Parasitic weeds cause huge annual losses to food production globally. A small number of species from the genera Cuscuta, Orobanche, Phelipanche and Striga have proliferated across many agroecological zones. Their control is compromised due to the lack of efficacy of conventional herbicides and their rapid adaptation to new resistant crop cultivars. A broad range of studies suggest consistent reductions in parasitic weed densities owing to increased spatial (intercropping) and temporal diversity (crop rotation). However, to date, no synthesis of this body of research has been published. Here we report the results of a meta-analysis using 1525 paired observations from 67 studies across 24 countries, comparing parasitic weed density and crop yields from monocrop and more diverse cropping systems. We found both spatial and temporal crop diversification had a significant effect on parasitic weed density reduction. Furthermore, our results show effects of spatial diversification are stronger in suppressing parasitic weeds than temporal effects. Furthermore, the analysis indicates intercrops which alter both microclimate and soil chemistry (e.g. Crotalaria, Stylosanthes, Berseem clover and Desmodium) are most effective in parasitic weed management. This analysis serves to underline the viability of crop diversification as a tool to enhance food security globally.
Weeds currently represent the most significant factor limiting agricultural production, with crop yield reductions attributable to weeds estimated at 34% globally1,2. Amongst the most serious weeds are a small number of parasitic plants of the genera Cuscuta, Orobanche, Phelipanche and Striga have proliferated, impacting food production worldwide3,4,5. Parasitic weeds disproportionately affect subsistence farming in the developed world6, exacerbating food insecurity and confounding poverty alleviation initiatives. Simplification of cropping systems has been recognised as a key driver of agricultural weeds in general7. This is also the case for parasitic weeds, which predominantly affect low-diversity agricultural systems, with large-scale monocultures providing a continuous supply of host plants, facilitating their spread5,8.
Agrodiversity describes diversity within varieties and species of cultivated crops, crop-management systems and techniques, as well as insect and soil biodiversity9,10. A key element of agrodiversity is the diversity of cultivated crop species11, including genetic diversity at the varietal and landrace level12. Techniques used to enhance crop diversity include crop rotation13, intercropping14, relay cropping15, cover cropping16 and the use of cultivar mixes of the same species12. Under a broad range of conditions it is possible to maintain yields whilst reducing the use of chemical fertilisers and herbicides17,18. Furthermore, diversification has been shown to stabilise and increase yields when compared with less diverse systems19. Such effects have been demonstrated on field, landscape and national scales18,20,21 as well as across climatic gradients22,23.
Diversification of crop rotations has been shown to have a significant positive effect on weed control24,25. For example, Weisberger et al.7 found an average weed density reduction of 49% in diverse crop rotations, compared with monocrops. Similarly, the role of intercrops in the suppression of weeds has been demonstrated across a wide range of crop types within both tropical and temperate biomes (e.g.26,27,28,29). However, in some studies this effect has been less evident, with significant variability in results between crops, years and locations (e.g.30,31,32).
Parastic weeds damage crops when they attach to the host plant, and before they become visible above ground. The unique rootlike haustorium of a parasitic weed penetrates the host plant’s vascular system, and allows the parasite to assimilate nutrients and water33. Host plant attachment reduces the photosynthetic requirements of parasitic weeds either completely in the case of holoparasites such as Cuscuta or partially in the case of hemiparasites such as Orobanche and Striga33.
Conventional weed management typically targets above-ground growth so that techniques such as herbicide application and mechanical weeding are frequently ineffective when applied to parasitic weeds. Methods of parasitic weed control must therefore focus on the reduction of germination and primary growth. Mechanisms to reduce parasitic weed recruitment include alteration of soil chemistry34,35, germination in the absence of available hosts (suicidal germination)36,37 and altering soil microclimate38,39,40,41.
Intercropping can suppress weeds, including parasitic species, through several mechanisms. The effect of niche complementarity has been observed in intercrops, particularly for cereal-legume combinations, because legumes facilitate increased input of fixed N2 cropping systems whilst not affecting uptake N uptake for the associated cereal crop42. Increased resource use efficiency by intercrops through differing nutrient requirements between crops has also been shown to assist in weed suppression. For example, Haugaard-Nielsen et al.28 found enhanced interception of N by when intercropped with pea, compared to barley monocrop, which resulted in reduced weed incidence. Another important mechanism is the allelopathic effects of some crops on weeds when grown in rotation (e.g.43,44).
