Norges landbrukshøgskoleAgricultural University of Norway

DOCTOR SCIENTIARUM THESIS 2003:29

Studies of the availability of soil phosphorus (P) and potassium (K) in organic farming systems,

and of plant adaptations to low P- and K-availability

Undersøkelser av tilgjengeligheten av P og K i jorda i økologiske dyrkingssystemer,

og av planters tilpasning til lav tilgjengelighet av disse næringsstoffene

ANNE-KRISTIN LØES

Department of Soil and Water SciencesAgricultural University of Norway

P.O. Box 5028, N-1432 Ås

ABSTRACT

In organic farming systems, the purchase ofnutrients is firmly restricted as compared to con-ventional farming systems. This will often cause anegative nutrient budget on the field level, especi-ally in cases with low animal density, where morenutrients are removed in yields than are appliedin fertilisers. Studies on five organic dairy farmsin Southern Norway, that were aiming at self-suf-ficiency, revealed that the concentrations ofammonium-acetate lactate (AL) soluble phospho-rus (P) had decreased with time. The farms hadbeen organically managed since 1986 or earlier. Itwas found that the higher the initial P-AL concen-tration, the greater was the decrease that occur-red, calculated per year. Out of 156 topsoil sam-ples, only 6 had low P-AL values (< 25 mg P kg-1

soil) at the second soil sampling. Nevertheless, inthe long run, a supply of P will be required inorganic farming systems. For potassium (K), nocorresponding decrease in K-AL was found, butfor this nutrient a considerable proportion of thecultivated land had low concentrations (< 65 mg Kkg-1 soil) on all farms. The concentrations of acid-soluble K were also low to medium. With soil thatdoes not have a high ability to replace K taken upby plants, a supply of K will be required in orga-nic farming systems. Hence, the development ofnutrient supplies that may fulfil the standards oforganic farming is very important. Furthermore,organic farmers must strive even harder thantheir conventional colleagues to reduce nutrientlosses caused by for example erosion or duringanimal manure storage and spreading.

A basic principle in organic farming is the cyclingof nutrients within the farming system. With lowanimal density, plant material may be used asmanure (green manure or mulching). The fate ofP, K and nitrogen (N) was studied when choppedplant material was used as a mulch in vegetablegrowing. Application of 9-10 tonnes plant drymatter (DM) ha-1 increased vegetable yields by26% in the year of application, and grain yieldswere increased by 0.6 tonne ha-1 in the subse-quent year. However, when the nutrient uptake innon-mulched plants was subtracted, only 15-20%of the N, P and K applied in mulch was recoveredby the vegetable crops.

Much of the N was probably lost by leaching orgaseous losses, whereas most of the surplus P andK were recovered in soil. The mulch-method is

suited for relatively small farms with low animaldensity where there is no capacity to use all thefarmland for vegetables. About 2/3 of the landshould be used for mulch production, and 1/3devoted to vegetables. A systematic crop rotationmust be used to avoid an uneven distribution ofnutrients within the farm.

Studies in low-P growing media have shown thatplants may react to low P supply by adaptationssuch as increased root-shoot ratio, relatively long-er and thinner roots or longer root hairs. Plantshave a remarkable ability to adapt to changes inenvironmental conditions, such as a variablenutrient supply. In order to assess whether suchadaptations might be useful for organic farmingsystems, this topic was studied by growing accessi-ons of spring wheat and barley released duringthe period 1900-2000, in the field at both opti-mum and limited nutrient supply, as well as in thelaboratory in low-P nutrient solution, where roottraits were recorded. Nutrient uptake and grainyields varied significantly, but the relations bet-ween nutrient uptake and root traits were weak.Modern accessions produced higher yields thanolder ones, mostly because they had a higher har-vest index and were more resistant to fungal dis-ease. In cereal breeding, root traits should berecorded and the root length is important. As nosignificant differences in SRL were found amongbarley or wheat accessions in our study, I suggestthat root characterisation may be simplified byestimating root length from the root weight andan appropriate value for the specific root length(SRL, m root g-1 root DM). Measuring SRL is verytedious.

Most studies that have shown large effects in rootor root hair growth, as a consequence of low Psupply, have either not confirmed the obtainedresults under a controlled environment in thefield, or the plants were not grown to maturity.Commonly, the P concentration in such studieshas been below a level that is relevant for practi-cal farming. Such low levels would most probablyhave led to crop failure in the field. My studiesindicate that root morphological adaptations tolow P concentrations in soil are of theoreticalinterest, as they demonstrate the ability of plantsto survive in contrasting environments. However,for the conditions found in agricultural soil withinthe Nordic countries, such adaptations probablyhave little significance.

SAMMENDRAG

I økologisk landbruk er det sterke begrensningerpå hvor mye næringsstoffer som kan tilføres går-den utenfra, og på skiftenivå blir det lett en nega-tiv næringsbalanse – mer næring fjernes i avlingenn det tilføres i gjødsel. Dette gjelder særlig pågårder med lav husdyrtetthet. Undersøkelser påfem gårder i Sør-Norge som hadde drevet økolo-gisk melkeproduksjon med minst mulig fôrinn-kjøp siden 1986 eller lenger tilbake, viste at inn-holdet av fosfor (P) i jorda målt som P-AL(ammonium-acetat laktat løselig) sank over tid,også lang tid etter omlegging. Jo høyere innholdetvar ved første gangs måling, jo større var ned-gangen beregnet per år. Av til sammen 156 jord-prøver fra matjordlaget på de fem gårdene vardet bare 6 som hadde lave verdier for P-AL (< 25mg P kg-1 jord) ved andre gangs prøvetaking.Likevel vil det på sikt være behov for tilførsler avP i økologisk landbruk. For kalium (K) var detikke noen tilsvarende nedgang i K-AL, men fordette næringsstoffet var nivået lavt (< 65 mg K kg-1 jord) på en betydelig del av arealet på allegårdene, og innholdet av tyngre tilgjengelig K varogså lavt eller middels høyt. På jord som ikke haren god frigjøring av K fra naturens side vil detderfor bli nødvendig å tilføre K-holdig gjødseleller jordforbedringsmateriale ved økologiskdrift. Det er stort behov for å utvikle næringstil-førsler som kan tilfredsstille de kravene som stil-les til økologisk produksjon. Videre må man gjøreen enda sterkere innsats for å redusere nærings-tap fra gården, f eks ved erosjon eller fra husdyr-gjødsel, enn det som er vanlig ved konvensjonelldrift.

Økologisk landbruk baserer seg på å sirkulerenæringsstoffer som finnes på gården. Ved begren-set husdyrhold er det aktuelt å bruke plantemate-riale som gjødsel (grønngjødsling, jorddekke). Vihar undersøkt hva som skjer med P, K og nitro-gen (N) når opphakket plantemasse ble brukt somjorddekke til grønnsaker (hvitkål, rødbeter). Meden plantemasse tilsvarende ca 1 tonn tørrstoff til-ført per dekar økte avlingene med 26%, og etter-virkningen i bygg året etter var 60 kg korn perdekar. Sammenliknet med næringsopptaket iplanter uten jorddekke ble lite av næringen ijorddekket tatt opp i grønnsaker.

Opptaket tilsvarte bare 15-20% av N, P og K inn-holdet i jorddekket. N går lett tapt til luft ellervaskes ut med nedbør, men for P og K ble meste-

parten av den overflødige næringa lagret i jorda.Jorddekke-metoden er godt egnet på mindre går-der med lite husdyrhold, der det ikke er kapasitettil å dyrke grønnsaker på hele arealet. Man bru-ker da ca 2/3 av arealet til produksjon av plante-materiale til gjødsling, og 1/3 til grønnsaker. Deter viktig å ha et systematisk vekstskifte for atikke næringsstoffene skal bli ujevnt fordelt.

Undersøkelser i svært P-fattige dyrkingsmedierhar vist at planter ofte reagerer på lav P-tilgangmed å øke rot-skudd forholdet, utvikle lengre ogtynnere røtter, lengre rothår og liknende tilpas-ninger. Planter har en betydelig evne til å tilpasseseg ulike miljø, herunder en varierende nærings-tilgang. For å undersøke om slike tilpasningerkan være nyttige også i økologiske dyrkingssy-stem, ble sorter av bygg og vårhvete fra perioden1900-2000 dyrket i felt med optimal og begrensetnæringstilførsel, og i næringsløsning med lavt P-innhold. Sentrale rotegenskaper ble målt inæringsløsning, og i felt med begrenset næringstil-førsel. Både næringsopptak og kornavling varier-te betydelig, men det var lite samsvar mellomnæringsopptak og rotegenskaper hos de ulike sor-tene. Gjennomgående ga moderne sorter høyereavling enn eldre sorter både ved optimal ogbegrenset næringstilgang. Viktige årsaker til detteer en høyere kornandel (”harvest index”), og enbedre soppresistens. I kornforedlingsarbeid børman likevel være oppmerksom på linjenes rot-egenskaper, spesielt rotlengden. Mine undersø-kelser viste at denne egenskapen kan måles indi-rekte på en enkel måte, ved hjelp av rotvekten.Det var ingen sikre sortsforskjeller i rotfinhet,også kalt spesifikk rotlengde (antall meter rot perg rot-tørrstoff) verken i bygg eller hvete. Den spe-sifikke rotlengden er svært arbeidskrevende åmåle.

De fleste undersøkelsene som har vist store effek-ter av et næringsfattig miljø på veksten av røttereller rothår, har ikke prøvd ut resultatene frakontrollerte betingelser i felt eller dyrket plantenefram til høsting. I mange tilfeller har P-konsen-trasjonen som forsøkene er utført ved vært myelavere enn det som er mulig i praktisk jordbrukuten å få misvekst. Våre undersøkelser tyder påat rotmorfologiske tilpasninger til lave P-konsen-trasjoner kan være viktige for planters evne til åoverleve i svært forskjellige miljø. Men i praksis,for de forholdene som gjelder i dyrka jord iNorden, har slike tilpasninger sannsynligvis litenbetydning.

