The experiment was conducted at the ICAR-Indian Agricultural Research Institute, New Delhi (28°64′ N latitude, 77°15′ E longitude, and an altitude of 228 meters above mean sea level) under triplicated split-plot design during the winter (rabi) seasons of 2021–22 and 2022–23. The main-plot treatments comprised of different crop establishment systems, while the sub-plot treatments included five weed management strategies (Table 1). The wheat crop was sown following the harvest of the preceding maize crop, maintaining a maize–wheat cropping system consistently applied on the experimental site for the past 5 years. Among the weed management treatments, the unweeded check (UWC) served as a control, allowing natural weed infestation without any intervention. The experimental field featured sandy loam soil (sand 58.6%, silt 22.8%, and clay 18.6%) with moderate water-holding capacity, an even topography, and a well-functioning drainage system. Soil analysis of the upper 150 mm layer revealed low organic carbon content (0.41%), low available nitrogen (221.3 kg ha⁻¹), medium levels of available phosphorus (18 kg ha⁻¹) and potassium (241.1 kg ha⁻¹), and a slightly alkaline reaction (pH 7.8).

The climatic conditions varied significantly during the 2 years of the study period (Figure 1). In the first year, higher rainfall was recorded during the early stages of crop growth, while the later stages experienced relatively lower rainfall. Conversely, in the second year, the rainfall was more pronounced during the later stages of crop ontogeny. Temperature trends also differed between the years, as a sudden rise in temperature was observed during the grain-filling stage of 2021–22, which contrasted with the second year, where temperatures during this critical stage remained fairly stable.

Wheat crop was sown in rows spaced 22.5 cm apart during both years of the study. For EBRC treatment, an elevated-bed planter was used (67.5 cm top and 30 cm furrow). Three rows of wheat variety “HD-3226” were sown on each bed keeping a row-to-row spacing of 22.5 cm using 100 kg seed ha⁻¹ to ensure an optimal plant population. Prior to sowing, the seeds were treated with the carbendazim at 2 g kg⁻¹ of seed to prevent fungal infections. The recommended fertilizer application of 150 kg N ha⁻¹, 60 kg P₂O5 ha⁻¹, and 40 kg K₂O ha⁻¹ was done through urea, di-ammonium phosphate, and muriate of potash, respectively. Full doses of phosphorus and potassium, along with one third dose of the nitrogen, were applied at the time of sowing. The remaining nitrogen was top-dressed in two equal splits, once at the active vegetative stage (40 days after sowing, DAS) and second at the flowering stage (80 DAS). Herbicides were applied at recommended doses at 2 days after sowing as pre-emergence and 25 days after sowing as post-emergence for effective control of weeds.

A quadrat measuring 0.5 m × 0.5 m was thrown randomly at four different locations within each plot for sampling at 40 DAS, and weeds were counted species-wise, and collected and categorized into sedges, and narrow-leaved and broad-leaved weeds, which were later summed up as total number of weeds. After air drying, the categorized weed species were dried in oven at a temperature of 65°C for 48 h to record the dry matter accumulation by different weed flora. Moreover, the different indices of weeds were calculated to evaluate the efficiency of different herbicides. These indices are mentioned as follows:

Numerous growth indices, such as leaf area index (LAI), crop growth rate (CGR), relative growth rate (RGR), and net assimilation rate (NAR), were computed using the following standard equations:

Data on weeds and wheat crop were analyzed by the analysis of variance (ANOVA) technique for a split-plot design using PROC GLM in SAS 9.3 (SAS Institute, Cary, NC). Weed populations were transformed through the square-root transformation method [(x + 0.5)^½] before ANOVA to reduce higher variation and for more precision. The species-wise populations of weeds, wheat grain yield, were subjected to Levene's test for homogeneity of variance. The error variances for almost all parameters (i.e., weed and wheat grain yield) were homogeneous over the years, indicating that the uniformity in error variance was significant. The significance of treatment means was appraised using Tukey's honest significant difference (HSD) test at p ≤ 0.05. The genotype + genotype by environment (GGE) biplot analysis was carried out to determine the effects of treatment combinations in controlling the weed population across both the years (Yan and Kang, 2011).

The predominant weed flora at the experimental site consisted of two narrow-leaved weed (NLW) species, four broad-leaved weed (BLW) species, and sedges, as detailed in Table 2. A pooled ANOVA over 2 years revealed 3.6% lower Chenopodium murale population in the second year compared to the first year of investigation. However, the densities of other weed species showed no statistically significant variation between the 2 years.

