soil mapping and soil monitoring: state of progress and...

EUROPEAN SOIL BUREAU RESEARCH REPORT NO. 6

Soil Mapping and Soil Monitoring in France. King, Stengel and Jamagne 63

Soil Mapping and Soil Monitoring: State ofProgress and Use in France

1King D.2Stengel P.1Jamagne M.

1 INRA, Departement Environment Et Agronomie, Unite de Science du Sol – SESCPF,Avenue de lla Pomme de Pin, BP20619 ARDON, 45166 Olivet Cedex, FRANCE

2 INRA, Direction Scientific Environnement Forêt et Agriculture, Domaine St Paul - SiteAgroparc, 84914 AVIGNON Cedex 9, FRANCE

IntroductionThe Service for the Study of Soils and thePedological Map of France (French acronym:SESCPF) was founded in 1968. Within the INRA(National Institute of Agronomic Research), itensures the coordination of most soil mapping andmonitoring programmes in France. This is madepossible by data received from a number of otherpublic or private organisations that are partners inthese programmes: research institutes (CNRS:National Centre for Scientific Research,Universities), professional developmentorganisations (Chambers of Agriculture, NationalForest Bureau), land development companies anddesign offices, etc. Work is carried out at therequest and with the support of the Agriculture andEnvironment Ministries, as well as localgovernments (Regional Councils).

France is composed of 56% arable land, 28%wooded surfaces, 8% of surfaces that are unusedor are destined to protect natural environments andfinally 8% surfaces that have been built upon(IFEN, 1999). Current trends indicate a loss ofland to building and infrastructure of close to40,000 hectares (ha) (100,000 acres) per year, theequivalent of one French Department every 10years. This trend is undoubtedly one of the primemenaces for the “soil” heritage. It is also anindicator of a large number of other pressures onsoils and may cause their irreversible degradation.

Soil mapping programmes are a valuable tool forland management. In spite of this, there are

currently significant reductions in the resourcesallocated to these programmes.


Soil Mapping and Soil Monitoring in France. King, Stengel and Jamagne64

At the same time, increasing demands are beingmade that are more diversified and morespecialised in terms of agricultural andenvironmental functions of soils. A systematicinventory of the country is no longer sufficient torespond to these demands. In addition there are amultiplicity of actions that must be coordinated atthe national scale in conjunction with Europeanprogrammes.

The aim of this paper is to describe several ofthese actions and briefly demonstrate their use inlong-term natural resource managementprogrammes. More detailed information can befound in the soil chapter of the last Frenchenvironmental report (IFEN, 1999) and in an in-depth review of research over the past 30 years(Bornand, 1997).

Soil MappingThe “Pedological Map of France” (CPF)programme has been in existence for more than 25years. It has led to the acquisition of a largequantity of data on the soil resources of thecountry and to a basic understanding of thetypology and spatial variability of the principal soilsystems (Jamagne et al., 1995).

This programme was the reason for the creation ofSESCPF, leading to the development of a nationalplatform for coordination and cooperation in thefield of spatial analysis of soil. By the end of 1998,27 maps at a scale of 1:100,000 had beenpublished and 13 additional maps were beingprepared, accounting for about 15% of the territory(Figure 1).



Figure 1: State of progress in 1999 of CPF and IGCS programmes.

The resources allocated to this programme,however, are not sufficient to achieve the totalcoverage of French territory in the medium term. Itwas thus decided to reorient the CPF programmeto give preference to the detailed formalisation ofsoil distribution laws within zones representativeof the main French pedological systems. Inessence, this means that old data will becomputerized and new acquisitions will be made inlittle prospected regions. This programme isintended as a scientific support for inventory

programmes carried out at broader scales includingthat described below.The “Soil Inventory, Management andPreservation” (IGCS) programme conducted bythe Agriculture Ministry and INRA since 1990aims to prepare a map with an associated databaseat a scale of 1:250,000 for each of the FrenchRegions (Bornand et al., 1989, Jamagne et al.,1995). Up-to-now, 3 regions have been completed,as well as 12 Departments, accounting for about40% of the country (Figure 1).



