Skip to navigation Skip to content

HYPRES updates

HYdraulic PRoperties of European Soils

Version 1 of HYPRES is now complete and comprises approximately 1800 soil profiles with about 5500 individual soil samples with hydrological data. These data are from over 120 soil units (according to the modified FAO soil legend, CEC, 1985). Summary data on the composition of HYPRES is given in Tables 1 and 2. The RAWRET and RAWK tables hold about 198 000 and 120 500 data pairs of theta/h and K/h respectively. In total, HYPRES has about 25 MBytes of soil hydrological, pedological and environmental data.

A review of the profile data contained within the HYPRES to potentially augment the SPADE/M database in 2012 provided the basis for revisions to the HYPRES database to correct inconsistencies in data format, to check and validate potential errors and to standardize the data formats throughout.

There was a substantial effort to convert all the various locally derive georeferences to a common standard INSPIRE compliant ETRS 1989 projection system where possible and the data within key fields had their terminology standardised throughout including revising the soil classification to the FAO 1994 classification. Unrealistic values or duplicate data were checked against the original data wherever possible and amendments made. A total of 111 records in the BASICDATA table were found to have no soil horizon data and were deleted leaving 1679 records. As there was a proposal to investigate the role of soil parent material and land use on the hydrological properties of the HYPRES soils separate fields within HYPRES detailing the land use and parent material were created where these had previously been held solely within the BASICDATA.SITEDESCRIP field. These were recorded using a hierarchical approach to facilitate analyses at different levels of detail. This constitutes Version 1.1 of the HYPRES database.

Table 1: Approximate counts Table 2: Number of profiles per Major Soil Group:
  • Soil units 95
  • Soil profiles 1777
  • Soil horizons (including replicates) 5521
  • Cambisols 512
  • Fluvisols 421
  • Gleysols 209
  • Luvisols 186
  • Podzols 158
  • Phaeozems 53
  • Histosols 30

The soil hydraulic data was derived by various methods resulting in a marked imbalance in the number of data pairs for the soil samples in HYPRES. Therefore, it was necessary to standardise these data prior to the development of pedotransfer functions to reduce the possibility of statistical bias. The volumetric soil water content, theta, and hydraulic conductivity, K, as functions of pressure head, h, were parameterised with the equations derived by van Genuchten (1980). The nonlinear least-squares optimisation program RETC (van Genuchten et al., 1991) was used to predict the unknown Mualem-van Genuchten parameters (theta r, theta s, Ks, alpha, l and n) simultaneously from measured water retention and hydraulic conductivity data. The subscripts r and s refer to residual and saturated values and alpha, l and n are parameters that determine the shape of the curves.

Once the parameterisation was completed, the optimised Mualem-van Genuchten model parameters were used to generate water content and hydraulic conductivity values for the following  selected pressure head values:  0, -10, -20, -50,  -100, -200, -250, -500, -1000, -2000, -5000, -10000, -15000 and -16000 cm. In this way all soil horizons, regardless of the number of original measured data points, could be represented by an equal weight in the process of development of class pedotransfer functions. The derived data are stored in the HYDRAULIC_PROPS table.

Figure 1: database structure

database structure - image needs to be added back in

As the primary aim of the project was to derive a series of pedotransfer function capable of predicting soil hydrological characteristics, work was initiated to provide a standardised dataset from which these functions can be derived. This involved firstly standardising the disparate systems of particle size distributions used by the network partners and secondly, standardising the contributed soil hydrological properties using Mualem-van Genuchten equations.

Particle size

The particle size data within the Soil Geographical Data Base of Europe follows the FAO system and in order to make HYPRES compatible with this dataset, it was decide to standardise the HYPRES texture data on these classes. This standardisation involved the estimation of the proportions of particles in the 2-50 mm range from those datasets where silt was defined as the fraction between 20 and 60 mm or 20 and 63 mm and where sand was defined as the fraction between 20 and 2000 mm. Four methods were used to describe the cumulative particle size curves and to derive the equations to be used in the interpolation (Nemes et al., Geoderma 90 pp187-202) These were loglinear, the Gompertz (a type of logistic curve), splines and a novel similarity procedure. This latter procedure involved comparing the cumulative particle size distribution with a well quantified reference set in which the proportions of particle less than 50 mm were known. The cumulative particle size fractions which were common to both datasets were compared and an iteration procedure was used to select at least 10 of the most appropriate curves from the reference set. The average proportions of particles in the 2-50 mm range were then read from the reference set and applied to the deficient dataset.

The accuracy of the four methods were tested against two datasets from HYPRES where the cumulative particle size curve was well quantified, that is, a dataset from northern Germany and one from The Netherlands. Selected points on the curves were omitted, each interpolation technique was applied to estimate this 'missing' value and the results compared. In general, the mean squared errors were higher for the loglinear method, followed by splines, Gompertz and finally the similarity procedure. However, the Gompertz method could not be used where there were less than four particle size classes (for example in the Scottish dataset) as it has four parameters which need to be optimised. The splines were only reliable where the points used in the interpolation were in close proximity, and so in many cases, the similarity procedure proved to be the most effective. The predicted particle size distributions were subsequently stored in the SOIL_PROPS Table and used to group the soils into texture classes.

