1. Homepage
  2. Output Data
  3. Model information and data caveats from ISIMIP Fast Track

Model information and data caveats from ISIMIP Fast Track


For more information about the model and their configuration please use the links behind the model names or get in contact with the groups directly.

Remarks on the bias-corrected GCM input data:

This data has two versions. The initial version was used to force impact models in the ISIMIP Fast Track phase and was only being used for the ISIMIP2 catch-up experiments for sectors that were already part of the Fast Track. Note, however, that this data has been removed from the ISIMIP2a protocol.

General remark on crop models pirrww variable:

The irrigation water demand (=potential irrigation water withdrawal) reported as „pirrww“ by the crop models includes different assumptions about the efficiency of irrigation, i.e. the reported pirrww is partly higher than the water assumed to be directly available to the plants because of losses due to transportation and inefficient irrigation. To harmonize the reported pirrww assuming a project efficiency of 100% the pirrww data provided by pDSSAT have to be multiplied by 0.75 for maize, soy and wheat. For LPJmL assumed project efficiencies varied spatially but independent of the crop. The reported pirrww data have to be multiplied by the project efficiencies given in irrigation_project_efficiencies.nc. For the other models there is no modification required as the modelling groups already assumed an irrigation use efficiency of 100%.

  • DBH
    • Contact: tangqh@igsnrr.ac.cn
    • Caveats or remarks:
      - CO2 concentration is set as constant throughout the simulation period.
      - possible bug in the model code for downscaling daily radiation to hourly ones. May result in unreliable estimates of hourly radiation, and thus the daily runoff output at about 60°N.
  • DIVA
  • GEPIC
  • H08
    • Contact: hanasaki@nies.go.jp, masaki.yoshimitsu@nies.go.jp
    • Caveats or remarks:
      Basic model description
      - Time step: Daily
      - Meteorological forcing: Rainf, Snowf, Tair, Wind, Qair, LWdown, SWdown, PSurf (See ALMA Convention for full names)
      - Energy balance: Yes
      - Evaporation scheme: Bulk formula
      - Runoff scheme: Saturation excess and non-linear baseflow
      - Snow scheme: Energy balance
      - Vegetation dynamics:No
      - CO2 effect: No
      Reference: Hanasaki et al. (2008a, 2008b) http://direct.sref.org/1607-7938/hess/2008-12-1007, http://direct.sref.org/1607-7938/hess/2008-12-1027

      Some caveats of output files
      1) Soilmoist50: The file contains soil water in 1-meter soil layer, not 50cm.
      2) Adomuse, Adomww, Airruse, Airrww, Pirruse, and Pirrww: Unit was converted from the original [kg s-1] to [kg m-2 s-1] using respective 0.5X0.5deg cell area.
      3) Amanuse and Amanww: Amanuse includes Aelecuse because H08 does not separate electricity water from manufacturing water. Similarly, Amanww includes Aelecww because of the same reason. See also http://isimip.wordpress.com/2012/11/13/h08-water-demand-projections/

      Caveats of model setup:
      The reservoir operation module was disabled in this ISI-MIP Fast Track simulation.
  • Hybrid
    • Contact: adf10@cam.ac.uk
    • Caveats or remarks:
      All simulations were run according to the ISIMIP simulation protocol with a 500 year spin-up and dynamic vegetation (i.e. plant type distributions change in response to forcings). Model implementation is largely as described in Friend, A.D. & White, A. (2000). Evaluation and analysis of a dynamic terrestrial ecosystem model under preindustrial conditions at the global scale. Global Biogeochemical Cycles 14(4), 1173-1190.
  • IMAGE
    • Contact: elke.stehfest@pbl.nl, kathleen.neumann@pbl.nl
    • Caveats or remarks:
      For overall information on IMAGE 2.4 see the IMAGE website and MNP (2006), Integrated modelling of global environmental change. An overview of IMAGE 2.4. Netherlands the global scale. Global Biogeochemical Cycles 14(4), 1173-1190.
      Environmental Assessment Agency (MNP), Bilthoven, The Netehrlands

      Input climate data:
      The reservoir operation module was disabled in this ISI-MIP Fast Track.
      The IMAGE crop model calculates climate-controlled distribution of major crops and crop productivity using the FAO (GAEZ) crop suitability approach. For this the crop model requires temperature, precipitation, cloud cover and daytime temperature as initial data.