CONTENTS

Background and objectives 3

Soil and cycling 4

Maintenance of soil fertility 4

Assessment of soil P and K concentrations by long-term organic farming 5

Soil sampling and analysis 5Soil P concentrations 5Soil K concentrations 5Sustainability of the P and K management in organic farming systems 6

Organic plant production with plant material used as nutrient input 8

Effect on yields of vegetables and subsequent barley, and the recovery of mulch N, P and K 8Nutrient budgets reflected in changes in soil P and K concentrations 9Sustainability of the P and K management by chopped plant mulch vegetable 10growing

Plant adaptations to limited P and K availability in soil 11

The interest of plant adaptations in organic farming systems 11

Yield reductions in organic production due to limitations in nutrient supply 11Plant growth retardation by nutrient deficiency 12Root morphological traits increasing the nutrient uptake 12Studies of the specific root length 12

Relations between root traits and nutrient uptake in field for 12

selected Norwegian accessions of spring wheat and barleyExperimental design 12Root traits as related to nutrient uptake in the field 13Grain yield levels and the usefulness of modern accessions in organic 13farming systems

The relevance of low-P studies in controlled environment for organic farming systems 14

Acknowledgements and personal comments 16

References 17

1

APPENDIX

Papers I-V

The present doctoral thesis is based on the following papers I-V, which will be referred to by their Romannumerals.

I Løes, A. K. and A.F. Øgaard, 2001. Long-term changes in extractable soil phosphorus (P) in organic dairy farming systems. Plant and Soil 237, p 321-332

II Løes, A. K. and A.F. Øgaard, 2003. Concentrations of soil potassium after long-term organic dairy production. Accepted by International Journal of Agricultural Sustainability, vol. 1 (in press)

III Riley, H., A.K. Løes, S. Hansen and S. Dragland, 2003. Yield responses and nutrient utilization with the use of chopped grass and clover material as surface mulches in an organic vegetable growing system. Biological Agriculture and Horticulture 21, p 63-90

IV Løes, A.K. and T.S. Gahoonia. Genetic variation in specific root length in Scandinavian wheat and barley accessions. Submitted to Euphytica International Journal of Plant Breeding.

V Løes, A.K., E.M. Færgestad, T.S. Gahoonia, H. Riley and M. Åssveen. The impact of root traits, nutrient uptake, age of accession, growth period and resistance to fungal disease for cereal production with limited nutrient supply and pesticide use.Submitted to Crop science

Permission to print Papers I, II and III was kindly granted by Plant and Soil/Kluwer AcademicPublishers, The Netherlands, International Journal of Agricultural Sustainability/Channel ViewPublications, UK and Biological Agriculture and Horticulture/Academic Publishers, UK, respectively.

2

Background and objectives

This doctor scientiarum thesis has been written tocontribute to the further development of organicfarming, which in Scandinavia and Germany isoften called “ecological” farming. The oldestresults included in the study were recorded morethan 20 years ago, when organic farming was arare farming practice in Norway, carried out byidealistic farmers who often practised biodynamicprinciples. The extent of organic farming practi-ce, described as “alternative” farming, was asses-sed to 151 farms in 1983, comprising 0.05% of thetotal farmland (Skøien, 1983). Since then, thenumber of organic farms has increased to 2303,comprising 3.3% of the farmland in 2002 (Debio,2003). In 1991, the organic farming method recei-ved a legal status by the EU Council Regulation2092/91 that defines standards for organic plantproduction. The Norwegian government aims at a10% fraction of the total farmland being organi-cally managed within 2010 (Ministry ofAgriculture, 2003). Organic farming is not anymore as exclusive as it once used to be. In a socio-logical study (Vartdal, 1993), the growth of orga-nic farming was found to have much in commonwith a classical innovation. Organic farmers couldbe grouped into innovators, early adopters, earlymajority, late majority and late-comers. It isreasonable to conclude that at present, organicfarmers are comprised of the three categories firstmentioned, and that the “early majority” consti-tutes an increasing proportion. A survey amongthe 202 Norwegian farmers managing their farmorganically in 1992, revealed that food quality,sustainable management of resources, and carefor the environment were their main reasons toconvert to organic agriculture. However, for 10%of the farmers, increased income was an impor-tant reason to convert (Løes, 1992). A govern-mental subsidy for conversion was introduced in1990, and premium prices on organic milk in1995. Hence, most probably the number of far-mers that are motivated by a higher income whenthey decide to convert has increased notablyduring the last decade. There is a risk that suchfarmers will go back to conventional farmingpractice if the premium prices are reduced. Tostabilise the further growth of organic farming,and to ensure that the method continues to deve-lop in a sustainable direction, there is a need tostrengthen the ideological base of the farmingpractice. The term sustainability is excellentlydescribed in a principal aim of the IFOAM(International Federation of Organic AgriculturalMovements) Basic standards, which states thatorganic production shall “.. support the establish-ment of an entire production, processing and dis-tribution chain which is both socially just andecologically responsible” (IFOAM 2002; my ita-lics). Such an aim requires a profound understan-ding of organic production systems, including theinteractions of the many factors that constitute

them. Knowledge of the management of plantnutrients will contribute to the development of amore sustainable organic agriculture, in additionto the essential agronomic significance of suchknowledge. Sustainability is a principal goal oforganic agriculture, but conceptual thinking mustbe followed by practical action. Organic farminghas reached an extent where a critical debate ofthe sustainability of this farming practice isrequired. The present thesis will contribute inthis debate.

The main part of the experimental work presen-ted here was conducted within the framework of astrategic inter-institutional research programme,”Nutrient supply to organic farming systems withsmall amounts of animal manure” (1998-2002). Aformer study on farm level was also included,where the initial samples were taken as early as in1983. The Norwegian Research Council providedfunding for both these activities. The first aim ofthe experimental work was to assess the sustaina-bility of organic farming systems with regard tophosphorus (P) and potassium (K) management.Secondly, I wanted to assess the potential of plantadaptations to low nutrient supply in organic far-ming systems. Genetic variability in cereal roottraits, and the relation between root traits andnutrient uptake by low nutrient supply in thefield, was the object of research in this part of thework.

The following questions have been the main topicsin the studies included in the thesis:

– What happens with the soil concentrations of Pand K after long-term organic management? Isthe present management of these nutrient resour-ces in organic farming systems sustainable?(Papers I and II)– What happens with the surplus P and K that isadded to the soil when amounts of chopped plantmaterial required for a satisfactory yield are usedas mulch in vegetable growing? Can P and K beredistributed within a stockless organic farmingsystem as chopped plant mulch, with acceptablelosses of these nutrients? (Paper III)– Does the specific rot length (= root fineness)vary significantly among cereal accessions?(Paper IV)– Can significant genetic variation be foundamong Norwegian accessions of spring wheat andbarley with respect to root traits that are impor-tant for the uptake of nutrients in field? (Paper V)– Are modern accessions of cereals better adaptedthan older accessions to the growing conditions inorganic production systems? (Paper V)

3

Soil and cycling

Maintenance of soil fertilityIn organic farming, a basic principle states thatfarmers should “maintain and increase long-termfertility and biological activity of soils using local-ly adapted cultural, biological and mechanicalmethods as opposed to reliance on inputs”(IFOAM 2002; my italics). This statement empha-sises that agricultural soil should not be regardedas a passive, inert medium where the nutrientsrequired for plant growth must be purchased andapplied. On the contrary, satisfactory soil fertilitycan be achieved by cycling of the farm nutrientresources. On a line drawn between these con-trasting ideas of soil and fertilisation, conventio-nal management is closer to a linear model, whe-reas organic management is closer to a cyclingmodel (Figure 1).

Soil provides physical support, water and airavailability for plant roots, as well as chemicaland biological support for plant growth and deve-lopment. Plant nutrients are released by weathe-ring of minerals, and by mineralisation of organicmatter. Recalcitrant soil organic matter has alarge impact on soil physical properties, and thesoil biota has a significant impact on plantgrowth. Cultivated soil is tilled and fertilised toensure satisfactory yield levels. The essential goalof self-sufficiency implies that organic farmers tryto minimise all kinds of purchased inputs to thefarm. Not only purchased fertilisers, but also fod-der and bedding material contain large amounts

of nutrients. Due to sale of products, and inevi-table internal losses of plant nutrients, the nutri-ent cycle is not closed, and it may be questionedwhether the goal of increased soil fertility can bereached. In Norway, professor emeritus ErlingStrand has repeatedly expressed strong criticismof organic farming systems for being non-sustai-nable because soil reserves of nutrients are redu-ced (e.g. Strand, 1997, Strand, 1999). Anotherextreme position is taken by some organic farmerswho advocate that fertilisation is not required aslong as a satisfactory soil structure is achieved,e.g. by the “Kemmink”-system (Søgaard, 1997;Hansen, 2000). Here, the soil is repeatedly deeplytilled during the growing season, and all machinesare driven along fixed lines in the field. There isno doubt that fertiliser demand is increased bypoorer soil structure caused by tractor traffic. Ina combined fertilisation and soil compaction tre-atment, amounts of aerated slurry containing 105kg N, 109 kg K and 17 kg P ha-1 y-1 was requiredto achieve the same yields of grass-clover ley withnormal tractor traffic, as in the uncompacted andunfertilised treatment (Hansen, 1996). The weat-hering and mineralisation of nutrients in the soilmay be notable, as have been shown in some long-term field experiments referred by Lieblein andSolberg (1996). These authors argue that releaseof nutrients from the soil may balance the nutri-ent deficits in organic farming systems. With thishot-tempered debate as a background, changes insoil concentrations of P and K by long-term orga-nic farming were studied to assess if the goal ofmaintained or increased soil fertility was achie-ved.

