Received: 20 February 2021 Accepted: 15 June 2021 Published online: 28 August 2021 DOI: 10.1002/agj2.20781 A R T I C L E C r o p E c o n o m i c s , P r o d u c t i o n , a n d M a n a g e m e n t Value of integrating perennial forage seed crops in annual cropping sequences Nityananda Khanal1 Rahman Azooz3 Noabur Rahman1 Henry Klein-Gebbinck1 Jennifer K. Otani1 Calvin L. Yoder2,4 Talon M. Gauthier2 1 Agriculture and Agri-Food Canada, Beaverlodge Research Farm, Beaverlodge, AB, Canada 2 Peace Region Forage Seed Association, 904-102 Ave., Dawson Creek, BC, Canada 3 Retired, Agriculture and Agri-Food Canada, Beaverlodge Research Farm, Beaverlodge, AB, Canada 4 Retired, Alberta Agriculture and Forestry, Alberta, Canada, Current Address: Peace Region Forage Seed Association, 904-102 Ave., Dawson Creek, BC, Canada Correspondence Nityananda Khanal, Agriculture and Agri- Food Canada, Beaverlodge Research Farm, Beaverlodge, Alberta, Canada. Email: nityananda.khanal@agr.gc.ca Assigned to Associate Editor Terry Griffin. Abstract Cropping sequences integrating perennial forages and annual crops can generate mul- tidimensional agroeconomic and environmental benefits. A 4-yr cropping sequence study was conducted from 2013 to 2016 to evaluate the relative merits of various cropping sequences. Three forage seed crops (creeping red fescue [Festuca rubra L.], alsike clover [Trifolium hybridum L.], and red clover [Trifolium pratense L.]) and four annual food crops (wheat [Triticum aestivum L.], canola [Brassica napus L.], barley [Hordeum vulgare L.], and pea [Pisum sativum L.]) were used to gener- ate eight different cropping sequence treatments, which were split into three levels of nitrogen (N): 0, 45, and 90 kg ha−1. The seed yields of different crops were expressed as canola equivalent yield (CEY) and summed for 4-yr cropping sequences. Eco- nomic benefits for 4-yr cropping sequences were calculated using the input–output price scenarios of individual years. The highest CEY, gross return, and gross margin over partial variable costs were obtained from creeping red fescue-based cropping sequences, irrespective of N application rates. Based on CEY and gross margins, the merit order is ranked as creeping red fescue-based sequences > high diversity annual crop sequences ≈ alsike clover-based sequence ≈ wheat–canola alternating sequence > continuous canola sequence. The clover-based sequences had positive effects on the performance of succeeding crops of wheat and canola. Among the annual crop-based sequences, economic profitability increased with the diversifica- tion from continuous canola through wheat–canola alternation to pea–barley–wheat– canola in the sequence. Further study investigating soil health and greenhouse gas emissions parameters is warranted to elucidate multidimensional agroecosystem ser- vices of diversified perennial forage seed crop-based cropping sequences compared with annual crops-based sequences. Abbreviations: AC, alsike clover; Bar, barley; BCR, benefit–cost ratio; BD, bulk density; Can, canola; CEY, canola equivalent yield; CF, creeping red fescue; In, inoculated; MBCR, marginal benefit–cost ratio; NUE, N use efficiency; OM, organic matter; RC, red clover; SED, standard error of differences of means; Sn×N, interaction between cropping sequences and N; Sn, cropping sequences; Whe, wheat. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2021 The Authors. Agronomy Journal published by Wiley Periodicals LLC on behalf of American Society of Agronomy 4064 wileyonlinelibrary.com/journal/agj2 Agronomy Journal. 2021;113:4064–4084. https://orcid.org/0000-0003-3847-3930 mailto:nityananda.khanal@agr.gc.ca http://creativecommons.org/licenses/by-nc-nd/4.0/ https://wileyonlinelibrary.com/journal/agj2 http://crossmark.crossref.org/dialog/?doi=10.1002%2Fagj2.20781&domain=pdf&date_stamp=2021-08-28 KHANAL ET AL. 4065 1 INTRODUCTION The Peace River region of western Canada is a northern agri- cultural frontier undergoing changes both in extensive mar- gin through clearing new land for farming (Hamley, 1992; Haarsma, 2014) and in intensive margin through cropping systems intensification with tight rotation of wheat (Triticum aestivum L.) and canola (Brassica napus L.) (Gill, 2018). The agroecosystems in the crust-forming, runoff prone, platy- structured Luvisol (Lavkulich & Arocena, 2011) are under increasing pressure with high reliance on external inputs and low diversity monoculture. The perennial forage seed crops that constituted cornerstone pioneer conditioners for fragile, poorly developed Luvisolic soil (Broersma et al., 1997; Guitard, 1965; Stewart, 1933) and break crops in the cropping sequences are losing ground (Wong, 2015) in favor of few annual crops. The loss of functional diver- sity in the cropping systems, coupled with global warming, can partly be attributed to the rapid evolution of herbicide- resistant weeds (Beckie et al., 2019) and outbreaks of crop diseases such as club-root [caused by Plasmodiophora bras- sicae Woronin, teleomorph Gibberella zeae (Schw.) Petch] in canola (Strelkov et al., 2020) and Fusarium head blight (caused by Fusarium graminearum Schwabe) in wheat (Turk- ington et al., 2011) in the region. Although there is a per- ception of higher economic returns from wheat and canola, the integration of forage seed crops in cropping sequences can enhance resilience and sustainability of the production systems. A number of studies conducted in the Peace River region have shown diverse benefits of integrating forage crops in the cropping sequences. Those benefits include soil conser- vation (Van Vliet & Hall, 1991), soil structure development (Broersma et al., 1997; Pawluk, 1980), nitrogen fixation and mineralization (Broersma et al., 1996; Rice, 1980), and break in crop disease cycle (Soon et al., 2005). Therefore, these ben- efits have yield-enhancing effects on succeeding crops (Hoyt, 1990; Hoyt & Leitch, 1983). By virtue of high root/shoot ratio and perennial growth (Bolinder et al., 2007; Thiagarajan et al., 2018), forage seed crops can provide multiple agroecologi- cal services in the fragile Luvisolic soils. However, there is a lack of studies integrating cumulative systems productivity and comparative economic advantage of long-term cropping sequences including forage seed crops and annual field crops. To bridge the knowledge gap, a study was designed to eval- uate the efficiencies of eight different crop sequences com- prising three forage seed crops (creeping red fescue [CF] [Fes- tuca rubra L.], alsike clover [AC] [Trifolium hybridum L.], and red clover [RC] [Trifolium pratense L.]) and four major annual food crops (wheat [Whe] [Triticum aestivum L.], bar- ley [Bar] [Hordeum vulgare L.], canola [Can], and pea [Pisum sativum L.]). The objective of the study was to assess cumu- lative systems productivity and comparative economic advan- Core Ideas ∙ Integration of perennial forage seed crops in crop- ping sequences can be productive and profitable. ∙ Prevailing commodity prices have bearing on the relative profitability of the crops. ∙ Forage legumes can replace external N application for two succeeding crops. ∙ Complementary studies help to elucidate agroe- cosystem services of perennial seed crops. tage of long-term cropping sequences that included forage seed crops and annual field crops. The underlying hypothesis of the study was that an increase in diversity in the cropping sequence and integration of perennial legume and grass seed crops results in an increase in cumulative systems productiv- ity in terms of seed and biomass yields, and enhanced prof- itability in terms of higher gross revenue and lower variable costs, hence higher gross margins. Informed evidence-based cropping sequences tailored to the coexistence of forage seed crops and annual crops hold promise for resilient crop farming able to withstand production risks while contributing to a sus- tainable agri-industry. The study outcomes can be useful for designing productive, agroecologically resilient, environmen- tally friendly, and economically profitable cropping systems, which are readily adaptable by farmers in the Peace River region and other forage seed producing regions of the prairie provinces of Canada. 2 MATERIALS AND METHODS 2.1 Study location, soil, and weather The study was conducted at Beaverlodge Research Farm of Agriculture and Agri-Food Canada (55˚12′ N, 119˚24′ W) in the Peace River region of western Canada from 2013 to 2016. The landscape is a gentle east-facing slope with Dark Gray Luvisol of moderately fine texture, named Berwyn (Alberta Agriculture and Forestry, 2020). The baseline soil physico- chemical properties at the onset of the study in 2013 are pro- vided in Table 1. The agroclimatic conditions are characterized by a short growing season with 117 d of frost-free period (20 May– 15 September), long days (up to 17.5 h), warm day tem- peratures (monthly maximum averages of 16–22 ˚C) and cool night temperatures (monthly minimum averages of 3– 9 ˚C). Total annual precipitation averages around 433 mm, of which over 60% occurs during growing season from May to September (Government of Canada, 2020). The monthly 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4066 KHANAL ET AL. T A B L E 1 B as el in e so il p h y si co ch em ic al p ro p er ti es an d fe rt il it y st at u s o f ex p er im en ta l si te s at d if fe re n t d ep th s in 2 0 1 3 D ep th Ba sic pr op er tie s Ex tr ac ta bl e nu tr ie nt s pH BD O M Sa nd Si lt C la y N H 4+ -N N O 3– - N PO 43– -P SO 42– -S cm g cm − 3 % m g k g − 1 0 – 1 5 5 .4 3 1 .3 5 .5 5 3 1 .3 4 6 .0 2 2 .7 5 .9 6 9 .3 1 3 5 .8 6 1 9 .8 1 5 – 3 0 6 .1 3 1 .5 4 .3 4 2 8 .1 3 6 .1 3 5 .8 3 .1 2 3 .9 8 9 .1 7 1 7 .4 3 0 – 6 0 6 .4 3 1 .6 3 .6 1 2 3 .3 3 0 .7 4 6 .0 2 .4 2 2 .0 1 3 .6 5 3 6 .6 No te . B D , b u lk d en si ty (g cm − 3 ); O M , o rg an ic m at te r. F I G U R E 1 Monthly averages of growing season precipitation (mm), and temperatures (minimum, maximum, and average in ˚C) in the experimental years compared with the 30 yr averages (1986–2016) at Beaverlodge Research Farm, Alberta averages of growing season temperatures and precipitation in the experimental years and 30 yr of averages are presented in Figure 1 and show the growing season weather varies between years. For example, the weather pattern of 2013 was similar to the 30-yr average with slightly higher well-distributed precip- itation, while 2014 and 2015 were relatively drier years with erratic distribution of precipitation over the growing season. Contrarily, the 2016 growing season was exceptionally wet, especially in June (reproductive period) and August (maturity period), affecting crop harvest. 2.2 Experimental design The experiment was designed as a split plot with four repli- cates in which eight different cropping sequences (Sn) consti- tuted the main plots factor (Table 2), and three levels of nitro- gen (N) application constituted subplot factors. Test crops in the cropping sequences included three perennial forage seed crops (CF, RC, and AC), and four annual field crops (Can, Pea, Bar, and Whe). The cropping sequence treatments were spilt into N input of 0, 45, or 90 kg N ha−1 applied at nonlegume phases of the rotations. 