      Simulation.
      For conducting the simulations the IMAGE crop model uses the following ISI-MIP climate input variables: daily mean temperature (TAS), precipitation (PR), shortwave downward radiation (rsds) . The latter is used to derive cloud cover information from. Further, the mean maximum temperature during the growing period is used as a proxy for daytime temperature (important for photosynthesis). As this one is not available from ISI-MIP climate database we calculated this variable ourselves. For this we added the difference between the monthly mean and the respective monthly mean over the historical period (1961-1990) to the historic trend of daytime temperature (period 1961-1990) that is used in IMAGE (based on 1961-1990 CRU data).
      For tas, pr, rsds we used monthly mean values from the ISI-MIP climate database. These monthly mean values are used to calculate monthly mean values for a 30 year period. For example, the mean temperate value for September 1990 is the mean value of all September months between 1976 and 2005. Potential yields in IMAGE are determined by climate and soil moisture condition (soil moisture, soil water holding capacity). Soil nutrient limiting factors are not accounted for. For the fully irrigated scenarios no limitation in irrigation water availability is assumed.

      Calibration and simulations:
      IMAGE was calibrated to match FAO national crop yield statistics for the period 1970 – 2005. Crop yields are calculated for an interval of five years, starting with 1970. Crop yields for the years in between are the result of a linear interpolation.
      A detailed description of the IMAGE crop model is provided in Leemans, R. and A. M. Solomon (1993). "Modeling the potential change in yield and distribution of the earth's crops under a warmed climate." Climate Research 3: 79-96
  • JeDi
  • JULES
    • Contact: rutger.dankers@metoffice.gov.uk, dbcl@ceh.ac.uk
    • Caveats:
      Simulations were performed with JULES version 3.0, but with added functionality. This includes some new diagnostics, the ability to read in time-varying CO2 concentrations, and a scheme to disaggregate daily meteorological forcing to 3-hourly values, which is based on the scheme used in IMOGEN (Huntingford et al., GMD, 2010). The model ran internally with a half-hourly timestep.

      Note that the ISI-MIP runs were run at HadGEM2-ES resolution (1.875 deg longitude by 1.25 deg latitude) but that the runoff fields were regridded to 0.5x0.5 degree for river flow routing using the TRIP routing scheme. This makes river discharge the only JULES output variable that is available on the higher resolution grid.

      Ancillaries were based on HadGEM2-ES standard ancillaries, where possible (see e.g., Jones et al., GMD, 2011). Parameters for TRIFFID were based on HadCM3C (see e.g., Falloon et al., Biogeosciences, 2012), with slightly adjusted parameters for broadleaf and needleleaf tree plant functional types to better match present-day vegetation distributions.

      The same runs were used to provide output to both the water and biome sectors. This means the water sector results use potential natural vegetation (as simulated by TRIFFID) rather than the actual present-day vegetation distribution. In other words, there is no human land use.

      The model was spun-up as prescribed in the simulation protocol. The model was started from a 1950 dump from a HadGEM2-ES historical run and ran for 5 spinup cycles of 30 years using the equilibrium solution of the vegetation model TRIFFID and a constant CO2 concentration of 280 ppm. Next, the model was run for 185 years using dynamic (standard) TRIFFID and increasing CO2 concentrations as specified in the protocol. The historical runs started in 1951 and output was provided to the archive starting from 1971. Note that a spin up was performed for HadGEM2-ES forcing only;historical runs forced by all other GCMs used the same initital state spun-up using the HadGEM2-ES forcing.

      In addition to the standard ISI-MIP runs we have performed additional sensitivity experiments, some of these are described in the ISI-MIP protocol. These include runs with constant CO2 concentrations, static vegetation distribution, and runs non-bias-corrected driving data from the HadGEM2-ES climate model. Please contact us if you want to know more about these runs.

