Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

NATIONAL CENTRE FOR EARTH OBSERVATION

VIEWS: 17 PAGES: 36

									                                        2




NATIONAL CENTRE
FOR EARTH OBSERVATION
NCEO HIGHLIGHTS OF 2009/10
and contributions from the Centre for
Earth Observation Instrumentation




www.nceo.ac.uk
Foreword
The National Centre for Earth Observation is one of the NERC’s research
and collaborative centres. Our core activity is fundamental science: the
challenge of understanding and predicting the behaviour of planet Earth
as a complex, multi-component coupled system using observations from
space and mathematical models. This year’s NCEO stories illustrate:
•    The value of long term quality controlled data sets to monitor climate trends

•    The challenges of converting photon counts to geophysical parameters

•    Improvements in the understanding of Earth system processes such as
     atmosphere/ocean gas exchanges, spatial and temporal changes of glaciers and
     sea ice, seismic hazards

•    The importance of data assimilation to test models stringently and to initialise
     for prediction

This year our core science programme has been augmented by a number of new projects
awarded in response to the NCEO Mission Support call and managed jointly with NERC.
The main awards are described briefly in this brochure.

This year we have also developed closer ties with our sister programme the Centre for
Earth Observation Instrumentation (CEOI). CEOI’s main activities are described.

One very exciting development is the International Space Innovation Centre (ISIC) at
Harwell. NCEO is driving the development of the EO aspects of the centre. The vision is
for ISIC to have global presence, a ‘shop window’ promoting the best of UK science and
technology to customers worldwide encouraging inward investment and developing new
markets. ISIC will exploit the presence of the newly established ESA Facility at Harwell,
delivering greater value from our ESA investments.

I would like to thank my NCEO colleagues for the great science they are doing. I am
especially grateful to Jan Fillingham for the photographs in this brochure and to our
guest editor Felicity Richardson, University of Bristol.

                                                                            Alan O’Neill
                                                                            NCEO Director
Table of Contents
Monitoring the changing planet
2    Realising long-term datasets of land surface temperature
4    The challenges of estimating biomass and carbon stores in sub-tropical woodlands
5    Operational deployment of SEVIRI Fire radiative power product
6    The sea level rainbow: when is a sea level trend really a trend?
7    Accelerated thinning of Pine Island glacier

Understanding Earth system processes
8    The 2009 L’Aquila earthquake: faulting and implications for seismic hazards
9    The 2010 Haiti earthquake – InSAR constraints of fault slip extent and future
     seismic hazard
10 The recent Karonga earthquake sequence in the southern East African Rift
11   Determining wetland methane emissions from spaceborne observations of methane
     and gravity
12 How accurate are the radiative properties of ice clouds derived from the CloudSat
   and Calipso satellites?
14 Improved understanding of changes in the global water cycle
15 Radiative forcing from persistent aircraft contrail cirrus case study
16 Physical Models of air-sea gas exchange and whitecaps
17   Incorporation of melt ponds into an Arctic sea ice model

Data assimilation
18 Data assimilation for modelling the carbon cycle
19 Deployment of a preliminary version of a convective-scale ensemble
   prediction system
20 Efficient nonlinear data assimilation
21 Assimilating Atlantic MOC changes, upper ocean temperatures and sea level

Data and model intercomparisons
22 SPARC CCMVal Report
24 Prediction of storms by ensemble prediction systems

Legacy datasets and quality assurance
26 NCEO’s lasting legacy datasets
27 Successful airborne campaign in the Arctic



28 Mission support
30 Centre for Earth Observation Instrumentation
32 International Space Innovation Centre
    MONI TORING T HE
    CH A NGING PL A NE T
    REALISING LONG-TERM DATA SETS OF LAND SURFACE TEMPERATURE
    John Remedios, Olof Zeller, Ed Comyn-Platt and Darren Ghent, University of Leicester; Chris Taylor,
    Heather Ashton and Richard Ellis, Centre for Ecology and Hydrology, Wallingford.

    Land surface temperature (LST) is a key parameter,                    demonstrate that the timing and amplitude of the bias are due to
    measurable from space, which both acts as a                           an overly strong control of soil water on evapotranspiration, which
    mediator of surface-atmosphere processes and is in                    develops in the model in June. Similar behaviour is found in a belt
    turn reflective of surface and atmosphere changes.                    from England down to Southern France. The next step is to use the
                                                                          LST data to improve the depiction of soil water and evaporation
    Evapotranspiration and surface energy budgets of
                                                                          in the climate model and see how it affects projections of future
    surfaces with low foliage e.g. grasslands or bare soils
                                                                          heat waves and droughts. This requires further investigation and
    are strongly influenced by soil moisture and surface                  utilisation of high quality LST data and improvements in the model
    soil temperatures. In turn, land cover changes force                  generated by rigorous intercomparison of models and data.
    local changes in surface temperature and soil moisture
    conditions whilst greenhouse gas radiative forcing and
                                                                                                               John Remedios
    atmosphere circulation changes can drive long-term,
    global averages of surface temperature.                                                                    ‘Research in NCEO is
                                                                                                               identifying some of the
    In the work performed within the NCEO, research has                                                        key factors that govern
    concentrated on two main aspects related to climate. Firstly,                                              the ability to model
    generation of high accuracy Earth Observation (EO) data sets for                                           evapotranspiration
    LST from the Along Track Scanning Radiometer (ATSR) series of                                              correctly.’
    instruments has been a high priority, to be followed by equivalent
    analyses for geostationary observations. The ATSR instruments
    extend backwards to 1991 (although current data quality for LST
    limits data sets to the 1995 period onwards) and forwards to
    the ENVISAT mission and to the Sentinel-3 operational mission
    to be launched in 2013. Analysis of ATSR data therefore offers
    the capacity to produce climate length time series of LST for
    comparison to models. Secondly, comparisons of LST with the UK          References:

    land surface-atmosphere model, JULES, have been performed with          Ghent, D., J. Kaduk, J. Remedios, and H. Balzter,
                                                                             Data assimilation into land surface models: the
    a view to understanding better the surface interface control in the      implications for climate feedbacks. International
    model. Research in NCEO is identifying some of the key factors           Journal of Remote Sensing, accepted 2010
    that govern the ability to model evapotranspiration correctly,          Noyes, E. J., G. Sòria, J. A. Sobrino, J. J. Remedios,
    particularly in drought stress related regimes.                          D. T. Llewellyn-Jones, and G. K. Corlett, (2007)
                                                                             AATSR land surface temperature product algorithm
    Figure 1 shows LST derived from AATSR for the snow event for             verification over a WATERMED site, Adv. Space Res.,
                                                                             39, 171–178
    January 8th 2010 using an improved LST algorithm developed
                                                                            Taylor, C. M., (2010), Feedbacks on convection from an
    within NCEO. The new LST retrieval is supported by ESA for
                                                                              African wetland, Geophys. Res. Lett., 37, L05406
    implementation to future processing of AATSR data and it also
                                                                            Taylor, C. M., P. P. Harris, and D. J. Parker (2010),
    forms the basis for the initial algorithm specification for the Sea       Impact of Soil Moisture on the Development of a
    and Land Surface Temperature Radiometer (SLSTR) on Sentinel-3             Sahelian Mesoscale Convective System: A Case Study
    which will have a wider swath than the ATSR instruments.                  from the AMMA Special Observing Period, Quart J
                                                                              Roy Meteorol Soc, 136, 456–470
    The second figure shows a comparison of satellite LST data              Ellis, R. J., C. M. Taylor, G. P. Weedon, N. Gedney, D. B.
    (MODIS) with JULES simulations. The model LST is 6K larger                Clark, and S. Los (2009), Evaluating the Simulated
                                                                              Seasonality of Soil Moisture with Earth Observation
    than the observations during August, with the model (red)                 Data, J. Hydromet., 10(6), 1548–1560
    over-predicting LST and under-predicting evapotranspiration             Taylor, C. M. (2008), Intra-seasonal land-atmosphere
    during July/August. This comparison helps us to identify errors           coupling in the West African monsoon, J. Climate,
    in the simulation of European summertime climate and how                  21(24), 6636–6648
    it will change in the coming decades. Sensitivity simulations



2
                                                                                                          3




                                                   Figure 1. LST derived for the UK from AATSR
                                                   data on January 8th 2010. The LST uses a high
                                                   resolution biome map including urban area
                                                   definitions. There is a small residual effect due to
                                                   clouds on this image; cloud detection over land
                                                   will be addressed by a new NCEO study led by
                                                   Edinburgh and including Swansea, Leicester and
                                                   King’s College London.




