Nitrogen release from white clover (Trifolium repens L.) residue and ensuing plant–soil utilisation by arable rotations

White clover is grown in monoculture for seed production, and, after seed harvest, the crop residue decomposes and becomes a source of nitrogen (N) to the ensuing crop in the rotation. This research aimed to quantify the amount of N accumulated in white clover biomass during seed production, the time of N release from that residue, and its subsequent utilisation in the plant–soil system. Two field experiments were conducted in commercial white clover seed crops established via conventional cultivation (“Conventional”) or direct drill (“Direct Drill”) in Canterbury, New Zealand. After seed harvest, the Conventional clover was replaced with a kale seed crop, whereas the Direct Drill clover was taken for a second season of seed production. For 238 days after seed harvest, the amount of N was measured in white clover residue, ensuing crop biomass and soil mineral N (NH₄⁺, NO₃⁻) in 0–10-, 10–20-, and 20–40-cm depths. Fallow subplots were created to enable estimation of the effect of N uptake by the ensuing crop. Soil N mineralisation was estimated using the acetylene (C₂H₂) reduction assay method. Total soil N (%) was determined at 0 and 238 days after seed harvest. Residue N content decreased in an asymptotic manner, over thermal time, with readily decomposable components (lamina, petiole, and floret) mineralised first. Residue-N had been 70% mineralised 760°Cd after harvest and was 95% decomposed after 1,600°Cd. Under the ensuing crop, total soil N increased by 226–232-kg N/ha, which was consistent with the quantity of N supplied as clover residue. In fallow plots, the total soil N increase was ~50% less than the increase under actively growing plants. It was estimated that unaccounted-for N had leached below the 40-cm depth. Ensuing crop N uptake was of comparable magnitude to the N available in both experiments. It was estimated that, due to N inputs, N from clover residue would be mineralised in the spring following seed harvest. This research concluded that the white clover seed crop residue rapidly released N, which was estimated to not be at risk of loss to the environment, provided the crop rotation proceeded without an extended fallow period.

1 Introduction

White clover (Trifolium repens L.) seed production is an important part of the crop rotation practiced by arable growers. Seed production of white clover from arable systems is necessary, so that, as a nitrogen (N) fixing species, white clover can be included in high-yielding pastures for diverse livestock production systems. New Zealand produces ≥50% of the global white clover seed supply, followed by Denmark, the USA, Uruguay, Argentina, and Australia (FAR, 2021; Mather et al., 1995). For seed production, the legume is grown in monoculture, and the seed is harvested in late summer. In New Zealand, during the 2023/24 season, 7,640 ha of white clover was grown to produce certified seed (Seed&Grain NZ, 2024). Seed yields average 470-kg/ha (Chynoweth et al., 2015), with specialty growers achieving yields of 750 to >1,000-kg/ha (Freeman, 1985; Clifford, 1985). Canopy management of the crop, via topping, grazing, or herbicide application, is a practice intended to enhance seed yield through increasing the number of inflorescences per unit area (Chakwizira et al., 2011). Nitrogen accumulated in clover biomass during seed production serves as a source of N for the next crop in the rotation after the clover crop is harvested. After 1 or 2 years of seed production, the clover crop is terminated by being sprayed and/or cultivated before the next crop is sown.

During seed production, the white clover plant builds a leaf canopy via horizontal proliferation of stolons outward from a central crown and taproot. The stolons produce leaves (lamina borne by a petiole) and flowers (inflorescence borne by a peduncle). These plant components have different N content (%). Components associated with photosynthesis (lamina) have N contents between 4% and 7% (Olykan et al., 2021; Sturite et al., 2006), whereas structural or storage components (petiole, stolon, root) have 2%-4% N, depending on plant maturity (Hay et al., 1985; Sturite et al., 2006; Cowling, 1961). The N contents of the reproductive components, inflorescence, and peduncle, are not well investigated or reported. Seed N content ranges from 4% to 5% (Whelan and White, 1985; Kumar and Goh, 2000a). Quantification of the contribution of each of these components to total biomass N is needed to inform post-harvest residue additions to soil. At harvest, the clover biomass (lamina, petiole, peduncle, and inflorescence) above the cutting height of the combine harvester is threshed to remove seed, with the remainder of the biomass, termed offal, spread behind thecombine, forming a residue N pool. The stolons, crown, and roots below the cutting height remain intact, allowing the clover to regrow. This biomass only contributes N to the clover residue N pool after crop termination.

The decomposition of plant residue is a microbially controlled process, where the timing of nutrient release depends on residue quality, soil conditions, residue management, and climatic conditions (Coûteaux et al., 1995; Kumar and Goh, 2000b). Residue quality influences the biodegradation of the material, with the water-soluble fractions degraded first (sugars, proteins, organic acids, etc.), then cellulose, hemicellulose, and finally lignin (Knapp et al., 1983; Harper and Lynch, 1981; Collins et al., 1990). The residue C:N ratio dictates the degradability of the plant material, with a C:N below 20–30 associated with faster decomposition rates due to reduced microbial competition for N (Janzen and Kucey, 1988). Edaphic factors that affect the microbial population and function, particularly temperature, moisture, and pH, also influence the rate of residue decomposition (Paul and Clark, 1989). A pH of 4.4–7.2 (Olewski et al., 2011), soil temperature of 25°C–35°C (Stanford et al., 1975) and soil moisture content of 60% water holding capacity (Pal and Broadbent, 1975) have been reported as being favourable for microbial mineralisation and therefore residue decomposition. Microbial access to the residue also affects the decomposition rate, with surface applied residue exposed to stronger fluctuations in temperature and moisture compared with residue incorporated into soil (Kumar and Goh, 2000b; Cogle et al., 1989). Crop-specific data (e.g., the quality of each plant component and its contribution to the residue N pool) are needed to apply these decomposition concepts to a white clover seed crop to predict the timing of N release post-harvest across different landscapes.

The amount of N within the aboveground biomass (AGB) of a white clover seed crop has been reported as 270-kg N/ha (Kumar and Goh, 2000a). This serves as a source of N to the following crop in the rotation. However, to reduce asynchrony between soil N supply and plant demand, an alignment between residue N release and ensuing crop uptake is needed (Crews and Peoples, 2005). Incorporation of white clover green manure, or lamina and petiole material, has resulted in N disappearance from plant material and detection in soil N pools within 3 months of application (Hauggaard-Nielsen et al., 1998; Shi, 2013; McCurdy et al., 2013). The rate of residue-N loss was faster in warmer summer soil conditions that were conducive to microbial decomposition compared with a sevenfold slower rate of loss during winter (McCurdy et al., 2013). Recovery of clover N in the ensuing crop biomass has been reported to range from 10% to 80% and 39% to 70% retained in the soil (Kumar and Goh, 2000b; Ladd and Amato, 1986). Pathways for clover N loss from the soil, when in residue form, have been reported to occur via ammonia volatilisation, nitrous oxide emission, dissolved organic N, and inorganic N transport below 1 m depth (Kumar, 2000; Van der Weerden et al., 1999; Rasmussen et al., 2008; Kušlienė et al., 2013; Adams-Pattinson and Pattinson, 1985). There are concerns that legume monocultures increase the risk of nitrate leaching when crops are terminated and the paddock is left fallow (Vogeler et al., 2019; Lilburne et al., 2003; Cookson et al., 2002). This risk persists into winter, with clover residue mineralisation reported to continue at incubation temperatures <5°C (Cookson et al., 2002). However, net immobilisation of mineralised residue N into the microbial biomass was also observed under winter conditions (Cookson et al., 2002), which emphasises the importance of synchronising N supply with crop demand to avoid winter N leaching. Hence, the amount of biomass-N grown, the time of residue N release post-harvest, and its utilisation in the plant–soil system, in the context of a legume seed crop, requires quantification through a field-scale N balance.

This study aimed to address the aforementioned knowledge gaps by quantifying the legacy effect of N, accumulated in white clover biomass during seed production, on soil and plant N pools during the growth of the subsequent crop in the rotation. Thus, the objective of this research was to quantify the amount and timing of N release from clover seed crop residues and N uptake by the ensuing crop. To do this, two field experiments were established, of white clover monocultures with different post-harvest management strategies, that represented current on-farm practices. It was hypothesised that (1) the canopy management treatment would have no effect on the quantity or time of N release from white clover seed crops, (2) the thermal time required for residue N release would be consistent with other residues of similar quality, and (3) the plant N uptake by the ensuing crop in the rotation will reduce estimated N losses from the plant–soil system, compared with a fallow treatment.

