**3. Results**

#### *3.1. Flowering Period, Plant Growth, and Soil Nitrate Content*

Experiments showed that maize flowering started between 60 and 70 days after sowing (DAS) and lasted for a short period between nine (FAK 2018) and 21 days (ET 2018) (Table 3). Beside of yellow sweet clover (MC), all IFP provided an additional period with flowers. Alfalfa (MA) was buried during the mechanical weeding process in FAK. Common vetch (MV) had the earliest flowering initiation of all IFP at all locations and years. On average, 40 DAS flowering started. However, flowering did not last for a long time in 2018 in ET and TH. After nine and 28 days of flowering, the common vetch was infested by *Erysiphaceae* (TUL. & C. TUL.) spp. and the plants died. Since the flowers of summer squashes I–III (MS1, MS2, MS3) and nasturtium (MN) were below the cutting height of the crop chopper, these IFP build up flowers even after maize harvest. In case of the common beans (MB1, MB2), it was observed that the first flowers appeared very late, more than 60 DAS. They even flower after maize anthesis and sometimes shortly before harvest.

**Table 3.** Days after sowing (DAS) until begin of flowering (SBF) and end of flowering (SEF), and the whole flowering time of maize and the different intercropped flowering partner (IFP) treatments (Δ = flowering period in days) for Ettlingen (ET), Tachenhausen (TH), and Forchheim am Kaiserstuhl (FAK) in 2018 and TH and FAK in 2019. The flowering periods of maize in the intercropping treatments corresponded to those of the maize in the control. For reasons of simplicity, the intercropped treatments only show the flowering periods of the IFP.


Common bean II (MB2) not available in 2019. M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II. ¶ In mixtures the first flowers in the crop stand were counted as begin flowering, regardless which of the mixture partners flowered, and for end of flowering, vice versa ‡ IFP still flowers after harvest. If common bean flowered until harvest, harvest date was set as end of flowering due to the fact that all flowers are in the harvested material.

The combined cultivation of maize and IFP did not affected the final height of maize in ET 2018 (Table 4). At ET no significant influence of N-level (*p* = 0.055) or intercropped seed placement (*p* = 0.326) on final maize plant height were observed in 2018 (Table 4). The IFP also did not show differences on plant height (*p* = 0.817). At TH (*p* = 0.943) and FAK (*p* = 0.225), there were also no significant changes in plant height in 2018 (Tables 5 and 6). In 2019, TH (*p* < 0.001) and FAK (*p* < 0.001) had significant differences in plant height. At TH, the highest values for maize plant height were found in the control, without any intercropping partner (M) with 341 cm, while the height of the maize in the intercropping plots with nasturtium (MN), summer squash I and II (MS1, MS2), and mixture I (MM1) was not significantly different from the control (M). In FAK, only common vetch (MV) and mixture II (MM2) were significantly different from the control (M), with 253 cm and 259 cm compared to 291 cm (Table 6).

In 2018, the DMY was significantly influenced by N fertilization in ET (*p* = 0.004) (Table 4). Full fertilizer application yielded the highest DMY, while there was no difference whether 100% of the required N is applied or 50% (Table 4). In 2018, no differences were measured whether the IFP was sown in the rows or between. The strongest influence on DMY was caused by the IFP used (*p* < 0.001). Significant lower yields compared to the control (M) were found by intercropping with alfalfa (MA), yellow sweet clover (MC) and common vetch (MV). Intercropping with nasturtium (MN) or common beans (MB1, MB2) showed no significant difference compared to 15.6 Mg ha−<sup>1</sup> DMY in the control. At TH significant IFP influences on DMY in both years were determined (*p* = 0.003 and <0.001) (Table 5). In 2018, common vetch (MV) showed with 11.5 Mg ha−<sup>1</sup> the lowest DMY, which were not different from the control (M) in 2018, while in 2019 common vetch (MV) and mixture II (MM2) had significant lower DMY compared to the control (M), with 17.9 Mg ha−1. Significant differences in the DMY were only found in 2019, not in 2018 (Table 6). Significantly lower DMY were found for common vetch (MV), summer squash III (MS3), and mixture II (MM2) in 2019.

