grain legumes in intercropping systems can provide a more sustainable source of N to cropping systems through biological N2 fixation (Crews and Peoples, 2004; Brooker et al., 2015; Bedoussac et al., 2015). Intercropping legumes with cereals can also minimize N losses commonly associated with legume monocrops through uptake of soil inorganic N by the cereal and slower N mineralization during decomposition due to higher cereal C/N ratios (Hauggaard-Nielsen et al., 2003). A number of grain legumes were used in intercropping systems such as common vetch (Vicia sativa subsp. sativa L.), peas (Pisum sativum subsp. sativum var. arvense L.) etc (Anil et al., 1998; Brooker et al., 2015; Bedoussac et al., 2015). Faba bean has not been extensively used in intercropping systems (Dordas and Lithourgidis 2011). Faba bean is a cool-season, N2-fixing, drought-tolerant, annual legume grown for forage and grain but is seldomly grown mainly due to its late maturity, especially in intercropping systems (Jensen et al., 2010; Lithourgidis and Dordas 2010). Faba bean is an excellent candidate crop to provide protein and starch for animal feed in many areas of the world. Moreover, this crop can tolerate harsh growth conditions and can be used to marginal agricultural lands and also to low-input agricultural systems (Jensen et al., 2010; Koepke and Nemecek, 2010).
The yield of an intercropping system depends on several factors including the seeding density, the crop species that are used and their genotypes, the management practices and also by the spatial arrangements (intercropping pattern) of the individual crops (Herbert et al., 1984; Hauggaard-Nielson et al., 2001; Esmaeli et al., 2011; Brooker et al., 2015; Bedoussac et al., 2015). Mixed cropping and 1:1 row arrangement (one row of each crop species) are common intercropping systems (Agegnehu et al., 2006). However, there are limited reports about the effect of various spatial arrangements on productivity of intercropped grain legumes with cereals which can be increased under different spatial arrangements which can affect the competitive ability of the crop species (Banik et al., 2006; Esmaeili et al., 2011).
Intercropping has a particular advantage since the different species that are used do not compete for the same resource niche and they use the environmental resources in a complimentary way (Hauggaard-Nielsen et al., 2001; Bedoussac and Justes, 2011). The advantage of cereal-legume intercrops comes from the complementary use of N sources by the components of the intercrops (Bedoussac and Justes, 2010; Bedoussac et al., 2015). Intercropping is quite important for the low-input cropping systems where N is a limited resource. Under low soil N level legumes have higher interspecific competitive ability compared with the cereal component which have higher competitive ability under high N level (Bedoussac and Justes, 2011; Hauggaard-Nielsen and Jensen, 2001; Monti et al., 2016). Intercropping systems of cereal-legumes can allow the optimal use of soil and atmospheric N sources and at the same time maintains high yield and quality especially under low soil N levels and minimize potential environmental impacts which occur under intensive agricultural systems (Pelzer et al., 2012). The soil N pool is maintained because of the use of legumes in the intercropping systems (Hauggaard-Nielsen et al., 2001; Bedoussac and Justes, 2011). In addition, the part of N derived from fixation by legume can be further increased in intercrops as the cereals are more competitive for N that they take up N from the soil and forces the legume to take up N from N2 fixation due to low NO3 concentrations as symbiotic N2 fixation is negatively affected by that chemical fertilizers (Hauggaard-Nielsen et al., 2009).
There are many indices that are used in previous studies to determine the advantages of intercropping systems compared with the monocrops (Bedoussac and Justes, 2011; Dhima et al., 2007; Weigelt and Jolliffe, 2003). However, most of these are based on dry matter and the use of the same indices based on N yield was not explored to determine the effects of intercropping in N utilization efficiency and also in the use of N yield to determine the competition between the different species in intercropping systems.
