Introduction

Research on intercropping with varying crop species for cover crops has yielded mixed results regarding component performance. Notably, intercropping for cover cropping is argued to be beneficial when the dominant crop receives a temporary competitive advantage (Amossé et al. 2013; Mhlanga et al. 2016; Brooker et al. 2020; Dzvene et al. 2023). However, this approach disadvantages the productivity of the cover crop, leading to poor uptake of the practice. Cover crop plant populations are reduced due to heavy competition and shading from the dominant crop (Mhlanga et al. 2016). Therefore, optimizing the planting timing and stand density of the cover crop can enhance its biomass productivity. Intercropping a cover crop with maize (Zea mays L.) to provide soil cover during the early, mid, or late growing seasons constitutes a management practice known as live mulching (Liedgens et al. 2004; Sigdel et al. 2021; Dzvene et al. 2023). This approach represents a sustainable and climate-smart strategy for diversifying rainfed monocropping systems and enhancing agroecosystem services in semi-arid areas.

The practice of intercropping legumes with cereal food crops to ameliorate soil biochemical properties in rainfed smallholder and subsistence farming systems is gaining popularity (Tsubo et al. 2003; Cong et al. 2015; Garland et al. 2017). Numerous studies have assessed the feasibility of integrating grain legumes such as pigeon pea (Cajanus cajan), groundnuts (Arachis hypogea), cowpea (Vigna unguiculata), and beans (Phaseolus vulgaris) into maize-dominated cropping systems to bolster food security (Mucheru-Muna et al. 2010). Additionally, researchers have explored intercropping options involving forage legumes, grasses, and brassicas to augment both the quality and quantity of biomass production for livestock feed, as well as residual soil biochemical properties (Hassen et al. 2017; Javanmard et al. 2020; Kutamahufa et al. 2022).

Conversely, the latter approach receives lower priority, as living mulch has no direct economic gains in smallholder cropping systems (Dzvene et al. 2022). Farmers are concerned with harvesting grains or fruits, which can be exchanged for cash or directly utilized in the household. They need to gain the scientific understanding behind cover crop intercropping for improved biodiversity by providing agroecosystem benefits. For instance, grain legumes like cowpea are commonly employed in semiarid environments owing to their adaptability and low fertility requirements; nodulation can further enhance soil nitrogen (Jeranyama et al. 2000; Li et al. 2007). Sunn hemp (Crotalaria juncea) shares similar adaptability traits with cowpea (Jeranyama et al. 2000). However, only a few quantitative studies have investigated or optimized its intercropping potential for improving the improvement of soil biochemical properties.

The advantages of intercropping include the provision of essential ecosystem services such as the reduction of soil erosion and nutrient losses, increased nutrient use efficiency, addition of soil nitrogen (N) through N2 symbiotic fixation with legume cover crops, and improved soil quality (Garland et al. 2017; Javanmard et al. 2020; Sigdel et al. 2021; Kutamahufa et al. 2022). Legume subordinate crops are most often selected for preventing nitrogen losses and for biological nitrogen (N2) fixation, which may reduce nitrogen inputs required for the subsequent crop, such as maize (Jeranyama et al. 2000; Hartwig and Ammon 2002; Li et al. 2007). Legumes can potentially fix nitrogen (N2), a portion of which will be available for high-nitrogen-requiring crops such as maize (Jeranyama et al. 2000; Mucheru-Muna et al. 2010). In areas where excess nitrogen is already a problem, the use of ground covers may provide a sink to tie up some of this excess nitrogen and hold it until the next growing season, when a crop that can make use of the nitrogen might be planted (Hooda et al. 1998). These possibilities provide the incentive for looking at the effect of various crop species on soil erosion, nitrogen budgets, weed control, nutrient availability, and other pest management and environmental problems (Hartwig and Ammon 2002). However, there is a need to evaluate the intercropping of leguminous subordinate crops on soil properties under rainwater harvesting.

In-field rainwater harvesting (IRWH) tillage is a widely used adaptation technique to rainfall variability and fluctuations among South African smallholder farming homesteads, aimed at conserving soil moisture and extending crop water availability (Bothma et al. 2012; Botha et al. 2015). The technique is recommended for use on shallow duplex clay soils, particularly those with sloping topographies in semiarid areas (Hensley et al. 2000). This is because in-field rainwater harvesting is ineffective on coarse-textured soils with high hydraulic conductivity (Rockstrom 2000). Crop residue retention is not only a requirement for the IRWH technique’s mulching component but also significantly impacts nutrient availability (Michael et al. 2021), as well as improving soil structure texture, and reducing hydraulic conductivity (Lampurlanés and Cantero-Martínez 2006). In South Africa, farmers in arid and semiarid areas possess arable lands with poor-textured soils that retain little water. Therefore, applying organic mulch becomes a priority for soil water retention while improving texture, structure, and nutrient availability. Moreover, the South African smallholder farming system is crop-livestock integrated (Dzvene et al. 2022), and current management involves continuous crop residue removal due to livestock feeding (Vanlauwe et al. 2014; Dzvene et al. 2019). Thus, there is a need to evaluate the intercropping of forage legumes to increase biomass production, offset the effects of carbon (C) removal, improve soil biochemical properties and ultimately crop yield. Several studies have highlighted the positive impacts of IRWH on physical soil properties such as soil moisture (Dunkerley 2002; Al-Seekh and Mohammad 2009; Botha et al. 2015). In these studies, mulching role was essential in enhancing the soil’s water content under IRWH.

