APPENDIX 10
Memorandum submitted by the Met Office
1. The Met Office is the National Meteorological
Service of the United Kingdom and leads the world in weather and
climate prediction. It is an Executive Agency of the Ministry
of Defence and became a Trading Fund in 1996. The Met Office's
Hadley Centre for Climate Prediction and Research was established
in 1990, building on 20 years of research into climate variability
and climate change prediction. The work of the Hadley Centre on
which this submission is based, is largely funded by the Global
Atmosphere Division of Defra, with additional resources from the
Ministry of Defence and the European Commission. The scientific
results have been published in the peer reviewed literature (eg
Nature) or are being prepared for publication. Related evidence
has also been submitted to the Select Committee for Environment
and Rural Affairs review of climate change.
EXECUTIVE SUMMARY
2. The Met Office's Hadley Centre is able
to provide some key scientific results that inform the main questions
raised in the Environment Audit Committee's inquiry into climate
change. In summary, the issues covered below are:
(a) The acceleration of climate change
by interaction with biological processes
3. The last Inter-Governmental Panel on
Climate Change (IPCC) report concluded that the world is committed
to some change in climate over the next 40 years (and beyond)
due to man's emissions of greenhouse gases in the past; because
of inertia in the climate system. The choices that we make over
the next 20-30 years will determine changes in climate in the
latter half of this century. It has been recognised for some time
that the sensitivity of the climate system to changes in emissions
is likely to depend on how the natural carbon cycle responds to
climate change. The Hadley Centre were the first to include the
feedbacks between climate change and the carbon cycle in a realistic
climate model. We showed that the rise in global mean surface
land temperature between 2000 and 2100 could be around 3C
greater when the climate is allowed to interact with the carbon
cycle, compared to the previous model estimates, which omit the
link. These results are some of those that underpin the UK Government's
policy on Kyoto and succeeding negotiations. They are also relevant
to the practicalities of carbon monitoring. Understanding the
feedback of biological processes on global warming will be important
in deciding whether the greenhouse gas "safe limit"
required by UNFCC is scientifically valid.
(b) The impacts of forests on climate
change
4. The beneficial effect on climate of the
additional carbon sinks created by afforestation and reforestation
may be, at least partially, offset by changes in the surface reflectivity
as dark trees replace land cover that is lighter in colour. Consequently,
in many areas, the climate benefits of planting extra trees will
not be as great as their carbon "sink" potential suggests.
This is an important consideration in designing a regulatory framework
and in assessing the feasibility of emissions trading.
(c) Analysis and modelling of carbon
for the European land surface
5. Through a number of EU projects, scientists
at the Hadley Centre and throughout Europe are working to provide
the best estimates of carbon sources and sinks both historically
and in real time. This information will provide key support to
those monitoring and managing greenhouse gas emissions.
(d) Responsibility for mitigation
6. The Brazilian proposal and other similar
mechanisms provide frameworks that could be used to assign future
responsibility for mitigation to those with greatest responsibility
for past climate change. The Hadley Centre and other scientists
around the world are working together to come up with a robust
methodology to quantitatively estimate how future emissions reductions
might be divided between nations in an equitable way, should such
approaches be adopted by the international community. This information
will underpin negotiations post Kyoto, and inform negotiations
on contraction and convergence.
7. We would be able to present some of these
and other new results at the forthcoming Defra G8 scientific conference
at the Met Office's Hadley Centre, as recently announced by the
Prime Minister.
(a) The acceleration of climate change
by interaction with biological processes
Introduction to the carbon cycle
8. Carbon is continuously cycled between
reservoirs in the ocean, on the land, and in the atmosphere, where
it occurs primarily as carbon dioxide. On land, carbon occurs
primarily in living biota and decaying organic matter. In the
ocean, the main form of carbon is dissolved carbon dioxide and
small creatures, such as plankton. The largest reservoir is the
deep ocean, which contains close to 40,000 Gt C, compared to around
2,000 Gt C (Gigatons of Carbon) on land, 750 Gt C in the atmosphere
and 550 Gt C in the upper ocean. The atmosphere, biota, soils,
and the upper ocean are strongly linked. The exchange of carbon
between this fast-responding system and the deep ocean takes much
longer (several hundred years).