A test of data transfer to the European system isunder way in the Côte d’Or Department inBurgundy (Finke et al., 1998). A second aim ofthe programme is to carry out detailed studies, at ascale of 1:10,000, of small size sample landscapes.Monitoring the agricultural and environmentalfunctioning of these zones provides a referencebase that can be generalized and applied to similarsoil systems identified at a scale of 1:250,000(Favrot, 1987; Favrot and Lagacherie, 1993).

These two programmes do not cover all mappingactivities. Several regional or nationalorganizations have started mapping programmes atsmaller scales, e.g. 1:50,000 in the Centre Region,1:25,000 in the Aisne Department, typology offorest stations, regional typologies, etc. Inaddition, there are many local studies, but it isdifficult to estimate the extent of this information(Bornand, 1997).

Figure 2: Location of monitoring sites of the OQS and RENECOFOR programmes.



Finally, there is the synthesis map at a scale1:1,000,000 that has been revised in theframework of European projects (Jones et al.,1998, Le Bas et al., 1998). Even so, this work isinsufficient to meet current needs and the expectedcoverage at the 1:250,000 scale will lead to athorough revision of the 1:1,000,000 geographicdatabase.

Soil Monitoring

There are two main French soil-monitoringnetworks. The first deals with long-term changesin cultivated land and natural non-wooded spaces.The second involves forests.

The main purpose of the “Soil QualityObservatory” (OQS) is to assess the presentsituation of soils and monitor their changes inorder to improve on and implement a soilpreservation policy (Martin, 1993). Eleven siteseach of about 1 ha were chosen on the basis oftheir representative nature for soil and land use.There is a minimum set of parameters that aresystematically measured at all sites of the network.For some specific degradation problems,additional measurements may be conducted atsome sites, e.g. light fraction C and N,mineralisable C and N, microbial biomass, soilenzymes, earthworms. The recommended timestep is five years.

On one agricultural site, continuous pig slurryapplication has been shown to increasesignificantly the organic carbon content and that ofsome metals (Cu, Zn). In an acid brown soil undera mature spruce stand, a decrease of total elementsin the organic layers, and of the exchangeableelements in organo-mineral horizons has beenobserved, which could have severe consequencesfor future forest nutrition.

A decrease of the lead concentration in forest littermay be considered as an encouraging sign in thecampaign to render gasoline greener. Otherongoing work involves the definition of biologicalindicators and development of a sampling strategyfor soil microbial biomass and determination ofsoil fauna .

The RENECOFOR programme is a long-termmonitoring system for forest ecosystems (Figure2). It was created by the National Forest Bureau in1992 in order to extend the system for monitoringthe health status of forests (Ulrich, 1995). It is theFrench part of a set of permanent parcels installedin 34 European countries. It covers highly variedareas of forests.

Two soil profiles are systematically studied andfertility is monitored. With respect to the"Cataenat" sub-system (total acid load ofatmospheric origin), atmospheric deposits havebeen measured in 27 parcels since 1993, andmeasurements are conducted on soil solutionstaken from 20 and 70 cm depth from 17 parcels.

Other monitoring systems including soils havebeen implemented, but they are concerned withother components of the environment, especiallywater resources. In addition, databases have beencreated from results of old analyses of soilssampled separately from, or in the context of, localprogrammes (for example, see next section on theNational Base of Land Analyses, French acronym:BNAT). These data are localised in space and intime and provide information on medium- andlong-term changes. It has not yet been decided,however, whether to continue the systematiccollection of the data from the laboratoriesconcerned.

Soil DatabasesIn 1990, a unique database structure was adoptedfor all nationally integrated programmes. Thesystem is called DONESOL and includes threeparts (Gaultier et al., 1993):1. Point data obtained from observations and

measurements made on soil profiles.2. Descriptive data of soil mapping units, soil

typology units of soils and horizons.Additional data, provided by expert advice,take into account the spatial variability withinthese entities. Digitized contours are includedin a GIS linked with DONESOL.

3. Metadata that indicate bibliographicreferences of the studies as well as theirlocation and precision (Favrot, 1994). Overthe past several years, considerable effort hasbeen devoted to cataloguing all detailed scalepedological studies. This work is currentlybeing extended by studies at medium andlarge scales. A directory of soil mappingprofessionals has also been published (Favrot,1997).