There are contrasting reports on the effect of IRWH on soil biochemical properties. A study by Singh et al. (2012) investigated the impact of rainwater harvesting combined with afforestation on soil properties, tree growth, and the restoration of degraded hills. After five years, there was an increase in soil pH, soil organic carbon, electrical conductivity, ammonium nitrogen (NH4+-N), nitrate (NO3−-N), and extractable phosphate (PO4-P) down the slope of their study sites (Singh et al. 2012). After five years, the increases in these soil properties were not attributed to rainwater harvesting but rather to a mass movement of material from the upper to lower slope. This resulted in the accumulation of salts and nutrients transported along with water from upper to lower slope positions (Singh et al. 2012). Similarly, Li et al. (2006) also observed similar trends of nutrient accumulation from the upper slope to the lower slope due to the positive effects of IRWH, as it enabled soil water retention and nutrient mobilization that enhanced vegetation cover through the turnover of roots and litter. Another study by Liu et al. (2009) in semiarid China, where mulching coupled with no-till practice was used as rainwater harvesting, observed an increase in soil organic carbon by 2.7% compared to the conventional tillage practice where a maize crop was planted. Al-Seekh and Mohammad (2009) reported a 5% increase in soil organic carbon due to rainwater harvesting on runoff, sediments, and soil properties. A more recent study by Mduduzi (2015) in KwaZulu Natal and Eastern Cape provinces of South Africa observed no clear trend in the effect of IRWH on exchangeable bases, soil pH, and micronutrients across all study sites. A short-term study by Posthumus and Stroosnijder (2010) showed no significant impact of rainwater harvesting on selected biochemical properties. A comprehensive review shows that the benefits of IRWH on crop yield increases and improved soil biochemical properties can be realized when the technique is used in conjunction with soil conservation techniques such as living mulching with cover crops, minimum and zero tillage (no-till).

Intercropping with a legume in a cereal-based system is critical for achieving the benefits of improved dominant crop yield and enhanced soil chemical properties. Legume intercropping can benefit significantly through improved nutrient and water cycling efficiency, enhanced climate regulation, wildlife habitat, and increased aesthetic, educational, and recreational value opportunities (Franzluebbers et al. 2014). Many changes associated with soil properties under perennial crops are mainly driven by limited soil disturbance and increased organic matter inputs from roots and rhizodeposits compared to annual crops (Franzluebbers 2015). The main critical aspects of intercropping management are selecting crop species and establishing an appropriate intercropping time (Sigdel et al. 2021). Berti et al. (2017) noted a 14% and 10% reduction in corn and soybean yields in response to intercropping camelina at V1-V3 of the corn and V1-V2 of the soybean growth stages, respectively. The choice of subordinate crop species may depend on the desired benefit and cost. Promoting a balance of favourable nutrient conditions and preventing competition for resources between the dominant crop and the intercropping subordinate crop species selected for live mulching is important. Therefore, it is essential to integrate sustainable and optimal practices under IRWH that will ensure improved soil chemical properties and soil moisture concurrently, ultimately achieving higher crop yields of arable land in the arid and semiarid regions of South Africa. This study’s rationale for intercropping sunn hemp, an annual crop species, by varying its planting times and stand densities was to optimize its biomass productivity for improving dominant crop yield. The study aims to investigate the effects of intercropping sunn hemp with maize at different timings and stand densities on soil chemical properties, soil water content, above-ground biomass, and maize grain yield. We hypothesized that the optimal sunn hemp intercropping period and planting density management in a maize-based system would benefit soil chemical properties through organic matter addition, which would eventually translate to improved crop yields of the main crop.

Materials and methods

Study site

The sunn hemp intercropping experiment was conducted at Kenilworth Experimental Farm (29°01′S, 26°09′E, 1354 m) near Bloemfontein (Kenilworth, Bainsvlei ecotope), Free State province, South Africa, during 2019/20 and 2020/21 summer growing seasons. The soil at the site is classified as Bainsvlei form (Soil Classification Working Group 2018), similar to Chromic Stagnic Plinthic Cambisol (WRB soil groups 2014). The experimental site is located in a semiarid agro-ecological zone with well-drained, red-brown soils with less than 1% topsoil organic matter with a very deep profile (>2000 mm), sandy-loam soils with a soft plinthic B2 horizon (Soil Classification Working Group 2018). The soil had a clay, sand, and silt fraction of 8.5%, 85%, and 7%, respectively, at the start of the experiment. The soil of the experimental site is also characterized as moderately acidic, with an average 0–30 cm depth of pH of 5.2, NH4-N concentrations of 10.3 mg kg−1, NO3-N concentrations of 11.2 mg kg−1, and available phosphorous concentrations of 7 mg kg−1 in the upper 300 mm horizons. The mean exchangeable base values for sodium, potassium, calcium, and magnesium were 22, 142, 336, and 100 mg kg−1, respectively. The region receives 528 mm annual rainfall (±155.6 mm), and annual mean minimum and maximum temperatures are 11.0 °C and 25.5 °C, respectively, with monthly standard deviations of ±0.8–2.0 °C and ± 1.2–3.1 °C. The main rainy season lasts from October to April, with some rain falling during May, August and September.