2 Materials and methods

2.1 Experimental layout

Two experiments were situated in commercial seed crops of the white clover cultivar "Romena", grown to produce seed during the 2022/23 season. The experiments were in two 5-ha paddocks at PGG Wrightson Seeds Kimihia Research Site in Canterbury, New Zealand (43°37′10.0″ S 172°30′03.6″ E). The dominant soil type of the area is an immature pallic sedimentary soil with a silt texture, according to New Zealand Soil Classification (Hewitt, 2010). The commercial grower had established one paddock via conventional methods (ploughing and tillage) and the other via direct drilling. The experiments within each of these paddocks are subsequently referred to as the “Conventional” and “Direct Drill” experiments, respectively. The experiments are considered as two discrete experiments for statistical purposes to avoid pseudo replication. Both paddocks were sown on 25 March 2022 in 30-cm row spacing at a sowing rate of 3-kg/ha. Experimental areas for the Conventional and Direct-Drill experiments occupied a 24 m × 25 m area in each paddock, with plots placed uniformly on either side of a tramline (consistent pathway for tractors). For each experiment, there were eight plots, 25-m long (parallel with clover drill rows) and 2.5-m wide, which gave four replicates.

The Conventional and Direct-Drill experiments both had two treatments: an unsprayed control (“control”) and a chemical canopy management treatment (“sprayed”) that received 1.5 L/ha of Argosy® (25 g/L diflufenican and 250 g/L bromoxynil) and 1.5 L/ha of Relay® Super S (680 g/L 2,4-D ester), respectively. The sprayed treatment mirrored the management of the commercial crop in the host paddock and was applied on 7 September 2022. The impact of the herbicide treatment was followed during the present study to determine the effect on biomass production, seed yield, and potential effects on soil N (Anderson et al., 2024a). The Conventional and Direct Drill experiments were desiccated twice for seed harvest, on 17 and 24 January 2023, with Reglone® (200 g/L diquat) at 3 L/ha per application. Seed harvest, with a specialised plot combine, was completed on 30 January 2023, with harvest residue spread evenly across each plot width.

2.2 Legacy experiment

After seed harvest, the experiments were continued in their original areas. The commercial grower’s crop rotation resulted in the Conventional white clover experiment being replaced with a kale (Brassica oleracea spp. Acephala L.) seed crop and the Direct Drill experiment being taken for a second year of white clover seed production. The kale was direct drilled counter-diagonally at 4-kg/ha in 50-cm row spacings on 3 February 2023. To assess plant N uptake, fallow subplots were imposed using Buster® (200 g/L glufosinate-ammonium) at a rate of 10 L/ha on 16 March 2023 to kill all vegetation. The subplots were 2.5 m × 2.5 m and placed randomly within each plot. These fallow areas are referred to as “Exclusion Zones”.

In the Conventional experiment, the white clover regrew after seed harvest until 24 March 2023, whereupon the clover regrowth was selectively terminated with Versatill™ PowerFlo™ (600 g/L Clopyralid) at 0.5 L/ha. Inorganic N was added to the Conventional experiment as Cropzeal® Boron Boost at 250-kg/ha (46-kg N/ha) on 2 February, and as SustaiN® at 100-kg/ha (46-kg N/ha) on 12 and 20 April 2023. Further details of the agronomic management of both experiments can be found in Supplementary Table 1 and Supplementary Table 2.

2.3 Measurements

A summary of the soil nutrient profile and soil bulk density of each experimental site is reported in Supplementary Table 3. A climate summary of the experimental period is given in Supplementary Table 4.

2.3.1 Soil

Soil cores of 2.2 cm diameter and 0–10-, 10–20-, and 20–40-cm depth increments were taken from the exclusion zones and the wider plots to determine mineral N content (NH₄⁺ and NO₃⁻). Sampling occurred weekly for 75 days post-harvest from 21 February until 6 April 2023. Following this, sampling occurred fortnightly until 25 September 2023, except for the month of July where, due to soil temperatures <10°C, only one soil sampling occurred (Lloyd and Taylor, 1994). Mineral N was determined following the methods of Blakemore et al. (1987) and Clough et al. (2001). Briefly, 4 g of fresh soil was extracted with 40 mL of 2-M potassium chloride (KCl) and aliquots measured via flow injection analysis using a FIAstar 5000 twin channel analyser (FOSS Analytical, Hillerød, Denmark).

Total soil N (%) was determined via the Dumas combustion method using a Rapid N Exceed (Elementar Analysis System, Langenselbold, Germany). This was measured in samples from 0–10-, 10–20-, and 20–40-cm depth increments taken on 26 January 2023, following crop desiccation, and at the end of measurements on 25 September 2023, 238 days post-harvest.

2.3.2 Plant

The N contents (%) of the remaining white clover harvest residue and ensuing crop biomass were determined monthly. The aboveground and belowground white clover residues were removed from a 30-cm-long, 15-cm-wide, and 20-cm-deep area encompassing one historic drill row of the seed crop. The biomass of the ensuing crop (kale or second-year white clover) was then removed from this area. No residue samples were taken from exclusion zones. Samples were washed over a fine mesh sieve to remove soil contamination and then dried at 65°C for at least 48 h to determine dry matter content. Dried samples were passed through a Retsch ZM 200 mill (Retsch GmbH, Haan, Germany) fitted with a 2-mm sieve and then analysed for crude protein content (CP%) using near-infrared spectroscopy (NIRS DS 2500 analyser, FOSS Analytical, Hillerød, Denmark). A factor of 6.25 was used to convert CP% to N% (Féret et al., 2021).

2.3.3 White clover component contributions to residue N

The N contents (%) of the aboveground biomass (AGB) components at harvest time were determined from a 0.36 m² quadrat cut from each plot. A subsample, taken as 20% of the quadrat sample, was segmented into lamina, petiole, stolon, peduncle, and flower head. Seeds were separated from the flower heads by hand threshing on ridged rubber mats. The remaining flower head material was referred to as “florets”, and N% was determined via NIRS as with the other components. Seed samples were processed through a TissueLyser II (Qiagen, Dusseldorf, Germany), and then the N content was determined via Dumas combustion as described above. Lamina, petiole, stolon, peduncle, and floret components were ground and analysed for N content by NIRS. The crown and taproots of five individual plants per plot were removed, by augering, washed and dried, and then processed for root and crown N%. Root and crown biomass were estimated based on relationships with AGB established during measurements from the previous clover seed production period of the experiments (Anderson, 2025). Linear regressions were fitted to the data, and the root:AGB ratio was estimated using Equation 1 (R² = 0.94), and the crown:AGB ratio was estimated using Equation 2 (R² = 0.79), where x is AGB (kg DM/ha).

$\begin{array}{ll}Root:ABG Ratio=(4.8\times {10}^5)x^{-1.98}& (1)\end{array}$

$\begin{array}{ll}Crown:ABG Ratio=(1.0\times {10}^8)x^{-2.77}& (2)\end{array}$

2.3.4 Timepoints for N pool comparisons

The quantity of N at the beginning of the experiment, “Start N”, refers to the sum of all measured N pools at the time of seed harvest. These pools included white clover biomass below the cutting height of the combine harvester, the combine offal residue, and soil mineral N in the 0-10-, 10-20-, and 20-40-cm depths. When the white clover was terminated, referred to as “Termination N”, the quantity of N present was the sum of N remaining in combine offal residue, the white clover regrowth since harvest and its belowground components, the subsequent crop, and soil mineral N to 40-cm depth. At the end of the experiment, referred to as “Final N”, the quantity of N was the sum of N remaining in white clover residue, the subsequent crop and soil mineral N to 40-cm depth.

2.4 Thermal time

Thermal time was taken as the mean of the daily maximum and minimum air temperature (°C) minus the base temperature (0°C), summed over a given period. Air temperature data were measured at each experiment or taken from Broadfield’s weather station (43°37′29.2"S, 172°28′1.95"E).