2.3 Crop management The experiment was established through zero-till air-seeding (Cross Slot with Varmax, Cross Slot IP Limited) on 30-cm row spacing. In the establishment year of 2013, a uniform rates of N and phosphorus (P) at 90 and 60 kg ha−1, respec- tively, was applied in all plots. In subsequent years, the fertil- ized split-plot treatments of annual crops at nonlegume phase received in-row granular urea during seeding, while the peren- nial creeping red fescue received a fall broadcast of granu- lar urea. Various preseed and postemergence herbicides were applied for weed control (Table 3). After the production years, the perennial forage seed crops of clovers and creeping red 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4067 T A B L E 2 Cropping sequence treatments included in the study Crops and yr Cropping sequence Cropping sequence abbreviation 2013 2014 2015 2016 Can–Can–Can–Can S1 Canola Canola Canola Canola CF–CF–CF–Can S2 Fescue Fescue Fescue Canola InRC–RC–Whe–Can S3 Red clover Red clover Wheat Canola InAC–AC–Whe–Can S4 Alsike clover Alsike clover Wheat Canola RC–RC–Whe–Can S5 Red clover Red clover Wheat Canola AC–AC–Whe–Can S6 Alsike clover Alsike clover Wheat Canola Pea–Bar–Whe–Can S7 Pea Barley Wheat Canola Whe–Can–Whe–Can S8 Wheat Canola Wheat Canola Note. Can, canola; CF, creeping red fescue; InRC, inoculated red clover (inoculated with Rhizobium biofertilizer prior to seeding); RC, red clover; Whe, wheat; InAC, inoculated alsike clover (inoculated with Rhizobium biofertilizer prior to seeding); AC, alsike clover; Bar, barley. T A B L E 3 Herbicide applications made in experimental plots Year Herbicides applied Crop 2013 Roundup (glyphosate) (Bayer) Wheat, pea, canola Basagran (bentazon) (BASF) Creeping red fescue, red clover, alsike clover, pea 2014 Infinity (pyrasulfotole and bromoxynil) (Bayer) Creeping red fescue, barley Roundup (glyphosate), Assure II (quizalofop p-ethyl) (AMVAC Chemical Corporation) Canola 2015 Infinity (pyrasulfotole and bromoxynil) Creeping red fescue, wheat Roundup (glyphosate), Assure II (quizalofop p-ethyl) Canola 2016 Roundup (glyphosate) Canola fescue were terminated by mowing and application of glyphosate on the active regrowth in autumn and early spring seasons. Dates of major agronomic operations are presented in Table 4. 2.4 Data acquisition and analyses 2.4.1 Economic analyses Crop biomass and seed yield were recorded for each crop every year. Input costs that varied across the treatments were calculated based on the prices in purchase receipts from local suppliers. Operation costs, which are the cost of applying the inputs that varied across the treatments, were derived from custom rate surveys of Alberta (Alberta Agriculture and Forestry, 2016). The inputs and their application costs that varied across the treatments are referred to as partial variable costs (Table 5; Supplemental Table S1). For supplementary information, crop-specific fixed costs and common variable costs of production were adapted from the crop production plans of Saskatchewan (Saskatchewan Agriculture, 2016) and Manitoba (Manitoba Agriculture and Resource Development, 2016) (Supplemental Table S2). The fixed costs including building repair, property taxes, business overhead, machinery depreciation, building depreciation, machinery investment, building investment, and land investment were the same for all types of crops. Common variable costs comprised labor and management, storage costs, crop insurance premiums, chem- icals (herbicide, fungicide), and nonnitrogenous fertilizers (P, potassium [K], and sulfur [S]). Except for storage costs that differed between various crops, all other common variable costs were the same for all listed crops. The storage costs of annual crops were up to CAN$11.95 ha−1 higher than that of forage seed crops (Supplemental Table S2), while the latter would incur seed cleaning costs, which were not identified in the production plans. In this way, the differences in posthar- vest costs would be minimal between the crops used in the study. Therefore, postharvest costs were not included in the gross margin analysis of the cropping sequences. Commod- ity prices were obtained from Agriculture Financial Services Corporation, which were originally collected from Alberta Agriculture and Forestry, and other industry sources (Agricul- ture Financial Services Corporation, 2019). Economic returns were calculated under low and average price scenarios of the annual food crops and perennial seed crops for comparison between the cropping sequences (Table 6). The yearly min- imum and average prices were derived using the monthly 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4068 KHANAL ET AL. T A B L E 4 Seeding, harvest, and herbicide application dates for different crops grown in 4-yr cropping sequences from 2013 to 2016 Crop sequences Crop Seeding date Herbicide application Harvest date 2013 Can–Can–Can–Can Canola 31 May 30 May 23 Sept. CF–CF–CF–Can Fescue 31 May 4 June – InRC–RC–Whe–Can Red clover 31 May 4 June – InAC–AC–Whe–Can Alsike clover 31 May 4 June – RC–RC–Whe–Can Red clover 31 May 4 June – AC–AC–Whe–Can Alsike clover 31 May 4 June – Pea–Bar–Whe–Can Pea 31 May 30 May 24 Sept. Whe–Can–Whe–Can Wheat 31 May 30 May 24 Sept. 2014 Can–Can–Can–Can Canola 16 May 30 May 2 Sept. CF–CF–CF–Can Fescue – 30 May 23 July InRC–RC–Whe–Can Red clover – 30 May 2 Sept. InAC–AC–Whe–Can Alsike clover – 30 May 2 Sept. RC–RC–Whe–Can Red clover – 30 May 2 Sept. AC–AC–Whe–Can Alsike clover – 30 May 2 Sept. Pea–Bar–Whe–Can Barley 16 May 30 May 2 Sept. Whe–Can–Whe–Can Canola 16 May 30 May 2 Sept. 2015 Can–Can–Can–Can Canola 21 May 30 May 10 Sept. CF–CF–CF–Can Fescue 15 May 23 May 22 July InRC–RC–Whe–Can Wheat 21 May 23 June 11 Sept. InAC–AC–Whe–Can Wheat 21 May 23 June 11 Sept. RC–RC–Whe–Can Wheat 21 May 23 June 11 Sept. AC–AC–Whe–Can Wheat 21 May 23 June 11 Sept. Pea–Bar–Whe–Can Wheat 21 May 23 June 11 Sept. Whe–Can–Whe–Can Wheat 21 May 23 June 11 Sept. 2016 Can–Can–Can–Can Canola 26 May 13 May, 23 June 10 Sept. CF–CF–CF–Can Canola 26 May 13 May, 23 June 10 Sept. InRC–RC–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. InAC–AC–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. RC–RC–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. AC–AC–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. Pea–Bar–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. Whe–Can–Whe–Can Canola 26 May 13 May, 23 June 10 Sept. Note. Can, canola; CF, creeping red fescue; InRC, inoculated red clover; RC, red clover; Whe, wheat; InAC, inoculated alsike clover; AC, alsike clover; Bar, barley. selling prices of food grain and forage seed in the particular year in which the crop was produced in the cropping sequence. Various agronomic and economic performance variables used in the comparison are listed and defined in Table 7. In the 4-yr cropping sequences, wheat was a common phase crop grown in six sequences of the third year, and canola was a common phase crop grown in all eight cropping sequences of the fourth year. Agronomic N use efficiency (NUE; change in seed yield per kg of N application) and partial factor pro- ductivity (seed yield per kg of N application) was assessed for individual crops of wheat and canola. 2.5 Statistical analyses Data analysis was performed with PROC GLIMMIX in SAS (Version 9.4) (SAS Institute, Inc.). The factors, rotation, and N were fixed, whereas blocks and their interactions with 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4069 T A B L E 5 P ar ti al v ar ia b le co st u se d fo r th e cu m u la ti v e g ro ss m ar g in an al y si s o f 4 -y r cr o p p in g se q u en ce s co m p ri si n g v ar io u s cr o p s fr o m 2 0 1 3 to 2 0 1 6 U re a co st Fe rt ili zi ng co st To ta lp ar tia lv ar ia bl e co st C ro pp in g se qu en ce Se ed co st Se ed in g co st 0 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 0 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 H er bi ci de co st H ar ve st co st 0 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 C A N $ h a− 1 C an – C an – C an – C an 2 8 2 2 2 0 9 7 4 5 3 8 0 8 1 8 7 3 7 3 0 1 6 0 7 7 7 1 ,1 8 8 1 ,5 4 3 C F – C F – C F – C an 6 9 9 9 9 7 4 5 3 8 0 8 1 8 7 3 7 3 5 0 1 2 0 4 5 3 8 6 4 1 ,2 1 9 In R C – R C – W h e– C an 1 3 4 1 5 4 9 7 3 4 7 5 9 7 1 8 5 5 5 5 5 0 1 2 0 5 7 3 8 6 0 1 ,1 1 0 In A C – A C – W h e– C an 1 5 7 1 5 4 9 7 3 4 7 5 9 7 1 8 5 5 5 5 5 0 1 2 0 5 9 6 8 8 3 1 ,1 3 3 R C – R C – W h e– C an 1 3 4 1 5 4 9 7 3 4 7 5 9 7 1 8 5 5 5 5 5 0 1 2 0 5 7 3 8 6 0 1 ,1 1 0 A C – A C – W h e– C an 1 5 7 1 5 4 9 7 3 4 7 5 9 7 1 8 5 5 5 5 5 0 1 2 0 5 9 6 8 8 3 1 ,1 3 3 P ea – B ar – W h e– C an 2 7 0 2 2 0 9 7 4 5 3 8 0 8 1 8 7 3 7 3 0 1 6 0 7 6 5 1 ,1 7 6 1 ,5 3 1 W h e– C an – W h e– C an 2 4 0 2 2 0 9 7 4 5 3 8 0 8 1 8 7 3 7 3 0 1 6 0 7 3 5 1 ,1 4 6 1 ,5 0 1 No te . C an , ca n o la ; C F , cr ee p in g re d fe sc u e; In R C , in o cu la te d re d cl o v er ; R C , re d cl o v er ; W h e, w h ea t; In A C , in o cu la te d al si k e cl o v er ; A C , al si k e cl o v er ; B ar , b ar le y. T h e p ar ti al v ar ia b le co st s in cl u d ed th e co st o f in p u ts su ch as se ed , fe rt il iz er an d h er b ic id es ,a n d th ei r ap p li ca ti o n co st s su ch as se ed in g ,f er ti li zi n g ,h er b ic id e ap p li ca ti o n an d h ar v es t co st s th at v ar ie d b et w ee n th e cr o p p in g se q u en ce s u n d er th re e N le v el s. V ar ia b le h er b ic id e in p u t an d ap p li ca ti o n co st s ar e co m b in ed in th e h er b ic id e co st ca te g o ry in th e ta b le . V ar ia b le in p u t co st s ar e ca lc u la te d b as ed o n p u rc h as e re ce ip ts fr o m lo ca l su p p li er s. V ar ia b le in p u t ap p li ca ti o n co st s w er e ad ap te d fr o m cu st o m ra te s su rv ey s o f th e P ea ce re g io n , co n d u ct ed b y A lb er ta A g ri cu lt u re an d F o re st ry (2 0 1 6 ). C ro p -s p ec if ic d et ai ls o f p ar ti al v ar ia b le co st s, fi x ed co st s an d co m m o n v ar ia b le co st s, an d cu m u la ti v e co m m o n p ro d u ct io n co st s o f cr o p p in g se q u en ce s ar e g iv en in th e S u p p le m en ta l T ab le s S 1 , S 2 , an d S 3 , re sp ec ti v el y. cropping sequences and N were random. The analysis for gross margins and canola equivalent yield (CEY) proceeded in two steps. First, maximum likelihood estimation was used to determine inclusion of two variance structures, one as a split plot and the other as a split plot with an additional term to account for possible heterogeneity in the variances among cropping sequences. The identity link was used because data were assumed to follow the normal distribution. Akaike infor- mation criterion was used to determine adequacy of the model. Also, normality of the residuals and adequacy of the model were reviewed through examination of the residual plots. Following the choice of the appropriate model, the sec- ond step in the analysis involved the ANOVA by using a restricted maximum likelihood estimation method. Kenward– Roger correction for denominator df was used to adjust for apparent heterogeneity in variances. Means separation was performed using Tukey’s procedure. 