Experimental design and field management

A split plot design was employed with three intercropping times as the main plot, and three stand densities as the subplot with three replicates. The standard maize developmental stage system was used to identify the vegetative stage (from seedling emergence VE to physiological maturity PM) (Ritchie et al. 1993). The intercropping time levels were simultaneously with maize planting (P1), V15 maize growth stage (P2), and R1 maize growth stage (P3). The stand densities were 16 plants m−2 (D1-low), 32 plants m−2 (D2-medium), and 48 plants m−2 (D3-high). Sole maize and sole sunn hemp were included, where sole sunn hemp was also planted at the respective three stand densities only at P1. Maize population in sole and as the component was fixed at four plants m−2, i.e., with an inter-row spacing of 1 m and in-row of 0.36 m. The sunn hemp in both intercrop and sole was planted in five rows with a 30 cm row spacing in the runoff strip of the in-field rainwater harvesting (IRWH) technique. The technique combines basin tillage and no-till for rainfall water harvesting from a 2 m wide catchment that is kept no-till, and crop planting is done along 1 m basin areas that serve as infiltration zones (Botha et al. 2003). In the sole sunn hemp treatment, there were no plants in the place where the maize was usually planted on the sides of the basin area. As a result, this study’s intercropping system was similar to strip intercropping, with two rows of maize in a 1 m strip and five rows of sunn hemp in a 2 m strip. The main plots were 180 m2 (12m width, 15 m length), and the subplots were 60 m2 (12m width, 5 m length). The intercrop components were sown in an additive series in both seasons (Connolly et al. 2001).

Rainfed maize fertilizer applications were based on a potential yield of 5000 kg ha−1 as determined by the Fertilizer Society of South Africa (FSSA 2007). Maize (cv. Pioneer P2432R) and sunn hemp (cv. local) were fertilized with 200 kg ha−1 2:3:2 (22) NPK (equivalent of 13 kg N ha−1, 19 kg P ha−1, and 13 kg K ha−1). No topdressing was applied on the sunn hemp cover crop. To meet the N requirements, a top dressing of 250 kg ha−1 lime ammonium nitrate (LAN 28% N, i.e., 70 kg ha−1 N) was split in half and applied to maize plots four and seven weeks after emergence. Crops were planted at relatively high densities and thinned to the required densities two weeks after emergence. Planting of SM, SSH, and simultaneous intercropping with maize planting took place on 3 December 2019 and 23 November 2020 for the respective 2019/20 and 2020/21 growing seasons. Simultaneously intercropping with maize planting and SSH treatments, crops grew for the entire growing season. The treatments were terminated at maize physiological maturity (PM) (30 March 2020 and 20 March 2021). Sunn hemp intercropping at the V15 maize growth stage was planted on 16 January 2020 and 8 January 2021, and both were terminated on 16 April 2020 and 7 April 2021, respectively. Sunn hemp intercropped at the R1 maize growth stage was planted on 7 February 2020 and 1 February 2021 in each growing season and was terminated at the same time as the sunn hemp intercropped at the V15 maize growth stage. Weeds were manually controlled throughout the season and spotted maize beetle (Astylus atromaculatus) infestation on maize during the reproductive phase was controlled with Dursban emulsifiable concentrate (EC) 480 g/l (44.6% w/w), as needed. Crop harvesting was done by hand, and maize and sunn hemp stover were left in the field. Therefore, the quantities and qualities of sunn hemp residues retained varied according to the management aspects.

The establishment of IRWH plots began during the summer of December 2018. The land was initially conventionally prepared with a ripper, mouldboard plough, and disc. This was followed by a single sheer mouldboard plough with a basin-forming implement to construct the basins in an E-W direction on the field with an N-S slope. The runoff strips in the plots were manually raked to smooth the topsoil. The IRWH plots were established with a 2:1 basin to runoff strip (Tesfuhuney 2012). The subsequent tillage was minimized. No-till for maize sowing was planted in tramlines (1.0 m wide) adopted from previous IRWH techniques in the Free State, South Africa (Van Rensburg et al. 2012).

