1. Introduction
Intercropping (IC) is the ancient practice of growing together two or more crops in one field during the entire growing season or at least in a timeframe [
1,
2]. IC generally has higher and more stable yields compared to sole crops (SC) [
3,
4,
5,
6,
7] because of the increased efficiency in using environmental resources by the intercropped species [
8,
9,
10,
11] and/or the reduction of weeds [
12,
13,
14,
15], and pests and diseases [
1,
16,
17].
Besides agronomic benefits, intercropping systems also deliver several agroecological services through enhanced biodiversity and synergistic effects between plants, so IC finally improves the resilience and sustainability of the cropping systems (CS) [
18,
19].
For these reasons, the Common Agricultural Policy (CAP) encourages farmers’ uptake of diversification measures and includes IC as a part of the eco-schemes [
20]. What is more, intercropping cereals with grain legumes can contribute to the EU goal of reducing nitrogen (N) fertilizer use in Europe and spreading organic farming in at least 25% of the European agricultural area. However, IC adoption is hindered by technical challenges among which sorting is a key constraint. Separating the yield of the companion crops is not necessary if the mixture is to be used as animal feed and IC has been proved to be more successful for forage than for grain production [
21,
22]. Therefore, IC for forage production could be more easily adopted by European farmers. Nonetheless, a large and sustained success of IC will not only rely on the numerous advantages of barley–field bean mixtures but will also depend on the quality (i.e., characteristics affecting consumption, nutritional value, and the final performance of livestock [
23]) of the mixed forages, because the livestock industry requests high-quality forages [
24].
Generally, growing a legume with a cereal has the potential for enhancing the quality because legumes are a good source of crude protein (CP) but produce low dry matter (DM) and, conversely, cereals provide high DM with low protein content [
3,
25,
26,
27].
However, other factors modify the forage quality, the most important of which is the harvest time [
28,
29] because plants continually change their composition as they mature; the plant cell wall content rises, and indigestible lignin accumulates, thus lowering forage quality [
27].
Also, nutrient availability (i.e., soil fertility and fertilization) may impact forage quality [
30]. For example, Carr et al. [
31] found forages more concentrated in crude protein (CP) in high- rather than low-soil-N environments, while Bedaso et al. [
32] observed a significant increase in the CP in herbage following manure application.
Angeletti et al. [
33] examined the effects of IC and different N rates on the forage production of barley and field bean and demonstrated that forage production was higher at the early dough stage than at the heading stage, even if the optimal rate of N at this stage was two-fold (i.e., 100 kg ha
−1) than the earlier harvesting. However, the literature reports a general shift to productivity more than to quality as the crops mature both in IC [
29,
34] and in SC barley [
35,
36] and field bean [
37].
So, we can hypothesize that intercropping and N fertilization could have not improved nutritive characteristics consistently with forage production [
33].
Furthermore, Angeletti et al. [
28] demonstrated that different barley and field bean ratios can affect the quality of herbage and silage as obtained by mixing the forage of the two crops. Anyway, in the combined forage of crops grown in IC, the actual proportion of the legume in the mixture cannot be directly manipulated like in the mixtures, as it is not only the result of the seeding rate ratios but is also shaped by N fertilization, by the stage of harvest, and by the competitive relationships established between the species [
22].
To the best of our knowledge, only a few studies have addressed the effects of N fertilization on the quality of forages of IC harvested at different maturities (i.e., growth stages).
To fill this gap, we analyzed the forages obtained in the experiment performed by Angeletti et al. [
33] with the following objectives: (i) evaluate how forage quality parameters develop across the crop cycle, i.e., from heading to early dough; (ii) set up the most suitable N-fertilizer rate for adequate quality forages; (iii) determine if barley–field bean intercropping improves the nutritional value of forage compared to the sole crops.