Figure 2. Comparison of average monthly
LST from MODIS satellite observations with a
simulation of JULES using standard Met Office
parameters, forced by observations over a region
of Northern France (Ashton, Taylor and Ellis, in
preparation).
    MONI TORING T HE
    CH A NGING PL A NE T
    THE CHALLENGES OF ESTIMATING BIOMASS AND CARBON STORES
    IN SUBTROPICAL WOODLANDS
    Mathew Williams, Casey Ryan, Emily Woollen and John Grace, University of Edinburgh

    Miombo woodlands are the dominant ecosystem                          biomass dynamics across landscapes. International efforts to
    in Southern Africa, containing significant carbon                    reduce emissions of carbon from deforestation, and to alleviate
    stocks in the vegetation and soils. This carbon is                   poverty by maintaining forest cover, rely on timely and accurate
    lost to the atmosphere when woodlands are cleared                    information of this kind.
    for agriculture or felled for charcoal production.
    Deforestation rates are high, but poorly quantified
    and understood. These woodlands are also subject to
    other disturbances - they are regularly burned - and
    so they present a varied, patchy mosaic landscape
    that complicates mapping of carbon stocks. At the
    University of Edinburgh, we have been investigating
    ways of accurately measuring ecosystem carbon
    stocks and their changes over time, using a
    combination of ground based surveys and Earth
    observation (EO) techniques.
    Using EO is difficult in woodlands, because a dense grass
    understory lies beneath a sparse tree canopy, where most
    vegetation carbon is stored. Traditional approaches, using
    optical satellite imagery to estimate biomass using a measure
    of “greenness”, are problematical because “greenness” cannot
    differentiate between grass and trees. We are using two
    approaches to overcome this problem. Our field work has identified
    critical periods in late October when trees come into leaf while
    grass remains dead. Optical imagery from this period can then
    be used to its greatest potential for mapping tree biomass. The      Figure 1. The field studies were incorporated into an African calibration
                                                                         of ALOS PALSAR, a radar sensor. The figure shows (a) HH and (b) HV
    second approach is to use radar imagery to map biomass. Radar
                                                                         backscatter (σ0) plotted against field-measured above ground biomass
    backscatter does not respond to grass biomass, only to structural    (AGB, Mg/ha) for four sites combined, with the x-axes shown with
    elements of trees. The two methods allow independent estimates       conventional and log10 scales. Second order log regression lines are fitted.
    of biomass to be generated.                                          The results suggest that a widely applicable general relationship exists
                                                                         between AGB and L-band backscatter for lower-biomass tropical woody
    However, EO data in themselves do not provide direct estimates       vegetation.
    of biomass and carbon stocks. Instead, they must be calibrated
    against ground data, including direct measurements of stem
    numbers and sizes. Most of the wood carbon is in the stems
    of a few, very large trees, which are unevenly distributed
    across the landscape. Thus, ground sampling, even with a large
    effort, may not deliver accurate estimates of carbon stocks.            References
    We have developed sampling techniques that help capture the             Mitchard, E. T. A., S. S. Saatchi, I. H. Woodhouse, G.
                                                                             Nangendo, N. S. Ribeiro, M. Williams, C. M. Ryan, S.
    heterogeneity of the distributions of trees across the landscape,
                                                                             L. Lewis, T. R. Feldpausch, and P. Meir, (2009), Using
    using nested plots for ground-based sampling. These samples              satellite radar backscatter to predict above-ground
    can then be used to calibrate EO data, which in turn can upscale         woody biomass: A consistent relationship across four
    results to the landscape, to produce reliable carbon maps. The           different African landscapes, Geophys. Res. Lett.,
                                                                             36, L23401, doi:10.1029/2009GL040692
    next challenge is to develop procedures for monitoring carbon/




4
                                                                                                                                                         5




OPERATIONAL DEPLOYMENT OF SEVIRI FIRE RADIATIVE PRODUCT
Martin Wooster, Gareth Roberts, Patrick Freeborn and Weidong Xu, King’s College London

A new operational satellite product has been generated                   ‘gridded’ product including attempts at some bias adjustment
to aid research, monitoring and forecasting of the                       (Freeborn et al., 2009). We are working with colleagues in the EU
effects of the many millions of km² of grassland and                     FP7 Monitoring Atmospheric Composition and Climate (MACC)
forest burned annually around the globe. Fire is one of                  project to exploit these products as source terms in the prototype
                                                                         GMES Core Atmospheric Service (www.gmes-atmosphere.eu/
the most widespread terrestrial ecosystem disturbance
                                                                         services/gac/fire). The products will be generated from the current
agents and a key way in which humans affect the
                                                                         Meteosat series until ~ 2016, and the design of Meteosat Third
Earth’s atmosphere. Wildfire activity shows large inter-                 Generation includes ‘fire dedicated’ spectral channels that will
annual variations, and quantification is most commonly                   enable further product improvements after that system becomes
undertaken by estimating the mass of vegetation                          operational.
burned, usually calculated via EO-derived burned area
measures coupled with estimates of the amount of
fuel consumed per unit area. Unfortunately the latter
parameter is often highly uncertain, and in any case
the approach only provides data once the fire has
occurred and gives no information on the rate at which
smoke is being released to the atmosphere, limiting the
ability to couple this to atmospheric transport models.
To support atmospheric modelling, monitoring and short-term
forecasting, we have developed an alternative approach based
on Fire Radiative Power (FRP) measurements derived from
geostationary satellites. This provides near real-time information
on fuel consumption rates, and working with scientists at
EUMETSAT headquarters and the Land Surface Analysis Satellite
Application Facility (landsaf.meteo.pt/) we have developed this
into a Meteosat SEVIRI operational product.
Two versions are available, a native high resolution product
(Roberts et al., 2009), and a spatio-temporally aggregated




                                                                         Figure 2. Continental map of the annual sum of FRP (in MW) derived for
                                                                         2004 at 10 grid cell resolution for Africa. Angola, Sudan and Central African
                                                                         Republic are particular areas where fire emissions peak.




                                                                            References
                                                                            Freeborn, P. H., M. J. Wooster, G. Roberts, B. Malamud,
                                                                              and W. Xu, (2009), Development of a virtual active
                                                                              fire product for Africa through a synthesis of
Figure 1. The Global Fire Emissions service of the EU FP7 MACC project        geostationary and polar orbiting satellite data,
provides FRP-derived biomass burning emissions estimates as input for         Remote Sensing of Environment, 113, 1700–1711
atmospheric monitoring and modelling. Figure shows mean daily FRP           Roberts, G., M.J. Wooster, and E. Lagoudakis, (2009),
density (mW/m²) calculated at 125 km gird cell resolution from global        Annual and diurnal African biomass burning
FRP data of 13 May 2010. See www.gmes-atmosphere.eu/data/ for                temporal dynamics, Biogeosciences, 6, 849–866
atmospheric products derived using these data.
    MONI TORING T HE
    CH A NGING PL A NE T
    THE SEA LEVEL RAINBOW: WHEN IS A SEA LEVEL TREND
    REALLY A TREND?
    Chris W. Hughes, Simon D. P. Williams and Joanne Williams, National Oceanography Centre, Liverpool

    Sea level does not simply rise and fall uniformly over                             changes from place to place. We find that there is a rainbow in the
    the ocean, in fact some regions may fall while others                              ocean: high frequencies (~30 days) dominate near the equator, with
    are rising. This is a result of changes to the Earth’s                             progressively lower frequencies dominating at higher latitudes, up to
                                                                                       ~100 days at 30 degrees north and south. When these spectra are
    gravity field and flexing of the solid Earth as mass
                                                                                       translated into colours, they result in a rainbow, with blues near the
    moves around the planet, and of changes in ocean                                   equator, progressing to reds at higher latitudes.
    currents. In order to make decisions about how to
    protect coasts around the world, we need to know not                               This means that we cannot use a simple spectral model to estimate
                                                                                       trend errors. Instead, we have fitted a 6-parameter model of the
    just the global mean sea level change, but the regional
                                                                                       spectrum at each point in the ocean, which captures the observed
    pattern too. Unfortunately, model predictions of such                              variation in shape of the spectrum. Using this, we can determine
    patterns do not agree with one another. The models                                 where the observed trends are significant and where they are not.
    need to improve, and to do this we need observations                               This shows that there is a significant trend in global mean sea level
    to see where they differ. But first we need to know                                (3.2 mm/year over the period 1995–2007, with a statistical error of
    whether the measured and modelled trends are                                       0.1 mm/year), but also that there are significant regional differences
    significant. You can fit a trend to any time series, but if                        in sea level trend. For example, sea level around the North Atlantic
    the time series is mostly oscillations with a very small                           coast, and around the northern and eastern Pacific has risen
                                                                                       significantly more slowly than the global average (in fact it has
    trend, the value obtained can be meaningless. In order
                                                                                       fallen in some regions). Knowing these patterns are meaningful,
    to quantify whether the fitted trend is significant, we                            we can now seek to determine what needs to be improved in ocean
    need to account not only for the size of the oscillations,                         models in order to reproduce these patterns.
    but also their spectrum: high frequency oscillations
    will average out well, but long-period oscillations will
    not, and can contaminate the trend estimate.
    We have used 12 years of satellite altimeter measurements of
    sea level to look at how the spectrum and amplitude of variability