Additionally, a suite of alternative tactics can also help mitigate yield losses from parasitic weeds, including the use of resistant crop varieties45,46,47 and post attachment tolerance of parasitic weeds by host crops48. Combinations of crops, intercrops, rotation crops and varieties thereof may therefore manage or mitigate the effects of parasitic weeds in any number of ways listed above. A huge number of field trials have been undertaken to analyse the effectiveness of these approaches to reducing parasitic weed densities and mitigating yield losses. However to date there has been no synthesis of this work and, to our knowledge, no analysis has been undertaken of the relative effects of rotation and intercropping on economically significant parasitic weeds. Such an analysis will improve our ability to manage parasitic weeds through improving our understanding of the efficacy of different components of agrodiversity.
Here we present the results of a meta-analysis of the effects of crop diversity on parasitic weeds using an extensive data set derived from laboratory, field, farm and landscape studies. This represents the first quantitative synthesis of the effects of crop diversification on parasitic weeds and associated crop yields. We address the following questions: Does crop diversity, expressed as the incorporation of additional crops within a system, affect parasitic weed density or crop yield? In terms of management factors, what are the strongest predictors of variation in parasitic weed density and crop yield? Which are the best-performing combinations of crops/intercrops and/or rotation crops in terms of weed reduction and yield increase? Finally, as an ancillary analysis, we analyse the effect of climate and altitude on weed densities.
A pilot study was undertaken to determine terms for the main search. This comprised the systematic search for relevant studies of electronic databases: Web of Science, Scopus and AGRICOLA using a range of Boolean search terms. The pilot study used combinations of provisional terms in conjunction with the genera: Striga and Orobanche (being among the most economically significant parasitic weed genera). The number of returns for each search combination, accompanied by an assessment of relevance based on the title of each study, indicated their relevance. This determined the final list of terms for inclusion, as some terms were too broad and returned too many unrelated results. Search combinations returning very high (e.g. > 400) numbers of records with a very large proportion of non-relevant studies indicated that the term was too broad and was omitted from the main search (e.g.: “Taxon” AND inter*, “Taxon” AND Legum*).
The choice of taxa for inclusion in the main search was determined by a review of economically significant parasitic plants using several sources33,49,50. The list was then subject to triage based on the nature of their parasitism, removing weeds affecting woody crops (trees and shrubs). Genera that returned no results for the search combinations were removed from the main search. In the case of genera containing high numbers of economically important species (e.g.: Cuscuta, Striga), the genus was included as a search term alone without going to the species level. Widely adopted synonyms at the family and genus level were also included. Appendix 1 details search combinations used for the pilot, results, list of taxa, synonyms, and full search methodology.
Main searches were performed in February 2021. Search strings were constructed as follows: First, the name for each parasitic weed taxon was used (Aeginetia, Alectra, Christisonia, Cuscuta, Grammica, Orobanche, Phelipanche, Scrophulariaceae, Striga). Each taxon name was then added to the following search term combinations: cover AND crop, taxon name AND Intercrop, taxon name AND trap*, taxon name AND push AND pull, taxon name AND companion, taxon name AND conservation AND agriculture *, taxon name AND integrated weed management, taxon name AND cultural AND control, taxon name AND suicidal*, taxon name AND legume, taxon name AND no AND till, taxon name AND zero AND till.
Additional searches were performed between May 2021 and February 2022 by manually searching for citations within relevant sections of 20 review studies of control methods for all economically significant parasitic weed taxa. The full list of review studies consulted is included in Appendix 2. Experts in the field of parasitic weed agronomy were also contacted to identify possible sources of data (including primary data) and to verify the thoroughness of our literature coverage. The list of studies and subsequent data were updated periodically as additional sources became available.
Criteria for inclusion of studies
Studies were included if they fulfilled the following relevance criteria:
Subjects studied: Any annual parasitic weed species, any combination of host and intercrops.
Treatment used: Intercropping or rotation cropping.
Study type: Any primary landscape, laboratory, field trials, farm trials, pot, bag and rhizotron studies with appropriate comparators.
Data type: Continuous data with means, information on sample sizes, available/calculable measures of variance or sufficient information to impute values.
Response(s): Host yield (t ha−1/kg ha−1), stover yield (t ha−1), weed dry weight (t ha−1/g pot/g plant/gm2), weed/weed seed density (per petri dish/pot/plant/M2/log10M2/density/severity score), percentage weed reduction/ratio (versus control/from original density).
Comparator: Appropriate controls: experimental units in which no intercrop was grown with the host crop, or monocrop/fallow/bare earth in the case of rotation studies.
Weed density and yield data were standardised to weeds/m−2 or t ha−1, respectively. Where reported, the long or short rainy season was also recorded. In the case of data presented in graph form, numeric data were extracted using data extraction software ‘im2graph’51. Data from studies were recorded to either intercrop or rotation cropping systems, as the mechanisms of impact of these on both parasitic weed density and yield are ecologically distinct.
Coordinates for study locations were directly extracted where available, or were estimated based on central coordinates of place names and extracted using Google maps52. In a handful of instances where it was not possible to determine separate coordinates for locations very close together (e.g. villages), data were aggregated and mean values calculated.