3 RESULTS 3.1 Seed yield of wheat and canola as third and fourth phase crops At all N levels, effect of Sn on wheat grain yield was sta- tistically significant (p = .002). Three cropping sequences including red clover (RC–RC–Whe–Can), alsike clover (AC–AC–Whe–Can) and pea (Pea–Bar–Whe–Can) produced significantly higher wheat yields than Whe–Can alternat- ing sequences (Table 8). The effect of N input levels (p = .4239) and their interaction with the crop sequences (Sn×N) (p = .1312) were not significant. In the absence of a supplemental N application, wheat crop preceded by bien- nial stand of red clover produced yield increases of up to 45 and 21% compared with wheat crop preceded by Whe–Can and pea–Bar sequences, respectively (Figure 2). Similarly, wheat crop preceded by biennial stand of alsike clover pro- duced yield increases of up to 49 and 24% compared with wheat crop preceded by Whe–Can and Pea–Bar sequences, respectively. The carryover effect of preceding forage legume seed crops was further pronounced in the canola crop following wheat in the fourth phase of the cropping sequences. The effects of Sn (p < .0001), N (p < .0001), and Sn×N (p = .042) were significant. Canola yields were significantly higher in alsike clover-based cropping sequences compared with all annual crop-based sequences and creeping red fescue-based sequences (Table 8). Closely following, canola yield in the red clover-based cropping sequence was statistically on par with the alsike clover-based sequence as well as the pea-based and Whe–Can alternating sequences. The continuous canola and creeping red fescue-based sequences produced the lowest canola yields. Canola yields increased significantly as the N 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4070 KHANAL ET AL. T A B L E 6 Average and low price values of output (food grains of annual crops and seeds of perennial forage crops) used for economic analyses of different crop sequences from 2013 to 2016 Crop Year Average prices of outputs Low prices of output CAN$ kg−1 Canola 2013 0.53 0.35 2014 0.39 0.32 2015 0.43 0.37 2016 0.43 0.36 Pea 2013 0.29 0.22 Wheat 2013 0.22 0.16 2015 0.19 0.18 Barley 2014 0.20 0.16 Creeping red fescue 2014 1.63 1.58 2015 1.69 1.65 Red clover 2014 1.87 1.80 Alsike clover 2014 2.75 2.57 Note. Commodity prices were obtained from Agriculture Financial Services Corporation (2019) that were originally collected from Alberta Agriculture and Forestry and other industry sources. T A B L E 7 Agronomic and economic performance variables and their definitions within the study context Variables Description Canola equivalent yield (CEY) Noncanola yield × price ratio of noncanola/canola (Liu et al., 2019) Partial variable cost Input costs and operational costs that vary between the crop species included in the cropping sequences Total partial variable cost Cumulative variable costs of a 4-yr cropping sequence Gross return Cumulative product of crop seed yield and corresponding prices of the commodities over 4-yr cropping sequences Gross margin Gross return − total partial variable cost (cost of inputs and application costs that differed between the treatments) Partial factor productivity for N Seed yield/N used (kg kg−1) Biomass nitrogen use efficiency (NUE) (biomass yield at Nx − biomass yield at N0)/Nx (kg kg−1), where Nx is nitrogen applied (45 and 90 kg ha−1) and N0 refers to the absence of nitrogen Agronomic NUE (seed yield at Nx − seed yield at N0)/Nx (kg kg−1) Agronomic NUE for CEY for 4-yr cropping sequences (CEY at Nx − CEY at N0)/Nx (kg kg−1) Economic N use efficiency (gross revenue at Nx − gross revenue at N0)/N cost at Nx ($ $−1) level increased from 0 to 45 kg ha−1 and from 45 to 90 kg ha−1. In the absence of N fertilizer, canola yield was signifi- cantly less in the creeping red fescue-based cropping sequence compared with alsike and red clover-based sequences (Figure 3). The continuous canola and Whe–Can sequences were on a par with the creeping red fescue sequence, while the diverse annual crop-based sequence (Pea–Bar–Whe–Can) was statistically on a par with alsike and red clover-based cropping sequences for canola yield. At a N rate of 45 kg ha−1, the creeping red fescue-based cropping sequence had significantly less canola yield than all the rest of the cropping sequences. At a N rate of 90 kg ha−1, the creeping red fescue- based and continuous canola sequences had significantly less canola yield than that of the alsike clover-based cropping sequence. The rest of the cropping sequences were on a par with both low and high yielding cropping sequences. With- out N fertilizer application, canola yield from forage legumes- based sequences were 40–70% higher than those from annual crop sequences (Figure 3). The carryover effect diminished in magnitude as the application rate of N increased. Based on the yield comparisons, biennial legume seed crops of red and alsike clovers replaced the N fertilizer requirement for succeeding wheat and canola crops by at least 90 and 45 kg ha−1, respectively. 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4071 T A B L E 8 Main effect of different cropping sequences and N rates on wheat as a third phase crop, canola as a fourth phase crop, and total biomass yield of 4-yr cropping sequences from 2013 to 2016 Wheat yield Canola yield Total biomass yield kg ha−1 Cropping sequences Can–Can–Can–Can – 2,190 c 24,695 bc CF–CF–CF–Can – 1,459 d 16,665 e InRC–RC–Whe–Can 4,313 a 2,650 a–c 21,826 c InAC–AC–Whe–Can 3,256 b 2,464 bc 17,153 de RC–RC–Whe–Can 4,341 a 2,905 ab 21,037 cd AC–AC–Whe–Can 4,199 a 3,197 a 22,773 c Pea–Bar–Whe–Can 3,918 a 2,530 bc 28,410 ab Whe–Can–Whe–Can 3,213 b 2,448 bc 29,859 a p-values .0002 <.0001 <.0001 SEM 208.9 151.9 806.4 SED 295.4 214.8 1140 N rates 0 kg ha−1 3,768 1,607 c 20,087 c 45 kg ha−1 3,911 2,435 b 23,177 b 90 kg ha−1 3,940 3,400 a 25,143 a p-values .4239 <.0001 <.0001 SEM 130.3 84.7 352.9 SED 184.3 119.8 499.0 Note. Can; CF, creeping red fescue; InRC, inoculated red clover; RC, red clover; Whe; InAC, inoculated alsike clover; AC, alsike clover; Bar, barley; SED, standard error of the difference in means. Means separation was performed using Tukey’s procedure. The values followed by the same letter, overlapping lettering range (such as a–c), or without any letters are not significantly different (p > .05). 3.2 Biomass productivity of 4-yr crop sequences Cumulative crop biomass yields of 4-yr cropping sequences were generally higher with higher levels of N application. Within a given level of N application, annual crop sequences had significantly higher aerial biomass yields than that of the sequences containing perennial crops (Table 8). Thus, the effects of Sn (p < .0001), N (p < .0001), and Sn×N (p < .00001) were highly significant (Figure 4). For total aerial biomass yield, annual crop-based sequences were sig- nificantly higher than all forage seed crop-based sequences. Overall, there was significant increase in biomass yield with the increase in N application. 3.3 Cumulative canola equivalent yield of 4-yr cropping sequences The CEY (noncanola yield multiplied by price ratio of non- canola to canola) was calculated under average and minimum price scenarios. The crop sequences had a significant effect on CEY under both minimum (p < .0001) and average price sce- narios (p < .0001) (Table 9). At the minimum price scenario, creeping red fescue-based cropping sequences had signifi- cantly higher CEY than continuous canola, Whe–Can alterna- tion, and clover-based sequences, while being statistically on par with pea-based annual cropping sequences. At the average price scenario, the CEYs of all annual crop-based sequences were statistically similar to creeping the red fescue-based sequence, while all clover-based sequences had significantly lower CEY. Under both minimum and average price scenar- ios, CEY increased significantly (p < .0001) with the increase in N levels from 0 through 45 to 90 kg ha−1. The interactions between the cropping sequences and N rates were significant at the minimum price level (p < .0001) (Figure 5a), but not significant at average price levels (p = .2530) (Figure 5b). In general, across three N levels, the cropping sequence with creeping red fescue had higher CEY followed by annual crop- based sequences, while the clover-based sequences had lower CEY. 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4072 KHANAL ET AL. T A B L E 9 T h e 4 -y r cu m u la ti v e ca n o la eq u iv al en t y ie ld (C E Y ), g ro ss re v en u e, an d g ro ss m ar g in in fl u en ce d b y ei g h t d if fe re n t cr o p p in g se q u en ce s an d th re e N le v el s w it h m in im u m an d av er ag e co m m o d it y p ri ce s o f se ed s o f fo ra g es an d g ra in s o f an n u al cr o p s fr o m 2 0 1 3 to 2 0 1 6 C EY G ro ss re ve nu e G ro ss m ar gi n M in .c om m od ity pr ic e Av g. co m m od ity pr ic e M in .c om m od ity pr ic e Av g. co m m od ity pr ic e M in .c om m od ity pr ic e Av g. co m m od ity pr ic e k g h a− 1 C A N $ h a− 1 C ro p p in g se q u en ce s C an – C an – C an – C an 7 ,9 8 3 b c 7 ,8 1 6 ab 2 ,7 5 8 b 3 ,4 1 1 ab 1 ,5 9 1 c 2 ,2 5 5 b C F – C F – C F – C an 1 0 ,6 5 6 a 9 ,4 9 6 a 3 ,5 7 3 a 3 ,7 6 6 a 2 ,6 5 6 a 2 ,7 6 3 a In R C – R C – W h e– C an 5 ,7 7 6 cd 5 ,3 4 3 c 2 ,0 2 6 c 2 ,2 6 4 c 1 ,1 3 6 c 1 ,3 8 6 c In A C – A C – W h e– C an 6 ,0 1 4 cd 5 ,6 2 8 b c 2 ,1 2 6 c 2 ,3 6 6 c 1 ,1 7 2 c 1 ,4 3 5 c R C – R C – W h e– C an 5 ,4 9 2 d 5 ,1 5 7 c 1 ,9 4 8 c 2 ,2 2 0 c 1 ,1 1 1 c 1 ,3 7 8 c A C – A C – W h e– C an 7 ,8 8 5 b -d 7 ,4 6 9 a– c 2 ,8 4 1 b 3 ,1 7 9 b 1 ,9 9 6 b 2 ,3 6 4 ab P ea – B ar – W h e– C an 9 ,0 9 7 ab 8 ,6 5 5 a 3 ,1 7 5 ab 3 ,8 3 6 a 1 ,9 2 5 b 2 ,5 6 8 ab W h e– C an – W h e– C an 8 ,1 8 9 b c 8 ,1 7 4 a 3 ,0 2 8 b 3 ,6 6 2 ab 1 ,9 0 4 b 2 ,5 6 8 ab p- v al u es < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 S E M 5 6 4 .3 4 4 7 .0 1 1 0 .8 1 2 8 9 2 .4 1 0 7 S E D 7 9 8 .1 6 3 2 .2 1 5 6 .8 1 8 1 1 3 0 .6 1 5 0 .9 N ra te s 0 k g h a− 1 6 ,3 7 0 c 6 ,0 1 9 c 2 ,2 6 4 c 2 ,6 0 0 c 1 ,6 0 5 b 1 ,9 3 8 b 4 5 k g h a− 1 7 ,5 4 9 b 7 ,1 7 4 b 2 ,6 6 3 b 3 ,0 6 8 b 1 ,6 5 2 b 2 ,0 5 4 b 9 0 k g h a− 1 8 ,9 9 0 a 8 ,4 5 8 a 3 ,1 2 6 a 3 ,5 9 5 a 1 ,8 0 2 a 2 ,2 7 7 a p- v al u es < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 < .0 0 0 1 .0 0 2 8 < .0 0 0 1 S E M 3 5 1 .2 1 9 3 .1 5 5 .8 6 4 .5 4 7 .2 5 4 .7 S E D 4 9 6 .7 2 7 3 .1 7 8 .9 9 1 .1 6 6 .7 7 7 .4 No te .C an ,c an o la ; C F ,c re ep in g re d fe sc u e; In R C ,i n o cu la te d re d cl o v er ; R C ,r ed cl o v er ; W h e, w h ea t; In A C ,i n o cu la te d al si k e cl o v er ; A C ,a ls ik e cl o v er ; B ar ,b ar le y ; S E D ,s ta n d ar d er ro r o f th e d if fe re n ce in m ea n s. M ea n s se p ar at io n w as p er fo rm ed u si n g T u k ey ’s p ro ce d u re . T h e v al u es fo ll o w ed b y th e sa m e le tt er o r o v er la p p in g ra n g e (s u ch as a– c an d b – d ) ar e n o t si g n if ic an tl y d if fe re n t (p > .0 5 ). 