      Notes on the output (see also the headers of the netcdf files):
      - JULES does not provide potential evapotranspiration (PotEvap).
      - Evaporation from soil (ESoil) includes both transpiration and bare soil (soil surface) components.
      - Root zone soil moisture (RootMoist ) is gridbox available moisture above wilting point, as defined in the ALMA convention.
  • LMM
  • LPJ-GUESS
    • Contact: stefan.olin@nateko.lu.se, thomas.pugh@kit.edu
    • Caveats or remarks:
      LPJ-GUESS results show potential yields, unlimited by nutrient or management constraints. Therefore they are not suitable for best-estimate economic forecasts, although they may have some utility for this purpose if the potential maximum yield is being explored. They are best suited to studies which try to advance scientific understanding of the processes being represented. As the yields are not calibrated to observations of actual yields, absolute changes in yield should not be used. Instead, the use of relative changes in yield (e.g. yield_2100 / yield_2005) is appropriate. These can be applied to an observed baseline if necessary.
      Further notes: Planting dates adaptation is allowed within ±15 days of the calculated optimum 1970 values. The sum of potential heat units required for maturity is dynamically adapted to the prevailing local climate (i.e. cultivar dynamically adjusted to climate). Only one growing season per year is simulated. See Lindeskog et al., Earth Syst. Dynam. Discuss., 4, 235-278, 2013 for further details of LPJ-GUESS parameterisations.
  • LPJmL
    • Contact:
      christoph.mueller@pik-potsdam.de (Agriculture)
      sibyll.schaphoff@pik-potsdam.de (Biomes)
      jens.heinke@pik-potsdam.de (Water)
    • Caveats or remarks:
      All sectors: Input data
      LPJmL uses daily input data on daily mean temperature (tas), precipitation (pr), shortwave downward radiation (rsds) and net longwave radiation. Net longwave radiation is not provided by ISI-MIP, so we compute that from longwave downward radiation (rlds) and surface temperatures.

      Modeling protocol

      Agriculture: Data have been computed for 67420 pixels (CRU land mask) with LPJmL from 1951-2099 in a transient simulation run, using a 200-year spinup to get soil temperatures into equilibrium (natural vegetation and soil carbon pools are neglected here so spinup could be short), recycling the first 30-years of that time series for the spinup phase (again, this would be different for the biome and water applications of LPJmL).
      I've been using the latest LPJmL (as described in Bondeau, et al., (2007); Fader, et al., (2010); Waha, et al., (2012)) version with more complex soil hydrology as needed for the permafrost implementation (Schaphoff et al., 2013).
      National cropping intensity has been calibrated to FAO statistics as described in Fader et al. (2010) but with a linear LAI-FPAR model for maize (Zhou, et al., 2002) and maximum intensity levels for maize at LAImax of 5, minimum root-to-shoot ratios at maturity of 10% (based on insights from the AgMIP wheat and maize pilots & literature). Sowing dates have been computed as in Waha et al. [2012], but have been kept constant after 1951. The model decides internally whether to grow winter or spring wheat on wheat areas. It has a preference for winter wheat, but if winters are too long, it will grow spring varieties (Bondeau, et al., 2007).
      The soil data are taken from Harmonized world soil database (FAO/IIASA/ISRIC/ISSCAS/JRC, 2012). The classification is based on the USDA soil texture classification (http://edis.ifas.ufl.edu/ss169), the hydraulic soil parameters (saturated hydraulic conductivity [mm/h], water content at wilting point/field capacity/saturation) are derived from Cosby et al. [1984] and the thermal parameters (thermal diffusivity (mm^2/s) at wilting point/15%/field capacity (100% whc)/ wilting point/saturation (all water)/ saturation (all ice)) from Lawrence and Slater (2008), the suction head (mm) in Green-Ampt equation following Rawls, Brakensiek and Miller (1983).

      Biomes:

      Water:

      Outputs available:

      AG:
      - Data is provided for the 12 crops implemented in LPJmL, which is all of the 4 priority crops of ISI-MIP and some of the 2nd priority list (millet, rapeseed, sugar beet, sugar cane) and additionally crops (field peas, cassava, sunflower, groundnuts and managed grassland)
      for the primary variables:
      • Yields in tDM/ha
      • Irrigation water applied in mm/year assuming full irrigation without constraints on availability
      as well as for the following diagnostic variables:
      • None of the N-related variables, since N is not explicitly accounted for in LPJmL so far
      • Anthesis dates are not explicitly modeled.
      • Planting dates as computed by LPJmL [Waha, et al., 2012], constant after 1951 [day of year]
      • Maturity date [day after planting/sowing]
      • AET [mm/year]
      • Total above ground biomass [tDM/ha]
      • Growing season precip [mm/growing_season_day]
      • Growing season rsds [W/m2/growing_season_day]
      • Growing season GDD0 [GDD0/growing_season_day]

      BI:

      WA:
  • Mac-PDM.09
    • Contact: simon.gosling@nottingham.ac.uk
    • Caveats or remarks:
      A detailed description of the model is provided in the following reference: Gosling SN and Arnell, NW (2011) Simulating current global river runoff with a global hydrological model: model revisions, validation and sensitivity analysis. Hydrological Processes 25: 1129-1145. doi: 10.1002/hyp.7727

