Impact model: DLEM

Sector
Biomes
Region
global

Dynamic Land Ecosystem Model. (http://www.geog.com.cn/EN/10.11821/xb201009001)

Information for the model DLEM 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.

Person responsible for model simulations in this simulation round
Hao Shi: haoshi@rcees.ac.cn, 0000-0001-8250-0567, Chinese Academy of Sciences (China)
Hanqin Tian: hanqin.tian@bc.edu, 0000-0002-1806-4091, Boston College (USA)
Output Data
Experiments: obsclim_histsoc_default, counterclim_histsoc_default, obsclim_histsoc_1901co2, counterclim_1901soc_default, obsclim_1901soc_default, counterclim_histsoc_obsco2
Climate Drivers: 20CRV3, 20CRV3-ERA5, 20CRV3-W5E5, GSWP3-W5E5
Date: 2023-03-14
Basic information
Model Version: DLEM v4.0
Model Output License: CC0
Reference Paper: Other References:
Resolution
Spatial aggregation: regular grid
Horizontal resolution: 0.5’ x 0.5’
Temporal resolution of input data: climate variables: daily
Temporal resolution of input data: co2: annual
Temporal resolution of input data: land use/land cover: annual
Temporal resolution of input data: soil: constant
Input data
Observed atmospheric climate data sets used: GSWP3-W5E5 (ISIMIP3a)
Emissions data sets used: Atmospheric composition (ISIMIP3a)
Socio-economic data sets used: Historical land transformation
Land use data sets used: Historical, gridded land use
Climate variables: tasmax, tas, tasmin, rsds, pr
Spin-up
Was a spin-up performed?: Yes
Spin-up design: Following the ISIMIP3a protocol and using 100 years of spinclim data which is identical with the first 100 years of the counterclim data; the spin up length is up to a maximum of 12,000 years; deviations of 0.5 g C m-2 yr-1, 0.5 mm H2O m-2, yr-1 and 0.1g N m-2 yr-1, respectively, for carbon, water and nitrogen tolerances are used.
Natural Vegetation
Natural vegetation partition: Using the SYNMAP
Natural vegetation dynamics: The fraction of each natural vegetation type within a grid cell is proportional to the non-agricultural land
Natural vegetation cover dataset: SYNMAP
Soil layers: 10 layers
Management & Adaptation Measures
Management: Fertilizer and manure applications; wood harvesting; irrigation; crop harvesting; pasture and rangeland grazing
Extreme Events & Disturbances
Key challenges: Extreme events such as pests, fires, water logging and frost damage are not considered.
Model set-up specifications
How do you simulate bioenergy? i.e. what pft do you simulate on bioenergy land?: All bioenergy land are treated as cropland.
How do you simulate the transition from cropland to bioenergy?: Not considered.
How do you simulate pasture (which pft)?: C3 and C4 grasslands.
Key model processes
Dynamic vegetation: Prescribed.
Nitrogen limitation: Determined by the leaf C:N ratio.
Co2 effects: CO2 regulates net assimilation ("A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species." Planta 149(1): 78-90).
Light interception: The radiation transfer follows Chen et al. 1999 ("Daily canopy photosynthesis model through temporal and spatial scaling for remote sensing applications." Ecological Modelling 124(2–3): 99-119).
Light utilization: Beer's law.
Phenology: Dependent on the ratio of LAI over the maximum LAI.
Water stress: Derived as the root fraction-weighted sum of water potential limit in each soil layer.
Heat stress: Temperature constraints both Vcmax (an exponential function) and stomatal conductance (a quadratic function).
Evapo-transpiration approach: Transpiraiton and canopy evaporation is derived using the Penman-Monteith equation.; soil evaporation is calculated as the product of potential evaporation and a soil moisture scalar.
Differences in rooting depth: Static rooting distribution.
Root distribution over depth: The root fraction decreases exponentially along the rooting depth.
Closed energy balance: Yes.
Coupling/feedback between soil moisture and surface temperature: The phase change of soil water depends on soil temperature.
