Скачать презентацию 3 5 Climate Change Prediction 3 5 Скачать презентацию 3 5 Climate Change Prediction 3 5

36dbbfa2b539c8a1484ec331d0e06187.ppt

  • Количество слайдов: 92

3. 5 Climate Change Prediction 3. 5 Climate Change Prediction

3. 5 Climate Change Prediction (i) Climate Modeling (ii) Detection and Attribution of Climate 3. 5 Climate Change Prediction (i) Climate Modeling (ii) Detection and Attribution of Climate Change (iii) Climate Change Predictions

(i) Climate Modeling: The Need for Climate Models • • 3 Test of understanding (i) Climate Modeling: The Need for Climate Models • • 3 Test of understanding Evaluation of response Prediction of climate change Attribution of causes of climate change

Elements of a Climate Model • Atmospheric and oceanic circulation – Equations of motion Elements of a Climate Model • Atmospheric and oceanic circulation – Equations of motion for a fluid (air or water). These represent Newton's laws, mass conservation for the fluid and some thermodynamic relationships – They take the form of nonlinear partial differential equations. • Atmospheric radiation budget – Radiation absorbed, transmitted, reflected and scattered by each level of the atmosphere, in each wavelength band – Sensitive to the composition of the atmosphere, which varies in time and position • Hydrology, and water phase changes – Cloud processes on scales of 10's to 1000's of km – Sea ice and snow cover • Chemical reactions in the atmosphere and ocean – Affect composition, which feeds back on the radiation balance and the biosphere. • Energy flow in rocks and soils – Critical in determining the surface temperature and magnitude of the OLR – Sensitive to the nature of the soil and the soil moisture, which is strongly varying in time and space • Biosphere – Responses of plant growth and ocean plankton development to climatic changes and changes in CO 2 4

A ‘Hierarchy’ of Climate Models • AGCM – Atmospheric General Circulation Model – Simulates A ‘Hierarchy’ of Climate Models • AGCM – Atmospheric General Circulation Model – Simulates atmosphere but prescribes the oceans and land surface • OGCM – Ocean General Circulation – Simulates the ocean circulation, but with a simple atmosphere sufficient to provide surface wind stress and heat supply • AOGCM – Coupled Atmosphere-Ocean General Circulation Model – Used extensively in climate change experiments 5

Example Model Processes Relative humidity change Clouds Cloud formation Number of cloud drops Horizontal Example Model Processes Relative humidity change Clouds Cloud formation Number of cloud drops Horizontal motion of air Cooling of air Meteorology Onset of rainfall Ascent of air Rainfall rate Buoyancy change of air Radiation Start Change in solar radiation reaching surface Rate of absorption of solar radiation Change in surface moisture 6 Change in temperature of surface air Evaporation rate Surface/vegetation Response of vegetation

Model Structure • Discretisation – Splitting continuous quantities up into discrete units that can Model Structure • Discretisation – Splitting continuous quantities up into discrete units that can be acted on by the driving processes – Necessary because a model can carry information only at a fixed number of points – Averaging over large ranges • Examples – Spatial (lat, lon, altitude) – Aerosol and cloud particles (usually just mass) – Wavelengths (wavelength bands) 7 Typical model resolution is 2 o x 20 altitude levels, equivalent to 250 x 250 km x 1 km

Example of Discretisation • Simulation of a wind-blown cloud of pollution Discretised concentration Real Example of Discretisation • Simulation of a wind-blown cloud of pollution Discretised concentration Real Effect of wind distance Pollution blown by the wind as represented on the grid Discretisation causes reduction in ‘resolution’ (detail) Changes of the discretised quantity are not the same as those of the real quantity 8

Parameterisation • Simplification of processes in terms of simpler equations with physically or empirically Parameterisation • Simplification of processes in terms of simpler equations with physically or empirically derived parameters (which can be ‘tunable’) • Example for clouds: – Rainfall assumed to occur when the liquid water content of the cloud reaches a prescribed value – Reality is a highly complex interaction of different sized droplets, ice crystals, hail, etc. • Parameterisations capture the essence of real processes but they can be inaccurate and unreliable when used to make predictions under new conditions • Almost all processes are parameterised in climate models 9