9. The ocean takes up carbon dioxide when
it is cold, at higher latitudes, and releases it near the tropics.
Photosynthesis takes carbon dioxide from the atmosphere and transfers
it to vegetation, while respiration releases carbon dioxide back
into the atmosphere. Although natural transfers of carbon dioxide
are approximately 20 times greater than those due to human activity,
they are in near balance, with the magnitude of carbon sources
closely matching those of the sinks. The additional carbon resulting
from human activity has raised levels of atmospheric carbon dioxide
by 30% over the last 150 years.
10. Changes in climate have a significant
effect on the carbon cycle. Increases in atmospheric carbon dioxide
concentration increase plant photosynthesis and the amount of
carbon stored in vegetation. However, increases in temperature
also lead to increases in plant and soil respiration rates, which
tend to reduce the size of the terrestrial carbon store. In some
regions, the changes in climate can also reduce plant photosynthesis
and reduce the ability of vegetation to sequester carbon.
Model predictions
11. Climate models predict that, as future
atmospheric carbon dioxide concentrations increase, due to fossil
fuel emissions and deforestation, the temperature of the planet
will also increase. This temperature increase is currently estimated
in two stages. Firstly, a model of the carbon cycle is used to
calculate the future atmospheric concentrations of carbon dioxide.
Secondly, the climate change is calculated using a separate global
climate model. However, in reality, climate change will alter
the, much larger, natural carbon cycle (see above) and this can
feed back on the climate change itself. Warming soils may emit
more carbon, and die back of vegetation may return carbon dioxide
to the atmosphere. A warmer ocean will take up less carbon dioxide
from the atmosphere. Furthermore, vegetation patterns move in
response to climate change. For instance, the tree line is predicted
to move poleward in the northern hemisphere. For the first time,
the Hadley Centre has coupled a representation of the carbon cycle
to a full climate model and made predictions of climate change
that incorporate climate-induced changes in the carbon cycle.
This has led to some radical new insight into the climate system.
12. Fig. 1 shows the atmospheric carbon
dioxide concentration predicted by the coupled carbon-cycle climate
model using greenhouse-gas emissions prior to present day and
IPCC business-as-usual (IS92a) emissions thereafter.
Fig 1 Simulated atmospheric concentrations (parts
per million by volume) of carbon dioxide when the two-way interaction
between climate and the carbon cycle is included. For comparison,
the results obtained when climate is not allowed to feed back
onto the carbon cycle are also shown. Prior to 1990, historical
emissions were used. Beyond 1990, emissions followed those in
the IPCC IS92a scenario.
Fig 2 Simulated global-mean temperature rise
over land with and without carbon-cycle feedback, as described
in the figure above.
13. The present-day carbon dioxide concentration
simulated by the model is in good agreement with observations
and the seasonal cycle of atmospheric carbon dioxide is also well
simulated, providing confidence in the future projections produced
with this new model. During the 21st century, the carbon dioxide
concentration in the coupled carbon-cycle climate model increases
faster than that predicted by previous models which neglected
carbon-cycle feedbacks. As a result, the rise in global mean surface
land temperature between 2000 and 2100 (below) is around 3C
greater when the climate is allowed to interact with the carbon
cycle.
14. The total global changes in soil and
vegetation carbon are shown in Fig 3. Maps of the change in terrestrial
carbon content between 1860 and 2100 are shown in Fig 4. The model
predicts that, in the second half of this century, vegetation
carbon storage in South America will begin to decline as a result
of the die back of the Amazon forest, which is caused by regional
warming and drying (direct anthropogenic deforestation is not
included). Around the middle of the century, the land biosphere
as a whole switches from being a weak sink for carbon to a strong
source, mainly due to the rapid loss in soil carbon beyond 2050.