The objectives of the present work were: (i) to evaluate faba bean and barley in two monocropping systems and four intercropping systems for dry matter and N yield and protein content, (ii) to study the effect of different intercropping systems on the growth rate of the two species, (iii) to assess the effect of competition among the component species of the different intercropping systems using the dry matter yield and N yield, and (iv) to assess the economic advantage of each intercropping treatment.
Materials and methods
Study site and crops management
A field experiment over two growing seasons (2009-2010 and 2010-2011) was established at the University Farm of Thessaloniki in northern Greece (22o59’ E, 40o32’ N). The experiment was established in a loam (L) soil (Typic Xerorthent) with pH of 7.8, organic matter content of 14 g kg-1, N-NO3 40 mg kg-1, P (Olsen) 29 mg kg-1, and K 180 mg kg-1 (0 to 30 cm depth). Seedbed preparation included moldboard plough, disk harrow, and cultivator. Nitrogen and P2O5 at 80 and 40 kg ha-1, respectively, were incorporated into the soil as diammonium phosphate (20-10-0) before sowing. The same field was used in both years but different plots were used each year. The field was uniform in terms of cultivation history and cropping systems that were used. There were no pesticides that were used and also no irrigation during both growing periods. Climatic data during the two growing seasons of the experimentation are given in Figure 1.
Faba bean (cv. Polycarpe) and barley (cv. Thessaloniki) as well as intercrops of faba bean with barley in one seeding ratio (50:50) based on seed weight, under four different spatial arrangement, were sown around the end of November in both growing seasons. The seeding rates that were used were for faba bean and barley monocrop 120 and 160 kg ha-1, respectively, whereas the seeding rates for each intercrop were 60 and 80 kg ha-1 of faba bean and barley respectively. The treatments were the following:
1. Faba bean monocrop
2. Barley monocrop
3. Faba bean – Barley 1:1 alternate rows (FB:B 1:1)
4. Faba bean – Barley 2:2 alternate rows (FB:B 2:2)
5. Faba bean – Barley 2:1 alternate rows (FB:B 2:1)
6. Faba bean – Barley mixed in the same row (FB:B mixed)
The corresponding seeding rate of monocrops was 34 and 432 seeds per m2 for faba bean and barley respectively. The corresponding seeding rate for intercrops was 17 and 216 seeds per m2 of faba bean and barley respectively.
The row spacing was 25 cm and seeds of both species in each intercrop were sown simultaneously. The experimental design was a randomized complete block with six treatments (two monocrops and four intercrops of faba bean with barley) replicated four times. Plot size was 5 m x 2 m and plots were separated by a 2.5 m buffer zone. All crops were kept free of weeds by implementing hand hoeing, where necessary. The nodulation was verified by examining the root system of the faba bean plants and observing nodules that were formed during the growing period.
Chlorophyll and plant height
Chlorophyll content readings (SPAD units) were taken with a hand-held dual-wavelength meter (SPAD 502, Chlorophyll meter, Minolta Camera Co., Ltd., Japan). For each plot 20 young fully expanded leaves were used at 6 weeks after tillering (anthesis). The instrument stored and automatically averaged these readings to generate one reading per plot. Furthermore, plant height was determined by measuring the height of 10 randomly selected plants per plot and per species from the soil to the top of the plant and getting an average value for each plot. Plant height was determined at 3 weeks after tillering and at the end of barley anthesis (about 6 weeks after tillering).
Growth rate, dry matter, and N yield and N Utilization Efficiency
For the determination of growth rate of the different species used in the intercropping treatments, faba bean and barley were sampled four times at 0, 3, 6, and 9 weeks after tillering (WAT). During these sampling barley was at tillering, jointing, booting and milk stage respectively. Also faba bean at the same periods was at stem elongation, first flower open, development of fruit and ripening of fruit respectively. The area that was harvested was 0.5 m2 of each plot. More specifically, plants were cut to the ground level manually and the different species were separated by hand to determine fresh weight of each species in each plot. Dry matter yield was determined by harvesting an area of 2 m2 from each plot at pod-setting stage of faba bean (which was at milk stage of barley) about mid-May of each growing season. The samples were dried at 70 oC to constant weight to determine the relative water content.