Cloud Parameterisation RH Scheme: Assumes clouds form wherever the Relative Humidity is above a Cloud Parameterisation RH Scheme: Assumes clouds form wherever the Relative Humidity is above a certain value CW Scheme: Treats Cloud Water as a ‘prognostic’ model variable and distinguishes ice and water clouds, and the different precipitation from them All schemes have adjustable parameters that can be tuned to reproduce climatological cloud cover 10 However, in a double-CO 2 experiment, the RH scheme produced a 5 K warming, CW produced 3 K and CWRP produced 2 K. This result shows the problem with key climate model parameterisations! CWRP Scheme: As the CW scheme, but accounts for the change in cloud reflectivity with water content

Climate Models vs. Weather Forecast Models • Can be the same model – The Climate Models vs. Weather Forecast Models • Can be the same model – The UK ‘Unified Model’ is used for Met Office weather forecasts and climate prediction • Climate models are mostly ‘free running’ – Day-to-day weather patterns not used, but average ‘climatic’ state should be OK • Weather forecast models are ‘nudged’ to match observations as much as possible –‘Data assimilation’ 12

Evolution of climate models (IPCC 2007) Grid size resolution 13 Evolution of climate models (IPCC 2007) Grid size resolution 13

Evolution of climate models (IPCC 2007) How has model performance improved? 14 IPCC (2007) Evolution of climate models (IPCC 2007) How has model performance improved? 14 IPCC (2007)

Two Major Problems with Early Climate Simulations • Ocean heat ‘flux adjustments’ – A Two Major Problems with Early Climate Simulations • Ocean heat ‘flux adjustments’ – A non-physical ‘adjustment’ to ocean heat content to account for incomplete ocean physics (failure to resolve narrow ocean currents, such as found in the N Atlantic) • Cloud responses – Clouds remain one of the largest uncertainties in climate response simulations – Cloud feedbacks still responsible for a large part of inter-model differences – IPCC 2007 15

Model Comparisons With Observations • Models do not simulate the current weather, but only Model Comparisons With Observations • Models do not simulate the current weather, but only a climatological state consistent with the prescribed forcings (greenhouse gas content of the atmosphere, aerosols, etc. ) • Need to evaluate models against average climate over, say, 1 year • Can also look at ‘typical’ seasonal cycle or typical El Nino variations, but not for any particular year 16

Climatological Temperature • Absolute error generally < 2 o. C • Slight general cold Climatological Temperature • Absolute error generally < 2 o. C • Slight general cold bias Labelled contours: climatological SST and surface air temp Colours: mean model error from several models 17 IPCC (2007)

Climatological Precipitation General pattern very good Obs Model Dry bias: problems modelling monsoon Errors Climatological Precipitation General pattern very good Obs Model Dry bias: problems modelling monsoon Errors in Indo-Pacific warm pool affects ability of model to capture teleconnections (El Nino) IPCC (2007) 18

Summary of Climatological Experiments from AR 4 • Confidence in model simulations has improved Summary of Climatological Experiments from AR 4 • Confidence in model simulations has improved since previous IPCC (2001). • Increased confidence from models no longer needing ‘flux adjustments’ – These models are able to maintain stable climates over centuries – Some biases and long-term trends remain – Tropical precipitation a problem – Clouds remain a key uncertainty in models 19

20 th Century Climate Variability 58 models driven by changes in natural and anthropogenic 20 th Century Climate Variability 58 models driven by changes in natural and anthropogenic forcings Obs Mean of models IPCC (2007) 20

Simulation of ENSO • Climate models have substantially improved spatial representation of pattern of Simulation of ENSO • Climate models have substantially improved spatial representation of pattern of SST anomalies in S Pacific - Better physics - Increased resolution • Some even used to forecast ENSO • SST gradients in equatorial Pacific still not well captured - Thermoclines too diffuse • Most models produce ENSO variability on timescales faster than observed • Helped by further increases in model resolution? 21