                                                                                       Figure 2. The spectrum of sea level variability, as it would be seen
                                                                                       translated into a spectrum of light. Bright regions indicate a lot of
                                                                                       variability and dark regions indicate little variability. Colours show the
                                                                                       dominant frequency, with blue representing a period of ~30 days and red
                                                                                       ~100 days. The rainbow effect in the tropics shows how high frequencies
                                                                                       dominate near the equator with progressively lower frequencies further
                                                                                       away, although this relationship breaks down at higher latitudes.


    Figure 1. This shows how the sea level trend at each point differs from the           References
    global average of 3.2 mm/year, over the period 1995-2007. Contours enclose            Hughes, C. W., and S. D. P. Williams, (2010), The colour
    regions of trends which are significantly higher (black contour) or lower (red         of sea level: The importance of spatial variations
    contour) than the global average. It is clear that sea level in some regions has       in spectral shape for assessing the significance of
    been rising faster than the global average, while in other regions it has been         trends, J. Geophys. Res. (submitted)
    rising more slowly or even falling over this period.




6
                                                                                                                                                                   7




                               Accelerated thinning of Pine Island glacier

ACCELERATED THINNING OF PINE ISLAND GLACIER
         Duncan Wingham and David Wallis, University College London
                                              Andrew Shepherd, University of Leeds
Duncan Wingham and David Wallis, University College London; Andrew Shepherd, University of Leeds




Figure 1. Thinning rate in the environs of the PIG in (a) 1995, and (b) 2006. Also shown are the boundaries of the glacier central trunk (dashed black line) and
 Figure 1. Thinning rate in the environs of the PIG in (a) 1995, and (b) 2006. Also shown are the boundaries
the glacier tributaries (black line) and the 1996 grounding line (thick black line).
of the glacier central trunk (dashed black line) and the glacier tributaries (black line) and the 1996 grounding
line (thick black line).
We have used ERS-2 and ENVISAT satellite radar                                               Andrew Shepherd
altimetry to examine spatial and temporal changes in
                                                                                                                ‘The pattern of thinning
 We have used ERS-2 Pine Island Glacier, West
the rate of thinning of theand ENVISAT satellite radar                            References
                                                                                                                has both accelerated and
 altimetry to examine period 1995 temporal changes in
Antarctica, during the spatial and to 2008 (Wingham
 the rate of
et al., 2009). thinning of the Pine Island Glacier, West
                                                                                                                spread inland
                                                                                  Wingham, D. J., D. W. Wallis, and A. … if the Shepherd,
 Antarctica, during the period 1995 to 2008 (Wingham                                                            acceleration
                                                                                  (2009), Spatial and temporal evolution ofcontinues at its
                                                                                                                               Pine Island
                                                                                                                present rate, the main trunk
The pattern2009). The both accelerated and spread inland
 et al., of thinning has pattern of thinning has both                             Glacier thinning, 1995-2006, Geophysical Research
                                                                                                                of the glacier will be afloat
to encompass tributaries flowing into inland to encompass
 accelerated and spread the central trunk of the                                  Letters, 36
                                                                                                                within some 100 years.’
 tributaries the central trunk, central trunk of the glacier.
glacier. Withinflowing into thethe average rate of volume loss
 Within the central trunk, the average ± 0.3 km3/yr in
quadrupled from 2.6 ± 0.3 km3/yr in 1995 to 10.1 rate of volume
                                              3
2006. The region of lightly grounded ice km /yr in 1995 to 10.1
 loss quadrupled from 2.6 ± 0.3 at the glacier terminus is
             3                                                                                                             Andrew Shepherd
 ± 0.3 km /yr in and the changes inland are consistent with
extending upstream, 2006. The region of lightly grounded
 ice at the prolonged disturbance to the ice flow, upstream,
the effects of a glacier terminus is extending such as the                                                      “The pattern of thinning
 and the changes inland are consistent with the
effects of ocean-driven melting. If the acceleration continues at                                                 has both accelerated
 effects of a prolonged disturbance to the ice flow,
its present rate, the main trunk of the glacier will be afloat within                                            and spread inland…if
 such as the effects of ocean-driven melting. If the
some 100 years, six times sooner than anticipated.                                                                   the acceleration
 acceleration continues at its present rate, the main
                                                                                                                continues at its present
 trunk of the editor’s will be afloat within some
This work was an glacierhighlight and #1 weekly download in 100                        References
                                                                                                                 rate, the main trunk of
 years, was widely reported than anticipated.
GRL, and six times sooner in the media, including BBC News at                          Wingham, D. J., D. W. Wallis, and A. Shepherd, (2009),
                                                                                                                the glacier will be afloat
10 and The Times.                                                                       Spatial and temporal evolution of Pine Island
                                                                                                               within some 100 years.”
 This work was an editor’s highlight and #1 weekly                                      Glacier thinning, 1995-2006, Geophysical Research
 download in GRL, and was widely reported in the                                         Letters, 36
 media, including BBC News at 10 and The Times.
    UNDERS TA NDING E A R T H
    S YS T EM PROCES SES
    THE 2009 L’AQUILA EARTHQUAKE: FAULTING AND IMPLICATIONS
    FOR SEISMIC HAZARD
    Richard Walters, John Elliott, Philip England and Barry Parsons, University of Oxford; James Jackson,
    University of Cambridge




                                                                                                        Figure 1. An interferogram of the
                                                                                                        L’Aquila earthquake, draped over digital
                                                                                                        topography. Each of the multicoloured
                                                                                                        interference fringes in the interferogram
                                                                                                        represents a contour of ground motion
                                                                                                        towards or away from the satellite. The
                                                                                                        area SW of the Paganica fault (shown
                                                                                                        by the black line) moved away from
                                                                                                        the satellite by around 25 cm, and the
                                                                                                        area NE of the fault moved towards the
                                                                                                        satellite by around 8 cm. These data
                                                                                                        were used to model the earthquake
                                                                                                        and to identify the Paganica fault as its
                                                                                                        source.


    On the 6th April 2009 a magnitude 6.3 earthquake                      hazard. In addition, the L’Aquila earthquake occurred in an area
    struck L’Aquila, the capital city of the Abruzzo region,              with a marked seismic deficit relative to GPS-measured strain
    central Italy, killing around 300 people and making                   accumulation (Hunstad et al., 2003) and we determined that the
    tens of thousands of people homeless. L’Aquila is                     recent earthquake can only have reduced this deficit by a small
                                                                          amount. These results were reported in Geophysical Research
    a medieval city, nestled in the Apennines, and it
                                                                          Letters in a paper that was selected as the Editor’s Choice in the
    has a long history of similar earthquakes that have
                                                                          issue in which it appeared. The work also stimulated a number
    repeatedly damaged the city.                                          of newspaper and journal articles, e.g. in Nature and American
    We worked with partners at University College London, the             Scientist.
    University of Leeds and Istituto Nazionale di Geosifica e
    Vulcanologia, Italy, combining field observations of the surface
    rupture, geomorphological analysis, Interferometric Synthetic
    Aperture Radar (InSAR) and seismological measurements to
    study the earthquake. We modelled the L’Aquila earthquake using
                                                                            References:
    InSAR and seismic body waves, enabling the Paganica fault to be
                                                                            Hunstad, I., G. Selvaggi, N. D’Agostino, P. England, P.
    identified as the source of the event, and finding that this SW-
                                                                             Clarke, and M. Pierozzi, (2003), Geodetic strain in
    dipping normal fault slipped by ~0.6-0.8 m during the earthquake.        peninsular Italy between 1875 and 2001, Geophys.
    Due to its weak surface features, the Paganica fault had previously      Res. Lett., 30(4), 1181, doi:10.1029/2002GL016447
    been somewhat overlooked in the literature and was thought by           Walters, R. J., J. R. Elliott, N. D’Agostino, P. C.
    most workers to be less active than other nearby faults.                 England, I. Hunstad, J. A. Jackson, B. Parsons, R. J.
                                                                             Phillips, and G. Roberts, (2009), The 2009 L’Aquila
    We also modelled how the L’Aquila earthquake imparted stress             earthquake (central Italy): A source mechanism and
                                                                             implications for seismic hazard, Geophys. Res. Lett.,
    changes on other nearby faults, and found that several had               36, L17312, doi:10.1029/2009GL039337
    been brought closer to failure, representing future seismic