Forage quality at harvest was determined by measuring the total N content with the Kjeldahl method and CP was calculated by multiplying the N content by 6.25. N utilization efficiency (NUtE) for biomass accumulation was calculated according to López-Bellido and López Bellido, 2001 using the following formula NUtE = DM/N, where DM is the dry matter at harvest and N is the total N that was taken up by the crop both based on the same area of land (m2).
Competition indices for dry matter and N yield
One of the commonly used index is the land equivalent ratio (LER) and is used as the criterion for intercropping advantage as both faba bean and barley were desired species in the intercrops. In addition, LER indicates the efficiency of intercropping system for using the resources of the environment compared with monocropping. The value of unity is considered the critical value for this index. Therefore, when LER is greater than one the intercropping favours the growth and yield of the intercropped species, whereas when LER is lower than one the intercropping negatively affects the growth and yield of the species (Dhima et al., 2007). The LER was calculated as:
LER = (LERfb + LERb), LERfb = (Yfbi/Yfb), LERb = (Ybi/Yb),
where Yfb and Yb are the yields of faba bean and barley, respectively, as monocrops and Yfbi and Ybi are the yields of faba bean and barley, respectively, as intercrops.
Also the land equivalent ratio for nitrogen yield (LERN) was determined modified according to Mead and Wiley (1980) which indicates a possible N yield advantage of intercrops, as follows:
LERN =NYbi/NYb +NYfbi/NYfb
where NYb and NYfb are the crop N yields for barley (b) and faba bean (fb) grown in monocrops and NYbi and NYfbi are the yields of the crops grown in intercrops. A LERN>1 shows a N yield advantage of the intercropping system whereas a LERN<1 indicates a N yield disadvantage. The LERN is the sum of the partial LERN of the individual crops in the intercrop. The partial LERN indicates the relative competitive ability of individual crops regarding N yields in intercrops.
The relative crowding coefficient (K) for dry matter and N yield is used to measure the relative dominance of one species over the other in an intercrop (Ghosh, 2004). The K was calculated as:
K = Kfb · Kb, Kfb = YfbiZbi/(Yfb-Yfbi)Zfbi, Kb = YbiZfbi/(Yb-Ybi)Zbi,
where Zfbi is the sown proportion of faba bean in intercrop and Zbi the sown proportion of barley in intercrop. When K is greater than one there is a yield advantage, when K is equal to one there is no yield advantage, and when it is less than one there is a disadvantage in efficient resource use resulting in relative yield loss. Similar to LER, K was calculated on N yield basis by replacing dry matter yield by N yield.
Aggressivity (A) is often used to indicate how much the relative yield increase in ‘a’ crop is greater than that of ‘b’ crop in intercropping (Dhima et al., 2007). The aggressivity is derived from the equation:
Afb = (Yfbi/YfbZfbi) – (Ybi/YbZbi), Ab = (Ybi/YbZbi) – (Yfbi/YfbZfbi),
if Ab = 0, both crops are equally competitive, if Ab is positive then barley species is dominant, if Ab is negative then barley species is the dominated species. Similar to LER, A was calculated on N yield basis by replacing dry matter yield by N yield.
System productivity index (SPI) is an index used for assessing intercrops as it standardizes the yield of the secondary crop (barley) in terms of the primary crop (faba bean) (Agegnehu et al., 2006) and is calculated as:
SPI = [(Yb/Yfb) · Yfbi] + Ybi
where Yb and Yfb are the mean yield of barley and faba bean in monocrop and Ybi and Yfbi are the mean yield of barley and faba bean in mixed culture. In addition, SPI was calculated on N yield basis.