Extreme Weather • Climate models are not weather forecast models, so they can’t simulate Extreme Weather • Climate models are not weather forecast models, so they can’t simulate individual events during a long simulation (of perhaps 100 years) • We need to test the models’ variability • • 22 Temperature: Simulation of hot and cold extremes has improved, with large regional discrepancies. Rainfall: Frequency of intense events and amount of precipitation during them are underestimated. Extra-tropical storms: These are storms affecting mid-latitude regions, such as northern Europe. These are well captured by models – improved since 2001. Tropical cyclones: Frequency and distribution captured well by some models – improvement since IPCC 2001

(ii) Detection and Attribution of Climate Change • Anthropogenic climate change occurs against a (ii) Detection and Attribution of Climate Change • Anthropogenic climate change occurs against a backdrop of natural climate variability • Internal variability – Climate variability not forced by external agents – All time-scales (weeks to centuries) • Externally forced variability – Natural (volcanic, solar) – Anthropogenic (greenhouse gases, aerosols) – not forgetting. . . Changes in natural variability • Detection of anthropogenic climate change within all this other climate variability is a statistical “signal-tonoise” problem 23

Definitions • Detection – Demonstrating that an observed change is significantly different (in a Definitions • Detection – Demonstrating that an observed change is significantly different (in a statistical sense) from that which can be explained by natural internal climate variability – Detection does not imply an understanding of the causes • Attribution – The isolation of cause and effect 24

Problems with Attribution • Reality – Statistical analysis of observational record – Demonstrate that Problems with Attribution • Reality – Statistical analysis of observational record – Demonstrate that observed changes are: • unlikely to be due entirely to internal variability • consistent with estimated/anticipated responses (models) • inconsistent with alternative explanations (models) • Limited data and imperfect model – Proof of cause and effect (100% agreement) impossible – Relies on rejecting alternatives – Incomplete knowledge means that “new alternatives” are still emerging 25

Measures of Confidence Used by IPCC 2007 26 Measures of Confidence Used by IPCC 2007 26

Requirements for Successful Detection and Attribution • Good data – Sufficient coverage to identify Requirements for Successful Detection and Attribution • Good data – Sufficient coverage to identify main features of natural variability – So far, surface and upper air temperatures have been used – Other climate variables used for ‘qualitative’ assessment (changes broadly consistent) 27

climate quantity Example of Need for Quality Data time 28 climate quantity Example of Need for Quality Data time 28

climate quantity The Need for Long Data Records time 29 climate quantity The Need for Long Data Records time 29

Beware of Correlations! 30 Beware of Correlations! 30

Beware of Correlations! • Temperatures have increased since 1700 to present • The number Beware of Correlations! • Temperatures have increased since 1700 to present • The number of pirates has decreased since 1700 to present Does this mean lack of pirates is causing climate change? ? ? The existence of a correlation does not indicate a causal mechanism 31

Quantifying Internal Climate Variability • From the instrumental record – Relatively short (compared to Quantifying Internal Climate Variability • From the instrumental record – Relatively short (compared to 30 -50 year period of interest) – Coverage incomplete, and varies with time • Paleoclimatic data – Reconstructions of climate before anthropogenic perturbations – Poor resolution and global coverage – Contains unknown external forcings • GCM ‘control’ runs over long periods (1000 years) 32

The Magnitude of Modelled Natural Variability 3 climate models run with no external forcings. The Magnitude of Modelled Natural Variability 3 climate models run with no external forcings. All variability is due to internal climate processes. These simulations are compared with observations in the right-hand panels. No evidence for model ‘natural variability’ anything like recent changes 33

A reminder of what we are dealing with: Estimated Forcings since pre-industrial times (IPCC A reminder of what we are dealing with: Estimated Forcings since pre-industrial times (IPCC 2007) 34

Radiative Forcing • Definition: A change in the net radiation at the top of Radiative Forcing • Definition: A change in the net radiation at the top of the atmosphere due to some external factor.