Studies in which there were no reported controls for the main treatment, or where data were not presented in a useable form were rejected. However, measures of variance were not reported in 53% of intercrop and 50% of rotation studies. Rejection of this proportion of studies due to missing variance risks the loss of significant volumes of data53. Furthermore, such omission can result in both losses of statistical power and errors in parameter estimates54 as well as a risk of bias toward studies that report significant results55. We, therefore, imputed missing variances as this has been shown to improve the reliability of meta-analysis53. Imputation was undertaken using the “mice” package in R using the predictive mean matching method56. This method was chosen as it selects values from the complete studies in the dataset predicted to be closest to values which are missing57. Other methods produced imputed values which were either not realistic or were negative (e.g. Random sample, Linear regression). Imputed values were estimated by averaging across ten iterations undertaken for each missing variance.
Climate and altitude
Climate data were obtained from the WorldClim2 dataset58. Climate variables recorded were mean annual rainfall, mean annual temperature and precipitation seasonality. Precipitation seasonality is defined as the coefficient of variation of mean monthly precipitation59. Altitudes for individual study sites were obtained from the SRTM 90m Digital Elevation Database v4.160,61 and were extracted using QGIS62.
Analyses were undertaken using linear mixed effect models weighted by the variance of effect sizes of each measure. Linear mixed effect models were used to test the overall effects of the cropping system on weed density and yield across studies. This was done by grouping the aggregated response data by cropping system.
Linear mixed-effect models were also used to identify the effect of management factors on weed density and yield across studies. Two groups of factors were included in these models with effect size as the response (weighted by the study variance), and study ID included as a random effect. Significance tests were conducted using Satterthwaites approximation in the R-package lmerTest63. This approach yields lower Type I error rates than alternatives64.
The effect size was estimated as Hedge’s g and its variance (standardised mean difference). This was done by calculating the difference between the treatment and control (weed density, weed dry weight or host yield) divided by the pooled standard deviation using the “compute.es” package in R65.
Linear models were used to determine the effect of rainfall CV, mean annual temperature, mean annual rainfall, and altitude on parasitic weed density and crop yield. These analyses were undertaken using a subset of studies where initial weed density was not deemed to have been manipulated (i.e.: farm, field trial or landscape). We also used local polynomial regression (LOESS) to visualize the effects of climatic variables.
A total of 3722 bibliographical references were retrieved using our search strategy. An initial assessment of the relevance of each study was made based on the title and abstract of each paper. This reduced the list to 83 original studies directly relating to the effect of either intercrops or rotation crops on parasitic weed density. After examining the full text of these papers, 67 were deemed to fulfil the inclusion criteria and provide all information needed. The remaining 16 were rejected as having either no experimental control or insufficient detail regarding the effects of response variables. The full list of studies included in the meta-analysis is included in Appendix 2.
The final dataset encompassed research across 24 countries and 89 localities (Fig. 1) and yielded 1525 individual data points. In terms of weed and crop diversity, it included 11 parasitic weed species, 70 varieties across 18 host crops and 115 intercrop rotation varieties across 105 trap crops (Appendix 3). Contingency tables for both intercrops and rotation crops are shown in Appendix 4.
The studies are predominantly located across sub-Saharan Africa, with a smaller number in North Africa and the Middle East, the Indian subcontinent and China and only three conducted in the United States and Europe. This distribution reflects the severity of the problem of parasitic weeds affecting annual crops across these regions, driving research efforts in search of solutions.
Our analysis reveals strong overall effects of both intercropping and crop rotation on weed density reduction and crop yields (Table 1). Consistent reductions in weed densities are associated with the use of intercrops across a diverse range of crops (Fig. 2A). Crop yields are also generally higher for intercrops (Fig. 2B), albeit for a smaller range of crop families. The use of multiple crops in the rotation has a consistently negative effect on weed density for a comparably large range of crops (Fig. 2C). Figure 2D indicates higher yields for rotation crops, in particular with respect to Pedaliaceae (sesame), Malvaceae (cotton) and Amaryllidaceae (garlic/onion). Indications of the relative size of these overall effects is provided in Table 2.
Analysis of effect sizes (Hedges g) indicated broadly similar mean effect sizes for both systems, with marginally greater weed reduction for rotation cropping and yield increase for intercropping (Fig. 3A,B). The number of crops used in rotation, denoted as diversity, did not have any significant effect on the percentage change in weed density for the mixed effect model (Table 1). Similarly, the mixed effect model for diversity did not show significant differences in weed reduction effect size between the numbers of rotation crops used (Table 2).