Moreover, Banik et al. (2000) reported that the actual yield loss (AYL) index, gave more precise information than other indices used about the competition between and within component crops and the behaviour of each species in intercropping. AYL is calculated with the following formula (Banik et al., 2000):
AYL = AYLfb + AYLb,
AYLfb = {[(Yfbi/Zfbi)/(Yfb/Zfb)] – 1}, AYLb = {[(Ybi/Zbi)/(Yb/Zb)] – 1}
AYL can have positive or negative values indicating an advantage or disadvantage accrued in intercrops when the main objective is to compare yield on a per plant basis. Similar to LER and the other indices, AYL was calculated on N yield basis by replacing dry matter yield by N yield.
Economic indices
None of the abovementioned competition indices provides any information about the economic advantage of an intercropping system. For this reason, the monetary advantage index (MAI) and the intercropping advantage (IA) index were calculated according to Banik et al. (2000) and Ghosh (2004) as follows:
MAI = (value of combined intercrops) x (LER-1)/LER (15)
IA = IAb + IAfb, IAfb = AYLfb · Pfb, IAb = AYLb · Pb
Pfb is the commercial value of faba bean silage (the current price is €44 per Mg), and Pb is the commercial value of barley silage (the current price is €32.5 per Mg). Value of combined intercrops was calculated as: (Yfbi · Pfb) + (Ybi · Pb).
Statistical analyses
The data (dry matter and N yield, plant height, chlorophyll, crude protein concentration and the indices) were analyzed by the ANOVA method according to a 2×7 factorial approach (growing season treatments) with a split plot arrangement with four replications per treatment combination. The growing seasons were considered as the main plots and the seven treatments the split plots (Steel et al. 1997). A combined analysis over growing season was carried out. Least significant difference post hoc procedure was used for testing the differences between treatment means averaged over the two growing seasons, since no statistically significant interaction between growing seasons and treatments was detected.
For the determination of the growth rate dry matter was measured at four different dates (0, 3, 6, and 9 WAT). Dry matter yield changes in cereals and faba bean were evaluated by regressing dry matter yield of each plant species against sampling time. Linear, quadratic, hyperbolic, and logarithmic equations were tested for their suitability to describe the relationship between dry matter response and time. The equation with the highest value of the adjusted coefficient of determination ( ) and the smallest standard error of estimate was selected as the most appropriate (Hair et al., 1995). In these regression equations, dry matter (t ha-1) was the dependent variable (y) and dates (weeks after tillering, WAT) the independent variable (x). All regression analyses were performed using 8 pairs of (x, y) values (four sampling date for each of the two growing seasons and the mean value of each sampling data over the two growing season are presented in the graphs). The significance level of all hypotheses testing procedures was preset at P<0.05. SPSS (1998, version 17) software was used for the analysis of variance (ANOVA).
Results
The weather conditions during the 2009 and 2010 growing seasons were quite similar as the rainfall and the average temperature were similar. Rainfall was 321 mm for 2009 and 306 mm for 2010 growing seasons (Fig. 1). In addition, as the ANOVA for the dry matter yield and the other characteristics showed that there was no treatment by growing season interaction and the treatment means averaged across growing seasons are presented.
Growth rate
The dry matter yield of barley and faba bean was increased from 0 up to 6 WAT and then declined in most treatments. The r2 comparisons among the models tested showed that the quadratic equation (y = a + bx – cx2) had the best fit for dry matter yield over time (Fig. 2). Τhe initial growth of the two crop species was different in monocrops (α value of 2.441 and 1.054 for barley and faba bean respectively), and the growth rate of barley and faba bean in monocrops (b values 1.095 and 1.483 respectively). In the case of barley there was no significant difference in initial growth in most intercropping treatments with the exception of the FB:B 2:1 treatment which showed lower initial growth (1.709). In the case of faba bean the initial growth was higher in the monocrop compared with the intercropping systems. In addition, faba bean showed higher initial growth at the FB:B 1:1, FB:B 2:1, and FB:B 2:2 treatments than at the FB:B mixed intercropping system. The faba bean showed greater growth rate when it grew as monocrop than when it grew in intercrops (Fig. 2).