Soil sampling and analysis

From the experimental plot, three random samples were collected from the top horizon within a 1 m2 area up to a depth of 30 cm using a 5 cm corer for both the basin and runoff sections in each IRWH plot. The collected soil samples from each plot were homogenized to form a composite sample. The samples were air-dried and then passed through a 2 mm sieve. Afterward, they were stored for laboratory analysis. The samples were collected annually at the end of the growing season, following the harvest of maize and sunn hemp, in the 2019/2020 and 2020/2021 seasons. This approach aimed to consider the impact of intercropping treatments on soil organic matter decomposition and soil fertility. The soil samples underwent analysis using standard methods. For instance, soil pH was measured with a pH meter in a 1:2.5 (v/v) soil-to-water suspension, following the procedure outlined in the Agricultural Laboratory Association of Southern Africa (AgriLASA 2004). Exchangeable cations (K, Ca, Mg, and Na), micronutrients (Cu, Mn, Zn, and Fe), and plant-available phosphorus (P) in soil samples were extracted using a modified Ambic 2 procedure (Thompson 1995) and determined using an Inductively Coupled Plasma Emission Spectrograph (ICP-OES) (Varian 710-ES). Total nitrogen (N) was determined in the air-dried soil samples through dry combustion using the LECO Truspec-CNS analyzer (LECO 2003). Soil organic matter (SOM) was quantified using the loss on ignition (LOI) method, as outlined by Okalebo et al. (2002).

Soil water

The soil water content in the root zone was monitored at intervals of 7 to 14 days starting from planting and at a depth of 0.30 m. Within each experimental plot, two neutron water meter steel access tubes were inserted into the centers of the basin and runoff strips, reaching a depth of 1.8 m. This resulted in 78 access tubes used for soil water measurement. A neutron water meter (NWM) (Campbell Pacific Nuclear model 503, CA USA, 1994) recorded neutron counts along the access tubes. These NWM counts were then converted into volumetric soil water content (cm3 cm−3) using two calibration regression equations developed on-site for the specific soil type under investigation (Chimungu 2009). Soil water content in millimeters (mm) was subsequently calculated as the product of volumetric water content (cm3 cm−3) and soil depth (mm).

Above-ground biomass and grain yield

Sunn hemp was sampled from a 1 m2 quadrant randomly placed in the center of the plot. All the sampled plants were oven-dried at 65 °C for 72 hours until a constant weight was achieved to determine dry biomass. Two rows of maize, each 5 m in length (covering an area of five m2), were manually harvested in two sections of the basin area when they reached maturity. This was done from the center of each plot (net plot) and was carried out at ground level using a sickle. The harvested maize plants were weighed after drying to ascertain the aboveground biomass. Maize cobs were then removed, and threshing was performed using a hand-powered thresher to determine grain yield. The mass of maize grain was calculated based on a water content of 12%.

Statistical analysis

The data were analyzed using SAS John’s Macintosh Project (JMP) ® Pro 14 statistical software (SAS Institute Inc., Cary, NC). Factors S, P and D were considered the three main factors when testing the effect of sunn hemp intercropping time (P), stand density (D) and growing season (S) and their interactions. If significant, post-hoc tests using the least significant difference (LSD) test at P ≤ 0.05 tested for differences between means. The three-way analysis of variance (ANOVA) was computed based on the statistical model as follows (Eq. 1).

$${Y}_{ij k m}=\mu +{\beta}_i+{A}_j+{\delta}_{ij}+{B}_k+{AB}_{ik}+{\omega}_{ij k}+{C}_m+{AC}_{jm}+{BC}_{km}+{AB C}_{jkm}+{\varepsilon}_{ij k m}$$
(1)

where:

Yijkm:

Response level

μ:

general mean

βi:

block effect

Aj:

season effect

δij:

the season random error (Error a)

Bk:

main plot/ treatment effect

ABik:

interaction effect between the season and the main plot

ωijk:

subject error (Error b)

Cm:

sub-plot effect

ACjm:

interaction effect between season and sub-plot

BCkm:

interaction effect between main plot and sub-plot

ABCjkm:

the three way (Factors A*B*C)

εijkm:

sub-plot random error effect (Error c) was used to test the cropping systems effects on the measured parameters

Results

Table 1 shows the analysis of variance (ANOVA) results for the intercropping effect of sunn hemp into maize at different planting periods (P) and stand densities (D) during the two experimental growing seasons (S) on the measured soil chemical properties, above-ground biomass and maize grain yield. The ANOVA results indicate that the growing season was the main factor influencing the sunn hemp intercropping effects on most of the measured soil chemical properties (Table 1). Extractable nutrients, Zn and Fe, above-ground biomass and maize grain yields were more sensitive to the interactions of the growing season (S), intercropping period (P) and stand density (D) as shown in Table 1.