Latent heat: Derived from ecosystem evapotranspiration.
Sensible heat: Subtracting net radiation by latent heat and ground heat.
How do you compute soil organic carbon during land use (do you mix the previous pft soc into agricultural soc)?: When land use change occurs, the previous PFT SOC is mixed into the current PFT SOC.
Do you separate soil organic carbon in pasture from natural grass?: Yes.
Do you harvest npp of crops? do you including grazing? how does harvested npp decay?: There are harvest and grazing. Harvested NPP is partitioned into three product pools, with a turnover time of 1 year,10 years, and 100 years, respectively.
How do you to treat biofuel npp and biofuel harvest?: Treated the same as crops.
Does non-harvested crop npp go to litter in your output?: Yes.
Causes of mortality in vegetation models
Age/senescence: Daily mortality is a function of the maximum age.
Fire: Not invoked.
Drought: Drought is not treated specifically. The water stress depends on soil water availability.
Insects: Not considered.
Storm: No.
Stochastic random disturbance: No.
NBP components
Fire: No.
Land-use change: Yes. Most of the above-ground biomass is partitioned into different product pools and part is emitted into the atmosphere. The below-ground biomass is left and goes into litter or woody debris.
Harvest: Yes. Treated as the same as land conversion but the PFT is kept unchanged.
Other processes: Note that NBP loss due to soil erosion and leaching is subtracted from the land carbon pools.
Model output specifications
Output format: Following the ISIMIP3a specification.
Output per pft?: No. Output per grid.
Land-use change implementation
Is crop harvest included? if so, how?: Yes. Crop yields go into product pools while the residual is returned by a fixed fraction.
Is cropland soil management included? if so, how?: Not in this version.
Is grass harvest included? if so, how?: No.
Is shifting cultivation included?: No.
Is wood harvest included? if so, how?: Yes. The harvest fraction adopts data from LUHv2h data.
Which transition rules are applied to decide where agriculture is conducted?: Using prescribed crop distribution maps.
Person responsible for model simulations in this simulation round
Hanqin Tian: hanqin.tian@bc.edu, 0000-0002-1806-4091, Boston College (USA)
Additional persons involved: Jia Yang: (jzy0010@tigermail.auburn.edu), Wei Ren: (wei.ren@uky.edu), Shufen Pan: (panshuf@auburn.edu)
Output Data
Experiments: I, II, IIa, III
Climate Drivers: None
Date: 2017-09-18
Basic information
Model Version: v2.0
Reference Paper: Main Reference: Tian, Hanqin, Guangsheng Chen, Chaoqun Lu, Xiaofeng Xu, Daniel J. Hayes, Wei Ren, Shufen Pan, Deborah N. Huntzinger, and Steven C. Wofsy et al. North American terrestrial CO2 uptake largely offset by CH4 and N2O emissions: toward a full accounting of the greenhouse gas budget. Climatic change,129,,
Reference Paper: Other References:
Resolution
Spatial aggregation: regular grid
Horizontal resolution: 0.5°x0.5°
Temporal resolution of input data: climate variables: daily
Temporal resolution of input data: co2: annual
Temporal resolution of input data: land use/land cover: annual
Temporal resolution of input data: soil: constant
Input data
Simulated atmospheric climate data sets used: IPSL-CM5A-LR, GFDL-ESM2M, MIROC5
Emissions data sets used: Atmospheric CO2 concentration
Other human influences data sets used: N-Fertilizer (ISIMIP2), Nitrogen deposition (ISIMIP2)
Other data sets used: Land-sea mask
Additional input data sets: Vegetation data and nitrogen deposition are from the SYNMAP BIOME data, provided by MstMIP project (https://daac.ornl.gov/NACP/guides/NACP_MsTMIP_Model_Driver.html)
Climate variables: tasmax, tas, tasmin, rsds, pr
Exceptions to Protocol
Exceptions: selected monthly output only, grid-level output, rather than PFT level output
Spin-up
Was a spin-up performed?: Yes
Spin-up design: Before the transient simulation, we have equilibrium run and model spin-up. In the equilibrium run, we used the average climate condition during 1661-1690, and other input data (CO2 concentration, nitrogen deposition, vegetation type, land use and land management) at the level of year 1660 to run the model. We assumed the equilibrium state was reached when the differences of carbon, nitrogen, and water content at grid level are less than 0.5 g C/m2, 0.5 g N/m2, and 0.5 mm between two consecutive 50-year periods. The model determines how many model years the equilibrium run takes. The 100-year spin-up run follows equilibrium run. During this stage, model randomly selected climate conditions from the 30-year period of 1661-1690 to drive the model, while other driving forces (CO2 concentration, nitrogen deposition, land use and land cover) were kept at the level of 1661. When the equilibrium run and spin-up were done, the model started the transient run from 1661 to the last year of climate data. The transient run is forced by all the time-series of driving factors.
