Diversifying crop rotation sequences with grain legume/cereal intercrops and legume/grass cover crops

Abstract

Diversification of cropping systems creates multiple benefits for the sustainable development of agroecosystems. Notably, introducing intercrops and cover crops in crop rotations allows efficient use of resources and promotes the synergy between ecosystem processes and functions (e.g. enhancing soil quality, preventing risk of build-up of pests and diseases). We used field experiments to assess how grain yield, crop biomass and weed biomass in cereals were influenced by introducing grain legume sole crops and legume/ cereal intercrops as preceding crops, as well as by integrating cover crops during periods between the main crops. The results showed either insignificant or positive effects of the tested crop diversification measures on subsequent cereal productivity. Along with other known benefits such as reduced need for nitrogen fertilisation, reduced risk of nutrient losses and soil erosion, our findings support that increased crop diversity can improve resource use efficiency and contribute to more sustainable agriculture.

Full text

Aspects of Applied Biology 146, 2021
Intercropping for sustainability: Research developments and their application
43
Diversifying crop rotation sequences with grain legume/cereal
intercrops and legume/grass cover crops
By CAROLINA RODRIGUEZ, ERIK STEEN JENSEN and GEORG CARLSSON
Department of Biosystems and Technology, Swedish University of Agricultural Sciences,
Box 103, 230 53 Alnarp, Sweden
Summary
Diversification of cropping systems creates multiple benefits for the sustainable
development of agroecosystems. Notably, introducing intercrops and cover crops in crop
rotations allows ecient use of resources and promotes the synergy between ecosystem
processes and functions (e.g. enhancing soil quality, preventing risk of build-up of pests
and diseases). We used eld experiments to assess how grain yield, crop biomass and weed
biomass in cereals were inuenced by introducing grain legume sole crops and legume/
cereal intercrops as preceding crops, as well as by integrating cover crops during periods
between the main crops. The results showed either insignicant or positive eects of the
tested crop diversication measures on subsequent cereal productivity. Along with other
known benets such as reduced need for nitrogen fertilisation, reduced risk of nutrient
losses and soil erosion, our ndings support that increased crop diversity can improve
resource use eciency and contribute to more sustainable agriculture.
Key words: Crop diversication, preceding crop, cover crops, intercropping, legumes
Introduction
Diversifying conventional farming systems and implementing alternative crops, creates multiple
benets for agro-ecosystems and can reduce the negative environmental impacts of agriculture.
Indeed, increasing crop diversication using legumes can enhance land productivity, reduce the
need for nitrogen (N) fertiliser inputs and associated fossil energy use (Jensen et al., 2010), and
prevent the build-up of pests and diseases (Lin, 2011). Furthermore, crop diversication supports
the expansion of new niche markets, reducing the economic risks for farmers, thus improving the
local economy (Meynard et al., 2018) and stimulating more resilient farming and food systems
(Bullock et al., 2017). To investigate how crop diversication inuences the performance of a crop
sequence, with focus on the subsequent cereal crop, this paper presents results from three dierent
eld experiments. The experiments were designed to assess the eects of grain legume sole crops
and grain legume-cereal intercrops as preceding crops, with and without the integration of cover
crops, on subsequent cereal crop grain yield, crop biomass and weed biomass.
Materials and Methods
Three eld experiments were carried out at SITES Lönnstorp research station, SLU, Alnarp
(55.65N, 13.06E) in a loam soil in 2015–2017 (Fig.1). Each experiment integrated several degrees
of crop diversication. Experiment 1 (Exp. 1) included the forage legumes lucerne (Medicago