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4073 F I G U R E 2 Grain yield of wheat as third phase crop in four-year cropping sequences under three N rates in 2015. The eight cropping sequences included in the study were S1 = Can–Can–Can–Can, S2 = CF–CF–CF–Can, S3 = InRC–RC–Whe–Can, S4 = InAC–AC–Whe–Can, S5 = RC–RC–Whe–Can, S6 = AC–AC–Whe–Can, S7 = Pea–Bar–Whe–Can, and S8 = Whe–Can–Whe–Can, and were grown from 2013 to 2016. In 2015, wheat was not grown in S1 and S2 sequences. The three N rates were 0, 45, and 90 kg ha−1. Vertical bars indicate SEM (n = 4). Means separation was performed using Tukey’s procedure. Columns in cropping sequences without any letters are not significantly different (p > .05). AC, alsike clover; Bar, barley; Can, canola; CF, creeping red fescue; InAC, inoculated alsike clover; InRC, inoculated red clover; RC, red clover; Sn, cropping sequences; SED, standard error of the difference between means; Sn×N, interaction between cropping sequences and N; Whe, wheat F I G U R E 3 Grain yield of canola as fourth phase crop in 4-yr cropping sequences under three N rates in 2016. The eight cropping sequences included in the study were S1 = Can–Can–Can–Can, S2 = CF–CF–CF–Can, S3 = InRC–RC–Whe–Can, S4 = InAC–AC–Whe–Can, S5 = RC–RC–Whe–Can, S6 = AC–AC–Whe–Can, S7 = Pea–Bar–Whe–Can, and S8 = Whe–Can–Whe–Can, and were grown from 2013 to 2016. The three N rates were 0, 45, and 90 kg ha−1. Vertical bars indicate standard error of means (n = 4). Cropping sequences with the same letter or overlapping lettering range (such as a-c and a-d) above the error bar within each N rate do not differ significantly within the N input levels (p > .05). AC, alsike clover; Bar, barley; Can, canola; CF, creeping red fescue; InAC, inoculated alsike clover; InRC, inoculated red clover; RC, red clover; Sn, cropping sequences; SED, standard error of the difference between means; Sn×N, interaction between cropping sequences and N; Whe, wheat F I G U R E 4 The 4-yr cumulative biomass yield of eight different cropping sequences at three levels of N application from 2013 to 2016. The 4-year cropping sequences included in the study were S1 = Can–Can–Can–Can, S2 = CF–CF–CF–Can, S3 = InRC–RC–Whe–Can, S4 = InAC–AC–Whe–Can, S5 = RC–RC–Whe–Can, S6 = AC–AC–Whe–Can, S7 = Pea–Bar–Whe–Can, and S8 = Whe–Can–Whe–Can. The three N rates were 0, 45, and 90 kg ha−1. Vertical bars indicate standard error (n = 4). Means separation was performed using Tukey’s procedure. Columns in cropping sequences within each N rate followed by the same letter are not significantly different (p > .05). AC, alsike clover; Bar, barley; Can, canola; CF, creeping red fescue; InAC, inoculated alsike clover; InRC, inoculated red clover; RC, red clover; Sn, cropping sequences; SED, standard error of the difference in means; Sn×N, interaction between cropping sequences and N; Whe, wheat 3.4 Cumulative gross revenue and gross margins over partial variable costs Under the minimum price scenario, significant effects were exhibited by cropping sequences (p < .0001), N levels (p < .0001), and their interactions (p = .0055) for the cumulative gross revenue. The creeping red fescue-based sequence stayed on par with alsike clover-based sequence and stood significantly higher than the rest of the sequences for the gross revenue (Table 10). Overall, the cumulative gross revenue increased significantly with the increase in N levels. Significant effects of cropping sequences (p < .0001) and N levels (p = .0028) were also observed on cumulative gross margins. The cumulative gross margin of the creeping red fescue-based sequence stood out significantly higher than all crop sequences, while some of the clover-based sequences and the continuous canola sequence remained the lowest. The interactions of cropping sequences and N levels were nonsignificant (p = .073) for the gross margins. Under the average price scenario of annual crops, the cumu- lative gross revenue of different 4-yr cropping sequences dif- fered significantly (p < .0001), with significant (p < .0001) increase in gross revenue with the increase in N levels. The interactions between the cropping sequences and N levels were also significant (p = .0061) for the gross revenue. The cumulative gross revenue of the annual crop-based sequences and creeping red fescue-based sequence did not differ 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4074 KHANAL ET AL. T A B L E 1 0 Cumulative gross revenue and gross margin of various 4-yr cropping sequences under different N application rates from 2013 to 2016 Minimum commodity price Average commodity price Crop sequence Gross revenue Gross margin Gross revenue Gross margin CAN$ ha−1 0 kg ha−1 Can–Can–Can–Can 2,323 g–l 1,551 2,889 d–h 2,127 CF–CF–CF–Can 2,858 c–h 2,377 3,000 d–g 2,459 InRC–RC–Whe–Can 1,660 l 1,047 1,826 i 1,226 InAC–AC–Whe–Can 1,911 j–l 1,215 2,145 g–i 1,425 RC–RC–Whe–Can 1,751 kl 1,191 1,978 hi 1,413 AC–AC–Whe–Can 2,549 e–k 1,982 2,824 e–h 2,286 Pea–Bar–Whe–Can 2,687 d–j 1,831 3,252 c–f 2,380 Whe–Can–Whe–Can 2,375 f–l 1,645 2,885 e–h 2,186 45 kg ha−1 Can–Can–Can–Can 2,677 d–j 1,490 3,317 c–f 2,141 CF–CF–CF–Can 3,411 b–d 2,521 3,576 b–e 2,622 InRC–RC–Whe–Can 2,126 h–l 1,224 2,382 f–i 1,491 InAC–AC–Whe–Can 2,116 h–l 1,130 2,400 f–i 1,389 RC–RC–Whe–Can 1,960 i–l 1,110 2,234 g–i 1,378 AC–AC–Whe–Can 2,758 c–i 1,901 3,083 c–g 2,254 Pea–Bar–Whe–Can 3112 b–g 1842 3,760 a–e 2,474 Whe–Can–Whe–Can 3,145 b–f 2,001 3,794 a–d 2,680 90 kg ha−1 Can–Can–Can–Can 3,275 b–e 1,732 4,027 a–c 2,495 CF–CF–CF–Can 4,451 a 3,071 4,722 a 3,207 InRC–RC–Whe–Can 2,291 h–l 1,138 2,582 f–i 1,441 InAC–AC–Whe–Can 2,352 f–l 1,172 2,553 f–i 1,490 RC–RC–Whe–Can 2,132 h–l 1,032 2,449 f–i 1,343 AC–AC–Whe–Can 3,214 b–e 2,106 3,629 b–e 2,551 Pea–Bar–Whe–Can 3,727 ab 2,101 4,494 ab 2,851 Whe–Can–Whe–Can 3,565 bc 2,065 4,308 ab 2,838 p-values .006 .073 .006 .138 SEM 152.8 129.8 177.1 150.8 SED 216.1 183.5 250.5 213.3 Note. Can, canola; CF, creeping red fescue; InRC, inoculated red clover; RC, red clover; Whe, wheat; InAC, inoculated alsike clover; AC, alsike clover; Bar, barley; SED, standard error of the difference in means. Means separation was performed using Tukey’s procedure. The values followed by the same letters or overlapping range (such as c–h, g–l, etc.) or without any lettering are not significantly different (p > .05). significantly (Table 10). Similarly, for the cumulative gross margins, significant effects of crop sequences (p < .0001) and N levels (p < 0.0001) were observed, while the interac- tion between the two factors was nonsignificant (p = .1378). The cumulative gross margin of the creeping red fescue- based sequence was significantly higher than that of contin- uous canola and clover-based sequences, while other annual crop-based sequences did not differ significantly from that of creeping red fescue. The cumulative gross margin of crop- ping sequences at 90 kg N ha−1 was significantly higher than that of lower N levels, while the cumulative gross margin val- ues of cropping sequences did not differ significantly between 0 and 45 kg N ha−1. The forage legumes-based sequences showed stagnant gross margins with the increase in N appli- cation (Table 10). The annual crop-based sequences showed an increase in gross margin with the increase in N application. 3.5 Nitrogen use efficiency under different cropping sequences Agronomic NUE and partial factor productivities were cal- culated for wheat and canola as common phase crops for the third and fourth years of the cropping sequence cycle. Wheat 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4075 F I G U R E 5 The 4-yr cumulative canola equivalent yields (CEY) of eight cropping sequences at three levels of N application from 2013 to 2016. The upper graph (a) and lower graph (b) present the CEY with minimum and average prices of seeds of forages and grains of annual crops. The 4-yr cropping sequences included in the study were S1 = Can–Can–Can–Can, S2 = CF–CF–CF–Can, S3 = InRC–RC–Whe–Can, S4 = InAC–AC–Whe–Can, S5 = RC–RC–Whe–Can, S6 = AC–AC–Whe–Can, S7 = Pea–Bar–Whe–Can, and S8 = Whe–Can–Whe–Can. The N rates were applied at 0, 45, and 90 Kg ha−1. Vertical bars indicate standard error (n = 4). Means separation was performed using Tukey’s procedure. Columns in cropping sequence under each N-rate followed by the same letter are not significantly different (p > .05). AC, alsike clover; Bar, barley; Can, canola; CF, creeping red fescue; InAC, inoculated alsike clover; InRC, inoculated red clover; RC, red clover; Sn, cropping sequences; SED, standard error of the differences in means; Sn×N, interaction between cropping sequences and N; Whe, wheat crop was absent from creeping red fescue-based and contin- uous canola sequences. At 45 kg N ha−1, wheat crop under Whe–Can–Whe–Can sequence showed significantly higher NUE compared with clover-based sequences (19.32 vs. less than −2.91, p = .034) (Table 11). The diversified annual crop- based sequence Pea–Bar–Whe–Can (NUE = 6.13) and inoc- ulated (In) red clover-based sequence InRC–RC–Whe–Can (NUE = 7.07) appeared on a par with both Whe–Can–Whe– Can and the rest of the clover-based sequences. Though sta- tistically nonsignificant, the increase in N from 45 to 90 kg ha−1 exhibited differential trends for NUE values between the treatments. Clover-based sequences generally displayed neg- ative values for the NUE of wheat. Canola crop after wheat in the sequences showed positive agronomic NUE at both 45 and 90 kg N ha−1 under all cropping sequences. At 45 kg N ha−1, canola under Whe–Can–Whe–Can and InRC–RC–Whe–Can sequence showed significantly higher NUE compared with creeping red fescue-based sequence (31.06 and 25.35 vs. 1.84, p = .034). The rest of the treatments stood on par with crop- ping sequences, having both higher and lower NUE values. The increase in N from 45 to 90 kg ha−1 resulted in differ- ential trends, but convergence in NUE values with no signif- icant differences between the treatments. Annual crop-based sequences have generally higher agronomic NUEs than that of the forage-based sequences. For both wheat and canola, partial factor productivities of N application were higher at 45 than at 90 kg ha−1 for all corre- sponding cropping sequences. Wheat had higher partial factor productivity than canola at 45 kg ha−1, but those differences narrowed at 90 kg ha−1. The cropping sequence effects were nonsignificant for partial factor productivity of wheat. The cropping sequences had significant effects on partial factor productivity of canola at both 45 and 90 kg N ha−1 (p < .001) in that the creeping red fescue-based sequence had the lowest partial factor productivity value followed by the continuous canola sequence. Various NUE indices showed differences between the cropping sequences. The cropping sequences had significant effects on biomass NUE. At both N levels, Whe–Can–Whe– Can and CF–CF–CF–Can had significantly higher biomass N use efficiencies than most of the clover-based and continuous canola sequences (>37.6 vs. 18.3, p < .01), with the exception of the sequence InRC–RC–Whe–Can (Table 12). With two exceptions of close values, the increase in N from 45 to 90 kg ha−1 resulted in a decrease in biomass NUE values within the crop sequences. For most of the cropping sequences, gross revenue-based economic NUE improved with the increase in N applica- tion up to 90 kg ha−1. At both 45 and 90 kg N ha−1, and under both minimum and average commodity price