Net Radiation Net radiation = Incoming - Outgoing Positive net radiation n Incoming > Net Radiation Net radiation = Incoming - Outgoing Positive net radiation n Incoming > Outgoing Negative net radiation n Outgoing > Incoming

Positive & Negative Forcing • Positive forcing warming • Negative forcing cooling Positive & Negative Forcing • Positive forcing warming • Negative forcing cooling

Forcing and Feedbacks Radiative forcing (external) Internal response (including feedbacks) Climate system Forcing and Feedbacks Radiative forcing (external) Internal response (including feedbacks) Climate system

Forcing and Feedbacks • “Forcing” is produced by an external process, e. g. – Forcing and Feedbacks • “Forcing” is produced by an external process, e. g. – Changes in solar flux – Volcanic eruptions – Human actions • A feedback is a response to temperature changes – Example: Increased water vapor due to warming

More Anthropogenic increases in greenhouse gases are considered forcings Increases in greenhouse gases that More Anthropogenic increases in greenhouse gases are considered forcings Increases in greenhouse gases that are caused by temperature changes are feedbacks

 • The same gas can be involved in forcings and feedbacks, e. g. • The same gas can be involved in forcings and feedbacks, e. g. , CO 2 • Forcing: – CO 2 increase from burning of fossil fuels • Feedback – temp decay CO 2

Comparing Causes of Temperature Change • Assumption: Larger radiative forcing larger effect on temperature Comparing Causes of Temperature Change • Assumption: Larger radiative forcing larger effect on temperature • Comparisons follow Source: Intergovernmental Panel on Climate Change (IPCC)

Positive Radiative Forcings • Largest – by far: increased greenhouse gases – Increase is Positive Radiative Forcings • Largest – by far: increased greenhouse gases – Increase is almost entirely anthropogenic

Long-Lived Greenhouse Gases Gas Forcing (Wm-2) CO 2 1. 66 CH 4(methane) 0. 48 Long-Lived Greenhouse Gases Gas Forcing (Wm-2) CO 2 1. 66 CH 4(methane) 0. 48 N 2 O (nitrous oxide) Halocarbons 0. 16 Total 2. 64 0. 34

More About Greenhouse Gases Radiative transfer model Adding greenhouse gas reduces outgoing longwave radiation More About Greenhouse Gases Radiative transfer model Adding greenhouse gas reduces outgoing longwave radiation (OLR) at top of atmosphere

Initial Equilibrium Top of atmosphere Absorbed Shortwave OLR Now, add greenhouse gas Keep temperatures Initial Equilibrium Top of atmosphere Absorbed Shortwave OLR Now, add greenhouse gas Keep temperatures fixed

Reduced Upward Flux Top of atmosphere Absorbed Shortwave OLR Reduced Upward Flux Top of atmosphere Absorbed Shortwave OLR

Net Downward Flux Top of atmosphere Net Flux Result: A positive radiative forcing Net Downward Flux Top of atmosphere Net Flux Result: A positive radiative forcing

Negative Radiative Forcings Largest: Increase in sulfate aerosols n Mostly anthropogenic Negative Radiative Forcings Largest: Increase in sulfate aerosols n Mostly anthropogenic

Anthropogenic Sulfate Aerosols • Coal and diesel fuel contain sulfur • Burning of these Anthropogenic Sulfate Aerosols • Coal and diesel fuel contain sulfur • Burning of these fuels produces sulfur dioxide (a gas) • In the atmosphere, this gas is converted into particles

Effect of Anthropogenic Sulfate Aerosols on Temperature • Direct effect – The aerosols themselves Effect of Anthropogenic Sulfate Aerosols on Temperature • Direct effect – The aerosols themselves reflect sunlight – This is similar to the effect of volcanic aerosols • Indirect effect – Sulfate aerosols act as condensation nuclei – This increases the droplet concentration in clouds – Result: Increased cloud albedo • Both effects tend to increase the Earth’s albedo

Evidence for Indirect Effect Bright streaks areas of enhanced albedo Cause: Emissions from ships Evidence for Indirect Effect Bright streaks areas of enhanced albedo Cause: Emissions from ships Streaks called “ship tracks”

Cause of Ship Tracks • Ship exhaust contains aerosols • The aerosols cause more Cause of Ship Tracks • Ship exhaust contains aerosols • The aerosols cause more droplets to form • Cloud albedo is increased