8
                                                                                                                                                      9




THE 2010 HAITI EARTHQUAKE – INSAR CONSTRAINTS OF FAULT SLIP
EXTENT AND FUTURE SEISMIC HAZARD
John Elliott and Barry Parsons, University of Oxford; Zhenhong Li, University of Glasgow

On the 12th January a magnitude 7.0 earthquake
occurred in Haiti west of the capital Port-au-Prince.
The relatively large magnitude and shallow depth of
faulting at such close proximity to a large population
housed in poorly constructed buildings resulted in the
second largest death toll due to an earthquake in over
a century. About 230,000 people lost their lives, with
300,000 injured and over a million homeless. The death
toll was comparable to the Sumatra earthquake and
tsunami of 2004, despite the Haitian earthquake being
500 times less powerful.
The initial seismological solutions indicated a predominately
strike-slip fault that ruptured at a shallow depth of about 10 km.
The earthquake occurred on the east-west Enriquillo fault on
which motion occurs parallel to the fault. GPS observations had
                                                                          Figure 2. ALOS interferogram covering the portion of the fault rupture
shown that the fault was accumulating strain at 7 mm/yr (Manaker          west of Port-au-Prince overlain on digital elevation topography. Coloured
et al., 2008). Using InSAR observations of the ground deformation,        fringes indicate surface deformation in the satellite line-of-sight.
we were able to locate more precisely the section of the fault that       Cycles of colour from blue to red indicate the ground has moved away
slipped. The important observation regarding future seismic hazard        from the satellite. SAR data from the Japanese Space Agency (JAXA).
                                                                          Detailed modelling remains to be completed, but the location of the
is that the portion of the fault due south of the capital did not
                                                                          fringes indicates the fault rupture occurred somewhat to the west of
rupture in this event. Therefore, this portion of the fault represents    Port-au-Prince along the Enriquillo fault, and that the section of the
a continuing seismic hazard, the accumulated elastic strain for           fault immediately south of the capital did not slip in the earthquake and
the most easterly section of the fault not having been released in        therefore represents a heightened seismic hazard.
this earthquake. Furthermore, stress modelling shows that this
unruptured portion has been brought closer to failure as a result of
this year’s earthquake.




                                                                            References:
                                                                            Ali, S. T., A. M. Freed, E. Calais, D. M. Manaker, and
                                                                              W. R. McCann (2008), Coulomb stress evolution in
                                                                              Northeastern Caribbean over the past 250 years
                                                                              due to coseismic, postseismic and interseismic
                                                                              deformation, Geophysics Journal International, 174,
                                                                              904-918, doi:10.1111/j.1365-246X.2008.03634.x
                                                                            Hough, S. E., Bilham R. G., (2006), After the Earth
                                                                             Quakes: Elastic Rebound on an Urban Planet.
                                                                             Oxford: Oxford Univ. Press. 321
                                                                            Manaker, D. M., E. Calais, A. M. Freed, S. T. Ali, P.
                                                                             Przybylski, G. Mattioli, P. Jansma, C. Prepetit,
                                                                             and J. B. de Chabalier, (2008), Interseismic Plate
Figure 1. Map of recent earthquakes and GPS vectors for Hispaniola with      coupling and strain partitioning in the Northeastern
the main shock and largest aftershock indicated to the west of Port-au-      Caribbean, Geophysics Journal International, 174,
Prince. Earthquakes are magnitude 5 and above from the Global CMT            889-903, doi:10.1111/j.1365-246X.2008.03819.x
catalogue for the period 1973–2008.
     UNDERS TA NDING E A R T H
     S YS T EM PROCES SES
     THE RECENT KARONGA EARTHQUAKE SEQUENCE IN THE
     SOUTHERN EAST AFRICAN RIFT
     Juliet Biggs, University of Oxford; Edwin Nissen, Tim Craig and James Jackson, University of Cambridge




                                                                                       Figure 1. Location of the 4 largest earthquakes in the 2009
                                                                                       Karonga Malawi earthquake sequence from teleseismic
                                                                                       records. The earthquakes were located on the western shore
                                                                                       of the Lake, 50 km from the major rift-bounding Livingstone
                                                                                       Fault. The white box is the location of the interferograms on
                                                                                       the right which show deformation patterns from the satellites
                                                                                       ALOS and Envisat. They are displayed such that each fringe
                                                                                       (red to blue) represents 2.8 cm of motion away from the
                                                                                       satellite in the satellite line of sight. The top panel contains
                                                                                       deformation from the complete earthquake sequences
                                                                                       whereas the lower panel only contains deformation from the
                                                                                       last earthquake of the sequence which was the largest. The
                                                                                       patterns are consistent with the rupture of a shallow, west-
                                                                                       dipping fault represented by the black line.

     In December 2009, a shallow earthquake sequence                        The geomorphology of the area is dominated by the 100-km long
     hit the Karonga region of northern Lake Malawi.                        Livingstone Fault; the down-thrown block to the west of the fault
     Between 6–19th December, four earthquakes occurred                     has created a topographic low, occupied by Lake Malawi and
     with magnitudes over 5.5, accompanied by a five                        its sediments. The Karonga earthquakes lie 50 km west of the
                                                                            rift-bounding Livingstone Fault, just on the shore of Lake Malawi.
     further events with magnitudes in the range 5.0–5.2.
                                                                            This demonstrates that the down-thrown block of the Livingstone
     Over 1000 houses collapsed, a further 2900 were
                                                                            Fault is not intact. It is actively breaking up, reflecting either the
     damaged, 300 people were wounded, and 4 were                           temporal and spatial migration of activity or the release of stresses
     killed. Although these earthquakes were of moderate                    within it.
     size only, the southern East African Rift has an
     unusually large thickness (35-40~km) of crust within                   The deformation patterns seen in the satellite images are
                                                                            consistent with the rupture of a shallow, west-dipping fault.
     which earthquakes can occur. This has resulted in
                                                                            Although magmatism and dike intrusion are important components
     wide tilted basins and extremely long faults with
                                                                            of continental rifting even in immature sections of the East
     the potential for magnitude 7–8 normal-faulting                        African Rift System, and the earthquakes did not follow a simple
     earthquakes.                                                           mainshock-aftershock pattern, we see no evidence for the
     In conjunction with ESA’s Changing Earth Science Network, we           involvement of magmatic fluids.
     used the satellites Envisat and ALOS to acquire radar images of the
     ground deformation associated with these earthquakes, which are
     among the first to be studied with satellite interferometry in the
     East African Rift. We use seismology and satellite interferometry        References:
     to obtain earthquake source parameters and combined this with            Biggs, J., E. Nissen, T. Craig, J. Jackson, and D. P.
     information on rift structure from geomorphology and seismic               Robinson, (2010) Breaking up the hanging wall of
     profiles providing an excellent opportunity to study active faulting       a rift-border fault: the 2009 Karonga Earthquakes,
                                                                                Malawi, Geophysical Research Letters, 37
     within a rift setting.




