REVIEWARTICLEShallow non-inversion tillage in organic farming maintains cropyields and increases soil C stocks: a meta-analysisJulia Cooper1 & Marcin Baranski1 & Gavin Stewart1 & Majimcha Nobel-de Lange1 &Paolo Bàrberi2 & Andreas Fließbach3 & Josephine Peigné4 & Alfred Berner3 &Christopher Brock5 & Marion Casagrande4 & Oliver Crowley6,7 & Christophe David4 &Alex De Vliegher8 & Thomas F. Döring6,9 & Aurélien Dupont10 & Martin Entz11 &Meike Grosse12 & Thorsten Haase13 & Caroline Halde11,14 & Verena Hammerl15 &Hilfred Huiting16 & Günter Leithold5 & Monika Messmer3 & Michael Schloter15 &Wijnand Sukkel16 & Marcel G. A. van der Heijden17 & Koen Willekens8 &Raphaël Wittwer17 & Paul Mäder3Accepted: 27 January 2016# The Author(s) 2016. This article is published with open access at Springerlink.comAbstract Reduced tillage is increasingly promoted to im-prove sustainability and productivity of agricultural systems.Nonetheless, adoption of reduced tillage by organic farmershas been slow due to concerns about nutrient supply, soilstructure, and weeds that may limit yields. Here, we compiledthe results from both published and unpublished researchcomparing deep or shallow inversion tillage, with various cat-egories of reduced tillage under organic management.Shallow refers to less than 25 cm. We found that (1) divisionof reduced tillage practices into different classes with varyingdegrees of intensity allowed us to assess the trade-offs be-tween reductions in tillage intensity, crop yields, weed inci-dence, and soil C stocks. (2) Reducing tillage intensity inorganic systems reduced crop yields by an average of 7.6 %relative to deep inversion tillage with no significant reductionin yield relative to shallow inversion tillage. (3) Among thedifferent classes of reduced tillage practice, shallow non-inversion tillage resulted in non-significant reductions in yield* Julia CooperJulia.Cooper@ncl.ac.uk1 School of Agriculture, Food and Rural Development, NewcastleUniversity, Kings Road, Newcastle NE1 7RU, UK2 Scuola Superiore Sant’Anna, Institute of Life Sciences, PiazzaMartiri della Libertà 33, 56127 Pisa, Italy3 Research Institute of Organic Agriculture (FiBL), Ackerstrasse,CH-5070 Frick, Switzerland4 ISARA-Lyon, Université de Lyon, 23 rue Jean Baldassini,69007 Lyon, France5 Organic Farming, Justus Liebig University Giessen,Karl-Glöckner-Str. 21c, 35394 Giessen, Germany6 Organic Research Centre (ORC), Elm Farm,Newbury, Berkshire RG20 0HR, UK7 Centre for Agri-Environmental Research, School of Agriculture,Policy and Development, The University of Reading Earley Gate,Reading RG6 6AR, UK8 Plant Sciences Unit, Crop Husbandry and Environment, Institute forAgricultural and Fisheries Research (ILVO), Merelbeke, Belgium9 Department of Agronomy and Crop Science, Humboldt UniversityBerlin, Albrecht-Thaer-Weg 5, 14195 Berlin, Germany10 Agriculture and Territories, Chambres d’agriculture de Bretagne, 4Avenue du Chalutier Sans Pitié, BP 10540, 22195 PlerinCedex, France11 Department of Plant Science, University of Manitoba,Winnipeg, Canada R3T 2N212 University of Kassel, Department of Organic Farming and Cropping,Nordbahnhofstr. 1a, 37213 Witzenhausen, Germany13 Landesbetrieb Landwirtschaft Hessen, Beratungsteam ÖkologischerLandbau, Kölnische Str. 48-50, 34117 Kassel, Germany14 Département de Phytologie, Université Laval, Québec City, CanadaG1V 0A615 Helmholtz Zentrum München (HMGU), Research UnitEnvironmental Genomics, Ingolstädter Landstr. 1,85764 Neuherberg, Germany16 Applied Plant Research (WUR-APR), Wageningen University andResearch Centre, Lelystad, The Netherlands17 Plant-Soil Interactions, Institute for Sustainability Sciences,Agroscope, Reckenholzstrasse 191, CH-8046 Zürich, SwitzerlandAgron. Sustain. Dev. (2016) 36:22 DOI 10.1007/s13593-016-0354-1relative to deep inversion; whereas deep non-inversion tillageresulted in the largest yield reduction, of 11.6 %. (4) Usinginversion tillage to only a shallow depth resulted in minimalreductions in yield, of 5.5 %, but significantly higher soil Cstocks and better weed control. This finding suggests that thisis a good option for organic farmers wanting to improve soilquality while minimizing impacts on yields. (5) Weeds wereconsistently higher, by about 50 %, when tillage intensity wasreduced, although this did not always result in reduced yields.Keywords No-till . Organic farming . Conservation tillage .Conservation agriculture . Meta-analysis . Crop yield .Weeds . Soil C . Reduced tillage .Minimum tillage1 Introduction1.1 Current status of reduced tillage practicesin the organic sectorReduced tillage intensity is one of the key components ofconservation agriculture systems promoted by the Food andAgriculture Association of the United Nations to conserve,improve, and make more efficient use of natural resources(Food and Agriculture Organization of the United Nations2015). The other two essential components of conservationagriculture are maximum soil cover and diversified crop rota-tions. Conservation agriculture practices not only reduce soildegradation but also contribute to sustained agricultural pro-duction, particularly in areas where soils are fragile and at riskof declining quality (Hobbs et al. 2008).The use of diverse crop rotations and mulching are funda-mental concepts within organic agriculture (Lampkin andMeasures 2001); however, implementation of reduced tillagepractices is less commonly accepted. Bàrberi (2006) explainsthat tillage is important in organic agriculture for a number ofreasons including incorporation of organic residues into thesoil to facilitate more rapid mineralization and release of nu-trients to the crop. Biomass incorporation into the soil mayalso reduce some soilborne pest and pathogen loads (Liebmanand Davis 2000). One of the most important roles of tillage inorganic systems is for the control of weed populations (Peignéet al. 2007). Although weed control without herbicides is pos-sible, there are challenges when combining organic practiceswith reductions in tillage intensity, and frequently crop yieldsare compromised.The real and perceived challenges of reducing or eliminat-ing tillage within organic systems has led to slower adoptionof reduced and no-till in the organic community in Europecompared with the conventional sector (Mäder and Berner2012). There are no precise figures available on numbers oforganic farmers practicing conservation agriculture in Europe;however, numbers are expected to be very low. A survey inGermany found that only 6 % of the sample group (367 arableorganic farmers) tilled the soil without a plough and that 22 %were using shallow (less than 15 cm) tillage (Fig. 1; Wilhelmet al. 2011). Furthermore, a recent targeted survey in Europeidentified organic farmers practicing components of conserva-tion agriculture and indicated a broad diversity of reducedtillage practices among this group (Peigné et al. 2015).Eighty-nine percent of surveyed organic farmers practicingconservation agriculture used some form of reduced tillagedefined as any tillage shallower than the standard convention-al ploughing practice and/or a non-inversion method, but only27 % of the organic farmers practicing conservation agricul-ture used no-till. The survey highlighted the challenges inweed management with reduced tillage, and also the needfor deep tillage at some phases of the rotation to incorporategreen manures and ley crops.As highlighted by Mäder and Berner (2012), since the1990s, several important experiments were established inEurope to investigate the use of reduced tillage systems underorganic practices. In addition, results from some studies in theMediterranean region (e.g., Bilalis et al. 2012b) and NorthAmerica (e.g., Luna et al. 2012) have recently been published.In this study, we compiled results of published data, as well asraw data obtained from collaborators in Europe and NorthAmerica, to allow a comprehensive meta-analysis of experi-mental data on the use of reduced tillage in organic farmingsystems.1.2 Study objectivesThe study was undertaken with the overall aim to identifyoptimal management practices for successful implementationof reduced tillage in organic farming systems. Note that “suc-cessful implementation” implies maximum yields, minimumweed incidence, and maximum SOC stocks.The following specific questions were addressed in themeta-analysis:1. What is the magnitude of the effect of reduced tillageintensity on crop yields in organic systems?2. Is this effect consistent across all environments (soil typesand climatic zones)?3. Are there certain management practices that can be usedto enhance production under reduced tillage intensity inorganic systems, i.e., crop rotation, crop choice (currentand previous year), use of mechanical weeding?4. Are the impacts of reduced tillage on yield in organicsystems stable over time?5. Is it really weed incidence that is causing yield reduc-tions? Or could there be other factors?6. Does using reduced tillage intensity in organic systemsincrease soil organic C above the levels already achievedby organic practice? 22 Page 2 of 20 Agron. Sustain. Dev. (2016) 36:22 2 Reduced tillage in organic systems2.1 Impact on crop yieldsReducing tillage intensity in conventional farming systemscan impact crop yields although the direction and magnitudeof this effect varies with climate, crop rotation, and soil type(Soane et al. 2012). A 2010 meta-regression study found anoverall yield reduction across conventional cropping systemsusing conservation tillage in Europe of 4.5 % (Van den Putteet al. 2010). More recently, Pittelkow et al. (2014) concludedthat implementing no-till practices without the other two cor-nerstones of conservation agriculture (residue retention andcrop rotation) resulted in yield reductions across all climatesof ~10 %, although gains in yield of almost 10 % wereachieved when no-till was combined with residue retentionand crop rotation in dry climates. Implementing reduced till-age methods in organic farming may be more challenging dueto issues with delayed and limited mineralization of nutrientsfrom organic matter which cannot be addressed by inputs ofsynthetic fertilizers, as well as increased pressure from weeds(Peigné et al. 2007).It is particularly important that the yield implications ofreducing tillage in organic systems are quantified, since cropyields under organic management for many crops may alreadybe significantly lower than conventional. Meta-analyses bySeufert et al. (2012) and Ponisio et al. (2015) indicated thatyields of organic cereals were from 19 to 26 % lower thanconventional. However, these authors highlighted the poten-tial to minimize this yield gap by using multicropping anddiverse rotations. It is essential that the implementation ofreduced tillage in organic systems does not further limitFig. 