44
Fig. 1. Illustration of the crop sequences in eld experiments. Precrop (preceding crops) were: faba bean,
pea and oat sole crops in Exp. 1; sole crops and grain legume-cereal intercrops of faba bean, pea and oat in
Exp. 2; and sole crops and grain legume-cereal intercrops of faba bean, pea (including a pea varietal mixture)
and wheat in Exp. 3. Cover crops included lucerne (Exp. 1 and 2), red clover (Exp. 1 and 2), legume-grass
mixtures (Exp. 1) and oil radish (Exp. 3).
sativa L.; CCluc) and red clover (Trifolium pratense L.; CCRC) and forage legume-grass mixtures
(a 50/50 mixture of lucerne and cocksfoot, Dactylis glomerata L.; CCgrass-luc, and a 50/50 mixture
of red clover and timothy, Phleum pratense L.; CCgrass-RC) under-sown in the preceding crop. The
preceding crop was a sole crop of either faba bean (Vicia faba L.), pea (Pisum sativum L.), or oat
(Avena sativa L.). Thereafter, the cover crops were left in the eld and their aboveground biomass
harvested twice during the subsequent growing season before they were destroyed mechanically
direct before establishing winter wheat (Triticum aestivum L.). Experiment 2 (Exp. 2) included forage
legumes (CCluc and CCRC) under-sown in the preceding crop (sole crops of faba bean, pea and oat, and
intercrops of each grain legume and oat), and destroyed in the subsequent spring before establishing
spring barley (Hordeum vulgare L.). Finally, in experiment 3 (Exp. 3), oil radish (Raphanus sativus
L.) was sown after harvest of the preceding crop (sole crops of faba bean, pea and spring wheat, a
mixture of two pea varieties (PeaVar-mix), and intercrops of the grain legumes and spring wheat), and
destroyed in the subsequent spring before establishing spring barley. The subsequent cereals were
fertilized during spring 2017 with 100 kg N ha-1 (winter wheat) and 60 kg N ha-1 (spring barley).
In Exp. 1, preceding crops and undersown cover crops were sown on 22 April 2015. The preceding
crops and cover crops in Exp. 2 were established on 22 April 2016, which was the same date as
the establishment of the preceding crops in Exp.3. Both subsequent cereals received a herbicide
treatment (Ariane S, 2 L ha-1) while the spring barley also received a fungicide treatment (Flexity,
0.3 L ha-1) during May 2017. Each experiment was a complete randomized block design with four
replicates and consisted of 12 m × 2 m plots for each crop sequence. Half of each plot surface (6
m × 2 m) was sown with the oil radish cover crop in Exp. 3, to compare crop and weed biomass
in subsequent spring barley with and without preceding cover crop. All the experiments were part
of the European project LEGATO (Legumes for the Agriculture of Tomorrow). Crop grain yield,
crop biomass and weed biomass were measured at harvest of the subsequent cereal. In Exp. 3,
grain yield of spring barley was measured in the entire experimental plot, without distinguishing
the eect of the cover crop. Analyses of variance were performed, with signicance level P<0.05
on linear mixed eect models R package lmerTest (Kuznetsova et al., 2017). Preceding crops and
cover crops were considered as xed eects while block was considered as random eect. Multiple
comparisons were tested using Tukey HSD method.
Results
There was no signicant eect of the preceding crops and cover crops on crop or weed biomass or
grain yield in the subsequent winter wheat in the Exp.1 (Table 1). On the contrary, the cover crops
clearly inuenced grain yield of spring barley in Exp. 2. For all the treatments, the grain yield was