4

Organic Conventional

Figure 1. Organic and conventional farming oriented according to contrasting ideas of nutrient management. To theright, the agricultural system is perceived as an industrial unit where required nutrients are purchased to maximisethe output for sale, often with a separation of arable and animal husbandry. To the left, the agricultural system isperceived as a cycling of the inherent nutrient resources, where the cycling creates a surplus of outputs for sale.Small amounts of nutrients may be purchased (- - -) when strictly required.

Assessment of soil P and K concentrationsby long-term organic farming

In a case study of 12 farms in conversion to orga-nic farming, significant reductions in the soil con-centrations of plant-available P and K was foundfrom 1989 to 1995 (Løes and Øgaard, 1997). Theinputs of both P and K to the farming systems arereduced during conversion, when purchased fer-tilisers are gradually omitted and the amount ofpurchased concentrates and other feed often sig-nificantly decreased (Kerner, 1993). Hence, thedecreases were not surprising, and we suggestedthat the levels of available P and K would gradu-ally stabilise on a lower level adapted to the lowernutrient inputs. In stead of waiting several yearsand sample the same farms again to assess whet-her this hypothesis was right, we chose five otherfarms where organic dairy production had occur-red for several years to study changes in soil Pand K concentrations. Dairy farms were chosenbecause stockless organic farms are rare inNorway. If the soil nutrient concentrates werefound to decrease within these systems, withnutrient budgets being close to zero on the farmlevel, they may be expected to decrease morerapidly in systems with larger nutrient deficits. Inorganic animal husbandry systems, the nutrientdeficits are usually increased by decreasing lives-tock density (Eriksen et al., 1995). When manureis legislated for use in organic farming and avai-lable for purchase, the farm level nutrient budgetmay show large surpluses as found for horticultu-ral systems by Watson et al. (2002), but the fieldlevel nutrient deficiencies will be very large instockless systems if no nutrients are purchased.

Soil sampling and analysis

On five dairy farms where a set of initial soil sam-ples was available from the period 1983-1990, asecond set of samples was collected from 1996 to1998. In order to obtain a reliable comparison,the sampling areas (about 80 m2) were the sameon both sampling dates. Both topsoil (0-20 cm)and subsoil (20-40 cm) was sampled, and analysedfor ammonium-acetate lactate (AL)-soluble P andK. This extraction method is the common analysisto assess the need for supply of these nutrients inNorway. For P, much more is usually extractedby the AL-method than any crop takes up duringone season. Hence, the P-AL extraction reflectssoil reserves of P in addition to the immediate Pavailability from the soil. For potassium, K fracti-ons that are not extracted by the AL-solution alsodeliver K to crops. Extraction with nitric acid (K-HNO3) was used to estimate the soil’s capacity todeliver K from K reserves. Acid-soluble K is theK extracted by nitric acid minus the K extractedby AL-solution. Farm level nutrient budgets wereestimated from annual accounts.

Soil P concentrations

Because of inevitable internal losses (Nolte andWerner, 1994), we have reason to believe that onthe field level, averaged over the crop rotationthree farms had P deficits (Paper I). One farmhad a low surplus, and one had a surplus compa-rable to conventional dairy farms. On all farms,the average topsoil P-AL concentration decreasedfrom the first to the second sampling (Paper I).The rate of decrease was found to be related tothe soil P-AL concentration at the first sampling,so that by a higher initial level, the decrease pro-ceeded more rapidly (Figure 2, steeper lines). Theaverage P-AL level was still medium (26-65 mg Pkg-1 dry soil) or high (66-150), and only 6 out of156 topsoil samples had a low P-AL concentration(< 25) at the second sampling. The results indicatethat if the farm management is pursued unchang-ed, the average P-AL value in topsoil will appro-ach a critically low level (Paper I) of ca 30 mg Pkg-1 dry soil within 15-45 years on four of the fivefarms (number 1, 2, 4 and 5 in Figure 2). Forconventional farming, a P-AL concentration of 70mg P kg-1 soil has been assessed as the best com-promise between the aims of high soil fertility andlow risk of algae growth due to P pollution ofwater bodies (Krogstad and Løvstad, 1987). Asthe N inputs in organic farming is generally lowerthan in conventional, it is possible that the opti-mal P-AL level in soil is somewhat lower in orga-nic farming systems. But it may be dangerous touse such speculations as an excuse for depletingsoil nutrient reserves. From our study, we conclu-ded that regular records of soil P concentrationsare required in organic farming, and that the Pinputs should be increased when the average levelis approaching the lower part of the mediuminterval. P inputs applicable in organic farmingsystems may be increased amounts of purchasedfodder, compost from household waste, rockphosphate, bone meal or even human urine (notyet legislated). On the last farm (no 3 in Figure2), the average topsoil P-AL concentration was119 mg P kg-1 dry soil. The most sustainable wayto handle the soil P reserves in such cases is toproceed with negative P balances on field leveluntil a more sustainable level is reached.

In the topsoil on farm 5, that has been managedbio-dynamically since 1932, a larger part of AL-soluble P than is usually found in cultivated soilswas organic P (Paper I). Another interestingresult was the remarkably even P-AL concentrati-ons in topsoil and subsoil (20-40 cm) on the samefarm. This indicates a deep topsoil layer, mostprobably due to the gentle soil cultivation practi-ce where much tillage has been done by horse,which has resulted in a satisfactory soil structure.This may well have facilitated a large earthwormactivity, as has been found by long-term biodyna-mic farming in farming system comparisons inSwitzerland (Pfiffner et al., 1993).

5

Soil K-concentrations

The inevitable internal losses of K are much hig-her than for P (Nolte and Werner, 1994). Hence,most probably all farms had K-deficits on thefield level, averaged over the crop rotation(Paper II). For plant-available K in soil, no signi-ficant decreases were found from the first to thesecond sampling. The average K-AL concentrati-ons in topsoil at the second sampling weremedium high (65-155 mg K kg-1 dry soil). Between26% and 45% of the topsoil samples had low K-AL concentrations (< 65), implying that the Kavailability may be limiting crop production on asignificant part of the farmland on all farms if Kis not provided in fertilisers or released from soilK reserves (Paper II). With respect to K reser-ves, the acid-soluble K-concentrations were gene-rally medium to low or low and mineralogicalanalysis by X-ray diffractometry did not revealsignificant amounts of K-releasing minerals. Noneof the samples studied contained more than 6%illite, and the content of K-feldspars was only 8-13%. Hence, the ability of the soil to replenish Ktaken up by plants on the farms studied here wasgenerally small. A significant decrease in acid-soluble K was found in both topsoil and subsoilon farm 4, which had a low livestock density andthe largest K deficits. As was found for P-AL, thelargest decreases occurred on the fields with thehighest initial K concentrations on this farm(Paper II).

As was discussed for P, it may be questionedwhether the concentration levels indicating “low”,“medium” or “high” availability of K are appro-priate for organic farming systems, where the Navailability is restricted. However, there are notenough data available from organic production to

evaluate to which extent the conventional criteriamay be used for organic systems. Hence, forpragmatic reasons we have used the well-establis-hed conventional criteria. This topic should befurther studied.

The K concentrations in soil indicated that thereis a serious risk of K deficiency in a significantpart of the farmland on each of the farms thatwere studied (Paper II). Severe K deficiencysymptoms have not yet been observed. This doesnot imply that the yield levels have not been redu-ced by the K availability. This may be the case inK demanding crops and with dry conditions thatdecrease the K transport rate in soil both by dif-fusion and mass flow. A limited K availability insoil may also have a negative impact on theamount of biologically fixed nitrogen (N) in thefarming systems. This is because ryegrass is moreefficient in the competition for K than red clover(Mengel and Steffens, 1985) and white clover(Salomon, 1999), which are essential legumes inNorwegian organic farming systems. Hence, thefarm management should be changed also withrespect to K. It is important to consider how allnutrient losses can be minimised. Storage andhandling of fodder and manure, as well as thetiming of soil tillage and management of crop resi-dues during winter should be considered in thiseffort. An example is how care is taken of therunoff from platforms where solid manure is com-posted, and the winter recreation paddocks forcattle. In addition, on farms producing significantamounts of cash crops the crop rotation may bechanged. For instance, in stead of K demandingpotatoes, grass or legume seed may be cropped.Further, the K inputs to the farming systems maybe increased as described for P above.

6

0

20

40

60

80

100

120

140

160

Years from first sampling

Farm 1

Farm 2

Farm 3

Farm 4

Farm 5

0 2 4 6 8 10 12

AL

-extr

acta

ble

P,

mg

kg

s

oil

-1

Figure 2. Change in average topsoil (0-20 cm) P-AL concentration on five long-term organic dairy farms during 7-13years of continued organic management.

Sustainability of the P and K management inorganic farming systems

Our studies were done on dairy farms with avarying degree of self-sufficiency with regard tonutrients. All farmers aimed at being self-suffici-ent, but the farmland size was very restricted ontwo of the farms and some fodder was purchasedto ensure a sufficient milk production level. Acomparatively larger purchase of fodder is usual-ly linked to a higher animal density, as may beseen from the present study where the size offarmland per milking cow varied from 2.8 to 0.75ha, and the nutrient deficits were much larger inthe first case (Paper II). Our results show thateven on organic farms where only milk and somemeat was sold, the soil P concentration decreasedwith time. In self-sufficient farming systems witha significant amount of cash crops, the K concen-trations in soil were also reduced.