scenar- ios, the economic NUE did not differ significantly between the cropping sequences. Creeping red fescue-based cropping sequence had the most consistent values across all condi- tions, having numerically higher values than that of other crop sequences under the average price scenario. For example, at 90 kg N ha−1 and under average commodity prices, the creep- ing red fescue-based cropping sequence showed CAN$5.47 gross return to every $1 spent on N fertilizer compared with $4.52, $3.94, and $3.61 for Whe–Can–Whe–Can, Pea–Bar– Whe–Can, and continuous canola sequences, respectively. Compared with continuous canola and Whe–Can cropping sequences, the creeping red fescue-based cropping sequence had less production cost by $324 and $282 ha−1, respectively. The cost savings in the creeping red fescue-based cropping sequence, compared with continuous canola, is a result of lower seed price and less seeding frequency due to peren- nial crop ($502 vs. $168 ha−1) and no harvest cost in the 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4076 KHANAL ET AL. T A B L E 11 A g ro n o m ic N u se ef fi ci en cy (N U E ) an d p ar ti al fa ct o r p ro d u ct iv it y o f w h ea t an d ca n o la g ro w n as th e th ir d an d fo u rt h p h as e cr o p s u n d er v ar io u s 4 -y r cr o p p in g se q u en ce s fr o m 2 0 1 3 to 2 0 1 6 A gr on om ic N U E Pa rt ia lf ac to r pr od uc tiv ity W he at (2 01 5) C an ol a (2 01 6) W he at (2 01 5) C an ol a (2 01 6) C ro pp in g se qu en ce 45 kg N ha − 1 90 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 45 kg N ha − 1 90 kg N ha − 1 k g k g − 1 C an – C an – C an – C an – – 1 8 .8 8 ab 1 6 .7 – – 5 7 .2 a 3 5 .9 b C F – C F – C F – C an – – 1 .8 4 b 2 2 .7 – – 2 4 .5 b 3 3 .5 b In R C – R C – W h e– C an 7 .0 7 ab 2 .7 3 2 5 .3 5 a 1 8 .2 9 5 .1 4 6 .8 7 1 .7 a 4 1 .4 ab In A C – A C – W h e– C an − 5 .3 8 b − 3 .8 5 1 5 .3 7 ab 1 6 .0 7 2 .8 3 5 .3 6 3 .7 a 4 0 .2 b R C – R C – W h e– C an − 2 .9 1 b − 4 .1 5 1 4 .3 8 ab 1 3 .9 9 1 .6 4 3 .1 7 2 .3 a 4 2 .8 ab A C – A C – W h e– C an − 5 .1 2 b − 0 .7 8 1 5 .5 0 ab 2 1 .0 9 1 .9 4 7 .7 7 7 .0 a 5 1 .7 a P ea – B ar – W h e– C an 6 .1 3 ab 9 .8 9 2 1 .4 7 ab 2 3 .8 8 4 .3 4 9 .0 6 2 .7 a 4 4 .4 ab W h e– C an – W h e– C an 1 9 .3 2 a 7 .6 1 3 1 .0 6 a 2 6 .1 8 4 .1 4 0 .0 6 5 .2 a 4 3 .1 ab p- v al u es .0 3 4 .0 5 9 .0 1 1 .0 9 9 .1 0 1 .0 8 5 < .0 0 0 1 .0 0 1 S E M 5 .6 5 3 .7 4 5 .0 9 3 .3 7 5 .5 9 3 .5 0 6 .3 8 3 .8 0 S E D 7 .9 9 5 .2 8 7 .2 0 4 .7 7 7 .9 1 4 .9 5 9 .0 2 5 .3 7 No te .C an ,c an o la ; C F ,c re ep in g re d fe sc u e; In R C ,i n o cu la te d re d cl o v er ; R C ,r ed cl o v er ; W h e, w h ea t; In A C ,i n o cu la te d al si k e cl o v er ; A C ,a ls ik e cl o v er ; B ar ,b ar le y ; S E D ,s ta n d ar d er ro r o f th e d if fe re n ce in m ea n s. In 2 0 1 5 ,w h ea t w as n o t g ro w n in C an – C an – C an – C an an d C F – C F – C F – C se q u en ce s. M ea n s se p ar at io n w as p er fo rm ed u si n g T u k ey ’s p ro ce d u re . T h e v al u es fo ll o w ed b y sa m e le tt er s, o v er la p p in g ra n g e o f le tt er s o r w it h o u t le tt er in g ar e n o t si g n if ic an tl y d if fe re n t (p > .0 5 ). 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4077 T A B L E 12 B io m as s, ec o n o m ic an d ag ro n o m ic N u se ef fi ci en ci es (N U E s) o f v ar io u s 4 -y r cr o p p in g se q u en ce s fr o m 2 0 1 3 to 2 0 1 6 Bi om as sN U E Ec on om ic N U E A gr on om ic N U E fo r C EY M in im um pr ic e Av er ag e pr ic e M in im um pr ic e Av er ag e pr ic e 45 kg 90 kg 45 kg 90 kg 45 kg 90 kg 45 kg 90 kg 45 kg 90 kg C ro pp in g se qu en ce N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 N ha − 1 k g k g − 1 $ $ − 1 k g k g − 1 C an – C an – C an – C an 1 6 .4 b c 1 7 .6 a– c 2 .6 2 3 .1 7 3 .0 2 3 .6 1 7 .4 2 ab 7 .4 2 8 .4 4 b c 8 .4 4 b c C F – C F – C F – C an 3 7 .6 ab 3 0 .5 a 4 .7 3 5 .0 9 5 .0 6 5 .4 7 7 .0 0 ab 1 2 .2 0 1 8 .4 4 a 1 6 .8 0 a In R C – R C – W h e– C an 4 1 .1 a 1 9 .4 a– c 5 .1 9 6 .1 7 2 .8 1 3 .3 6 1 4 .4 ab 1 4 .2 4 7 .8 0 b c 7 .7 5 b c In A C – A C – W h e– C an 1 5 .5 c 5 .3 c 2 .2 7 2 .8 2 1 .8 3 1 .8 2 6 .3 8 b 6 .4 9 3 .9 3 c 4 .1 5 c R C – R C – W h e– C an 1 8 .3 b c 9 .2 b c 2 .3 2 2 .8 4 1 .6 9 2 .0 9 6 .8 3 b 6 .9 1 4 .7 4 c 4 .8 2 c A C – A C – W h e– C an 1 5 .6 c 1 7 .7 a– c 2 .3 2 2 .8 8 2 .9 6 3 .5 8 6 .5 0 b 6 .6 2 8 .2 4 b c 8 .2 5 b c P ea – B ar – W h e– C an 2 7 .0 ab 1 9 .8 a– c 3 .1 5 3 .7 7 3 .3 1 3 .9 4 8 .8 2 ab 8 .7 5 9 .2 3 b c 9 .0 6 b c W h e– C an – W h e– C an 4 2 .0 5 a 2 4 .3 ab 5 .7 1 6 .7 3 3 .7 8 4 .5 2 2 1 .8 a 1 5 .8 0 1 4 .2 6 ab 1 0 .6 4 ab p- v al u es .0 0 8 .0 0 2 .0 9 1 .1 0 4 .0 5 3 .0 5 4 .0 1 0 .1 0 1 < .0 0 0 1 < .0 0 0 1 S E M 6 .4 4 3 .5 6 1 .0 3 1 .1 9 0 .6 7 0 .7 7 3 .1 3 7 2 .7 5 2 1 .5 0 1 .3 4 S E D 9 .1 1 5 .0 4 1 .4 6 1 .6 8 0 .9 4 1 .0 8 4 .4 3 6 3 .8 9 2 2 .1 2 1 .8 9 No te . C E Y , ca n o la eq u iv al en t y ie ld ; C an , ca n o la ; C F , cr ee p in g re d fe sc u e; In R C , in o cu la te d re d cl o v er ; R C , re d cl o v er ; W h e, w h ea t; In A C , in o cu la te d al si k e cl o v er ; A C , al si k e cl o v er ; B ar , b ar le y ; S E D , st an d ar d er ro r o f th e d if fe re n ce in m ea n s. M ea n s se p ar at io n w as p er fo rm ed u si n g T u k ey ’s p ro ce d u re . T h e v al u es fo ll o w ed b y sa m e le tt er s, o v er la p p in g ra n g e o f le tt er s o r w it h o u t le tt er in g ar e n o t si g n if ic an tl y d if fe re n t (p > .0 5 ). 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4078 KHANAL ET AL. establishment year ($160 vs. $120 ha−1), while the creep- ing red fescue-based cropping sequence had additional her- bicidal termination cost ($120 ha−1) after the last produc- tion year. Clover crops did not receive fertilizer in the seed production year, resulting in the saving in fertilizer cost by $211 ha−1 in the plots of 90 kg N ha−1 in clover-based crop- ping sequences, compared with continuous canola and Whe– Can cropping. In the same plots, total cost savings due to less seed and lower seeding cost as perennial crop, lower fertilizer cost, and lower harvest cost (having no harvest in the estab- lishment year) in clover-based cropping sequences compared with continuous canola sequence was $433 ha−1. With the exception of InRC–RC–Whe–Can, the economic NUE val- ues of clover-based sequences were generally lower than other crop sequences. Under a minimum price scenario, the agronomic NUE for CEY differed significantly (p = .01) between the cop sequences at 45 kg N ha−1, but did not differ at 90 kg N ha−1 (p = .1). At 45 kg N ha−1, three clover-based sequences had significantly lower NUE for CEY (<6.83) than that of Whe–Can–Whe–Can (21.8), while the rest of the sequences remained on a par with both the highest and lowest NUE val- ues. In the average price scenario, the creeping red fescue- based sequence had significantly higher NUE for CEY than all crop sequences except for Whe–Can repeating sequences at both 45 kg N ha−1 (18.44 vs. <9.23) and 90 kg N ha−1 (16.80 vs. <9.06). The InRC–RC–Whe–Can sequence was an exception in all results showing contrast with the other three clover-based sequences. 4 DISCUSSION This study examined the agronomic and economic values of integrating perennial forage crops in the annual crop-based cropping sequences. Eight diverse cropping sequences under study included four annual crops (Whe, Can, Pea, and Bar), and three perennial forage seed crops (CF [turf-grass], AC, and RC [forage legumes]). The eight cropping sequences fall under three broad categories, as follows: (a) exclusively annual crop-based sequences, (b) perennial turf-grass seed crop-based sequence, and (c) perennial legume seed crop- based sequences. The annual crop-based sequences varied in diversity levels from continuous canola through Whe–Can alternations to diverse cropping sequence of Pea–Bar–Whe– Can. Similarly, the perennial turf-grass seed crop of three years was followed by canola, and perennial legume seed crops of two years were followed by wheat and canola in the sequences. The agronomic and economic outcomes of the study are discussed in terms of N economy and break crop effects, biomass productivity and contribution to soil car- bon (C) stock, and economic viability of different cropping sequences. 4.1 Nitrogen economy and break crop effects In this study, Whe–Can alternating sequence (Whe–Can– Whe–Can) had the most recurrent wheat and canola crops, with only one break crop in the sequence. Without regards to the N levels, overall mean yield of wheat under cropping sequences that included biennial seed crop of red clover (RC– RC–Whe–Can) and alsike clover (AC–AC–Whe–Can) was up to 35% higher than that of alternating Whe–Can sequence. Similarly, the mean yield of wheat crop preceded by Pea–Bar in sequence (Pea–Bar–Whe–Can) was 22% higher than that of Whe–Can alternating sequence. Thus, the mean yield of wheat in the sequences preceded by biennial crops of clovers was about 11% higher than that of wheat under the diverse annual crop sequence containing pea as a legume crop. The break crop effects of preceding crops on wheat in our study are similar to those reported by Angus et al. (2015). They reported that the mean yield increase of wheat varied from 0.5 t ha−1 after oats to 1.2 t ha−1 after grain legumes, while other broad-leaf crops such as canola, mustard [Brassica juncea (L.) Czern.], and flax (Linum usitatissimum L.) have intermediate effects on following wheat crop. Two successive break crops resulted in 0.1–0.3 t ha−1 higher yield than after a single break crop (Angus et al., 2015). In our study, strong effects of cropping sequences masked the effects of N supple- mentation to the wheat crop. The yields of wheat under unfer- tilized clover-based sequences were statistically similar to those at 90 kg N ha−1 under Whe–Can alternating sequences. This could be attributed to both N fixation and nonnitroge- nous break crop effect of the 2 yr of leguminous clover crops before the wheat crop in the cropping sequence. In a previous study conducted in the Peace River region, annual N fixation by seed crops of alsike clover and red clover was estimated in the range of 20.8–143.0 kg ha−1 and 15.3–77.3 kg ha−1, respectively (Rice, 1980). Intuitively, the wheat yield results in our study imply that integration of clover seed crops in the tight sequences of wheat and canola can replace the need of external N application for the succeeding crop of wheat. Canola was the fourth phase common crop for all cropping sequences in our study. Except for two crop sequences, Can– Can–Can–Can and CF–CF–CF–Can, wheat was the immedi- ately preceding crop of canola for the rest of the six crop- ping sequences. Preceding crop effects of legumes and break crops over 2–3 yr in the sequence evidently passed down to canola. Without regard to the N levels, overall mean yield of canola under clover seed crop-based sequences (RC– RC–Whe–Can and AC–AC–Whe–Can) was up to 46 and 31% higher than that of continuous canola and alternat- ing Whe–Can sequences, respectively. Similarly, the mean yields of canola preceded by Pea–Bar–Whe in sequence (Pea– Bar–Whe–Can) and alternating Whe–Can