Increasing crop rotational diversity can enhance cereal yields

Commodity cropping has generally focused on increasing yields while overlooking its contribution to environmental degradation, climate change, and biodiversity loss. Global food production, therefore, threatens people, the planet, and agriculture itself. Agroecological solutions that maintain crop yields while minimizing external inputs have been proposed as a way forward. Diversifying agriculture through crop rotation, using a greater number of crop species in sequence, is a promising practice. However, global staple crop production often relies on short rotations of two species, or even monoculture, resulting in regional crop diversity loss. Farmers have long known that grain yields decline with monoculture or short rotations, requiring high external inputs to maintain production. Soil fertility and nutrient use efficiency improve with increased species and functional group diversity, enhancing soil microbial biomass, water holding capacity, soil carbon, and nitrogen availability and uptake. Increased diversity also decreases weed, pest, and disease pressure, reducing the need for fertilizers and crop protection inputs. A recent meta-analysis found crop rotational diversity (CRD) resulted in higher yields with low nitrogen input, especially when legumes were present. However, the extent to which diverse rotations maintain grain yields, compensate for reduced fertilizer inputs over time, and affect different grain crop species remains unclear. Most CRD studies compare monoculture to diverse rotations, focus on single sites, are short-term, or include only a single or few species in their diverse rotation. Therefore, it’s unclear how crop production benefits and fertilizer dependency develop with increasing CRD, whether species diversity or functional group diversity is more impactful, and how these effects vary over time and under contrasting fertilization levels for different cereals. Much of our understanding of diversity-productivity relationships comes from long-term grassland experiments, demonstrating increased plant biomass production with increased plant species or functional groups. This increase is explained by species selection effects, niche differentiation, and facilitation between species (niche complementarity). It’s hypothesized that this would be replicated in arable ecosystems, with crop yields increasing with CRD, but this hasn’t been widely verified across diverse levels. Grassland experiments differ from arable ecosystems in several key aspects: species mixes, levels of nutrient input and soil disturbance, and type of diversification (spatial vs. temporal). These differences hinder direct translation of grassland findings to arable systems. Focusing on small grain cereals and maize, we hypothesized that diversifying rotations would raise rainfed grain yields within years of implementation, with yield benefits increasing gradually over time due to the buildup of ecosystem functions. We expected this to be stronger with higher functional group richness, but also that specific functional groups would be more influential than others. We also expected CRD benefits to be higher under lower external nitrogen fertilization rates.

The introduction thoroughly reviews existing literature on the impacts of crop rotation diversity on yields, focusing on the gaps in knowledge regarding the long-term effects, the influence of different functional groups, and the interaction with nitrogen fertilization levels. It highlights the limitations of previous studies, such as their short durations, limited scope of species and diversity levels, and the challenges of extrapolating from grassland experiments to arable systems. The authors correctly point out that most studies haven't explored the full spectrum of CRD levels and have not differentiated between the role of species diversity and functional group richness. The literature review also sets the stage for the hypothesis of the current study, emphasizing the expected long-term benefits of CRD, its greater effectiveness at lower nitrogen levels, and the potential influence of specific functional groups.

To quantify the relationship between cereal yield and crop rotational diversity (CRD) over time, the researchers collected 27,460 rainfed, annual crop yield observations from 32 long-term experiments (10–63 years) across North America and Europe. The experiments were selected based on two criteria: availability of yield data from at least two rotation treatments with different CRD levels for a focal crop species (maize in North America and spring/winter small grain cereals in Europe) and comparable management practices across the different treatments. CRD was measured using two metrics: species diversity (a modified inverse Simpson's diversity index) and functional richness (number of functional groups—annual cereals, annual legumes, annual broadleaf crops, and ley). External nitrogen (N) input was classified as 'low' (below local recommendations) or 'high' (at or above recommendations). Mean-centered yields for each indicator crop were calculated within each site, accounting for the long-term within-site average across all CRD treatments and N input levels. Gaussian mixed-effects models were used to analyze the mean-centered yields, including CRD, time, and fertilization level as fixed effects, and interactions between these factors. Model selection was based on the Akaike Information Criterion (AIC). To assess the effect of specific functional groups, separate models were run using binomial variables for the presence of legumes, ley, and broadleaf crops. Calendar year was included as a random effect to account for technological advances, and site was a random effect nested within each experiment. Residual diagnostics were used to assess the models' performance. The data were analyzed separately for the three indicator crop groups due to geographical and response differences reported in previous studies.