Do these results confirm the criticism that organicagriculture is non-sustainable? Referring to theIFOAM principles cited above, sustainabilityimplies social justice and ecological responsibility.There is no doubt that the social justice is astrong motivating factor for the farmer aiming atself-sufficiency (Løes, 1992; Vartdal, 1993). Atextbook in organic soil cultivation states that “Ina global perspective, there is hardly enoughresources and energy to introduce industrialisedfarming practice all over the world” (Hansen andMcKinnon 1999, my translation). Mineral Presources are scarce (Lægreid et al., 1999), andwith the present use of P fertilisers all knownresources will be consumed within 250 years.Hence, mineral phosphates should be used in soilswhere they would cause significant yield increase,and contribute to reduce hunger problems.Mineral K resources are abundant, both as mine-rals and in seawater, but the recovery and distri-bution of mineral K requires energy. With respectto the ecological responsibility on a global scale,the organic farmer takes regard by using localresources and thereby avoiding pollution andenergy use during production and transport ofmineral fertilisers. By avoiding excess fertilisati-on, with negative environmental impacts, she isalso ecologically responsible on smaller scale. Ona farm scale level, to define an ecologicallyresponsible level of soil fertility we may ask atwhat level the crop production is reasonable incomparison to the required input of energy.Extremely low yield levels as compared to conven-tional production can not be accepted, but exactminimum yield levels are difficult to set. Criteriato evaluate achieved organic yield levels and towhich extent they may be assessed as satisfactorywith regard to the input/output ratio of energyand other resources is the aim of a recently star-ted research programme (Eltun, 2002).

When evaluating nutrient management on farmlevel, a complicating factor is that a possibledecrease in soil fertility and yield levels may pro-ceed slowly and for long be masked by yield levelvariations due to climatic conditions. Analysingwhether changes in the farm management are infact an adaptation to a decreased yield level isalso complicated. To decrease farm nutrient defi-cits, the farmer may increase the amount of pur-chased fodder, or reduce the number of calvesraised for beef production. However, the reasonfor more fodder may just as well be an increasedincome, and fewer calves may be due to a demandof reduced work.

From my point of view, there is a risk that strong-ly motivated organic farmers may over-emphasisethe aim of self-sufficiency, reasoned by ecologicalresponsibility on a global level. In consequence,on the local level the farming practice may not beregarded as sustainable in a long-time perspecti-ve. Hence, organic farmers have to monitor theirsoil fertility status regularly, and manage nutrientresources carefully. Organic farming should cont-inue to focus on optimising the use of soil reser-ves. Some soils have large reserves of P becauseof a high content of organic matter and/or deca-des of surplus P fertilisation. Some soils containminerals that release significant amounts of K. Insuch cases, negative budgets of P and/or K onfield level will be sustainable. Other soils are lessfertile with respect to nutrients. In general, thereis a need to close the nutrient cycles, so that a lar-ger proportion of the sold nutrients is returned toagriculture. There is an urgent need to study hownutrient sources like human urine and wastesfrom household and industry can be processedand applied in organic agriculture. It is essentialthat such products appear in a form that does notcompromise soil quality aspects like the content ofheavy metals and pesticide residues. However, theEU- as well as national regulations for organicfarming systems must be changed to be adapted toa larger degree of nutrient transfer from societyto agriculture.

Currently, there is a limited use of mineral K-fer-tilisers and rock phosphate in organic farmingsystems, but the amounts that are used are smallas compared to conventional systems. Suchresources should be regarded as valuable remedi-es for the least fertile soils assessed on a globalscale, rather than being used in excess in wealthycountries. With respect to N, biological fixationwill always be the major input in organic farmingsystems, and it must be emphasised that this dis-cussion is not intended as suggestion to generallylegislate the use of mineral fertilisers in organicfarming.

7

Organic plant production with plantmaterial used as nutrient input

So far, organic farming in Norway has mainlybeen comprised of animal husbandry systems,mostly roughage based productions (dairy, sheep,and beef). The demand for organically producedvegetables and cereals is increasing, and quite afew conventional farmers consider a conversion toorganic plant production only if this can be donewithout establishment of animal husbandry. Onepossible design of such a system is to producevegetables for sale one a part of the farmland,whereas the rest is used for production of plantmaterial that is used as nutrient input for thecash crops. Chopped plant material used asmulch combines weed control and nutrient supplyfor vegetables (Paper III). The method is analternative way of organising the redistribution ofnutrients within a farm, as compared to producti-on of fodder and recycling of nutrients to soil byanimal manure. The large amounts of plant mate-rial commonly used as mulch to achieve an effici-ent control of weeds as well as a significant yieldincrease, implies field level nutrient balances withlarge P and K surpluses on a small part of thefarmland. Up to 3 times the size of the vegetableland must be used for mulch production, depen-dent on how early in the growing season the firstapplication must be made (Paper III). On thisland, there will be a corresponding nutrient defi-cit. There was a need to study whether the P andK not taken up in vegetables remain in soil, toassess if the mulching system is sustainable withregard to P and K management. By mulching withN-rich plant material, there is a risk of losses ofammonia (NH3) and nitrous oxide (N2O), asshown by Larsson et al (1998). This problemshould be solved before the method is widespre-ad; however, that was not the task of the presentstudy.

From 1998 to 2001, field experiments were carri-ed out on Norwegian Institute of Crop Science,Research Centre Apelsvoll division Kise (PaperIII) to study the effect of chopped plant mulch onyields of vegetables with a varying demand fornutrients, when various amounts and types ofplant material were used. The fate of P and Kapplied in mulch constituted a special study wit-hin these experiments.

Vegetables with a different demand for nutrients,red beet with a moderate and white cabbage witha high demand, were grown from 1998 to 2001 ona soil with low to moderate concentrations of P-AL, K-AL and acid-soluble K (K-HNO3 minus K-AL). Each year, the effect of mulch applicationson vegetable yields was studied (Paper III). Oneor two applications of 9-10 t DM of chopped redclover or grass material per hectare were used.Residual effects on subsequent spring barleycrops were measured in 1999-2002. Yield levels

and contents of N, P and K in marketable pro-ducts, crop residues and mulch material, wererecorded each year. Residual levels of mineral Nin the soil were measured each year after harvest.Soil concentrations of P and K were measuredpreliminarily in 1998 and in more detail in 1999.The mulch treatments tested each year weremodified, according to experience gained in theprevious year. At harvest, the weights of marke-table products and crop residues were recordedand samples of both were taken for analysis ofDM content and N-, P- and K-concentrations.Only the marketable products were removed fromthe field. Leaves and stems were incorporatedinto the soil by rotovating soon after harvest, andthe plots were ploughed in the following spring.

The nutrient management was evaluated by appa-rent recovery of nutrients, which is the amount ofN, P or K in above ground plant material in tre-atments receiving no mulch (control crop) sub-tracted from the amounts in treatments receivingmulch. The recovery was expressed as a percenta-ge of the amount applied. The recovery may beunderestimated if the control crop takes up morenutrients from soil that the mulched crop. On theother hand, there is also a risk that the recoveryis overestimated, because the mulched crop infact takes up more nutrients from the soil thanthe control crop, because the mulched crop growsfaster. However, not subtracting the nutrientstaken up in the control crop would imply that allnutrients in mulched crops originated from themulch, which is obviously not the case. It must bekept in mind that the apparent recovery will beinfluenced by the soil fertility. In fertile soil, theapparent recoveries will be relatively larger dueto a higher nutrient uptake in the control crop.

Effect on yields of vegetables and subsequent bar-ley, and the recovery of mulch N, P and K

One mulch application of about 10 tonne of drymatter (DM) ha-1 increased the yields of bothcrops significantly. On average, the saleableyields were increased from 27 to 33 tonne freshweight (FW) ha-1 of red beet, and from 44 to 56tonne of white cabbage. For conventional farmingin this district, 30-40 tonne of red beet and 60-80tonne of cabbage per hectare are typical yields, sothe mean yield levels obtained with mulch seemsatisfactory. However, the average apparent reco-veries of mulch derived nutrients in abovegroundplant parts were only 13%, 14% and 18% of N, Pand K, respectively, as was reflected by largeamounts of surplus P and K in the nutrient bud-gets shown in Table 1. Some 3-10% of the N sup-plied in mulch was found as mineral N at 0-60 cmsoil depth after harvest, and subsequent barleyyields were increased by on average 620 kg ha-1,

8

or 20%. In spite of the effect on subsequent cere-als, the low N use efficiency indicates that muchN was lost by NH3 volatilisation, or by denitrifi-cation. Such nutrient losses represent a seriouschallenge for organic farming systems, aiming atbeing environmentally friendly. It should not beaccepted that harmless N2 is biologically fixed toplant protein, and later transferred to gases thatare harmful with respect to climate change (N2O)or nutrient enrichment of the atmosphere andprecipitation (NH3). Efforts must be taken toincrease the utilisation of biologically fixated Nwithin organic farming systems.

The yield increases varied much from year toyear (Paper III). By the lowest yield level thatwas obtained in the control treatment, 34 tonneha-1 of white cabbage (experimental year 2000),the yield increased to 42 tonne ha-1 by one appli-cation of chopped clover material. By the highestyield level that was achieved, in 2001, the cabba-ge yield increase was from 52 to 64 tonne ha-1.A large variation in yield levels without mulchapplication was to be expected, because the yieldsare then dependent on the release of nutrients bymineralisation of organic matter and weatheringof soil minerals. Both these processes are influen-ced by the weather conditions. It might have beenexpected that fertilisation by mulch applicationwould stabilise the yield levels, but on the contra-ry, the yield increase was proportional to the con-trol yield level, so that the effect of mulch waslargest in the years when the yield level was gene-rally highest. This result probably reflects thatthe N availability from mulch is at least as vari-able as is the soil N-mineralisation. The develop-mental stage of the plants used for mulching alsohas a large impact on the rate of decomposition.The recovery of nutrients was greatest in the caseof mulch material that was broken down slowly,but still had a comparatively high concentrationof N (cocksfoot; Dactylis glomerata L.).