Dry matter, N yield and crude protein
Monocrop of barley gave higher dry matter yield compared with faba bean monocrop (Table 1). Also, in most cases, dry matter yield was affected significantly in the different intercropping treatments. The greatest dry matter yield was obtained from FB:B 2:1 intercrop (6.06 Mg ha-1) followed by barley monocrop (5.95 Mg ha-1). In particular, intercrop of FB:B 2:1 produced on average about 13, 23, and 39% more dry matter yield than FB:B 2:2, 1:1 and FB:B mixed intercrop, respectively (Table 1). Faba bean proportion was affected by the intercropping treatment, whereas was higher at the FB:B 2:1 intercropping treatment.
Nitrogen yield which was measured as N accumulated in shoots followed a different trend in the case of intercrops and monocrops as it was higher in the faba bean monocrop and lower in barley monocrop and the FB:B mixed intercropping system (Table 2). Among the intercropping treatments the highest N accumulated yield was found in the case of FB:B 2:1 followed by the FB:B 2:2 treatment. The faba bean proportion was lower when measured as dry matter compared with the N yield, because N accumulated by faba bean was higher compared with the dry matter due to higher N concentration of faba bean. The range that faba bean proportion increased as N yield and was from 18% up to 52% compared with the dry matter proportion that was in the range of 9-34% (Tables 1, 2). NUtE was the lowest in the FB monocrop and the highest at the barley monocrop, and it was showed significant differences among the intercropping systems (Table 2).
Crude protein concentration and protein yield (CP) was the highest in the faba bean monocrop (198.1 g kg-1 of DM and 992.6 kg ha-1, respectively) (Table 2). On the other hand, barley had the lowest CP concentration (69.8 g kg-1 of DM). In addition, there was an increase in the CP concentration in the intercropping treatments and especially in FB:B 2:2 and FB:B 2:1. As regards the protein yield per area basis (CP yield), the highest protein yield for intercropping treatments was found in FB:B 2:1 (736 kg ha-1) followed by the FB:B 2:2 (595 kg ha-1) (Table 2).
Plant height and chlorophyll determination
Plant height of faba bean and barley did not show significant differences in the different intercropping treatments compared with the monocrop (Table 3). In contrast, chlorophyll content showed difference between faba bean intercrops and monocrop as it was higher in the monocrop. However, the trend was different in barley where there were no significant differences between the different treatments (Table 3).
Competition indices
The LER value for FB:B intercrops was greater than one only in the case of FB:B 2:1 (Table 4). In this case, total LER was significantly higher than 1.00. In all cases, partial LER of faba bean was lower than 0.5 in all intercropping treatments. Also in all treatments partial LER of barley was above 0.5, and there was no difference among the different treatments.
When LER was calculated on the basis of N yield the trend was similar with the dry matter and the highest value found in the FB:B 2:1 followed by the FB:B 2:2, the lowest value was found at the FB:B mixed intercropping system (Table 5). The LERN was higher compared with the LER based on the dry matter. Partial LERN for faba bean was in the same range as in dry matter but in the case of barley there was difference as LERN was higher compared with the dry matter.
Relative crowding coefficient values followed a similar trend with the LER values in the case of faba bean. In particular, the K values were below one in the case of faba bean in all treatments (Table 4). However, in barley the K value was above one, whereas the highest K total value was found in the case of FB:B 2:1. The situation was similar in the case of K calculated on N basis as the partial K for faba bean was below one and for barely was above one. Total K was higher in the case of FB:B 2:1 in both cases measured in dry matter and N yield (Table 4, 5). The results of aggressivity conformed to those of the K as barley was the dominant species (Abarley positive) in all intercrops and Afaba bean had negative values (Tables 6, 7).