Table 1 Analysis of variance (ANOVA) results for the effect of sunn hemp intercropping period (P) and stand density (D) on soil chemical properties of the Plinthic Cambisol, above-ground biomass and maize grain yield after two growing seasons (S)
Full size table

Soil organic matter

Sunn hemp intercropping period and different stand densities had no significant effect (P > 0.05) on SOM accumulation between the growing seasons (Table 1). However, the interaction effect of sunn hemp intercropping period and stand density on SOM contents was significant (Table 1). The impact of the planting period of the sunn hemp (P) seems to have a more substantial influence on the SOM accumulation compared to the different sunn hemp stand densities, as indicated by significantly different SOM contents. The effect was more visible when sunn hemp was planted together with maize (P1) and when sunn hemp was planted at the R1 maize growth stage (P3) at a constant stand density of 16 plants m−2 (Table 3). Sunn hemp managed with P1 was allowed to mature, and at maturity, sunn hemp exhibited a high biomass quantity, which can slow mineralization and results in a high SOM value (Fig. 4). However, the short growing season and competition from maize hampered the growth of the P3 sunn hemp treatment, resulting in low biomass quantity (Fig. 4). When the two growing seasons were compared, significant differences were noted in SOM concentrations across the two growing seasons (Table 2). Soil organic matter increased significantly from 2.14 g/kg in 2019/2020 to 2.99 g/kg in 2020/2021.

Table 2 The effect of the growing season of intercropping sunn hemp with maize on selected soil chemical properties of the Plinthic Cambisol
Full size table

Soil pH and macronutrients (N, P, K, Ca, Mg)

Sunn hemp intercropping at various maize growing stages and stand densities had no effect on pH between the growing seasons (Table 1). All the experimental plots’ soils were acidic, with pH values ranging from 4.54 to 4.99. However, the pH of the soil in the experimental plots with sunn hemp residue retention increased from 4.66 in 2019/2020 growing season to 4.77 in the 2020/2021 growing season (Table 2). After two seasons of treatment application, the soil pH values were lower compared to their initial status. Sunn hemp intercropping management had no significant effect (P > 0.05) on soil pH, N, P, Ca, and Mg concentrations in the experimental plots between the growing seasons (Table 1).

The effect of the growing seasons was significant on the concentrations of N, K and Ca (Table 2). Only Mg was not significantly different across the two growing seasons. N and K increased significantly from 0.124 g/kg and 178.3 mg/kg in 2019/2020 to 0.153 g/kg and 225.6 mg/kg in 2020/2021 respectively (Table 2). In contrast, Ca decreased significantly from 325.2 mg/kg in 2019/2020 to 286.8 mg/kg in 2020/2021. Overall, after two seasons of treatment application, P, K, and Mg increased whereas Ca decreased.

Micronutrients (Na, Cu, Mn, Zn, Fe)

Sunn hemp intercropping at various maize growing stages and stand densities had no effect on most of the micronutrients across the growing seasons (P > 0.05) except Zn (Table 1). Extractable Zn was the most affected soil chemical property by the management of sunn hemp. Across the two seasons, significantly higher Zn accumulated when sunn hemp was planted at the V15 maize growth stage (P2) (results not shown). Furthermore, the planting period’s interaction with the stand density and season was significant on the extractable Zn (Table 1). Planting sunn hemp at the V15 maize growth stage with a stand density of 32 plants m−2 registered significantly higher Zn contents than other combinations across the two seasons (Table 3). Higher Zn concentrations were recorded at the end of the first season (S1) when sunn hemp was planted at early maize vegetative growth (P2) (results not shown). Combining all the treatment factors shows that planting sunn hemp at V15 maize growth stage with a stand density of 32 plants m−2 during the first season resulted in significantly higher Zn contents (Table 4). These significant observations on extractable Zn seem to have been triggered by planting the sunn hemp at V15 maize growth stage (P2). Significantly higher concentrations of Fe were registered across different sunn hemp density treatments at the end of the first season compared to the second season (Fig. 1).

Table 3 Sunn hemp intercropping period (P) and stand density (D) interaction effect on SOM and extractable Zn of the Plinthic Cambisol
Full size table
Table 4 Sunn hemp intercropping growing season (S), planting period (P) and stand density (D) interaction effect on extractable Zn (mg/kg) of the Plinthic Cambisol
Full size table
Fig. 1
figure 1

The interaction effect of sunn hemp intercropping stand density (D) changes with growing seasons (S) on the Plinthic Cambisol’s extractable Fe (mg/kg). (D1-low: 16 plants m−2; D2-medium: 32 plants m−2; D3-high: 48 plants m−2; S1-growing season 1: 2019/20, S2-growing season 2: 2020/21). Different letters on the bars indicate the statistical significance at LSD (0.05). Error bars indicate ± standard deviation

Full size image

The results confirmed the significant effect of higher cover crop stand densities on net Fe removal due to crop harvest (Fig. 3). The concentrations of Na, Fe, and Mn were significantly (P < 0.05) affected by the growing seasons (Table 1). Na and Fe concentrations were significantly reduced from 5.55 to 3.80 and 65.23 to 36.64 mg/kg, from 2019/2020 to the 2020/2021 growing season respectively, while concentrations of Mn were significantly increased from 13.86 to 22.26 mg/kg during the same period (Table 2). A net decrease in Na was noted after two seasons of treatment application. At the end of the first and second seasons, the initial extractable Na content at the beginning of the trial was significantly reduced by 74.8% and 82.7%, respectively.

