Edited by: Patricia Dorr De Quadros, University of Waterloo, Canada
Reviewed by: Esmaeil Rezaei-Chiyaneh, Urmia University, Iran; Sajjad Raza, Nanjing University of Information Science and Technology, China
This article was submitted to Soil Biology, Ecosystems and Biodiversity, a section of the journal Frontiers in Soil Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Intercropping is an ancient agricultural management practice quickly re-gaining interest in mechanized agricultural systems. Mechanized management practices have led to a decreased biodiversity at the macro- and micro-fauna levels. These agricultural practices have also resulted in the degradation of soil and long-term inefficiencies in land, water, and nutrients. The inland Pacific Northwest (iPNW) of the United States of America is a wheat-dominated cropping system. The integration of winter and spring legumes and oilseeds has improved the biodiversity and nutrient-use efficiency of the cropping systems. This article examines the feasibility of pea-canola (peaola) intercropping in dryland production systems of the iPNW. In two site years, small plot peaola trials were established near Davenport, WA. Overall, the land equivalence ratio (LER) of peaola was found to be 1.46, showing an increase in efficiency of the system. Increasing the N fertilizer application rates did not affect peaola yield, indicating that peaola has low demand for N inputs. The effects of peaola on insects and bacterial diversity were examined on replicated large scale strip trials. Peaola was found to have significantly greater numbers of beneficial insects than the monoculture controls. There were no significant differences between the diversity of the soil bacterial communities found in peaola vs. pea and canola monocultures. However, we found that the strict core soil bacterial microbiome of peaola was larger than the monocultures and included core members from both the canola and pea soil microbiomes. In conclusion, the widespread adoption of peaola would likely increase the biodiversity and increase the land use efficiency of dryland production systems in the iPNW.
Most industrial agriculture systems are monocultures with the only feasible option for increased diversity being crop rotation. Subsistence agriculture on the other hand has long relied on multispecies systems (
Peas (
Canola has been shown to be a useful tool for improving the water and nutrient efficiency as a nutrient catch crop and increasing the water infiltration when grown in rotation (
Due to the ability of legumes to host symbiotic bacteria that conduct biological N fixation, they have frequently been included in intercropping systems. A number of studies that assessed the effects of N rate on peaola productivity have been conducted (
In addition to the questions regarding N rate, intercropping systems should improve production through increasing the ecological diversity, increasing the resilience, and increasing the resource-use efficiency (
This study has three principal objectives. The first objective was to assess the LER of winter and spring peaola in the iPNW. The second was to assess the effect of N fertilizer rate on the relative yields of peas and canola as well as LER in winter peaola. The third objective was to assess the changes in insects and soil microbial communities in peaola vs. the corresponding monocultures. We hypothesize (
The small plot (1.7 × 9 m) trials were conducted near Davenport, WA, and were seeded and harvested with small plot research equipment. The small plot experiment was laid out using a randomized complete block design with four plot replicates per treatment combination. The two controls were monoculture canola with 67 kg N ha−1 and monoculture peas with 0 kg N ha−1. The three treatments were peaola at different N rates (0, 33, and 67 kg N ha−1). N applications were made using urea-ammonium nitrate and streamed on using a CO2 backpack sprayer in the spring. During the 2020 growing season, these applications were made just prior to precipitation. However, in the 2021 growing season, there was little to no spring precipitation and the fertilizer was simply applied in March.
A Fabro double disk no-till drill was used to seed the plots into no-till winter wheat chemical fallow. The winter pea variety Goldenwood (ProGene Plant Research, LLC) was used in both the monoculture and the intercropping plots, while Plurax (Rubisco Seeds) was used as the canola variety of choice. Both Goldenwood and Plurax have been successfully grown in the iPNW. The peas and canola were planted in the same row at the same time for both the 2020 and the 2021 cropping years. Typically, peas are planted later in the fall than canola in Eastern Washington. However, in this study, the planting date was a compromise between peas and canola with a late August planting date. Grassy weed herbicide applications were made in the spring of 2020 and 2021. The whole plot yield was sampled, and the peas and canola were separated using an M-2B clipper mill from A. T. Ferrell & Company Bluffton, Indiana.