Hydraulic properties

The hydraulic properties of some 3890 soil horizons were parameterised using the Burdine solution of the Mualem-van Genuchten equations. The data were parameterised in textural groups using a modified version of the RETC code [van Genuchten et al., 1991]. This semi-automated procedure involved visual inspection of the goodness-of-fit. When the parameterisation was complete, the parameters were then used to generate soil moisture retention and unsaturated hydraulic conductivity data for 14 pressure heads. In this way, soil horizons with data derived from the evaporation method and those where the data were from simple desorption techniques could be successfully combined with a minimal risk of statistical bias in the data. These 14 points and the parameters are stored in the HYDRAULIC_PROPS Table and were subsequently used in the derivation of the pedotransfer functions.

Pedotransfer

The work on the derivation of the class pedotransfer functions was presented by Henk Wösten. The class pedotransfer functions were derived from the 14 theta/h and K/h data pairs calculated for each individual soil horizon in each of 11 texture groups. The texture classes are those described within the Soil Geographic database, that is, coarse, medium, medium-fine, fine and very fine for both topsoils and subsoils plus an organic class, giving 11 in total. The soil moisture retention and unsaturated hydraulic conductivity curves were plotted in textural groups, outliers were identified and removed from subsequent calculations. The geometric means and standard deviations were calculated for each of the 14 generated points within the texture class groupings giving average curves and a measure of their variability. In most cases there was little difference between the curve fitted through these geometric mean values and a new parameterisation of the data using the Mualem-van Genuchten equations thus also allowing the derivation of pedotransfer functions of Mualem-van Genuchten parameters for each texture class.

Spatial distribution of soil hydrological properties

Early in the project it was decided that we should strive to develop HYPRES in such a way that it was compatible with existing EU-wide soils datasets and that we should explore the possibility of linking our derived pedotransfer functions to the 1: 1 000 000 Soil Geographical database. These links were formalised at the second annual workshop held in Orléans during October 1996 and collaborative work during year 3 gave us the ability to express the results from HYPRES spatially. Christine Le Bas (INRA, Orléans) standardised the disparate geo-references to a common system and prepared maps showing the topsoils textures of the dominant soil taxonomic unit within each soil mapping unit depicted on the 1: 1 000 000 scale soil map of Europe. The class pedotransfer functions were then used to calculate retained water capacities such as the total available water capacity for each texture class (that is, the water retained between field capacity and permanent wilting point). These values were modified according to the depth of the topsoils to give a cumulative profile available water content for the soil and the results plotted on a map of Europe. This demonstrates that HYPRES can indeed be linked with other EU-wide soil datasets.

Future of HYPRES

The inclusion of this Network within the framework of the European Soil Bureau (ESB) means that the HYPRES database and the derived pedotransfer functions will become one of the core soils databases of the European Soil Bureau. The decision to join with the ESB was made at the workshop in Orléans in October 1996 but prior to that, HYPRES was being specifically designed to complement these existing databases wherever possible. By linking to a GIS, HYPRES is capable of being combined with a wide range of non-soils related spatial data. In future, it is envisaged that the database (and its derived data) will be applied to a wide range of environmental issues as well as to many small and large scale strategic research projects. The flexibility of the database means that new methods of parameterisation or of deriving pedotransfer functions can be easily incorporated allowing it to remain current and active over a prolonged period of time.

References

Appendix 1: List of official participants in the EU funded project 'Using existing soil data to derive hydraulic parameters for simulation modelling in environmental studies and in land use planning'