In total, between the middle of the 19th century and the end of
the 21st century, the combined effects of climate change and increases
in atmospheric carbon dioxide concentration are predicted to reduce
global soil and vegetation carbon storage by around 100 Gt C.
15. Approaches to the UNFCC require greenhouse
gas concentrations to be established at levels that prevent dangerous
anthropogenic interference with the climate system. As we have
seen above, the concentration of the main anthropogenic greenhouse
gas, carbon dioxide, is dependent not only on anthropogenic production
but also on complex carbon cycle interactions. Furthermore, the
choice of levels will depend on decisions about land use, and
no choice can eliminate natural climate variability. The choice
also needs careful scientific scrutiny so that the implications
of the choice on land use, climate variability and climate change
are fully understood.
16. Because this is the first time the two-way
interaction between climate change and the carbon cycle has been
included in a full climate model, there is much uncertainty in
the results. Future work will look at the sensitivity of the model
to the representation of vegetation, soils and ocean carbon, and
improve these to increase the confidence in our predictions.
Fig 3 Simulated changes in the global total
soil and vegetation carbon content (Gt C) between 1860 and 2100
Fig 4 Patterns of change in the carbon content
of soil (top) and vegetation (bottom) predicted by the carbon
cycle-climate model between 1860 and 2100 (Not printed, information
is available from http://www.met-office.gov.uk)
(b) Climate effect on forestation
17. The Kyoto Protocol allows emissions
of greenhouse gases to be offset by the establishment of new forests
planted since 1990. However, will these forests actually slow
down climate change? The Hadley Centre climate model has been
used to quantify the effects of growing dense evergreen coniferous
forests at all the locations north of 30N that are capable
of sustaining them (Fig 5).
18. The results were compared with a situation
in which these locations were instead used as arable cropland.
The amount of extra carbon stored in the newly forested areas
(the sequestration potential) is shown below (Fig 5a). However,
trees not only absorb carbon dioxide, they have other effects
on climate. In particular, because they reflect different amounts
of sunshine than the underlying surface, they can alter the amount
of sunlight that is absorbed. Dark green forests absorb more of
the incoming solar radiation than arable cropland and will tend
to warm the planet. Estimates have been made of how much the new
forests would alter the climate through this mechanism.
19. The effect is greatest during the winter
months when large unforested areas are covered in highly reflective
snow, but when much of a forest canopy would remain above the
snow line. To compare the effect on climate of surface reflectivity
changes with that due to the capacity of the trees to sequester
carbon, the reflectivity effect has been expressed as equivalent
amounts of carbon emissions. A map of the equivalent emissions
is shown in Fig 5b.
Fig 5 (a) Estimated carbon uptake if suitable
arable land north of 30N were to be replaced with trees.
(b) The additional effect on climate of the changes in surface
reflectivity when trees are planted on suitable arable land north
of 30N, expressed as equivalent carbon emissions. (c) The
difference between the two diagrams above. Negative values show
where the net effect of planting trees is to warm climate. (Not
printed, information is available from http://www.met-office.gov.uk)
20. As expected, regions where the surface
reflectivity effect is most important are at high northern latitudes
in areas that have a winter covering of snow. In some boreal forest
locations, the changes in reflectivity reverse the beneficial
effects on climate from the uptake of carbon dioxide from the
atmosphere. In many other areas, the changes in reflectivity still
offset a large fraction of the sequestration potential.
21. These estimates have many uncertainties,
notably, the predictions of snow amount and surface reflectivity.
The calculations are also for a present-day climate, and changes
in temperature and atmospheric carbon dioxide concentration will
alter the results. However, the results do clearly show that the
beneficial effect on climate of the additional carbon sinks created
by afforestation and reforestation may be, at least partially,
offset by changes in the surface reflectivity as dark trees replace
land cover that was lighter in colour. Consequently, in many areas,
the climate benefits of planting extra trees will not be as great
as their carbon "sink" potential suggests.