2. Materials and Methods
This study is part of a larger experiment that compared barley and field bean sole crops and intercropping, carried out in two consecutive years (from November 2014 to May 2015 and from November 2015 to May 2016) at the Research Centre of the Department of Agriculture, Food, and Environment of the University of Pisa, Italy (43°40 N, 10°19 E), on a Xerofluvent soil with loam texture, medium organic matter content (i.e., 2.1% determined following the Walkley and Black method), and (i.e., 1 g kg
−1 total N, Kjeldahl method) [
33]. In the two years, precipitations were similarly distributed from sowing to heading and from heading to early dough and were about 10% lower than the long-term mean for the area [
33]. While in the previous paper we reported the forage yields together with the key indices for IC evaluation (i.e., Land Equivalent Ratio, Aggressivity Index, Competitive Ratio, and Competitive Balance Index), here we present the bromatological parameters and the nutrient yields of the forages to provide insight into the effects of the treatments on the quality of obtained fodder biomass.
Therefore, the site characteristics, experimental design, and crop management refer to the same research of Angeletti et al. [
33]. However, briefly, we compared the quality of forages obtained at two stages of harvest (SH) (i.e., heading and early dough of barley) from three cropping systems (CS) (i.e., barley sole crop, field bean sole crop, and intercropping of barley and field bean in a 1:1 additive design) and five N-fertilizer rates (NR) (i.e., 0, 50, 100, 150, and 200 kg N ha
−1). Pure crops were fertilized similarly to the intercrops to balance the experimental design and make evaluations and comparisons more consistent. We used N rates from 0 to 200 kg ha
−1 because about 150 kg ha
−1 is the common amount in Central Italy [
38] and 160 kg ha
−1 did not affect field bean yields [
39].
The experimental design was a randomized complete block design, replicated three times, so each year 15 treatments were tested (resulting from 3 CS × 5 NR) through 45 plots (15 × 3 replicates) harvested twice a season.
On November 18 and 12 in 2014 and 2015, respectively, 400 seeds m−2 with a 16 cm row spacing of barley cultivar Ketos (Gotic × Orblonde) × (12,813 × 91H595) and 70 seeds m−2 with a 32 cm row spacing of field bean cultivar Chiaro di Torrelama were sown for the sole crops while 400 + 70 seeds m−2 of barley and field bean were sown in the IC.
Phosphorus and potassium fertilizers were distributed during pre-sowing (i.e., 150 kg ha
−1 of P
2O
5 and K
2O kg ha
−1). Nitrogen application was not conducted on unfertilized (control) plots, while urea was applied to fertilized plots split into two applications: 25 kg N ha
−1 at sowing and the remaining (i.e., 25, 75, 125, and 175 kg ha
−1) at the stem elongation of barley (BBCH 30) [
40], (i.e., on 14 February 2015 and 22 February 2016) as well as for sole and intercropped crops. Field bean sole crops received N fertilization concurrently with the cereal.
In both years, the crop growth was monitored referring to the BBCH scale [
36] to precisely identify the right growth stages for N fertilizations and harvest samplings.
The plants of both sole and intercrops were harvested twice: (i) at the beginning of heading—BBCH 51 of barley, while field bean at first pods visible (BBCH 70): on 21 and 20 April in 2015 and 2016, respectively; (ii) at early dough—BBCH 83 of barley, while field bean at final pods development (BBCH 79): on 10 May 2015 and 4 May 2016.
The plants were manually harvested at about 5 cm above the ground and those of IC were separated into components crops (barley and field bean). Samples were then oven-dried at 65 °C to a constant weight to measure dry weights (DW) and to determine dry matter (DM) concentration, milled to pass a 1 mm sieve, and analyzed to assess the bromatological traits of the forages. The concentrations of crude protein (CP), ether extracts (EE), ash, neutral-detergent fiber (NDF), acid-detergent fiber (ADF), and acid-detergent lignin (ADL) were determined according to Martilotti et al. [
41].
The concentration of non-fibrous carbohydrates (
NFC) was estimated using the formula described by
NRC [
42]:
And that of total digestible nutrients (
TDN) was calculated according to Lithourgidis et al. [
43]:
To estimate the fiber quality, the relative feed value (
RFV) was calculated according to Rohweder et al. [
42] from the estimates of dry matter intake (
DMI) and digestible dry matter (
DDM) as follows:
where
DMI (% of body weight) = 120/
NDF% and
DDM = 88.9 − (0.779 ×
ADF%).
RFV provides an estimation of the nutritional value of forage compared with a full bloom alfalfa which has
RFV = 100.