Soil water

During the first season (2019/20) in the basin side (maize), the highest soil water (80 mm) was recorded at approximately 14 DAP in the P3D3 treatment (Fig. 2c). The lowest water content was recorded at approximately 84 DAP in all the treatment plots irrespective of the intercropping period or plant densities. Generally, P2 treatments recorded high soil water contents as compared to P1 and P3 throughout the season. In contrast, P1 recorded lower soil water content throughout the season as compared to P2 and P3, regardless of the stand density. At the start of the second season (2020/21), all the experimental plots had higher soil water content than the start of the first season (2019/20). The highest soil water content (84 mm) was recorded at approximately 28–36 DAP in the P3D3 treatment (Fig. 2f). The lowest soil water content (33 mm) was recorded at approximately 48 DAP in the P1D3 treatment (Fig. 2e). Throughout the season, irrespective of the plant density, there is no clear trend among the intercropping periods in which one sustained high or low soil water contents (Fig. 2).

Fig. 2
figure 2

Changes in basin side (maize) soil water content at the 0–30 cm soil profile depth as influenced by the sunn hemp planting period and stand density management during the 2019/20 (a, b, & c) and 2020/21 (d, e, & f) growing seasons [Simultaneous intercropping at maize planting (P1), intercropping at the V15 maize growth stage (P2), and intercropping at the R1 maize growth stage (P3), planting at D1: (16 plants m−2); D2: (32 plants m−2); and D3: (48 plants m−2)]. RF = rainfall (mm) on the right-hand side Y-axis

Full size image

During the first season (2019/20) in the run-off side (sunn hemp), the highest soil water (85 mm) was recorded at approximately 30 DAP in the P3D3 treatment (Fig. 3c). The lowest water content was recorded, approximately 84 DAP in all the treatment plots, irrespective of the stand densities or intercropping period. Throughout the whole season, P3 recorded relatively higher soil water contents than P1 and P3, regardless of the stand density. Contrastingly, P1 treatments recorded lower soil water content than P2 and P3 treatments throughout the season. At the start of the second season (2020/21), all the experimental plots had higher soil water content than the start of the first season (2019/20). The highest soil water content (88 mm) was recorded at approximately 36 DAP in the P3D3 cropping system (Fig. 3f). The lowest soil water content (27 mm) was recorded approximately 48 DAP in the P2D2 treatment (Fig. 3e). Throughout the season, irrespective of the stand density, there is no clear trend among the intercropping periods in which one sustained high or low soil water contents (Fig. 3). Generally, the highest soil water content was recorded in the P3D3 cropping system in both the basin and the runoff side in both seasons (Figs. 2 and 3).

Fig. 3
figure 3

Changes in runoff side (sunn hemp) soil water content at the 0–30 cm soil profile depth as influenced by the sunn hemp planting period and stand density management during the 2019/20 (a, b, & c) and 2020/21 (d, e, & f) growing seasons [Simultaneous intercropping at maize planting (P1), intercropping at the V15 maize growth stage (P2), and intercropping at the R1 maize growth stage (P3), planting at D1: (16 plants m−2); D2: (32 plants m−2); and D3: (48 plants m−2)]. RF = rainfall (mm) on the right-hand side Y-axis

Full size image

Above-ground biomass and grain yield

The intercropping period significantly affected both the sunn hemp and maize above-ground biomass (Table 1). Planting sunn hemp and maize simultaneously, as well as planting sunn hemp at the V15 maize growth stage, had significantly higher above-ground biomass as compared to planting at R1 maize growth stage (results not shown). The effect of planting sunn hemp at different stand densities was significant only on the sunn hemp biomass (Table 1). The above-ground biomass for the maize and sunn hemp were not significantly different for the two seasons (Table 1). The interactive effects of the intercropping period and different densities of sunn hemp were significant on both the maize and sunn hemp above-ground biomass (Table 1). Moreover, the interactive effects of the intercropping period of sunn hemp and season was also significant on both maize and sunn hemp above-ground biomass (Table 1). The interaction effects of the intercropping period with the sunn hemp stand density and season show that the simultaneous planting of both maize and sunn hemp resulted in more above-ground biomass of both maize and sunn hemp (Figs. 4 and 5).

