Experimental evaluation of biological regeneration of arable soil: The effects of grass-clover leys and arbuscular mycorrhizal inoculants on wheat growth, yield, and shoot pathology

Abstract

Wheat yields have plateaued in the UK over the last 25 years, during which time most arable land has been annually cropped continuously with short rotations dominated by cereals. Arable intensification has depleted soil organic matter and biology, including mycorrhizas, which are affected by tillage, herbicides, and crop genotype. Here, we test whether winter wheat yields, mycorrhization, and shoot health can be improved simply by adopting less intensive tillage and adding commercial mycorrhizal inoculum to long-term arable fields, or if 3-year grass-clover leys followed direct drilling is more effective for biological regeneration of soil with reduced N fertiliser. We report a trial of mycorrhization, ear pathology, and yield performance of the parents and four double haploid lines from the Avalon x Cadenza winter wheat population in a long-term arable field that is divided into replicated treatment plots. These plots comprised wheat lines grown using ploughing or disc cultivation for 3 years, half of which received annual additions of commercial arbuscular mycorrhizal (AM) inoculum, compared to 3-year mown grass-clover ley plots treated with glyphosate and direct-drilled. All plots annually received 35 kg of N ha^-1 fertiliser without fungicides. The wheat lines did not differ in mycorrhization, which averaged only 34% and 40% of root length colonised (RLC) in the ploughed and disc-cultivated plots, respectively, and decreased with inoculation. In the ley, RLC increased to 52%. Two wheat lines were very susceptible to a sooty ear mould, which was lowest in the ley, and highest with disc cultivation. AM inoculation reduced ear infections by >50% in the susceptible lines. In the ley, yields ranged from 7.2 to 8.3 t ha^-1, achieving 92 to 106% of UK average wheat yield in 2018 (7.8 t ha^-1) but using only 25% of average N fertiliser. Yields with ploughing and disc cultivation averaged only 3.9 and 3.4 t ha^-1, respectively, with AM inoculum reducing yields from 4.3 to 3.5 t ha^-1 in ploughed plots, with no effect of disc cultivation. The findings reveal multiple benefits of reintegrating legume-rich leys into arable rotations as part of a strategy to regenerate soil quality and wheat crop health, reduce dependence on nitrogen fertilisers, enhance mycorrhization, and achieve good yields.

Keywords: arbuscular mycorrhizal fungi (AMF); conservation tillage; grass-clover leys; no-tillage; regenerative agriculture; systemic induced resistance.

Figure 1

Aerial view (June 2018) of the experimental plots arranged in four blocks (A-D) running downslope at Spen Farm, Tadcaster (53° 51' 44” N; 1° 20' 35”W). The image shows the wheat direct drilled into a 3-year ley compared to plots that have been cultivated and cropped with wheat for 3 successive years, with Skyfall wheat guard rows and the surrounding field which is not part of the trial. Map data © 2018 Google.

Figure 2

Ley and tillage effects on wheat yields in the 2018 harvest, averaging across all six genotypes and across the treatments with and without mycorrhiza inoculum added. Bars represent the tillage treatments: ploughed, minimally tilled, and direct drilled into ley. (A) wheat tiller height (cm) (ploughed n = 48, min tillage n = 52, and ley n = 24), (B) grain-filling (g / 100 grains) (ploughed n = 48, min tillage n = 48, ley n = 24), and (C) total grain yield (corrected for 15% moisture, t / ha ⁻¹) (ploughed n = 48, min tillage n = 47, and ley n = 24). The error bars show ± 1 SE. Results of Tukey's HSD test results of significant differences are shown (*p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 3

Effects of ploughing, disc- cultivation, and direct drilling into a 3-year ley, on wheat, averaging across all 6 genotypes, and across the treatments with and without mycorrhiza inoculum added, on (A) AMF colonisation [expressed as % root length colonisation by arbuscules, hyphae, and vesicles (ploughed n = 39, min tillage n = 39, ley n = 5)] and, (B) sooty ear mould (infected ears ha ⁻¹) (ploughed n = 48, min tillage n = 52, and ley n = 24). The error bars show ± 1 SE and are displaced to the right for vesicles for clarity. Results of significant Tukey's HSD test results for differences between means are shown (*p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 4

