1. Homepage
  2. Impact Model Settings & Characteristics
  3. Impact model: FLake-IGB

Impact model: FLake-IGB

Sector
Lakes (local)
Region
local

FLake is a 1-dimensional bulk hydrodynamic lake model. It simulates water temperature and vertical mixing in lakes and was specifically designed to parameterize inland waters in climate models and numerical weather prediction systems. FLake is based on a two-layer parametric representation of the vertical temperature structure. The upper layer is treated as well-mixed and vertically homogeneous. The structure of the lower stably stratified layer, the lake thermocline, is parameterized using a self-similar representation of the temperature profile. The version used here has been slightly modified from the original to accept longwave radiation as a direct input instead of calculating it from cloud cover. More information can be found at http://www.flake.igb-berlin.de/.

Information for the model FLake-IGB is provided for the simulation rounds shown in the tabs below. Click on the appropriate tab to get the information for the simulation round you are interested in.

Confirm data for simulation round ISIMIP2b
Person responsible for model simulations in this simulation round
Georgiy Kirillin: kirillin@igb-berlin.de, Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), Berlin (Germany)
Tom Shatwell: tom.shatwell@ufz.de, Helmholtz Centre for Environmental Research (UFZ-Magdeburg) (Germany)
Additional persons involved: Tom Shatwell
Output Data
Experiments: II, III, VIII (only historical and future periods)
Climate Drivers: None
Date: 2020-04-29
Basic information
Model Version: FLake
Simulation Round Specific Description: FLake is forced by 2 m air temperature, 10 m windspeed, incoming solar radiation, surface humidity, downwelling longwave radiation. Input variables were disaggreagted from daily to 3-hourly resolution. Humidity [mb] was calculated as ps*huss/(.622+huss) / 101.300, and was truncated at saturation if the diurnal temperature changes would have caused supersaturation.
Reference Paper: Main Reference: Mironov, D. et al. Parameterization of lakes in numerical weather prediction. Description of a lake model. COSMO Technical Report No. 11, Deutscher Wetterdienst,None,,2008
Reference Paper: Other References:
Resolution
Spatial aggregation: lake
Additional spatial aggregation & resolution information: FLake was applied to specific case study lakes, forced with meteorology for the 0.5 x 0.5 degree grid cell in which the respective lake is located.
Additional temporal resolution information: The model was forced with meteorology at 3-hour resolution to account for diurnal heating and cooling cycles, which affect stratification. To do this, the daily inputs supplied by ISIMIP were disaggregated to 3-hour resolution as described in Shatwell et al 2019, Hydrology and Earth-System Sciences, 23: 1533-1551.
Input data
Simulated atmospheric climate data sets used: IPSL-CM5A-LR, HadGEM2-ES, GFDL-ESM2M, MIROC5
Observed atmospheric climate data sets used: EWEMBI
Additional input data sets: Limnological data from the case study lakes were used to parameterise the model (e.g. measured light extinctinon, morphology etc.)
Climate variables: huss, sfcWind, tasmax, tas, tasmin, rlds, rsds, ps
Additional information about input variables: Daily input data were disaggregated to hourly values as described in Shatwell et al 2019, Hydrol. Earth Syst. Sci. 23(3): 1533-1551.
Spin-up
Was a spin-up performed?: Yes
Spin-up design: One-year spin-up period: spin-up is initialized with water temperature = annual mean air temperature. Then the first 365 days of the simulation period are simulated, and the system state on day 365 is used to initialize the main simulation, which starts again at day 1.
Extreme Events & Disturbances
Additional comments: The model was parameterised using local lake information: mean lake depth, from hypsographic information, truncated to 60 m for very deep lakes, the lake fetch, estimated as the square root of the lake area, and the light extinction coefficient, which was measured directly, estimated from Secchi depth, or from maximum lake depth. Two parameters in the model were calibrated to observed water temperature in the local lakes: 1) the extinction coefficient was allowed to vary by 30% above/below the long term average observed extinction, 2) the profile relaxation time parameter c_relax_C was allowed to vary between 0.001 and 0.9. Fitting was performed using a pseudo-random search algorithm and the Levenberg-Marquardt algorithm. For calibration the model was forced with GSWP3-EWEMBI historical data from ISIMIP2a. The fitted parameters were used in the simulation runs for ISIMIP2b.
Methods
Snow melt: Snow was not modelled
Calibration
Was the model calibrated?: Yes
Which years were used for calibration?: 1979-2016, depending on data availability for each lake
Which dataset was used for calibration?: GSWP3-EWEMBI
How many catchments were callibrated?: 62 local lakes were calibrated. Exceptions due to lack of data were Alqueva, Taupo, Waahi, Zlutice




Website consultancy, design and implementation:
Matthias Brück
brueck.io

Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution 4.0 International license.