10
                                                                                                                                                      11




DETERMINING WETLAND METHANE EMISSIONS FROM SPACEBORNE
OBSERVATIONS OF METHANE AND GRAVITY
Anthony Bloom and Annemarie Fraser, University of Edinburgh

Atmospheric concentrations of methane (CH4), a                             emission estimates, strongly suggesting that the recent observed
potent greenhouse gas, have increased steadily over                        change in CH4 concentration were due mainly to changes in
the past two centuries. However, in the 1990s and                          wetland emissions.
early 2000s, concentrations started to level off until                     This work is funded by United Kingdom Natural Environmental
2007 when they resumed their global rise. We used                          Research Council studentship NE/F007973/1 and the National
satellite observations to investigate the reasons for                      Centre for Earth Observation. The project was performed in
these most recent changes in CH4 concentration.                            collaboration with Christian Frankenberg, SRON Netherlands
                                                                           Institute for Space Research, Utrecht, Netherlands.
Wetlands are the largest single source of CH4, accounting
for approximately 40% of the total source. In our work, we
quantified the size, distribution, and variability of CH4 emissions
from wetlands. Wetland emissions depend primarily on the soil
temperature and the soil water level. Observed variations in
CH4 measured by the SCIAMACHY satellite instrument provide
information about changes in wetland emissions of CH4. We used
surface temperatures from a weather model and soil water levels
inferred from GRACE satellite observations of small changes in
gravity. By combining all these data, we found that CH4 emissions
from tropical wetlands are best described by changes in water
level, while high latitude wetland emissions are most sensitive
to changes in soil temperature. Our results also showed that
tropical wetlands account for more than half of the total wetland
source. Over the period 2003-2007 we found that global CH4
wetland emissions increased by 7%, largely due to an increase in           Figure 2. Spaceborne observations of Methane (CH4) and Gravity were
temperatures over northern mid-latitude wetlands.                          used to quantify global and zonal contributions of CH4 emissions from
                                                                           wetlands (Tg y-1) between 2003 and 2007 (Bloom et al., 2010): the change
We used our new wetland emissions, along with global estimates             in 2007 is also shown as a percentage of 2003 emissions.
of the other sources of CH4, as input to a computer model of
atmospheric chemistry and transport, and compared to a similar
model calculation using older emission estimates. We found
that our wetland emissions were generally better at reproducing
the size and changes in CH4 concentrations than other wetland




                                                                             References:
                                                                             Bloom, A. A., P. I. Palmer, A. Fraser, D. S. Reay,
Figure 1. Global mean atmospheric methane (CH4, ppb) concentrations            C. Frankenberg, (2010), Largescale controls of
from 1984 to 2009. The growth rate dropped in the late 20th century, and       methanogenesis inferred from methane and gravity
while no significant growth was observed in the first years of the 21st        spaceborne data. Science, 327, 322–325
century, a renewed growth has taken place since 2007.
     UNDERS TA NDING E A R T H
     S YS T EM PROCES SES
     HOW ACCURATE ARE THE RADIATIVE PROPERTIES OF ICE CLOUDS DERIVED
     Nicky Chalmers, Robin J. Hogan, Julien Delanoë and Richard P. Allan, Department of
     Meteorology, University of Reading

     Ice clouds play a key role in the climate system                       and precipitation properties simultaneously. This will be used to
     via their interaction with both shortwave radiation                    produce official products from the EarthCARE satellite, due to be
     from the sun and longwave radiation emitted by                         launched by ESA and JAXA in 2013 and carrying Doppler radar,
     the Earth’s surface. We have developed and applied                     lidar, multi-spectral imager and a broad-band radiometer on the
                                                                            same platform.
     a rigorous method to derive the amount of ice
     present and the size of the particles in ice clouds by
     combining the signals from NASA’s CloudSat radar                                                            Julien Delanoë
     (wavelength 3 mm) and Calipso lidar (wavelength
                                                                                                                 ‘With the support of both
     0.5 µm). But how accurate is this for determining
                                                                                                                 NERC and the European
     cloud radiative properties and how important is it                                                          Space Agency (ESA), we are
     to use both radar and lidar? To test this we have                                                           currently developing an
     taken retrievals over the ocean with only ice clouds                                                        algorithm that would
     present and run the Edwards-Slingo radiation scheme                                                         use the same synergistic
     to calculate the upwelling shortwave and longwave                                                           approach to also retrieve
                                                                                                                 liquid cloud, aerosol and
     fluxes at the top of the atmosphere. These have
                                                                                                                 precipitation properties
     been compared to measurements from the CERES                                                                simultaneously.’
     radiometer in the same train of satellites as CloudSat
     and Calipso. The results are shown in Figure 1. The
     longwave fluxes are in excellent agreement, with
     the predicted flux on average only 0.3Wm-2 less than
     the measurements, and a root-mean-squared (rms)
     difference of 14Wm-2. The predicted shortwave flux
     is 4Wm-2 larger than the measurements, on average,
     with an rms difference of 71Wm-2. The larger
     scatter in the shortwave is most likely due to three-
     dimensional scattering effects.
                                                                              References:
     Figure 2 shows the contrasting situation when we                         Chalmers, N., R. J. Hogan and R. P. Allan, (2009),
     estimate the cloud properties from solely the CloudSat radar.             Investigating the radiative impact of clouds using
                                                                               retrieved properties to classify cloud type. Current
     In this case we use an empirical relationship to estimate ice             Problems in Atmospheric Radiation (IRS 2008),
     water content from the radar signal and the air temperature, and          1100, 525–528
     without the lidar we must estimate particle size purely based on         Delanoë, J., and R. J. Hogan, 2008: A variational
     temperature. In this case, the biases increase to 10Wm-2 in the           scheme for retrieving ice cloud properties
                                                                               from combined radar, lidar and infrared
     longwave (rms difference 47Wm-2) and 48Wm-2 in the shortwave
                                                                               radiometer. J. Geophys. Res., 113, D07204, doi:
     (rms difference 110Wm-2). The larger scatter in both parts of the         10.1029/2007JD009000
     spectrum illustrates that the synergy of radar and lidar instruments     Delanoë, J., and R. J. Hogan, 2009: Combined
     is really needed to pin down the water content and particle size          CloudSat-CALIPSO-MODIS retrievals of the
     in ice clouds, necessary to calculate their radiative properties and      properties of ice clouds. J. Geophys. Res., in press

     hence determine their effect on the climate system.                      Hogan, R. J., M. P. Mittermaier and A. J. Illingworth,
                                                                               2006: The retrieval of ice water content from radar
     With the support of both NERC and the European Space Agency               reflectivity factor and temperature and its use in the
                                                                               evaluation of a mesoscale model. J. Appl. Meteorol.
     (ESA), we are currently developing an algorithm that would use            Climatology, 45, 301–317
     the same synergistic approach to also retrieve liquid cloud, aerosol




12
                                                                                                                                                                13




              How accurate the radiative properties of of clouds derived from
          How accurate areare the radiative properties ice ice clouds derived from
                             CloudSat and properties of ice clouds derived from
              How accurate are the radiativeCalipso satellites?
        How accurate are the radiative properties of ice clouds derived from
                         thethe CloudSat and Calipso satellites?
                             the CloudSat Calipso satellites?
                        the CloudSat and and Calipso satellites?
              Nicky Chalmers, CALIPSO SATELLITES? Richard P. Allan
FROM THE CLOUDSAT AND Robin J. Hogan, Julien Delanoë Richard P. AllanAllan
          Nicky Chalmers, Robin J. Hogan, Julien Delanoë and and Richard P.
         NickyNicky Chalmers, Robin J. Hogan, Julien Delanoë and Richard P. Allan
               Chalmers, Robin J. Hogan, Julien Delanoë and
                                       Department of Meteorology, University of Reading
                                    Department of Meteorology, University of Reading
                                        Department of Meteorology, University of Reading
                                    Department of Meteorology, University of Reading




            Figure 1. Comparison of CERES measurements of upwelling flux with the values predicted by the Edwards-
     Figure 1. Comparison of CERES measurements of upwelling flux with the values predicted by the Edwards-
      Figure 1. Comparison of CERES measurements of upwelling flux with the values the variational CloudSat-Calipso retrieval scheme
                                                                                     predicted       Edwards-Slingo radiation
            Slingo radiation of CERESon ice-cloud profiles theupwelling by the with the predicted codethe on ice-cloud profiles     run
     Slingo radiationComparison run CERES profiles of upwelling flux with the valuesvalues retrieval scheme of of
    Figure 1. Comparison code on ice-cloudmeasurementsfrom variational CloudSat-Calipso predicted by the Edwards-
              Figure 1. code run of measurements from of                                       flux                            by       Edwards-
      from the variational CloudSat-Calipso retrieval scheme of Delanoe and Hogan (2008, 2010). The shortwave part of the spectrum is shown in the left panel
            Delanoe and Hogan on2010). ice-cloud shortwave partthe spectrum is shown in the left retrieval scheme of
                            code The colours 2010). The profiles from the the CloudSat-Calipso occurred panel logarithmic
                                                                                        of variational
     Delanoe and Hogan run (2008, on The profiles from theofvariational spectrum is shown retrieval scheme and the
               radiation the right.code ice-cloud shortwave part                                                           in the left panel of
    Slingo Slingo radiation (2008, run indicate how frequently each combination of measurement CloudSat-Calipso (note the and the scale),
      and the longwave in                                                                                 and prediction
     longwave in indicatesHogan (2008, 2010). The shortwave Calipsoofeach spectrum is of measurement and prediction
            longwave and right. colours
              Delanoe in the The The colours indicate frequently each combination of measurementpanel panel and
                                                                                            the combination in the left and and the
    Delanoe and the right.1.4-km along-trackindicate how how frequentlyspectrum is shownshown in the leftprediction the
      where a “count” Hogan a(2008, 2010). The shortwave part of the data.
                                                       average of the CloudSat and
                                                                                   part
     occurred in the the the The colours indicate where how frequently combination of measurement and ofand CloudSat
            occurred in the logarithmic scale), how frequently each 1.4-km along-track measurement the prediction
              longwave right. right. The colours indicate a “count” indicates a 1.4-km along-track average prediction
    longwave (note(notelogarithmic scale), where a “count” indicates aeach combination ofaverage of the CloudSat
     and Calipso data.data. logarithmic scale), where a “count” indicates a 1.4-km along-track average of the CloudSat
            and(note the logarithmic scale), where a “count” indicates a 1.4-km along-track average of the CloudSat
    occurred Calipso
              occurred (note the
              and Calipso
    and Calipso data. data.
      