1 Implements commonlyused by organic farmerspracticing conservationagriculture in Europe: aResearchers at the ResearchInstitute of Organic Agriculture inSwitzerland examining a chiselplough with stern (star) roller. bAroller crimper used for destroyingcover crops and leys in organicno-till systemsAgron. Sustain. Dev. (2016) 36:22 Page 3 of 20 22 yields; therefore, environments and management practicesthat allow organic yields to be maintained or increased underreduced tillage need to be identified.2.2 Impact on weedsReduced tillage concentrates weed seeds in surface soil layersand creates increased opportunities for germination and emer-gence, leading to increases in weed abundance (Légère et al.2011). As a consequence, direct weed control methods such asmechanical weeding in organic systems, need to be particular-ly effective to prevent a progressive buildup of the weedseedbank and increased weed infestation in the long term(Melander et al. 2013). This problem is exacerbated in organicfarming, where use of synthetic herbicides is forbidden.As an alternative to herbicides, organic farmers rely onmechanical methods as well as well-planned and diversifiedcrop rotations to keep weeds under reasonable control(Amossé et al. 2013). Diversification of the crop rotationand appropriate timing of management interventions createsa sequence of ecological disturbances which impedes (i) com-pletion of the life cycle for the vast majority of weed speciesand (ii) selection of “crop mimics,” i.e., weed species welladapted to the prevailing disturbance regime (Bàrberi 2002).Limiting tillage in organic reduced tillage systems makesweed control still more challenging. Researchers at theRodale Institute in Pennsylvania and the USDA inBeltsville, MD, have been addressing this problem by devel-oping systems for organic farmers that combine reduced till-age with weed suppressive cover crops, mulches, or associat-ed crops. These systems can be effective at limiting annualweed growth in the subsequent crop; however, control of pe-rennial weeds remains a challenge (Mirsky et al. 2012).Researchers in the American Midwest are also developingsystems that use weed suppressing cover crops which aredestroyed in a timely fashion by a roller crimper or sicklebar mower; in some cases, a preceding cover crop of rye canresult in almost complete elimination of weeds in the subse-quent crop (Silva 2014).The problem of perennial weeds in organic no-till systemsmay result from a dramatic shift in weed community compo-sition so that relative abundance of grasses, wind-disseminated annuals, biennials, and perennials becomes pro-gressively higher in a short time (Bàrberi and Lo Cascio2001). The lack of severe soil disturbance allows the second-ary succession of in-field vegetation to go beyond the initialstage characterised by annuals (Zanin et al. 1997). This wasobserved by Armengot et al. (2015) who reported a trendtoward higher populations of perennial weeds over time in along-term organic reduced tillage experiment. These ongoingproblems have led to the adoption of hybrid systems whereno-till is used for selected annual crops in the cropping se-quence, e.g., spring crops like soya bean and corn, withreduced tillage during other crop phases across a rotation(Carr et al. 2013).2.3 Impact on soil C stocksThe increase in surface soil C concentrations under no-till hasbeen well documented (West and Post 2002); however, evi-dence for increases in the quantity of C stored (C stocks) insoils under no-till has been less consistent. Depth of samplingis crucial with studies where sampling is done to a shallowerdepth more likely to detect an increase in stocks under no-till(e.g., Angers and Eriksen-Hamel 2008), while deeper sam-pling usually results in few differences compared to full inver-sion tillage (Baker et al. 2007). Soil C can become stratifiedunder no-till with concentrations higher than conventional till-age in the top soil layer, but lower concentrations at deeperdepths, resulting in no net difference in stocks (Luo et al.2010).Organic farming practices have also been shown to in-crease soil C concentrations (Gattinger et al. 2012) which isattributed to the higher rates of C inputs in organic systemsfrom ley crops and manure/compost inputs. It is thereforepossible that merging both farming approaches into organicreduced tillage systems could result in further enhancement ofsurface soil C concentrations and potentially soil C stocks aswell. There have been few studies on organic reduced tillagesystems, but Mäder and Berner (2012) showed that the use ofreduced tillage in organic farming systems compared to con-ventional tillage further enhanced soil quality indicators suchas organic carbon, microbial activity, and soil structure in theuppermost soil layer. In general, greatest differences were ob-served when comparing conventional plough treatments (fullinversion) with shallow non-inverting cultivation. Enhancedsoil microbial activity within organic reduced or no-till sys-tems is particularly important because it can contribute to im-proved nutrient uptake by arbuscular mycorrhizal fungi (Köhlet al. 2014) and a more efficient cycling of nutrients (Benderand Van Der Heijden 2014). This enhancement of soil biolog-ical activity under organic reduced tillage systems may be onestrategy to improve crop nutrient supply and ultimately yields,within organic systems.3 Material and methodsData used in the analysis was sourced from raw field trial datafrom experiments where tillage intensity was reduced underorganic management. In addition, published data from re-duced intensity tillage trials in organic systems from bothrefereed and non-refereed sources was used. In all cases, datawas double checked to ensure that raw data used in the anal-ysis was not duplicating published results. Where both raw 22 Page 4 of 20 Agron. Sustain. Dev. (2016) 36:22 and published results from the same experiment existed, theraw data was used in the analysis.3.1 Sourcing and compilation of raw field trial dataPreviously unpublished data from a total of 15 field trials wasprovided by partners and associates of the TILMAN-ORGproject (www.tilman-org.net) (Table 1). Trial managers wereprovided with a spreadsheet for entering the details of the fieldtrials including site information, experimental design, trialmanagement, and annual results for key response variables.Values for response variables were provided as means for eachtreatment and information on the number of replications (N)and variability of the mean (standard error or standard devia-tion) were recorded. Data on crop yields, soil C, and weedswere compiled. Yield data was compiled as marketable yields,or total yields when marketable yield data was not available.Soil C values were recorded as stocks in g m−2. Where soil Cwas provided as a concentration and bulk densities wereavailable, stocks were calculated by multiplying the Cconcentration by the bulk density and sampling depth. If soilC was only provided as a concentration, stocks werecalculated using the pedotransfer function for bulk densityreported by Gattinger et al. (2012) and the value was convert-ed to a stock as described above. C stock comparisons be-tween each paired control and experimental treatment werealways based on the same sampling depth. Sampling depthsranged from 0–15 to 0–35 cm. If stocks or concentrations formore than one layer were reported (e.g., 0–15 and 15–30 cm),then the stocks for each layer were calculated and summed to auniform depth.Weed data were also compiled. Each study did not use thesame measure of weed pressure with three different possiblemeasurement units identified: biomass (g m−2), cover (%), ordensity (# m−2). We selected data based on just one of theseunits from a given experiment, using biomass where availableas the best measure of weed pressure, followed by density andcover if biomass data was not available. Initially, weed datawere divided into four different plant types: annual monocots,annual dicots, perennial monocots, and perennial dicots. Toallow comparison of the weed data across experiments, valuesfor each of the different plant types for a given measurementunit were summed to create a new response variable: weedincidence.For crop yields and weed data the effect size was cal-culated as the ratio between the experimental and controltreatments. This resulted in an effect size that was stan-dardized and unitless; therefore, it was not necessary tohave the same units of measurement when combining datafrom different experiments. This allowed inclusion of datareported using any of the three weed measurement unitsused in the experimental studies. For the soil C data, sinceall measurements were in the same units (g m−2) the effectsize was calculated as the mean difference. This allowedpresentation of the effect in actual units of mass per area.Data from all the experimental years available were includ-ed in the dataset.3.2 Published data sourcing and selectionA literature survey of peer-reviewed published literaturein the ISI-Web of Science and CAB Abstracts (Ovid) be-tween 1910 and 2013 was conducted to identify papersreporting results from studies using reduced tillage in or-ganic farming. The following search terms and their var-iations were used in various combinations: reduced/mini-mum/shallow/chisel, tillage/ploughing/plowing, intensity,no-till/direct