A similar trend to that of LER, aggressivity, and K was also observed for AYL. In particular, AYLbarley was higher than AYLfaba bean and also AYLbarley had positive values in all intercropped treatments, whereas the partial AYL of faba bean was negative (Table 6). The AYLtotal values were positive only in the case of FB:B 2:1 intecropping system (Table 6). The higher system productivity index (SPI) was found in FB:B 2:1 intercrop (6.45) compared with the other treatments (Table 6). Similarly, AYLNtotal and SPI for N had higher values in the FB:B 2:1 intercrop (Table 7). General, all indices that were used for N yield were higher than the same indices based on dry matter yield.
Economic indices
The Intercropping Advantage (IA) for barley indicated that all intercrops were economic advantageous for barley as they had positive values, however for faba bean all partial IA were negative indicating economic disadvantage (Table 8). In addition, IAtotal was the highest (3.22) in FB:B 2:1 intercrop. A similar trend was noted when the MAI values were observed and the MAI values were positive only in the FB:B 2:1 treatment (Table 8).
Discussion
Growth rate
Different arrangement of the intercropping species can affect the growth rate of faba bean in the intercrops and this could be attributed to the differences in competition between the two species that are grown together in the different intercropping systems (Lithourgidis et al., 2006; Brooker et al., 2015; Bedoussac et al., 2015). Moreover, the less effect of faba bean on growth rate of barley and the lower dry matter proportion of faba bean in intercrops compared with barley, could be attributed to the fact that barley is more competitive species than faba bean, which agrees with other studies where they compared other cereals (Lithourgidis et al., 2010, 2011b).
Dry matter, N yield and crude protein
The greatest dry matter yield was obtained in the intercropping treatment FB:B 2:1 and the lowest at the FB:B mixed system. This possible indicates that changing the arrangement of sowing the different species can affect the competition between the two species. In many cases, it has been reported that yields of intercrops between legumes and cereals were intermediate or even lower than yields of monocrops due to competition between the intercropped species (Vandermeer, 1990; Brooker et al., 2015; Bedoussac et al., 2015). However, there are several reports which show that intercrops can have higher dry matter yield compared with the monocrops (Chapagain and Riseman, 2014; Banik et al., 2006). The corresponding proportions of faba bean were 16, 23, 34 and 9% (vs. the expected 50%) in the intercropping systems FB:B 1:1, 2:2, 2:1 and mixed respectively. It seems that the different arrangement of the intercropping system can change the contribution of the two species due to the differences in competition. Also, the observed decrease of faba bean contribution in the dry matter yield of the intercrops compared with the expected 50% could be attributed to the stronger competitive nature of one species over the other (Anil et al., 1998; Lithourgidis et al., 2011b). Furthermore, a possible explanation for the low faba bean contribution in the intercrops is that barley produced many tillers and therefore showed much higher competitive ability than faba bean. In the case the intercropping system FB:B 2:1 which show higher dry matter yield compared with the rest intercropping systems that the competition was reduced and the two species grew better than in the other treatments.
Nitrogen yield was the highest at the faba bean monocrop due to the higher N concentration followed by the FB:B 2:1 intercropped compared with the other intercropping treatments which was quite lower. In addition, the proportion of faba bean contribution was much higher when was calculated on N basis compared with the dry matter basis. This is due to the higher N concentration that faba bean has and also higher dry matter contribution at the FB:B 2:1 treatment.
Faba bean monocrop had the lowest NUtE value (31.5) because of the higher N concentration which caused higher N uptake and the lower dry matter yield than the other crops. This means that the faba bean monocrop produced the lowest biomass per kilogram of N that was taken up than the other crops (Fageria and Baligar, 2005). In contrast, barley monocrop and the FB:B mixed and 1:1 intercrops had higher NUtE values than the other treatments (63.46 and 60.68, respectively). Therefore, barley and its intercrops with faba bean need a lower amount of N to produce the same amount of dry matter, which can be attributed to the higher N concentration in the tissues of faba bean and to that most of the N that is taken up is through N fixation. Similar response was found in other studies were faba bean and other legumes such as common vetch and pea showed lower NUtE, and also in some of the intercrops NUtE was lower than in the cereal monocrops (Lithourgidis and Dordas, 2010; Dordas and Lithourgidis, 2011).