             
             
             
             
             
             
             
             
             
             
             
             
             
             
             
             
             

      Figure 2. As As 1 As 1 the radar-only ice water content water content retrieval scheme of Hogan et al. (2006) coupled to
            Figure Fig. Fig. but for the radar-only ice water content et al. (2006) coupled to the Kristjansson et (2006) coupled to
     Figure 2. Fig.2. but for but1for the radar-only iceretrieval scheme of Hogan retrieval scheme of Hogan et al.al. (2000) effective radius
     the Kristjansson same et al. in for Met Officeradius model. parameterization,same of usedusedal. (2006) coupled to
      parameterisation, the 1 Fig. (2000)the the radar-only parameterization, retrieval scheme of in the et Office climate
            the As Fig. etbut 1 the radar-only climate ice water content the the same as et in the Met Office climate
              Figure 2. As al. used
    Figure 2. Kristjansson asforbut(2000) effective water content retrieval scheme as HoganHoganMetal. (2006) coupled to
                                           effective ice radius
            model.
     model. Kristjansson et al. effective radius parameterization, the same as used used Met Office climate
              the
    the Kristjansson et al. (2000)(2000) effective radius parameterization, the same as in the in the Met Office climate
    model.model.
             
             
   UNDERS TA NDING E A R T H
   S YS T EM PROCES SES
            Improved understanding of changes in the global water cycle
   IMPROVED UNDERSTANDING OF CHANGES IN THE GLOBAL WATER CYCLE
       Dr. Richard P. Dr Viju Dr Viju John2 Dr. Brian Soden3, Dr. Dr. William Ingram2,
   Dr. Richard P. Allan1, Allan1, John2, Dr. Brian,Soden3, Dr. Igor Zveryaev4, Igor Zveryaev4, Dr.                                          William
   Dr. Peter Good2, Dr. Beate Liepert5 2, Dr. Peter Good2, Dr. Beate Liepert5
                                Ingram




    Figure 1. The linear sensitivity of frequency of daily rainfall intensity to sea surface temperature (%/K) for observations (SSM/I, HadISST), atmosphere model
   Figure 1. The linear sensitivity of frequency of daily rainfall intensity to sea surface temperature (%/K) for
    experiments (ensemble mean) and a Clausius Clapeyron experiment where precipitation rises at the rate 7%/K (right). Both models and observations show
    a marked increase in the frequency of intense rainfall events (above the experiments with warmer sea surface temperatures. The observed response
   observations (SSM/I, HadISST), atmosphere model98th percentile bin)(ensemble mean) and a Clausius Clapeyron
    is 2-3 times larger than the model ensemble rises at the rate the upper end of Both th     models and observations show left panel. For
   experiment where precipitation mean response and is at 7%/K (left). the range in model simulated responses, shown in the a marked
                      the frequency
   increase insee Allan et al. (2010). of intense rainfall events (above the 98 percentile bin) with warmer sea surface
    further details
   temperatures. The observed response is 2-3 times larger than the model ensemble mean response and is at
    Impacts of future the range the water cycle are
   the upper end of changes in in model simulated responses, shown in the right panel.to variation. A Special Focus
                                                                                   and thermodynamic contributions For further details see
   Allan et al. (2010).                                                            Issue of Environmental Research Letters was published in 2010 in
    of paramount importance to societies and the
                                                                                                with Liepert
    ecosystems upon changes in the water cycle are of Drpartnership fromDrNorth from North West Research Associates.
   Impacts of future which they depend. However, model                                                             West Research Associates.
                                                                                     Liepert together the community on an important problem:
                                                                                   This brought
    projections importance to societies                             that
   paramount are highly uncertain and it is vitaland the This brought together the community on an important
                                                                                   what are the anticipated changes in the water cycle?
    robust responses in which of the water cycle are
   ecosystems upon aspects they depend. However, problem: what are the anticipated changes in the
    identified using the observational record and it is
   model projections are highly uncertain grounded vital water cycle and was published in 2010.
    by robust responses in aspects a the water of
   thatsound physical principles. Using ofcombinationcycle
                                                                                      References:
    satellite data, reanalysis the observational record References of Meteorology/NCAS-Climate, University
   are identified using products and climate model
   grounded by sound physical principles. Using a
                                                                                      1
                                                                                        Department
    simulations, current changes in precipitation and their
   combination of satellite data, reanalysis products and
                                                                                1        of Reading, UK; 2Met Office, Exeter, UK; 3RSMAS,
                                                                                                         of        Meteorology/NCAS-Climate,
                                                                                 Department of Miami, USA; 4P.P.Shirshov Institute,
                                                                                         University
    links to model simulations, current changes
   climate the Earth’s energy budget were revealed. In in University of Reading, UK; Met Office, Exeter, USA
                                                                                                                         2
                                                                                         Moscow, Russia, Northwest Research Associates, UK;
                                                                                                            5

    particular, the tendency links to the to become
   precipitation and their for wet regionsEarth's energy
                                                                                3                                                      4
                                                                                      Allan, University of Miami, USA; P.P.Shirshov
                                                                                 RSMAS, R. P., B. J. Soden, V. O. John, W. Ingram,
    wetter were revealed. In particular, the tendency for Institute,Good (2010), Current Changes in Tropical
   budget and dry regions drier appears robust in models
                                                                                                                            5
                                                                                         P.    Moscow, Russia, Northwest Research
    and observations over tropical regions. Intensification                              Precipitation,
   wet regions to become wetter and dry regions drier Associates, USA submitted to Environmental
   appears robust in models the wet ocean regime                                         Research Letters. 5, 025205, doi:10.1088/1748-
    of rainfall was identified over and observations over
   tropical regions. Intensification of rainfall was Allan, 9326/5/2/025205               R. P., B. J. Soden, V. O. John, W. Ingram, P.
    using daily model and satellite data.                                             Allan, R. P. and I. I. Zveryaev (2010), Variability in
   identified over the wet ocean regime using daily Good (2010), Current Changes in Tropical
                                                                                         the summer season hydrological cycle over the
    Results from satellite data.
   model and this research were quoted in the ‘Copenhagen                                Atlantic-Europe region Environmental Research
                                                                                Precipitation, submitted to1979–2007, Int. J. Climatol.,
    Diagnosis’ used in the Copenhagen COP-15 climate change                     Letters. press 5,
                                                                                         in                   025205,            doi:10.1088/1748-
    conference. Observed changes in water cycle quoted in the
   Results from this research were variables over the                           9326/5/2/025205
                                                                                      Allan, R. P. (2009), Examination of Relationships
   "Copenhagen Diagnosis" useddocumented, including
    north Atlantic-Europe region were also in the Copenhagen                             between Clear-Sky Longwave Radiation and Aspects
                                                                                          R. P. and I. I. Hydrological Cycle Variability
   COP-15 climate change conference. Observed Allan, of the AtmosphericZveryaev (2010), in Climate in
    precipitation, water vapour and evaporation; explanations for                        Models, Reanalyses, and Observations, J. Climate,
                                                                 the north
   changes in water cycle variables over to large-scale the summer season hydrological cycle over the
    regional changes are less well understood compared                                   22, 3127–3145
   Atlantic-Europe region were also documented, Atlantic-Europe region 1979-2007, Int. J. Climatol., in
    responses and further work is necessary to separate the dynamic
   including           precipitation,        water         vapour         and press
   evaporation; explanations for regional changes are
   less well understood compared to large-scale Allan, R. P. (2009), Examination of Relationships
14 responses and further work is necessary to separate                          between Clear-Sky Longwave Radiation and Aspects
   the dynamic and thermodynamic contributions to of the Atmospheric Hydrological Cycle in Climate
                                                                                                                                                         15




RADIATIVE FORCING FROM PERSISTENT AIRCRAFT CONTRAIL
CIRRUS CASE STUDY
Dr. Jim Haywood1, Dr. Richard P. Allan2, Jorge Bornemann1, Piers M. Forster3, Peter N. Francis1, Sean Milton1,
Dr. Gaby Radel2, Alexandru Rap3, Keith P. Shine2, and Robert Thorpe1




Figure 1. A sequence of thermal infra-red satellite imagery showing linear and coil-shaped contrails over the north sea (bright shades denote reduced
thermal emission to space due to high cloud and contrails) which subsequently drifted toward southern England, evolving into a ubiquitous blanket of
cirrus cloud.