Information about mapping and monitoringprogrammes can be found on the INRA OrleansWorld Wide Web site (http://www-sescpf.orleans.inra.fr/public/). One objective is todevelop this site particularly to include themetadata described above.

Other databases have been prepared in conjunctionwith, or in parallel to, the DONESOL database. Incontrast to the above-mentioned programmes,these databases have been compiled in the context



of targeted research projects, including ASPITET,SOLHYDRO and BNAT.ASPITET: In 1994, INRA a research programmebegan entitled "Implications of PedologicalStratification on the Interpretation of TraceElements Contents" (French acronym: ASPITET).The aim of this programme was to acquirereference data on the natural contents of traceelements in soils. The work entailed taking intoaccount soil types and geological parent materials.

The population of samples studied up to the end of1998 was 1310 pedological horizons,corresponding to 706 distinct sites. Total traceelement concentrations were determined for eachsample (systematically Cd, Cr, Cu, Mn, Ni, Pb, Zn,often As, Co, Hg, Se, Ti).

The results show that the general degree ofcontamination by human activities is very low incomparison to the stocks of natural trace elementsin soils.

Nevertheless, there are three types of exceptions incultivated lands:

(i) regions in which grapevines have beenplanted and those devoted to arboriculture,that are almost always contaminated withcopper;

(ii) parcels fertilised with sewage sludge havinga large heavy metal load;

(iii) parcels located in proximity to oreprocessing factories (affected byatmospheric pollution).

Figure 3: Organic matter content of the surface horizon in cultivated lands,extracted from the BNAT.



SOLHYDRO: A project was started in 1998 tocompile an analytical database of the hydraulicproperties of French soils. The purposes of thisdatabase are to:

(i) assemble all data on these propertiesdetermined by different researchlaboratories, as well as any additionalanalytical or descriptive data;

(ii) make these data available to researchers forcomparison of methods of measurement andfor preparation of pedotransfer functionsusing the range of French soils;

(iii) prepare nationally recognised references forusers.

There are links between this database, DONESOL,and the European HYPRES system (Wösten et al.,1998).

BNAT: More than 200,000 soil analyses arecarried out by private companies each year. Afeasibility study was conducted from 1990 to 1994to assess recovery of these data and determinewhether to include them in a database termed‘Base Nationale des Analyses de Terre’ (NationalBase of Land Analyses) (Walter et al., 1997,Schvartz et al., 1998).

The data have been collated by township and bydistricts to protect the rights of the owners of theanalyses. Each variable can be expressed in theform of maps or statistical tables and thesedocuments have led to the confirmation of spatialstructures that are often known but neverquantified, especially for variables that are difficultto determine in conventional mapping work(Figure 3). The value of this database is that itfacilitates the analysis of possible changes in soilproperties over time. The detection of thesechanges will require a longer time span.

The collection and structuring of analytical dataobtained by private parties is a valuable source forregional or national reviews. In addition to theBNAT, the AGREDE programme conducted byADEME and INRA recovered close to 12,000analyses of heavy metals obtained between 1992and 1997 relating to studies before fertilisationwith sewage sludge from water treatment stations.Maps currently being edited display the generalquality of agricultural soils and reveal zones ofanomalies, either natural or induced by localpollution.

Use of Existing Soil DataExamples of the use of the soil databases describedabove were selected on the basis of themes directlyinvolving the soil (erosion), agricultural

production (quality of products), or themanagement of natural resources (aquifers, rivers).The examples were also selected to demonstratenational or regional applications that show thevalue of the different databases in different ways.

Among the various threats to soil, erosion isclearly one of the most visible, in terms of bothagriculture (loss of land) and the environment(water pollution, damage to and by urban and roadinfrastructures). At the request of the EnvironmentMinistry, two national mapping programs werestarted.

The first was an inventory of mudflows based oninsurance claims filed for natural catastrophesbetween 1985 and 1995 (Figure 4). The secondwas a map of the “soil erosion” probability,produced by combining the various factorsresponsible for this erosion in a GIS (LeBissonnais et al., 1998): crusting and soilerodability determined from the 1:1,000,000database, plant cover taken from CORINE landcover, slopes and precipitation (quantity andintensity).