(c) Carbon Assimilation and Modelling
of the European Land-Surface (CAMELS) and ocean (CASIX)
22. Under the Kyoto Protocol to the United
Nations Framework Convention on Climate Change, Annex I countries
are permitted to offset emissions of CO2 by changing land use
and land management to increase carbon accumulation. Methods include:
establishment of new forests (aforestation or reforestation),
forest management, cropland management, grazing land management
and revegetation. Loses of carbon from deforestation are also
accounted for. The related sources and sinks of CO2 must be reported
in a "transparent and verifiable manner". The Met Office's
Hadley Centre is leading an EU project called CAMELS, which will
provide key support to EU countries in meeting their obligations
under Kyoto, through the following products:
(i) Best estimates and uncertainty bounds
for the contemporary and historical land carbon sinks in Europe
and elsewhere, isolating the effects of direct land-management.
(ii) A prototype carbon cycle data assimilation
system (CCDAS) exploiting existing data sources (eg flux measurements,
carbon inventory data, satellite products) and the latest terrestrial
ecosystem models, in order to produce operational estimates of
"Kyoto sinks".
23. CAMELS will pioneer a highly innovative
method of estimating contemporary carbon fluxes, involving the
assimilation of observed data into terrestrial carbon cycle models.
The new scheme will be used to address the following questions:
— Where are the current carbon sources
and sinks located on the land and how do European sinks compare
with other large continental areas? The aim is to provide a consistent
estimate of the European land carbon sink by making intelligent
use of all of the existing data-sources.
— Why do these sources and sinks exist,
ie what are the relative contributions of CO2 fertilisation, nitrogen
deposition, climate variability, land management and land-use
change?
— How could we make optimal use of
existing data sources and the latest models to produce operational
estimates of the European land carbon sink?
24. One of the main products, which will
be made available to EU policy makers, will be high-resolution
maps of the European land carbon sink, which can be broken down
into the relative contributions arising from land management (as
covered under the Kyoto protocol) and other environmental factors.
25. A parallel system is also being developed
to estimate oceanic carbon uptake in real time. This is being
done as a close collaboration between the Met Office and the NERC
Centre of Observation of Air-Sea Interactions and Fluxes (CASIX).
It will exploit the Met Office's world-leading capability in operational
ocean modelling by assimilating real-time observations into the
Met Office FOAM (Forecasting Ocean-Atmosphere Model) system. In
combination with estimates of the terrestrial carbon cycle from
CAMELS, it will eventually be possible to establish an integrated,
near real time assessment of carbon sources and sinks on global
and regional scales.
(e) Using past responsibility for climate
change to estimate the share of future mitigation efforts
26. Future negotiations will require agreement
on how to divide effort on mitigation. One suggested method of
doing this is contraction and convergence. Another is the "Brazilian
proposal" (suggested by the Brazilians during negotiations
of the Kyoto protocol). Although this proposal was not adopted,
the Subsidiary Body on Scientific and Technical Advice (SBSTA)
requested that the methodological and scientific aspects of the
proposal be further studied.
27. The basis of the Brazilian proposal
is that future mitigation burdens should be divided up according
to past responsibility for climate change, evaluated using one
of a variety of indicators (such as temperature). The ad-hoc group
for the modelling and assessment of contributions to climate change
(MATCH, which includes Hadley Centre participation) is following
this up by improving the robustness of the calculations and assessing
the uncertainties more rigorously. The Hadley Centre has developed
its own simple tool for estimating the proportion of responsibility
to climate change indicators. The results are found to depend
on a range of scientific parameters and policy choices.
| Policy choices |
Scientific parameters |
| Start year for emissions | Chemistry model
|
| End year for emissions | Climate model
|
| Year for responsibility calculation | Type of responsibility calculation
|
| Choice of greenhouse gases | Choice of greenhouse gases
|
| Choice of climate indicator | Selection of historic emissions dataset
|
| Inclusion of aerosol |
Fig 6 An illustrative calculation of responsibility based
only on carbon dioxide emissions, mid-range (Bern) carbon cycle
parameters and a climate model tuned to HadCM3. The chosen indicator
is global mean temperature rise. It is important to recognise
that the values listed in the Figure are for a single set of scientific
and policy parameters (apart from emissions start year, of which
four are chosen). Choosing different but equally valid parameters
would alter the results.