The nutrient yields of crude protein (CPY), ether extracts (EEY), neutral-detergent fiber (NDFY), acid-detergent fiber (ADFY), non-fibrous carbohydrate (NFCY), and total digestible nutrients (TDNY) per unit area were calculated by multiplying the DM yield (and reported by Angeletti et al. [
33]) per unit area and the corresponding concentrations, as above determined.
Hereafter, we define “Combined” as the forage produced by barley and field bean, either as sole crops or intercropping, together in the equivalent unit area (i.e., one hectare of intercropping and the sum of half a hectare of barley and field bean sole crops) as also reported by Angeletti et al. [
33]. Due to the additive design of IC, the data obtained by barley and field bean as sole crops are from a two-fold large unit surface area and were averaged to be compared to IC data, which were conversely summed because they were obtained by barley and field bean on a single unit surface area. In this way, the comparison between the two cropping systems is theoretically between the forage of a mixture of the two SC and the forage from intercropping.
The data on concentrations of combined forages were back-calculated from the nutritive yield of each bromatological trait of combined SC or IC divided by the DW of combined SC or IC forage.
The year and its interactions were not found significant by the analysis of variance (ANOVA) for any of the studied parameters; so, a combined ANOVA over the two years was performed and data were presented as the two-year average, as the Barlett’s test had previously revealed the homogeneity of variance over the two years. Data were analyzed with a split–split–plot design with NR as the main plots, CS as the subplots, and SH as the sub-subplots. The two SH were allocated in the sub-subplots so that successive observations on the same plots permitted a more precise evaluation of the treatment and its interactions [
44]. In
Supplementary Tables S1 and S2, the ANOVA results for the bromatological traits and for the forage nutrient yields of barley, field bean, and combined crops are reported. For significant main effects or interactions, the values were separated at the 0.05 probability level following Tukey’s HSD Test.
4. Discussion
This research corroborated that the quality of the forage obtained by barley and field bean was different and, generally, worse in the cereal compared to the legume, mainly because of a lower concentration of CP and NFC and a greater fiber concentration [
45].
However, the treatments imposed modified the quality of the forage.
First, the bromatological traits of the forage obtained by all the crops (i.e., barley, field bean, and combined crops) improved from heading to the early dough stage of harvest. In this transition, the CP, EE, and fiber components decreased while the quality indices, TDN and RFV, increased. The improvement of the quality across the crop cycle was also confirmed by a rise in the non-fibrous carbohydrates (NFC) at the later stage of harvest. This result was somewhat unexpected as the nutritive quality was supposed to decline with maturity [
45], because of a reduced leaf-to-stem ratio. In our research, even if the proportion of leaves in forage declined as the plant matured from heading to early dough [
22], the presence of the grains, lacking at the first but present in the second stage, likely contributed to the reduction in the fiber and to the qualitative improvement of the forages. Grains have low and highly digestible fiber and may have partially or completely offset the declining quality of the vegetative plant parts (leaves and stems). In addition, we previously demonstrated that the temporal relationships between barley and field bean happened, so that the legume was the dominant species at early dough [
33]. Thus, at this stage, a higher proportion of legume biomass was present which likely contributed to boost the quality of the forages of the IC. Moreover, this finding could also have been related to the additive design adopted in the present research. This design was chosen to maintain the same plant population and spacing to appropriately detach intra- and interspecific competition and correctly evaluate and compare the outcomes from IC and SC [
46,
47]. Similar results have also been observed in wheat–field bean [
48] and barley–field bean [
49] intercrops with an additive design.
Second, nitrogen fertilization also caused significant variations in the bromatological traits of the forages. As expected, the most marked changes were recorded in the cereal. In barley, the concentrations of the CP, EE, and fiber components increased while those of ash and NFC decreased, and consequently also the quality indices worsened, due to the N applications. The crude protein concentration was increased with N rates higher than 50 kg ha
−1, confirming the findings of Kwabiah et al. [
50] who stated that 60 kg ha
−1 of nitrogen are needed to optimize the CP content of barley. However, a lower leaf/stem ratio with increasing N rates in barley, as reported by Pampana et al. [
22], is likely the main factor responsible for the increase in the fibrous fraction of cellulose, hemicellulose, and lignin (i.e., NDF and ADF).