Total Anthropogenic Effect on Climate • Total Anthropogenic Climate Forcing = sum of all Total Anthropogenic Effect on Climate • Total Anthropogenic Climate Forcing = sum of all anthropogenic forcings • Mainly, greenhouse gases (+) + sulfate aerosols (-)

Net Anthropogenic Radiative Forcing (1750 – 2005) Best Estimate: 1. 6 Positive. 2 W/m Net Anthropogenic Radiative Forcing (1750 – 2005) Best Estimate: 1. 6 Positive. 2 W/m

Carbon Dioxide • From fossil fuel burning • ~60% contribution to total radiative forcing Carbon Dioxide • From fossil fuel burning • ~60% contribution to total radiative forcing • Atmospheric concentration increased from 280 ppm in 1750 to 380 ppm in 2005 (36%) • 1999 – 2005 CO 2 fossil fuel / cement emissions increased by ~3% / yr • Today’s CO 2 concentration has not been exceeded during the past 420, 000 years and likely not during the past 20 million years. • The rate of increase over the past century is unprecedented, at least during the past 20, 000 years 58

Methane Trends IPCC (2007) Factor 2. 5 levelling off of upward trend not understood Methane Trends IPCC (2007) Factor 2. 5 levelling off of upward trend not understood 59

Trends in Halocarbons Radiative forcing peaked in 2003 – now beginning to decline 60 Trends in Halocarbons Radiative forcing peaked in 2003 – now beginning to decline 60

Using Forcing-Response Relationships for Detection and Attribution • Use the temporal and spatial variation Using Forcing-Response Relationships for Detection and Attribution • Use the temporal and spatial variation of the different forcings • Can separate natural and anthropogenic influences only if spatial and temporal responses are known – Climate record: Different responses are superimposed – impossible to separate – Climate model: Study responses to individual forcings 61

Example of model Forcing-Response Patterns Solar Volcanic Well mixed GHGs Direct sulfate Temp change Example of model Forcing-Response Patterns Solar Volcanic Well mixed GHGs Direct sulfate Temp change 18901999 (o. C / century) 66 Ozone Total

Can natural forcings explain Global Warming? • A climate model including only natural forcings Can natural forcings explain Global Warming? • A climate model including only natural forcings (solar + volcanic aerosol) does not explain the temporal change in global mean temperature IPCC (2007) 67

Can natural forcings explain Global Warming? Models with both natural and anthropogenic forcings do Can natural forcings explain Global Warming? Models with both natural and anthropogenic forcings do far better IPCC (2007) 68

Regional response to natural and anthropogenic forcings 69 Regional response to natural and anthropogenic forcings 69

Detection of natural and anthropogenic signals Contribution from GHGs, other anthropogenic and natural foircings Detection of natural and anthropogenic signals Contribution from GHGs, other anthropogenic and natural foircings to temperature changes between 1990 s and 1900 s. 70

Conclusions • It is extremely unlikely (<5%) that the global pattern of warming during Conclusions • It is extremely unlikely (<5%) that the global pattern of warming during last 50 years can be explained without external forcing. • Greenhouse gas forcing has very likely caused most of warming over last 50 years. • It is likely that there has been a substantial anthropogenic contribution to surface temperature increases in every continent except Antarctica since the middle of the 20 th century. • Recommend read summary and conclusions to Chapter 9 IPCC AR 4. 71

(iii) Climate Change Projections (IPCC 2013) • Climate model experiments • Projections of future (iii) Climate Change Projections (IPCC 2013) • Climate model experiments • Projections of future climate (IPCC AR 5) and • ‘Geo-engineering’ IPCC AR 5 - Chapter 12 72

Radiative Forcing ~2. 3 W/m 2 Radiative Forcing ~2. 3 W/m 2

Radiative Forcing: aerosols • See notes 74 Radiative Forcing: aerosols • See notes 74

Climate Projections • It is not possible to make deterministic projections of future climate Climate Projections • It is not possible to make deterministic projections of future climate change. • It is not even possible to make projections of all possible outcomes (as is “done” in medium range weather forecasting). • Projections are uncertain because: (i) They are primarily dependent on scenarios of future anthropogenic and natural forcings that are uncertain. (ii) incomplete understanding and inprecise models of climate system. (iii) Presence of internal variability.

