Lecture 3: Lithosphere and climate models

Truncated earth science

• Earth is 4.54 ± 0.05 billion years old

• Early Earth’s atmosphere had no oxygen and was mostly nitrogen and CO2

• Earliest life

  • 3.85 billion years ago – organic carbon found in rocks

  • bacteria found 3.46 billion years ago in Western Australia.

• Precambrian lasted from 4.5 billion to 540 million years ago when complex life started during the Cambrian

• Made up of lots of small proto continents

  • Tectonic and volcanic activity has reshaped the landmasses and the atmosphere, ocean and cryosphere many times during the Earth’s history

Plate tectonics

• The Earth’s plates are constantly changing through time

• There have been a few supercontinents

  • Pangaea 250Ma

  • Gondwana 200 Ma

• But also times when the continents are more spread out (like today)

• The distribution of the continents (and oceans) impacts the climate

Milankovitch cycles (external solar radiation)

• The orbit and tilt of the Earth changes overtime

• Determines how close the Earth is to the Sun, affecting the amount of solar radiation travelling to Earth (driving interglacial cycles

  • Eccentricity

    • Highly elliptical orbit

    • Nearly circular orbit

    • Obliquity

      • Axial tilt

      • Ellipse orbit

    • Precession

      • Precession of orbit

Why are we interested in past climate?

• Paleoclimate archives (e.g. ice cores, corals, marine and lake sediments, speleothems, tree rings, borehole temperatures, stalagmites, lake cores, soils) permit the reconstruction of climatic conditions before the instrumental era

• Establishes long-term context for the climate change
  beyond the past 150 years and the projected changes in the future

  • Methodical measurements of temperature on earth only started in ~1850

  • To understand natural climate variability and change through time we reconstruct past temperatures using proxies (which record a response to past temps or other climate variables (e.g. rainfall, winds))

Temperature and climate proxies

• Can be physical, chemical or biological

• Physical proxies

  • Colour, grain size, magnetism, texture

• Chemical proxies

  • Could be the concentration of elements, or chemical characteristics of those elements

• Biological proxies

  • Include the remains of living organisms and includes pollen, diatom and charcoal

CO2 feedback (warming)

External forcing (increase in insolation) → warmer oceanic temperatures (either direct or through ice sheet loss) → ocean CO2 solubility reduced → CO2 released from oceans into atmosphere → higher atmosphere CO2 concentrations drive greenhouse effect

CO2 feedback (cooling)

External forcing (decrease in insolation) → colder oceanic temperatures → ocean CO2 solubility increased → CO2 absorbed from atmosphere into oceans → lower atmosphere CO2 concentrations reduce greenhouse effect

Climate feedback loops

• Positive feedbacks

  • albedo

  • warming oceans

  • methane from wetland/permafrost and the oceans (clathrates)

  • land use - deforestation

  • bushfires

• Negative feedbacks

  • weathering

  • increased cloudiness

  • biological productivity

  • ocean solubility of CO2

• Climate models need to take these feedbacks into account

Paleoclimate modelling intercomparison project (PMIP)

• Same models as CMIP, but run using boundary conditions determined from paleoclimate archives

  • e.g. CO2 concentrations

• Model past climates which are significantly different from modern climates to test the model’s abilities to model these very different climate conditions

• Originally focused on mid-Holocene and Last Glacial Maximum, but extended to other time periods (Last Interglacial and Pliocene Warm Period)

• Targeting periods of time that we have enough proxy data to compare w

  • May provide analogues for the future

What is a climate model?

• A climate model is a computer program that model Earth or its climate system using a horizontal (latitude-longitude) or vertical grid (height or pressure or ocean depth)

• The model can conclude a number of different processes (e.g. ocean-atmosphere)’

• Climate model is made of:

  • governing equations (parameterisation)

  • forcing conditions

    • natural (solar, volcanoes)

    • human (greenhouse gas emissions, aerosol, land use change)

  • initial boundary conditions

• Global circulation models (GCM) which is a supercomputer produced model output

Hierarchy of climate models (increasing complexity)

• Simple

  • Energy Balance Models (EBM)

• Intermediate

  • 2D EBMs

  • simple Global Circulation Models (GCMs)

  • coupled Atmosphere-Ocean Global Circulation Models (AOGCM) (physical processes only)

  • fully coupled Earth Systems Models (ESM) (usually include biogeochemical processes and carbon cycle)

• Complex

  • Coupled Model Inter-comparison Project (phase) 6 (CMIP6)

*All models have biases (different parametrisation of processes) so combine results of the models to produce an average (or model ensemble)

Regional climate models

• Most global climate models (GCM) are relatively coarse resolution (~200 km)

• To get higher resolution you need to use a regional coupled model (RCM)

  • e.g. <10 km – which is nested within a coarse GCM – provide the boundary conditions.

  • This is called downscaling.

Stretched grid models

• Run for the entire globe, but have a higher resolution over the area of interest

• e.g. Conformal Cubic Atmospheric Model (CCAM)

Why do we run regional models?

• Regional models display:

  • Improved topography

  • Coastlines

  • Extremes in data

• They may not improve on large scale features inherited from the driving global model

Coupled Model Intercompatison Project (CMIP)

• computer-based models of the Earth's climate, where different parts (atmosphere, oceans, land, ice) are "coupled" together, and interact in simulations

• a collaboration of different worldwide modelling groups who run Coupled Models and run historical model runs and then future SSP scenarios

  • by comparing model outputs, they can
    look for agreement and disagreement (or combine the outputs to produce a Model Ensemble to cancel biases)

Evaluation of a CCAM model (example)

• AGCD gridded dataset is compared to model

• Compared the performance of the downscaled models and host models

• Also examines:

  • Extremes

  • Seasonality

  • Wet/dry days

Paleo climate models

• Can test models by comparing paleoclimate data

  • e.g. the last glacial maximum when it was colder

• Help to determine the climate sensitivity

  • e.g. temperature increase for a doubling of CO2

• Tests whether we understand all the
  climate processes and feedbacks between the
  different systems