The fractioning of the plant biomass in ET 2018 showed no significant influence of the N-Level on the share of maize, IFP and weed (*p* = 0.504, *p* = 0.067 and *p* = 0.198) (Table 4). The use of an IFP did

not reduce the share of maize (*p* = 0.341). The IFP with the highest share in the harvested biomass were the common beans (MB1, MB2). They had a share of 3.45 and 4.28%, while nasturtium (MN) with 0.61% had a low share of the harvested biomass. With a share of more than 87.0% maize was the main yield component. At TH in 2018 a significant influence by IFP on the share of maize biomass was observed (*p* < 0.001) (Table 5). The smallest proportions of maize were found with 94.5% when intercropped with common vetch (MV). In 2019 there was also a significant influence of the IFP on the maize proportion (*p* < 0.001). Smallest maize proportion was found under intercropping with both mixtures (MM1, MM2) and with common bean I (MB1). FAK in 2018 showed no significant influence on maize proportion by IFP (*p* = 0.221) (Table 6). In this year also there was a weed share between 1.55% (MS2) and 6.67% (MB2). The highest proportion of IFP in the harvested biomass was found in common vetch (MV) in 2018 and mixture II (MM2) in 2019 at 5.78% and 3.96%, respectively. Both years and all locations showed that intercropping maize leads to maize proportions of >81.8% in the harvested biomass.

**Table 4.** Growth parameters plant height (cm), dry matter yield (DMY, Mg ha−1), dry matter content (DMC, %), and the share of maize, intercropped flowering partners (IFP), and weed of the DMY from section harvest (%); for the three factor levels N-level, seed placement of IFP and the different IFP treatments at Ettlingen (ET) in 2018.


Values with the same letter within one parameter indicate non-significant differences within the three factor levels (N-Level, Seed placement and IFP) (HSD-test, α = 5%). \*\*\* *p* ≤ 0.001, \*\* *p* ≤ 0.01, \* *p* ≤ 0.05. M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II.


**Table 5.** Growth parameters plant height (cm), dry matter yield (DMY, Mg ha−1), dry matter content (DMC, %), and the share of maize and intercropped flowering partners (IFP) of the DMY from section harvest (%); for the different IFP treatments at Tachenhausen (TH) 2018 and 2019.

Values with the same letter within one parameter indicate non-significant differences within the IFP (HSD-test, α = 5%). \*\*\* *p* ≤ 0.001, \*\* *p* ≤ 0.01, \* *p* ≤ 0.05. Common bean II (MB2) not available in 2019. M Maize (Control), MA Maize + Alfalfa, MC Maize Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II.

In 2018, in ET, the soil NO3-N content was mainly influenced by the N-level (*p* < 0.001), in 0–30 cm and 30–60 cm depth after harvest, and at end of vegetation period, respectively. IFP showed in ET no influence after harvest (Figure 1). The seed placement also did not show an influence at any layer and both dates.

At the end of the vegetation period, the layers from 0–30 cm and 30–60 cm showed significant differences by the IFP used (*p* = 0.044 and *p* = 0.014) (Figure 1). Common bean II (MB2) had the significant highest amount with 80.4 kg ha−<sup>1</sup> NO3-N in the upper layer, while in the middle layer both beans (MB1, MB2) did not differ significantly from the control (M). In TH, no significant influence by IFP on soil NO3-N content was found in any soil layers, for the two dates and in 2018. In 2019, mixture II (MM2) had significant lower NO3-N contents with 3.14 kg ha−<sup>1</sup> compared to 8.41 kg ha−<sup>1</sup> NO3-N in the control (M) at the end of the vegetation period in 30–60 cm. Also, at the 60–90 cm layer common vetch (MV) had higher NO3-N contents than control (M), 3.25 compared to 1.49 kg ha−<sup>1</sup> NO3-N.


**Table 6.** Growth parameters plant height (cm), dry matter yield (DMY, Mg ha−1), dry matter content (DMC, %) and the proportion of maize, intercropped flowering partners (IFP) and weed of the DMY from section harvest (%); for the different IFP treatments at Forchheim am Kaiserstuhl (FAK) 2018 and 2019.

Values with the same letter within one parameter indicate non-significant differences within the IFP (HSD-test, α = 5%). \*\*\* *p* ≤ 0.001, \*\* *p* ≤ 0.01, \* *p* ≤ 0.05. Common bean II (MB2) not available in 2019. M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II.

Differences at FAK after harvest in 2018 were only detectable in the deepest layer. Common vetch (MV) and common bean II (MB2) showed increased contents of 103% and 125% compared to the control (M), respectively. At the end of the vegetation period, intercropping with yellow sweet clover (MC) showed NO3-N contents of 17.1 kg ha−1, while the control had 9.48 kg ha−<sup>1</sup> in the upper layer.