Across different crop establishment systems (Table 2), densities of Phalaris minor, Avena ludoviciana, and Coronopus didymus were 11.8, 10.8, and 11.0% lower under the elevated-bed system (EBRC) compared to double zero-tillage (DZRC), respectively. The densities of A. ludoviciana and Chenopodium album were statistically comparable between EBRC and conventional intensive tillage with residue incorporation (ITRI) treatments. No significant differences were noticed in the populations of Anagallis arvensis and Medicago indica across the crop establishment systems.

Among weed management options, the lowest populations of four dominant BLW species, namely, C. didymus, C. album, C. murale, and M. indica, were recorded under treatment WFC, followed by Pyro-MetCarf, while the population densities of two major narrow-leaved weed species, namely, P. minor and A. ludoviciana, were 67.6% and 46.5% lesser in WFC compared to SulfoMet, respectively, although the differences between WFC and Pyro-MetCarf treatments were non-significant. Furthermore, the sequential application of Pyro-MetCarf consistently resulted in the lowest populations of P. minor, A. ludoviciana, C. didymus, C. album, C. murale, and M. indica, compared to standalone applications of Pyro or SulfoMet.

The category-wise weed density (Table 3) was significantly influenced by the year, crop establishment system, weed management options, and their interactions. In the second year of experimentation, the overall densities of NLWs (11.0 %), BLWs (3.4%), sedges (6.6%), and total weeds (6.1%) were significantly lower compared to the first year. Similarly, among the crop establishment systems, EBRC showed lower densities of BLWs (7.1%), sedges (11.5 %), and total weeds (10.9%) compared to DZRC. Contrarily, non-significant differences were observed in NLW population across the crop establishment systems. The Pyro-MetCarf treatment reduced the NLW, BLW, and sedges density by 178%, 143.2%, and 58% as compared to UWC, respectively. However, the best treatment was WFC, and it remained statistically superior to all other treatments. Notably, for controlling sedges population, Pyro alone and pyro-MetCarf were statistically at par. The Pyro-MetCarf treatment remained at par to WFC for controlling diverse weed flora.

For weed dry matter (Table 4), no significant differences were observed due to the years and crop establishment systems. However, numerically, the lowest total weed dry matter and dry matter of NLWs, BLWs, and sedges were recorded in the EBRC, compared to DZRC and ITRI. The weed-free check (WFC) exhibited the lowest dry matter accumulation (Tables 3, 4) among all treatments. Dual-stage spray treatment of Pyro-MetCarf resulted in the lowest dry matter of NLWs, BLWs, sedges, and total weeds, outperforming other herbicidal treatments. However, the WFC had resulted in lowest weed dry matter accumulation. The interaction effects for main plot × subplot (C × W), year × main plot (Y × C), and year × main plot × subplot (Y × C × W) were found to be non-significant.

Both weed control efficiency (WCE; 66.3%) and weed control index (WCI; 67%) were significantly higher during the first year of experimentation compared to the second year (60.6 and 60.3%, respectively). However, no significant differences in WCE and WCI were observed among the various crop establishment systems (Table 5). Among weed management options, the WFC achieved the highest WCE and WCI. However, out of the herbicidal treatments, the application of Pyro-MetCarf resulted in significantly higher WCE (80.3%) and WCI (79.4%), followed by sole pre-emergence application of Pyro, which recorded 72.7% WCE and 71.9% WCI. The next best treatment was SulfoMet, with WCE and WCI values of 64.3% and 66.95%, respectively.

To have greater insights in interactive effects, four patterns of GGE biplot viz. “which won where/what,” “mean vs. stability,” “ranking genotypes,” and “ranking environments” have been generated for weed density to identify best treatments, dominant weed species under specific treatments, average control of weed population, prevalent weed species, and best treatment combination for controlling weeds.

Across the 2-year data, non-significant differences were observed in leaf area index (LAI), dry matter accumulation (DMA) (Table 6), tiller count per meter row length (Table 7), crop growth rate (CGR) (Figure 3), relative growth rate (RGR) (mg g⁻¹ m⁻²) (Figure 4), net assimilation rate (NAR) (Figure 5), ear length (cm), and the number of seeds per ear (Table 8) across crop establishment methods and weed management options as well during the experimentation period. There were no significant variations in LAI at 40 DAS, DMA at 40, 80, and 120 DAS (Table 6), tiller count per meter row length at 40, 80, and 120 DAS (Table 7), CGR (Figure 3), RGR (mg g⁻¹ m⁻²) (Figure 4), and NAR (Figure 5) among the crop establishment systems. However, LAI values at 80 DAS (4.47) and 120 DAS (3.01) were significantly higher under the EBRC compared to DZRC. These LAI values for EBRC were statistically on par with those observed under ITRI. This indicates that the EBRC provided a favorable environment for enhanced crop growth during later growth stages.