Natural Vegetation
Natural vegetation partition: The DLEM assume up to four natural vegetation types coexist in one grid. The DLEM did not simulate “natural vegetation”, but use SYNMAP BIOME classification data to drive our model, which provided the percentage of various vegetation types in each grid. For cropland, we use a crop system map to determine crop types (eg. wheat, rice, corn, etc) and rotation.
Natural vegetation dynamics: The fraction of natural vegetation was estimated according to cropland area in each grid.
Natural vegetation cover dataset: SYNMAP
Management & Adaptation Measures
Management: For cropland simulation, we used prescribed data on nitrogen fertilizer use and irrigation to force DLEM model.
Extreme Events & Disturbances
Key challenges: The current version of DLEM is capable of catching drought, heat wave and extreme cold, but unable to simulate flooding impact on ecosystems.
Key model processes
Dynamic vegetation: no
Nitrogen limitation: yes
Co2 effects: yes, Farquhar/Collatz photosynthesis
Light interception: big-leaf apporach
Light utilization: Farquhar/Collatz photosynthesis
Phenology: Fixed annual phenology based on satellite observation
Water stress: Influence on photosynthesis, carbon allocation, evapotranspiration, and soil biogeochemical processes
Heat stress: Influence on photosynthesis and repisration
Evapo-transpiration approach: Penman-Monteith equation, with the consideration of soil moisture availability
Differences in rooting depth: no
Root distribution over depth: PFT-specific root distribution by following Zeng 2001, "Global vegetation root distribution for land modeling"
Closed energy balance: DLEM does not calculate energy balance at land surface
Coupling/feedback between soil moisture and surface temperature: yes
Latent heat: yes, latent heat is based on ET (Penman-monteith equation)
Sensible heat: no
Causes of mortality in vegetation models
Age/senescence: yes, DLEM set a maximum age (AGEmax) for each PFT type. In each year, DLEM assume 1/AGEmax of this vegetation die due to aging effect.
Fire: no
Drought: no
Insects: no
Storm: no
Stochastic random disturbance: no
Other: no
Remarks: DLEM has a process-based fire module to estimate burned area, fire emissions, and fire mortality. However, we did not switch on Fire module in the ISIMIP simulation.
NBP components
Fire: no
Land-use change: yes, DLEM assumes that part of biomass transferred to litter and woody debris pool; part of them transferred to product pools; and the left directly released to the atmosphere due to burning.
Harvest: 1. harvested crop biomass tranferred to product pool. Crop residue and straw biomass tranferred to litter pool. 2. no forest harvest. 3. no grassalnd harvest.