2.5 Prediction of soil N mineralisation

During data analysis following the completion of the field experiments, it was decided that an estimate of the daily soil N mineralisation rate was needed. Both experiments were on the same soil type. Thus, the Direct Drill experiment was selected due to the site being ~100 days after white clover seed harvest. Daily soil N mineralisation was estimated using the acetylene (C₂H₂) reduction assay method detailed in Hatch et al. (1990); Jarvis et al. (1996), and Jarvis et al. (2001). Briefly, soil cores of 2.2-cm diameter × 10-cm depth were taken randomly across the Direct Drill paddock on 20 May 2024. Four of these cores (30% SMC) were sealed in a 500-mL glass jar, and 2% v/v of the headspace (including air-filled soil pore volume) was replaced with C₂H₂ (10 cm³ per jar) through a rubber septa fitted to the lid. Five incubation temperatures were selected to represent the range of 0–10-cm soil temperatures measured between January and September 2023: 5°C, 10°C, 15°C, 20°C, and 25°C. There were three jar replicates per incubation temperature. The difference in soil mineral N, determined using 2-M KCl extraction, before and after 14 days of incubation in the presence of C₂H₂ nitrification inhibition, equated to net mineralisation of NH₄⁺. An asymptotic relationship was fitted to the data (incubation temperature by daily mineralisation rate). Soil temperature had been measured hourly in the 0–10-cm soil depth of each treatment in both experiments using CS650 time domain reflectometry probes (Campbell Scientific, Utah, USA). Field temperature data were applied to the asymptotic incubation relationship to estimate daily soil N mineralisation in both experiments during the residue decomposition phase.

2.6 Estimation of N leaching—pore volume solute transport

A soil water balance was used to estimate the potential leaching of NO₃⁻. Porosities of the 0–10-, 10–20-, and 20–40-cm layers were calculated using soil bulk density (ρ_b) of the given soil depth (Supplementary Table 3) and soil particle density (ρ_p), which was assumed to be 2.65 g/cm³. The pore volume (cm³) of each soil depth was given by multiplying porosity by the volume of each soil depth based on a 1-m² area. Total pore volume of the soil profile to 40 cm was the sum of the pore volumes from each depth. Initial soil moisture, measured as volumetric water content (ϴ_v), was used to determine the initial water-filled pore space. The pore volume equivalents of daily millimeter of evapotranspiration lost and rainfall received were used to calculate the daily net-change in water-filled pore space.

2.7 Estimation of initial 30 days of residue mineralisation

An incubator experiment that replicated the soil moisture and rainfall conditions measured in the field was completed to determine the rate of residue disappearance during the first 4 weeks after harvest. Soil temperature at the 0–10-cm depth during this period averaged 18.5 (± 2°C, SD); thus, samples were incubated at 20°C. During the first 4 weeks after harvest, the plots received 0, 5, 25, and 56 mm of rainfall, respectively. Dried and ground (2 mm) Conventional clover AGB samples, taken on 16 January 2023, were placed in emptied 12 cm × 9 cm teabags (Dilmah Ceylon Tea Company, PLC) at a rate of 6.3 g DM/bag, equivalent to the 9,500-kg DM/ha loading of plant material at harvest time. There were twelve 500-mL glass jars that were each filled with 300 g of 4-mm sieved soil retained from the Conventional experiment. One teabag was placed on top of the soil in each jar and added to the incubator (without lids). Three replicates were assigned to each incubation period, 1, 2, 3, or 4 weeks (Table 1). On day 1 of each week, deionised water was applied using a syringe fitted with a 24G 0.55 mm × 25 mm needle positioned above the jar using a clamp. Water was fed through under gravity at a rate of 30 mL/h. After incubation, the teabags were removed and the clover residue extracted. Residue was dried at 65°C, weighed, and the N content measured using NIRS. The rate of residue disappearance was calculated from a linear regression of residue mass remaining versus soil thermal time. Soil thermal time was calculated as noted above with a base temperature (T_b) of 0°C and was thus 20°Cd per day of incubation. The proportion of residue N remaining, according to thermal time, was given as Equation 3, where y is the amount of residue N mass remaining (%) and x is thermal time (°Cd). This was used to predict the amount of residue N remaining in harvest residue between harvest and termination in both Conventional and Direct Drill experiments.

Table 1

Table 1. The incubation treatments of Conventional clover residue samples incubated at 20 °C for allotted periods (days incubated) and water applied (mL) at a rate equivalent to that received as rainfall at the experimental site between 30 January and 2 March 2023.

$\begin{array}{ll}y=-0.11x+107& (3)\end{array}$

2.8 Statistical analysis

Statistical analyses were performed using Genstat® 22 (VSN International Ltd, 2022). Linear or exponential curves were fitted to individual plot data. Treatment means were taken from the replicates and compared using analysis of variance (ANOVA) and a significance level of α = 0.05. All data were tested for normality, homogeneity of variance, and independence. No data transformation was required during analysis. Graphs were created using SigmaPlot® 15 (Grafiti LLC 2022). The contribution of each N pool at discrete timepoints was predicted using a log-linear generalised linear model due to measurements being conducted on different dates. This allowed for a comparison between pools, treatment and time based on predicted values, their associated errors, and the confidence interval of those predicted means. Each pool was fitted with a binomial model (Poisson distribution) to examine the proportion of N per pool out of the total N. Treatment and time were then combined into separate models to allow contrast calculations between time and plot and produce chi-square prediction P-values to assess the relationship significance.

3 Results

3.1 Quantity of white clover N

In the Conventional experiment, prior to harvest desiccation (16 January 2023), the total AGB N yield was 228-kg N/ha (± 17; numbers in brackets represent SEM) across the control and sprayed treatments (P = 0.53). This equated to 28-kg N/t DM (± 0.5). At harvest, the C:N ratio of the AGB averaged 25.6 (± 0.8) and did not differ between treatments (P = 0.75). Flower heads, separated into seed and floret material, contained 5.33% N (± 0.10) and 2.09% N (± 0.1), respectively. The floret components contributed 11% (± 0.02) and 15% (0.04) of residue-N mass in the control and sprayed treatments, respectively (P = 0.02). Export of seed N accounted for 67-kg N/ha (± 3) with the remaining combine offal residue-N averaging 125-kg N/ha (± 10), which was not different between treatments (P = 0.47). The lamina, petiole, peduncle, and dead material components in the residue, contributed 42% (± 1.4), 21% (± 0.8), 11% (± 0.7), and 12% (± 1.4) of the residue-N, respectively (Table 2).

Table 2

Table 2. The N concentration (%) of the white clover above ground biomass (AGB) components immediately prior to harvest desiccation and their contribution to AGB N (kg N/ha), averaged across control and sprayed treatments, in Conventional and Direct Drill seed crops grown in Canterbury, New Zealand.

In the Direct Drill experiment, 1 day following harvest desiccation (20 January 2023), total AGB N yield averaged 230-kg N/ha (± 11) across control and sprayed treatments (P = 0.10). Thus, the clover crops contained, on average, 27-kg N/t DM (± 0.4). At harvest, the AGB C:N ratio averaged 22.2 (± 1.2) and did not differ between treatments (P = 0.33). Flower heads, separated into seed and floret material, contained 5.03% N (± 0.1) and 2.09% N (± 0.1), respectively. After 19-kg N/ha (± 3) of seed N was exported, the N content of combine offal residue-N averaged 125-kg N/ha (± 7) between treatments (P = 0.49). The lamina, petiole, peduncle, floret, and dead material components contributed 44% (± 3), 28% (± 1), 12% (± 1.5), 9% (± 1.5), and 8% (± 1.5) to the residue-N mass, respectively (Table 2).

3.2 Timing of residue N release

In the Conventional experiment, the decomposition of clover residue comprised two distinct periods: between harvest and termination, when clover components below combine cutting height remained intact and were contributing to clover regrowth, and between termination and final measurements (Figure 1). Following termination, N embodied in the clover was added to the residue-N pool that already contained the combine offal residue-N following harvest. The AGB regrowth after harvest contained an average of 72-kg N/ha (± 8) between treatments (P = 0.16). The total clover N added to the residue-N pool following termination averaged 103-kg N/ha (± 6), including 12-kg N/ha of root and crown N. At termination, the control residue-N had lost 84% (± 5) of its N content since harvest, whereas sprayed residue-N had lost only 63% (± 6) of its N content (P < 0.001). The time taken to 95% decomposition did not differ between treatments (P = 0.38) and averaged 1,520°Cd (± 46) after termination (2,240°Cd after harvest).

Figure 1

Figure 1. The decline of residue N (kg N/ha) in white clover versus thermal time since harvest (°Cd) in Conventional and Direct Drill seed crops grown in Canterbury, New Zealand. The equations refer to the regressions fitted to their respective datapoints and time periods, where 0°Cd is seed harvest. The black arrow on the Conventional graph indicates the time of crop termination.

The decomposition of clover residue comprised only one period in the Direct Drill experiment because the clover was taken for a second season of seed production (Figure 1). The time taken to 95% decomposition did not differ between treatments (P = 0.86) and averaged 1,670°Cd (± 54). When the Direct Drill exclusion zones were created with Buster® on 16 March 2023, the clover in those areas was terminated. Therefore, the residue-N pool in the subplots received, on average, 123-kg N/ha (± 16) of total plant N, derived from regrowth, at termination time (P = 0.51). This was added to an average of 48-kg N/ha (± 12) of remaining combine offal, which had lost, on average, 57% (± 0.1) of its N content since harvest.