Among the weed management options, the weed-free check (WFC) consistently recorded the highest values for LAI (1.63, 5.07, and 3.27), DMA (72.5, 807.0, and 1253.2 g m⁻²), and tiller count (83.4, 119.0, and 125.3 per meter row length) at 40, 80, and 120 DAS, respectively, compared to all herbicidal treatments studied. The treatment where co-application of herbicides has been done (Pyro–MetCarf) demonstrated significantly higher LAI (1.60, 4.92, and 3.19), DMA (68.2, 775.1, and 1213.3 g m⁻²), and tiller counts (82.0, 115.9, and 112.4 m⁻¹ row length) at 40, 80, and 120 DAS, respectively. Furthermore, this treatment recorded 53.8% higher LAI, 29.4 5% higher dry matter, and 19.9% more number of tillers compared to UWC after 80 DAS. During the early growth stage (0–40 DAS), the highest crop growth rate (CGR), relative growth rate (RGR), and net assimilation rate (NAR) were observed under the WFC (2.41 g m⁻² day⁻¹, 4.28 mg g⁻¹ day⁻¹, and 0.72 g m⁻² day⁻¹, respectively), followed closely by Pyro–MetCarf application (2.27 g m⁻² day⁻¹, 4.18 mg g⁻¹ day⁻¹, and 0.66 g m⁻² day⁻¹, respectively). Non-significant differences were seen at later stages of crop growth (40–80 DAS and 80–120 DAS) in RGR and NAR. At later growth stages (80 and 120 DAS), wheat plants approach maturity with reduced vegetative growth and biomass accumulation, leading to stabilization of RGR and NAR values. Moreover, canopy closure and uniform environmental conditions minimize treatment effects, resulting in non-significant differences among herbicide treatments.

However, CGR was significantly higher under Pyro–MetCarf and the weed-free check (WFC) during these periods, highlighting the superior weed suppression and crop growth benefits of these treatments.

The number of effective tillers, ear length, and the number of grains per ear of wheat were not affected significantly due to temporal variations (year effects) or crop establishment systems as well. However, imposed weed management practices had a significant impact on all yield attributes (number of effective tillers, ear length, and number of grains per ear, Table 7). The weed-free check (WFC) recorded the highest values for yield parameters, including number of effective tillers (110.38), ear length (12.49 cm), and number of grains per ear (61.44). In contrast, the lowest values for these attributes were observed in the UWC, which were 22.1, 18.2, and 7.6% lower than WFC in number of effective tillers, ear length, and number of grains per ear, respectively. Dual-stage application of Pyro-MetCarf significantly enhanced yield attributes, recording 105.6 effective tillers per meter row length, 12.0 cm ear length, and 60.2 number of grains per ear. These values were statistically comparable to those achieved under the WFC, indicating the efficacy of Pyro-MetCarf in optimizing wheat yield attributes through effective weed control.

The differences in wheat biological yield were not reached to significant levels because of temporal variations and different crop establishment methods studied as demonstrated by a pooled analysis of 2-year data (Table 8). Among the crop establishment systems, EBRC produced the highest grain yield (5.55 t ha⁻¹), followed by ITRI at 5.45 t ha⁻¹, and DZRC at 5.29 t ha⁻¹ (Figure 6). The treatment WFC recorded the highest grain yield (6.27 t ha⁻¹), while the UWC obtained the lowest crop yield (4.37 t ha⁻¹). Among the sequential herbicide treatments, Pyro-MetCarf recorded the highest grain yield (5.77 t ha⁻¹). Over the 2 years, Pyro-MetCarf resulted in a mean grain yield increase of 24.3% compared to the UWC, highlighting its effectiveness in improving yield through better weed control efficacy.

The weed index (WI), which reflects the percentage reduction in crop yield due to crop-weed competition, showed that the highest yield loss (30.3%) occurred in the UWC, underscoring the substantial negative impact of uncontrolled weed growth on crop productivity. The WFC served as the standard control in which no weed growth was permitted, thereby representing an ideal weed-free environment where crop yield was not affected by weed interference (Figure 7). Among the sequential herbicide treatments, the lowest weed index (8.02%) was recorded with Pyro-MetCarf, closely followed by SulfoMet at 12.7%. The highest weed index among herbicide treatments was observed with Pyro alone (15.9%), suggesting that while it provided some degree of weed suppression, it was less effective than the sequential applications. These findings highlight the critical role of integrated weed management, particularly the use of sequential herbicides such as Pyro-MetCarf in minimizing yield losses caused by weed competition.

