Journal cover Journal topic
Nonlinear Processes in Geophysics An interactive open-access journal of the European Geosciences Union

Journal metrics

  • IF value: 1.321 IF 1.321
  • IF 5-year<br/> value: 1.636 IF 5-year
  • SNIP value: 0.903 SNIP 0.903
  • SJR value: 0.709 SJR 0.709
  • IPP value: 1.455 IPP 1.455
  • h5-index value: 20 h5-index 20
Nonlin. Processes Geophys., 23, 447-465, 2016
© Author(s) 2016. This work is distributed
under the Creative Commons Attribution 3.0 License.
Research article
09 Dec 2016
Constraining ecosystem model with adaptive Metropolis algorithm using boreal forest site eddy covariance measurements
Jarmo Mäkelä1, Jouni Susiluoto1, Tiina Markkanen1, Mika Aurela1, Heikki Järvinen2, Ivan Mammarella2, Stefan Hagemann3, and Tuula Aalto1 1Finnish Meteorological Institute, P.O. Box 503, 00101 Helsinki, Finland
2Department of Physics, P.O. Box 48, University of Helsinki, 00014 Helsinki, Finland
3Max Planck Institute for Meteorology, Bundesstraße 53, 20146 Hamburg, Germany
Abstract. We examined parameter optimisation in the JSBACH (Kaminski et al., 2013; Knorr and Kattge, 2005; Reick et al., 2013) ecosystem model, applied to two boreal forest sites (Hyytiälä and Sodankylä) in Finland. We identified and tested key parameters in soil hydrology and forest water and carbon-exchange-related formulations, and optimised them using the adaptive Metropolis (AM) algorithm for Hyytiälä with a 5-year calibration period (2000–2004) followed by a 4-year validation period (2005–2008). Sodankylä acted as an independent validation site, where optimisations were not made.

The tuning provided estimates for full distribution of possible parameters, along with information about correlation, sensitivity and identifiability. Some parameters were correlated with each other due to a phenomenological connection between carbon uptake and water stress or other connections due to the set-up of the model formulations. The latter holds especially for vegetation phenology parameters. The least identifiable parameters include phenology parameters, parameters connecting relative humidity and soil dryness, and the field capacity of the skin reservoir. These soil parameters were masked by the large contribution from vegetation transpiration.

In addition to leaf area index and the maximum carboxylation rate, the most effective parameters adjusting the gross primary production (GPP) and evapotranspiration (ET) fluxes in seasonal tuning were related to soil wilting point, drainage and moisture stress imposed on vegetation. For daily and half-hourly tunings the most important parameters were the ratio of leaf internal CO2 concentration to external CO2 and the parameter connecting relative humidity and soil dryness. Effectively the seasonal tuning transferred water from soil moisture into ET, and daily and half-hourly tunings reversed this process.

The seasonal tuning improved the month-to-month development of GPP and ET, and produced the most stable estimates of water use efficiency. When compared to the seasonal tuning, the daily tuning is worse on the seasonal scale. However, daily parametrisation reproduced the observations for average diurnal cycle best, except for the GPP for Sodankylä validation period, where half-hourly tuned parameters were better. In general, the daily tuning provided the largest reduction in model–data mismatch.

The models response to drought was unaffected by our parametrisations and further studies are needed into enhancing the dry response in JSBACH.

Citation: Mäkelä, J., Susiluoto, J., Markkanen, T., Aurela, M., Järvinen, H., Mammarella, I., Hagemann, S., and Aalto, T.: Constraining ecosystem model with adaptive Metropolis algorithm using boreal forest site eddy covariance measurements, Nonlin. Processes Geophys., 23, 447-465, doi:10.5194/npg-23-447-2016, 2016.
Publications Copernicus
Short summary
The land-based hydrological cycle is one of the key processes controlling the growth and wilting of plants and the amount of carbon vegetation can assimilate. Recent studies have shown that many land surface models have biases in this area. We optimized parameters in one such model (JSBACH) and were able to enhance the model performance in many respects, but the response to drought remained unaffected. Further studies into this aspect should include alternative stomatal conductance formulations.
The land-based hydrological cycle is one of the key processes controlling the growth and wilting...