Xueming Yang

This study presents results from the first 5 years of an organic cropping trial in Ontario, Canada, where legume cover crops were the primary nitrogen source in a soybean-winter wheat/cover crop-corn rotation. Treatments included cover crop termination using moldboard plow (MP) or chisel plow (CP), a no-cover crop control under conventional production (CK-C), and four cover crops including summer-seeded crimson clover (CC, Trifolium incarnatum L.), summer-seeded hairy vetch (HV, Vicia villosa L. Roth), summer-seeded red clover (RC_ss, Trifolium pratense L.), and frost-seeded red clover (RC_fs). Summer-seeding occurred after wheat harvest (July–August), and frost-seeding occurred in early spring (March–April). At cover crop termination, average aboveground cover crop biomass ranged from 5.9 to 8.1 Mg ha⁻¹, while accumulated biomass nitrogen ranged from 155 to 193 kg ha⁻¹. Corn grain yields were 11.6 Mg ha⁻¹ for MP and 10.2 Mg ha⁻¹ for CP tillage-termination method; and 13.3 Mg ha⁻¹ for CK-C, 10.9 Mg ha⁻¹ for RC_fs, 10.6 Mg ha⁻¹ for HV, 10.2 Mg ha⁻¹ for CC, and 9.5 Mg ha⁻¹ for RC_ss. Organic winter wheat yields were nitrogen-limited, averaging 27% lower than CK-C. Winter wheat yields were 10–15% lower in the RC_fs than in other summer-seeded cover crop treatments. Soybean yields were largely unaffected by the treatments. It was concluded that summer-seeded legume cover crops are an effective primary nitrogen source for corn, but not as effective for the winter wheat phase of the soybean-winter wheat-corn rotation.

Summer-seeding legume cover crops can reduce erosion, mitigate nitrate leaching losses during the non-growing season, and provide bio-accumulated nitrogen (N) to the following crop. Very little information is available, however, on the most effective methods for terminating summer-seeded cover crops, or on the amount of N they can provide to subsequent crops. We therefore evaluated the impacts of selected legume cover crops and cover crop termination-tillage methods on corn (Zea mays L.) N credit and grain yield in a soybean (Glycine max Merr.) - winter wheat (Triticum aestivum L.)/cover crop - corn rotation over 3 consecutive years on a clay loam soil in southwestern Ontario. The cover crops were summer-seeded after wheat harvest, and included hairy vetch (Vicia villosa L. Roth), red clover (Trifolium pratense L.), white clover plus crimson clover mix (incarnatum and Trifolium alba L.), and a no cover crop control (CK). The cover crops were nested within termination-tillage method, including moldboard plow-down in fall, and herbicide spray-down in spring under strip-tillage or no-tillage then corn planting. Nitrogen fertilizer to corn was applied via side-dress at 200 kg N ha^-1 to the CKs, and at 100 kg N ha^-1 to the cover crops. Averaged over 3 years, above-ground biomass N level followed the pattern: hairy vetch > white clover plus crimson clover mix > red clover for fall plow-down termination. There were no significant differences in cover crop biomass N levels for spring termination. Corn response to cover crop was significantly affected by termination-tillage method and cover crop species, with poor corn stands after no-tillage, red clover and white plus crimson clover. Fall plow-down termination provided a range in N credit of 80-85 kg N ha^-1, which was significantly greater than the N credit under herbicide spray-down in spring. Under spring strip-tillage and no-tillage, hairy vetch produced significantly greater corn grain yields than red clover and white plus crimson clover mix.