Nutrient budgets reflected in changes in soil Pand K concentrations

Without mulch application, there was no statisti-cally significant change in the topsoil P- or K-concentrations from spring to late autumn (Table5, Paper III). For comparison with the surplusamounts of P and K applied in mulch, the con-centrations have been expressed on an area basis,multiplying the concentrations by the soil volumeand an average bulk density of 1.45 kg dm-3 aftercorrecting for an average gravel content of 15%.In mulched treatments, the increase in topsoil P-AL amounted to 25 kg P ha-1 in red beet and 15 inwhite cabbage, which is somewhat less than halfof the surplus P applied in mulch (Table 1). Withthe larger surplus in red beet, the fraction of Pthat could be accounted for as soil P-AL wasslightly larger, 41% as compared to 34% in cab-bage. It was interesting to see that the P taken upin saleable products in the control treatments wit-hout mulch did not cause significant decreases inthe soil P-AL concentrations. This indicates thatsoil P-AL was replenished during the growing sea-son.

For K, the increases in topsoil K-AL and acid-soluble K were slightly larger than the surplusadded in mulch. The K-AL increases correspon-ded to 209 kg K ha-1 in red beet and 174 in cabba-ge, and the acid-soluble K increases to 252 kg Kha-1 in red beet and 199 in cabbage. Hence, thetotal increase in extracted soil K (K-AL plus acid-soluble K) was larger than the K surplus for bothcrops. The amount of K not accounted foramounted to 139 kg K ha-1 for red beet and 81 forcabbage. The amounts of K taken up by plants inthe control plot (Table 1) were considerably lar-ger than the decreases in extracted soil K in theseplots, which amounted to 40 kg K ha-1 for red beetand 87 for cabbage. Hence, the experimental soilhad a capacity to replenish some of the

9

Red beet White cabbage

No mulch Two mulches No mulch Two mulches

P BUDGETP in mulch 0 71 0 71P removed 8 10 19 27Surplus of P -8 61 -19 44

K BUDGETK in mulch 0 490 0 490K removed 90 168 135 198Surplus of K -90 322 -135 292

Table 1. Field level nutrient budget for P and K (kg ha-1) by two mulch applications to vegetables in 1999. P and Kremoved is the uptake of nutrients in plant parts removed from the field.

K taken up by plants. In part, the unexpectedlyhigh soil K-concentrations in autumn in the mulchtreatments may be due to analytical inaccuracy,but it may also be questioned if the mulching hasstimulated soil biological processes and thereby atransformation of mineral K to K extractable withAL-solution or nitric acid.

In the subsoil, no changes in P or K concentrati-ons were found during the season. As the subsoillayer measured here was rather thick (25-60 cm),the P-AL values varied more at this depth than inthe topsoil, and because the surplus of P was lowas compared to that of K, possible increases insubsoil P concentrations may have been obscu-red. However, the unchanged subsoil K valuessuggest that there was probably no transport ofthis nutrient from topsoil, and as K is more mobi-le in soil than P, this indicates that the nutrientsapplied in mulch were stored in the topsoil. Thesurplus P that could not be accounted for as P-AL may have been bound in soil faunal biomass(and hence was not extracted by AL-solution), orit may have been adsorbed to soil minerals in aform not extractable by AL-solution. The surplusrequired to increase the P-AL concentration by 1mg kg-1 of topsoil, was 7.6 kg for red beet and 11for cabbage. These values correspond well withvalues found by Øgaard (1995) in a long-termexperiment with mineral P-fertiliser in a soil withan initial P-AL value of 50 mg P kg-1. In thatstudy, a surplus of 6.5 kg P ha-1 increased the P-AL concentration by 1 mg kg-1 topsoil (0-20 cm).For a 25 cm layer, the amount of P surplusrequired to increase P-AL by 1 mg would havebeen approximately 8 kg, which is comparable tothe results found in the present study.

Sustainability of the P and K management bychopped plant mulch vegetable growing

The main arguments in the disfavour of usingplant material as a fertiliser are a low recovery ofnutrients, as well as the negative environmentalimpacts of the gaseous N-losses. If the P and Knot recovered in crop yields are lost to waterbodies, the P may increase algae growth and bothP and K losses will reduce the soil fertility of thefarming system. However, losses of P and K may

not be very likely. No indications of losses werefound in the present study. It is possible that los-ses may occur on light textured soils with less abi-lity to absorb excess nutrients, especially by hig-her levels of soil P and K concentrations than wasfound in the present study. However, in generalthe mulch will prevent soil erosion and runoffboth directly because of the soil cover and indi-rectly because of a better soil structure and incre-ased infiltration rate. Hence, it is possible that Pand K losses may be reduced as compared to baresoil. The mulch application will increase the fer-tility of the field(s) used for vegetables, as wasshown by the significant increases in subsequentcereal yields (Paper III). If large areas are avai-lable that are well suited for mulch productionbecause there is no better alternative use of thatland (e.g. former pasture), the mulch system maybe an analogue to the former farming systemsbased on nutrient import from outlying fields.The system may be of interest for farmers wherethe farmland size is relatively large in comparisonto the amount of work that is available per gro-wing season. Cultivated land has to be croppedeach year, and selling roughage is not an alterna-tive for organic plant production systems becausethis would cause too large nutrient deficits.Producing nutrient inputs for redistribution wit-hin the farm requires little labour and invest-ments as compared to establishing an animal hus-bandry production.

The gaseous losses of N from vegetable manuremay be reduced by a dressing of soil (Janzen &McGinn, 1991) or chopped wood (Larsson, 1997).Such amendments require extra work, and a soildressing will decrease the weed control effect ofthe mulch. Nevertheless, possibilities of decrea-sing gaseous N-losses should be further studied.They warrant special attention since organic far-ming aims to be environmentally sound.Amendments that increase the N recovery fromchopped plant mulch may also be used to increasethe recovery from other green manure plants. Inconclusion, if the utilisation efficiency for N canbe increased, the chopped plant material mulchsystem seems to be a sustainable method to redis-tribute plant nutrients, ensure an agronomicallysound crop rotation and produce cash crops wit-hin stockless organic farming.

10

Plant adaptations to limited P and K availability in soil

When searching the web for the term ”plantadaptations”, 58 858 pages were found. On ofthese, from The Missouri Botanical Garden,explained the meaning of plant adaptations asThe special characteristics that enable plantsand animals to be successful in a particular environment. Plant adaptations to extreme clima-tic conditions, calcifuge and acidifuge plants andadaptation to high salt concentrations in soil weredescribed; all of which may be studied in wildspecies as a part of the general botany. However,crop plant adaptations to low nutrient concentra-tions in soil remains a topic of agronomic interest.Many studies are published within this field,including such various topics as the impact ofmycorrhiza, root exudations and uptake mecha-nisms at the cell level. In this thesis I have focus-sed on traits related to root morphology. No otheradaptations to limited nutrient concentrations willbe discussed; however, an overview has been pro-vided in Løes (1999). The presentation is restric-ted to spring wheat and barley, which are impor-tant crops in organic farming systems in NorthernEurope.

The interest of plant adaptations inorganic farming systems

Yield reductions in organic production due tolimitations in nutrient supply

Because of the aim of self-sufficiency, and restric-tions posed on purchase of nutrients, the nutrientsupply is generally limited in organic farming sys-tems as compared to in conventional farming. Ina Danish study comparing 36 pairs oforganic/conventional mixed dairy farms, the useof fertilisers and manure was considerably higheron the conventional farms (Halberg andKristensen, 1997). On average 31 tonne of animalmanure plus 148 kg N ha-1 y-1 was applied on aconventional field, as compared to 21 tonne ofanimal manure ha-1 y-1 on an organic. Most proba-bly, some of the conventional fertiliser was com-pound and hence increased the input of P and Kin the conventional system. The yield level was21-37% lower in organically produced cereals and12-18% lower in fodder beets and grass-clovercrops. Nutrient availability was suggested as animportant reason for these differences.Comparable yield level differences were found inNorwegian studies (Kerner, 1994; Ebbesvik,1997), and a positive relation was found betweenorganic ley yields and the amount of N applied inmanure. In a farming systems comparison study,the organic cereal yields were 60% of the conven-

tional, on average for 2000-2002, 3.33 t ha-1 y-1

(Korsæth et al, 2003). No nutrients except frombiologically fixed N were applied to the organicsystem, whereas in the conventional plant pro-duction treatment, 31 kg P, 88 kg K and 123 kg Nha-1 y-1 were applied in mineral fertiliser on avera-ge for the four-year crop rotation (Korsæth et al.,2001). These results demonstrate that in bothmixed animal and cash crop organic farming sys-tems, plant growth is restricted as compared to inconventional systems, and that one importantreason for the yield difference is the nutrientavailability.

Because of the lower level of P and K inputs toorganic farming systems, such systems will berelatively more dependent on the soil’s ability torelease phosphate and potassium ions by minera-lisation of organic matter and weathering of mine-rals. These processes may be stimulated by a rela-tively better soil structure (Mäder et al., 2002)and a higher biological activity in soil (Oberson etal., 1996) in organic farming systems. However,in general the plant growth will relatively often belimited by the nutrient supply in organic farmingsystems as compared to conventional.