seeding, organic/ecological, farming/sys-tems/agriculture/management. The initial search providedaround 180 papers published from 1986 to 2013 in scien-tific journals from the Web of Science. More relevantpapers were found by searching through the reference listsof papers already selected for the meta-analysis and rec-ommendations of experts in tillage research. We also ex-tracted data from non-peer reviewed sources for the anal-ysis. Field trials that were possibly relevant for the meta-analysis were found via webpages (http://www.orgprints.org/) dedicated to research in organic agriculture as wellas recommendations of tillage researchers. These sourceswere provided by the study authors who submitted thesesand reports, often in their native language, for inclusion inthe analysis.All studies were scrutinized and only included if theymet the following selection criteria: (i) experiment underorganic management for at least 3 years prior to the dateof response measurement; (ii) at least two levels of tillageintensity included as a treatment; (iii) no “mixing” oftreatments, i.e., only tillage varied between experimentaltreatments; and (iv) included climatic zones found inEurope (Table 2). This screening process resulted in 26published studies being identified (20 peer reviewed and 6non-peer reviewed; see Table 1).Details on the field trials identified from published sources,and the values of the response variables were entered into thesame database used to compile the raw data from trials asexplained in Sect. 3.1.3.3 Characterization of the dataWe assigned the tillage treatment factor in each experi-ment to a class based on the level of tillage intensity.The six tillage classes in order of decreasing intensitywere as follows: deep inversion (greater than or equal to25 cm depth), double-layer ploughing (inversion of thesoil to a depth of ~15 cm and loosening to ~30 cm(Gruber and Claupein 2009; Vakali et al. 2011), shallowAgron. Sustain. Dev. (2016) 36:22 Page 5 of 20 22 Table1Detailsoffieldexperimentswherepublished(authoryearcitation)andunpublished(TILMAN-ORG)datawassourcedfortheanalysisandnumbersofyearsdataincludedintheanalysisfromeachtrialCitationTrialname,institutionCountryaClimatezonebSoiltypeLatitude,longitudeTillagetreatmentsStartcExp.yearsdResponsevariablesNeLewisetal.2011RussellE.LarsonAgriculturalResearchCenterUSADfbSilt loam40°42′43.2″N,77°56′38.4″Wshallowinversion(<25cm),non-inversion(10–25cm)20032006SoilC4Vaismanetal.2011IanN.MorrisonResearchFarmandcertifiedorganicfarmnearOxbow,UniversityofManitobaCanadaDwbSandyloam,loam49°29′N,98°0′Wand49°13′N,102°10′WNon-inversion(<10cm),no-till20072008–2009Cropyield,weeds4ShirtliffeandJohnson2012KernenResearchFarm,UniversityofSaskatchewanCanadaDfbClay loam52°09′N,106°33′WDeepinversion(≥25cm),no-tillNdf2009Cropyield,weeds4TILMAN-ORGCarman,UniversityofManitobaCanadaDwbSandyloam49°29′53.20″N,98°01′47.10″WNon-inversion(<10cm),no-till20082009–2012Cropyield4TILMAN-ORGFarmingSystemandTillageexperimentAgroscope(FAST)I/II,AgroscopeSwitzerlandDfbLoam47°26′20″N,8°31′40″EShallowinversion(<25cm),non-inversion(<10cm)2009/20102010–2012Cropyield,weeds4Delateetal.2010Neely-KinyonFarm,IowaStateUniversityUSADfaSilty clayloamndNon-inversion(10–25cm),no-tillnd2010CropyieldndDelateetal.2011Neely-KinyonFarm,IowaStateUniversityUSADfaSilty clayloamndNon-inversion(10–25cm),no-tillnd2011SoilC,weedsndTILMAN-ORGScheyern,HelmholtzZentrumMünchen(HMGU)GermanyCfbLoam48°29′51″N,11°26′39″EDeepinversion(≥25cm),non-inversion(10–25cm),non-inversion(<10cm)19921993–2012Cropyield2Kainzetal.2005Scheyern,TechnischeUniversitätMünchen(TUM)GermanyCfbSiltloam48°29′51″N,11°26′39″EDeepinversion(≥25cm),shallowinversion(<25cm)19921995–2004Cropyield,soilCndKochandGaberle2010ndGermanyCfbSiltloamndDeepinversion(≥25cm),double-layerploughing,shallowinversion(<25cm),non-inversion(10–25cm)19941995–2008Cropyield,soilCndEmmerling2007Eichenhof,UniversityofTrierGermanyCfbSilty clayloamndDeepinversion(≥25cm),double-layerploughing,non-inversion(10–25cm)19951997Cropyield,soilC2TILMAN-ORGOrganicArableFarmingExperimentGladbacherhof(OAFEG),JustusLiebigUniversityGiessenGermanyCfbSilty clayloam50°24′N,8°15′EDeepinversion(≥25cm),shallowinversion(<25cm),non-inversion(10–25cm)19981998–2009Cropyield,soilC,weeds4Vakalietal.2011EcologicalSoilManagementproject,StiftungOekologieundLandbau(SOEL)GermanyCfbSilty clayloam49°49′23″N,8°5′24″EDeepinversion(≥25cm),double-layerploughing,non-inversion(10–25cm)nd1999–2001Cropyield,weeds2PaffrathandStumm2010ndGermanyCfbSandyloamndDeepinversion(≥25cm),non-inversion(10–25cm)19982000–2006Cropyield,weedsnd 22 Page 6 of 20 Agron. Sustain. Dev. (2016) 36:22 Table1(continued)CitationTrialname,institutionCountryaClimatezonebSoiltypeLatitude,longitudeTillagetreatmentsStartcExp.yearsdResponsevariablesNeDittmannandZimmer2010ndGermanyCfbLoamysandndShallowinversion(<25cm),non-inversion(10–25cm)19982001–2002Cropyield,soilC4Reinickeetal.2010ndGermanyCfbSiltloamndDeepinversion(≥25cm),double-layerploughing19982003Cropyield,weeds4TILMAN-ORGFrickTillagetrial,ResearchInstituteofOrganicAgriculture(FiBL)SwitzerlandDfbClay47°30′N,8°01′EShallowinversion(<25cm),non-inversion(<10cm)20032003–2009Cropyield,weeds8TILMAN-ORGKerguehennec(CRABretagne)FranceCfbLoam47°53′10.39″N,2°44′4.44″EDeepinversion(≥25cm),shallowinversion(<25cm),non-inversion(10–25cm)20032003–2007,2010–2011Cropyield,soilC,weeds3TILMAN-ORGG10,InstituteforAgriculturalandFisheriesResearch(ILVO)BelgiumCfbSandyloam50°59′12″N,3°47′11″EDeepinversion(≥25cm),non-inversion(10–25cm)20052005–2008Cropyield,soilC4TILMAN-ORGThil,ISARA-LyonFranceCfbSandyloam45°49′9.44″N,5°2′2.62″EDeepinversion(≥25cm),shallowinversion(<25cm),non-inversion(10–25cm),non-inversion(<10cm)20052005–2011Cropyield,soilC,weeds3GruberandClaupein2009KleinhohenheimandGoldenerAcker,UniversityofHohenheimGermanyCfbSandyclayloamndDeepinversion(≥25cm),double-layerploughing,shallowinversion(<25cm),non-inversion(10–25cm)19992005–2007Cropyield,weedsndTILMAN-ORGG9,InstituteforAgriculturalandFisheriesResearch(ILVO)BelgiumCfbSandyloam50°59′12″N,3°47′11″EDeepinversion(≥25cm),non-inversion(10–25cm)20062006–2009Cropyield,soilC4Lehocká,2009ResearchInstituteofPlantProductionSlovakiaCfbLoamndShallowinversion(<25cm),non-inversion(10–25cm)20062007SoilC3TILMAN-ORGFrankenhausen1,UniversityofKasselGermanyCfbSiltloam51°24′30.36″N,9°27′13.36″EDeepinversion(≥25cm),shallowinversion(<25cm)20072008–2011Cropyield,weeds4TILMAN-ORGBroekemahoeve-BASIS,WageningenUniversityandResearchCentre,AppliedPlantResearch(WUR-APR)The Netherl-andsCfbSandyloam52°16′12.0″N,5°18′36.0″EDeepinversion(≥25cm),non-inversion(10–25cm),non-inversion(<10cm)20092009–2012Cropyield,weeds4Wangetal.2011MountainHorticulturalCropsResearchStationUSACfbSandyloam35°25′39″N,82°33′21″WDeepinversion(≥25cm),no-till19942009SoilC4TILMAN-ORGMuriTillagetrial,ResearchInstituteofOrganicAgriculture(FiBL)SwitzerlandDfbLoam47°17′N,8°19′EDeepinversion(≥25cm),non-inversion(<10cm)20092009–2011Cropyield4,8TILMAN-ORGDuchyHomeFarmEcodynTrial,OrganicResearchCentre(ORC)UKCfbSiltyclayloam51°39′15″N,2°09′39″WShallowinversion(<25cm),non-inversion(<10cm)20102010–2012Cropyield,weeds3TILMAN-ORGAeschTillagetrial,ResearchInstituteSwitzerlandDfbLoam47°28′N,7°34′EDeepinversion(≥25cm),non-inversion(<10cm)20102010–20124Agron. Sustain. Dev. (2016) 36:22 Page 7 of 20 22 Table1(continued)CitationTrialname,institutionCountryaClimatezonebSoiltypeLatitude,longitudeTillagetreatmentsStartcExp.yearsdResponsevariablesNeofOrganicAgriculture(FiBL)Cropyield,soilC,weedsTILMAN-ORGNaffertonFactorialSystemComparision(NFSC),NewcastleUniversityUKCfbSandyloam54°59′27.39″N,1°53′53.78″WDeepinversion(≥25cm),non-inversion(<10cm)20122012Cropyield,weeds4Díaz-Pérezetal.2008UniversityofGeorgiaUSACfaSandyloamndDeepinversion(≥25cm),no-till20042005Cropyield,weeds4Teasdaleetal.2012NorthFarm,USDA-ARSBeltsvilleAgriculturalResearchCenterUSCfaSiltloam39°02′05″N,76°54′28″WNon-inversion(10–25cm),no-till20082008–2010Cropyield,weeds4Luna2003VegetableCropsCorvallisFarm,OregonStateUniversityUSACsbSiltloamndDeepinversion(≥25cm),non-inversion(10–25cm)19981999Cropyield,weeds4Maddenetal.2004ndUSACsaClay loamndDeepinversion(≥25cm),no-till19992000–2001Cropyield,weedsndDiTizioetal.2008Experimentalfarm(Viterbo),UniversityofTusciaItalyCsaclayloamNdDeepinversion(≥25cm),shallowinversion(<25cm)20012006SoilC3BilalisandKaramanos2010ndGreeceCsaSiltloamndShallowinversion(<25cm),no-till20062006–2007Cropyield,soilC4Bilalisetal.2011AgriculturalUniversityofAthensGreeceCsaClay loam34°58′N,23°43′Eand38°35′N,21°25′EShallowinversion(<25cm),non-inversion(10–25cm),no-till20062008–2009Cropyield4Bilalisetal.2010AgriculturalUniversityofAthensGreeceCsaClay loam34°58′N,23°43′EShallowinversion(<25cm),non-inversion(10–25cm),no-till20062009–2010Cropyield,soilC4Bilalisetal.2012aAgriculturalUniversityofAthensGreeceCsaClay loam34°58′N,23°43′EShallowinversion(<25cm),non-inversion(10–25cm),no-till20062009–2010Weeds4Bilalisetal.2012bAgriculturalUniversityofAthensGreeceCsaClay loam38°35′N,21°25′EShallowinversion(<25cm),non-inversion(10–25cm)20062010–2011Cropyield4Collinsetal.2011WashingtonStateUniversityUSACsbSandyloamndNon-inversion(10–25cm),no-till20102010Cropyield,weedsndaAcronymsaccordingISO3166–1bKoppenclimateclasses,seeTable2forfullnamescYearinwhichtillageexperimentbeginsdYearfromwhichdatawasincludedinthemeta-analysiseNumberoffieldreplicationsfNodata 22 Page 8 of 20 Agron. Sustain. Dev. (2016) 36:22 inversion (less than 25 cm depth), non-inversion (10–25cm depth), non-inversion (less than 10 cm depth), and no-till (Fig. 2). We also collected information from the fieldtrials and selected published data to further describe theenvironmental and management factors for each study.Environmental variables included as factors were soiltype and climate defined in nine climatic zones asdescribed in Quemada et al. (2013) (Table 2). Soil typeswere grouped into three categories according to USDAtexture classes (Soil Survey Division Staff 1993): heavy,which included all soils with a clay content greater than40 % (sandy clay, silty clay, clay); light, which includedall soils with less than 40 % clay and greater than 50 %sand (sand, loamy sand, sandy loam, sandy clay loam);and loamy, which included all other soils (silt, silty loam,medium loam, clay loam, silty clay loam). Managementfactors included in the analysis were crop rotation class,mechanical weeding class, and main crop and previouscrop class (Table 2). Crop rotations were assigned toone of four types: intensive arable (i.e., no ley crops);arable with ley periods; intensive horticulture (i.e., noley crops); horticulture with ley periods. For the purposesof this study, we defined a ley period as a full season of asoil building crop such as grass/clover. Crops grown forsoil building purposes for less than a year were consid-ered green manures. Main crops and previous crops weredivided into three classes: combinable crops (e.g., winterand spring grains, maize, legumes); non-combinable crops(e.g., leaf vegetables and root vegetables including pota-toes); cover crops/leys (as described above).3.4 Data analysisMany of the published studies did not provide an estimate ofvariance which meant it was not possible to use a weightedmeta-analysis approach for this analysis; therefore, results forunweighted analyses only are presented in this paper. Datawere analyzed using unweighted meta-analysis techniques inthe R statistical software package (www.r-project.org; RDevelopment Core Team 2011). An observation pairconsisted of a data point for a designated control treatmentand a data point for an experimental treatment, for eachexperimental year, crop, and fertilization management. Twopreliminary subsets of data were identified. The first setincluded all experiments with deep inversion tillage (greaterthan or equal to 25 cm depth) as a treatment. For this set, deepinversion tillage was designated as the control treatment andall other treatments were comparedwith this control. A secondset of data included all experiments with shallow inversiontillage (less than 25 cm; as shown in Armengot et al. 2015Fig. 1b) as a treatment but without a deep inversion treatment.For the second set, shallow inversion tillage was designated asthe control treatment and all other treatments were comparedwith this control. Double-layer ploughing was classified asintermediate in intensity between deep inversion and shallowinversion. Only five experiments included double-layerploughing as a treatment, and these were all part of the firstset of data that also included deep inversion as a control. Itshould be noted that 5 out of the 41 experiments were includedin both sets of data, if they included deep inversion and shal-low inversion, as well as other tillage treatments. This was thecase for the OAFEG (Giessen, Germany), Kerguehennec(Bretagne, France), and Thil (ISARA, France) trials as wellas for experiments in Germany reported by Koch and Gaberle(2010) and Gruber and Claupein (2009).The effect size for each yield and weed incidence observa-tion pair was calculated as the response ratio (r=Xe /Xc),where Xe is the experimental treatment mean and Xc is thecontrol mean of each variable. For the soil C data, since allmeasurements were in the same units (g m−2), the effect sizewas calculated as the mean difference (MD=Xe−Xc).Fig. 2 Hierarchy of tillage intensities used to classify experimentaltreatments. Treatments within the box outlined in a solid line were partof the first set; the connecting arrow indicates the control treatment forthis set. Treatments within the box outlined in a dashed line were part ofthe second set; the connecting dashed arrow indicates the controltreatment for this setAgron. Sustain. Dev. (2016) 36:22 Page 9 of 20 22 The outliers among response ratios were rejected using arobust statistical method in which a “Tukey fence” wasestablished and all values outside the interquartile range wereconsidered as outliers (Hoaglin et al. 2000). The range was cal-culated fromQ1−k×IQR toQ3+k×IQR, whereQ1 is the low-er quartile point,Q3 is upper quartile point, IQR is the interquar-tile range (Q3-Q1), and k is a non-negative constant (here 1.5).The significance of effect sizes was tested using linear mixedeffects models in the “lme4” package in R with random effectsspecified as experiment/year. The fixed effects of tillage class orclimate were tested using this model. The effect of tillage classwithin a given soil type or crop class was tested after firstsubsetting the data by soil type or crop class and then using amixed effects model as described above. For all analyses, least-square means and confidence intervals were generated using the“lsmeans” package in R. Means were considered significantlydifferent from zero if the 95 % CI did not overlap zero, anddifferent from one another if their 95 % CIs were non-overlapping (Hedges et al. 1999). The results of each mean ratio(r) were expressed as % effect size (e), where e= (r−1)×100.To investigate the relationship between weed incidence andcrop yields, we modeled the relationship between crop yieldresponse ratio and weed response ratio using linear regressionin the “ggplot2” package in R. This fits a straight line throughthe set of n points to minimize the sum of squared residuals ofthe model (that is, vertical distances between the points of thedata set and the fitted line).4 Results and discussion4.1 Crop yield effects of reduced tillage intensity in organicsystemsWe found that on average across all environments and manage-ment practices yields were 7.6 % lower when tillage intensitywas reduced in organic systems compared to deep inversiontillage (total 873 observation pairs from 21 studies; Fig. 3a).This is greater than the 2.8 % average yield reduction reportedfor conventional reduced tillage systems in Europe (Van denPutte et al. 2010). We expected to see yield reductions fromreduced tillage in organic systems, as reported by Mäder andBerner (2012) in their review of European research; however,the effects were not obviously related to our classification oftillage intensity. When we considered the various classes ofreduced tillage compared to deep inversion (Fig. 3a), we foundthat the reductions in yield for both double-layer ploughing(7%) and shallow inversion tillage (5.5%) were relatively smalland for shallow non-inversion tillage the reduction of 3 % wasnon-significant. Although the tillage methods in Fig. 3a areTable 2 Environmental and cropmanagement factors and classeswithin each factor for the raw trialdata and selected published dataFactor ClassesSoil type (USDA texture class) Clay, clay loam, loam, loamy sand, sandy clay loam, sandyloam, silt loam, silty clay loambEuropean climate zone (Koppen climate class) Mediterranean (Csa, Csb), humid subtropical (Cwa, Cwb, Cfa),humid oceanic (Cfb), humid continental (Dfa, Dfb, Dwb)aCrop rotation class Arable with ley periods, intensive arable (no ley crops),intensive horticulture (no ley crops)Mechanical weeding (per growing season) None, once, 2–4 times, 5 or more timesCrop/previous crop classesLegume or grass/legume mixture grown for>12 monthsLegume ley<12 months legume cover crop primarily forsoil protection/improvementLegume cover crop<12 months cover crop primarily for soilprotection/improvementNon-legume cover cropAnnual cash crop of vegetable where aboveground parts are marketedLeaf vegetableAnnual cash crop of vegetable where belowground parts are marketedRoot vegetableBarley, wheat, rye, oats, spelt sown in theautumnWinter small-grain cereal/oilseedBarley, wheat, rye, oats, spelt sown in the springor summerSpring small-grain cereal/oilseedAnnual cash crop of legume Peas/beansGrain or silage maize or sorghum Maize/sorghumaAs used in Quemada et al. (2013)b As described in Soil Survey Division Staff (1993) 22 Page 10 of 20 Agron. Sustain. Dev. (2016) 36:22 arranged in order of declining intensity from top (double-layer)to bottom (no-till), there is no clear trend of increasing yieldreductions with reductions in tillage intensity. Surprisingly,deeper non-inversion tillage (10–25 cm) resulted in a significant-ly greater reduction in yield (11.6 %) relative to deep inversiontillage than shallow non-inversion (less than 10 cm) which didnot significantly reduce yield. This contradicts the results of themeta-regression carried out by Van den Putte et al. (2010) whoconcluded that yields were higher when tillage depth was in-creased in non-inversion systems. Soil type may be influencingthe results in our study. Closer inspection of the dataset indicatedthat the six studies included in the shallow non-inversion setrepresented only soils with loam or sandy loam textures. Thedeep non-inversion set included soils with a higher clay content,e.g., sandy clay loam, silty clay loam, and silt loam soils, as wellas loam and sandy loam. Arvidsson et al. (2013) carried out adetailed study on the impacts of tillage depth in non-inversionsystems in Sweden and found a slight negative correlation be-tween crop yield and clay content. This was attributed to chang-es in soil structure at depth in deep non-inversion tillage systems,which disturbed transport of water and nutrients. In our study,this impact may also account for the large yield reductions fordeep non-inversion tillage relative to shallow non-inversion.For the second subset of data where shallow inversiontillage was the control treatment, yields on average were4.2 % lower when tillage intensity was reduced, but thisdifference was not significant (total 206 observation pairsfrom 14 studies; Fig. 3b). In spite of the relatively largenumber of observation pairs and studies included, confi-dence intervals for this subset were still large and do notallow any firm conclusions to be drawn.4.2 Influence of climate and soil type on yields in organicreduced intensity tillage systemsIt has been suggested that the success of reduced intensity tillagesystems in organic farming is largely dependent on local envi-ronmental conditions, including climate and soil type (Vian et al.2009; Krauss et al. 2010). To investigate this, we conductedseparate analyses of yield effects for each climate and soil class.Figure 4a shows the overall effect of each climate classwhen tillage intensity is reduced relative to deep inversion.The dataset was primarily comprised of experiments fromthe humid continental zone; therefore, the yield reduction forthis climate class (8 %) is similar to the yield reduction for thewhole dataset. For the three studies that were conducted in thehumid oceanic climate there is a less than one percent reduc-tion in yield, while results from the Mediterranean zone showlarge yield reductions, although these represent only two stud-ies (Luna 2003 and Madden et al. 2004) so should not beconsidered as broadly representative of tillage effects onyields in this climate zone.As noted in the first analysis, there are no significant effectsof reduced tillage intensity in the subset of data where shallowinversion tillage is used as the control (Fig. 4b). The trend to-ward increases in yield in the Mediterranean climate zone isinteresting and deserves further investigation. In these regions,reducing tillage intensity may increase the amount of crop res-idues on the soil surface which improves soil and water conser-vation and enhances crop yields (Soane et al. 2012). This is abenefit frequently ascribed to conservation agriculture systemsunder conventional management, and it is reasonable to assumethat improvements in soil water relations under organicFig. 