The large-scale replicated spring peaola strip trials (11 × 61 m) were established on 9 April 2020 near Colfax, WA. The large-scale strip trials included 4 replicates of canola monoculture, pea monoculture, and peaola. Placement of replicates was randomized. The strips were direct seeded into stubble from the previous year's winter wheat crop using a no-till Cross Slot drill. A winter pea variety (Goldenwood from ProGene Plant Research, LLC) was used, as there was a concern that early and aggressive growth of spring peas would outcompete the early stages of canola growth, The spring-type canola was a Clearfield canola variety from DynaGro 200 CL. Fertilizers were applied in furrow at planting with 101 kg N ha−1 applied to the canola monoculture, 51 kg N ha−1 applied to the peaola, and 0 kg N ha−1 applied to the monoculture peas. Beyond (imazomox) and select (clethodim) herbicides were applied in late May. The strips were harvested on 14 September of 2020 and weighed using a weigh wagon. As the harvested pea-canola mix was dumped into the weigh wagon, a small amount (~1 kg) was sampled from the grain stream using a polyvinyl chloride (PVC) pipe. The peas and canola were separated using an M-2B clipper mill from A. T. Ferrell & Company Bluffton, Indiana. The peas and canola were then weighed individually and were applied to the overall grain yield which was used to calculate the relative pea and canola yield on a per hectare basis.
Land equivalence ratio was calculated using Equation 1, where
Insect samples were collected at the Colfax location in the spring of 2020 as the small plot experiments in Davenport were not considered suitably large enough to conduct an adequate insect sampling. The large-scale strips were oriented roughly north to south lengthwise. The insect samples were taken 10 m from the north end of the strips and 3 m east side of the plot to ensure a uniform sampling location between plots. The insects were identified and categorized into a functional group. The functional groups were pollinators (Hymenoptera in the Apoidea superfamily and Diptera in the Syrphid family), parasites (Hymenoptera in the families Braconidae and Ichneumonidae), predators [Araneae (spiders) and Coleoptera in the family Coccinellidae (ladybeetles)], and herbivores (Hemiptera in the families Aphidae, Miridae, and Pendatomidae, all larval Lepidoptera, and Coleoptera in the family Curculionidae). For analyses, these pollinators, parasites, and predators were classified broadly as beneficial arthropods, while the herbivores were classified as pest arthropods.
Soil samples were collected on 14 July 2020 from the large-scale strip trials located near Colfax, WA, to a depth of 10 cm. This corresponded with early flowering of canola. Three samples were taken within each replication (four canola monoculture, four pea monoculture, and four peaola) toward the middle of the plot, resulting in twelve samples per treatment. Once collected, the samples were put in a cooler and transported to WSU where they were kept at −20°C until DNA extraction.
The DNA was extracted using a Kingfisher DNA extraction machine following the Earth Microbiome Project's protocol for the QIAGEN® MagAttract® PowerSoil® DNA KF Kit. A no-soil blank was added to each extraction plate to control for cross-contamination. A high-sensitivity dsDNA quantification was performed using a Qubit following the manufacturer's protocol. The amplification of the 16S V4 region was done using the primers 515F: 5′-GTGCCAGCMGCCGCGGTAA-3′ and 806R: 5′-GGACTACHVGGGTWTCTAAT-3′ using the Thermo Scientific DreamTaq DNA Polymerase following the manufacturer's instructions. The thermocycler program was denaturation at 95°C for 3 min; 30 cycles of 95°C for 45 s, 50°C for 60 s, and 72°C for 90 s; final elongation at 72°C for 10 min; and an infinite hold at 15°C. An agarose gel electrophoresis was performed to confirm the presence of correctly sized amplicons at ~300 bp.
The DNA was sent to Michigan State University's Research Technology Support Facility for an Illumina Amplicon sequencing of the 16S V4 region on the MiSeq v2 Standard platform, resulting in 250-bp paired end reads. The ZymoBIOMICS Microbial Community Standard II (Log Distribution) was included in place of our extraction negative. A negative control was added by the sequencing center.
The sequences were analyzed using the QIIME2 version 2021.8 on WSU's Kamiak High Performance Computing Cluster. The bacteria were classified using “qiime feature-classifier classify-sklearn” with the Silva 138 99% OTUs from 515F/806R classifier found on the QIIME2 data resources page. The mitochondria and chloroplasts were filtered out before analyzing the diversity metrics. The samples were analyzed at a depth of 10,201 in QIIME2 to determine the diversity of the microbial communities. The analysis of our alpha-rarefaction plot proved to be sufficient in showing the full diversity of our samples. Identification of the bacterial core microbiome was done using “qiime feature-table core-features” with the mitochondria and chloroplasts filtered out. We chose to use the strict bacterial core microbiome with core members being present in 100% of the tested samples. Boxplots were made using the raw data generated by the QIIME2 version 2021.8 using the R version 4.0.3 in the RStudio version 1.2.5001.