Species / Plant Functional Types (PFTs)
List of species / pfts: Tundra (T); Boreal Broadleaf Deciduous Forest (BBDF); Boreal Needleleaf Evergreen Forest (BNEF); Boreal Needleleaf Deciduous Forest (BNDF); Temperate Broadleaf Deciduous Forest (TBDF); Temperate Broadleaf Evergreen Forest (TBEF); Temperate Needleleaf Evergreen Forest (TNEF); Temperate Needleleaf Deciduous Forest (TNDF); Tropical Broadleaf Deciduous Forest (TrBDF); Tropical Broadleaf Evergreen Forest (TrBEF); Deciduous Shrub (Dshrub); Evergreen Shrub (Eshrub); C3 grassland (C3G); C4 grassland (C4G); Cropland
Comments: Cropland (14 types including: Barly, Cassava, Cotton, Corn, Millet, Potato, Rapeseed, Rice, Sorghum, Soybeen, Sugarcane, Sunflower, Wheat, Oat), provided by Hanqin Tian - 30-10-2015
Model output specifications
Output format: DLEM output is based on per grid-cell area. when calculating global or regional total, one should multiple by grid area.
Output per pft?: DLEM output is based on grid-cell area. We did not provide output for each PFT type.
Land-use change implementation
Is crop harvest included? if so, how?: yes. Grain was be harvested at the end of growing season
Is cropland soil management included? if so, how?: no
Is grass harvest included? if so, how?: no
Is shifting cultivation included?: no
Is wood harvest included? if so, how?: no
Carbon-cycle benchmarking
Does your model reach a (near) steady state after spin up (characterized by nbp of < 0.2 pgc y-1)? (yes/no, provide number): yes. global NBP < 0.1 Pg C yr-1.
What is your modeled nbp for the 1990-2000 decade? is it within 1.2 +/- 0.8 gtc/yr (1-sigma) of observed data from o2/n2 trends (keeling and manning 2014) for 1990-1999 (yes/no, provide number): yes.
Person responsible for model simulations in this simulation round
Hanqin Tian: hanqin.tian@bc.edu, 0000-0002-1806-4091, Boston College (USA)
Additional persons involved: Jia Yang: (jzy0010@tigermail.auburn.edu), Wei Ren: (wei.ren@uky.edu), Shufen Pan: (panshuf@auburn.edu)
Output Data
Experiments: historical
Climate Drivers: None
Date: 2016-04-28
Basic information
Model Version: v2.0
Model Output License: CC BY 4.0
Reference Paper: Main Reference: Tian, Hanqin, Guangsheng Chen, Chaoqun Lu, Xiaofeng Xu, Daniel J. Hayes, Wei Ren, Shufen Pan, Deborah N. Huntzinger, and Steven C. Wofsy et al. North American terrestrial CO2 uptake largely offset by CH4 and N2O emissions: toward a full accounting of the greenhouse gas budget. Climatic change,129,,
Reference Paper: Other References:
Resolution
Spatial aggregation: regular grid
Horizontal resolution: 0.5°x0.5°
Temporal resolution of input data: climate variables: daily
Temporal resolution of input data: co2: annual
Temporal resolution of input data: land use/land cover: annual
Temporal resolution of input data: soil: constant
Input data
Observed atmospheric climate data sets used: GSWP3, PGMFD v2.1 (Princeton), WATCH (WFD), WATCH-WFDEI
Additional input data sets: Vegetation data and nitrogen deposition are from the SYNMAP BIOME data, provided by MstMIP project (https://daac.ornl.gov/NACP/guides/NACP_MsTMIP_Model_Driver.html)
Climate variables: tasmax, tas, tasmin, rsds, pr
Exceptions to Protocol
Exceptions: selected monthly output only, grid-level output, rather than PFT level output
Spin-up
Was a spin-up performed?: Yes
Spin-up design: Before the transient simulation, we have equilibrium run and model spin-up. In the equilibrium run, we used the average climate condition during 1901-1930, and other input data (CO2 concentration, nitrogen deposition, vegetation type, land use and land management) at the level of year 1901 to run the model. We assumed the equilibrium state was reached when the differences of carbon, nitrogen, and water content at grid level are less than 0.5 g C/m2, 0.5 g N/m2, and 0.5 mm between two consecutive 50-year periods. The model determines how many model years the equilibrium run takes. The 100-year spin-up run follows equilibrium run. During this stage, model randomly selected climate conditions from the 30-year period of 1901-1930 to drive the model, while other driving forces (CO2 concentration, nitrogen deposition, land use and land cover) were kept at the level of 1901. When the equilibrium run and spin-up were done, the model started the transient run from 1901 to the last year of climate data. The transient run is forced by all the time-series of driving factors.