The combination of these inputs was carried outwith a simple empirical model using grids with astep of 250 m. The results were then collectedaccording to catchment or to small agriculturalregions and presented for each season (Figure 5).

The two documents facilitate inter-regionalcomparisons that define the nature and intensity ofthe erosion. These documents are compared toregional studies (King et al., 1998) or sent toregional authorities so that they can promotesensitivity campaigns and proposals for combatingthe problem.Trace element contents in soils result from thecombination of natural processes (heritage fromparent rock) and direct or indirect inputs resultingfrom human activities. Results of the ASPITETprogramme together with work done incooperation with the AGREDE programme haveidentified areas containing high natural levels ofmetals that are often higher than regulatorythresholds for sewage sludge fertilisation (Baize,1997).

For example, Figure 6 shows four zones in theSeine-et-Marne Department having abnormalconcentrations of Zn (higher than 80 mg/kg),corresponding to fertilisation sites close to fourrailway stations at which the composts wereunloaded (Baize and Paquereau, 1997). This typeof study clearly pinpoints zones in whichfertilising with sewage sludge should be controlledor prohibited.



Figure 4: Map of mudflows on the French territory (1985-1995).

The database resulting from the IGCS programmeis rarely used alone to express a soil property. Inmost cases soil information is combined with otherdatabases, in particular those of climate,topography and agriculture. Combinations arecreated using dynamic models that reproduce pastevents by simulation (agrometeorological models,hydrological models). The maps show risks relatedto a change in agricultural practices or to climaticchange, in the form of frequency analyses.

This is a topical area of research and the number ofstudies is high (MAPA, 1998): for example, risk ofnitrate leaching in Brittany (Saby et al., 1999),irrigation management in the Beauce (Cousin etal., 1998), vulnerability to fertilising, fertilisingwith urban composts (Legros et al., 1991), qualityof agriculture production ( Monnet and Gaiffe,1998), diversification of cultivation after removalof grapevines (Bornand et al., 1994).

In most of these studies, the soil has a pivotal rolein the results, but it is difficult to determine theweight given to it in the final decision. This resultsfrom the modelling methods that use otherenvironment variables and from sociological andpolitical factors that are necessarily present indecision-making processes. This observation hasalso been verified in relation to the use ofmonitoring networks. For example, the objectivesconcerning air pollution are to reduce theemissions of pollutants that affect ecosystems atlong distance. The concept of critical loads enablesthe effects of atmospheric fallout to be quantifiedin terms of forest decline and pH of streams.

In France, the ADEME has implemented and ledregional and national research to respond to theparticular needs of the Environment Ministry at thetime of revision of international protocols (e.g.Dambrine et al., 1998, Thomas et al., 1998). Thenetwork data used in this case are an essentialsupport for international comparisons.



Figure 5: Maps of seasonal erosion risk in France.

PerspectivesPerspectives are discussed in terms of threeaspects: soil mapping, soil monitoring and thedistribution of information.

In the field of mapping, surveying at the scale of1:250,000 is continuing in order to obtainexhaustive and integrated information for thewhole of France. This scale, however, isinsufficient to respond to all kinds of socialdemands that are highly varied and requireprecision. There are no national plans to allocateconsiderable resources to respond to this demandfor precise information.

On the other hand, it has been proposed toreinforce 1:100,000 existing studies by organisingthe results obtained in knowledge bases.

In this way, local governments and institutions,rural land managers or design offices can consultthe data and apply this knowledge to the areas withwhich they are concerned. In addition, there is anincrease in the needs for mapping at highlydetailed scales, i.e. down to the level ofagricultural plots.

This has resulted from progress in techniques ofspatial positioning and their potential to usealongside associated agricultural techniques(precision farming).



Figure 6: Map of estimated zinc concentrations for the Seine-et-MarneDepartment. [Interpolation by kriging.]

It is thus planned to reinforce research on thesenew techniques, in particular, the use ofgeophysics and digital elevation models. Finally,this multi-scale approach, from the plot to theregion, requires continuation of research into scaletransfer methods (e.g. Lagacherie et al., 1997).Concerning soil monitoring, French systems arejudged satisfactory for forestry but insufficient foragriculture. There is thus a need to develop a newsystem that includes two levels.