28. In the illustration shown in Fig 6 the indicator
is temperature, and one parameter (namely the year for the responsibility
calculation) is varied. These results are affected by the timing
of emissions from each country or country group and delays in
the carbon cycle and climate system. Other parameter choices would
lead to different results. For example, if non fossil fuel greenhouse
gas emissions are included then the contributions of less developed
nations, with a greater reliance on agriculture, tend to be increased.
Another complexity is choosing the time lag between the last year
that emissions are included and the year of the responsibility
calculation. For indicators such as temperature there is a delay
between emissions and their effect on climate change so that making
the time difference between the last emissions and the year for
the responsibility calculation too short means that not all of
the climate change resulting from past emissions will have been
realised. Work is currently taking place to look at the sensitivity
of the result to a greater number of parameter choices and to
better understand the robustness of the assumptions and datasets
on which the method is based.
(f) Other relevant work
29. The Hadley Centre is involved in three new EU projects,
GEMS, CarboEurope and CarboOcean, which will use observations
and models to develop monitoring systems for CO2 and other environmental
variables. The ultimate aim is to have an operational monitoring
system. As part of this and other developments models will be
improved, enhancing our capability to predict future changes.
The Hadley Centre's climate models are being extended to allow
a fuller, more quantitative understanding of the earth system
as a whole, than has so far been possible. For example, the models
are currently being developed to allow the impact of changes in
mineral dust and oceanic iron to be assessed over the 21st century
(see section d above). By including such processes in climate
models it will be possible to better quantify and reduce the uncertainties
in climate projections and so provide focused advice to policymakers
on a wider range of issues and options.
30. It has been suggested that fertilisation of parts
of the ocean with a solution of iron could reduce atmospheric
carbon dioxide by enhancing phytoplankton growth and so increasing
the drawdown of carbon dioxide into the ocean. However, modelling
studies show that this approach is unlikely to significantly reduce
global atmospheric carbon dioxide, as other factors would quickly
come into play to limit phytoplankton growth. There is also good
reason to be concerned about side-effects of iron fertilisation
on the marine biosphere, for example reduced oxygen levels would
have an adverse impact on fish and other marine animals.
31. There is reason to believe that the natural supply
of iron to the ocean from mineral dust may change as a result
of global warming. The resulting phytoplankton changes may feed
back on the climate through emission of organic sulphur compounds
which modify cloud properties. The Hadley Centre's climate models
are currently being developed to provide a well-founded quantitative
estimate of the importance of this effect.
32. Aviation is not yet a major contributor to climate
change; however, aircraft emissions are growing rapidly. Furthermore,
although the contribution to climate change from aircraft carbon
dioxide emissions may be relatively straightforward to calculate,
aircraft also affect climate in many other ways, which are much
more uncertain. At the altitude where commercial aircraft cruise,
they will create condensation trails, which can either disappear
quickly or linger for hours depending on the meteorological conditions,
or, in many circumstances, develop into cirrus clouds. Contrails
and cirrus clouds have a warming effect on climate, but its magnitude
is very uncertain; IPCC show a range of 20 between the low and
high estimates of radiative forcing due to aircraft in 1992. It
will be important that this range of uncertainty is narrowed,
using a combination of experimental observations from aircraft,
theoretical calculations (for example of cirrus ice crystal radiative
properties) and climate modelling.
33. Other work in the Hadley Centre is directly relevant
to:
— Calculating the allowable greenhouse gas emissions
that would lead to various levels of climate change.
— Determining the impact of mitigation actions on
future climate (including stabilisation).
— Predicting dangerous climate change.
— Quantify the uncertainty in the predictions, which
will enable adaptation and mitigation ideas to be combined more
effectively with risk assessment methodologies.
28 October 2004
| 