## *3.2. Quality Parameters*

In 2018, the DMC was neither significantly influenced by N-level (*p* = 0.536), nor by seed placement of the IFP (*p* = 0.294) (Table 4). Only the IFP itself had a significant influence on the DMC (*p* = 0.012). With 34.0% nasturtium intercropping (MN) showed the highest DMC, which did not differ from the control. All DMC were in a range were ensiling could take place. At FAK (Table 6), no differences were found in 2018 (*p* = 0.096), but the DMC for some IFP were over the recommended ensiling maximum of 40% DMC due to the hot and dry weather, which accelerated the ripening. For TH in 2018, significant differences were observed (*p* = 0.003), DMC of yellow sweet clover (MC) intercropped plots were significantly higher from the control (M). In 2019, no significant differences were found between the IFP.

**Figure 1.** NO3-N (kg ha−1) of soil samplings for the different intercropped flowering partner (IFP) treatments (M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II) in Ettlingen (ET) (2018 **A1**), Tachenhausen (TH) (2018 **A2**, 2019 **B1**) and Forchheim am Kaiserstuhl (FAK) (2018 **A3**, 2019 **B2**) after harvest and at the end of the vegetation period (End.Veg.Ped.; =end of growing season, when the weather causes a plant growth stop) for the three depths 0–30 cm (green bars), 30–60 cm (white bars), and 60–90 cm (gray bars). Bars with the same letter within one depth and one sampling date indicate non-significant differences (HSD-test, α = 5%).

The N-level had a significant influence on the content of CP in the biomass at harvest time (*p* < 0.001) (Appendix A Table A1). With increasing N-level there was a significant increase in CP. The highest content was achieved at 100% N-level, with 8.91%. The placement of the IFP seeds did not have an influence on CP (*p* = 0.236), while the influence of IFP was significant (*p* < 0.001). The only statistical influence between the IFP could be found between common vetch (MV), with 9.97% and all the other IFP (Figure 2). Control (M) achieved 7.08%. In 2018 at TH, nasturtium (MN) and both summer squashes (MS1, MS2) had lower CP contents than the control. In the second season, nasturtium (MN) and summer squash I and III (MS1, MS3) showed once again significantly reduced CP contents compared to the control (6.09%), while mixture II (MM2) showed an increase.

**Figure 2.** CP (crude protein), CF (crude fat), CX (crude fiber), CA (crude ash), and NfE (nitrogen-free extracts) content (% of DM) for the different intercropped flowering partner (IFP) treatments (M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II) in Ettlingen (ET) (2018 **A1**), Tachenhausen (TH) (2018 **A2**, 2019 **B1**), and Forchheim am Kaiserstuhl (FAK) (2018 **A3**, 2019 **B2**). Bars with the same letter within one parameter indicate non-significant differences (HSD-test, α = 5%).

For FAK in both years, CP differences were measured (Figure 2), while in 2018, the summer squash I (MS1) had lower CP contents than the control. In 2019, common vetch (MV) had higher CP contents. No other intercropping treatment showed significant changes.

Neither the N-level (*p* = 0.407, *p* = 0.694, and *p* = 0.324) nor the seed placement (*p* = 0.854, *p* = 0.725, and *p* = 0.870) had significant influences on CF, CX, and CA content in ET 2018 (Table A1). The IFP also did not have an influence on the CF content (*p* = 0.178) (Figure 2). Common vetch intercropping (MV) showed higher CX contents than the control (M), while nasturtium (MN) and both beans (MB1, MB2) had comparable CA contents with the control. The NfE content was influenced by the N-level (*p* = 0.003) but not by the seed placement of the IFP (*p* = 0.959) (Appendix A Table A1). A high fertilization leads to lower NfE contents. The NfE behaved in the opposite way to the CP. Common vetch (MV) showed the significant lowest NfE contents in ET (Figure 2). In TH there were no significant differences between the IFP in 2018 for CF, CX, CA, and NfE, while in 2019, significant differences were observed for all parameters. For CF, none of the IFP differs from the control, while for CX and CA common bean I (MB1) had significantly higher amounts. For the NfE, all IFPs, except alfalfa intercropping (MA), had lower contents than the control. In FAK, only CF (*p* = 0.008) and CA (*p* = 0.007) showed significant differences in 2018. For CF, common bean I (MB1) showed higher contents than the control (M), while for CA, alfalfa (MA) had higher contents. In the second year, CX, CA, and NfE (*p* = 0.018, *p* = 0.031, and *p* = 0.004) showed significant differences, but none of the IFP was different from the control, except the summer squash III (MS3), which had a lower amount of NfE.