45
signicantly higher (between 3–15%) in the treatments including forage legumes (CCluc or CCRC)
as cover crops compared to treatments without cover crop (Table 1). Preceding crop, on the other
hand, did not have any signicant eect on subsequent barley biomass or grain yield in Exp. 2.
Weed biomass in the spring barley crop was relatively low, and not aected by the preceding crop
or cover crop treatments (Table 1).
Table 1. Dry matter yield of subsequent cereals and weed biomass in crop sequences of
experiments 1 and 2 during 2015–2017. Data are presented as means (n = 4) with standard
errors (± SE). Means with dierent letters indicates signicant dierence at P<0.05 (uppercase
latter indicates signicance eect of cover crops). The abbreviations indicate treatments with no
cover crop (NoCC), lucerne cover crop (CCluc), red clover cover crop (CCRC), lucerne-cocksfoot
mixture cover crop (CCgrass-luc) and red clover-timothy mixture cover crop (CCgrass-RC)
Aboveground biomass and grain dry matter yields (Mg ha-1)
Exp. 1 Exp. 2
Wheat Weed Barley Weed
Treatments Biomass Grain Biomass Biomass Grain Biomass
Faba NoCC 10.1±0.9 5.8±0.4 2.0±0.3 10.1±0.2 6.7±0.2B 0.2±0.1
Faba + CCluc 11.5±1.6 6.1±0.9 1.8±0.6 10.5±1.7 7.3±0.1A 0.1±<0.1
Faba + CCRC 12.7±2.3 6.1±0.7 1.5±0.5 10.7±0.6 7.2±0.2AB 0.1±<0.1
Faba + CCgrass-luc 12.1±2.2 6.2±0.8 1.6±0.5
Faba + CCgrass-RC 11.2±1.9 5.4±0.8 1.8±0.5
Pea NoCC 12.7±1.9 6.4±0.6 1.3±0.6 8.6±0.7 6.4±0.3B 0.3±0.1
Pea + CCluc 9.1±1.2 4.7±0.2 2.5±0.2 11.3±0.2 7.1±0.2A 0.2±<0.1
Pea+ CCRC 10.6±1.6 5.0±0.7 2.1±0.6 10.0±0.9 7.1±<0.1A 0.3±0.2
Pea + CCgrass-luc 11.6±0.7 6.0±0.6 1.7±0.3
Pea + CCgrass-RC 9.1±0.8 4.8±0.3 2.4±0.3
Oat NoCC 12.4±1.0 6.4±0.8 1.1±0.2 9.2±1.0 6.6±0.2B 0.3±0.2
Oat + CCluc 11.0±0.7 5.3±0.5 2.0±0.3 10.9±0.7 6.8±0.1AB 0.2±0.1
Oat + CCRC 7.9±1.2 4.8±0.6 2.7±0.4 11.5±1.0 7.2±0.1A 0.2±0.1
Oat + CCgrass-lec 11.2±1.9 5.9±0.7 1.6±0.5
Oat + CCgrass-RC 9.5±1.4 4.8±0.4 3.0±0.5
Faba/Oat NoCC 9.9±0.8 6.4±0.1B 0.1±<0.1
Faba/Oat + CCluc 9.4±0.3 6.9±0.2AB 0.2±0.1
Faba/Oat + CCRC 9.6±0.4 7.4±0.2A 0.1±<0.1
Pea/Oat NoCC 8.5±0.2 6.5±0.1B 0.2±0.1
Pea/Oat + CCluc 11.3±0.4 7.0±0.1A 0.2±0.1
Pea/Oat + CCRC 11.2±1.5 6.9±0.2A 0.3±0.1