sequence were 16 and 11% higher than that of continuous canola sequence, 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4079 respectively. The cropping sequence effects of annual crop- ping systems in our study corroborate the previous study find- ings in western Canada (Gill, 2018; Harker et al., 2015; Liu et al., 2019). In a recent study conducted in loamy Luvisol in the Peace region, Gill (2018) reported that one or two break crops of pea, barley, or flax resulted in 19.4 and 7.2% increase in canola and wheat yield over the unbroken sequence of canola and wheat, respectively. Similarly, in a study con- ducted in five locations including the Peace River region in western Canada, canola grown in an alternating sequence after two break crops in rotation yielded 22% higher com- pared with continuous canola (Harker et al., 2015). Another cropping sequence study conducted in three sites in west- ern Canada showed that replacing cereal wheat with pulse crop lentil (Lens culinaris Medik.) resulted in increase in the system productivity, expressed as canola equivalent yield, by 33%, which is mainly attributable to a N-related benefit of pre- ceding lentil crop to canola (Liu et al., 2019). Hegewald et al. (2018) reviewed extensively the effects of preceding crops on seed yield of oilseed rape (Brassica napus L.) and inferred that three break crops resulted in a consistent yield benefit of 0.47 t ha−1 of oilseed rape, and that much of the yield advantage was attributed to the reduction of pressures of insect pests, dis- eases, and weeds. Integration of forage legume seed crops in the annual cropping sequences thus provides both nitrogenous and nonnitrogenous benefits. Yield of canola after a 3-yr perennial stand of creeping red fescue seed crop was lowest of all crop sequences across all levels of N fertilizer applied. However, the yield gap of canola diminished with the increase in N level. A study by White and Schmid (1963) showed a N dose-dependent yield response of annual crops that followed after perennial grasses. One of the causes of poorer yield of canola after creeping red fescue is believed to be sod binding condition created by dense roots of creeping red fescue in the zero-tillage practice adopted in the experimental plots. Sod binding has been reported as the major cause of rapid decline in seed production of creeping red fescue with the increase in age of the stand (Sharp, 1965). Another effect on canola yield after creeping red fescue was presumably the result of appropriation of soil N by microor- ganisms to act on the roots of creeping red fescue causing N deficiency to canola crop at low fertility level. A study by Hodge et al. (2000) showed competition between roots and soil microorganisms for nutrients in the soil. Decomposi- tion of high C/N ratio plant materials such as nonleguminous straw and roots can cause N immobilization and temporary N deficiency for crop plants (Jingguo & Bakken, 1997; Kor- saeth et al., 2002). The narrowing gap of canola yield between creeping red fescue-based and other cropping sequences, with the increasing rates of N application, suggests N-limiting con- ditions for canola in the creeping red fescue-based cropping sequence. The cropping sequences had differential effects on var- ious measures of NUE. The Whe–Can and creeping red fescue-based cropping sequences had higher biomass NUE and agronomic NUE (based on CEY) than most of the clover- based and continuous canola sequences. However, the gross revenue-based economic NUE did not differ significantly between the cropping sequences. The agronomic and biomass NUE results in our study implied that cropping systems with high frequency of grass species (such as wheat and creep- ing red fescue) had higher NUEs. These results align with a 12-yr study from five cropping systems in the semiarid prairie region of western Canada, which showed that contin- uous wheat and Whe–Can–Whe–Pea cropping sequences had higher NUE for grain production than a sequence containing lentil green manure followed by wheat (St. Luce et al., 2020). 4.2 Biomass productivity and contribution to soil C stock In our study, annual crop-based cropping sequences produced higher aerial biomass than perennial crop-based sequences. The 4-yr average biomass productivity across different fer- tility regimes of W–C alternating sequence was the high- est (29,859 kg ha−1) followed in order by pea-based (Pea– Bar–Whe–Can) (95%), continuous canola (83%), and clover- based sequences (57–76%), while the creeping red fescue- based sequences produced the least biomass (56%). In a com- parative study between annual and perennial biomass crops, Hallam et al. (2001) also reported higher biomass productiv- ity in annual crops. By using a probabilistic modeling com- bined with a trait database of annual and perennial crops, Vico and Brunsell (2018) showed that perennial seed crops had more extensive root systems contributing to more stable yields than annuals. A study by Bolinder et al. (2007) showed that about half of the net primary productivity of perennial forages lie below ground with an average shoot–root ratio of 1.6, while the belowground net primary productivity of annual crops was about 20% with a shoot–root ratio of >5. A recent study further corroborated those findings and reported that within the soil depth of 1 m, average root/shoot ratio of pea, barley, wheat, and canola were 0.211, 0.200, 0.217, and 0.369, respectively, while that of forages averaged at 0.868 with the values of 0.618 in establishment year and 1.213 in the production years (Thiagarajan et al., 2018). Applying those root/shoot ratio factors to the aerial biomass to estimate the below-ground biomass productivity in our study, the perennial forage seed crop-based sequences produced over 45% higher biomass below ground than the annual crop-based sequences over a 4-yr cycle. Comparing 2-yr annual cropping sequences of Bar–Can with perennial cropping sequences of timothy (Phleum pratense L.) and orchardgrass (Dactylis glomerata L.), Lasisi et al. (2018) reported five to sixfold higher root 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4080 KHANAL ET AL. biomass in perennials than that of annual cropping sequences. Compared with the results of their study, the root biomass of perennials is underestimated in our study. The contribu- tion of roots to relatively stable soil C pools is estimated to be 2.3 times higher than that of shoot-derived plant mate- rial (Kätterer et al., 2011). Therefore, perennial forage crops in the cropping sequences can be instrumental in increasing soil organic C, which in turn enhances water holding capac- ity, infiltration, and microbial activities, rendering the soil more favorable for crop production (reviewed by Reeves et al., 1997). Soil organic matter is an indicator of soil natural cap- ital that buffers yield variance against adverse weather while reducing reliance on external inputs (Cong et al., 2014). Inclu- sion of forage crops, especially the legumes in the crop rota- tions, is conducive to higher productivity of the following crops by virtue of nitrogenous-fertilizer economy, increased soil organic matter, improved soil quality, disruption of dis- ease cycle and insect pest epidemics, and shifts in the weed population away from arable crop weeds (Entz et al., 1995, 2002). In the Luvisolic soil of the Peace River region and other parts of the Boreal ecoregion, lack of soil organic matter has been recognized as a major limiting factor in crop production. Global climate change has added further risks and uncertainty associated with biotic and abiotic stresses. In this challenge, crop diversification with perennial forage seed crops can be one of the most effective sustainable solutions to improve crop productivity while protecting the environment. 4.3 Economic viability of different cropping sequences The comparative economic advantage of different cropping sequences was analyzed in terms of gross revenue, gross mar- gin, and CEY at minimum and average prices of the grains of annual crops and seeds of perennial crops of creeping red fescue, and alsike and red clovers. Under both price scenar- ios, the CEY, gross revenue, and gross margin of creeping red fescue-based sequences were higher than that of annual crop-based cropping sequences. These benefits of creeping red fescue are attributed to the several-fold higher price of fescue seeds and lower partial variable costs of production compared with food grains of annual crops. During the study period, the output prices of creeping red fescue were 4–5 times higher than that of canola, about 9 times higher than that of wheat, 8–11 times higher than that of barley, and 6–7 times higher than that of pea. In terms of partial variable cost of production, creeping red fescue-based cropping sequences have a cost savings of $282–$324 ha−1. Between the annual crop sequences, the CEY, gross revenue and gross mar- gins increased with the increase in diversity from continuous canola through Whe–Can alternation to Pea–Bar–Whe–Can sequences. These results align with the findings of Smith et al. (2013), where profitability of cropping sequences increased with the integration of break crops of wheat or flax in the con- tinuous canola and pea rotations. In line with our study find- ings with annual crop-based cropping sequences, benefits of crop diversification with leguminous crops were further eluci- dated by MacWilliam et al. (2014). In their study, replacement of a wheat crop from Can–Whe–Whe–Whe sequence by legu- minous crops lentil and dry pea resulted in higher economic return accompanied by reduced environmental effects pertain- ing to global warming and resources use, ecosystem quality, and human health. Cropping sequences based on clover seed crops had lower CEYs, gross revenues, and gross margins than that of annual crops sequences. Contrarily, although the alsike and red clover seeds fetched five-to-seven times higher prices than canola grains coupled with saving in production costs in clover- based sequences by up to $433 ha−1, lower seed productiv- ity of clovers bore less economic advantage in the cropping sequences compared with annual crop-based and creeping red fescue-based cropping sequences. These results are in contrast with the finding of three case study regions across Europe, in which integration of forage legumes into cropping systems increased gross margins (Reckling et al., 2016). The results of another study conducted in the Palouse region of the U.S. Pacific Northwest (Wieme et al., 2020) were also in contrast to the results of our study with forage legumes. In that study, 8-yr organic cropping sequence integrating perennial alfalfa (Medicavo sativa L.) enhanced the profitability of the organic grain cropping systems. In the Peace River region of Canada, slow establishment and vernalization requirement for peren- nial seed crops render the establishment year as a seasonal lag without any seed production. This causes a loss of 1 yr in the cropping sequence, which can have implications on the prof- itability. To compensate for this loss, some farmers underseed clover and grass seed crops with annual crops in the Peace River