Fig. 4
figure 4

a The interaction effect of sunn hemp intercropping period (P) and stand density (D) on sunn hemp biomass (Mg/ha) at the end of the 2020/21 growing season after harvest. b The interaction effect of sunn hemp intercropping period (P) and season (S) on sunn hemp biomass (Mg/ha) at the end of the 2020/21 growing season after harvest (P1: simultaneous sunn hemp and maize planting; P2: sunn hemp planted at the V15 maize growth stage; P3: sunn hemp planted at R1 maize growth stage; D1-low: 16 plants m−2; D2-medium: 32 plants m−2; D3-high: 48 plants m−2; S1-growing season 1: 2019/20, S2-growing season 2: 2020/21). Different letters on the bars indicate the statistical significance at LSD (0.05). Error bars indicate ± standard deviation

Full size image
Fig. 5
figure 5

a The interaction effect of sunn hemp intercropping period (P) and stand density (D) on maize biomass (Mg/ha) at the end of the 2020/21 growing season after harvest. b The interaction effect of sunn hemp intercropping period (P) and season (S) on maize biomass (Mg/ha) at the end of the 2020/21 growing season after harvest. c The effect of sunn hemp intercropping period (P), stand density (D) and season (S) on maize grain yield (Mg/ha) at the end of the 2020/21 growing season after harvest. d The interaction effect of sunn hemp stand density (D) and season (S) on maize grain yield (Mg/ha) at the end of the 2020/21 growing season after harvest. (P1: simultaneous sunn hemp and maize planting; P2: sunn hemp planted at the V15 maize growth stage; P3: sunn hemp planted at R1 maize growth stage; D1-low: 16 plants m−2; D2-medium: 32 plants m−2; D3-high: 48 plants m−2; S1-growing season 1: 2019/20, S2-growing season 2: 2020/21). Different letters on the bars indicate the statistical significance at LSD (0.05). Error bars indicate ± standard deviation

Full size image

The effects of sunn hemp intercropping period, density and season were all significant on the maize grain yield (Table 1). Planting sunn hemp and maize simultaneously as well as planting sunn hemp at the V15 maize growth stage had significantly higher maize grain yield as compared to planting sunn hemp at R1 maize growth stage (Fig. 5). Treatments where sunn hemp was intercropped at low and medium densities had significantly higher maize grain yields as compared to treatments where sunn hemp was intercropped at high densities (Fig. 5). The maize grain yield was significantly lower at the end of the second season than the first (Fig. 5). The interactive effects of intercropping sunn hemp at different stand densities and season was significant on the maize grain yield (Table 1). In both seasons, the grain yield at the second and third sunn hemp densities was significantly different (Fig. 5).

Discussion

Effect of sunn hemp intercropping period, stand densities and their interactions on selected soil properties after two seasons

The lack of a significant effect of sunn hemp management strategies on SOM accumulation after two growing seasons can be attributed to the short-term nature of the study. A study by Wang et al. (2015) comparing intercropping and monocultures reported similar findings after 10 years. The significant effect of the interaction between sunn hemp intercropping period and stand density on SOM contents (Table 1) indicates that the planting period of the sunn hemp (P) seems to exert a more substantial influence on SOM accumulation as compared to different sunn hemp stand densities. Early planting of sunn hemp likely resulted in a higher biomass yield of the sunn hemp residues, consequently yielding significantly higher SOM contents (Fig. 4). The significant SOM increase due to higher sunn hemp biomass demonstrates that crop mixtures can build resilient agricultural cropping systems by enhancing SOM stocks (Poeplau and Don 2015; Thapa et al. 2022). A remarkable increase of SOM across the two seasons highlights the positive effects of no-till systems in conjunction with crop residue retention under IRWH. Several field studies have shown that, in conjunction with residue retention, no-till systems result in a more significant accumulation of SOM matter in the surface layers (Govaerts et al. 2007). Generally, crop residues with slow decomposition rates maintain a surface cover mulch over an extended period, which helps reduce weed pressure and conserve soil and water more effectively (Thapa et al. 2022).

Generally, sunn hemp management strategies have no effect on soil pH, consistent with observations by Kutamahufa et al. (2022), who claimed that the accumulation of different pools of organic matter during the early years of establishing interactive cropping systems is usually too low to affect fundamental soil chemical properties like pH. Generally, the acidic nature of the soils at the experimental site provides a soil pH range that is not ideal for optimal nutrient availability for most crops, including maize (Kutamahufa et al. 2022). However, the increase in the pH of the soil in the experimental plots with sunn hemp residue retention increasing by 0.11 units after two seasons shows the positive effects of using a legume in enhancing the soil’s biochemical properties (Gura et al. 2022). At the end of the first season, planting sunn hemp at a stand density of 48 plants m−2 (D3) significantly enhanced soil pH, further confirming legumes’ positive effects on soil biochemical properties. The lower soil pH after two seasons of treatment application can be due to the acidifying effect of no-till adoption (Turmel et al. 2014) under the IRWH technique in this study. The increased acidity under no-till adoption is attributed to the SOM accumulation from the decomposition of residues (Franzluebbers and Hons 1996). However, some additional acidifications could also be attributed to the nitrification effect of fertilizer N that was applied at the beginning of the trial.