46
For most of the treatments in Exp. 3, preceding crops and cover crops showed an eect on the dry
weight biomass of the spring barley (Table 2). In the case of preceding crops followed by the oil
radish cover crop, spring barley biomass was greater after faba bean (12.6 Mg ha-1) than after wheat
(10.4 Mg ha-1) and faba bean/wheat (10 Mg ha-1). In terms of the cover crop eect, the spring barley
biomass was 9–20 % higher in treatments with the cover crop, and the dierence was signicant
in most of the treatments (the exception was after the faba bean-wheat intercrop). Grain yield after
faba bean and pea varietal mixture was signicantly higher than after wheat and faba bean/wheat.
Considering weed biomass, only wheat with cover crop showed a signicant reduction of the weed
biomass in the subsequent cereal compared to the same preceding crop without cover crop.
Table 2. Dry matter yield of subsequent cereal and weed biomass in crop sequences of
experiment 3 during 2016–2017. Data are presented as means (n = 4) with standard errors
(± SE). Means with dierent letters indicates signicant dierence at P<0.05 (uppercase
letters indicate signicant eects of cover crops, while lowercase letters indicate eects of the
preceding crop). The abbreviations indicate treatments with no cover crop (NoCC), oil radish
cover crop (CCOR) and mixture of two pea varieties (PeaVar-mix)
Aboveground biomass and grain dry matter yields (Mg ha-1)
Exp. 3
Barley Weed
Treatments Biomass Grain Biomass
Faba NoCC 10.7±0.5B 7.0±0.1a 0.1±<0.1
Faba + CCOR 12.6±0.2Aa 0.1±<0.1
Pea NoCC 9.9±0.6B 6.8±0.2ab 0.2±0.1
Pea+ CCOR 11.4±0.4Aab 0.1±0.1
Wheat NoCC 9.5±0.7B 6.1±0.3b 0.2±0.1A
Wheat + CCOR 10.4±0.3Ab <0.1±<0.1B
Faba/Wheat NoCC 9.8±0.4 6.1±0.1b 0.2±<0.1
Faba/Wheat + CCOR 10.0±0.4b 0.3±0.1
PeaVar-mix NoCC 10.2±0.6B 6.9±0.2a 0.2±<0.1
PeaVar-mix + CCOR 11.6±0.4Aab 0.1±0.1
PeaVar-mix/Wheat NoCC 9.0±0.2B 6.5±0.1ab 0.2±<0.1
PeaVar-mix/Wheat + CCOR 10.8±0.4Aab 0.2±0.1
Discussion
Our results conrmed the positive eect of diversifying cereal-based cropping systems by including
grain legumes or legume-based cover crops in the crop sequence. Legumes are known to increase the
total pool of soil nitrogen though the addition of biologically xed N in crop residues (Hauggaard-
Nielsen et al., 2009). This N input can improve the yields of subsequent crops in the rotation if
the increased soil N availability is in synchrony with the N demand of the subsequent crop. The
results from Exps 2 and 3 in our study showed several cases of increased cereal grain yields after
legume cover crops or grain legume preceding crops, which indicates that more soil N might have
been available to the cereal in these treatments than when no legume was present either as cover
crop or as preceding main crop. However, the amount of soil mineral nitrogen being available for