Plant growth retardation by nutrient deficiency

During vegetative growth there is a large demandfor N in meristematic tissues where cell divisionoccurs, because of synthesis of proteins andnucleic acids. Hence, the extent of vegetativeplant growth is largely determined by N-availabi-lity (Mengel and Kirkby, 2001). By limited N-availability, the initial response will be a decrea-sed leaf elongation rate (Sinclair and Vadez,2002). Photosynthate will be accumulated asstarch and fructans in leaves and stems whereasthe protein content will be depressed. The sameprocesses will occur in the plant if P or K availa-bility is retarding the vegetative growth (Mengeland Kirkby, 2001). P is essential for the synthesisof nucleic acids, and K acts both as a stimulatorof plant enzymes and a regulator of osmotic pres-sure, which is important during cell elongation.As a consequence, plant growth may be signifi-cantly reduced because of unbalanced or general-ly limited nutrient availability without that anysymptoms of nutrient deficiency can be observed.This retardation in plant growth may be seen as aprimary plant adaptation to limited availability ofP and K.

Which macronutrient that limits growth in orga-nic farming systems will vary according to thecrop as well as to local soil and climatic conditi-ons. When N-mineralisation and biological N-fixation is high, the availability of P or K maybecome the limiting factor of growth. However, ina Nordic climate, both N-mineralisation and fixa-tion will commonly be restricted by climatic con

11

ditions. A significantly lower rate of N-fixationwas found in red clover grown in NorthernNorway as compared to the southern part of thecountry; 128 as compared to 210 kg N ha-1 y-1,respectively (Nesheim and Øyen, 1994). Hence, itis well possible that N-availability is limiting plantgrowth in Norwegian organic farming systems to alarger extent than in countries with a more favou-rable climate. There may be a potential to increa-se the biological N-fixation, e.g. by more efficientRhizobium strains, or crop cultivars adapted tocold climate. A variety of lucerne with higher N-fixation capacity has been released (Barnes et al.,1988; referred in Sinclair and Valdez, 2002).Recycling of human urine may also have a largepotential to reduce N-deficiencies in future orga-nic farming. Hence, plant adaptations to limitedP and K availability in soil are of interest in orga-nic farming systems in spite of that N may oftenbe the most limiting nutrient.

Root morphological traits increasing the nutrientuptake

Root architecture is of major importance forplant nutrient uptake. By increased root densityin soil, a larger interface between soil minerals,organic matter particles and plant roots is develo-ped. By deeper rooting, a larger soil volume willbe exploited. Root hairs enhance the nutrient-absorbing surface of the root manifold. A relati-vely large root mass, with relatively thin rootsthat spread their way through a large soil volume,would be a useful plant adaptation to limitednutrient availability. This may be achieved by

allocation of a relatively larger amount of photo-synthate to root growth. The root DM/shoot DMratio has been widely used as a measurement ofthis allocation. Alternatively, the root length per gtotal or above ground plant matter may be used(Nielsen, 1983), commonly abbreviated L*. Thistrait is less dependent on the developmental stageof the plant than the root/shoot ratio. Withrespect to spring wheat and barley, significantgenetic variability in root length, rooting densityand root hair length (RHL) has been found inmany former studies, referred in Paper IV and V.

Studies of the specific root length

In less fertile soil, where the subsoil also containsnutrients, a high root density is essential for thenutrient uptake and the root length is probablythe most important root trait with such conditi-ons. Measuring the root length requires muchwork, and hence is often avoided in cereal bree-ding. However, the specific root length (SRL, m g-1 root DM) did not vary significantly among thecereal accessions studied in this thesis (Paper IV).As shown in Figure 3, the root length is closelyrelated to the root weight. Hence, root lengthscan be estimated with sufficient reliability by theroot weight and an appropriate SRL value. Itshould be observed that this value will be lowerfor roots grown in soil as compared to roots fromhydroponics (Paper IV), because some soil parti-cles will stick to the root surface even after clea-ning. Calculating the root length instead of mea-suring it, may facilitate further root studies incereals.

12

y = 0,20x + 0,73

R2 = 0,56

y = 0,32x - 0,19

R2 = 0,74

0

2

4

6

8

10

12

0 10 20 30 40 50

Root weight, mg

Wheat

Barley

Roo

t le

ng

th, m

Figure 3. Relation between dry weight andlength of scanned rootsamples of 17 wheat and35 barley accessionsgrown in nutrientsolution with sub-

optimal P availability.

Relations between root traits and nutri-ent uptake in field for selectedNorwegian accessions of spring wheatand barley

A study of winter wheat cultivars released in theperiod 1969-1988 (Foulkes et al., 1998) showedthat the oldest of these cultivars had a relativelyhigher ability to take up N from soil without Nsupply. Opposite to this, the most recent cultivarshad a relatively higher uptake of N applied in fer-tiliser. This demonstrates that cereal cultivarsmay in fact gradually become adapted to higherlevels of fertilisers, and possibility exists that thismay hamper root traits essential for nutrientuptake from less fertile soils. To avoid such adap-tations, the selections of cereal cultivars in fieldshould be done with modest supplies of nutrientsto the soil.

As root traits are especially tedious to measure inthe field, screening of root traits with controlledconditions in nutrient solution could simplify thebreeding of nutrient efficient cultivars.Bertholdsson (2000) found a close relation betwe-en seminal root length and nutrient efficiency infield when growing seedlings in hydroponics withoxygen stress. On this background, we decided totest whether genetic variation in some root mor-phological traits of importance for nutrient upta-ke could be found in Norwegian accessions ofspring wheat and barley. Further, we wanted tostudy if such differences measured in controlledenvironment could be verified by studies of nutri-ent uptake and grain yield in the field.

Experimental design

52 accessions of spring wheat and barley wereselected to represent the cultivars bred and grownin Norway during the period 1900-2000. TwoNorwegian wheat accessions, and five Swedishbarley accessions selected for organic farmingwere included. On the basis of root traits recor-ded in circulation low-P nutrient solution, 20accessions (9 of wheat and 11 of barley) were cho-sen for comparing nutrient uptake and grainyields in field by limited nutrient supply and pes-ticide use. The accessions with maximum or mini-mum values for L* or RHL in nutrient solutionwere chosen, to be able to reveal differences inthe field. The selection of 20 accessions includedold land races and accessions released in themiddle of the century, but the proportion of notyet released accessions was larger than it was inthe selection of 52 accessions. The above groundDM (ADM) and uptake of N, P and K was recor-ded at 2 weeks interval from 18 June to 16

August, and by the final harvest in earlySeptember. As all the 52 accessions were grownby optimal nutrient supply prior to the nutrientsolution experiment, grain yields could be compa-red by different nutrient supply. Further detailson cereal accessions, nutrient supply, chemicalanalyses etc. are shown in Papers IV and V.

Root traits as related to nutrient uptake in the field

Significant genetic variation in L* was found inbarley, but not in wheat (Paper V). For bothcereal species, significant genetic variation wasfound in RHL both in nutrient solution and fieldby low nutrient supply. However, there was noclose relationship between L*, RHL in nutrientsolution or RHL in field, and the nutrient uptakein field. The nutrient uptake in ADM during thegrowing season varied significantly between acces-sions, but the accessions that most often had thehighest uptake of N, P and K were not the sameaccessions that produced the highest grain yieldat the final harvest. At each sampling, the uptakeof N, P and K in ADM were closely related.

Grain yield levels and the usefulness of modernaccessions in organic farming systems

By optimal supply, barley produced higher grainyields than wheat, on average 5.6 as compared to4.6 t ha-1. By limited supply, the average wheatand barley yields were comparable; 3.6 and 3.8 tha-1, respectively. The ranking between accessionswith respect to grain yield was comparable withoptimal and limited nutrient supply. The highestgrain yields were achieved by modern accessionsthat were fairly resistant to the dominating fungaldiseases. In wheat, this was mildew (Blumeriagraminis (Oudem.) J.J. Davis), and in barley, itwas scald (Rhynchosporium secalis (Oudem.)J.J.Davis). A higher harvest index contributednotably to the increased grain yields for modernaccessions. Modern accessions, especially linesselected for organic farming, performed best ofthe accessions studied here. However, some olderaccessions possessed interesting traits e.g. resis-tance to mildew (wheat cv. Snøgg), or a highnutrient uptake during the season (wheat cv.Møystad, barley cv. Lise) that may be useful infurther breeding. The study indicates that inNorwegian accessions of spring wheat and barley,genetic differences in nutrient uptake are present,however, these differences will be of little signifi-cance for the grain yields that can be achieved inorganic farming systems as compared to otherfactors.

13

The relevance of low-P studies in con-trolled environment for organic far-ming systems

Several studies have shown that plants respond toa decreased nutrient availability by producingrelatively more roots. The root/shoot ratio wasfound to increase for different plant species byreduced P-availability, especially in wheat andryegrass (Föhse et al., 1988). In a laboratoryexperiment comparing barley varieties, Römerand Schenk (1998) found longer roots by low P-conditions, on average 40 m roots per plant, ascompared to 25 m by optimum P-fertilisation. Inthe same study, the root mass increased from onaverage 112 mg DM per plant for optimum P con-ditions, to 165 mg by low P, after 20 days ofgrowth in a climate chamber (Schenk, 1992). Theaverage total plant mass was equal, 412 mg DMper plant by both P levels. This indicates that theplants invested relatively more of their dry matterproduction in root growth by low P-availability.