The linear mixed-effects models did not detect significant differences in effect sizes for the majority of factors (Table 2). This does not mean the rotation had no effect on the responses, but that effect sizes did not differ greatly enough between the factors. Mean effect sizes for both weed reduction and yield were in fact greater than 0.5 for over 75% of factors tested (Table 2).
Our models indicated that weed, crop and intercrop species, as well as intercrop variety, had significant effects on weed density effect sizes in intercropping systems. Weed and intercrop species also had a significant effect on yield effect sizes in intercropping systems (Fig. 4A,B). Mixed-effects models for crop rotation also indicated significant effect sizes for weed and host crop species and host crop variety. Notable effects on weed reduction included, inter alia, Desmodium and Stylosanthes in intercropping and maize, wheat and cotton in rotations. Mixed effect models for factors pertaining to yield in rotation systems did not indicate any individual significance for effect sizes.
In terms of climatic factors, rainfall seasonality (CV), mean annual rainfall and altitude were significant factors for both weed density and yield for intercropping systems (see Table 1 and Fig. 5A,B). For rotation cropping, rainfall seasonality, mean annual rainfall, mean temperature and altitude were significant factors in determining weed density. Mean annual rainfall and mean temperature were significant factors for yields.
A clear negative relationship is seen between log weed density and mean rainfall (Fig. 5A). In addition, an increase in rainfall variability is linked to increases in weed density up to an intermediate level, beyond which densities appear to drop off (Fig. 5B). There are two clear peaks in weed density around zero and 1250 m above sea level, relating generally to the distribution of Orobanche and Striga species respectively (Fig. 5C). Peak mean temperatures for parasitic weeds is around 20 °C, with very few recorded observations below approximately 17.5 °C. Many species have significant densities in excess of 20 °C however (Fig. 5D).
Egger’s tests for funnel plot asymmetry indicated a significant degree of heterogeneity within the effect sizes of the data set (random-effects model: p = < 0.0001, mixed-effects meta-regression model p = 0.0449). This indicates that the distribution of effect sizes for studies included in this meta-analysis differs sufficiently from that expected to suggest a bias in the reporting of results. The additional fail-safe N test undertaken indicated however that the impact of any potential bias within the data was low (Rosenberg significance Level = < 0.0001, fail-safe N: 311,129, Rosenthal significance Level = < 0.0001, fail-safe N: 447,309), Orwin fail-safe N = 1517.
Our results demonstrate that crop diversification has consistent effects in reducing parasitic weed density and increasing crop yield. Effects are significant for increases in both spatial (intercropping) and temporal (rotation cropping) crop diversification, though there are notable differences between the two systems. The linear mixed-effects models show the greater effect of weed suppression and yield for intercrops.
The significant effect of crop diversification on weed density is supported by several comparable meta-analyses. For example, in reductions of weed densities in general7,69, increased crop yields due to intercropping19 and improved yield stability17 noted that intercropping. Meta-analyses of agroforestry (which can also be considered a form of diversification) have also found reductions in parasitic and non-parasitic weeds70, and crop yield increases71.
A recent, meta-analysis of weed responses to crop diversification by Weisberger et al.7 found that weed reduction correlated with temporal diversity expressed as the variance of sowing dates between different crops. The metric of temporal crop diversification can encompass elements of intercropping (such as relay cropping) as well as rotation cropping. However, our results suggest that the effects of spatial diversification are stronger than temporal in suppressing parasitic weeds.
Our results further suggests that soil microclimate and host crop pre-attachment resistance effects may be stronger than effects more clearly attributable to rotation such as alteration of soil N2. Suicidal germination and allelopathy can occur within both intercropping and rotation cropping systems and could therefore be equally important mechanisms. Different combinations of crops and intercrops will produce different combinations of effects influencing weed density. Intercrops combining strong shading properties and favourably affect soil N2 show particularly strong effects in reducing parasitic weed density here, such as Crotalaria ochroleuca, Stylosanthes, Berseem clover and Lupin. Likewise, crops affording shade with allelopathic properties, antagonistic to parasitic weeds, such as Fenugreek72 have large effect sizes in both rotation and intercropping studies. Desmodium is effective in three ways, shading, enhancing N2 and stimulating suicidal germination by root exudates72,73, reflected by its’ significant effect size in this analysis.
Publication bias, in particular the potential over-reporting of significant results, can compromise the validity of the results of meta-analyses74. The Egger’s tests undertaken indicated a significant level of potential publication bias within the dataset, supported by the strong concurrence of results from a wide combination of systems, crops and weed species in terms of general trends. Although caution should be exercised in the inference of fail-safe N values, the results of the fail-safe N tests indicate that the data are sufficiently robust in terms of the impact of potential bias74.