In the Conventional experiment, residue-N followed a weak exponential decline between harvest and termination, according to thermal time since harvest. Thereafter, the residue followed an asymptotic manner of decay. Residue-N declined in a similar asymptotic manner in the Direct Drill experiment.

3.3 Utilisation of residue N in the plant–soil system

In the Conventional plant–soil system, following seed export, there was an average of 198-kg N/ha (± 18) of Start N present in both treatments (Figure 2A). For the control treatment, Start N was composed of 62% as residue-N, 2% as soil mineral N (0–40 cm depth), and 36% as living clover material. The sprayed treatment had 65% as residue-N, 2% as soil mineral N (0–40-cm), and 33% in living clover biomass. Then, 45 days after seed harvest (16 March 2023, 756°Cd), at crop termination, there was an average of 188-kg N/ha (± 12) of N in both treatments. The regrowth of clover contributed 54% and 38% of N present in the control and sprayed treatments at this time, respectively. Upon termination, this regrowth N was added to the residue-N pool that equated to 12% and 27% of N present at this time. Soil mineral N had increased by an average of 46-kg N/ha (0–40 cm) since harvest: This increased soil mineral N contribution equated to 31% and 33% of N present in the control and sprayed plots, respectively. Kale plants had taken up an average of 5-kg N/ha (± 1), thus contributing an average 3% of N present at termination. By the following spring, 238 days after harvest (2,590°Cd), 226- and 327-kg N/ha were accounted for in the control and sprayed treatments, respectively (Final N). Remaining residue-N, having surpassed 95% degradation, was negligible and contributed 3% of Final N. In the control and sprayed plots, soil mineral N contributed 10% and 16% of the Final N, 80% of which was at 0–10-cm depth. The kale biomass had accumulated 195 (± 39) and 270-kg N/ha (± 55) in the control and sprayed treatments (P = 0.11) and contributed 86% and 82% of N present at the end of the experiment, respectively. Nitrogen fertiliser added 134-kg N/ha to the system within 80 days after seed harvest (1 May 2023).

Figure 2

Figure 2. Nitrogen (kg N/ha) in N pools at seed harvest (Start N), clover regrowth termination (Termination N), and at final measurement 238 days after seed harvest (25/09/2023, Final N) in Conventional (A) and Direct Drill (B) experiments grown in Canterbury, New Zealand. The drop lines in the Conventional (A) graph indicate the addition of fertiliser-N (kg N/ha) between these time points.

In the Direct Drill plant–soil system, the Start-N equated to an average of 227-kg N/ha (± 11) in both treatments (Figure 2B). This was composed of 55% as residue-N, 7% in the top 40 cm of soil as mineral N, and 38% within living clover biomass. Termination N equated to 252- and 196-kg N/ha in the control and sprayed treatments, respectively. Harvest residue contributed 27% in the control treatment and 14% in the sprayed treatment; clover regrowth had taken up 46- and 25-kg N/ha since seed harvest in the control and sprayed treatments, giving a clover N pool of 136 (± 24) and 111-kg N/ha (± 23), respectively. Clover N contribution to Termination N was 54% and 57% in the control and sprayed treatments, respectively. Soil mineral N had also increased by ~30-kg N/ha (0–40 cm), increasing its contribution to 20% and 29% in the control and sprayed treatments, respectively. Due to the herbicide applications involved in the crop’s agronomy (Supplementary Table 2), clover N accumulation was frequently interrupted. Peak clover N content was measured 119 days after seed harvest (29 May 2023) and averaged 188-kg N/ha (± 83) between treatments. Final N in the Direct Drill system (2,590°Cd, 239 days after harvest) averaged 99-kg N/ha in the control and sprayed treatments, respectively. Remaining residue-N was >95% degraded and made a negligible contribution (2.8%) to Final N. Soil mineral N averaged 32-kg N/ha (0–40 cm) and contributed 33% to Final N in both treatments. The N embodied in the clover (above and below ground) equated to 65% of Final N, with the clover biomass containing 64-kg N/ha (± 7), with no difference between treatments. This left approximately 125-kg N/ha (± 30), or 50% and 64% of Termination N in control and sprayed treatments, respectively, unaccounted for between day 45 (16 March 2023) and day 238 (25 September 2023) after seed harvest.

3.4 Total soil N

In the Conventional experiment, total soil N increased (P = 0.007) from 0.16% N (± 0.01) at seed harvest to 0.18% N (± 0.01) at final measurement, 238 days after seed harvest, in the 0–10-cm depth but was not different between treatments (P = 0.88). Under kale growth, total soil N increased by 226-kg N/ha (± 68), which was 188-kg N/ha more than the 38-kg N/ha (± 61) increase under the exclusion zones (P = 0.04). In the 10–20-cm depth, there was no increase in total soil N (P = 0.65) under kale and there was no difference when compared with the exclusion zones (P = 0.95), averaging 0.16% N (± 0.01). The 20–40-cm depth had an increased N concentration under kale (P = 0.06) of 303-kg N/ha (± 168), but this did not differ between treatments (P = 0.60) or exclusion zones (P = 0.68) and averaged 0.07% N (± 0.01).

In the Direct Drill experiment, total soil N also increased (P<0.001) in the 0–10-cm depth from 0.21% (± 0.001) N at seed harvest to 0.23% N (± 0.001) at the end of the experiment, but this did not differ between treatments (P = 0.79). Under clover growth, total soil N increased (P<0.001) in the 0–10-cm depth by 232-kg N/ha (± 31), which was greater (P = 0.04) than the accumulation under the exclusion zones of 126-kg N/ha (± 28). There was no increase in total soil N (P = 0.53) in the 10–20-cm depth under clover, which averaged 0.20% N (± 0.01), and there was no difference between the clover and the exclusion zone treatments (P = 0.09). There was no increase in total soil N in the 20-40-cm depth (P = 0.26) under clover or the exclusion zone (P = 0.91), which averaged 0.09% N (± 0.01).

3.5 Soil N mineralisation

The C₂H₂ reduction assay completed on soil from the Direct Drill experiment enabled estimation of the daily soil N mineralisation rate (Equation 4), according to soil temperature (x, °C).

$\begin{array}{ll}{C}_2{H}_2 Daily mineralisation (kg N{H}_4^{+}-N h{a}^{-1})=0.109+0.205({1.112}^x)& (4)\end{array}$

Increasing soil temperature did not affect the mineralisation rate until incubation temperature exceeded 15°C (P<0.001), and averaged 0.69-kg NH₄-N/ha/day (± 0.1) between 5°C and 15°C. At 20°Cd and 25°Cd, daily mineralisation averaged 2.13 (± 0.3) and 2.98-kg NH₄-N/ha/day (± 0.4), respectively. In the Conventional and Direct Drill experiments, the total net mineralised N estimated for the measurement period was 217- and 222-kg NH₄-N/ha over the 0–10-cm depth.

3.6 Estimation of N leaching losses

In the Conventional experiment, soil mineral N (Figure 3B) in the exclusion zones began to decline in May, after harvest. One pore volume of drainage had occurred by day 105 since seed harvest (15 May 2023), with a second pore volume of drainage by day 172 (21 July 2023, Figure 3A), after 449 and 625 mm of cumulative rainfall since harvest, respectively. Although not demonstrated in Figure 3B, prior to the first pore volume of drainage, soil mineral N in the 20–40-cm depth increased whereas in the 0–10- and 10–20-cm depths it decreased. Between 2 May and 8 August, soil mineral N declined by an average of 98-kg N/ha (± 6) under exclusion zones. This indicated possible transport of N down the profile and beyond the 40-cm depth of measurement. The decrease in soil mineral N in the Conventional plots coincided with the uptake of N by kale plants, which by the time of the first pore volume of drainage had taken up an average of 89-kg N/ha (± 8) and 151-kg N/ha (± 14) by the time of the second pore volume of drainage.

Figure 3

Figure 3. Cumulative pore volumes in the Conventional experiment (A) and soil mineral N over 0–40-cm depth (B) of the wider plot growing kale and under exclusion zones (ExZone), and the Direct Drill cumulative pore volume (C) and soil mineral N (D) between seed harvest on 30 January 2023 and final measurements on 25 September 2023 in Canterbury, New Zealand. The drop lines indicate the time of the first and second pore volumes of drainage per experiment.