Both canola and pea yields were significantly higher in 2020 than in 2021 (
Significance of year, cropping system, and
2020 | Canola | 67 | 2,198 | 0 | 1.00 |
2020 | Pea | 0 | 0 | 2,752 | 1.00 |
2020 | Peaola | 0 | 2,029 | 2,011 | 1.65 |
2020 | Peaola | 34 | 1,704 | 1,667 | 1.38 |
2020 | Peaola | 67 | 991 | 2,698 | 1.43 |
2021 | Canola | 67 | 933 | 0 | 1.00 |
2021 | Pea | 0 | 0 | 85 | 1.00 |
2021 | Peaola | 0 | 649 | 84 | 1.68 |
2021 | Peaola | 34 | 1,071 | 63 | 1.89 |
2021 | Peaola | 67 | 541 | 71 | 1.42 |
Year | . | ||||
Cropping system | |||||
NS | NS | NS | |||
Year X cropping system | NS | NS | |||
Year × |
NS | NS | NS |
Precipitation and temperature on a month-by-month basis for the 2019–2020 growing season and the 2020–2021 growing season. Spring (March, April, and May) precipitation was substantially higher in 2020 than 2021. Additionally, June and July average temperatures were warmer in 2021 than in 2020. This chart was developed using data from WSU AgWeatherNet (
Distribution of peaola LER across all treatments in 2020 and 2021. The year 2021, which had much lower yields, had a higher average LER across all peaola treatments.
The cropping system was shown to have a significant main effect on the canola yield, the pea yield, and the LER (
The relative yield of canola and peas in the canola system. The negative-sloped line from (0.1) to (1.0) indicates the LER of the monocultures, while the dashed line indicates to which degree the peas or canola are favored.
At the strip trial near Colfax, the LER of the peaola (1.37) was not significantly different from the LER of the monocultures (
Spring canola yield at Colfax location.
Peaola | 50 | 805 | 489 | 1.37 |
Canola | 101 | 778 | – | 1 |
Pea | 0 | – | 1,427 | 1 |
Cropping system | NS | NS |
Herbivores (mostly pea aphids) were significantly higher in pea-only plots (
Average counts of beneficial insects (and estimated standard errors) based on 2020 field survey. The bars with error bars that do not overlap are significantly different. Output estimates from negative binomial generalized linear mixed model.
Average counts of insect herbivores (and estimated standard errors) based on 2020 field survey. The bars with error bars that do not overlap are significantly different. Output estimates from negative binomial generalized linear mixed model.
The analysis of our microbial community standard revealed that we were able to detect the included bacteria at their appropriate abundance down to bacteria present at a relative abundance of 0.089%. The measures of α-diversity-Shannon diversity index, Observed Features, and Evenness-did not show any significant differences (
Results of Faith's Phylogenetic Diversity index for canola, pea, and peaola soils. The microbial communities in the monoculture pea soil are trending toward being richer than the monoculture canola soil as determined by the Faith's Phylogenetic Diversity index (Kruskal–Wallis Test, n1 = 12, n2 = 12,
The measures of β-diversity–Jaccard distance, Bray-Curtis distance, and unweighted UniFrac distance–did not show any significant differences (
Ordination showing the results of the Weighted UniFrac Distance comparison performed between the different cropping systems. A trend was found toward there being a difference in the community composition of the pea and canola monoculture soils (PERMANOVA,
When looking at the makeup of the strict bacterial core microbiome, we found that the peaola core microbiomes consisted of 34 members, the pea core microbiome consisted of 23 members, and the canola core microbiome consisted of 29 members (
Strict bacterial core microbiomes for canola, pea, and peaola.