Moreover, we previously reported that nitrogen fertilization modified the relationships between barley and field bean, so that the cereal contribution to the total biomass was enhanced by the N rate [
22], due to its improved competitiveness for nitrogen. Therefore, the higher proportion of the cereal biomass explained the worse quality of the forages with increasing N rates.
Similar variations, although smaller and often not significant, also occurred in field bean and combined forage, in accordance with the findings of Ghambari and Lee [
51]. This finding further corroborates the importance of the crop land share in balancing the quality of forage in cereal/legume intercrops.
Third, the bromatological traits of the tested forages were modified by the cropping system (i.e., sole crops vs. intercropping). However, the forages of barley and field bean were not similarly affected by CS; in the cereal, the IC, with respect to SC, increased the crude proteins and fibers and decreased the NFC, while in the legume, the reverse occurred; in other words, IC decreased the forage quality of barley while increasing that of the legume. As a result, the mixtures had higher CP content than barley, while lower than field bean, as also observed in former research [
21,
29].
In our research, the increase in the CP concentration of barley was favored by the proximity of a legume, as also indicated by Carr et al. [
52] who found that intercropping barley with pea enhanced the CP concentration of forage. Higher CP in intercropped forage was also revealed by Roberts et al. [
53], Martin et al. [
54], and Hall and Kephart [
55]. The improvement of N concentration in intercropped cereals has been ascribed to better light, water, and nutrient use efficiency [
33,
56,
57], and in some cases, to the transfer of fixed N from the legume [
58].
It is worth noting that we did not discover a significant interaction between nitrogen fertilization and the cropping system; this indicates that the CP increase occurred comparably with all the tested N rates. This means that different N amounts had similar effects on both IC and SC and implies that the advantage due to intercropping with field beans was not only limited to nitrogen availability.
Our results confirm that increased nitrogen availability increases the CP levels in the forage of sole cereals but may have minimal effect in cereal–legume stands unless the legume component is substantially reduced [
46].
The increase in the fiber components of IC barley was probably due to the reduction in the production of spikes prompted by field bean competition [
22,
33]. Because of the changes detected in the two species, in combined forage, the adoption of the intercropping system increased crude protein and fiber, and decreased the NFC and forage quality. Similarly, intercrops have been proved to improve neutral-detergent fiber and water-soluble carbohydrates [
59,
60,
61]. Notably, in our research, barley always showed values of NDF higher than 55% (i.e., the level above which Van Soest [
62] affirmed that they diminish voluntary intake of forage), but they were significantly reduced to acceptable levels (i.e., 29.4% in SC and 27.2% in IC) in combined forage, because of the lower values of field bean (about 37–40%).
Overall, the forage nutritive value, as expressed by the relative feed value (calculated through dry matter digestibility, ADF, and intake potential, NDF), increased from heading to early dough but was not improved by the other studied treatments (i.e., N rate and cropping system). However, in our research, the effect of the CS depended on the crop, because the RFV of barley was lowered by the IC crop system, while that of field bean was improved, which determined a comparable RFV in combined forage in both cropping systems (i.e., SC and IC). This confirms that the impact of the growth stage is often greater than other managements on forage quality [
52,
63,
64,
65].
Together with the bromatological traits (i.e., nutrient content) of the forages, the nutrient yield (i.e., yields of nutrients obtained per unit area) should also be considered for an efficient evaluation of the SH, NR, and IC effects on forage production, because forages generally make up the most consistent part of feed ratios (e.g., half of the feed rations of lactating cows) and therefore determine the animal units that can be fed.
Noticeably, in our research, the differences in nutrient yields came from the differences in biomass production rather than the nutrient concentration. With aging (i.e., from the heading to the early dough stage of harvest), the CPY increased in barley, field bean, and combined forage, but this was due to the increase in dry matter production [
33] because their concentrations were lower at the latest stage. This means that the increase in the dry matter yield more than counterbalanced the decrease in the concentrations of the CP, EE, NDF, and ADF. However, in barley, the increases during the same period occurred also for the EEY and NFCY but not for the fiber components. Higher plant biomass implies the “dilution” of nutrients, resulting in the decrease in nutrient elements [
65,
66,
67]. The difference in CP and fiber content could be a side effect of the presence of the grains in barley, which was the dominant species with higher competitiveness for light.
