future climate Flashcards
lecture
what are climate models?
Climate models” are mathematical representations of all these physical processes and feedbacks, written in computer code.
what are the uses of climate models?
Predicting climate change:
*Will future climate be wetter/drier, warmer/colder?
*Will there be more extreme weather? (droughts, floods, heatwaves, hurricanes)
*Testing hypotheses
*What will happen if…? e.g. geoengineering
Driving impacts studies
*How will changes in climate affect agriculture, water supply, flood defence….
what do climate models generate…
large scale data
models can be used to…
understand the past
what did a 2012 model of arctic sea ice volume project ?
In 2012 a model of Arctic sea ice volume projected a disappearance of sea ice by 2015
* They redid the model the 3 years later. The updated fit suggested sea ice would disappear in 2035.
a model is used to represent the…
earths climate
simpler models allows us to learn about….
climate effects and feedbacks
Simple “box model” based on the idea that …
incoming energy must balance outgoing energy.
In a simple model, surface temperature is a function of:
The solar constant
- The planetary albedo
- The emissivity of the atmosphere,
where do the simple model values come from?
- they are assumptions
a and e change as temp changes
dependant on climate feedbacks
what is the climate feedbacks for ice-albedo, water vapour and cloud feedback?
Ice-albedo feedback: positive but weak
Water vapour feedback: positive and
strong
*almost doubling the effect of just increasing CO2
- Cloud feedback: extremely uncertain, but likely positive
where do estimates of climate feedbacks come from?
complex climate models - there is no scientific way of calculating them
what is the least and most temp rise expected from doubling co2?
AR6 report says likely to be 2.5 - 4 degrees
and says it could e 2-5 degrees
for example, if we double CO2 concentrations, the change in radiative forcing (ΔF) is approximately….
3.7 Watts per square meter (Wm^-2), according to scientific estimates.
how large is climate sensitivity?
Ice-albedo feedback
Warmer→ less snow/ice → lower albedo
Water vapour feedback
Warmer→ more water vapour → stronger greenhouse effect
Cloud feedback
Complicated changes in low and high clouds → changes to both albedo and greenhouse effect
what do climate scientists refer to complex climate models as?
Climate scientists refer to complex climate models as “GCMs” Global Climate Model
We also have RCMs – Regional Climate Models.
what are earth system models?
“Earth System Models” (ESM) are complex climate models that includes biogeochemical components.
since the 1980s climate models have included more elements of the….
earth system
A climate model “slices and dices” the atmosphere into thousands of 3-D cubes, about 100 km ….
by 100 km and about 500 m deep
The size of the grid boxes has been gradually decreasing since……
the 1980s.
This is largely a response to increasing computing power.
Smaller grid boxes mean that the equations are solved more accurately.
Smaller grid also means more detail.
Regional Climate Models (RCMs) cover a smaller part of the globe but in more detail than a GCM.
Typically, ………… (how many km square grids)
25 km square grids or even smaller.
Information from outside the RCM domain is fed in from an appropriate GCM.
A major use of RCMs is providing inputs for impact studies (agriculture, hydrology etc.).
what processes are parameterised in most climate models?
Radiation, and its interaction with the atmosphere, clouds and the Earth’s surface.
Surface processes: the exchange of heat, moisture and momentum between the Earth’s surface and the atmosphere.
Layer cloud formation and precipitation.
Convective cloud formation and precipitation
Parameterised schemes in GCMs account for the ……
the main differences between models.
They account for some degree of the uncertainty in climate predictions.
All the parameterization schemes currently in use in GCMs are based on sound physical principles, including detailed observations.
The current UK Met Office supercomputer is a……
Cray XC40, which can do 23,000 trillions calculations per second with 480,000 cores.
Even on such a powerful computer, is takes a month to simulate 100 years of Earth system changes.
what does the output of climate models get used for?
advise on policies for many years to come
whats the largest source of uncertainty and differences between models?
for the parameterisations. There are no exact equations
Models have improved in many respects from what years?
2007 (orange) to 2013 (purple)
what generations of emission scenarios have been created so far? (like reports and stuff)?
Several generations of scenarios have been created so far:
Special Report on Emission Scenarios (SRES), published in 2000.
Representative Concentration Pathways (RCPs), published in 2010.
Shared Socio-economic Pathways (SSPs), published in 2013.
Those scenarios look at the demographic, socio-economic and technological drivers of the emissions of greenhouse gases and other species.
GINI definition
“GINI is ameasure of statistical dispersionintended to represent theincome inequality, thewealth inequality, or theconsumption inequality” From Wikipedia
The scenarios need to assume ….
how the population and many other factors might evolve
what are SSPs for?
SSP scenarios specify concentrations and corresponding emissions of greenhouse gases and aerosol. They are then used in climate model simulations that are extensively used by IPCC
what is a climate projection?
A climate projection is a climate simulation that extends into the future based on a scenario of future external forcing
* Doesn’t say the climate will be like this
what is a climate prediction?
Climate forecast, can only predict up to about 10 years ahead
temperature changes depending on scenario could increase to,….. / stay at….
Temperature could increase to 5.5° C above pre-industrial levels by 2100
* Or stay ~1.5° C above pre- industrial levels
Patterns are defined for 3 Global Warming Levels which include:
*1.5°, 2°, 4°
warming is mostly in…
high latitudes
warming is strongest over…. l
land
warming of 4° globally could be > 7° over …. the a
the Arctic
MORE precipitation in tropics and high….
latitudes
less precipitation in ..
sub tropics
wet gets…..
wetter
dry gets…..
drier
In warmest scenarios Artic sea ice disappears in …. what month
september
Little variation in sea level rise …. by what year
2100
*Between 0.5m and 1.0m
It takes until what year to see the
impacts of scenarios on sea level
2300
*Between 0.5m and 7m *Can’t rule out > 15m!
After 3000 years the ice sheet has completely melted causing ,,,,
a 7m rise in global sea level.
we have 4 degree heatwaves pretty much every…. y
year
extreme rainfall occurs…. what percent more often now
30 percent - nearly 3 times more at 4 degrees
droughts have increased in frequency by what percentage?
70 percent at 4 degrees - quadrupled
the european heatwave was in… what year
2003
during the european heatwave how man
uk climate projections have a very complicated process involving:
Very complicated process involving:
-Use of multiple climate models
-Use of an “emulator” to simulate even more possible variations
-Use of observed data to evaluate different model results
-Downscaling to the local scale
what are the 3 types of uncertainty in climate projections:
- Internal variability: Uncertainty due to natural variations in the Earth’s climate on annual to decadal scales.
- Scenario uncertainty: Uncertainty in future greenhouse gas and aerosol concentrations
- Model uncertainty: Uncertainty due to imperfect models.
the larger the region - the smaller the …
internal variability
internal variability uncertainty is due to
natural climate variability reduces with the length of a climate forecast.
major model uncertainties include:
Carbon cycle behaviour
*Cloud responses
*Distribution of water vapour and temperature changes in the vertical *Land surface changes
*Ocean heat uptake
model uncertainty is….
Differences in projections from different models
scenario uncertainties are…
Differences in projections from different scenarios
Scenario uncertainty dominates….
at longer timescales and larger spatial scale
* Other uncertainties are more important at shorter timescales and smaller spatial scales