region of western Canada. One limitation of our current study was that the perennial forage seed crops were seeded as sole crops in the establish- ment year. Therefore, the study did not represent the farm- ers’ strategy of yield compensation for establishment year of forage crops by intercropping with suitable annual crops as a catch crop. A number of studies report the use of cereal clover intercropping as forage, companion, or cover crops (Blaser et al., 2006; Gaudin et al., 2013; Schmidt et al., 2003; Thorsted, Olesen, et al., 2006; Thorsted, Weiner, et al., 2006, and literature cited therein). Further study should consider clover and annual crops intercropping in the establishment year as it is shown to be an agroecologically benign and economically viable option in cropping systems design. Differential input requirements for different crop species and differential prices offered for the output commodities were the major determinants of eco- nomic viability of different cropping sequences. The forage 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense KHANAL ET AL. 4081 seed crops generally mature earlier in the season than annual seed crops. This allows farmers to stagger the use of labor and equipment over the season, while at the same time diversify production and commodity portfolio for buffering against the marketing risks. Change in the commodity price scenario from minimum to average prices of annual grains and perennial forage seeds resulted in changes in the CEY, gross revenue, and gross mar- gin of the cropping sequences. However, relative economic advantage still held for creeping red fescue-based sequence under two different price scenarios of the commodities. We had further curiosity about how viable the cropping sequences would be if all fixed and variable costs of produc- tion were considered. For this purpose, we adapted cost of production data for relevant crops from the production plans of the neighboring province of Saskatchewan for annual crops and the next province of Manitoba for forage seed crops (Sup- plemental Tables 2 and 3). The fixed costs in those production plans included building repair, property taxes, business over- head, machinery depreciation, building depreciation, machinery investment, building investment, land investment, and common variable costs (excluding partial variable costs considered in our study) including labor and management, storage costs, crop insurance premium, chemicals (herbicide, fungicide), and fertilizer (P, K, S). Those fixed and variable costs, which were common for all crop kinds, would total between $2,132 and $2,156 ha−1, making maximum dif- ferences of $24 ha−1 between the 4-yr cropping sequences. Considering all these costs, only creeping red fescue-based sequences would have positive net returns and a benefit–cost ratio of greater than one under the minimum price scenario (Supplemental Figure 1). Changing the price scenario to average level would result in positive net returns with benefit–cost ratio >1 for alsike clover-based and diversified annual crop-based (Pea–Bar–Whe–Can) cropping sequences at all N levels. Increasing N application from 0 through 45 kg ha−1 to 90 kg N ha−1 would improve the profitability of most of the cropping sequences. It is intriguing how farmers have been practicing these cropping sequences in varying proportions over time. There must be some other motivational factors and implicit incentives such as opportunity costs for farmers to continue these cropping systems. Hence, the net benefits derived from secondary source data may not be the exact representation of the reality. In such circumstances, gross margin analysis may be a safer tool to evaluate relative advantage of various cropping sequences. Considering the total costs of production adapted from secondary data from other provinces, the relative economic advantage expressed in terms of net returns of different cropping sequences would remain the same as that revealed by gross margins generated from primary and secondary data of the Peace River region. Beneficial rotational effects of functionally different crops are extensively documented in various parts of the world. The use of evidence-based crop sequence designs tailoring the coexistence of forage seed and annual field crops has the potential to lead to resilient production practices contribut- ing to a more sustainable agri-industry throughout the Peace River region. There is a lack of studies integrating cumulative systems productivity, comparative economic advantage, soil health, and greenhouse gas emissions of longer-term crop- ping sequences that include forage seed crops and annual field crops. Further studies are needed to generate agronomic, eco- nomic, soil health, and environmental matrices for selected crop sequences, and integrate the systems performance in the Holos model or other whole farm models to provide ratio- nal decision support to the farmers in the Peace River region, while providing catalytic resources for other parts of Canada. 5 CONCLUSIONS This 4-yr cropping sequence study shows that perennial for- age seed crop-based cropping sequences can have higher pro- ductivity and profitability and greater potential to enhance soil C stock compared with annual crop-based sequences. Creeping red fescue-based cropping sequence consistently outperformed annual crop-based sequences in terms of var- ious productivity and economic indicators. The economic benefits were attributable to the reduction in variable costs of production due to lower seed price and less frequency of seeding, and higher gross revenue due to substantially higher sale price of creeping red fescue seeds compared with annual crops. Alsike and red clover as leguminous forage seed crops contributed to higher N economy to following crops wheat and canola in the cropping sequences, replac- ing the full amount of N fertilizer requirement for imme- diate succeeding wheat and partial amount of N fertilizer requirement for next succeeding crop canola. The economic advantage under a given price scenario stood in the order of creeping red fescue-based sequences > high diversity annual crop sequences ≈ alsike clover-based sequence ≈ wheat- canola alternating sequence > continuous canola sequence. In the face of prevailing global climate change, it is impor- tant to examine the agroecological dimensions of effects contributed by perennial forage crops through soil physical, chemical, and biological properties compared with annual cropping systems. This study is particularly relevant to the Peace River region where agriculture is continuously expanding to new frontiers, and land used for the produc- tion of forage seed crops is declining in favor of annual crop production, likely reducing soil organic matter, which is crucial to the resiliency of the cropping systems. Fur- ther studies are warranted to integrate the systems perfor- mance in terms of agronomic, economic, soil health, and environmental matrices for selected crop sequences to pro- vide rational decision support to the farmers and devising 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4082 KHANAL ET AL. appropriate agriculture-based climate solutions in the Peace River region. A C K N O W L E D G M E N T S This project was funded by Agriculture and Agri-Food Canada under the Growing Forward Agri-Innovation Program as AgriScience Project J-000656, which was led and funded by the Peace Region Forage Seed Association. The authors are grateful to Pat Ganselves, Mike Leighton, and the team of summer students who provided ongoing technical support to this research. AU T H O R C O N T R I B U T I O N S Nityananda Khanal: Data curation; Formal analysis; Method- ology; Project administration; Supervision; Visualization; Writing-original draft. Rahman Azooz: Conceptualization; Data curation; Investigation; Methodology. Noabur Rah- man: Data curation; Formal analysis; Validation. Henry Klein-Gebbinck: Formal analysis; Validation. Jennifer K. Otani: Conceptualization; Funding acquisition; Investigation; Methodology; Project administration; Supervision; Valida- tion. Calvin L. Yoder: Conceptualization; Methodology; Vali- dation. Talon M. Gauthier: Conceptualization; Funding acqui- sition; Project administration; Validation. C O N F L I C T O F I N T E R E S T The authors declare no conflict of interest. O R C I D Nityananda Khanal https://orcid.org/0000-0003-3847- 3930 R E F E R E N C E S Agriculture Financial Services Corporation. (2019). AgriStability price lists. https://afsc.ca/income-stabilization/agristability/pricing/ Alberta Agriculture and Forestry. (2016). Agriculture surveys: Data and trends related to the agriculture and agri-food industries. https://open. alberta.ca/dataset/1200-9814 Alberta Agriculture and Forestry. (2020). Alberta Soil Information Viewer. https://soil.agric.gov.ab.ca/agrasidviewer/ Angus, J. F., Kirkegaard, J. A., Hunt, J. R., Ryan, M. H., Ohlander, L., & Peoples, M. B. (2015). Break crops and rotations for wheat. Crop and Pasture Science, 66, 523–552. https://doi.org/10.1071/CP14252 Beckie, H. J., Shirriff, S. W., Leeson, J. Y., Hall, L. M., Harker, K. N., Dokken-Bouchard, F., & Brenzil, C. A. (2019). Herbicide-resistant weeds in the Canadian prairies: 2012 to 2017. Weed Technology, 34, 461–474. https://doi.org/10.1017/wet.2019.128 Blaser, B. C., Gibson, L. R., Singer, J. W., & Jannink, J. L. (2006). Optimizing seeding rates for winter cereal grains and frost-seeded red clover intercrops. Agronomy Journal, 98, 1041–1049. https://doi.org/ 10.2134/agronj2005.0340 Bolinder, M. A., Janzen, H. H., Gregorich, E. G., Angers, D. A., & Bygaart, V., & A, J. (2007). An approach for estimating net pri- mary productivity and annual carbon inputs to soil for common agricultural crops in Canada. Agriculture, Ecosystems & Environ- ment, 118, 29–42. https://doi.org/10.1016/j.agee.2006.05.013 Broersma, K., Juma, N. G., & Robertson, J. A. (1996). Net nitrogen min- eralization from a Gray Luvisol under diverse cropping systems in the Peace River region of Alberta. Canadian Journal of Soil Science, 76, 117–123. https://doi.org/10.4141/cjss96-018 Broersma, K., Robertson, J. A., & Chanasyk, D. S. (1997). The effects of diverse cropping systems on aggregation of a Luvisolic soil in the Peace River region. Canadian Journal of Soil Science, 77, 323–329. https://doi.org/10.4141/S96-013 Cong, R. G., Hedlund, K., Andersson, H., & Brady, M., (2014). Man- aging soil natural capital: An effective strategy for mitigating future agricultural risks? Agricultural Systems, 129, 30–39. https://doi.org/ 10.1016/j.agsy.2014.05.003 Entz, M. H., Baron, V. S., Carr, P. M., Meyer, D. W., Smith, S R., Jr., & McCaughey, W. P. (2002). Potential of forages to diversify cropping systems in the northern great plains. Agronomy Journal, 94, 240–250. https://doi.org/10.2134/agronj2002.0240 Entz, M., Bullied, W., & Katepa-Mupondwa, F. (1995). Rotational ben- efits of forage crops in Canadian prairie cropping systems. Journal of Production Agriculture, 8, 521–529. https://doi.org/10.2134/jpa1995. 