3 Overall effect on cropyields of reduced tillage and effectof each tillage method relative toa deep inversion tillage and bshallow inversion tillage. Meanvalues and 95 % confidenceintervals of the back-transformedresponse ratios are shown.Sample sizes (i.e., the number ofcontrol-treatment pairs)/numberof experiments are shown on theright of the confidence intervalsAgron. Sustain. Dev. (2016) 36:22 Page 11 of 20 22 management also occur. Further experiments on reduced tillageintensity in organic systems in the Mediterranean are needed toallow a more robust meta-analysis of the factors influencingcrop yield in organic reduced tillage systems in thisenvironment.The impact of soil type on yields where tillage intensity wasreduced relative to deep inversion was similar for the loamy soilgroup (−8 %; Fig. 5a) and the light soil group (−6 %; Fig. 5b).For the loamy soils, in this subset of data, there was some var-iation in the yield effect depending on the tillage class: double-layer ploughing, shallow inversion, and deep non-inversion till-age all significantly reduced yield, while the shallow non-inversion class of tillage did not result in any yield reduction.On the light soils, there was a trend toward yield reductionswhen either depth of non-inversion tillage was used. Van denPutte et al. (2010) also found the greatest yield reductions inno-till and reduced tillage systems on sandy soils. This may atfirst seem counterintuitive, but it is related to the challenge ofbuilding good soil structure in light, sandy soils. These soilslack the fine particles necessary to form the organo-mineralcomplexes that are the building blocks of soil aggregates(Bronick and Lal 2005), and are not de-compacted naturallyby the climate since they have no shrinkage and swellingeffects. Earthworm populations are also significantly lowerin soils with a sandy texture (Lapied et al. 2009), so they lackthese “ecosystem engineers” to build soil structure. TheseFig. 4 Overall effect of reducedtillage and effect of each climateclass on yields relative to a deepinversion tillage and b shallowinversion tillage.Mean values and95 % confidence intervals of theback-transformed response ratiosare shown. Sample sizes (i.e., thenumber of control-treatmentpairs)/number of experiments areshown on the right of theconfidence intervalsFig. 5 Overall effect of reducedtillage and effect of each tillagemethod on yields relative to deepinversion for a loam and b lightsoil types. Mean values and 95 %confidence intervals of the back-transformed response ratios areshown. Sample sizes (i.e., thenumber of control-treatmentpairs)/number of experiments areshown on the right of theconfidence intervals 22 Page 12 of 20 Agron. Sustain. Dev. (2016) 36:22 factors mean that without regular tillage, light soils can slumpand become compacted compared to loamy or clayey soils.Relative to shallow inversion, there was a significant re-duction in yields for the 14 observation pairs from the twoexperiments conducted on the light soils (17 %), but no reduc-tion for the loamy soil group (data not shown). There was onlyone study in the analysis that was conducted on heavy soils;for this case, yields were higher by 10 % when shallow non-inversion tillage was used rather than shallow inversion till-age. This supports the findings from the first dataset showingthat lighter soils tend to exhibit higher yield reductions thanheavy soils, as reported in Van den Putte et al. (2010).4.3 Impact of crop choice and rotational sequenceon success of reduced tillage intensity in organic systemsThe actual crop grown had some effect on the yield reductionrelative to deep inversion tillage (Fig. 6) but no effect for thesubset where shallow inversion tillage was the control (resultsnot shown). Combinable crops such as wheat and maize suf-fered the greatest average yield reductions (Fig. 6a; 8 %), andthis was particularly pronounced when deep non-inversiontillage was used when yield reductions were 13.5 %. Thisreflects the results of a Swedish study in conventional systemswhere the authors compiled results for 918 experimental yearsand found significantly lower yields for winter wheat in shal-low tillage systems as well as lower yields for sugar beet(Arvidsson et al. 2014). They concluded that for cereals,yields may be limited by the residues of the previous cropwhich could result in difficulties with crop establishmentand increased disease pressure. In organic systems where fun-gicides are not permitted, control of cereal fungal diseases isusually achieved by crop rotation and incorporation of resi-dues into the soil (Lampkin and Measures 2001). Without theoption of incorporating residues, disease pressure in organicreduced tillage systems may be higher than conventional sys-tems and contribute to the lower yields found in our study.Yield reductions for non-combinable crops were 6 %(Fig. 6b). The majority of the non-combinable crops includedin the study were root crops, either potatoes or carrots. Acloser investigation of tillage practices for these root cropsindicates that while depth of tillage may be shallower at theprimary tillage stage, in many cases (e.g., Scheyern trial),there are several subsequent tillage operations performed in-cluding rototilling prior to planting and ridging as many asthree times during crop growth. These practices could resultin soil conditions not unlike those in the control plots and alsoreduced weed incidence, both of which could explain whyroot crops do not experience as large a yield reduction as thecombinable crop category.Fig. 6 Overall effect of reducedtillage and effect of each tillagemethod on yields relative to deepinversion for a combinable, bnon-combinable, c cover crops/leys crop types. Mean values and95 % confidence intervals of theback-transformed response ratiosare shown. Sample sizes (i.e., thenumber of control-treatmentpairs)/number of experiments areshown on the right of theconfidence intervalsAgron. Sustain. Dev. (2016) 36:22 Page 13 of 20 22 In general, cover crops and leys were not affected by re-ductions in tillage intensity (Fig. 6c). There is some evidencethat reducing tillage intensity can benefit ley crops. For in-stance, grass-clover leys in reduced tillage plots of the Fricktillage trial profited in drought periods from better capillarywater supply from the ground, and produced more yield(Krauss et al. 2010). This was remarkable, as the establish-ment method for the ley in both the plough and the reducedtillage systems in the Frick trial was the same, indicating alasting effect of tillage treatments prior to ley establishment.In most cases, the choice of previous crop (data not shown)did not alter the effect on subsequent crop yields; a previouscombinable crop or cover crop/ley resulted in similar averageyield reductions of ~7–8 % relative to deep inversion. Thisdoes not confirm the widely held belief among organicfarmers that the mouldboard plough is essential for incorpo-ration of ley crops. Without the option of killing off a grass/legume cover with a herbicide, as commonly practiced amongthe non-organic no-till community, many organic farmers relyon deep inversion tillage to destroy the ley crop and minimizethe risk of grass becoming a weed in subsequent crops(Lampkin and Measures 2001). In practice, this may not benecessary or organic farmers may want to develop a “hybrid”system of tillage where some inversion tillage is still used atspecific stages of the rotation (e.g., for incorporation of a leyor for burying cereal residues for disease control) while otherreduced tillage options are implemented following some com-binable or non-combinable crops. Carr et al. (2013) describean organic zero till (ZT) system where soil is left undisturbedduring certain phases of the rotation, and strategically tilled atother phases, primarily to suppress weeds.4.4 Impact of time under reduced tillage on crop yieldsin organic systemsThere is a common perception that yield reductions followingthe adoption of reduced tillage will be ameliorated over time(Soane et al. 2012). This has been attributed to various factorsincluding short-term problems with soil compaction beforesoil structure has improved, reduced N availability in the shortterm, and operators initially lacking practical experience(Soane et al. 2012). To find out if this was the case in organicsystems, we plotted the effect size (% change from the control)versus the number of years that the tillage treatment had beenimplemented (Fig. 7). For both cases, there was no clear trendtoward a smaller yield effect over time. We did not haveFig. 