Crude protein concentration is used as a quality characteristic of forage crops and is always measured to evaluate a forage system and especially intercropping systems like the ones studied in the present study (Malézieux et al., 2009; Chapagain and Riseman 2014). The highest CP concentration was found in faba bean monocrop followed by the 2:1 and 2:2 FB:B intercrops, and the lowest in barley monocrop. CP concentration depends on the soil available N and it is possible that these differences could be larger on poor soils and with no N inputs (Bedoussac and Justes, 2010, 2011). In most cases, CP yield per area basis followed a similar trend with the CP concentration, and it was the highest in faba bean monocrop followed by the FB:B 2:1 intercrop. Also, in the different intercrops there were significant differences in CP yield which is because of the differences in dry matter yield and also in the different CP concentration. In contrast, others found no significant differences in CP yield between the different intercrops and their monocrops (Li et al. 2006; Lithourgidis et al., 2011b, Dordas et al., 2012; Chapagain and Riseman 2014). In addition, the increase in CP concentration and CP yield which is because of the increase in faba bean contribution was reported in other studies and this is one of the most important reason for including a legume in an intercropping system (Osman and Nersoyan, 1986; Berkenkamp and Meeres, 1987; Carr et al., 2004; Brooker et al., 2015; Bedoussac et al., 2015).
Plant height and chlorophyll determination
In all treatments, plant height was not affected by the intercropping treatment. Similarly, Agegnehu et al. (2006) reported that plant height of faba bean in monocropping was not affected by the intercrop treatment as plant height was similar to that of faba bean in intercrops with barley. In contrast, others found that intercropping increased plant height of broad bean compared with broad bean monocrop (Karlidag and Yildirim, 2007). One possible reason for these discrepancies is the competition between the two species that are grown at the same time. When there is competition plants normally grow higher to capture more light, whereas when there is lower competition the difference in the plant height can be lower which is possible the case for the present study.
Chlorophyll meters are quite useful and were used extensively as a quick way for determining the N level and they measure the degree of greenness of the leaves during the growing season (Dordas et al. 2008). In addition, the degree of green color and the stay green characteristic is quite important for higher yields and better quality with higher protein concentration. Indeed, in this work, faba bean monocrop which had high chlorophyll content produced high CP yield and had higher CP content. On the other hand, the chlorophyll values of barley were not affected by the intercrop treatment.
Competition indices for dry matter and N yield
In all intercrops the LERfaba bean was lower than 0.5, while the LERbarley was higher than 0.5, which indicates that there was an advantage for barley in these intercropping systems and a disadvantage for the faba bean (Mead and Willey, 1980). Yield advantage of intercropping over monocrops in terms of total LER was higher than 1.0 in the FB:B 2:1 intercrop (1.08). More specifically, this means that a monocropping system will require up to 8% more land area to produce the same yield of the intercropping system (FB:B 2:1 intercrop). This fact indicates greater land use efficiency of intercrops than monocrops (Midya et al., 2005; Agegnehu et al., 2006). This significant yield advantage of the particular intercropping system over monocrops can be due to better land utilization and better use of the environmental resources for plant growth because of the better arrangement (Banik et al., 2006; Chapagain and Riseman 2014). Similarly, high LER values in intercropping with different arrangement where reported in other studies (Chen et al., 2004, Banik et al., 2006).