However, these results do not imply that by arelatively lower nutrient concentration in field,the root mass will generally increase. As descri-bed above, the general response to deficiency inmacronutrients is decreased growth. Mollier andPellerin (1999) found a large and rapid decreasein leaf expansion in maize when establishing Pdeficiency. After a temporary short-term increa-se, the root growth was also significantly decrea-sed. The explanation was that with the rapiddecrease in leaf expansion, excess carbohydrateswere for a while available for increased rootgrowth. Thereafter, with reduced leaf area theamount of carbohydrates fixated by photosynthe-sis decreased, causing a significant decrease in

root growth. This study supports the view thatwith decreased nutrient supply, the total amountof plant material and thereby also the root masswill decrease. For wheat and barley accessionsgrown in field with different P-concentrations insoil achieved by long-term differences in P appli-cation (Holten, 2002), the average ADM increasedwith increasing soil P-concentrations up to thesecond highest level, where it flared (Table 2).This result was to be expected. The root mass,which was recorded indirectly by measurement ofroot in-growth in buried soil bags, increased byincreasing P-concentrations, even to a largerextent than the ADM production (Table 2). Thisis contradictory to the results of Föhse et al.(1988) as well as Schenk (1992). The reason forthe disagreement is probably different soil P-con-centrations. In the subsoil used by Schenk (1992),the initial concentration corresponded1 to < 5 mgP-AL kg-1soil. It was enriched with 13.6 mg P kg-1

in the low P treatment and 136 mg in the opti-mum P treatment, and hence the P-availabilitywas probably more limiting in the low-P treat-ment of Schenk (1992) than in the study of Holten(2002), see Table 2. In the subsoil used by Föhseet al. (1988), the P-CAL concentration “transfor-med” to P-AL1 varied from 6.5 to 82 mg P kg-1 soilfor the levels where shoot DM increased with Pfertilisation. It was especially from the concentra-tion levels corresponding to P-AL= 6.5 to P-AL =23 that the root/shoot ratios were found to decrease significantly, whereas from P-AL =23 toP-AL = 82, the ratio decreased only slightly forwheat and was in fact increased for ryegrass(Föhse et al. 1988). For a Carex-species, Powell(1974) found a constant root/shoot ratio by increasing soil P-concentrations from a value corresponding to 19 mg P-AL kg-1 soil, to a valueof 116 (Figure 4).

14

P-application2 Soil P-AL ADM IGR DM/ADM

kg ha-1 y-1 mg P-AL kg-1 soil g m-2 g * 1000 per g

0 35 354 0.2616 49 476 0.3732 87 604 0.4048 130 587 0.51

Table 2. Average values of aboveground dry matter (ADM) yields in mid-July for two accessions of spring wheat andfour of barley, and DM of in-growth of roots (IGR) relative to ADM per m2. Data recorded at ear emergence/anthesisgrowth stage. (Holten, 2002)

1) The soil P-concentration (Schenk 1992) was < 10 mg P2O5 extracted by calcium-acetate-lactate (CAL). According to Sibbesenand Sharpley (1997), one unit of CAL-soluble P corresponds to 1.8 units of P-AL, whereas one unit of Olsen-P corresponds to3.2 units of P-AL

2) Applied each year since 1966

15

19 23 30 52 116

Low P-concentrations in soil have also beenfound to cause relatively longer root hairs incereals (Gahoonia et al., 1999). Pictures sho-wing diminishing root hair lengths in a soil withP-concentration corresponding to 149 mg P-ALkg-1 soil, whereas with 45 mg P-AL kg-1 the roothairs thrived, contribute to the impression thatplants compensate lower P-concentrations in soilby increasing root or root hair growth.However, for the cereal accessions studied byHolten (2002), the root hairs were significantlylonger with 87 as compared to 35 mg P-AL kg-1

soil; on average for barley, 0.93 as compared to0.83 mm, respectively. This result is in line withthe fact that decreased nutrient supply causes ageneral reduction in growth.

In conclusion, the detailed experimental studiesof plant adaptations to low nutrient concentrati-ons in soil or hydroponics must be interpretedwith much care when used to generalise knowled-ge about plant growth in field. Increased rootgrowth or similar responses to decreased nutrientconcentrations should not be regarded as if suchplant adaptations may have a significant impactin organic farming systems with limited nutrientsupply. However, this somewhat negative conclu-sion should not be seen as a contradiction withacknowledging the importance of a satisfactorysoil structure and a high root density in all thefertile soil depths in organic farming. As mentio-ned in the introduction, these factors are essentialfor organic crop production.

Figure 4. Growth of root, shoot and the whole plant of Carex coriacea at soil P-concentrations corresponding to19, 23, 30, 52 and 116 mg P-AL kg-1 soil, assuming that the bulk density of the silt loam used in the pot experimentwas 1 kg dm-3 soil. Adapted from Powell (1974).

Acknowledgements and personal comments

The aim of the present thesis was to assess thesustainability of organic farming systems withregard to soil P- and K-concentrations, and tostudy some plant adaptations to low nutrient con-centrations. The close relation between a growingcrop plant and the soil surrounding its roots hasbeen a fascinating object for agronomy resear-chers for centuries, and the relevant literature istoo extensive for a single person to digest. Evenso, much is still obscure about how plants acquirethe nutrients they need to grow and develop. Atthe start of the work, I thought that after doingit, I would know this. Now, approaching the endof the work, I realise that I still do not know real-ly what happens when the nutrients change theirstate of condition from being a part of the soil, tobeing a part of the plant. However, I have lear-ned how complicated it is, how many processesthat are involved, and how many possibilitiesexist for plant adaptation to conditions of sub-optimal nutrient availability. Out of this multipli-city, I have focussed on root morphological cha-racteristics, which comprised working with pictu-res that contained much qualitative informationin addition to the numbers that could be derived.

As the planning of the strategic programme star-ted early in 1997, I devoted a period of more than6 years to produce this thesis. During this period,my small children turned into teen-agers, and myfamily, used to living on a shore with rabbits inthe garden, experienced that survival was alsowell possible in a large city like Copenhagen. Thisbig adventure in our lives happened when I stay-ed one year at the Plant nutrition and soil fertili-ty laboratory in the Department of agriculturalsciences at the Royal veterinary and agriculturaluniversity in Denmark. The Copenhagen periodwas very instructive and fruitful also from a sci-entific point of view.

I am grateful to a very large amount of kind andskilled people that have contributed in variousways so that my studies would give valid andinteresting results. Out of these I especiallyremember four personal comments, which havebeen important “maximes” during this work. Thefirst came from Ragnar Eltun, who stressed theimportance of achieving experience on variouslevels (farm level, field experiment and laborato-ry studies) within a PhD-study. The second camefrom Olav Arne Bævre, who stressed the impor-tance of growing healthy plants in controlled

experiments. For instance, they should not bedeficient in iron when the aim was to study Puptake. Tara S. Gahoonia provided the third bytelling me not to waste time on unnecessarydetails; indeed a good advice for a stubbornCapricorn (me). Finally, Niels Erik Nielsen ensu-red me that I would always look back at the peri-od of the PhD-study as the best in my life, and Ithink he is right. The confusing world crowdedwith different projects to be followed up has alre-ady invaded my office, my computer and mythoughts and the comfortable days when the PhD-thesis was the one and only thing that mattered tome (at least during working hours) is gone.

I want to thank all of you who have participated,in so many important ways. Without your contri-bution, this thesis would not have been a realitynow. In addition to those already mentioned, andmy dear co-authors Anne Falk Øgaard, Tara S.Gahoonia, Sissel Hansen, Hugh Riley, MauritzÅssveen and Ellen Mosleth Færgestad, I want tothank

– the friendly and hospitable farmers contribu-ting in the farm level case studies– the colleagues participating in soil sampling, byno matter what kind of weather– all people doing their best to take care of fieldexperiments and chemical analyses– Jon Magne Holten for engaged and high-qualitywork in his M Sc thesis, which has complementedmy own thesis in a very valuable way– all colleagues at NORSØK for their encourage-ment and interest– many friends in other institutions who alwayswere ready to discuss– my supervisor, Tore Krogstad for being helpfuland open-minded– my children Linde, Anje and Mathias, and myhusband Morten, for lots of practical assistance,and for being brave enough to move abroad, andpatient enough to still be at home when I am finis-hed– and the rest of my family for being helpful, andfor looking forward to celebrate this thesis foralmost a decade.

Financial resources have been provided from theNorwegian Research Council, from the Nordicacademy for advanced study (NorFa) and fromNorwegian Centre for Ecological Agriculture.

Tingvoll, July 2003Anne-Kristin Løes

16

References

Barnes, D.K., Sheaffer, C.C., Heichel, G.H., Smith, D.M.and Peaden, R.N. (1988) Registration of “Nitro” alfalfa.Crop Science 28, p 718.

Bertholdsson, N.O. (2000) Ökad stresstolerans- nyckeln tillen hög skörd och ett renare svenskt jordbruk (Increasedstress tolerance- a key to higher yields and less use of chemi-cals in agriculture). Sveriges Utsädesförenings Tidskrift(Journal of the Swedish Seed Association) 4, p 214-225(abstract in English).

Debio (Norwegian control and certification body fororganic agricultural production) (2003)http://www.debio.no/diverse/debstat

Ebbesvik, M. (1997) Økologisk eng. Viktige faktorer foravlingsnivå. Norsk senter for økologisk landbruk, rapport3-97, 44 s. (English summary).

Eltun, R. (2002) Organic cropping systems for higher andmore stable cereal yields. Proposal for a strategic instituteprogramme 2003 form Norwegian Research council (NFR),area “Bioproduksjon og foredling”, 21 p.

Eriksen, J., Hansen, J. F., Kyllingsbæk, A., Sibbesen, E.,Petersen, S. O., Djurhuus, J., Vinther, F. P., Heidmann,T., Olsen, P. and Hansen, B. (1995) Næringsstofbalancerpå markniveau i økologisk kvægbrug og planteavl. In J EOlesen and J Vester (eds): Næringsstofbalancer og energifor-brug i økologisk jordbrug - fokus på kvægbedrifter og plan-teavl, s 48-74. Statens planteavlsforsøg rapport 9,Landbrugs- og fiskeriministeriet, Lyngby, Denmark.

