Yonggen Zhang
Subsurface Hydrology Computational Group


Pedotransfer function

PTF can be used in small scale.
From Van Looy et al (2017)

GCM also need PTFs
(Pic from NOAA)

Schaap et al. (2001) developed a very popular pedotransfer function (PTF), Rosetta. It is based on artificial neural network analysis coupled with bootstrap re-sampling method which allows the estimation of van Genuchten water retention parameters (van Genuchten, 1980), saturated hydraulic conductivity (Ks), and their uncertainties. The Rosetta is available at http://cals.arizona.edu/research/rosetta/.

We developed a new Rosetta model that unify the water retention and Ks submodels into one. The models have less bias and better performance in terms of observed quantities. Both the Rosetta and the improved version are implemented in the python programming language, and the source code as well as additional documentation will be available at: http://www.cals.arizona.edu/research/rosettav3.html. A web-based user interface is being developed.

Recent Publications

  • Yonggen Zhang, Marcel G. Schaap, and Yuanyuan Zha, 2018. A High-Resolution Global Map of Soil Hydraulic Properties Produced by a Hierarchical Parameterization of a Physically-Based Water Retention Model, Water Resources Research, 54. https://doi.org/10.1029/2018WR023539
  • Yonggen Zhang, and Marcel G. Schaap, 2017. Weighted Recalibration of the Rosetta Pedotransfer Model with Improved Estimates of Hydraulic Parameters and Covariance Structures, Journal of Hydrology, doi: 10.1016/j.jhydrol.2017.01.004
  • Marcel G.Schaap, Feike J.Leij, Martinus Th.van Genuchten, 2001. Rosetta: a computer program for estimating soil hydraulic parameters with hierarchical pedotransfer functions, Journal of Hydrology. doi: 10.1016/S0022-1694(01)00466-8
  • Kris Van Looy, Johan Bouma, Michael Herbst, John Koestel, Budiman Minasny, Umakant Mishra, Carsten Montzka, Attila Nemes, Yakov Pachepsky, José Padarian, Marcel Schaap, Brigitta Tóth, Anne Verhoef, Jan Vanderborght, Martine van der Ploeg, Lutz Weihermüller, Steffen Zacharias, Yonggen Zhang, and Harry Vereecken, 2018. Pedotransfer functions in Earth system science: challenges and perspectives, Reviews of Geophysics., doi:10.1002/2017RG000581


Inverse modeling

Soil moisture movement evolution
in a vertical profile

In this project, we propose two methods of parameterizing vadose zone hydraulic properties. Each of these utilizes PTFs and k-means clustering based on kriged soil texture data. One approach entails homogeneous and the other heterogeneous clusters. Our approach preserves heterogeneity without introducing a large number of adjustable parameters. We illustrate it on a 14.5 m-deep zone of unsaturated sediments, measuring 50 × 50 m2 in area, at a semiarid site near Phoenix, Arizona.

Recent Publications

  • Yonggen Zhang, Marcel G. Schaap, Alberto Guadagnini, and Shlomo P. Neuman, 2016. Inverse Modeling of Unsaturated Flow Using Clusters of Soil Texture and Pedotransfer Functions, Water Resources Research., 52, 7631-7644, doi: 10.1002/2016WR019016


Data Assimilation

To be updated

Application of ensemble Kalman filter (EnKF) to Richards equation­ type unsaturated flow problems is challenging, because the relation of state and model parameters is strongly nonlinear.

In this project, we updated hydraulic conductivity in Restart EnKF, which rerun the nonlinear model from the initial time to obtain the updated state variables, in synthetic cases to explore the factors that may influence estimation results, including the initial estimate, the ensemble size, the observation error, and the assimilation interval. The current will be embedded into the Bayesian model averaging framework to enhance the model reliability and reduce predictive uncertainties.


Lattice Boltzmann Simulation of Capillary Trapping Process for CO2 Sequestration

The role of capillary trapping in CO2 storage
From Juanes et al. (2006) WRR

Carbon capture and storage is the only feasible and promising technology to reduce the global warming effects caused by carbon emissions while allowing continued large-scale use of fossil fuels. Deep saline aquifers have the largest identified storage potential for CO2. Injection of CO2 into deep saline aquifers leads to a multi-component, multiphase flow system, while capillary trapping of CO2 is an essential mechanism after the injection phase, in which CO2 becomes immobile. In this project, we investigated whether multiple drainage and imbibition of capillary trapping enhance the efficiency of geological CO2 storage in deep saline aquifers where super-critical conditions prevail.


Sources identification of organic pollution

Contaminant from different sources

Diagnostic ratios and factor analysis with nonnegative constraints (FA-NNC) were used to determine the pollution sources of polycyclic aromatic hydrocarbon (PAH) in different wetland components, including groundwater, surface water, sediments, and soils. FA-NNC combined with Latin Hypercube Sampling (LHS) simulation was shown to be an effective quantitatively method to identify sources, source contribution, and uncertainty of PAH contamination.

Recent Publications

  • Yonggen Zhang, Mingzhu Liu, Honghan Chen, and Guohua Hou, 2014. Source identification of polycyclic aromatic hydrocarbons in different ecological wetland components of the Qinkenpao Wetland in Northeast China, Ecotoxicology and environmental safety.,102, 160-167, doi: 10.1016/j.ecoenv.2014.01.030