In the Direct Drill experiment, the first pore volume of drainage occurred on day 106 since seed harvest (16 May 2023). The second pore volume of drainage was on day 172 since seed harvest (Figure 3C). Soil mineral N decreased in the 0–10- and 10–20-cm depths while increasing in the 20–40-cm depth (Figure 3D). Between 2 May and 8 August, the decline in inorganic-N averaged 78-kg N/ha (± 10) in the exclusion zones.

4 Discussion

This study aimed to quantify the legacy effect of N, accumulated in white clover biomass during seed production, on soil and plant N pools during the growth of the subsequent crop in the rotation. Nitrogen release from Conventional and Direct Drill clover residue, alongside ensuing crop N uptake, was monitored for 238 days after seed harvest to understand the quantity, timing of release, and utilisation of white clover N within the cropping rotation.

4.1 Quantity of N released from white clover residue

Both the Conventional and Direct Drill experiments had accumulated the same amount of N within white clover biomass by seed harvest time, which indicated that crop establishment method did not affect this parameter. Statistical comparison between these methods is prevented due to establishment methods being limited to one experiment each. Final clover biomass did not differ between the control and sprayed treatments in either experiment due to plasticity in plant growth allowing for recovery from the effects of herbicide application earlier in the season (Anderson et al., 2024a). The N concentration of each component was also unaffected by canopy management and was consistent across experiments. Therefore, canopy management did not affect the final N content of the crops at seed harvest time, despite the impacts on flowering and seed production (Anderson et al., 2024b). Less seed N was exported in the Direct Drill experiment due to an unplanned early desiccation for harvest, which resulted in similar quantities of combine offal residue N in both experiments. Had the Direct Drill crop reached full seed maturity, the mass of N removed may have been greater, leaving less N behind in clover residue. Both experiments demonstrated the white clover’s capacity to produce a similar amount of N per unit of biomass grown. These findings align closely with previous studies that also showed a similar amount of N grown per unit of biomass, regardless of clover management or intended end use (Kumar and Goh, 2000a; Lucas et al., 2010; Widdup et al., 2001).

4.2 Timing of N release from white clover residue

The rapid depletion of combine offal within 43 days of harvest (756°Cd) indicated that the clover components comprising the combine offal residue were readily accessible and of a labile form for microbial decomposition to occur. Lamina, petiole, and floret material contributed an average of 78% of combine offal N, and 43 days after harvest, a convergent amount of N had disappeared from the residue. Residue quality affects its microbial availability (Knapp et al., 1983; Harper and Lynch, 1981; Collins et al., 1990), and the different half-lives of each component produced the asymptotic decomposition of residue observed (Kumar and Goh, 2000b). The white clover’s composition of components produced a phased, asymptotic breakdown. The readily soluble material, such as lamina and petiole with high N content and low lignin, are the first to degrade (Douglas and Rickman, 1992). The more structural components that are higher in lignin or associated with carbohydrate storage, such as the peduncle and stolon, are expected to degrade more slowly. Similar trends were observed for roots of Fabaceae species, where decomposition was faster with increasing N content and a higher proportion of soluble carbohydrates (Aulen et al., 2012; Birouste et al., 2012).

The low C:N ratio of AGB at harvest meant that 95% disappearance of residue, including stolon and peduncle, was achieved within 1,560°Cd–1,670°Cd after the material was added to the residue-N pool. Singh et al. (2020) reported that hairy vetch (Vicia villosa), with a C:N ratio <10, had 80% mineralisation of residue-N within 800°Cd of termination. The C:N ratio of residue material affects the mineralisation rate, where a C:N ratio of <20–30 enhances the rate of nutrient release due to reduced competition for available N, maintaining high microbial activity for longer (Janzen and Kucey, 1988). Reporting the N disappearance from residue according to thermal time has broader applications than reporting it according to calendar days. As residue breakdown is a microbially controlled process that is affected by soil temperature, mineralisation rates are lower during winter than summer. McCurdy et al. (2013) observed 50% degradation of white clover green manure after 11 summer days whereas it took 74 winter days to achieve a similar result. Calendar days are perhaps more approachable for a grower, particularly without temperature sensing equipment, but consistent relationships between crop residue disappearance and thermal time have been given across a range of crop species, systems, and regions globally (Douglas and Rickman, 1992; Honeycutt and Potaro, 1990). Therefore, the white clover residue N disappearance relationships given in this research align with previous research and could be applied to long-term climate averages in any given crop production area to generally estimate the rate of N input into the soil in any given season. Other edaphic factors that affect microbial activity, such as soil pH, moisture, and residue placement, also impact mineralisation rates and need to be considered (Olewski et al., 2011; Pal and Broadbent, 1975; Kumar and Goh, 2000b; Cogle et al., 1989).

4.3 Utilisation of N in the plant–soil system

Ideally, the 233-kg N/ha (± 34) measured in kale biomass at the end of measurements in the Conventional experiment originated from the 228-kg N/ha added as white clover residue following harvest and termination. However, confounding effects from fertiliser application and soil N mineralisation must be considered. Without insight from ¹⁵N-labelled residue or fertiliser, it is difficult to confirm the effect of N fertiliser addition on soil mineral N and associated N pools. Conclusions are drawn with caution based on the data available. Between harvest and clover termination in the Conventional experiment, the 46-kg N/ha increase in soil mineral N to 40-cm depth could have come from the fertiliser application, the prompt release of residue N, or mineralisation of preexisting soil organic matter. As the clover had a C:N ratio of <25, rapid mineralisation of residue N and incorporation into the microbial biomass is expected. This means that fertiliser N was probably the primary contributor to soil mineral N between harvest and termination, reducing system demand for N mineralisation. Chakwizira et al. (2015) presented kale N uptake data that suggested >90% recovery of N from urea applications ≤100-kg N/ha, in addition to ~88-kg N/ha background soil mineral N, in kale growth 53 days after N was applied and 111 days after kale was sown. This would suggest that Conventional kale was utilising both the fertiliser N that was applied within 69 days of sowing, and preexisting soil mineral N. However, plant demand for N was <50-kg N/ha within 75 days after sowing, giving the observed increase in soil mineral N during this time (Figure 3B).

Nitrogen fertiliser applied after clover termination was likely taken up by kale, with measurements in May showing that kale N uptake equalled fertiliser N applied and soil mineral N was <20-kg N/ha to 40-cm depth. Therefore, the delivery of residue N via mineralisation from organic forms was likely to begin in the first month of spring. This was evident in the rapid assimilation of 110-kg N/ha in kale biomass paired with a 10–30-kg N/ha accumulation of soil mineral N in the 0–10-cm depth in September (Figure 3B). The C₂H₂ incubations demonstrated that mineralisation was lower when soil temperatures were below 10 °C. Therefore, in spring, with increasing soil temperatures, soil N mineralisation was likely the main contributor to the flush of mineral N. This aligns with Groffman et al. (1987) and suggests that a grower should expect the delivery of residue N in plant available form in the spring following harvest, despite the N disappearing from residue in autumn. Further evidence that the contribution of the clover seed crop legacy N is a longer-term resource given by Kumar et al. (2001), who reported that only 65% of N from a clover seed crop was recovered in plant uptake and soil solution within 1 year of harvest. The majority of N was likely embodied within the soil microbial biomass and available for release when conditions were favourable later in the cropping cycle (Crews and Peoples, 2005; Kumar et al., 2001). The effect of N fertiliser addition was confirmed in the soil mineral N pool within the Direct Drill experiment. No N fertiliser was applied, and subsequently, the harvest-termination increase in soil mineral N was lower than observed in the Conventional experiment. The flush of mineral N in September was also observed in the Direct Drill experiment, suggesting that this was the result of residue-N mineralisation. Any increase in soil mineral N prior to winter was likely a result of soil N mineralisation of preexisting organic N, with net N mineralisation favoured due to a lack of disruption from fertiliser N.