Crop establishment methods had a pronounced impact, resulting in significantly higher density and dry weight of narrow-leaved weeds (NLWs), broad-leaved weeds (BLWs), sedges, and total weed populations in the first year of experimentation compared to the second year. This trend can be attributed to favorable environmental conditions in the first year, particularly optimal precipitation, promoting multiple flushes of weed seed germination during the wheat-growing period (Figure 1). In contrast, the retention of crop residues as mulch proved highly effective in reducing weed density and dry matter accumulation. Maize residues applied on the soil surface functioned as a physical barrier, restrictin g light penetration to the soil, thereby suppressing weed germination and growth of numerous weed species infesting the wheat crop (Sharma et al., 2023). In addition, the residue covers likely induced physico-chemical and biological changes in the soil environment, further suppressing weed seed germination and emergence (Teasdale, 2000; Jabran and Chauhan, 2015).

The EBRC remained significantly more effective in reducing the densities of NLWs, BLWs, sedges, and total weeds compared to DZRC and ITRI systems (Sharma et al., 2023; Kumar and Kaur, 2024). The EBRC enhanced wheat growth by improving root density, increasing dry matter accumulation, and promoting a greater number of tillers (Kumar et al., 2023). These improvements facilitated better nutrient acquisition throughout the growing season. The higher tiller count in wheat strengthened its competitive ability, effectively suppressing weed growth on raised beds compared to flatbed systems, irrespective of the tillage method employed (Kumar and Kaur, 2024).

The influence of tillage on the weed seed bank remains a topic of ongoing debate. Some studies have reported negligible effects of tillage on increasing (Barberi et al., 2001) or decreasing (Murphy et al., 2006) weed seed bank density. Tillage practices affect the redistribution of weed seeds across the soil profile (0–30 cm) in varying ways. For instance, conventional tillage (CT) evenly distributes weed seeds within the plow zone, while zero tillage (ZT) results in accumulation of 60–90% of weed seeds in the top 0–5 cm of the soil (Hoffman et al., 1998). During the initial period from conventional to CA and no-tillage systems, the germinable weed seed bank tended to increase temporarily. However, later, no-tillage practices led to a significant reduction in weed seed bank density, by ~45–75%, compared to CT (Sergeja et al., 2024).

In addition, the dual-stage application of Pyro-MetCarf demonstrated remarkable efficacy in reducing weed densities. Over 2 years, this treatment achieved significant reductions in the densities of NLWs (64.1%), BLWs (58.9%), and sedges (41.2%) compared to the unweeded check.

Pyroxasulfone proved highly effective in controlling early-emerging NLWs, BLWs, and sedges by targeting them at the germination stage. This pre-emergence herbicide, characterized by its pyridine and oxazole rings, inhibits the biosynthesis of very-long-chain fatty acids, thereby arresting weed growth. On the other hand, the post-emergence combination of Met+Carf effectively controlled later-emerging NLWs and BLWs. A similar pattern was observed in the reduction of weed dry matter across NLWs, BLWs, and sedges. Higher WCE was recorded in elevated beds and intensive tillage with residue retention due to a lower weed density (Kaur et al., 2019; Kumar et al., 2022). Conversely, DZRC exhibited lower WCE and WCI, primarily due to the accumulation of a larger weed seed bank in the upper soil layer and the absence of soil disturbance in ZT practices.

The sequential application of Pyro-MetCarf demonstrated the highest WCE and WCI. This superior performance can be attributed to its ability to significantly reduce both the density and dry matter accumulation by weeds, ensuring balanced and sustained weed control throughout the crop growth period.