The land equivalent ratio for dry matter (LER) and for N yield is used as an indicator to quantify intercrop productivity of the intercropping system and is not used to interpret interference (Bedoussac and Justes, 2011). In the present study LERN of intercrops was up to 1.23 which means that intercrops can increase the use of N sources by up to 23%. In addition, in all treatments LERNB was >0.5 and LERNFB was <0.5 which also indicates that barley was much more competitive than faba bean to acquire N from the soil. This is probably because barley has a faster and deeper root growth and earlier N uptake and higher N demand as already observed by many authors for others cereal–grain legume intercrops (Hauggaard-Nielsen et al., 2003; Jensen et al., 2010; Bedoussac and Justes, 2011; Neugschwandtner and Kaul, 2015).
The results presented in this study are in agreement with other studies which showed values of LERN > 1 for cereal-legume intercrops (Hauggaard-Nielsen et al., 2009; Bedoussac and Justes, 2011; Neugschwandtner and Kaul, 2015; Monti et al., 2016). In addition, our results confirmed the advantage of the intercrop compared with the monocrop system for N acquisition more than for biomass production as reported by others (Bedoussac and Justes 2011). These results are probably because of the fact that the legume-cereal intercrops show higher efficiency due to the complementary use of mineral soil N and atmospheric N2 between the companion species (Bedoussac and Justes, 2010; Jensen et al., 2010). Cereals are more competitive than legumes for soil N and this forces the legume to rely on N2 fixation and to increase the percentage of N derived from air compared with the monocrop.
The complementary in the use of N resources was better for some intercropping treatments (i.e. FB:B 2:1) than for others (i.e. FB:B mixed). A greater share of faba bean biomass in the intercrops compared with the other intercropping treatments can explain the high N use complementarity found in FB:B 2:1 treatment (Bedoussac and Justes, 2011; Neugschwandtner and Kaul, 2015; Monti et al., 2016). Therefore, high LERN values are important for increasing the ecological role of intercropping (through the use of legumes) as a sustainable way to increase N into a low input agro-systems.
The K and Aggressivity values for barley were more than faba bean in all intercrops, which indicates a definite yield advantage of barley in the intercropping systems (Banik et al., 2000). In addition, total K was more than one in FB:B 2:1 indicating yield advantage over the monocrop (Ghosh, 2004; Dhima et al., 2007). Similar trend was found when K and Aggressivity were calculated on N yield basis.
The AYLbarley had positive values in all intercrops, which indicates a yield advantage for barley probably because of the positive effect of faba bean on barley in the intercrops. AYL is an index that can give more precise information about the inter- and intra-specific competition of the component crops than the other indices (Banik et al. 2000). Partial AYL can be used to quantify the loss or gain of dry matter yield due to the association with other species or the variation of the sowing density which cannot be obtained with the use of partial LERs. Therefore, it was observed a positive for barley and negative for faba bean which indicates an increase in dry matter yield of barley in the intercrops. Only the intercrop FB:B 2:1 had positive total AYL and this indicates an advantage of intercropping over monocrops which agrees with LER. When AYL was calculated on N yield it was found that AYLNFB was negative and the AYLNB was positive which indicates that barley was the dominant species in terms of N accumulated in the shoots compared with faba bean. This is probably because barley has a faster and deeper root system and take up more N earlier compared with faba bean which is based on N2-fixation from the atmosphere (Hauggaard-Nielsen et al., 2009; Bedoussac and Justes, 2011; Bedoussac et al., 2015).
The highest system productivity index (SPI) was found in FB:B 2:1 in which LER and K had also greater values. The values of SPI were high and largely determined by the FB:B intercrop which were not reduced much by intercropping indicating higher productivity of these intercrops (Agegnehu et al., 2006; Lithourgidis et al., 2011b).
Partial relative crowding coefficient, aggressivity, and partial actual yield loss values clearly indicated barley as dominant species in the intercrops of faba bean in the intercropping systems. Similarly, greater competitive ability of barley to exploit resources in association with chickpea, groundnut, red pea, common vetch and faba bean has been reported by other researchers (Ghosh, 2004; Banik et al., 2006; Lithourgidis et al., 2011b; Dordas et al., 2012; Dhima et al., 2007).