The first would be composed of a large number ofsites monitored with a time step of about 5 years.The aim of this would be to have an alert systemthat makes no prior judgments on pressures thatmay be exerted on soils within the coming years.The second would be applied to a limited numberof sites to follow fluxes at shorter time steps (day,month) and thereby be able to analyse and explainrecorded long-term changes.



In this context, the Environment Ministry hasstarted a research programme (GESSOL) whoseaim is to establish the scientific tools and bases toassess, monitor and even restore soil quality.

Finally, it is recognized that all current or futureprogrammes will be of interest only if the datagathered are distributed as widely as possible andin the most instructive manner possible. Thisimplies the continuation of research in the field ofcombining spatial data and in that of errorpropagation when different data sets are combined.In addition, there is need to develop methods forstructuring and distributing information, inparticular using modern computer technologies(Web, CD-ROM). It is also necessary to obtain

more information on user needs in order todevelop the tools that can most efficiently respondto these needs.

AcknowledgmentsThe authors thank all the people, partners in theprograms or members of their scientificcommittees for sending the information for thispublication, especially D. Arrouays, D. Baize, M.Bornand, J.C. Favrot, R. Hardy, P. Lagacherie, C.Le Bas, C. Schvartz and C. Walter.

ReferencesBaize D. (1997). Teneurs totales en éléments

traces métalliques dans les sols (France).Références and stratégies d'interprétation.INRA Editions, Paris. 410pp.

Baize, D. and Paquereau, H. (1997). Teneurstotales en éléments traces dans les solsagricoles de Seine and Marne. Etude andGestion des Sols. 4(2), 77-94.

Bornand, M., Arrouays, D., Baize, D. andJamagne, M. (1989). Méthodologie d’unecartographie régionale des sols à l'échelle du1/250 000. Science du Sol. 27 (1), 17-20.

Bornand, M., Legros, J.P.and Rouzet, C. (1994).Les banques régionales de données-sols.Exemple du Languedoc-Roussillon. Etude andGestion des Sols. 1, 67-82.

Bornand, M. (1997). Connaissance and suivi de laqualité des sols en France. MAPA-MATE-INRA. INRA Montpellier. 176pp.

Cousin,I., Vautier, A., Bruand, A., Duval, O. andNicoullaud, B. (1998). Recharge de la nappe deBeauce : influence du type de sol. Doc. Ronéo.INRA Science du Sol Orléans. 4pp.

Dambrine, E., Thomas, A.L., Party, J.P., Probst ,A., Boudot, J.P., Duc, M., Dupouey, J.L.,Gégout, J.C., Guérold, F., King, D., Landman,G., Maitat, O., Nicolaï, M., Pollier, B. andThimonier, A. (1998). Acidité des écosystèmesforestiers dans les Vosges gréseuses:distribution, évolution, rôle des dépôtsatmosphérique et conséquences biologiques.C.R. Acad. Agric. Fr. 84(5).,75-94.

Favrot, J.C. (1987). Etudes and recommandationspréalables au drainage: la methode desSecteurs de Reference. C.R. Acad. Agric. Fr.87-73 (4) 23-32.

Favrot J.C. and Lagacherie, Ph. (1993). Lacartographie automatisée des sols: une aide à lagestion ecologique des paysages ruraux.. C.R.Acad. Agric. Fr. 93-79 (5), 61-76.

Favrot, J.C. (1994). Note de synthèse sur le voletpédologique and sur le répertoire informatiséREFERSOLS. INRA Document, Montpellier.34pp.

Favrot, J.C. (1997). Répertoire des organismes andspécialistes intervenant en cartographie, enanalyse des sols and pédologie appliquée(France métropolitaine). INRA-MAPADocument. INRA, Montpellier. 88pp.

Finke, P., Hartwich, R., Dudal, R., Ibañez, J.,Jamagne, M., King, D., Montanarella, L. andYassoglou, N. (1998). Georeferenced SoilDatabase for Europe. Manual of procedures,version 1.0. EUR 18092 EN. Office of theOfficial Publications of the EuropeanCommunities, Luxembourg. 184pp.

Gaultier, J.P., Legros, J.P., Bornand, M., King, D.,Favrot, J.C. and Hardy, R. (1993).L'organisation et la gestion des donnéespédologiques spatialisées: le projetDONESOL. Revue de géomatique, 3, 235-253.