Effect sizes for both weed reduction and yield were significant (i.e. nonzero) for all models, and greater than 0.5 for over 75% of factors tested. The most notable effects were those of host crop and host crop variety, intercrop, and to a lesser extent rotation crop. Caution should be exercised with a simplistic, interpretation of effect sizes in terms of small, medium and large in quantitative studies75. However, these results clearly show individual crops which perform better than others. The notable effects of crop variety on parasitic weed density support studies of individual parasitic weeds45,46,47,76,77. This effect also supports the rationale of a broader effort to identify and breed crop varieties resistant or tolerant to a wide range of parasitic weed pests4,78.
Our models did not detect significant differences in effect sizes for the majority of management factors. This does not indicate these factors should be discounted, but just that effect sizes did not differ greatly enough between the individual elements of these factors. The effects of management factors on yield may not be directly related to weed density, as there is no way to demonstrate the link in this analysis. Other factors are likely involved in influencing yields, as it is clearly understood that different crops, intercrops and crop varieties produce different yields independently of weed density.
The significant negative effect of precipitation on parasitic weeds is the most notable climatic effect revealed within the analysis undertaken here. There was also some evidence of a role for precipitation variation. The importance of rainfall and soil moisture is also shown across reviews of future weed distribution trends (e.g.79), niche modelling80,81,82 and landscape-scale studies of parasitic weeds77. Drier, warmer climates across many areas of Eurasia, South and North America, combined with more erratic rainfall patterns will favour the spread of many of the most problematic parasitic weeds such as Striga and Orobanche. This underlines the importance of monitoring and biosecurity measures to prevent or contain the introduction into currently uninfested agricultural zones.
This meta-analysis underlines the important role that temporal and spatial crop diversification has in the reduction of economically important parasitic weeds. This effect is consistent across a wide range of geographic locations, crops, varieties and weed species. There is also strong evidence of the positive effect of diversification on crop yield, although this may involve factors other than weed reduction. This analysis further serves to underline the viability of crop diversification as a tool to enhance global food security. This will become increasingly relevant given projections of the future proliferation of many parasitic weeds to areas currently not under infestation driven by globalisation and climate change.
The concentration of studies undertaken in sub-Saharan Africa indicates, however, that crop diversification is still largely focused on subsistence farmers in the developing world. While this is an entirely valid concentration of efforts, increased research should focus on the effects of diversification on the industrial production of staple crops within agroecological zones possessing Mediterranean climates globally. This will likely happen in response to the evolution of global patterns of weed distributions. However, proactive research strategies informed by predictive risk modelling could help in gaining the upper hand in the crop-weed “arms race”.
The datasets generated and/or analysed during the current study are available in the ORDA repository: https://figshare.shef.ac.uk/articles/software/R_Scripts_for_Crop_diversification_and_parasitic_weed_abundance_a_global_meta-analysis/20443896/1 and https://figshare.shef.ac.uk/articles/dataset/Data_files_for_Crop_diversification_and_parasitic_weed_abundance_a_global_meta-analysis/20443923/1.
Chauhan, B. S. Grand challenges in weed management. Front. Agron. https://doi.org/10.3389/fagro.2019.00003 (2020).
Oerke, E. C. Crop losses to pests. J. Agric. Sci. 144, 31–43 (2006).
Samejima, H. & Sugimoto, Y. Recent research progress in combatting root parasitic weeds. Biotechnol. Biotechnol. Equip. 32(2), 221–240 (2018).
Aly, R. Conventional and biotechnological approaches for control of parasitic weeds. In Vitro Cell. Dev. Biol. Plant 43(4), 304–317 (2007).
Fernández-Aparicio, M., Delavault, P. & Timko, M. P. Management of infection by parasitic weeds: A review. Plants 9(9), 1184 (2020).
Rodenburg, J., Demont, M., Zwart, S. J. & Bastiaans, L. Parasitic weed incidence and related economic losses in rice in Africa. Agric. Ecosyst. Environ. 235, 306–317 (2016).
Weisberger, D., Nichols, V. & Liebman, M. Does diversifying crop rotations suppress weeds? A meta-analysis. PLoS One 14(7), e0219847 (2019).
Ejeta, G. The Striga scourge in Africa: A growing pandemic. In Integrating New Technologies for Striga Control: Towards Ending the Witch-hunt 3–16 (World Scientific, 2007). https://doi.org/10.1142/9789812771506_0001.
Netting, R. M. & Stone, M. P. Agro-diversity on a farming frontier: Kofyar smallholders on the Benue plains of central Nigeria. Africa 66(1), 52–70 (1996).
Pimentel, D. et al. Conserving biological diversity in agricultural and forestry systems. Bioscience 42, 354–362 (1992).
Khoshbakht, K. & Hammer, K. How many plant species are cultivated?. Genet. Resour. Crop Evol. 55(7), 925–928. https://doi.org/10.1007/s10722-008 (2008).