4.4 Leaching risk during the residue decomposition phase

Total soil N data gave an insight into whether the soil system captured the release of residue-N. An increase in total soil N, as measured in the 0–40-cm depth of soil, indicates an addition of N to the system that may not have been detected in soil net-mineral N or captured by plant biomass. Between harvest and final measurements at the end of September, both Conventional and Direct Drill experiments had an increase in total N for the 0–10-cm soil depth that was equivalent to the amount of N added in residue. However, this was only under actively growing plants: Total N increase under exclusion zones was on average 16% and 54% of that under kale and clover, respectively. The enhanced N capture under actively growing plants may be due to plant stimulation of the microbial biomass through root activity. Dhungana et al. (2023) demonstrated that root exudates sustained soil microbiota and their activity, particularly when species-specific host plants were present for longer in the system. This favours a specialised, complimentary community that is capable of rapidly decomposing plant material (Ayres et al., 2009). The greater N capture in the Direct Drill exclusion zones may have been due to carbon availability, with greater organic C availability increasing the capacity for microbial N immobilisation (Barret and Burke, 2000). Of the N not captured under exclusion zones, it is speculated that the N unaccounted for was transported below measurement depth in the soil. It is estimated that complete solute transport in a given depth requires two full pore volumes of drainage (Cameron and Haynes, 1986). Without direct measurement methods of the soil in-situ, the findings of this study are limited to the principles of solute transport. As both Conventional and Direct Drill soils received sufficient rainfall for two pore volumes of drainage, and soil mineral N decreased with increasing cumulative pore volumes under the exclusion zones, it indicates that inorganic-N had moved to below 40 cm. Francis et al. (1995) demonstrated that fallow periods after pasture termination prior to winter increased the potential for mineralisation of soil organic N and subsequently made N vulnerable to leaching, whereas active plant N uptake reduced leaching losses by 60%. This aligned with the findings of this research. However, transport below the 40-cm depth does not necessarily mean the N was leached from the soil profile. The soil types of both experiments have A and B horizons that are more than 1-m deep, combined (Manaaki, 2019a, Manaaki, 2019b, Manaaki, 2019c). Adams-Pattinson and Pattinson (1985) measured the soil mineral N to 1 m depth below an ex-white clover seed crop, and despite over 200-kg NO₃-N accounted for, the majority was located in the upper 60 cm of soil and losses beyond the 1-m depth were minimal within 6 months of harvest. The cumulative N transport during any given season, however, would depend on the amount of rainfall received and existing soil moisture conditions. The findings of the current study would suggest that, given the accumulation of total soil N despite conditions sufficient for leaching to occur, neither kale nor second-year clover lost significant amounts of N to the lower depths of this soil. Both crop species had the capacity to take up available soil N, with kale biomass reported to contain 500-kg N/ha with dry matter yields over 20-t DM/ha (Fletcher and Chakwizira, 2011). After a white clover seed crop, a farmer may also plant a cereal or pasture as required in their crop rotation, with these species expected to have roots that extend below 40-cm depth and take up N available in the soil. Equally, deep-rooted legumes will also preferentially use soil mineral N from throughout the soil profile and therefore reduce the expenditure of photo-assimilates on the exchange of N with rhizobia bacteria via symbiotic N₂ fixation (Russelle et al., 2001; Huang et al., 1975; Pate, 1976).

For the duration of the experiments, particularly over winter when soil temperatures were cooler, soil mineral N did not reach the ~200-kg NH₄⁺-N/ha that was predicted to be mineralised under the exclusion zones of the plots during this period. With leaching conditions in effect, and no depletion of total soil N, it suggests that the majority of predicted net N mineralisation was returned to organic-N forms through immobilisation. Wessels-Perelo et al. (2006) observed microbial biomass N turnover and found, depending on soil conditions and substrate, that 95–323-kg N/ha could be cycled through soil microbiota annually. Consequently, the predicted amount of net N mineralisation may have been accurate but microbial biomass capacity for recycling of this N was strong, resulting in no recovery of this mineralised N in soil solution. Had the system demanded N from soil organic matter, and had there been no alternative source, mineralisation could potentially have met these crop requirements, based on predicted rates. Further research investigating the contribution of soil N mineralisation under crops receiving legume monoculture residue-N inputs is needed without the confounding effects from fertiliser and crop management.

5 Conclusions

Managing the legacy N of a white clover seed crop requires an understanding of both N inputs and outputs within the plant–soil system. Clover biomass contained 28-kg N/t DM/ha, and once the N in seed was exported (5% N), the remaining N was available for decomposition following termination. Combine offal at harvest was readily decomposable with 70% of the embodied N mineralised within 50 days of harvest. However, it appeared that residue N was not plant-available until net soil organic matter (SOM) mineralisation occurred in the following spring. The presence of an actively growing crop post-harvest assisted in capturing residue N: Fallow areas showed that N was transported below 40-cm depth, which may have resulted in N leaching. Under a second-year white clover, seed crop or a kale crop leaching was negligible.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

FA: Visualization, Investigation, Conceptualization, Formal analysis, Writing – original draft. TC: Writing – review & editing, Methodology, Supervision, Conceptualization. MA: Writing – review & editing, Supervision, Conceptualization. DM: Writing – review & editing, Supervision, Conceptualization.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. The doctoral research of FA, a portion of which this paper is derived from, was funded by the Seed Industry Research Centre (SIRC), the Lincoln University Doctoral Scholarship and Lincoln University through departmental funding for PhD students. The Lincoln University Faculty of Agriculture and Life Sciences Postgraduate Publishing Bursary supported FA during preparation of this paper.

Acknowledgments

FA wishes to acknowledge Ruth Butler for assistance with the more complex statistical analyses, Don Heffer for soil sampling equipment maintenance, Elise Dupouy for sampling assistance, and the staff of PGG Wrightson Seeds Kimihia Research Site for hosting the experiments.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fagro.2025.1708560/full#supplementary-material

References

Adams-Pattinson J. A. and Pattinson J. M. (1985). Nitrate leaching losses under a legume-based crop rotation in central Canterbury, New Zealand. New Z. J. Agric. Res. 28, 101–107. doi: 10.1080/00288233.1985.10427002

Crossref Full Text | Google Scholar

Anderson F. C. (2025). The growth, development and nitrogen dynamics of white clover seed crops. Lincoln University, Lincoln, New Zealand.

Google Scholar

Anderson F. C., Andreucci M. P., Clough T. J., and Moot D. J. (2024a). Yield and dry matter accumulation of a white clover seed crop grown under chemical canopy management in Canterbury. Proc. Agron. Soc. New Z. 54.

Google Scholar

Anderson F., Andreucci M., Clough T., and Moot D. (2024b). “Effect of spring canopy management on flowering and seed yield components of a white clover seed crop,” in Adaptive agronomy for a resilient future. Eds. Lawes R., Flower K., Michael P., Mwenda G., and Singh B.(Albany, Western Australia, Australia). Proceedings of the 21st Australian Society of Agronomy Conference, 21–24 October, 2024.

Google Scholar

Aulen M., Shipley B., and Bradley R. (2012). Prediction of in situ root decomposition rates in an interspecific context from chemical and morphological traits. Ann. Bot. 109, 287–297. doi: 10.1093/aob/mcr259

PubMed Abstract | Crossref Full Text | Google Scholar

Ayres E., Steltzer H., Simmons B. L., Simpson R. T., Steinweg J. M., Wallenstein M. D., et al. (2009). Home-field advantage accelerates leaf litter decomposition in forests. Soil Biol. Biochem. 41, 606–610. doi: 10.1016/j.soilbio.2008.12.022

Crossref Full Text | Google Scholar

Barret J. E. and Burke I. C. (2000). Potential nitrogen immobilization in grassland soils across a soil organic matter gradient. Soil Biol. Biochem. 32, 1707–1716. doi: 10.1016/S0038-0717(00)00089-4

Crossref Full Text | Google Scholar

Birouste M., Kazakou E., Blanchard A., and Roumet C. (2012). Plant traits and decomposition: Are the relationships for roots comparable to those for leaves? Ann. Bot. 109, 463–472. doi: 10.1093/aob/mcr297

PubMed Abstract | Crossref Full Text | Google Scholar

Blakemore L. C., Searle P. L., and Daly B. K. (1987). Methods for chemical analysis of soils. New Z. Soil Bureau Sci. Rep. 80, 27–29.

Google Scholar

Cameron K. C. and Haynes R. J. (1986). “Retention and movement of nitrogen in soils,” in Mineral nitrogen in the plant-soil system. Ed. Haynes R. J. (Academic Press Inc, London, UK), 166–241.

Google Scholar

Chakwizira E., de Ruiter J. M., and Maley S. (2015). Effects of nitrogen fertiliser application rate on nitrogen partitioning, nitrogen use efficiency and nutritive value of forage kale. New Z. J. Agric. Res. 58, 259–270. doi: 10.1080/00288233.2015.1016538

Crossref Full Text | Google Scholar

Chakwizira E., Trethewey J., George M., Fletcher A. L., and Rolston M. P. (2011). Effects of spring management techniques on seed yield and yield components of two contrasting white clover varieties. Proc. Agron. Soc. New Z. 41, 133–148.

Google Scholar

Chynoweth R. J., Pyke N. B., Rolston M. P., and Kelly M. (2015). Trends in herbage seed production: 2004-2014. Proc. Agron. Soc. New Z. 45, 47–56. Available online at: https://www.agronomysociety.org.nz/2015-journal-papers.html.