Föhse, D., Claassen, N. and Jungk, A. (1988) Phosphorusefficiency of plants. I. External and internal P requirementand P uptake efficiency of different plant species. Plant andSoil 110, p 101-109.

Foulkes, M.J., Sylvester-Bradley, R. and Scott, R.K.(1998) Evidence for differences between winter wheat culti-vars in aqcuisition of soil mineral nitrogen and uptake andutilization of applied fertilizer nitrogen. Journal of agricul-tural science, Cambridge 130, 29-44.

Gahoonia T.S., Nielsen, N.E. and Lyshede, O.B. (1999)Phosphorus (P) acquisition of cereal cultivars in the field atthree levels of P fertilization. Plant and Soil 211, 269-281.

Halberg, N. and Kristensen, I. S. (1997) Expected cropyield loss when converting to organic dairy farming inDenmark. Biological Agriculture and Horticulture 14, 25-41.

Hansen, P. H. (2000) 13 års økologisk drift uden gødning.Økologisk jordbrug, 25.02.2000, s 14.

Hansen, S. (1996) Effects of manure treatment and soilcompaction on plant production of a dairy farm system con-verting to organic farimg practice. Agriculture, Ecosystemsand Environment 56, p 173-186.

Hansen, S. and McKinnon, K. (1999) Økologisk jordkul-tur. Landbruksforlaget, Oslo. 344 s.

Holten, J.M. (2002) Phosphorus uptake in six selectedScandinavian wheat and barley cultivars at low soil phos-phorus availability as related to root hair length. Dep. ofSoil and Water Sciences, Agricultural Univ. of Norway, 52 p.

IFOAM (International Federation of Organic AgriculturalMovements) (2002) http://www.ifoam.org

Janzen, H.H. and McGinn, S.M. (1991) Volatile loss ofnitrogen during decomposition of legume green manure. SoilBiology & Biochemistry 23, p 291-297.

Kerner, K. N. (1993) Næringshusholdning i økologisk land-bruk. Del 1: Handelsbalanser for nitrogen, fosfor og kaliumpå gårdene i 30 bruks-prosjektet. Faginfo 20, Statens fagtje-neste for landbruket, Ås, s 1- 32.

Kerner, K. N. (1994) Korn-, potet- og grønnsakproduksjon iøkologisk landbruk. Norsk Landbruksforskning 8, s 177-187.

Korsæth, A., Eltun, R. and Nordheim, O. (2001)Dyrkingssystemet på Apelsvoll. Forandring av forsøkspla-nen for dyrkingssystemene på Apelsvoll. In: U. Abrahamsen(ed.): Jord- og plantekultur 2001. Planteforsk (NorwegianInstitute of Crop Science) 1-2002, s 316-323.

Korsæth, A., Eltun, R. and Gaardløs, T. (2003)Dyrkingssystemene på Apelsvoll. Avlings- og avrenningsre-sultater fra dyrkingssystemene på Apelsvoll 2000-02. In: U.Abrahamsen (ed.): Jord- og plantekultur 2003. Planteforsk(Norwegian Institute of Crop Science) 1-2002, s 274-285.

Krogstad, T. and Løvstad, Ø. (1987) Fosfor i jord og vann.Jord og Myr 6, s 189-207.

Larsson, L. (1997). Evaluation of mulching in organicallygrown black currant (Ribes nigrum) in terms of its effects on the crop and the environment.Doctoral Thesis, SLU, Alnarp, 33 pp.

Larsson, L., Ferm, M., Kasimir-Klemedtsson, Å., &Klemedtsson, L. (1998). Ammonia and nitrous oxide emissions from grass and alfalfa mulches.Nutrient Cycling in Agroecosystems 51, p 41-46.

Lieblein, G. and S.Ø. Solberg, 1996. Hvordan fungereregentlig jord? Et bidrag i prosessen for å forstå næringsba-lanser og økologisk jordbruk. Forskningsnytt om økologisklandbruk i Norden 2-96, s 12-13.

Løes, A.K. (1992) Aktuelle omleggingsproblemer – en spør-reundersøkelse. Norsk senter for økologisk landbruk,Tingvoll, rapport 9-92, 29 s.

Løes, A.K. (1999) Opptak og funksjon av fosfor og kalium iplanter- med eksempler på tilpasninger som kan øke oppta-ket fra jorda. Norsk senter for økologisk landbruk,Tingvoll, 34 s.

17

Løes, A.K. and A.F. Øgaard (1997) Changes in the nutrientcontent of agricultural soil on conversion to organic farming,in relation to farm level nutrient balances and soil contents ofclay and organic matter. Acta Agricultura ScandinavicaSection B, Soil and Plant Science 47, p 201-214.

Mäder, P., Fliessbach, A., Dubois, D., Gunst, L., Fried,P. and Niggli, U. (2002) Soil Fertility and Biodiversity inOrganic Farming. Science 296, p 1694-1697.

Mengel, K. and Kirkby, E. A. (2001) Principles of plantnutrition 5th ed. Kluwer Academic Publishers,Dordrecht/Boston/London, 849 p.

Mengel, K. and Steffens, D. (1985) Potassium uptake ofrye-grass (Lolium perenne) and red clover (Trifolium pra-tense) as related to root parameters. Biology and Fertility ofSoils 1, p 53-58.

Ministry of Agriculture/ Det kongelige norske landbruksde-partement (2003) St.meld. nr 19 (1999-2000) Om norsklandbruk og matproduksjon, s 20. Also available athttp://odin.dep.no/ld/norsk/Ansvarsomraader/Okologisk-landbrukdok

Mollier, A. and Pellerin, S. (1999) Maize root systemgrowth and development as influenced by phosphorus defici-ency. Journal of Experimental Botany 50, p 487-497.

Nesheim, L. and Øyen, J. (1994) Nitrogen fixation by redclover (Trifolium pratense L.) grown in mixtures with timot-hy (Phleum pratense L.) at different levels of nitrogen fertil-ization. Acta Agricultura Scandinavica Section B-Soil andPlant Science 44, p28-34.

Nielsen, N.E. (1983) Plant parameters controlling the effici-ency of nutrient uptake from soil. In: Efficient use of fertili-zers in agriculture. Proceedings Symposium on Researchinto Agro-technical Methods aiming at increasing the pro-ductivity of crops. United Nations Economic Commission forEurope, Geneva, Switzerland, pp199-217.

Nolte, C. and Werner, W. (1994) Investigations on the nutri-ent cycle and its components of a biodynamically-managedfarm. Biological Agriculture and Horticulture 10, p 235-254.Oberson, A., Besson, J.M., Maire, H. and Sticher, H.(1996) Microbial processes in soil organic phosphorus trans-formations in conventional and biological cropping systems.Biology and Fertility of Soils 21, p 138-148.

Pfiffner, L., Mäder, P., Besson, J.-M. and Niggli, U. (1993)DOK-Versuch: vergleichende Langzeit-Untersuchungen in dendrei Anbausystemen bio-Dynamisch, organisch-biologisch undkonventionell. III Boden: Untersuchungen über dieRegenwurmpopulationen. Schweizerische LandwirtschaftlicheForschung 32(4), s 547-564.

Powell, C.L. (1974) Effect of P fertilizer on root morphologyand P uptake of Carex coriacea. Plant and Soil 41, p 661-667.

Römer, W. and Schenk, H. (1998) Influence of genotype onphosphate uptake and utilization efficiencies in spring bar-ley. European Journal of Agronomy 8, p 215-224.Salomon, E. (1999) Availability of potassium to clover andgrass from soils with different potassium fertilization histori-es. Doctoral thesis, Swedish University of AgriculturalSciences, Agraria 149, 180 p.

Schenk, H. (1992) Untersuchungen zur Phosphateffizienzvon Sommergerstensorten. Diplomarbeit, Institut fürAgrikulturchemie, Universität Göttingen, 92 s.

Sibbesen, E. and Sharpley, A.N. (1997) Setting and justifyingupper critical limits for phosphorus in soils. In: H. Tunney,O.T. Carton, P.C. Brookes and A.E. Johnston. (eds):Phosphorus loss from soil to water. CAB international, pp151-176.

Sinclair, T.R. and Vadez, V. (2002) Physiological traits forcrop yield improvement in low N and P environments. Plantand Soil 245, p 1-15.

Skøien, S. (1983) Alternativt jordbruk. NLVF-utredning 127(The Norwegian Agricultural Research Council), Oslo. 150 s.

Strand, E. 1997. Hvorfor 3000 nye økobønder i Norge?Bondebladet, 18.06.97, s 38.

Strand, E. 1999. Økologisk landbruk og sultproblemer.Aftenposten, 08.03.99, debattside.

Søgaard, C. 1997. Jordbehandling erstatter gødning.Forskningsnytt om økologisk landbruk i Norden 1-97, s 2-3.

Vartdal, B. (1993) Motivasjon og meistring av omlegging tiløkologisk jordbruk. Senter for bygdeforskning, rapport 3-93, based on a M.Sc. Thesis in sociology, University ofTrondheim, 182 p.

Watson, C. A., Bengtsson, H., Ebbesvik, M., Løes, A. K.,Myrbeck, Å., Salomon, E., Schroder, J. and Stockdale,E.A. (2002) A rewiew of farm-scale nutrient budgets fororganic farms as a tool for management of soil fertility. SoilUse and Management 18 supplement, Soil fertility in organi-cally managed soils, p 264-273.

Øgaard, A.F. (1995). Effect of phosphorus fertilization andcontent of plant-available phosphorus (P-AL) on algae-avai-lable phosphorus in soil. Acta Agriculturae Scandinavica Section B Soil and Plant Science 45, p 242-250.

18