I have a model (Miguez et al 2008) that defines wilting point as the soil water content below which there is no transpiration (from Johnson et al 1993). Field capacity is defined as the volume of water that remains in wetted soil after excess has drained. Neither are truly constants for a given soil, but they are commonly used as such in modeling.

Are there are any generic pedotransfer functions that can be used to estimate these parameters from sand/silt/clay fractions? These estimates will be used as priors in Bayesian data assimilation, and will thus be subsequently constrained by observations of soil moisture and transpiration.


  • Johnson et al 1993 The Implementation and Validation of Improved Land-Surface Hydrology in an Atmospheric General Circulation Model. J. Climate (pdf)
  • Miguez et al 2008 A semimechanistic model predicting the growth and production of the bioenergy crop Miscanthus giganteus: description, parameterization and validation. GCB-Bioenergy (pdf).
  • Wikipedia "Field Capacity" and "Wilting Point".

1 Answer 1


As far as I know, there’s no one definitive pedo-transfer function (PTF) but there have been several studies that train functions against multi-site databases, rather than just deriving site-specific functions. These PTFs typically define soil moisture-conductivity-suction relationships based on silt, sand and clay fractions from which you can read off field capacity (suction = -10 kPa) and wilting point (suction = -1500 kPa). Which PTF you use will depend on the soil moisture-conductivity-suction hydraulic description that your model uses.

For Clapp and Hornberger hydraulics there's the Cosby et al (1984) PTF. This is trained on data from the US. It's quite possible that there are more recent PTFs trained on more and better data, but this one is still in use (e.g., by the UK Met Office).

For van Genuchten hydraulics there's the Wosten et al (1999) PTF (see also Wosten et al, 2001). This is trained on the European HYPRES database, and you need organic percentage and bulk density in addition to the silt/sand/clay percentages. There’s also ROSETTA (Schaap et al, 2001), which uses several PTFs hierarchically and is trained on data from North America and Europe.

Note that these hydraulic properties are difficult to measure in situ and in the lab, so the fitted PTFs often explain quite a low fraction of the variance in the training data. You might be able to use this information to inform your priors, which should probably be quite conservative.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.