Hajjar, R., Jarvis, D. I. & Gemmill-Herren, B. The utility of crop genetic diversity in maintaining ecosystem services. Agric. Ecosyst. Environ. 123(4), 261–270 (2008).
He, H. M. et al. Crop diversity and pest management in sustainable agriculture. J. Integr. Agric. 18(9), 1945–1952 (2019).
Ofori, F. & Stern, W. R. Cereal–legume intercropping systems. Adv. Agron. 41, 41–90 (1987).
Tanveer, M., Anjum, S. A., Hussain, S., Cerdà, A. & Ashraf, U. Relay cropping as a sustainable approach: Problems and opportunities for sustainable crop production. Environ. Sci. Pollut. Res. 24(8), 6973–6988 (2017).
Hartwig, N. L. & Ammon, H. U. Cover crops and living mulches. Weed Sci. 50(6), 688–699 (2002).
Raseduzzaman, M. D. & Jensen, E. S. Does intercropping enhance yield stability in arable crop production? A meta-analysis. Eur. J. Agron. 91, 25–33 (2017).
Davis, A. S., Hill, J. D., Chase, C. A., Johanns, A. M. & Liebman, M. Increasing cropping system diversity balances productivity, profitability and environmental health. PLoS One 7(10), e47149 (2012).
Himmelstein, J., Ares, A., Gallagher, D. & Myers, J. A meta-analysis of intercropping in Africa: Impacts on crop yield, farmer income, and integrated pest management effects. Int. J. Agric. Sustain. 15(1), 1–10 (2017).
Abson, D. J., Fraser, E. D. & Benton, T. G. Landscape diversity and the resilience of agricultural returns: A portfolio analysis of land-use patterns and economic returns from lowland agriculture. Agric. Food Secur. 2(1), 1–15 (2013).
Renard, D. & Tilman, D. National food production stabilized by crop diversity. Nature 571(7764), 257–260 (2019).
Gaudin, A. C. et al. Increasing crop diversity mitigates weather variations and improves yield stability. PLoS One 10(2), e0113261 (2015).
Bowles, T. M. et al. Long-term evidence shows that crop-rotation diversification increases agricultural resilience to adverse growing conditions in North America. One Earth 2(3), 284–293 (2020).
Chauhan, B. S., Singh, R. G. & Mahajan, G. Ecology and management of weeds under conservation agriculture: A review. Crop Prot. 38, 57–65 (2012).
Nichols, V., Verhulst, N., Cox, R. & Govaerts, B. Weed dynamics and conservation agriculture principles: A review. Field Crop Res. 183, 56–68 (2015).
Banik, P., Midya, A., Sarkar, B. K. & Ghose, S. S. Wheat and chickpea intercropping systems in an additive series experiment: Advantages and weed smothering. Eur. J. Agron. 24(4), 325–332 (2006).
Workayehu, T. & Wortmann, C. S. Maize–bean intercrop weed suppression and profitability in Southern Ethiopia. Agron. J. 103(4), 1058–1063 (2011).
Haugaard-Nielsen, H., Ambus, P. & Jensen, E. S. Interspecific competition, N use and interference with weeds in pea barley intercropping. Field Crop Res. 70, 101–109 (2001).
Jensen, E. S. Intercropping of Cereals and Grain Legumes for Increased Production, Weed Control, Improved Product Quality and Prevention of N-losses in European Organic Farming Systems, Final Report on Intercrop Project (QLK5-CT-2002-02352) (Risø National Laboratory, 2006).
Arlauskienė, A., Šarūnaitė, L., Kadžiulienė, Ž, Deveikytė, I. & Maikštėnienė, S. Suppression of annual weeds in pea and cereal intercrops. Agron. J. 106(5), 1765–1774 (2014).
Szumigalski, A. & van Acker, R. Weed suppression and crop production in annual intercrops. Weed Sci. 53(6), 813–825 (2005).
Stoltz, E. & Nadeau, E. Effects of intercropping on yield, weed incidence, forage quality and soil residual N in organically grown forage maize (Zea mays L.) and faba bean (Vicia faba L.). Field Crop Res. 169, 21–29 (2014).
Sauerborn, J., Müller-Stöver, D. & Hershenhorn, J. The role of biological control in managing parasitic weeds. Crop Prot. 26(3), 246–254 (2007).
Jamil, M., Rodenburg, J., Charnikhova, T. & Bouwmeester, H. J. Pre-attachment Striga hermonthica resistance of New Rice for Africa (NERICA) cultivars based on low strigolactone production. New Phytol. 192(4), 964–975. https://doi.org/10.1111/j.1469-8137.2011.03850.x (2011).