Wheat performance averaged across genotypes on ploughed and minimally tilled plots, with added sterile carrier (- AM) or commercial AMF inoculum (+ AM) compared to direct drilling into ley at the 2018 harvest. (A) wheat tiller height (cm) (ploughed – n = 24, ploughed + n = 24, min tillage – n = 24, min tillage + n = 28, and ley n = 24), (B) grain weight (g / 100 grains) (ploughed – n = 24, ploughed + n = 24, min tillage – n = 24, min tillage + n = 23, and ley n = 24), and (C) total grain yield (corrected for 15% moisture, t ha ⁻¹) (ploughed – n = 24, ploughed + n = 24, min tillage – n = 24, min tillage + n = 24, and ley n = 24). The error bars show ± 1 SE. Results of significant Tukey's HSD test results of significant differences are shown on the graphs *p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 5

Wheat performance, averaging across genotypes, on ploughed and minimally tilled plots, with added sterile carrier (- AM) or commercial AMF inoculum (+ AM) compared to direct drilling into a 3-year ley at the 2018 harvest. (A) Mycorrhizal abundance (% colonisation of wheat roots by hyphae, arbuscules, and vesicles) (ploughed and disc cultivated n = 24, ley n = 14), and (B) Sooty ear mould (infected ears ha ⁻¹) (n = 24 in all treatments). The error bars show ± 1 SE. Results of significant Tukey's HSD test results of significant differences are shown on the graphs (*p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 6

Genotype effects of wheat lines Avalon, Cadenza, AxC22, AxC53, AxC57, and AxC69 on wheat performance, averaging across mycorrhiza, tillage and ley / arable treatments on (A) wheat tiller height (cm) (Avalon n = 29, Cadenza n = 29, AXC22 n = 21, AXC53 n = 22, AXC57 n = 20, and AXC69 n = 21), (B) grain weight (g / 100 grains) (Avalon n = 20, Cadenza n = 20, AXC22 n = 20, AXC53 n = 20, AXC57 n = 20, and AXC69 n = 20), and (C) total grain yield (corrected for 15% moisture, t ha⁻¹) (Avalon n = 29, Cadenza n = 29, AXC22 n = 21, AXC53 n = 22, AXC57 n = 20, and AXC69 n = 21). The error bars show ± 1 SE. Results of significant Tukey's HSD test results of significant differences are shown on the graphs (*p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 7

Genotype effects of wheat lines Avalon, Cadenza, AxC22, AxC53, AxC57, and AxC69 on (A) Mycorrhizal abundance (% colonisation of wheat roots by hyphae, arbuscules, and vesicles) and (B) Sooty ear mould (infected ears ha⁻¹) (n = 20 for each wheat genotype). The error bars show ± 1 SE. Results of Tukey's HSD test results of significant differences are shown on the graphs (*p = 0.05; **p = 0.001; ***p < 0.0001).

Figure 8

Cultivation and genotype effects on wheat yields from 2018 harvest. Wheat lines Avalon, Cadenza, AxC22, AxC53, AxC57, and AxC69 in ploughed and disc-cultivated plots with a sterile carrier (- AM) or mycorrhizal inoculum (+AM) or direct drilled into ley. (A) wheat height (cm) (n = 20 for each wheat genotype), (B) grain weight (g / 100 grains) (n = 20 for each wheat genotype), and (C) grain yield (15% moisture content correction, t ha⁻¹) (n = 20 for each wheat genotype). Error bars show ± 1 SE. The results of the three-way ANOVA testing for effects of genotype, mycorrhizal inoculum and tillage (ploughing vs. disc cultivated) and their interactions are presented for the permanent arable plots (excluding the ley). In parallel, the two-way ANOVA results are presented for effects of tillage (direct drilled ley, disc cultivated, and ploughed) and genotype, and their interaction, averaging across the mycorrhizal inoculum treatments, and including the ley plots. In each case, *p = 0.05; **p = 0.001; and ***p < 0.0001.

Figure 9

Effects of wheat genotype (Avalon, Cadenza, AxC22, AxC53, AxC57, AxC69), tillage (plough, disc cultivated, and direct drill), mycorrhiza inoculum (- sterile carrier, + commercial mycorrhizal inoculum), and ley on (A) total mycorrhizal abundance (% colonisation of wheat roots by hyphae, arbuscules, and vesicles) (n = 16–18 for each wheat line or parent), and (B) effects on the amount of sooty ear mould (ears ha⁻¹) (n = 20 for each wheat line or parent). Error bars show ± 1 SE. The results of the three-way ANOVA testing for effects of genotype, mycorrhizal inoculum and tillage (ploughing vs. disc cultivated), and their interactions are presented for the permanent arable plots (excluding the ley). In parallel, the two-way ANOVA results are presented for effects of tillage (direct drilled ley, disc cultivated, and ploughed) and genotype, and their interaction, averaging across the mycorrhizal inoculum treatments, and including the ley plots. In each case, *p = 0.05; **p = 0.001; and ***p < 0.0001.