The rapid growth and the forecast future expansion of                            forcing of around 7% of recent estimates of the persistent
the aviation industry mean that the potential climatic                           contrail radiative forcing due to the entire global aircraft fleet on
effects have received considerable attention over the                            a diurnally averaged basis. This work was also described in the
past decade. Persistent condensation trails (contrails)                          Sunday Times and the online Guardian.
and aviation-induced cirrus may impact significantly
the Earth’s radiative energy balance. A substantial
contribution was made to the identification, analysis                               References:

and rapid publication of a case study demonstrating
                                                                                    1
                                                                                     Forecasting Research and Development, Met Office,
                                                                                      Exeter, UK.
the formation of a blanket of cirrus cloud over the                                 2
                                                                                        Department of Meteorology/NCAS Climate,
southern UK relating primarily to a single aircraft                                      University of Reading, Reading, UK.
contrail.                                                                           3
                                                                                        Department of Environmental Science, University of
                                                                                         Leeds, Leeds, UK.
Using a combination of global and regional models, conventional                     Haywood, J. M., R. P. Allan, J. Bornemann, P. Forster,
observations and satellite datasets the contrail induced cirrus                      P. N. Francis, S. Milton, G. Rädel, A. Rap, K. P.
was found to produce a detectable heating effect at night due                        Shine and R. Thorpe (2009) A Case Study of the
                                                                                     Radiative Forcing of Persistent Contrails Evolving
to an enhanced greenhouse effect but a cooling by day due to
                                                                                     into Contrail-Induced Cirrus, J. Geophys. Res., 114,
increased reflection of sunlight back to space. The study estimated                  D24201,doi:10.1029/2009JD012650
that this single event may have generated a global-mean radiative
   UNDERS TA NDING E A R T H
   S YS T EM PROCES SES
   PHYSICAL MODELS OF AIR-SEA GAS EXCHANGE AND WHITECAPS
                         Physical models of air-sea gas exchange and whitecaps
    David Woolf and Lonneke Goddijn-Murphy, Environmental Research Institute, Thurso (ERI)
                      Physical models of air-sea gas exchange and whitecaps
                  David Woolf and LonnekeGoddijn-Murphy, Environmental Research Institute, Thurso (ERI)
    Transfer velocities are fundamental to the calculation              coverage. For this and other reasons, whitecap coverage is a
           David Woolf and LonnekeGoddijn-Murphy,                       common interest of geochemists and Thurso (ERI)
    of air-sea fluxes of carbon dioxide and other gases. Environmental Research Institute, space agencies; for example
                                                                           European of the Agency and the Surface Ocean
    Historically, gas transfer velocities have been the calculation representatives SpaceEuropean Space Agency and the Surface
                  Transfer velocities are fundamental to
                                        carbon dioxide and
                  of air-sea fluxes of relationship to windother gases. Ocean Lower AtmosphereStudy recently identified this as as a
    estimated from an empirical                                            Lower Atmosphere Study recently identified this a likely
                  Historically, gas transfer to the calculation European Space Agency collaboration (http://www.cost-
          Transfer velocities are fundamental velocities have been subject of collaboration (http://www.cost-735.org/meetings/
                                                                           likely subject of and the Surface Ocean
          of air-sea fluxes of carbon dioxide and velocity that Lower Atmosphere Study recently identified this as
                   physical from of empirical relationship to
    speed, but estimatedmodelsan gas transfer other gases. wind 735.org/meetings/meetings.html; Toulouse, a 30-31
                                                                        meetings.html; Toulouse, 30-31 March 2010). A “parametric
          Historically, gas physical velocitiesgas transfer velocity
                  speed, but contributions of have been
    recognise the separate transfer models of surface stirring likely March 2010).Acollaboration approach” to predicting
                                                                             subject of       “parametric (http://www.cost-
                  that from bubbles are preferable. Physical surface approach” to predicting whitecap Toulouse, by scientists
          estimated recognise empirical relationship to wind 735.org/meetings/meetings.html; coverage is being developed by
    by the wind and by        an the separate contributions of             whitecap coverage is being developed 30-31
                    but physical models of and by bubbles March at ERI and and National University of Ireland Galway.
          speed,stirring by the wind gas transfer velocity are scientists at ERINational University of Ireland Galway.
                                                                             2010).A “parametric approach” to predicting
          that recognise the wind speed dependence
    models predict thatthe separate contributions ofthat the wind
                  preferable.Physical models predict surface whitecap coverage is being developed by scientists
          stirringvelocity varies widely among gases varies
    of transfer speed dependence of transfer velocityof are widely and National University of Ireland Galway.
                     by the wind and by bubbles                     at ERI
          preferable.Physical models an assumption contrary
                  among gases of to predict that of
    different solubility, contrarydifferent solubility, the wind to an
                   dependence the traditional models. Therefore,
          speed assumption of transfer velocity varies widely
    the traditional models.ofof traditional application to an the
                                Therefore, the contrary of
          among gases of different solubility, models based on
                  application
                  models based on observations of convenient
    traditional observationstraditional models. Therefore,helium) to
          assumption of the of convenient tracers (e.g. the
          application different very different gases,is suspect.
                  very of
    tracers (e.g. helium) totraditional models 2, based on
                                 gases, includingCO including
          observations of convenient tracers (e.g. helium) to
           is suspect.
    CO2,very different gases, includingCO , is suspect.
                                                  2