47
the subsequent cereal crop may vary depending on the cover crop treatment, with lower levels in
mixtures of forage legumes and grasses compared to sole legume cover crops, since grasses may
diminish net nitrogen mineralisation. A potential tradeo thus exists between growing pure legume
cover crops for the benet of subsequent cereal productivity, and growing non-legume or mixed
cover crops to reduce the risk of N losses, Valkama et al. (2015) showed that pure legume cover
crops do not reduce N leaching.
There was an almost consistent positive eect of the oil radish cover crop on biomass production
of subsequent spring barley. Even though we could not distinguish potential eects of oil radish on
grain yield of the subsequent spring barley, the positive eect on barley biomass is an indication
of increased productivity that is likely to also increase grain yield. Oil radish established after the
harvest of a main crop and growing during autumn is known to reduce N losses (Aronsson et al.,
2016), and it is possible that mineralisation of the frost-killed oil radish biomass leads to increased
soil N availability for the subsequent spring cereal compared to when the soil is left uncovered
during autumn.
While previous research has shown that crop diversication through intercropping or integration
of cover crops can reduce the weed abundance in subsequent crops (Finney et al., 2016; MacLaren
et al., 2019), we did not nd any signicant reduction of weed biomass in the cereal crops grown
after the diversied crop sequences in our study. One reason for this may be that some of the
cover crop species tended to regrow in the subsequent crop, where the cover crop appeared as a
weed. Indeed, red clover was encountered among weeds in Exp. 1, which highlights the risk of
competitive cover crop species causing problems when reappearing as weeds later on in the crop
rotation (MacLaren et al., 2019).
Overall, our results showed that crop diversication through the integration of grain legumes,
intercrops and cover crops either had positive eects or did not inuence crop yields and weed
reduction in the studied crop sequences. Considering other known benets of crop diversication,
e.g. improved nitrogen use eciency, reduced risk of nutrient losses and positive eects on
associated biodiversity, also the lack of negative eects on crop yield and weed reduction should
be seen as an advantage in a whole-system assessment. In conclusion, increased crop diversity
across spatial and temporal scales can contribute to resource-ecient farming systems and reduce
economic risk for farmers, improving local economy and thus enhancing sustainable food systems.
Acknowledgements
The eld experiments used in this study were part of the LEGATO project (LEGumes for the
Ariculture of TOmorrow), which received funding from the European Union FP7 under grant
agreement no 613551. We acknowledge the SITES Lönnstorp Research Station for technical
management of the eld experiments.
References
Aronsson H, Hansen E M, Thomsen I K, Liu J, Øgaard A F, Känkänen H, Ulén B. 2016.
The ability of cover crops to reduce nitrogen and phosphorus losses from arable land in southern
Scandinavia and Finland. Journal of Soil and Water Conservation 71:41–55. http://dx.doi.org/
https://10.2489/jswc.71.1.41.
Bullock J M, Dhanjal-Adams K L, Milne A, Oliver T H, Todman L C, Whitmore A P, Pywell
R F. 2017. Resilience and food security: rethinking an ecological concept. Journal of Ecology
105:880–884. http://dx.doi.org/10.1111/1365-2745.12791.
Finney D M, White C M, Kaye J P. 2016. Biomass production and carbon/nitrogen ratio inuence
ecosystem services from cover crop mixtures. Agronomy Journal 108:39–52. http://dx.doi.org/
https://10.2134/agronj15.0182.

48
Hauggaard-Nielsen H, Mundus S, Jensen E S. 2009. Nitrogen dynamics following grain
legumes and subsequent catch crops and the eects on succeeding cereal crops. Nutrient Cycling
in Agroecosystems 84:281–291. http://dx.doi.org/https://10.1007/s10705-008-9242-7.
Jensen E S, Peoples M B, Hauggaard-Nielsen H. 2010. Faba bean in cropping systems. Field
Crops Research 115:203–216. http://dx.doi.org/10.1016/j.fcr.2009.10.008.
Kuznetsova A, Brockho P B, Christensen R H B. 2017. lmerTest Package: Tests in Linear
Mixed Eects Models. Journal of Statistical Software 82(13). http://dx.doi.org/https://10.18637/
jss.v082.i13.
Lin B B. 2011. Resilience in Agriculture through Crop Diversication: Adaptive Management for
Environmental Change. BioScience 61:183–193. http://dx.doi.org/10.1525/bio.2011.61.3.4.
MacLaren C, Swanepoel P, Bennett J, Wright J, Dehnen-Schmutz K. 2019. Cover Crop Biomass
Production Is More Important than Diversity for Weed Suppression. Crop Science 59:733–748.
http://dx.doi.org/https://10.2135/cropsci2018.05.0329.
Meynard J-M, Charrier F, Fares M H, Le Bail M, Magrini M-B, Charlier A, Messéan A.
2018. Socio-technical lock-in hinders crop diversication in France. Agronomy for Sustainable
Development 38:54. http://dx.doi.org/https://10.1007/s13593-018-0535-1.
Valk ama E , Le mol a R, K än kän en H , Turt ol a E. 20 15. Meta-analysis of the eects of undersown
catch crops on nitrogen leaching loss and grain yields in the Nordic countries. Agriculture,
Ecosystems and Environment 203:93–101. http://dx.doi.org/10.1016/j.agee.2015.01.023.