For use as feedstock in biogas plants, besides the DMY, the amount of produced biogas and methane are important factors. Only the N-level had a significant influence on the biogas yield (*p* < 0.001) (Appendix A Table A1). Significant decreases in biogas yields were found for 100% N fertilization compared to 0 and 50%, respectively, while the N-Level had no influence on methane yield (*p* = 0.714). Placement of IFP seed showed no influence on biogas and methane yield (*p* = 0.546 and *p* = 0.733) (Appendix A Table A1). The IFP had an influence on these two parameters in ET 2018 (*p* = 0.025 and *p* = 0.016) (Figure 3). Common vetch (MV) showed significantly lower biogas and methane yields, compared to the control (M). While the control achieved yields of 556 L kg−<sup>1</sup> oDMC (organic DMC) for biogas and 300 L kg−<sup>1</sup> oDMC for methane, intercropping with common vetch (MV) reduced these values to 539 L kg−<sup>1</sup> and 294 L kg−<sup>1</sup> oDMC, respectively. For TH, the response of the IFP on biogas and methane yield depended on the experimental year, while in 2018, biogas and methane yield were significantly increased by intercropping with common bean I (MB1), and in 2019, both yields were significantly decreased by intercropping with common bean I (MS1). In FAK, there were no significant differences on the biogas or methane yield for any of the IFP used, neither in 2018 nor in 2019.

In dairy cattle feeding, GE, ME, and NEL are important factors for evaluating the quality of the silage. At ET in 2018, there was only a significant influence of the N-level on the content of GE (*p* = 0.021), ME and NEL were not influenced by the level of N fertilization. The seed placement of the IFP had no influence on the feeding quality parameters (Appendix A Table A1). All three parameters showed that intercropping with common vetch (MV) significantly decreased the contents in ET.

In TH in 2018, ME and NEL showed significant differences, but no differences were found between the control (M) and any of the other IFP. The only significant difference was found between summer squash I intercropping (MS1) and common vetch intercropping (MV) (Figure 4). In 2019, except for the alfalfa intercropping (MA), all intercropping treatments had significant lower GE contents than the control (M). For ME and NEL, only intercropping with common bean I (MB1) showed significantly decreased contents. In FAK, no differences were observed in 2018 for any parameters. In 2019, only the GE showed significant changes. While intercropping with alfalfa (MA), yellow sweet clover (MC), common vetch (MV), and nasturtium (MN) did not differ significantly from the control (M), the use of the other IFPs significantly reduced the GE.

**Figure 3.** Biogas and Methane yields (L kg−<sup>1</sup> oDMC) for the different intercropped flowering partner (IFP) treatments (M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II) in Ettlingen (ET) (2018 **A1**), Tachenhausen (TH) (2018 **A2**, 2019 **B1**), and Forchheim am Kaiserstuhl (FAK) (2018 **A3**, 2019 **B2**). The upper end of each colored section in the stacked bars shows the respective value for the parameter's biogas and methane. Bars with the same letter within one parameter indicate non-significant differences (HSD-test, α = 5%).

**Figure 4.** GE (gross energy), ME (metabolizable energy), and NEL (net energy for lactation) (MJ kg−<sup>1</sup> DM) for the different intercropped flowering partner (IFP) treatments (M Maize (Control), MA Maize + Alfalfa, MC Maize + Yellow sweet clover, MV Maize + Vetch, MN Maize + Nasturtium, MS1 Maize + Summer Squash I, MS2 Maize + Summer Squash II, MB1 Maize + Common Bean I, MB2 Maize + Common Bean II, MS3 Maize + Summer Squash III, MM1 Maize + Mixture I, MM2 Maize + Mixture II) in Ettlingen (ET) (2018 **A1**), Tachenhausen (TH) (2018 **A2**, 2019 **B1**) and Forchheim am Kaiserstuhl (FAK) (2018 **A3**, 2019 **B2**). The upper end of each colored section in the stacked bars shows the respective value for the parameter's GE, ME, and NEL. Bars with the same letter within one parameter indicate non-significant differences (HSD-test, α = 5%).