  1. The Winand Staring Centre for Integrated Land Soil and Water Research Wageningen, the Netherlands Dr. J.H.M. Wösten
  2. Wageningen Agricultural University, Dept. of Soil Science and Geology Wageningen, the Netherlands Dr. H.W.G. Booltink
  3. Consejo Superior de Investigaciones Cientificas, Instituto de Recursos Naturales y Agrobiologia de Sevilla Sevilla, Spain Dr. F. Moreno
  4. Institut National de la Recherche Agronomique - Science du Sol Orléans, France Dr. A. Bruand
  5. Institut National de la Recherche Agronomique - Science du Sol Montpellier, France Dr. M. Voltz
  6. Soil Survey and Land Research Centre Silsoe, England Dr. J. Hollis
  7. Macaulay Land Use Research Institute Aberdeen, Scotland Dr. A. Lilly
  8. The Danish Institute of Plant and Soil Science, Department of Land Use Tjele, Denmark Dr. M. Greve
  9. Istituto di Idraulica Agraria Universita' degli Studi di Napoli "Federico II" Napoli, Italy Dr. N. Ronano
  10. Technische Universität Berlin, Institut für Ecology, Department of Soil Science Berlin, Germany Dr. R. Plagge
  11. Technische Universität Braunschweig, Institut für Geographie und Geoökologie Braunschweig, Germany Dr. B. Diekkrüger
  12. GSF - Forschungszentrum für Umwelt und Gesundheit, Institut für Bodenökologie Oberschleissheim, Germany Dr. E. Priesack
  13. Zentrum für Agrarlandschafts- und Landnutzungsforschung (ZALF) e.V., Institut für Hydrologie Müncheberg, Germany Dr. U. Schindler
  14. Bundesanstalt für Geowissenschaften und Rohstoffe Hannover, Germany Dr. V. Hennings
  15. Aristotle University of Thessaloniki, Soil Science Laboratory Thessaloniki, Greece Dr. K.P. Panayiotopoulos
  16. Agricultural University of Athens, Laboratory of Soils and Agricultural Chemistry Athens, Greece Dr. C. Kosmas
  17. Estaçäo Agronómica Nacional Lisboa, Portugal Dr. M. da Conceiçäo Goncalves
  18. Katholieke Universiteit Leuven, Institute for Land and Water Management Leuven, Belgium Dr. L Hubrechts

In addition three other Institutions played an active part in contributing data and in scientific discussions during the three years of the project, these are:

  1. Soil Fertility Research Institute Bratislava, Slovakia Dr B. Houšková
  2. Swedish University of Agricultural Sciences Department of Soil Sciences Uppsala, Sweden
  3. Department of Agriculture, Northern Ireland Belfast, Northern Ireland Dr A.J. Higgins

Finally, at the first annual workshop it was suggested that we create a list of references on the development and use of pedotransfer functions. Such a reference list has been compiled.


European Soil Bureau

The group has been formally recognised as the forth working group of the European Soil Bureau and will deal with Soil Hydraulic Parameters.

Relevant Publications:

  1. Lilly, A. 1995. A description of the data to be held in the European database of soil hydraulic properties. Manual. Version 2.0. October 1995. Winand Staring Centre. Wageningen. The Netherlands.
  2. Lilly, A. 1995. Notes on the data to be held in the Nederlands database of soil hydraulic properties and for contributions to the European database. Manual. Version 1.0. October 1995. Winand Staring Centre. Wageningen. The Netherlands.
  3. Lilly, A. 1995. Report on the first annual workshop of the EU funded project using existing soil data to derive hydraulic parameters for simulation models in environmental studies and in land use planning. October 1995. Winand Staring Centre. Wageningen. The Netherlands.
  4. Lilly, A. 1996. Report to SOAEFD on the work done under RO 011443 while on secondment to the Winand Staring Centre, Wageningen, The Netherlands in connection with the first year of the EU funded project (CHRX - CT94 - 0639) "Using existing soil data to derive hydraulic parameters for simulation models in environmental studies and in land use planning"
  5. Lilly, A. 1997. Report of the second annual workshop of the EU funded project Using existing soil data to derive hydraulic parameters for simulation models in environmental studies and in land use planning. In: The use of Pedotransfer in soil hydrology research in Europe. Proceedings of the second EU Workshop on Using existing soil data to derive hydraulic parameters for simulation models in environmental studies and in land use planning. Edited by Bruand, A., Duval,O., Wösten, J.H.M. and Lilly, A. October 1996. INRA. Orléans and EC/JRC Ispera.
  6. Lilly, A. 1997. A description of the HYPRES database (HYdraulic PRoperties of Europeans Soils). In: The use of Pedotransfer in soil hydrology research in Europe. Proceedings of the second EU Workshop on Using existing soil data to derive hydraulic parameters for simulation models in environmental studies and in land use planning. Edited by Bruand, A., Duval,O., Wösten, J.H.M. and Lilly, A. October 1996. INRA. Orléans and EC/JRC Ispera.
  7. Lilly, A. and Wosten, J.H.M. 1996. Development and use of a database of measured soil hydraulic properties for European soils. Workshop on Land Information Systems: Developments to planning the sustainable use of land resources. Hannover, November 1996.
  8. Nemes, A., Wösten, J.H.M., Lilly, A. and Oude Voshaar, J.H., 1999. Evaluation of different procedures to interpolate the cumulative particle-size distribution to achieve compatibility of soil databases. Geoderma, 90, 187-202.
  9. J.H.M. Wösten, A. Lilly, A. Nemes and C. Le Bas. 1999. Development and use of a database of hydraulic properties of European soils. Geoderma, 90, 169-185.
     

Learning & Resources

Tags


Printed from /learning/natural-resource-datasets/hypres/updates on 19/03/24 07:27:10 AM

The James Hutton Research Institute is the result of the merger in April 2011 of MLURI and SCRI. This merger formed a new powerhouse for research into food, land use, and climate change.