7 Effect of length oftreatment period on yield in unitsof percent change from thecontrol yields: a control = deepinversion tillage and bcontrol = shallow inversiontillage. Mean values and 95 %confidence intervals of the back-transformed response ratios areshown (open circles). Samplesizes (i.e., the number of control-treatment pairs)/number ofexperiments are shown aboveeach mean. Dots represent eachobservation pair 22 Page 14 of 20 Agron. Sustain. Dev. (2016) 36:22 enough observation pairs to investigate the interaction be-tween crop rotation and length of time that reduced tillagehad been implemented; however, Van den Putte et al. (2010)found that for cereal only rotations the relative yield graduallydeclined due to the buildup of pests and diseases. This mayalso be happening in our studies. Although organic rotationsare usually characterized by a high degree of diversity and leyperiods for prevention of weeds and disease and the regener-ation of soil fertility (Lampkin andMeasures 2001), in the firstdataset (using deep inversion tillage as a control) about 20 %of the observation pairs came from studies classified as “in-tensive arable”with no ley periods in the rotation. The secondsubset of data represented an even higher proportion of inten-sive arable experiments with more than twice falling into thiscategory.4.5 Are weeds higher under reduced tillage intensityin organic systems?Relative to both deep and shallow inversion tillage, reducingtillage intensity in organic systems increases weed incidence.This is shown in Fig. 8a where deep inversion tillage was usedas the control and 94 observation pairs were identified. Forthis set of data, the overall average increase in weed incidencewas 54 % with only shallow inversion and no-tillage showingno increase in weed incidence relative to the control; however,the number of observation pairs for no-till is very low leadingto a high degree of uncertainty in the estimate of effect size.Results were similar where shallow inversion tillage was thecontrol and 68 observation pairs were identified (Fig. 8b);here, the overall increase in weed incidence was 56 %, withno difference in weed incidence among the tillage classes.There is a long-standing assumption in the organic produc-tion sector that weed pressure will increase if tillage is re-duced, and that this will lead to reductions in crop yields(Mäder and Berner 2012). To check if this was the case inour study dataset, we plotted the relationship between themean effect size for yield and the mean effect size for weedincidence (Fig. 9). A negative correlation between the twoparameters would imply that lower yields were associatedwith higher weed incidence. This was the case when tillageintensity was reduced relative to deep inversion for double-layer ploughing (P=0.049) and shallow inversion (less than25 cm) (P=0.023), but not for non-inversion at either depth.This is interesting because the treatment with the greatest yieldreduction was deep non-inversion tillage (Fig 3a), yet the re-lationship between weed incidence and yield was not signifi-cant for this tillage class. In the second set of data (shallowinversion tillage control), there was a negative correlation be-tween weed incidence and yields only when deep non-inversion was adopted. However, for shallow non-inversion,this relationship was non-significant. These results suggestthat while weed incidence may be a factor contributing toyield reductions under reduced tillage, other factors may alsobe involved.Since many of our experiments had been running for arelatively short time (see Fig. 7), nutrient supply patternsand soil structural conditions may have had more of an impacton yields than weeds. Yields may be restricted due to delays orlimitations in nutrient supply as has been reported when re-duced tillage is implemented in conventional systems (Soaneet al. 2012). This may create even more of a challenge inorganic farming where no supplemental mineral fertilizer in-puts are allowed. Soil structural limitations, i.e., compaction inFig. 8 Overall effect on weedincidence of reduced tillage andeffect of each tillage methodrelative to a deep inversion tillageand b shallow inversion tillage.Mean values and 95% confidenceintervals of the back-transformedresponse ratios are shown.Sample sizes (i.e., the number ofcontrol-treatment pairs)/numberof experiments are shown on theright of the confidence intervalsAgron. Sustain. Dev. (2016) 36:22 Page 15 of 20 22 the short term, may also create suboptimal conditions for rootgrowth (Soane et al. 2012). Typically, these problems are moreprevalent in reduced tillage systems during the initial “transi-tion” period.In contrast, weed populations may build up over time oncereduced tillage is implemented (Armengot et al. 2015). In thefirst years after transition to reduced tillage, the weeds may nothave reached the critical threshold level where they start torestrict yields (Peigné, personal communication).Nevertheless, our study showed a clear trend toward higherweed incidence when tillage intensity was reduced (seeFig. 8). Since organic farmers have few options to controlweeds once they become a problem, this trend is of particularconcern in a system where tillage is not available as a weedcontrol method. In organic reduced tillage systems, farmerswill have to be extra vigilant to ensure that weeds do notbecome a problem and employ creative approaches includingcover crops, diverse rotations, timely mowing of problemweeds, and integration of smother and/or allelopathic cropsto manage weed populations (Anderson 2015; Jabran et al.2015).4.6 Impacts of reduced tillage intensity in organic systemson soil C stocksA preliminary analysis of the data for both subsets of datawhere soil C stocks were reported or calculatedwas conductedto test if the sampling depth (0–20, 0–25, 0–30 cm and notreported) had any effect on the size of the standard meandifference (results not shown). This showed that samplingdepth did not affect the size of the standard mean differencedemonstrating that it was valid to pool results from studieswhich measured stocks at different depths (although the samesampling depth was used for any given observation pair).There were a total of 184 observation pairs in the firstdataset where reduced tillage intensity was compared to deepinversion tillage. For this set of data, all reductions in tillageintensity increased soil C stocks by 143 g m−2 (Fig. 10a).There was little difference in the relative increase in C stocksdepending on the tillage class, although shallow non-inversion(less than10 cm) did not result in a significant increase in Cstocks relative to the deep inversion control. For the secondsubset of data where reduced tillage intensity was compared toshallow inversion tillage, there was no clear increase in soil Cstocks when tillage intensity was reduced (Fig. 10b).Gattinger et al. (2012) have already demonstrated that or-ganic farming practices can result in enhanced soil C stocks.On average, they report C stocks that are 198 g C m−2 higherin organic systems compared with conventional in the top15 cm. These gains were attributed to the higher rates of Cinputs in organic systems from ley crops and manure/compostinputs. Tillage practices were not considered in their study;therefore, these gains occurred over a range of tillage methodswith mouldboard ploughing assumed to be the most commonpractice. Our study demonstrates that further gains in soil Cstocks in the topsoil can be achieved in organic farming sys-tems by reducing tillage intensity in mouldboard plough-based systems, something that has also been shown in numer-ous studies in conventional systems (West and Post 2002).This increase in topsoil stocks, which results from higher Cconcentrations in topsoil, can result in improved soil physicaland biological quality, one of the key benefits of reducedtillage systems (Angers and Eriksen-Hamel 2008). Higher Cconcentrations result in a range of soil quality and ecosystemservice benefits including higher water infiltration rates, im-proved aggregate stability, reduced erosion risk, enhanced soilbiological activity, and improved soil nutrient cycling(Fließbach et al. 2007).However, gains in topsoil stocks of organic C in systemswith reduced tillage intensity do not always translate into netgains in stocks if deeper sampling depths are included (seeSect. 2.3). This is why scientists are increasingly cautiousabout attributing climate change mitigation benefits to no-tillsystems (see Powlson et al. 2014). Nevertheless, in addition toFig. 9 Relationship between themean effect size for yield and themean effect size for weedincidence relative to a deepinversion tillage and b shallowinversion tillage. Slopes weresignificant for a double-layerploughing (P= 0.049); shallowinversion tillage (P= 0.023) and bnon-inversion tillage (P= 0.030) 22 Page 16 of 20 Agron. Sustain. Dev. (2016) 36:22 the soil quality and ecosystem service benefits of increasedtopsoil C, reduced tillage systems also provide an indirectgreenhouse gas mitigation service through reductions in fossilfuel energy inputs which can be 20 kg C ha−1 year−1 lower inconventional systems when converting from deep inversiontillage to no-till (Johnson et al. 2007; Reicosky and Archer2007).5 ConclusionsThe establishment of a sustainable crop rotation is a basicprinciple of organic farming (Soil Association 2014) andone of the three cornerstones of conservation agriculture.Maintenance of surface cover through residues and covercropping are also commonly used by many organicfarmers at specific stages of the rotation. However, theadoption of minimal soil disturbance through reduced till-age intensity has always been perceived as a major hurdlelimiting adoption of the full conservation agriculturepackage in organic farming systems. The concern hasbeen that crop yields, which can already be lower in or-ganic farming, may be further reduced by weed competi-tion and delayed nutrient mineralization. This analysis hasshown that yield reductions do not always occur whentillage is reduced in organic farming. The use of shallownon-inversion methods resulted in no significant declinein yields compared to deep inversion tillage. Shallow in-version tillage also showed promise with minimal reduc-tions in yield (~5.5 %), non-significant increases