Google Scholar

Clifford P. T. P. (1985). “Effect of leaf area on white clover seed production,” in Producing herbage seeds-Grassland Research and Practice Series No. 2. Eds. Hare M. D. and Brock J. L.(New Zealand Grassland Association, Palmerston North, New Zealand), 37–40.

Google Scholar

Clough T. J., Stevens R. J., Laughlin R. J., Sherlock R. R., and Cameron K. C. (2001). Transformations of inorganic-N in soil leachate under differing storage conditions. Soil Biol. Biochem. 33, 1473–1480. doi: 10.1016/S0038-0717(01)00056-6

Crossref Full Text | Google Scholar

Cogle A. L., Saffigna P. G., and Strong W. M. (1989). Carbon transformations during wheat straw decomposition. Soil Biol. Biochem. 21, 367–372. doi: 10.1016/0038-0717(89)90145-4

Crossref Full Text | Google Scholar

Collins H. P., Elliott L. F., and Papendick R. I. (1990). Wheat straw decomposition and changes in decomposability during field exposure. Soil Sci. Soc. America J. 54, 1013–1016. doi: 10.2136/sssaj1990.03615995005400040013x

Crossref Full Text | Google Scholar

Cookson W. R., Cornforth I. S., and Rowarth J. S. (2002). Winter soil temperature (2-15°C) effects on nitrogen transformations in clover green manure amended or unamended soils; a laboratory and field study. Soil Biol. Biochem. 34, 1401–1415. doi: 10.1016/S0038-0717(02)00083-4

Crossref Full Text | Google Scholar

Coûteaux M. M., Bottner P., and Berg B. (1995). Litter decomposition, climate and litter quality. Trends Ecol. Evol. 10, 63–66. doi: 10.1016/S0169-5347(00)88978-8

PubMed Abstract | Crossref Full Text | Google Scholar

Cowling D. W. (1961). The effect of nitrogenous fertiliser on an established white clover sward. J. Br. Grassland Soc. 16, 65–68. doi: 10.1111/j.1365-2494.1961.tb00213.x

Crossref Full Text | Google Scholar

Crews T. E. and Peoples M. B. (2005). Can the synchrony of nitrogen supply and crop demand be improved in legume and fertilizer-based agroecosystems? A review. Nutrient Cycling Agroecosystems 72, 101–120. doi: 10.1007/s10705-004-6480-1

Crossref Full Text | Google Scholar

Dhungana I., Kantar M. B., and Nguyen N. H. (2023). Root exudate composition from different plant species influences the growth of rhizosphere bacteria. Rhizosphere 25, 100645. doi: 10.1016/j.rhisph.2022.100645

Crossref Full Text | Google Scholar

Douglas C. L. Jr. and Rickman R. W. (1992). Estimating crop residue decomposition from air temperature, initial nitrogen content, and residue placement. Soil Sci. Soc. America J. 56, 272–278. doi: 10.2136/sssaj1992.03615995005600010042x

Crossref Full Text | Google Scholar

Féret J.-B., Berger K., de Boissieu F., and Malenovský Z. (2021). PROSPECT-PRO for estimating content of nitrogen-containing leaf proteins and other carbon-based constituents. Remote Sens. Environ. 252, 112173. doi: 10.1016/j.rse.2020.112173

Crossref Full Text | Google Scholar

Fletcher A. L. and Chakwizira E. (2011). Developing a critical nitrogen dilution curve for forage brassicas. Grass Forage Sci. 67, 13–23. doi: 10.1111/j.1365-2494.2011.00830.x

Crossref Full Text | Google Scholar

Foundation for Arable Research (2021). New Zealand's Arable Industry: The backbone of food and farming in New Zealand for over 150 years. Available online at: https://www.far.org.nz/arable-industry-profile (Accessed September 14, 2022).

Google Scholar

Francis G. S., Haynes R. J., and Williams P. H. (1995). Effects of the timing of ploughing-in temporary leguminous pastures and two winter cover crops on nitrogen mineralization, nitrate leaching and spring wheat growth. J. Agric. Sci. 124, 1–9. doi: 10.1017/S0021859600071185

Crossref Full Text | Google Scholar

Freeman C. L. (1985). “Factors affecting scale of enterprise,” in Producing herbage seeds-Grassland Research and Practice Series No. 2. Eds. Hare M. D. and Brock J. L.(New Zealand Grassland Association, Palmerston North, New Zealand), 37–40.

Google Scholar

Groffman P. M., Hendrix P. F., and Crossley D. A. Jr. (1987). Nitrogen dynamics in conventional and no-tillage agroecosystems with inorganic fertilizer or legume nitrogen inputs. Plant Soil 97, 315–332. doi: 10.1007/BF02383222

Crossref Full Text | Google Scholar

Harper S. H. T. and Lynch J. M. (1981). The kinetics of straw decomposition in relation to its potential to produce the phytotoxin acetic acid. J. Soil Sci. 32, 627–637. doi: 10.1111/j.1365-2389.1981.tb01735.x

Crossref Full Text | Google Scholar

Hatch D. J., Jarvis S. C., and Philipps L. (1990). Field measurement of nitrogen mineralization using soil core incubation and acetylene inhibition of nitrification. Plant Soil 124, 97–107. doi: 10.1007/BF00010937

Crossref Full Text | Google Scholar

Hauggaard-Nielsen H., de Neergaard A., Jensen L. S., Høgh-Jensen H., and Magid J. (1998). A field study of nitrogen dynamics and spring barley growth as affected by the quality of incorporated residues from white clover and ryegrass. Plant Soil 203, 91–101. doi: 10.1023/A:1004350215467

Crossref Full Text | Google Scholar

Hay M. J. M., Nes P., and Robertson M. R. (1985). Effect of grazing management and season on nitrogen and phosphorus content of leaves and stolons of white clover in mixed swards. New Z. J. Exp. Agric. 13, 209–214. doi: 10.1080/03015521.1985.10426083

Crossref Full Text | Google Scholar

Hewitt A. E. (2010). New Zealand soil classification Landcare Reseach Science Series 1:3(New Zealand: Manaaki Whenua Press). doi: 10.7931/DL1-LRSS-1-2010

Crossref Full Text | Google Scholar

Honeycutt C. W. and Potaro L. J. (1990). Field evaluation of heat units for predicting crop residue carbon and nitrogen mineralization. Plant Soil 125, 213–220. doi: 10.1007/BF00010659

Crossref Full Text | Google Scholar

Huang C. Y., Boyer J. S., and Vanderhoef. L. N. (1975). Limitation of acetylene reduction (Nitrogen fixation) by photosynthesis in soybean having low water potentials. Plant Physiol. 56, 228–232. doi: 10.1104/pp.56.2.228

PubMed Abstract | Crossref Full Text | Google Scholar

Janzen H. H. and Kucey R. M. N. (1988). C, N and S mineralization of crop residues as influenced by crop species and nutrient regime. Plant Soil 106, 35–41. doi: 10.1007/BF02371192

Crossref Full Text | Google Scholar

Jarvis S. C., Hatch D. J., and Lovell R. D. (2001). An improved soil core incubation method for the field measurement of denitrification and net mineralization using acetylene inhibition. Nutrient Cycling Agroecosystems 59, 219–225. doi: 10.1023/A:1014401932276

Crossref Full Text | Google Scholar

Jarvis S. C., Stockdale E. A., Shepherd M. A., and Powlson D. S. (1996). Nitrogen mineralization in temperate agricultural soils: Processes and Measurement. Adv. Agron. 57, 187–235. doi: 10.1016/S0065-2113(08)60925-6

Crossref Full Text | Google Scholar

Knapp E. B., Elliott L. F., and Campbell G. S. (1983). Carbon, nitrogen and microbial biomass interrelationships during the decomposition of wheat straw: A mechanistic simulation model. Soil Biol. Biochem. 15, 455–461. doi: 10.1016/0038-0717(83)90011-1

Crossref Full Text | Google Scholar

Kumar K. (2000). Effects of different crop residues and management practices on soil nitrogen dynamics, yield and recovery of nitrogen by wheat. Lincoln University, Lincoln, New Zealand. Available online at: https://hdl.handle.net/10182/1486 (Accessed April 22, 2024).