Although it is well established that soil temperature has substantial effects on the agri-environmental performance of crop production, little is known of soil temperatures under living cover crops. Consequently, soil temperatures under a crimson clover and white clover mix, hairy vetch, and red clover were measured for a cool, humid Brookston clay loam under a corn–soybean–winter wheat/cover crop rotation. Measurements were collected from August (after cover crop seeding) to the following May (before cover crop termination) at 15, 30, 45, and 60 cm depths during 2018–2019 and 2019–2020. Average soil temperatures (August–May) were not affected by cover crop species at any depth, or by air temperature at 60 cm depth. During winter, soil temperatures at 15, 30, and 45 cm depths were greater under cover crops than under a no cover crop control (CK), with maximum increase occurring at 15 cm on 31 January 2019 (2.5–5.7 °C) and on 23 January 2020 (0.8–1.9 °C). In spring, soil temperatures under standing cover crops were cooler than the CK by 0.1–3.0 °C at 15 cm depth, by 0–2.4 °C at the 30 and 45 cm depths, and by 0–1.8 °C at 60 cm depth. In addition, springtime soil temperature at 15 cm depth decreased by about 0.24 °C for every 1 Mg·ha⁻¹ increase in live cover crop biomass. Relative to bare soil, cover crops increased near-surface soil temperature during winter but decreased near-surface soil temperature during spring. These temperature changes may have both positive and negative effects on the agri-environmental performance of crop production.

Infrared spectroscopy has the potential to rapidly analyse soil water-dissolved carbon and amino sugars. In this study, mid-infrared (MIR) and near-infrared (NIR) spectra collected from soil water extracts or from bulk soils were analysed with partial least squares regression (PLSr) to estimate the concentrations of water-dissolved carbon and amino sugars in diverse agricultural soils collected from five field sites in two western and two eastern Canadian provinces. The MIR-PLSr models developed from soil water extract spectra estimated hot-water (100 °C) dissolved carbon (HWDC) [R² = 0.97–0.70, ratio of prediction to deviation (RPDp) = 6.13–1.83] well, but the MIR-PLSr models did not estimate cold-water (21 °C) dissolved carbon (CWDC) well (R² = 0.82–0.50, RPDp = 2.35–1.42). The model estimates of HWDC at the multisite scale (all samples together) and for the two western Canada sites (R² = 0.97–0.93, RPDp = 6.13–3.68) surpass the modal estimates for the three eastern Canadian sites (R² = 0.81–0.70, RPDp = 2.28–1.83). The MIR- and NIR-PLSr models derived from bulk soil spectra both estimated HWDC well at the multisite scale (R² = 0.91–0.88, RPDp = 3.32–2.90) and for the western Canada sites (R² = 0.90–0.87, RPDp = 3.18–2.96). Models developed from hot-water extract spectra and bulk soil spectra resulted in poor estimates of soil amino sugars (R² = 0.74–0.21, RPDp = 1.99–1.12), except for the approximate quantitative estimation of muramic acid by models based on soil spectra at the western and the multisite scale (R² = 0.82–0.80, RPDp = 2.33–2.21). We concluded that MIR and NIR models at regional and multisite scales can be used as a tool to monitor HWDC but that additional research is required for estimating soil amino sugars.

Few studies have evaluated structural features of soil organic carbon (SOC) in different soil particle fractions, especially SOC changes induced by tillage, using Fourier transformed mid-infrared (MIR) spectroscopy. To make a contribution in this context, soil samples of a Brookston clay loam (mesic Typic Argiaquoll) with averaged pH and organic matter concentration at 7.28 and 43.9 g kg⁻¹, respectively, were collected from short-term no-tillage (NT97) and mouldboard plow (CT97) treatments initiated in 1997 and long-term no-tillage (NT83) and mouldboard plow (CT83) treatments initiated in 1983 under a corn-soybean rotation, and were separated into sand, silt, and clay fractions using sonication. Structural features of SOC in these soil fractions were investigated using curve-fitting analysis of mid-infrared (MIR) spectra. Aromatic C content was found to be greater in clay- than in sand-sized fractions, while aliphatic C content was higher in sand- than in silt- and clay-sized particles. With decrease in tillage intensity, the aliphatic C gradually increased in sand- and clay-sized fractions but not in the silt-sized fraction. The aliphatic C content in sand fraction was significantly greater in NT83 than CT83 (P < 0.05). The aromatic C in silt- and clay-sized fractions was greater in NT83 than in both CT soils, whereas aromatic C contents were higher in both CT soils than in NT83 soil. Significantly higher aromatic/aliphatic C ratio in CT83 than NT83 was found in sand-sized fractions, while the opposite trend was found in the silt-sized fraction. These findings were not apparent until the curve-fitting technique was employed, which has the capacity to quantify many overlapped bands in the spectra. This study demonstrates that the curve-fitting of MIR spectra advances the analysis of organic matter in soil samples.