      Data for Dis, Qsb and Qs for 1971 only in all simulations was corrupted during file writing; all other data are ok.
  • MATSIRO
  • MIASMA
    • Contact: p.martens@maastrichtuniversity.nl
    • Caveats or remarks:
      A detailed model description can be found in Martens, P. (1998). Health and Climate Change: Modelling the impacts of global warming and ozone depletion. Earthscan Publications Ltd., London, as well as in the model documentation on the website.
  • MPI-HM
    • Contact: tobias.stacke@zmaw.de, stefan.hagemann@zmaw.de
    • Caveats or remarks:
      A detailed model describtion can be found in: Stacke, T. & Hagemann, S. Development and evaluation of a global dynamical wetlands extent scheme Hydrol. Earth Syst. Sci., 2012, 16, 2915-2933
  • ORCHIDEE
    • Contact: patricia.cadule@lsce.ipsl.fr
    • Caveats or remarks:
      In the first set of ORCHIDEE simulations a land use pattern was erroneous applied. As the ISI-MIP protocol asks for a "pure natural vegetation run" the data set has been updated to correct for the problem in post-processing. Updated runs for the historical and rcp85 scenario were provided in January 2014 under the model name ORCHIDEE-NAT. Though, after the republish of the data in 2016 on the ESGF Node both versions will be available. Refrain from using the erroneous data.
  • PCR-GLOBWB
    • Contact: y.wada@uu.nl, dwisser@uni-bonn.de
    • Caveats or remarks:
      A detailed model description is given in: Wada, Y., Wisser, D., and Bierkens, M. F. P.: Global modeling of withdrawal, allocation and consumptive use of surface water and groundwater resources, Earth Syst. Dynam. Discuss., 4, 355-392, doi:10.5194/esdd-4-355-2013, 2013.
  • pDSSAT
    • Contact: jelliott@ci.uchicago.edu
    • Caveats or remarks:
      pDSSAT simulations for ISI-MIP use historical management data from SPAM, fertistat, and Potter et al. (2010). No tuning/calibration step is performed. Simulations include no adaptation: the planting window is fixed by the historical average for all future years as are all other management decisions (cultivar choice, fertilizer application, etc.).
  • UMU-WHOCCRAM
    • Contact: joacim.rocklov@umu.se
    • Caveats or remarks:
      Statistical malaria model that incorporate climate and socioeconomic parameters to map the spatial distribution in the disease using a spatial to temporal proxxy in climate change projections. A detailed model description has been provide in: Beguin A, Hales S, Rocklov J, Astrom C, Louis VR, Sauerborn R. The opposing effects of climate change and socio-economic development on the global distribution of malaria. Global Environ Chang. 2011;21(4):1209-14.
  • VECTRI
    • Contact: tompkins@ictp.it
    • Caveats or remarks:
      Dynamical malaria model that represents vectri larvae and adult growth cycles in addition to the parasite. Model optimized for falciparum transmitted by Gambiae in Africa. Can run at high spatial resolutions due to the temperature downscaling algorithm, the detailed surface hydrology and the incorporation of human population density at 5km resolution. A detail model description can be found in Tompkins A.M. and Ermert V, 2013: A regional-scale, high resolution dynamical malaria model that accounts for population density, climate and surface hydrology, Malaria Journal, DOI:10.1186/1475-2875-12-65. See online http://www.malariajournal.com/content/12/1/65
  • VIC
  • VISIT
  • WaterGAP
    • Contact: floerke@usf.uni-kassel.de, p.doell@em.uni-frankfurt.de
    • Caveats or remarks:
      Döll, P., Kaspar, F., Lehner, B. (2003): A global hydrological model for deriving water availability indicators: model tuning and validation. Journal of Hydrology, 270 (1-2), 105-134.
      Döll, P., Hoffmann-Dobrev, H., Portmann, F.T., Siebert, S., Eicker, A., Rodell, M., Strassberg, G., Scanlon, B. (2012): Impact of water withdrawals from groundwater and surface water on continental water storage variations. J. Geodyn. 59-60, 143-156, doi:10.1016/j.jog.2011.05.001
      Flörke, M., Kynast, E., Bärlund, I., Eisner, S., Wimmer, F., Alcamo, J. (2013): Domestic and industrial water uses of the past 60 years as a mirror of socio-economic development: A global simulation study. Global Environ. Change, 23, 144-156, http://dx.doi.org/10.1016/j.gloenvcha.2012.10.018




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.