Google Scholar

Kumar K. and Goh K. M. (2000a). Biological nitrogen fixation, accumulation of soil nitrogen and nitrogen balance for white clover (Trifolium repens L.) and field pea (Pisum sativum L.) grown for seed. Fields Crop Res. 68, 49–59. doi: 10.1016/S0378-4290(00)00109-X

Crossref Full Text | Google Scholar

Kumar K. H. and Goh K. M. (2000b). Crop residues and management practices: Effects on soil quality, soil nitrogen dynamics, crop yield, and nitrogen recovery. Adv. Agron. 68, 197–319. doi: 10.1016/S0065-2113(08)60846-9

Crossref Full Text | Google Scholar

Kumar K., Goh K. M., Scott W. R., and Frampton C. M. (2001). Effects of ¹⁵N-labelled crop residues and management practices on subsequent winter wheat yields, nitrogen benefits and recovery under field conditions. J. Agric. Sci. 136, 35–53. doi: 10.1017/S0021859600008522

Crossref Full Text | Google Scholar

Kušlienė G., Rasmussen J., and Eriksen J. (2013). “Leaching of dissolved nitrogen under white clover pure stand vs grass-clover mixture,” in 17th Symposium of the European Grassland Federation, 156–159.

Google Scholar

Ladd J. N. and Amato M. (1986). The fate of nitrogen from legume and fertilizer sources in soils successively cropped with wheat under field conditions. Soil Biol. Biochem. 18, 417–425. doi: 10.1016/0038-0717(86)90048-9

Crossref Full Text | Google Scholar

Lilburne L. R., Webb T. H., and Francis G. S. (2003). Relative effect of climate, soil, and management on risk of nitrate leaching under wheat production in Canterbury, New Zealand. Aust. J. Soil Res. 41, 699–709. doi: 10.1071/SR02083

Crossref Full Text | Google Scholar

Lloyd J. and Taylor J. A. (1994). On the temperature-dependence of soil respiration. Funct. Ecol. 8, 315–323. doi: 10.2307/2389824

Crossref Full Text | Google Scholar

Lucas R. J., Smith M. C., Jarvis P., Mills A., and Moot D. J. (2010). Nitrogen fixation by subterranean and white clovers in dryland cocksfoot pastures. Proc. New Z. Grassland Assoc. 72, 141–146. doi: 10.33584/jnzg.2010.72.2825

Crossref Full Text | Google Scholar

Manaaki W. (2019a). “Templeton_1a.1 soil report,” in S-Map Online – the digital soil map of New Zealand. Lincoln, New Zealand: Manaaki Whenua Landcare Research. Available online at: https://smap.landcareresearch.co.nz (Accessed August 30, 2023).

Google Scholar

Manaaki W. (2019b). “Wakanui_6a.1 soil report,” in S-Map Online – the digital soil map of New Zealand. Lincoln, New Zealand: Manaaki Whenua Landcare Research. Available online at: https://smap.landcareresearch.co.nz (Accessed August 30, 2023).

Google Scholar

Manaaki W. (2019c). “Flaxton_1a.1 soil report,” in S-Map Online – the digital soil map of New Zealand. Lincoln, New Zealand: Manaaki Whenua Landcare Research. Available online at: https://smap.landcareresearch.co.nz (Accessed August 30, 2023).

Google Scholar

Mather R. D. J., Melhuish D. T., and Herlihy M. (1995). Trends in global marketing of white clover cultivars Agronomy Society of New Zealand Special Publication No. 11/Grassland Research and Practice Series No. 6. 7–14. Christchurch New Zealand: Agronomy Society of New Zealand.

Google Scholar

McCurdy J. D., McElroy J. S., Guertal E. A., and Wood C. W. (2013). Dynamics of white clover decomposition in a southeastern Bermudagrass lawn. Am. Soc. Agron. J. 105, 1277–1282. doi: 10.2134/agronj2013.0058

Crossref Full Text | Google Scholar

Olewski J., Baddeley J., Watson C., and Baggs E. (2011). Does soil pH affect the decomposition of white clover? Aspects Appl. Biol. 109, 125–130.

Google Scholar

Olykan S. T., Lucas R. J., Black A. D., and Moot D. J. (2021). Refining foliage sampling protocols for white clover. J. New Z. Grasslands 83, 59–67. doi: 10.33584/jnzg.2021.83.3498

Crossref Full Text | Google Scholar

Pal D. and Broadbent F. E. (1975). Influence of moisture on rice straw decomposition in soils. Soil Sci. Soc. America J. 39, 59–63. doi: 10.2136/sssaj1975.03615995003900010018x

Crossref Full Text | Google Scholar

Pate J. S. (1976). “Physiology of the reaction of nodulated legumes to environment,” in Symbiotic nitrogen fixation in plants, vol. 7 . Ed. Nutman P. S. (Cambridge University Press, Cambridge, UK), 335–360.

Google Scholar

Paul E. A. and Clark F. E. (1989). Soil microbiology and biochemistry (Orlando, USA: Academic Press).

Google Scholar

Rasmussen J., Gjettermann B., Eriksen J., Steen Jensen E., and Høgh-Jensen H. (2008). Fate of ¹⁵N and ¹⁴C from labelled plant material: Recovery in perennial ryegrass-clover mixtures and in pore water of the sward. Soil Biol. Biochem. 40, 3031–3039. doi: 10.1016/j.soilbio.2008.08.025

Crossref Full Text | Google Scholar

Russelle M. P., Lamb J. F. S., Montgomery B. R., Elsenheimer D. W., Miller B. S., and Vance C. P. (2001). Alfalfa rapidly remediates excess inorganic nitrogen at a fertilizer spill site. J. Environ. Qual. 30, 30–36. doi: 10.2134/jeq2001.30130x

PubMed Abstract | Crossref Full Text | Google Scholar

Seed & Grain New Zealand (2024). Interim statistics for season 2023/2024 - Industry Information Seed_Stats 2023–2024 Season Finals. Available online at: https://www.nzgsta.co.nz/industry-info/seed_interim_stats-2023-2024-season-finals/ (Accessed August 19, 2025).

Google Scholar

Shi J. (2013). Decomposition and nutrient release of different cover crops in organic farm systems. University of Nebraska Graduate College, Lincoln, Nebraska, USA. Available online at: https://digitalcommons.unl.edu/natresdiss/75.

Google Scholar

Singh G., Dhakal M., Yang L., Kaur G., Williard K. W. J., Schoonover J. E., et al. (2020). Decomposition and nitrogen release of cover crops in reduced- and no-tillage systems. Agron. J. 112, 3605–3618. doi: 10.1002/agj2.20268

Crossref Full Text | Google Scholar

Stanford G., Frere M. H., and van der Pol R. A. (1975). Effect of fluctuating temperatures on soil nitrogen mineralisation. Soil Sci. 119, 222–226. doi: 10.1097/00010694-197503000-00007

Crossref Full Text | Google Scholar

Sturite I., Uleberg M. V., Henriksen T. M., Jørgensen M., Bakken A. K., and Breland T. A. (2006). Accumulation and loss of nitrogen in white clover (Trifolium repens L.) plant organs as affected by defoliation regime on two sites in Norway. Plant Soil 282, 165–182. doi: 10.1007/s11104-005-5697-3

Crossref Full Text | Google Scholar

Van der Weerden T. J., Sherlock R. R., Williams P. H., and Cameron K. C. (1999). Nitrous oxide emissions and methane oxidation by soil following cultivation of two different leguminous pastures. Biol. Fertility Soils 30, 52–60. doi: 10.1007/s003740050587

Crossref Full Text | Google Scholar

Vogeler I., Hansen E. M., Thomsen I. K., and Østergaard H. S. (2019). Legumes in catch crop mixtures: Effects on nitrogen retention and availability, and leaching losses. J. Environ. Manage. 239, 324–332. doi: 10.1016/j.jenvman.2019.03.077

PubMed Abstract | Crossref Full Text | Google Scholar

Wessels-Perelo L., Jimenez M., and Munch J. C. (2006). Microbial immobilisation and turnover of ¹⁵N labelled substrates in two arable soils under field and laboratory conditions. Soil Biol. Biochem. 38, 912–922. doi: 10.1016/j.soilbio.2005.07.013

Crossref Full Text | Google Scholar

Whelan G. A. and White J. G. H. (1985). Biomass accumulation and nitrogen fixation by a white clover seed crop. Agron. Soc. New Z. 15, 87–92. Available online at: https://www.agronomysociety.org.nz/1985-journal-papers.

Google Scholar

Widdup K. H., Purves R. G., Black A. D., Jarvis P., and Lucas R. J. (2001). Nitrogen fixation by caucasian clover and white clover in irrigated pastures. Proc. New Z. Grassland Assoc. 63, 171–175. doi: 10.33584/jnzg.2001.63.2423

Crossref Full Text | Google Scholar