Core Ideas: Tillage practices on surface soil temperature in the winter months were tested. No‐till soil responses slowly to changes of air temperature than tilled soil. Soil temperatures in winter were significantly greater in no‐till than tilled soils. Soil temperature affects soil microbial activity and hence impacts soil greenhouse gas (CO ₂ , N ₂ O, CH ₄ ) emissions and plant nutrient recycling. However, there is a lack of information on the effects of tillage practices on soil temperature in the surface soil layers in the winter months. Soil temperature and moisture in no‐tillage (NT) and fall moldboard plow (MP) of a Brookston clay loam in southwestern Ontario were measured on an hourly basis over the winter months (December to April) at soil depths (25, 75, and 150 mm) during 2012 to 2013, 2013 to 2014 and 2014 to 2015, respectively. Both soil temperature and moisture significantly varied with tillage practices over the winter months. The tillage and depth interaction occurred for soil moisture, but not soil temperature. Soil temperatures in December and January were significantly greater (+0.7°C) in NT than MP soil with a maximum divergence of 2 to 4°C in the 25‐mm depth for 4 to 5 d in January. In March and April, the soil temperature was about 0.7°C cooler for the NT than the MP soil. The soils were generally wetter in the NT than the MP plots and the difference was statistically significant from December to February. In general, the soil was warmer and wetter under NT than MP management in winter months for this clay loam soil whereas soils were cooler and wetter under no‐tillage in the spring.

We test the common assumption that organic carbon (OC) storage occurs on sand-sized soil particles only after the OC storage capacity on silt- and clay-sized particles is saturated. Soil samples from a Brookston clay loam in Southwestern Ontario were analysed for the OC concentrations in bulk soil, and on the clay (−1 soil. The OC concentrations on all three particle size fractions were significantly related to the OC concentration of bulk soil. However, OC concentration increased slowly toward an apparent maximum on silt and clay, but this maximum was far greater than the maximum predicted by established C sequestration models. In addition, significant increases in OC associated with sand occurred when the bulk soil OC concentration exceeded 30 g C kg⁻¹, but this increase occurred when the OC concentration on silt + clay was still far below the predicted storage capacity for silt and clay fractions. Since the OC concentrations in all fractions of Brookston clay loam soil continued to increase with increasing C (bulk soil OC content) input, we concluded that the concept of OC storage capacity requires further investigation.

Detailed information on the profile distributions of agronomically important soil properties in the planting season can be used as criteria to select the best soil tillage practices. Soil cores (0–60 cm) were collected in May, 2012 (before soybean planting), from soil transects on a 30‐yr tillage experiment, including no‐tillage ( NT ), ridge tillage ( RT ) and mouldboard plough ( MP ) on a Brookston clay loam soil (mesic Typic Argiaquoll). Soil cores were taken every 19 cm across three corn rows and these were used to investigate the lateral and vertical profile characteristics of soil organic carbon ( SOC ), pH, electrical conductivity ( EC ), soil volumetric water content ( SWC ), bulk density ( BD ), and penetration resistance ( PR ). Compared to NT and MP , the RT system resulted in greater spatial heterogeneity of soil properties across the transect. Average SOC concentrations in the top 10 cm layer were significantly greater in RT than in NT and MP ( P  =   0.05). NT soil contained between 0.8 and 2.5% (vol/vol) more water in the top 0–30 cm than RT and MP , respectively. MP soil had lower PR and BD in the plough layer compared to NT and RT soils, with both soil properties increasing sharply with depth in MP . The RT had lower PR relative to NT in the upper 35 cm of soil on the crop rows. Overall, RT was a superior conservation tillage option than NT in this clay loam soil; however, MP had the most favourable soil conditions in upper soil layers for early crop development across all treatments.