Yoneyama, K. et al. Nitrogen deficiency as well as phosphorus deficiency in sorghum promotes the production and exudation of 5-deoxystrigol, the host recognition signal for arbuscular mycorrhizal fungi and root parasites. Planta 227(1), 125–132. https://doi.org/10.1007/s00425-007-0600-5 (2007).
Sauerborn, J. Legumes used for weed control in agroecosystems in the tropics. Plant Res. Dev. 50, 74–82 (1999).
Ejeta, G. & Butler, L. G. Host-parasite interactions throughout the Striga life cycle, and their contributions to Striga resistance. Afr. Crop Sci. J. 1(2), 75–80. https://doi.org/10.4314/acsj.v1i2.69889 (1993).
Carsky, R. J., Singh, L. & Ndikawa, R. Suppression of Striga hermonthica on sorghum using a cowpea intercrop. Exp. Agric. 30(3), 349–358. https://doi.org/10.1017/s0014479700024467 (1994).
Hsiao, A. I., Worsham, A. D. & Moreland, D. E. Effects of temperature and dl-strigol on seed conditioning and germination of witchweed (Striga asiatica). Ann. Bot. 61(1), 65–72. https://doi.org/10.1093/oxfordjournals.aob.a087528 (1988).
Patterson, D. T. Effects of Environment on Growth and Reproduction of Witchweed and the Ecological Range of Witchweed (Monograph Series of the Weed Science Society of America, 1990).
Stewart, G. R. & Press, M. C. The physiology and biochemistry of parasitic angiosperms. Annu. Rev. Plant Biol. 41(1), 127–151. https://doi.org/10.1146/annurev.pp.41.060190.001015 (1990).
Anil, L., Park, R. H. P. & Miller, F. A. Temperate intercropping of cereals for forage: A review of the potential for growth and utilization with particular reference to the UK. Grass Forage Sci. 53, 301–317 (1998).
Mamolos, A. & Kalburtji, K. Significance of allelopathy in crop rotation. J. Crop Prod. 4, 197–218 (2001).
Khan, T. D., Chung, M. I., Xuan, T. D. & Tawata, S. The exploitation of crop allelopathy in sustainable agricultural production. J. Agron. Crop Sci. 191(3), 172–184 (2005).
Cissoko, M., Boisnard, A., Rodenburg, J., Press, M. C. & Scholes, J. D. New Rice for Africa (NERICA) cultivars exhibit different levels of post-attachment resistance against the parasitic weeds Striga hermonthica and Striga asiatica. New Phytol. 192(4), 952–963 (2011).
Rodenburg, J. et al. Do NERICA rice cultivars express resistance to Striga hermonthica (Del.) Benth. and Striga asiatica (L.) Kuntze under field conditions?. Field Crop Res. 170, 83–94 (2015).
Randrianjafizanaka, M. T., Autfray, P., Andrianaivo, A. P., Ramonta, I. R. & Rodenburg, J. Combined effects of cover crops, mulch, zero-tillage and resistant varieties on Striga asiatica (L.) Kuntze in rice-maize rotation systems. Agric. Ecosyst. Environ. 256, 23–33 (2018).
Rodenburg, J. et al. Genetic variation and host–parasite specificity of Striga resistance and tolerance in rice: The need for predictive breeding. New Phytol. 214(3), 1267–1280. https://doi.org/10.1111/nph.14451 (2017).
Nickrent, D. L. & Musselman, L. J. Introduction to parasitic flowering plants. Plant Health Instr. 13(6), 300–315 (2004).
Parker, C. Parasitic weeds: A world challenge. Weed Sci. 60(2), 269–276 (2012).
Shai Vaingast 2014. im2graph. Retrieved from: https://www.im2graph.co.il/free-downloads/windows-3264bit/ (2014).
Google Maps 2021. https://maps.google.com [Accessed February 2021–December 2022].
Kambach, S. et al. Consequences of multiple imputation of missing standard deviations and sample sizes in meta-analysis. Ecol. Evol. 10(20), 11699–11712 (2020).
Nakagawa, S. & Freckleton, R. P. Missing inaction: The dangers of ignoring missing data. Trends Ecol. Evol. 23(11), 592–596 (2008).
Idris, N. R. N. & Robertson, C. The effects of imputing the missing standard deviations on the standard error of meta analysis estimates. Commun. Stat. Simul. Comput. 38(3), 513–526. https://doi.org/10.1080/03610910802556106 (2009).
van Buuren, S. & Groothuis-Oudshoorn, K. mice: Multivariate imputation by chained equations in R. J. Stat. Softw. 45, 1–67 (2011).
van Buuren, S. Flexible Imputation of Missing Data (CRC Press, 2018).
Fick, S. E. & Hijmans, R. J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315. https://doi.org/10.1002/joc.5086 (2017).
