Parameterization of mires in a numerical weather prediction model Alla Yurova Hydrometeorological Centre of Russia

Mire (peatland): definition Drainage of water is blocked. Precipitation is retained. Water table is close to the surface (max 70 cm) Specific vegetation-Sphagnum moss, sedges Decomposition of organic matter is slowed-peat is formed

The spatial distribution of mires in Russia from the GIS "Peatlands of Russia" (Vompersky et al., 2005). Mires have a specific: Heat balance Moisture exchange

Global semi-Lagrangian NWP model SL-AV (Tolstykh, 2001) Operational NWP in Hydrometeorological Centre of Russia Resolution 0.72 о lat и 0.9 о lon, 50 vertical levels Dynamical core- semi-Lagrangian, semi-implicit, vorticity and divergence as prognostic variables, unstaggered horizontal grid Physical parameterizations- ALADIN/ALARO, including ISBA LSS In Siberia forecasts in summer are biased towards high air temperature and low relative humidity

Modifications done to the SL-AV model to simulate mire heat and water balance Multilayer soil heat transfer model with heat capacity and thermal conductivity from Wania et al. (2009) Water balance with MMWH Two schemes to simulate evapotranspiration(1-Lafleur et al., 2005;2-Weiss et al., 2006) Prescribed roughness length and albedo 1 2 ET-evapotranspiration РЕТ-potential ET

The Mixed Mire Water and Heat model MMWH (Granberg et al., 1999) W=P-E-q, q=l q ·i·K_h(z cat -z wt ), W-water content, P-precipitation, E-evapotranspiration, q-runoff, i-slope of the water table,·K_h-transmissivity coefficient, l q -lumped parameter z cat z wt

Sensible heatLatent heat Components of the heat balance from the eddy-flux measurements, standard model simulation (stand), and simulation with a new model (mire). Degero Srormyr mire, Sweden

Components of the radiation and heat balance from the standard model simulation (stand), and simulation with a new model (mire). July-August 2008, mire grid cells only, Western Siberia LW balance Sensible heat Latent heat Soil heat flux

Mean bias error (MBE) for forecasted temperature C, for the standard model (ref), for the saturated mire surface (satur) model, for the model with the Weiss et al. (2006) function for evapotranspiration (finevap), and for the model incorporating the Lafleur et al. (2005) function for evapotranspiration (canevap). July-August 2008, mire stations only, Western Siberia

. Mean absolute error (MAE) for forecasted temperature C, for the standard model (ref), for the saturated mire surface (satur) model, for the model with the Weiss et al. (2006) function for evapotranspiration (finevap), and for the model incorporating the Lafleur et al. (2005) function for evapotranspiration (canevap). July-August 2008, mire stations only, Western Siberia

Mean bias error (MBE) for forecasted relative humidity, for the standard model (ref), for the saturated mire surface (satur) model, for the model with the Weiss et al. (2006) function for evapotranspiration (finevap), and for the model incorporating the Lafleur et al. (2005) function for evapotranspiration (canevap). July-August 2008, mire stations only, Western Siberia

RMSE for forecasted relative humidity, for the standard model (reef), for the saturated mire surface (satur) model, for the model with the Weiss et al. (2006) function for evapotranspiration (finevap), and for the model incorporating the Lafleur et al. (2005) function for evapotranspiration (canevap). July-August 2008, mire stations only, Western Siberia

Conclusions: It is important to incorporate mires when forcasting weather in Siberia Heat balance partitioning has changed The mire parameterization has helped to reduce a large warm temperature bias in Western Siberia for the forecast for lead times of 12, 36 and 60h, but did not eliminate forecast bias for lead times of 24, 48 and 72h.

Future plans: Testing the model for winter conditions (freezing and thawing) Investigating the effect of mire drainage on local and regional weather conditions

Thanks for your attention! This work was financed from the RFBR grants а and а