As expected, N fertilization improved the nutrient yields of barley, whose CPY yields progressively increased with the N rate, and differently from the SH effect; in this case, it was because of the mutual effect of both biomass and concentration increases. Interestingly, there was no interaction with the CS treatment, implying that in both cases, N was the most limiting factor. Conversely, the yields of all other parameters were increased by the N rate more in SC than in IC, prompted by the different effects on the biomass [
22].
In field bean and combined forage, however, the increases in yield due to the N fertilizer were less evident and not linearly affected by N availability because legumes can obtain nitrogen from the biological fixation of atmospheric N and facilitate N availability to their companion crops in mixtures and associations [
68,
69,
70].
Also, the nutrient yield of forages differed between the two cropping systems (i.e., SC and IC). Overall, the yields of barley per unit area were higher when sole cropped but the differences were enlarged by N availability, highlighting other resources restriction, and confirming that the effects of IC are more important under low-N environments [
52,
71,
72,
73]. Different N availabilities may also explain the inconsistent literature findings, as Li et al. [
74] did not find significant differences in CP yield between IC and SC; but Carr et al. [
52] explained that increasing intercrop protein yield required a significant proportion of the legume DM in the mixture which is driven by lower N availability [
22].
The legume obtained a lower nutrient yield when intercropped, due to concurrent lower concentration and biomass production [
22], confirming the findings of Hauggaard-Nielsen and Jensen [
75] and Ghaley et al. [
76].
The effects that the cropping system and N rate caused on the nutrient yield were more marked than those recorded for the bromatological characteristics of the forages. Thus, from SC to IC, the protein and TDN yield increased by 46 and 29%, respectively, while from 0 to 150 N, it increased by 49 and 46%.
However, the combined forage permitted a higher nutrient yield when obtained in IC than in SC, further confirming that growing the crops in association has more advantages than just mixing the forages of the two monocrops.
5. Conclusions
The present research quantified the changes in the nutrient contents and yields of the forages of sole and intercropped barley and field bean harvested at two growth stages and grown with five N rates.
We found that bromatological traits worsened at early dough, even if the overall quality increased, mainly because of the improved content of the NFC. However, the nutrient yields were boosted, driven by higher biomass production.
Therefore, we can conclude that delaying harvesting can be feasible if the main aim is the self-sufficient feed production, because the increase in calculated CP yield was 1500 kg ha−1. The reverse is true if the forage should be used to fulfill deficiencies or integrate feeds and high nutrient levels are demanded to support high animal production.
Also, nitrogen fertilization influenced the quality of combined barley and field bean forage, causing in both cropping systems an increase in the fiber components, a decrease in the NFC, and a decrease in quality parameters. The most marked variations were recorded in the cereal rather than in the legume and were probably due to the increase that fertilization caused in the vegetative plant parts, rather than in the reproductive one.
We determined that no or a low-rate N application (as much as 50 kg ha−1) can better exploit the relative feed value of forages, while the CPY and TDNY are maximized with 150 kg N ha−1.
The adoption of an additive intercropping increased the concentration of crude protein and fiber, simultaneously decreasing the general quality of the forage. This was mainly driven by the cereal performance that reduced the production of ears, and therefore of grains, because of the competition exerted by the field bean.
We established that higher nutrient yields of all parameters can be achieved with IC adoption, thanks to the boosted biomass production.
In conclusion, we found that N fertilization reduced the NFCs and the quality expressed as the TDN, despite increasing the fiber concentration (NDF and ADF), whereas intercropping improved the CP concentration of the forage. N fertilization should not exceed 50 kg ha−1 to maximize the RFV, while 150 kg ha−1 are preferable to boost nutrient yields.
Overall, our results indicate that barley and field bean intercropped in Mediterranean environments should be harvested at the early dough stage to gain the highest CPY and relative feed value. However, for higher CP concentration, heading is recommended.
Finally, the present evaluation of the quality of forages from intercropping systems may improve the cultural agronomic acceptability of this practice, with farmers becoming more aware of the effects on their forages, contributing to the EU goals of larger IC adoption.