The study found that increased crop rotational diversity (CRD), measured by both species diversity and functional richness, significantly enhanced grain yields across all three indicator crops (spring small grain cereals, maize, and winter small grain cereals). The yield benefits increased over time, with the most substantial gains observed after 35 years. For instance, with low external nitrogen, the maximum yield gain after 35 years was 0.94 t/ha for spring small grain cereals, 1.32 t/ha for winter small grain cereals, and 4.19 t/ha for maize. However, winter small grain cereal yields showed a decline at the highest species diversity levels, both at low and high nitrogen inputs. The benefits of CRD were more pronounced under low external nitrogen input, especially for maize, suggesting that CRD enhances nutrient use efficiency. The analysis of functional richness indicated that rotations with two to three functional groups generally produced the highest yields. Specific functional groups, such as annual legumes (nitrogen-fixing crops), significantly contributed to the yield enhancement. The yield benefits from CRD were greater under low nitrogen inputs for all crops, particularly maize. Comparing the yield benefits from diversified rotations (at yield-maximizing diversity) with low nitrogen input to monocultures with high nitrogen input showed that, while high N monocultures had initially higher yields, the diversified rotations eventually surpassed the monoculture yields, highlighting the long-term benefits of diversification. This result underscores the potential for CRD to reduce nitrogen fertilizer dependence and associated environmental impacts. The models showed relatively low R-squared values, indicating that other factors beyond CRD influence yields. However, despite the limitations of the low R-squared values and the large confidence intervals, the robust and consistent trends observed across the study highlight the significance of CRD.

The findings address the research question by demonstrating a clear positive relationship between crop rotational diversity and cereal yields. This is particularly evident under low external nitrogen fertilization, suggesting that CRD enhances nutrient use efficiency and possibly other ecosystem services like pest regulation. The observed increase in yield benefits over time highlights the importance of long-term perspective in evaluating the effectiveness of CRD. The finding that functional richness often provides greater benefits than species diversity alone underscores the importance of including functionally diverse crops in rotations. The results support the hypothesis that CRD enhances nutrient-mediated benefits. Other secondary mechanisms could include improved pest regulation or soil water holding capacity. The relatively low R-squared values point to the complexity of factors influencing yields, emphasizing the need for site-specific assessments. The long-term benefits of CRD suggest it could be a key strategy in sustainable agriculture.

This study provides compelling long-term evidence that increasing crop rotational diversity enhances cereal yields, particularly under low nitrogen inputs. The benefits accrue over time and are often more strongly related to functional richness than species diversity. The results demonstrate the potential of CRD to improve nutrient use efficiency and reduce reliance on synthetic fertilizers, contributing to more sustainable agricultural practices. Future research should focus on identifying optimal combinations of species and functional groups for maximizing yields under different environmental conditions and exploring the relative contributions of different mechanisms underlying the CRD effects, such as improved nutrient cycling, pest regulation and water use efficiency.

The relatively low R-squared values in the statistical models indicate that factors beyond crop rotational diversity significantly influence grain yields. The study's large spatial and temporal scales, although beneficial for generalizability, might have introduced some heterogeneity. The unbalanced design of some experiments, with an unequal representation of diversity levels and functional groups, limited the ability to fully disentangle the effects of species diversity and functional richness. Differences in management practices across experiments might also have influenced the results, especially concerning pest and weed control. Finally, the economic and practical aspects of implementing diverse rotations were not fully addressed.