Laboratory evaporation experiments in undisturbed peat columns for determining peat soil hydraulic properties : [abstract for conference american geophysical union, fall meeting 2013, san francisco, usa, 09 – 13 december 2013]

http://agu-fm13.abstractcentral.com/s1agxt/com.scholarone.s1agxt.s1agxt/C844.
CONTROL ID: 1801605
TITLE: Laboratory evaporation experiments in undisturbed peat columns for determining peat soil hydraulic
properties
AUTHORS (FIRST NAME, LAST NAME): Ullrich Dettmann1, Enrico Frahm2, Michel Bechtold1
INSTITUTIONS (ALL): 1. Institute of Climate-Smart Agriculture, Thünen-Institut (TI), Braunschweig, Germany.
2. Liquid Flow, Department 1.53, Physikalisch-Technische Bundesanstalt (PTB), Braunschweig, Germany.
ABSTRACT BODY: Knowledge about hydraulic properties of organic soils is crucial for the interpretation of the
hydrological situation in peatlands. This in turn is the basis for designing optimal rewetting strategies, for assessing the
current and future climatic water balance and for quantifying greenhouse gas emissions of CO2, CH4 and N2O, which
are strongly controlled by the depth of the peat water table.
In contrast to mineral soils, the hydraulic properties of organic soils differ in several aspects. Due to the high amount of
organic components, strong heterogeneity, and shrinkage and swelling of peat, accompanied by changing soil volume
and bulk density, the applicability of standard hydraulic functions developed for mineral soils for describing peat soil
moisture dynamics is often questioned.
Hence, the objective of this study was to investigate the applicability of the commonly applied van Genuchten-Mualem
(VGM) parameterization and to evaluate model errors for various peat types. Laboratory column experiments with
undisturbed peat soils (diameter: 30 cm, height: 20 cm) from 5 different peatlands in Germany were conducted.
In numerical simulations using HYDRUS-1D the experimental data were used for an inverse estimation of the soil
hydraulic parameters. Using the VGM parameterization, the model errors between observed and measured pressure
heads were quantified with a root mean square error (RMSE) of 20 – 65 cm. The RMSE increased for soils with higher
organic carbon content and higher porosity. Optimizing the VGM 'tortuosity' parameter (τ) instead of fixing it to its
default of 0.5 strongly reduced the RMSE, especially for the soils that showed high pressure head gradients during the
experiment.
Due to the fact, that very negative pressure heads in peatlands occur rarely, we reduced the range of pressured heads in
the inversion to a 'field-relevant' range from 0 to -200 cm which strongly reduced the RMSE to 6 - 12 cm and makes the
VGM parameterization applicable for all investigated peat soils. For the field-relevant scale, especially for very wet
conditions, we demonstrate the importance of macro-pores by using a simple macro-pore approach, with only 1
additional parameter, i.e. the macro-pore fraction, which strongly reduced the RMSE down to 1 – 7 cm. Since τ has not
been identified as an important parameter for the field-relevant range, only 5 parameters were optimized in this
approach. This keeps the derivation of the parameters manageable and thus provides a model that is applicable to
practical issues.
KEYWORDS: 1875 HYDROLOGY Vadose zone, 1847 HYDROLOGY Modeling.
Additional Details
Previously Presented Material: EGU - Vienna 2013
Contact Details
CONTACT (NAME ONLY): Ullrich Dettmann
CONTACT (E-MAIL ONLY): ullrich.dettmann@ti.bund.de
TITLE OF TEAM:

Source: http://literatur.ti.bund.de/digbib_extern/dn053046.pdf