L5 Flashcards
Land-surface models/ vegetation models
- Computer based model
- Used to simulate land processes
- Numerical models that solves the coupled fluxes of water, energy and carbon between the land-surface and atmosphere, within a context of direct and indirect human forcings and ecological dynamics
- They are also known as terrestrial carbon models, dynamic global vegetation model (DGVM)
- Names come from configurations they have
Land surface models work for simplified configurations eg for:
- Meteorology
- Soil ecology
- Vegetation dynamics
- Hydrology
Land surface models also work for full-complexity configurations eg:
- These resolve many cycles/ patterns etc
How can we use land surface models?
independently or within an Earth system model
We can use land-surface models independently or within an Earth System Model
- An earth system model is a set of models that can be run independently or together to simulate the Earth global climate
- Land model is an important component of Earth System models
- Land- surface models can involve complex code all centred by the coupler which connects eg land ice to ocean
- These models are ran in high performance computing systems (HPC)
Examples of terrestrial feedbacks within the Earth system (components of land-surface model)
- Snow cover and climate
- Soil moisture- evapotranspiration- precipitation
- Land use and land cover change
- Carbon cycle
- Reactive nitrogen
- Chemistry-climate (BVOCs, O3, CH4, aerosols)
- Biomass burning (fires)
Land is a critical … of CO2 emissions
Sink
- Trees involved in the model produce CO2 etc
- These models can tell us implications of future emissions etc
Many land-surface models available
- Often ran in conjunction with one another
- 16 land-models participated in the 2022 carbon budget
- They vary in what they feature eg fires vs no fires
- Often have different configurations but are made by the same companies
How do land-surface models work in the global carbon budget?
- Take an average from the land models results
- Provide certain elements towards the carbon budget others eg gas reserve etc which are not in land-surface models
How are land models interdisciplinary?
- Started off very simple then added more ecosystem functions to the land model
- Evolved from land as a lower boundary to the atmosphere to land as an integral component of the Earth system
- Developed along with computer capability and research questions
What is the role of a land model in an Earth system model?
Land-atmosphere exchanges
- Energy, water vapour, CO2, dust, trace gases
Land surface states - Soil moisture, soil temperature, canopy temperature, snow water equivalent, C and N stocks Land surface characteristics - Soil texture, surface roughness, albedo, emissivity, vegetation type, LAI index
- When using a land model all these fluxes are passed from land to atmosphere whilst conserving each other
What are the main components of a land model?
- Surface energy fluxes
- Hydrology
- Biogeochemical cycles
The land surface model solves surface energy balance , surface water balance and carbon balance equations at each model timestep
surface energy balance inputs or forcings
- Direct solar and Incoming long wave radiation
surface energy balance outgoings
Response fluxes
- Reflected solar - Outgoing longwave
The land surface affects the energy balance through what three properties?
Albedo
Surface Roughness
Evapotranspiration
Land models can look at the … effect
temperature effect eg converting forest to cropland
albedo directly affects…
net radiation
trees are darker than grasses and have lower albedo
surface roughness affects…
sensible and latent heat fluxes
trees are tall and aerodynamically rough
grasses are warmer and drier
latent heat flux…
evaporatively cools
trees transpire more water than grasses
influence of deforestation on climate - if we took all forest and converted it to grassland (on the three properties)
- Albedo only shows a decrease in temperature
- Trees absorb more radiation as they are darker
- Evapotranspiration shows a warming effect
- Forests evapotranspire more than grass and cool
Surface roughness shows a warming effect
net response (annual mean)
- Deforestation both increases and decreases temperature depending on location
- Deforestation cools higher latitudes but warms tropical latitudes
surface water balance
- Also needs to resolve other water balances such as snow and soil water
- Precipitation comes in
- Various evaporation and runoff methods
- Models also needs to resolve change in soil moisture over time
surface carbon balance
- Model computes net ecosystem exchange (NEE)
- Depends on GPP
- GPP = gross primary productivity (photosynthesis)
- Losses out of the system from respiration, carbon lost through fire and land use change
- Resolves plant and soil carbon pools
land use change … total C storage
decreases
- When cropland is replaced - Forest stores more carbon than grasslands
Land modelling challenges: Land surface heterogeneity
- Models are ran at 100 x 100km resolution (low)
- Take into account heterogeneity by the process of subgrid tiling
Subgrid tilling
- Grid cell is divided by land unit
- Land units involve eg vegetated, lake, urban, glacier, crop
- Soil column is then placed into natural land units where soil occurs
- Patches are then defined in vegetated column into plant functional types
Urban land unit
- Features configurations eg roofs, sun walls, shade walls,
Glacier
- Elevation classes
Crops
Unrrig, irrig
How does land unit work?
- Each grid cell has a percentage of the type of land unit
- We don’t know eg where lakes will be, but we know the percentage of it lakes shall make up
- Information comes from satellite data
- Resolve energy etc for each land unit and provides an output per grid cell weighted depending on land unit percentage
- Land surface models are parameter heavy
Where do parametrisations in land models come from?
- Laboratory understanding
- Of plant physiological processes
- Empirical relationships
- From as large a sample of the real world as possible, from field and satellite observations - Optimality theory
- Plants are rational actors, on average
Sources of uncertainty in land-surface modelling
- Forcing (scenario) uncertainty, GHG emissions scenarios, land use etc
- Response (model) uncertainty from parameterisations, resolution etc
- Using multiple models corrects for this - Internal (natural or unforced) variability; initial value problems
- Running several temperatures eg air temperature
- Several experiments
- Response (model) uncertainty from parameterisations, resolution etc
Paths to reducing model uncertainty
model intercomparisons
model intracomparison
model benchmarking
model data-fusion
comparison to real- world manipulative experimens
model hierarchy
conclusions
- Land models enable scientists and policymakers to better understand and manage the Earth’s natural resources, and to address some of the most pressing environmental challenges for our time
- Land-surface models are a starting point for the science, but not the science itself
- Land surface models are very complex and multidisciplinary, however are easy to run, but hard to interpret