Climate Projections • The word “projection” is used to reflect the uncertainties and dependencies. Climate Projections • The word “projection” is used to reflect the uncertainties and dependencies. • Nevertheless as greenhouse gas concentations continue to rise we expect changes in the climate system to be greater than those already observed. • It is possible to understand future climate change using models to characterise likely outcomes and uncertainties under specific assumptions about future forcing scenarios.

Description of Scenarios • Previous IPCC reports based projected emissions on socioeconomic scenarios – Description of Scenarios • Previous IPCC reports based projected emissions on socioeconomic scenarios – from storylines based on future demographics and economic development, regionalisation, energy production and use, technology, agriculture, forestry, and land-use. Models were then forced with an appropriate level of GHGs and aerosols. • A new set of scenarios were created for AR 5 – so-called “Representative Concentration Pathways” – these are focused on the net radiative forcing rather than on the atmospheric constituents reflecting the multiple “pathways” that can result in he same radiative forcing. • They are defined by their net radiative forcing by 2100.

Description of Scenarios • Four scenarios are used: RCP 2. 6 is the lowest Description of Scenarios • Four scenarios are used: RCP 2. 6 is the lowest of the four peaks at 3. 0 Wm-2 and declines to 2. 6 Wm-2 by 2100. RCP 4. 5 (medium-low) stabilizes at 4. 2 Wm-2 by 2100. RCP 6. 0 (medium-high) stabilizes at 6. 0 Wm-2 by 2100. RCP 8. 5 (high) reaches 8. 3 Wm-2 by 2100, on a rising trajectory.

A 1 B - Rapid economic growth, population peak mid-century, balance across sources A A 1 B - Rapid economic growth, population peak mid-century, balance across sources A 2 – Heterogeneous world, continuously increasing population, self-reliance B 1 – Convergent world, population peak mid-century, clean technologies IS 92 a – “business as usual” (as understood in 1996), 1% growth in CO 2 concentration per year

80 80

Geo-engineering • Implementation of man-made or artificial enhancement of natural negative radiative forcing to Geo-engineering • Implementation of man-made or artificial enhancement of natural negative radiative forcing to counteract increase in GHGs • Could be alternative to mitigation or adaptation? Some proposed methods: - Artificial enhancement of the sulfur cycle through e. g. addition of iron to the oceans - Enhancemnet of marine cloud reflectance through increased sea-salt fluxes – i. e. ‘man-made sea spray’ - ‘Space mirrors’ - Injection of sulphate aerosol into the stratosphere – compare with volcanic eruptions 93

Geo-engineering Ocean-going sea-spray producers proposed by Latham and Salter 94 Geo-engineering Ocean-going sea-spray producers proposed by Latham and Salter 94

Geo-engineering For: “An opportunity to ‘buy time’ while technologies are improved to enable effective Geo-engineering For: “An opportunity to ‘buy time’ while technologies are improved to enable effective emissions reductions and development of non-fossil fuel energy sources” “Not an alternative to mitigation” Against: “At best, a last resort to preserve habitability when all else fails; at worst, dangerous interference with the Earth system with catastrophic consequences” “Insurance policies can encourage risky behaviour” 95

CO 2 Net RF Radiative forcing Mitigation and geo-engineering CO 2 Net RF Time CO 2 Net RF Radiative forcing Mitigation and geo-engineering CO 2 Net RF Time Albedo modification Geoengineering instead of mitigation 96 Time Albedo modification Geoengineering to ‘take edge off the heat’

Geo-engineering – open questions Only tackles radiative GHG impacts – e. g. ocean acidity Geo-engineering – open questions Only tackles radiative GHG impacts – e. g. ocean acidity would remain problem Who decides that we go ahead with it? Are the unknowns presented by uncertainty in the future climate state any more dangerous than those introduced by deliberate man-made interference? If we have the ability to act, should we do it sooner rather than later? Public/political perceptions – would the existence of these tools be seen as a carte-blanche for uncontrolled emissions? Unforeseen impacts – ‘perpetrator’ and ‘victim’ determinable? international conflicts? The case for Geoengineering by Dr. David Keith http: //www. ted. com/talks/view/id/192 97