Acidobacteriales | Acidobacteriales | |
Acidobacteriales | Acidobacteriales | Acidobacteriales |
Acidobacteriales | ||
Burkholderiales |
||
Gaiellales | ||
Frankiales | Gaiellales | |
Gaiellales | ||
Gaiellales | ||
Gaiellales | ||
Gaiellales |
Polyangiales |
|
Gaiellales | ||
Gaiellales | ||
Uncultured Acidobacteria | ||
Polyangiales |
Uncultured Acidobacteriales | |
Solirubrobacterales |
||
– | ||
– | ||
Uncultured Acidobacteria | Polyangiales |
– |
Uncultured |
– | |
Uncultured |
– | |
Solirubrobacterales |
– | |
– | – | |
– | Uncultured Acidobacteria | – |
– | Uncultured |
– |
– | Uncultured |
– |
– | – |
During the 2020–2021 cropping season, the location of the winter peaola trials experienced extreme meteorological drought when compared with the 2019–2020 cropping season (
Comparison the meteorological drought conditions from the last week of June in 2020 to the last week of June in 2021. The location of this trial in Davenport, WA, experienced a drought-free year during the 2020–2021 growing season. However, during the 2020–2021 growing season, Davenport experienced extreme drought. These maps were adapted from
The peaola system did not appear to benefit from increasing the rates of synthetic N fertilizer in either 2020 or 2021. In a review of legume-oilseed intercropping, Dowling et al. (
The lack of a response to increasing N rate should not be interpreted as conclusive evidence that peaola negates the need for N fertilization. A positive crop response to fertilizer inputs is dependent on the fertilizer being a limiting factor in production. As noted above, in 2021, the crop yield was most likely limited by extreme weather events rather than N supply. However, the lack of a positive effect from N fertilizer in 2019–2020 requires further explanation. Previous studies conducted in the region have shown that monoculture winter canola does not always respond to increasing N applications in a manner that would be expected (
Future research may choose to address the potential for transfer of N from peas to canola to be able to determine if plant-plant-microbe interactions are responsible for the increased LER with decreased synthetic N inputs of peaola. Such research would likely require the use of stable isotopes. Regardless of whether the N is transferred from the peas to the canola during the peaola cropping year, incorporating peaola as opposed to monoculture canola should provide rotational N, thereby reducing the dependence of the entire cropping systems on synthetic N. The reduced need for synthetic N inputs in the peaola system will serve to increase the adaptive capacity of the overall cropping system to the economic and supply chain stress, which may impact the availability and cost of synthetic fertilizers.
The relative yield of peas to canola showed that the winter peaola systems did not strongly favor either pea or canola yield (
The abundance of insects by class was shown to shift based on intercropping at Colfax in 2020. Peaola was shown to have significantly greater numbers of beneficial insects among the monoculture systems. Meanwhile, both the canola and the peaola had significantly lower herbivores than the pea monoculture. Whether or not these shifts in populations result in higher economic thresholds for pest insects in peaola over peas cannot be determined from these data. However, future work should look at developing the economic thresholds for insecticide applications in peaola as compared to canola and peas.
Considering that we saw no significant differences between the diversity of the peaola soil bacterial community and the soil bacterial communities of pea and canola, it can be concluded that intercropping did not increase the diversity of the soil bacterial community. This is not surprising, however, since in previous studies done in other intercropping systems, only slight changes were observed in the soil bacterial community (
To begin to test the hypothesis that canola can interact with microorganisms that are not normally available to it in monoculture, we will need to determine how the rhizosphere and root microbiomes are changing. In studies that have focused on how intercropping impacts the diversity of the bacterial community in the rhizosphere and root microbiomes, it has been found that they experience an increase in their diversity compared to their monoculture counterparts (
Peaola is a promising production strategy for the iPNW and other regions dominated by large-scale mechanized monoculture agriculture (
While not originally set forth as an objective, one of the most important findings from this trial is the role of peaola in apparent resistance to drought and heat stress. The data from 2021 highlight that in a year with drought and heat stress where one crop fails (peas), an intercropping scheme can provide a more productive system. Since extreme weather events cannot be easily predicted, planting intercrops can be considered a means of increasing adaptive capacity of the system or insuring potential loss. Previous research conducted on sunflower-soybean intercrops increasing moisture was shown to increase the LER (
The data presented in the study are deposited in the NCBI repository, accession number
IM designed and carried out the field trials as they relate to field agronomy, wrote the first draft of the manuscript, analyzed the yield, and LER data. MF and JP were responsible for designing and carrying out the microbial sampling plan as well as analyzing the microbial data and writing the sections regarding the soil microbiology. RC was responsible for carrying out the entomological sampling conducting the data analysis on the entomological data. All authors contributed to the article and approved the submitted version.
This research was supported in part by the WSU Center for Sustaining Agriculture and Natural Resources BIOAg Program (USDA-NIFA #2020-67013-30864 and USDA-NIFA-SARE #GW21-228).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.