IFEN, (1999). L’environnement en France. Ladécouverte. Editions La découverte & Syros etIFEN. Paris and Orléans. 480 pp.

Jamagne, M., Hardy, R., King, D. and Bornand,M. (1995). La base de données géographiquedes sols de France. Etude and Gestion des Sols.2(3), 153-172.



Jones, R.J.A., Buckley, B., Jarvis, M.G. (1998).European Soil Database: information accessand data distribution procedures. In: LandInformation Systems. Developments forplanning the sustainable use of land resources,Heineke, H.J., Eckelmann ,W., Thomasson, A.,Jones, R.J.A., Montanarella, L. and Buckley,B. (Eds). EUR 17729 EN. Office of theOfficial Publications of the EuropeanCommunities, Luxembourg.Ispra, Italy. 564pp.

King, D., Fox, D.M., Daroussin, J., Le Bissonnais,Y., and Danneels, V. (1998). Upscaling asimple erosion model from small areas to alarge region. Nutrient Cycling inAgroecosystems. 50, 143-149.

Lagacherie, P., Cazemier, D.R., Van Gaans,P.F.M. and Burrough, P.A. (1997). Fuzzy k-means clustering of fields in an elementarycatchment and extrapolation to a larger area.Geoderma. 77 (2-4), 197-216.

Le Bas, C., King, D., Jamagne, M. and Daroussin,J. (1998). The European Soil InformationSystem. In: Land Information Systems.Developments for planning the sustainable useof land resources, Heineke H.J., EckelmannW., Thomasson A., Jones R.J.A., MontanarellaL. and Buckley, B. (Eds). ESB-SAI-JRC, Ispra,Italy. 564pp.

Le Bissonnais, Y., Montie,C., Daroussin, J. andKing, D. (1998). Cartographie de l'aléa érosiondes sols de France. IFEN-MATE Report.Collection Etudes et Travaux No. 18, 60 pp.

Legros, J.P., Bornand, M. and Viron, J.C. (1991).Recherches des zones aptes à l'épandage decomposts urbains dans la région de Montpellier(Hérault, France). GIS Colloquium, Florac.INRA. 229-239.

MAPA. (1998). Report of the IGCS colloquium atPoitiers. Doc. Roneo. 12 pp.

Martin, S. (1993). L’observatoire de la Qualité desSols, un outil de gestion pour l’agriculture, uninstrument d’observation de la biosphère.Chambre d’Agriculture. Supplement No. 897,8-10.

Monnet, J.C. and Gaiffe, M. (1998). Le terroir, uneréalité géographique mise en évidence par descritères édaphiques. Etude et Gestion des Sols,5(1), 43-59.

Saby, N., Walter, C., Combo, S. and Quidu, O.,(1999). Constitution et thématisation de la basede données du référentiel pédologique du sudde l'Ille et Vilaine. Laboratoire de Science duSol ENSA Rennes. 52pp

Schvartz, C., Walter, C., Daroussin, J. and King,D. (1998). Statistical review of the soil testsmade in France and comparison with the1:1,000,000 soil map. 16th World Congress ofSoil Science, Montpellier, 20-26/08/1998.

Thomas, A.L., King, D., Dambrine, E., Couturier,A. and Roque, J. (1998). Modelling spatialvariability of soils with parameters derivedfrom relief and geologic materials in sandstonepart of the Vosges (Northeastern France).Geoderma. (in press).

Ulrich, E. (1995). Le reseau RENECOFOR:Objectifs et réalisations. Revue ForestièreFrançaise. Vol. 2. 107-121.

Walter, C., Schvartz, C., Claudot, B., Bouedo, T.and Aurousseau, P. (1997). Synthèse nationaldes analyses de terre réalisées entre 1990 et1994. Etude et Gestion des Sols. 4(3), 205-218.

Wösten, J.H.M., Lilly, A., Nemes, A. and Le Bas,C. (1998). Using existing soil data to derivehydraulic parameters for simulation models inenvironmental studies and in land useplanning. SC-DLO Report No. 156.Wageningen. The Netherlands. 106pp.

soil mapping and soil monitoring: state of progress and...

Documents