The GGE biplot analysis of weed population under different main plot and subplot combination revealed that R. dentatus was not controlled effectively by the combination of EBRC and DZRC along with varied herbicide application, whereas C. didymus and M. alba were controlled by these treatments when compared to UWC. This might have been due to shifting of weed flora from grassy weeds to more resistant R. dentatus along resistance to tolerate even higher dose of some herbicides (Chhokar et al., 2022). Furthermore, the population of P. minor was lowest implying effective control by different treatment combinations. The lowest population of P. minor might be owing to the inhibition of light penetration for germination and growth of weeds with residue retention combined with effectiveness of various herbicides (Ghosh et al., 2022a,b; Kaur and Singh, 2019). Herbicide resistance in R. dentatus and other broad-leaved weeds has been linked with increased metabolic resistance and altered translocation, as in case of P. minor for isoproturon resistance (Chhokar et al., 2022). Moreover, repeated use of single-site herbicides such as pyroxasulfone may select for cross-resistance in multiple weed species, necessitating integrated weed management strategies involving herbicide rotation and mixtures (Kaur and Singh, 2019). A. ludoviciana and A. arvensis were recorded similar mean population along with stability indicating the identical effectiveness of treatment combinations in controlling the population of these two weeds. Ranking of the environments analysis revealed ITRI in combination with sulfoMet was most effective treatment in controlling weed population significantly followed by DZRC and EBRC. The combination of ITRI and pyroxasulfone was inferior in controlling weed population indicating higher population of weeds. Relatively higher population of weeds might be attributed to lower control of some broad-leaved weeds by application of pyroxasulfone alone which otherwise could be effectively controlled by a combination of both pre- and post-emergence herbicides (Kaur and Singh, 2019; Samota et al., 2024).

During the second year of experimentation, a significant decline in yield was noticed. The unexpected rainfall at the grain-filling stage of wheat resulted in significant crop lodging and subsequent yield reductions. However, wheat cultivated on elevated beds demonstrated resilience to lodging and achieved higher yields. This advantage was attributed to better root proliferation, reduced weed density, and a more favorable microclimate that supported better crop growth under EBRC. Furthermore, the EBRC system significantly minimized the competition for critical resources, including space, light, nutrients, and moisture, and also benefitted from the weed-suppressive effects of maize residue retention (Raj et al., 2022; Kaur et al., 2024a). The maize residue cover acted as a physical barrier, which effectively suppressed weed germination and growth by limiting light penetration and altering soil microclimatic conditions (Bana et al., 2020). In addition, the retained residue improved soil health by enhancing moisture retention, moderating temperature fluctuations, and fostering beneficial microbial activity (Bana et al., 2023b).

These improvements allowed the wheat crop to establish better anchorage and deeper root systems, which facilitated enhanced nutrient and moisture acquisition. This, in turn, supported efficient photosynthesis, leading to increased leaf area, greater dry matter accumulation, and vigorous vegetative growth across the crop canopy (Duan et al., 2024). Moreover, the reduction in weed density and dry matter minimized competition for critical resources, such as water, nutrients, and space, while improving soil aeration in the EBRC system (Mondal et al., 2020). This favorable environment created ideal atmosphere for robust crop development, resulting in higher number of tillers, longer ear length, and an increased number of grains per ear (Table 8). These findings are in the line with the studies from Ghosh et al. (2022a), Sharma et al. (2023), and Kaur et al. (2024b), which emphasize the advantages of raised bed planting systems in enhancing wheat growth and productivity. Collectively, these factors underscore the superiority of the EBRC method in mitigating yield losses under adverse weather scenarios, such as rainfall-induced lodging, and in promoting higher and more sustainable wheat yields through improved weed suppression, resource use efficiency, and crop growth performance (Ghosh et al., 2022b).

Among the weed management strategies tested, the dual-stage application of Pyro-MetCarf emerged as the most effective option for controlling a wide spectrum of weed flora. This treatment provided a weed-free environment during the critical period of crop-weed interference, a key factor in minimizing yield losses and optimizing crop growth (Ghosh et al., 2022b; Kumar et al., 2022). The comprehensive weed suppression achieved through the sequential herbicide application not only minimized competition for vital resources but also created season-long idealistic environment for wheat growth, ultimately resulting in higher grain yields.

Despite these advancements in weed control, no significant differences in grain yield were observed across the various crop establishment systems within the short duration of the experiment. This lack of variation could be attributed to the limited time frame of the study as changes in soil health and productivity often require a longer period to manifest. The physical and biological properties of soil, including soil structure, microbial activity, and nutrient availability, evolve gradually over time (Kumar et al., 2023). Moreover, the accumulation of soil organic carbon, which is vital for sustaining long-term soil fertility and crop productivity, may take several years of consistent management practices to reach levels that significantly influence yields (Bana et al., 2023a). It is plausible that sustained experimentation over 4–5 years or more would reveal significant differences in grain yield among the crop establishment systems (Bana et al., 2020, 2023b). Continuous implementation of practices such as residue retention, elevated bed planting, or no-tillage could progressively enhance soil properties, improve water-use efficiency, and build resilience against biotic and abiotic stresses, ultimately contributing to sustainable and higher crop yields (Govaerts et al., 2007).