Natural Vegetation
Natural vegetation partition: The DLEM assume up to four natural vegetation types coexist in one grid. The DLEM did not simulate “natural vegetation”, but use SYNMAP BIOME classification data to drive our model, which provided the percentage of various vegetation types in each grid. For cropland, we use a crop system map to determine crop types (eg. wheat, rice, corn, etc) and rotation.
Management & Adaptation Measures
Management: For cropland simulation, we used prescribed data on nitrogen fertilizer use and irrigation to force DLEM model.
Extreme Events & Disturbances
Key challenges: The current version of DLEM is capable of catching drought, heat wave and extreme cold, but unable to simulate flooding impact on ecosystems.
Key model processes
Dynamic vegetation: no
Nitrogen limitation: yes
Co2 effects: yes, Farquhar/Collatz photosynthesis
Light interception: big-leaf apporach
Light utilization: Farquhar/Collatz photosynthesis
Phenology: Fixed annual phenology based on satellite observation
Water stress: Influence on photosynthesis, carbon allocation, evapotranspiration, and soil biogeochemical processes
Heat stress: Influence on photosynthesis and repisration
Evapo-transpiration approach: Penman-Monteith equation, with the consideration of soil moisture availability
Differences in rooting depth: no
Root distribution over depth: PFT-specific root distribution by following Zeng 2001, "Global vegetation root distribution for land modeling"
Closed energy balance: DLEM does not calculate energy balance at land surface
Coupling/feedback between soil moisture and surface temperature: yes
Latent heat: yes, latent heat is based on ET (Penman-monteith equation)
Sensible heat: no
Causes of mortality in vegetation models
Age/senescence: yes, DLEM set a maximum age (AGEmax) for each PFT type. In each year, DLEM assume 1/AGEmax of this vegetation die due to aging effect.
Fire: no
Drought: no
Insects: no
Storm: no
Stochastic random disturbance: no
Other: no
Remarks: DLEM has a process-based fire module to estimate burned area, fire emissions, and fire mortality. However, we did not switch on Fire module in the ISIMIP simulation.
NBP components
Fire: no
Land-use change: yes, DLEM assumes that part of biomass transferred to litter and woody debris pool; part of them transferred to product pools; and the left directly released to the atmosphere due to burning.
Harvest: 1. harvested crop biomass tranferred to product pool. Crop residue and straw biomass tranferred to litter pool. 2. no forest harvest. 3. no grassalnd harvest.
Species / Plant Functional Types (PFTs)
List of species / pfts: Tundra (T); Boreal Broadleaf Deciduous Forest (BBDF); Boreal Needleleaf Evergreen Forest (BNEF); Boreal Needleleaf Deciduous Forest (BNDF); Temperate Broadleaf Deciduous Forest (TBDF); Temperate Broadleaf Evergreen Forest (TBEF); Temperate Needleleaf Evergreen Forest (TNEF); Temperate Needleleaf Deciduous Forest (TNDF); Tropical Broadleaf Deciduous Forest (TrBDF); Tropical Broadleaf Evergreen Forest (TrBEF); Deciduous Shrub (Dshrub); Evergreen Shrub (Eshrub); C3 grassland (C3G); C4 grassland (C4G); Cropland
Comments: Cropland (14 types including: Barly, Cassava, Cotton, Corn, Millet, Potato, Rapeseed, Rice, Sorghum, Soybeen, Sugarcane, Sunflower, Wheat, Oat), provided by Hanqin Tian - 30-10-2015
Model output specifications
Output format: DLEM output is based on per grid-cell area. when calculating global or regional total, one should multiple by grid area.
Output per pft?: DLEM output is based on grid-cell area. We did not provide output for each PFT type.