O’Donnell, M. S. & Ignizio, D. A. Bioclimatic predictors for supporting ecological applications in the conterminous United States. US Geol. Surv. Data Ser. 691(10), 4–9 (2012).
Reuter, H. I., Nelson, A. & Jarvis, A. An evaluation of void filling interpolation methods for SRTM data. Int. J. Geogr. Inf. Sci. 21(9), 983–1008 (2007).
QGIS Development Team. QGIS Geographic Information System http://qgis.osgeo.org (Open Source Geospatial Foundation Project, 2020).
Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. lmerTest Package: Tests in linear mixed effects models. J. Stat. Softw. 82(13), 26. https://doi.org/10.18637/jss.v082.i13 (2017).
Song, C., Peacor, S. D., Osenberg, C. W. & Bence, J. R. An assessment of statistical methods for non-independent data in ecological meta-analyses. Ecology 101(12), e03184. https://doi.org/10.1002/ecy.3184 (2020).
Del Rey, A. C. compute.es: Compute Effect Sizes. R package version 0.2-2. https://cran.r-project.org/package=compute.es (2013).
R Core Team. R: A language and environment for statistical computing. http://www.R-project.org/ (R Foundation for Statistical Computing, 2020).
Wickham, H., Francois, R., Henry, L. & Müller, K. dplyr: A grammar of data manipulation. R package version 0.4. 3 (2015)
Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67(1), 48. https://doi.org/10.18637/jss.v067.i01 (2015).
Liebman, M. & Dyck, E. Crop rotation and intercropping strategies for weed management. Ecol. Appl. 3(1), 92–122 (1993).
Pumariño, L. et al. Effects of agroforestry on pest, disease and weed control: A meta-analysis. Basic Appl. Ecol. 16(7), 573–582 (2015).
Kuyah, S., Whitney, C. W., Jonsson, M., Sileshi, G. W., Öborn, I., Muthuri, C. W. & Luedeling, E. Agroforestry delivers a win-win solution for ecosystem services in sub-Saharan Africa. A meta-analysis (2019).
Evidente, A., Fernandez-Aparicio, M., Andolfi, A., Rubiales, D. & Motta, A. Trigoxazonane, a mono-substituted trioxazonane from Trigonella foenum-graecum root exudates, inhibits Orobanche crenata seed germination. Phytochemistry 68, 2487–2492 (2007).
Khan, Z. R. et al. Control of witchweed Striga hermonthica by intercropping with Desmodium spp., and the mechanism defined as allelopathic. J. Chem. Ecol. 28(9), 1871–1885 (2002).
Nakagawa, S. et al. Methods for testing publication bias in ecological and evolutionary meta-analyses. Methods Ecol. Evol. 13(1), 4–21 (2022).
Bakker, A. et al. Beyond small, medium, or large: Points of consideration when interpreting effect sizes. Educ. Stud. Math. 102(1), 1–8 (2019).
Scott, D. et al. Mapping the drivers of parasitic weed abundance at a national scale: A new approach applied to Striga asiatica in the mid-west of Madagascar. Weed Res. 60(5), 323–333 (2020).
Scott, D. et al. Identifying existing management practices in the control of Striga asiatica within rice–maize systems in mid-west Madagascar. Ecol. Evol. 11(19), 13579–13592 (2021).
Rubiales, D. & Fernández-Aparicio, M. Innovations in parasitic weeds management in legume crops. A review. Agron. Sustain. Dev. 32(2), 433–449 (2012).
Bir, M. S. H. et al. Weed population dynamics under climatic change. Weed Turfgrass Sci. 3(3), 174–182 (2014).
Mohamed, K. I., Bolin, J. F., Musselman, L. J. & Townsend, P. A. Genetic diversity of Striga and implications for control and modelling future distributions. In Integrating New Technologies for Striga Control—Towards Ending the Witch-Hunt (eds Ejeta, G. & Gressel, J.) 71–84 (World Scientific, 2007).
Mandumbu, R., Mutengwa, C. S., Mabasa, S. & Mwenje, E. Predictions of the Striga scourge under new climate in southern Africa. J. Biol. Sci. 17, 192–201. https://doi.org/10.3923/jbs.2017.194.201 (2017).
Mudereri, B. T. et al. Multi-source spatial data-based invasion risk modelling of Striga (Striga asiatica) in Zimbabwe. GIScience Remote Sens. 57(4), 553–571. https://doi.org/10.1080/15481603.2020.1744250 (2020).
This study was funded by ACCE a doctoral training partnership funded by the Natural Environment Research Council.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Scott, D., Freckleton, R.P. Crop diversification and parasitic weed abundance: a global meta-analysis. Sci Rep 12, 19413 (2022). https://doi.org/10.1038/s41598-022-24047-2