in weedincidence, and increased soil C stocks that provide addedbenefits from improvements in soil quality and ecosystemservice delivery.For all systems, the relationship between weed incidenceand crop yields was not as strong as expected, confirming thatother factors such as nutrient supply patterns and soil physicalproperties may be influencing crop productivity in organicreduced tillage systems.A useful outcome of this research for practitioners was theevidence showing that the double-layer plough resulted inyields, weed incidence, and soil C stocks similar to the shal-low inversion treatment, suggesting that there is no real ad-vantage to the double-layer plough compared with shallowinversion. Shallow inversion tillage may be a relatively easyoption for farmers to take up; although depending on soil typeand envisaged tillage depth, it may still be necessary forfarmers to invest in specialized machinery, such as a skimplough, in order to achieve consistently shallow depths ofinversion tillage.Concerns that weeds always limit crop yields in organicreduced tillage or no-till systems were not confirmed by ourstudy. However, the ~50 % higher levels of weed incidencewhen tillage was reduced relative to either deep or shallowinversion was cause for concern. Innovative farmers and re-searchers in Europe and North America are now working todevelop organic no-till and rotational tillage systems wherecover crops and surface residues are managed to suppressweeds. These systems use mechanical methods of cover cropdestruction (e.g., roller crimpers; see Fig. 1b) and no-till dril-ling equipment in place of herbicides to control weeds(Mirsky et al. 2012). Recent developments like automaticsteering technologies for inter-row weed hoeing using a cam-era proved the effectiveness of this practice to control weedsand to drastically reduce labor (Kunz et al. 2015).The poten-tial of using “complex” crop rotations to manage weeds inorganic no-till has also been demonstrated (Anderson 2015).Fig. 10 Overall effect on soil Cstocks of reduced tillage andeffect of each tillage methodrelative to a deep inversion tillageand b shallow inversion tillage.Mean values and 95% confidenceintervals of the back-transformedresponse ratios are shown.Sample sizes (i.e., the number ofcontrol-treatment pairs)/numberof experiments are shown on theright of the confidence intervalsAgron. Sustain. Dev. (2016) 36:22 Page 17 of 20 22 In both conventional and organic systems, the use of “stra-tegic tillage” at critical stages in the rotation to manage perni-cious perennial weeds or control residue-borne crop diseasesmay be appropriate (Dang et al. 2015); however, this shouldbe balanced against the negative effects on soil quality. Anyform of tillage may result in redistribution of carbon gains todeeper depths and mineralization of labile carbon fractions.This is a particular concern since most of the increases in soilC concentrations in no-till systems occur in the labile fractionwhich is most susceptible to mineralization when disturbed(Powlson et al. 2014). Moreover, the stable bio-pores whichhave been formed by roots and by earthworms are damagedby deep inversion tillage. Our data suggest that temporaryshallow inversion tillage is a good compromise, as we alsoobserved carbon gains in this system compared to deep inver-sion ploughing, with improved weed control relative to theother classes of reduced tillage. By placing the ploughingactivities in dry periods, in which vertically burrowing earth-worms move to the subsoil, negative effects on earthwormscan further be reduced.Further research is needed to quantify the trade-offs be-tween soil quality and crop productivity in organic reducedtillage systems that use these strategies for weed control. Apragmatic, rather than a dogmatic approach should be imple-mented to design optimized site-specific systems that will al-low the potentially additive benefits of conservation agricul-ture practices within organic farming systems to be realized.Acknowledgments The authors wish to acknowledge the efforts of allresearchers over the years who set up and maintained the various exper-iments included in this meta-analysis. This research was carried out with-in the framework of the TILMAN-ORG project (www.tilman-org.net)funded by CORE Organic II Funding Bodies, being partners of the FP7ERANet (www.coreorganic2.org).Open Access This article is distributed under the terms of the CreativeCommons At t r ibut ion 4 .0 In te rna t ional License (h t tp : / /creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided you give appro-priate credit to the original author(s) and the source, provide a link to theCreative Commons license, and indicate if changes were made.ReferencesAmossé C, JeuffroyM-H, Celette F, Goulevant G, David C (2013) Relay-intercropped forage legumes help to control weeds in organic grainproduction. Eur J Agron 49:158–167Anderson RL (2015) Integrating a complex rotation with no-till improvesweedmanagement in organic farming. A review. Agron Sustain Dev35(3):967–974. doi:10.1007/s13593-015-0292-3Angers DA, Eriksen-Hamel NS (2008) Full-inversion tillage and organiccarbon distribution in soil profiles: A meta-analysis. Soil Sci SocAm J 72(5):1370–1374. doi:10.2136/sssaj2007.0342Armengot L, Berner A, Blanco-Moreno JM, Mäder P, Sans FX (2015)Long-term feasibility of reduced tillage in organic farming. AgronSustain Dev. doi:10.1007/s13593-014-0249-yArvidsson J, Westlin A, Sorensson F (2013) Working depth in non-inversion tillage—effects on soil physical properties and crop yieldin swedish field experiments. Soil Tillage Res 126:259–266Arvidsson J, Etana A, Rydberg T (2014) Crop yield in Swedish experi-ments with shallow tillage and no-tillage 1983–2012. Eur J Agron52, Part B (0):307-315. doi:10.1016/j.eja.2013.08.002Baker JM, Ochsner TE, Venterea RT, Griffis TJ (2007) Tillage and soilcarbon sequestration—what do we really know? Agric EcosystEnviron 118(1–4):1–5. doi:10.1016/j.agee.2006.05.014Bàrberi P (2002) Weed management in organic agriculture: are we ad-dressing the right issues? Weed Res 42:176–193Bàrberi P (2006) Tillage: how bad is it in organic agriculture? In:Kristiansen P, Taji A, Reganold J (eds) Organic Agriculture: AGlobal Perspective. pp 295–303.Bàrberi P, Lo Cascio B (2001) Long-term tillage and crop rotation effectson weed seedbank size and composition. Weed Res 41:325–340Bender SF, Van Der HeijdenMGA (2014) Soil biota enhance agriculturalsustainability by improving crop yield, nutrient uptake and reducingnitrogen leaching losses. J Appl Ecol 52:228–239Bilalis DJ, Karamanos AJ (2010) Organic maize growth and mycorrhizalroot colonization response to tillage and organic fertilization. JSustain Agric 34(8):836–849. doi:10.1080/10440046.2010.519197Bilalis DJ, Karkanis A, Papastylianou P, Patsiali S, Athanasopoulou M,Barla G, Kakabouki I (2010) Response of organic linseed (Linumusitatissimum L.) to the combination of tillage systems, (minimum,conventional and no-tillage) and fertilization practices: Seed and oilyield production. Aust J Crop Sci 4(9):700–705Bilalis D, Karkanis A, Patsiali S, Agriogianni M, Konstantas A,Triantafyllidis V (2011) Performance of wheat varieties (Triticumaestivum L.) under conservation tillage practices in organic agricul-ture. Not Bot Horti Agrobo 39(2):28–33Bilalis D, Kakabouki I, Karkanis A, Travlos I, Triantafyllidis V, Hela D(2012a) Seed and Saponin production of organic quinoa(Chenopodium quinoa Willd.) for different tillage and fertilization.Not Bot Horti Agrobo 40(1):42–46Bilalis D, Karkanis A, Pantelia A, Patsiali S, Konstantas A, EfthimiadouA (2012b) Weed populations are affected by tillage systems andfertilization practices in organic flax (Linum usitatissimum L.) crop.Aust J Crop Sci 6(1):157–163Bronick CJ, Lal R (2005) Soil structure and management: a review.Geoderma 124(1–2):3–22. doi:10.1016/j.geoderma.2004.03.005Carr PM, Gramig GG, Liebig MA (2013) Impacts of organic zero tillagesystems on crops, weeds, and soil quality. Sustainability 5:3172–3201Collins D, Corbin A, Benedict C, Cogger C, Bary A, Burrows C,Miles C(2011) Reducing tillage with cover crops in western Washingtonorganic vegetable production: early research progress and futuredirection. Tilth Producers Q. 21.Dang YP, Moody PW, Bell MJ, Seymour NP, Dalal RC, Freebairn DM,Walker SR (2015) Strategic tillage in no-till farming systems inAustralia’s northern grains-growing regions: II. Implications foragronomy, soil and environment. Soil Tillage Res 152:115–123.doi:10.1016/j.still.2014.12.013Delate K, Cwach D, Butler J (2010) Evaluation of an Organic No-TillSystem for Organic Corn Production—Neely-Kinyon Farm Trial,2010. Annual Research Reports-2010 Armstrong Research andDemonstration Farm, Ames, IA, USADelate K, Cwach D, Mckern A, Schwarte K (2011) Evaluation of anOrganic No-Till System for Organic Corn Production—Neely-Kinyon Farm Trial, 2011. Annual Research Reports-2011Armstrong Research and Demonstration Farm, Ames, IA, USADi Tizio A, Lagomarsino A,Moscatelli MC, Marinari S, Mancinelli SGR(2008) The effects of system management on soil carbon dynamics.Lucrari stiintifice, seria B. Horticultura 51:645–650Diaz-Perez JC, Silvoy J, Phatak SC, Ruberson J, Morse R (2008) Effectof winter cover crops and no-till on the yield of organically-grown 22 Page 18 of 20 Agron. Sustain. Dev. (2016) 36:22 bell pepper (Capsicum annuumL.). In: Prange RK, Bishop SD (eds)Proceedings of the International Symposium on Sustainabilitythrough Integrated and Organic Horticulture. Acta Horticulturae.vol 767. pp 243–247Dittmann B, Zimmer J (2010) Ökologische Fruchtfolge Güterfelde. In:Schmidt H (ed) Öko-Ackerbau ohne tiefes Pflügen. Praxisbeispieleund Forschungsergebnisse. Verlag Dr. Köster, Berlin, pp 227–231Emmerling C (2007) Reduced and conservation tillage effects on soilecological properties in an organic farming system. Biol AgricHortic 24(4):363–377Fließbach A, Oberholzer HR, Gunst L, Mäder P (2007) Soil organicmatter and biological soil quality indicators after 21 years of organicand conventional farming. Agric Ecosyst Environ 118:273–284Food and Agriculture Organization of the United Nations (2015) What isconservation agriculture? http://www.fao.org/ag/ca/1a.htmlGattinger A, Muller A, Haeni M, Skinner C, Fliessbach A, Buchmann N,Mäder P, Stolze M, Smith P, Scialabba NE-H, Niggli U (2012)Enhanced top soil carbon stocks under organic farming. Proc NatlAcad Sci U S A 109(44):18226–18231Gruber S, Claupein W (2009) Effect of tillage intensity on weed infesta-tion in organic farming. Soil Tillage Res 105(1):104–111. doi:10.1016/j.still.2009.06.001Hedges LV, Gurevitch J, Curtis PS (1999) The meta-analysis of responseratios in experimental ecology. Ecology 80(4):1150–1156Hoaglin DC, Mosteller F, Tukey JW (2000) Understanding robust andexploratory data analysis. John Wiley & Sons, New YorkHobbs PR, Sayre KD, Gupta R (2008) The role of conservation agricul-ture in sustainable agriculture. Philos Trans R Soc Lond Ser B BiolSci 363:543–555Jabran K, Mahajan G, Sardana V, Chauhan BS (2015) Allelopathy forweed control in agricultural systems. Crop Prot 72:57–65. doi:10.1016/j.cropro.2015.03.004Johnson JM-F, Franzluebbers AJ, Weyers SL, Reicosky DC (2007)Agricultural opportunities to mitigate greenhouse gas emissions.Environ Pollut 150:107–124Kainz M, Gerl G, Lemnitzer B, Bauchenß J, Hülsbergen K-J (2005)Effects of different tillage systems in the long-term field experimentScheyern. 