The lack of significant effect of the management of sunn hemp on macro-nutrients may be due to the short-term nature of the study. A study by Wang et al. (2015) did not observe any significant effects of intercropping on soil fertility after 10 years. These observations suggest that changes in soil chemical properties may require a longer time for subtle changes to become evident due to intercropping. The significant increases noted in N and K across the two growing seasons were most probably associated with the decomposition of the sunn hemp residues retained on the soil surface. Sunn hemp have a lower C:N ratio (Reicosky et al. 1995), which can be easily decomposed, thereby returning a significant amount of nutrients to the soil, and contributing to sustainable soil fertility management (Gura et al. 2022).

Crop species such as sunn hemp, with higher decomposition rates, release nutrients, especially N, quickly, helping to meet all or part of the N requirements of subsequent crops (Poffenbarger et al. 2015; Thapa et al. 2018, 2022). Moreover, in the long-term, due to their faster decomposition rates, sunn hemp residues will stabilize greater proportions of residue-derived C and N into soil organic matter enhancing agricultural sustainability through greater microbial efficiency (Cotrufo et al. 2013). The non-significant effect of residues associated with P and Mg may also indicate that some nutrients may require a longer time to be affected, depending on the management system employed (Govaerts et al. 2007). A decrease associated with Ca may be due to a higher net removal of Ca by the growing maize and sunn hemp crops and a lack of sufficient replenishments through the decomposition of crop residues.

Significantly lower Fe and Na concentrations in the second season may be due to these nutrients being removed through crop harvest and lack of sufficient replacement through decomposition of crop residues. The significant increase in Mn after two seasons was possibly caused by crop residue decomposition, adding more Mn to the soil. A net decrease in Na was noted after two seasons, which may be a significant observation confirming the usefulness of a forage legume crop in reducing salinity levels. This agrees with an observation made by Gura and Mnkeni (2019), who discovered that incorporating soybean crop in the crop rotation sequence significantly reduced salinity levels.

The importance of crop mulches in retaining soil water was demonstrated by the high soil moisture at the start of the second season compared to the start of the first season. At the end of the first season, the above-ground biomass left on the soil surface after harvest retained soil moisture for the second season. The highest soil water content recorded in the P3D3 cropping system in both seasons may be attributed to the high above-ground biomass due to the high stand density of sunn hemp and less competition for water due to late planting. The high above-ground biomass late into the season maintained a surface cover mulch over an extended period compared to other treatments, which helped conserve soil water more effectively (Thapa et al. 2022).

Effect of sunn hemp intercropping period, stand densities and their interactions on above ground biomass and crop yield after two seasons

The significance of the intercropping period on above-ground biomass for both crops indicates that intercropping sunn hemp earlier in the season offers more benefits for the maize crop by providing N earlier, thus resulting and in higher maize biomass. Sunn hemp planted at different densities did not result in any significant effect on above-ground maize biomass. In contrast, a study by Nurgi et al. (2023) aimed at determining the effects of variety and plant density of a legume on yield components of crops and productivity, reported a significant difference in above-ground biomass for the two seasons under comparison. The results from this study suggest that intercropping sunn hemp earlier provides an opportunity for the main crop to have sufficient N throughout the season, hence the higher above-ground biomass.

After two seasons of applying the intercropping treatments, maize grain yield was significantly influenced by sunn hemp management strategies. The results indicate that intercropping sunn hemp earlier during the season offers more benefits for the maize crop by providing more N earlier, hence the significantly higher maize grain yields for the treatments where sunn hemp was planted together with maize. In contrast, sunn hemp intercropped at higher densities significantly reduced maize grain yields. This decrease in maize yield may be attributed to high competition for nutrients by the sunn hemp at high stand densities and the maize crop. The decrease in maize yield due to competition between maize and other forage crops has been frequently reported (Borghi et al. 2013; Secco et al. 2023). The interactive effects of intercropping sunn hemp at different stand densities with the season also highlighted the competition aspect between the maize crop and sunn hemp at high stand densities.

Conclusion

Sunn hemp planting date and stand density had no significant effect on most measured soil chemical properties between the two growing seasons. However, extractable Zn was the most sensitive to the management of sunn hemp, as indicated by the significant interaction effects of the sunn hemp planting period and density. The significant increases reported in SOM, N, K and Mn between the two growing seasons were most probably associated with the decomposition of the sunn hemp residues with varying qualities and quantities retained on the soil surface. Conversely, a decrease associated with Ca and Fe may be attributed to higher net removal of these nutrients by the growing maize and sunn hemp crops, coupled with subsequent lack of sufficient replenishments through decomposition of sunn hemp residues.

The retention of sunn hemp residues with varying quantities and qualities due to the planting periods and densities influenced short nutrient dynamics and soil water contents. Moreover, above-ground biomass and maize grain yields were also sensitive to the interactions of the growing season, intercropping period, and stand density. Significant changes in soil chemical properties may take longer to manifest, and future research should be conducted in agricultural regions with different soil mineral matrices. Additionally, sunn hemp management exerted varying abilities to fix nitrogen, scavenge nutrients, and improve soil organic matter. Hence, the evaluation of sunn hemp at various planting times and stand densities can determine the most effective combination for enhancing nutrient cycling and achieving sustainably higher crop yield.