0521 Gaudin, A., Westra, S., Loucks, C. E., Janovicek, K., Martin, R. C., & Deen, W. (2013). Improving resilience of northern field crop sys- tems using inter-seeded red clover: A review. Agronomy, 3, 148–180. https://doi.org/10.3390/agronomy3010148 Gill, K. S. (2018). Crop rotations compared with continuous canola and wheat for crop production and fertilizer use over 6 yr. Canadian Journal of Plant Science, 98, 1139–1149. https://doi.org/10.1139/ cjps-2017-0292 Government of Canada. (2020). Canadian climate nor- mals 1981–2010 station data: Beaverlodge CDA Alberta. https://climate.weather.gc.ca/climate_normals/results_1981_2010_e. html?stnID=2658%26dispBack=1%26month1=6%26month2=8 Guitard, A. A. (1965). Agriculture in the Peace – past, present, and future. In D. H. Calverley (Ed.), History is where you stand: A history of the Peace Region. South Peace Historical Society. https://calverley. ca/article/08-003-agriculture-in-the-peace-past-present-and-future/ Haarsma, D. G. (2014). Spatial analysis of agricultural land conver- sion and its associated drivers in Alberta [Master’s thesis, University of Alberta]. University of Alberta Library. https://doi.org/10.7939/ R34B2XD2T Hallam, A., Anderson, I. C., & Buxton, D. R. (2001). Comparative economic analysis of perennial, annual, and intercrops for biomass production. Biomass and Bioenergy, 21, 407–424. https://doi.org/10. 1016/S0961-9534(01)00051-4 Hamley, W. (1992). The farming frontier in northern Alberta. The Geo- graphical Journal, 158, 286–294. https://doi.org/10.2307/3060297 Harker, K. N., O’Donovan, J. T., Turkington, T. K., Blackshaw, R. E., Lupwayi, N. Z., Smith, E. G., Dosdall, L. M., Hall, L. M., Kutcher, H. R., Willenborg, C. J., Peng, G., Irvine, R. B., & Mohr, R. (2015). Canola cultivar mixtures and rotations do not mitigate the negative impacts of continuous canola. Canadian Journal of Plant Science, 95, 1085–1099. https://doi.org/10.4141/cjps-2015-126 Hegewald, H., Wensch-Dorendorf, M., Sieling, K., & Christen, O. (2018). Impacts of break crops and crop rotations on oilseed rape productivity: A review. European Journal of Agronomy, 101, 63–77. https://doi.org/10.1016/j.eja.2018.08.003 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://orcid.org/0000-0003-3847-3930 https://orcid.org/0000-0003-3847-3930 https://orcid.org/0000-0003-3847-3930 https://afsc.ca/income-stabilization/agristability/pricing/ https://open.alberta.ca/dataset/1200-9814 https://open.alberta.ca/dataset/1200-9814 https://soil.agric.gov.ab.ca/agrasidviewer/ https://doi.org/10.1071/CP14252 https://doi.org/10.1017/wet.2019.128 https://doi.org/10.2134/agronj2005.0340 https://doi.org/10.2134/agronj2005.0340 https://doi.org/10.1016/j.agee.2006.05.013 https://doi.org/10.4141/cjss96-018 https://doi.org/10.4141/S96-013 https://doi.org/10.1016/j.agsy.2014.05.003 https://doi.org/10.1016/j.agsy.2014.05.003 https://doi.org/10.2134/agronj2002.0240 https://doi.org/10.2134/jpa1995.0521 https://doi.org/10.2134/jpa1995.0521 https://doi.org/10.3390/agronomy3010148 https://doi.org/10.1139/cjps-2017-0292 https://doi.org/10.1139/cjps-2017-0292 https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?stnID=2658%26dispBack=1%26month1=6%26month2=8 https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?stnID=2658%26dispBack=1%26month1=6%26month2=8 https://calverley.ca/article/08-003-agriculture-in-the-peace-past-present-and-future/ https://calverley.ca/article/08-003-agriculture-in-the-peace-past-present-and-future/ https://doi.org/10.7939/R34B2XD2T https://doi.org/10.7939/R34B2XD2T https://doi.org/10.1016/S0961-9534(01)00051-4 https://doi.org/10.1016/S0961-9534(01)00051-4 https://doi.org/10.2307/3060297 https://doi.org/10.4141/cjps-2015-126 https://doi.org/10.1016/j.eja.2018.08.003 KHANAL ET AL. 4083 Hodge, A., Stewart, J., Robinson, D., Griffiths, B. S., & Fitter, A. H. (2000). Competition between roots and soil micro-organisms for nutrients from nitrogen-rich patches of varying complexity. Journal of Ecology, 88, 150–164. https://doi.org/10.1046/j.1365-2745.2000. 00434.x Hoyt, P. B. (1990). Residual effects of alfalfa and bromegrass crop- ping on yields of wheat grown for 15 subsequent years. Canadian Journal of Soil Science, 70, 109–113. https://doi.org/10.4141/cjss90- 012 Hoyt, P. B., & Leitch, R. H. (1983). Effects of forage legume species on soil moisture, nitrogen, and yield of succeeding barley crops. Cana- dian Journal of Soil Science, 63, 125–136. https://doi.org/10.4141/ cjss83-012 Jingguo, W., & Bakken, L. R. (1997). Competition for nitrogen during decomposition of plant residues in soil: Effect of spatial placement of N-rich and N-poor plant residues. Soil Biology and Biochemistry, 29, 153–162. https://doi.org/10.1016/S0038-0717(96)00291-X Kätterer, T., Bolinder, M. A., Andrén, O., Kirchmann, H., & Menichetti, L. (2011). Roots contribute more to refractory soil organic matter than above-ground crop residues, as revealed by a long-term field experiment. Agriculture, Ecosystems & Environment, 141, 184–192. https://doi.org/10.1016/j.agee.2011.02.029 Korsaeth, A., Henriksen, T. M., & Bakken, L. R. (2002). Tem- poral changes in mineralization and immobilization of N during degradation of plant material: Implications for the plant N supply and nitrogen losses. Soil Biology and Biochemistry, 34, 789–799. https://doi.org/10.1016/S0038-0717(02)00008-1 Lasisi, A. A., Akinremi, O. O., Tenuta, M., & Cattani, D. (2018). Below- ground plant biomass and nitrogen uptake of perennial forage grasses and annual crops fertilized with pig manures. Agriculture, Ecosystems & Environment, 268, 1–7. https://doi.org/10.1016/j.agee.2018.08. 006 Lavkulich, L. M., & Arocena, J. M. (2011). Luvisolic soils of Canada: Genesis, distribution, and classification. Canadian Journal of Soil Science, 91, 781–806. https://doi.org/10.4141/cjss2011-014 Liu, K., Johnson, E. N., Blackshaw, R. E., Hossain, Z., & Gan, Y. (2019). Improving the productivity and stability of oilseed cropping sys- tems through crop diversification. Field Crops Research, 237, 65–73. https://doi.org/10.1016/j.fcr.2019.03.020 MacWilliam, S., Wismer, M., & Kulshreshtha, S. (2014). Life cycle and economic assessment of Western Canadian pulse systems: The inclu- sion of pulses in crop rotations. Agricultural Systems, 123, 43–53. https://doi.org/10.1016/j.agsy.2013.08.009 Manitoba Agriculture and Resource Development. (2016). Guide- lines for estimating forage seed production costs 2016. https://www.gov.mb.ca/agriculture/farm-management/ production-economics/pubs/cop_crop_forageseed.pdf Pawluk, S. (1980). Micromorphological investigations of cultivated Gray Luvisols under different management practices. Canadian Jour- nal of Soil Science, 60, 731–745. https://doi.org/10.4141/cjss80- 082 Reckling, M., Bergkvist, G., Watson, C. A., Stoddard, F. L., Zander, P. M., Walker, R. L., Pristeri, A., Toncea, I., & Bachinger, J. (2016). Trade-offs between economic and environmental impacts of introduc- ing legumes into cropping systems. Frontiers in Plant Science, 7, 669. https://doi.org/10.3389/fpls.2016.00669 Reeves, D. W. (1997). The role of soil organic matter in maintaining soil quality in continuous cropping systems. Soil & Tillage Research, 43, 131–167. https://doi.org/10.1016/S0167-1987(97)00038-X Rice, W. A. (1980). Seasonal patterns of nitrogen fixation and dry matter production by clovers grown in the Peace River region. Canadian Journal of Plant Science, 60, 847–858. https://doi.org/10. 4141/cjps80-124 Saskatchewan Agriculture. (2016). Crop planning guide – 2016. https: //publications.saskatchewan.ca/#/products/77820 Schmidt, O., Clements, R. O., & Donaldson, G. (2003). Why do cereal- legume intercrops support large earthworm populations? Applied Soil Ecology, 22, 181–190. https://doi.org/10.1016/S0929-1393(02) 00131-2 Sharp, W. C. (1965). Effects of clipping and nitrogen fertilization on seed production of creeping red fescue. Agronomy Journal, 57, 252–253. https://doi.org/10.2134/agronj1965.00021962005700030008x Smith, E. G., Kutcher, H. R., Brandt, S. A., Ulrich, D., Malhi, S. S., & Johnston, A. M. (2013). The profitability of short-duration canola and pea rotations in western Canada. Canadian Journal of Plant Science, 93, 933–940. https://doi.org/10.4141/cjps2013-021 Soon, Y. K., Klein-Gebbinck, H. W., & Arshad, M. A. (2005). Residue management and crop sequence effects on the yield and brown girdling root rot of canola. Canadian Journal of Plant Science, 85, 67–72. https://doi.org/10.4141/P04-058 St Luce, M., Lemke, R., Gan, Y., McConkey, B., May, W., Con, C., Zentner, R., Wang, H., Kroebel, R., Fernandez, M., & Brandt, K. (2020). Diversifying cropping systems enhances productivity, stabil- ity, and nitrogen use efficiency. Agronomy Journal, 112, 1517–1536. https://doi.org/10.1002/agj2.20162 Stewart, G. M. (1933). Survey of the forage crop seed situation in the prairie provinces of Canada. Scientific Agriculture, 13, 687–697. https://doi.org/10.4141/sa-1933-0063 Strelkov, S. E., Hwang, S. F., Manolii, V. P., Turnbull, G., Fredua- Agyeman, R., Hollman, K., & Kaus, S. (2020). Characteriza- tion of clubroot (Plasmodiophora brassicae) from canola (Brassica napus) in the Peace Country of Alberta, Canada. Canadian Journal of Plant Pathology, 1, 155–161. https://doi.org/10.1080/07060661. 2020.1776931 Thiagarajan, A., Fan, J., McConkey, B. G., Janzen, H. H., & Campbell, C. A. (2018). Dry matter partitioning and residue N content for 11 major field crops in Canada adjusted for rooting depth and yield. Cana- dian Journal of Soil Science, 98, 574–579. https://doi.org/10.1139/ cjss-2017-0144 Thorsted, M. D., Olesen, J. E., & Weiner, J. (2006). Width of clover strips and wheat rows influence grain yield in winter wheat/white clover intercropping. Field Crops Research, 95, 280–290. https://doi.org/10. 1016/j.fcr.2005.04.001 Thorsted, M. D., Weiner, J., & Olesen, J. E. (2006). Above-and below- ground competition between intercropped winter wheat Triticum aes- tivum and white clover Trifolium repens. Journal of Applied Ecology, 43, 237–245. https://doi.org/10.1111/j.1365-2664.2006.01131.x Turkington, T. K., Clear, R. M., Demeke, T., Lange, R., Xi, K., & Kumar, K. (2011). Isolation of Fusarium graminearum from cereal, grass and corn residues from Alberta, 2001-2003. Canadian Journal of Plant Pathology, 33, 179–186. https://doi.org/10.1080/07060661. 2011.560189 Van Vliet, L. J. P., & Hall, J. W. (1991). Effects of two crop rotations on seasonal runoff and soil loss in the Peace River region. Cana- dian Journal of Soil Science, 71, 533–543. https://doi.org/10.4141/ cjss91-051 Vico, G., & Brunsell, N. A. (2018). Tradeoffs between water require- ments and yield stability in annual vs. perennial crops. Advances in 14350645, 2021, 5, D ow nloaded from https://acsess.onlinelibrary.w iley.com /doi/10.1002/agj2.20781 by C ochrane C anada Provision, W iley O nline L ibrary on [17/08/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1046/j.1365-2745.2000.00434.x https://doi.org/10.1046/j.1365-2745.2000.00434.x https://doi.org/10.4141/cjss90-012 https://doi.org/10.4141/cjss90-012 https://doi.org/10.4141/cjss83-012 https://doi.org/10.4141/cjss83-012 https://doi.org/10.1016/S0038-0717(96)00291-X https://doi.org/10.1016/j.agee.2011.02.029 https://doi.org/10.1016/S0038-0717(02)00008-1 https://doi.org/10.1016/j.agee.2018.08.006 https://doi.org