8. Wissenschaftstagung Ökologischer Landbau. Kassel,GermanyKoch W, Gaberle K (2010) Extensivierung der Grundbodenbearbeitungin einer auf Marktfruchtbau orientierten Vierfelder-Fruchtfolge(Bernburg). In: Schmidt H (ed) Öko-Ackerbau ohne tiefesPflügen. Praxisbeispiele und Forschungsergebnisse. Verlag Dr.Köster, Berlin, pp 232–236Köhl L, Oehl F, Van Der Heijden MGA (2014) Agricultural practicesindirectly influence plant productivity and ecosystem servicesthrough effects on soil biota. Ecol Appl 24(7):1842–1853. doi:10.1890/13-1821.1Krauss M, Berner A, Burger D, Wiemken A, Niggli U, Mäder P (2010)Reduced tillage in temperate organic farming: implications for cropmanagement and forage production. Soil Use Manag 26(1):12–20.doi:10.1111/j.1475-2743.2009.00253.xKunz C, Weber JF, Gerhards R (2015) Benefits of precision farmingtechnologies for mechanical weed control in soybean and sugarbeet—comparison of precision hoeing with conventional mechani-cal weed control. Agronomy 5(2):130–142Lampkin NH, Measures M (2001) Organic farm management handbook,4th edn. Welsh Institute of Rural Studies University of Wales,AberystwythLapied E, Nahmani J, Rousseau GX (2009) Influence of texture andamendments on soil properties and earthworm communities. ApplSoil Ecol 43(2–3):241–249. doi:10.1016/j.apsoil.2009.08.004Légère A, Stevenson F, Benoit D (2011) The selective memory of weedseedbanks after 18 years of conservation tillage. Weed Sci 59:98–106Lehocka Z, Klimekova M, Bielikova M, Mendel L (2009) The effect ofdifferent tillage systems under organic management on soil qualityindicators. Agron Res 7(Sp. Iss. 1):369–373Lewis DB, Kaye JP, Jabbour R, Barbercheck ME (2011) Labile carbonand other soil quality indicators in two tillage systems during tran-sition to organic agriculture. Renewable Agric Food Syst 26(4):342–353. doi:10.1017/s1742170511000147Liebman M, Davis AS (2000) Integration of soil, crop, and weed manage-ment in low-external-input farming systems. Weed Res 40:27–47Luna JM (2003) Conservation tillage systems for organic vegetable pro-duction. Organic Farming Research Foundation InformationBulletin. 12Luna JM, Mitchell JP, Shrestha A (2012) Conservation tillage for organicagriculture: Evolution toward hybrid systems in the western USA.Renewable Agric Food Syst 27(1):21–30. doi:10.1017/s1742170511000494Luo Z, Wang E, Sun OJ (2010) Can no-tillage stimulate carbon seques-tration in agricultural soils? A meta-analysis of paired experiments.Agric Ecosyst Environ 139:224–231Madden NM, Mitchell JP, Lanini WT, Cahn MD, Herrero EV, Park S,Temple SR, Van Horn M (2004) Evaluation of conservation tillageand cover crop systems for organic processing tomato production.HortTechnology 14(2):243–250Mäder P, Berner A (2012) Development of reduced tillage systems inorganic farming in Europe. Renewable Agric Food Syst 27(1):7–11. doi:10.1017/s1742170511000470Melander B, Munier-Jolain N, Charles R, Wirth J, Schwarz J, Van DerWeide R, Bonin L, Jensen P, Kudsk P (2013) European perspectiveson the adoption of nonchemical weed management in reduced-tillage systems for arable crops. Weed Technol 27:231–240Mirsky SB, Ryan MR, Curran WS, Teasdale JR, Maul J, Spargo JT,Moyer J, Grantham AM, Weber D, Way TR, Camargo GG (2012)Conservation tillage issues: cover crop-based organic rotational no-till grain production in the mid-Atlantic region, USA. RenewableAgric Food Syst 27(Special Issue 01):31–40. doi:10.1017/S1742170511000457Paffrath A, Stumm C (2010) Systemvergleich wendende und nichtwendende Bodenbearbeitung im Ökologischen Landbau(Auweiler). In: Schmidt H (ed) Öko-Ackerbau ohne tiefes Pflügen.Praxisbeispiele und Forschungsergebnisse. Verlag Dr. Köster,Berlin, pp 252–256Peigné J, Ball BC, Roger-Estrade J, David C (2007) Is conservationtillage suitable for organic farming? A review. Soil Use Manag23(2):129–144Peigné J, CasagrandeM, Payet V, David C, Sans FX, Blanco-Moreno JM,Cooper J, Gascoyne K, Antichi D, Bàrberi P, Bigongiali F, SurböckA, Kranzler A, Beeckman A, Willekens K, Luik A, Matt D, GrosseM, Heß J, Maurice Clerc, Dierauer H, Mäder P (2015) How organicfarmers practice conservation agriculture in Europe. RenewableAgric Food Syst. doi:10.1017/S1742170514000477Pittelkow CM, Liang X, Linquist BA, Groenigen KJV, Lee J, LundyME,Gestel NV, Six J, Venterea RT, Kessel CV (2014) Productivity limitsand potentials of the principles of conservation agriculture. Nature517:365–368. doi:10.1038/nature13809Ponisio LC,M’gonigle LK,MaceKC, Palomino J, DeValpine P, KremenC (2015) Diversification practices reduce organic to conventionalyield gap. Proc R Soc Lond Ser B Biol Sci 282:20141396. doi:10.1098/rspb.2014.1396Powlson DS, Stirling CM, Jat ML, Gerard BG, Palm CA, Sanchez PA,Cassman KG (2014) Limited potential of no-till agriculture for cli-mate change mitigation. Nat Clim Chang 4(8):678–683. doi:10.1038/nclimate2292QuemadaM, BaranskiM, LangeMNJN-D, Vallejo A, Cooper JM (2013)Meta-analysis of strategies to control nitrate leaching in irrigatedagricultural systems and their effects on crop yield. Agric EcosystEnviron 174:1–10Agron. Sustain. Dev. (2016) 36:22 Page 19 of 20 22 R Development Core Team (2011) R: A language and environment forstatistical computing. R Foundation for Statistical Computing. http://www.R-project.orgReicosky DC, Archer DW (2007) Moldboard plow tillage depth andshort-term carbon dioxide release. Soil Tillage Res 94:109–121Reinicke F, Heyer W, Christen O (2010) Langfristige Wirkungendifferenzierter Anbausysteme des Ökologischen Landbaus (BadLauchstädt). In: Schmidt H (ed) Öko-Ackerbau ohne tiefesPflügen. Praxisbeispiele und Forschungsergebnisse. Verlag Dr.Köster, Berlin, Germany, pp 242–246Seufert V, Ramankutty N, Foley JA (2012) Comparing the yields oforganic and conventional agriculture. Nature 485:229–232. doi:http://www.nature.com/nature/journal/vaop/ncurrent/abs/nature11069.html#supplementary-informationShirtliffe SJ, Johnson EN (2012) Progress towards no-till organic weedcontrol in western Canada. Renewable Agric Food Syst 27(1):60–67. doi:10.1017/s1742170511000500Silva EM (2014) Screening five fall-sown cover crops for use in organicno-till crop production in the Upper Midwest. Agroecol SustainFood Sys 38(7):748–763Soane BD, Ball BC, Arvidsson J, Basch G, Moreno F, Roger-Estrade J(2012) No-till in northern, western and south-western Europe: areview of problems and opportunities for crop production and theenvironment. Soil Tillage Res 118:66–87. doi:10.1016/j.still.2011.10.015Soil Association (2014) Soil Association organic standards farming andgrowing. Bristol, UKSoil Survey Division Staff (1993) Soil survey manual. vol Handbook 18.Soil Conservation Service, U.S. Department of AgricultureTeasdale JR, Mirsky SB, Spargo JT, Cavigelli MA, Maul JE (2012)Reduced-tillage organic corn production in a hairy vetch cover crop.Agron J 104(3):621–628. doi:10.2134/agronj2011.0317Vaisman I, Entz MH, Flaten DN, Gulden RH (2011) Blade roller-greenmanure interactions on nitrogen dynamics, weeds, and organicwheat. Agron J 103(3):879–889. doi:10.2134/agronj2010.0437Vakali C, Zaller JG, Koepke U (2011) Reduced tillage effects on soil prop-erties and growth of cereals and associated weeds under organic farm-ing. Soil Tillage Res 111(2):133–141. doi:10.1016/j.still.2010.09.003Van den Putte A, Govers G, Diels J, Gillijns K, Demuzere M (2010)Assessing the effect of soil tillage on crop growth: a meta-regression analysis on European crop yields under conservation ag-riculture. Eur J Agron 33(3):231–241. doi:10.1016/j.eja.2010.05.008Vian JF, Peigne J, Chaussod R, Roger-Estrade J (2009) Effects of fourtillage systems on soil structure and soil microbial biomass in organ-ic farming. Soil Use Manag 25(1):1–10. doi:10.1111/j.1475-2743.2008.00176.xWang Y, Tu C, Cheng L, Li C, Gentry LF, Hoyt GD, Zhang X, Hu S(2011) Long-term impact of farming practices on soil organic carbonand nitrogen pools and microbial biomass and activity. Soil TillageRes 117:8–16. doi:10.1016/j.still.2011.08.002West TO, Post WM (2002) Soil organic carbon sequestration rates bytillage and crop rotation: a global data analysis. Soil Sci Soc Am J66:1930–1946WilhelmB, Tiedemann L, Hensel O, Heß J (2011) GrundbodenbearbeitungimÖkolandbau - eine Betriebsumfrage. Paper presented at the Es gehtums Ganze: Forschen im Dialog von Wissenschaft und Praxis,Beiträge zur 11. Wissenschaftstagung Ökologischer Landbau,Justus-Liebig-Universität Gießen, 15–18 Mars 2011Zanin G, Otto S, Riello L, Borin M (1997) Ecological interpretation ofweed flora dynamics under different tillage systems. Agric EcosystEnviron 66:177–188 22 Page 20 of 20 Agron. Sustain. Dev. (2016) 36:22

TILMAN - the final report

Abstract

Extracted data

TILMAN - the final report

Abstract

Extracted data

Related items

Related items