                                                                                     Figure 2.Predictions of CO2transfer velocity. The
                                                                                 Figure 2. Predictions of CO2 transfer velocity. The points are based on
                                                                                     points are based on a physical model of coverage and
                                                                                 a physical model of gas transfer and observed whitecapgas transfer
                                                                                 winds during the “MAP of CO2transfer velocity.windsempirical
                                                                             Figure 2.Predictions whitecap coverage are based The during
                                                                                     and observed campaign”. The curves and                 on
                                                                                 relationships to wind a physical model of gas transfer
                                                                                     are “MAP campaign”.The curves are transfer
                                                                             points the based on speed. The physical model predicts based on
                                                                                   observed whitecap the curve wind speed. during
                                                                                     empirical relationships to “N00” winds much earlier by
                                                                             andvelocities scattered aroundcoverage andoriginatedThe physical
                                                                                   “MAP and colleagues.
                                                                                     model predicts transfer velocities based on
                                                                             the Nightingalecampaign”.The curves are scattered around
                                                                             empirical curve “N00”originatedmuch earlier by Nightingale
                  Figure 1.Some observations, established empirical the relationships to wind speed. The physical
                                                                             model predicts transfer
                  models and a physical model (“modified COARE”) of and colleagues. velocities scattered around
                                                   established empirical the
          Figure 1.Some observations, CO2 as a function of wind curve “N00”originatedmuch earlier by Nightingale
                  the transfer velocity of
           1. Some and a physical model (“modified COARE”) of
   Figuremodels observations, established empirical models and a physical and colleagues.
                                                                                     References
                  speed. The physical model is as consistent with References:
          (‘modified COARE’) of the of CO as of function of wind
   model the transfer velocitytransfer velocity a CO2 as a function of
                  these observations as any empirical model and is
                                                2
                                                                                     Callaghan, A., de Leeuw, L. Cohen, and and C. D.
          speed.much more consistent with the observed transfer Callaghan, A., G.G. de Leeuw, L. Cohen,C.
                       The model is consistent as consistent with References
   wind speed. The physicalphysicalasmodel is with these observations
   as any these velocity of is much more consistent with the observed is
           empirical model and dimethyl sulphide (not model and                        D. O’Dowd, (2008), Relationship of oceanic
                   observations as any empirical shown).                             O’Dowd,(2008), Relationship ofand wind whitecap
                                                                                                                                  oceanic
                                                                                       whitecap coverage to             speed
   transfer velocity of dimethyl sulphide (not shown). observed transfer Callaghan, A., G. to wind wind Cohen, and C. history.
          much more consistent with the                                              coverage                       L.
                                                                                                     de Leeuw,speed and wind D.                 history.
                                                                                       Geophysical Research Letters, 35
                                                                             O’Dowd,(2008),
                  Researchers at University and Southampton and ERI GeophysicalRelationship of oceanic whitecap
          velocity of dimethyl sulphide (not shown). of ERI have shown                               Research Letters, 35
   Researchers at University of Southampton
                  have shown that a physical model is consistent with
                                                                                    Goddijn-Murphy, speed and wind history.
                                                                             coverage to wind L., D. K. Woolf and A. Callaghan,
                                                                 gases of
           physical model is University of observations of and ERI                     (2010), Parameterizations and algorithms for oceanic
   that aResearchers at consistent with Southampton solubilities; Geophysical Research Letters, 35Woolf and A. Callaghan,
                  observations of gases of various                                     whitecap coverage, D.K.
                                                                              for Goddijn-Murphy, L., in preparation
                                                           can explain why
   various solubilities; for example, a physical modelconsistent with
          have shown that a physical model is explain why dimethyl (2010),Parameterizations and algorithms for oceanic
                  example, a physical model can
          observations of gases of gas) exhibits a relatively relatively Jeffery, C. coverage, in preparation Callaghan,
   dimethyl sulphide (a relatively solublesoluble gas)exhibits a for Goddijn-Murphy,D., I. S. Robinson and D. K. Woolf, (2010)
                  sulphide (a relatively various solubilities;                       whitecap L., D.K. Woolf and A.
                                                                                       Tuning a physically-based model of the air-sea gas
                  linearphysical model can explain why dimethyl (2010),ParameterizationsOcean Modelling,for oceanic
          example, a between transfer velocitytransfer velocity The wind
   linear relationship     relationship between and wind speed. and                    transfer velocity,
                                                                                                              and algorithms
                                                                                                                                   31(1–2), 28–35
          sulphide (a relatively soluble gas)exhibits a velocity to wind Jeffery, C.D.,in preparation and D.K. Woolf,(2010)
                  speed. The relationship of transfer relatively whitecap coverage, I.S. Robinson
   relationship of transfer velocity to wind speed is incomplete, since
                                                                                     Tuning a physically-based C. S. of the air-sea gas
          linear relationship between transfer physical and windof gas Nightingale, P. D., G. Marlin,modelLaw, A. J. Watson,
                  speed is incomplete, since velocity models
          speed.transfer velocity velocity imply a dependence onwhitecap transfer velocity, Oceanand Boutin and R. 28-35
                       of gas transfer imply a dependence on
   physical modelsThe relationship of transfer velocity to wind Jeffery, P. S. Liss, M. I. Liddicoat, J.D.K. Woolf,(2010)
                                                                                        C.D., I.S. Robinson Modelling, 31(1-2), C.
                                                                                       Upstill-Goddard, (2000), In situ evaluation of
          speed coverage,which since physical of bubbles, gas Tuning a physically-based model of the air-sea gas
   whitecap coverage, which quantifies the numbermodels of and
                    is incomplete, quantifies the number of bubbles,
                                                                                       air-sea gas exchange parameterizations using
                                                                                        velocity, Ocean Modelling, 31(1-2), S. Law,
                  and it whitecap a that not a simple whitecap transfernovel conservative G. Marlin, C. 28-35
          transfer thatis known coverage iswhitecap coverage
   it is known nowvelocity implynowdependence on function is not a Nightingale, P. D.,and volatile tracers. GlobalA. J.
                                               the speed. Observed
                                                    number of bubbles,
          coverage,which quantifiescoverage combined with a whitecap Watson, P. S. Liss, M. I. Liddicoat,J. Boutin and R. C.
                  simple function of
   of wind speed. Observed whitecap wind                                               Biogeochemical Cycles, 14(1), 373–387
                                                                                     Upstill-Goddard, Marlin, and their role
          and it coverage now that whitecap coverage is not a Nightingale, P. D., G.(2000),In C. S. Law, A.ofJ.
                  is known combined with a physical model
   physical model of the transfer velocity of carbon dioxide implies a of the Woolf, D. K., (1997), Bubbles situ evaluationin air-              air-sea
                     function of wind carbon dioxide implies a “spread” gas S. exchange. parameterizations and C. novel
          simpletransfer velocity of speed. Observed whitecap Watson, P. gasLiss, M. I. Liddicoat,J. Boutin and R.Global
                                                                                       sea   exchange In, The Sea Surface using
          coverage transfer values a any moderateany of the  or (especially)
   “spread” of transfer velocityvelocityat values at model moderate Upstill-Goddard, (2000),In situ evaluation of air-seaGlobal
                  of combined with                physical                     or conservative P. S. Liss and R. A. Duce, Cambridge
                                                                                       Change, Eds., and          volatile tracers.
          transfer velocity context of Earth observation, this places
                  (especially) carbon dioxide implies context of gas Biogeochemical Cycles, 14(1), 373–387 novel
   high wind speed. In the ofhigh wind speed. In the a “spread” Earth exchange Press, 173–205
                                                                                       University parameterizations using
   a large value observation, this places aany for whitecap on more
          of transfer velocity values methods moderate or conservative and volatile tracers.
                   on more accurate retrieval at        large value                                                                     Global
                  accurate retrieval methods for context of Earth Biogeochemical K.,(1997),Bubbles and
          (especially) high wind speed. In the whitecap coverage.For Woolf, D. Cycles, 14(1), 373–387 their role in air-sea
                            this places a large value coverageis a gas exchange. In, The Sea Surface and Global
          observation,and other reasons,whitecap on more
                  this
                  common methods of geochemists and                          Woolf, D. K.,(1997),Bubbles and their role in air-sea
          accurate retrieval interest for whitecap coverage.For space Change, Eds., P.S. Liss and R.A. Duce, Cambridge
                                                                                     University In, The Sea
          this and otherforreasons,whitecap coverageis a gas exchange. Press, 173-205Surface and Global
                  agencies;         examplerepresentatives of the
16
          common interest of geochemists and space Change, Eds., P.S. Liss and R.A. Duce, Cambridge
                                                                                                                                                17




INCORPORATION OF MELT PONDS INTO AN ARTIC SEA ICE MODEL
Daniel Feltham and Daniela Flocco, University College London

Arctic sea ice has an important role in the global
climate. Its high albedo means that a reduction in sea
ice coverage leads to a reduction in reflected solar
radiation and, therefore, to increased global warming.
Global Climate Models (GCMs) fail to predict the rapid
decrease in Arctic summer sea ice extent observed
in recent years. Our research suggests this is at least
partly due to the lack of an accurate representation of
melt ponds.
Melt ponds form on Arctic sea ice during the summer and early
autumn due to the accumulation of melt water formed from the
melting of snow and the upper layers of sea ice. Field observations   Figure 1. Between 1997–1998, SHEBA US field experiment measured the
of melt ponds, especially those taken during the Surface Heat         atmospheric and oceanic forcing of the ice cover and recorded the melt
Budget of the Arctic (SHEBA) year-long camp from autumn               processes taking place.
1997 to autumn 1998 [e.g. Perovich et al., 1999, Figure 1] have
provided information on how ponds form and evolve throughout
the melt season, until they freeze over at the end of autumn.
Satellite observations show that a decrease in successive years of
winter ice thickness is strongly correlated with the length of the
intervening melt season, [Laxon et al., 2003].
Melt ponds cover up to 50% of the ice during summer. The ponds
reduce the albedo of the ice, leading to increased warming. The
melting rate of sea ice beneath a melt pond is up to 2-3 times
greater than bare sea ice. The theoretical understanding of melt
ponds, based on mathematical models describing the physics of
melting and meltwater transport, has been advanced over the
last 6-7 years by researchers in UCL. Flocco and Feltham [2007]
                                                                      Figure 2. Basin-averaged climatology of Arctic sea ice thickness (1980–
developed the first, physically-based model of melt pond area         2001) taken from the sea ice component of a GCM with (blue) and without
evolution suitable for climate models and have now included this      (red) melt ponds.
model into the CICE sea ice model [Flocco et al., 2010]. CICE is
the Los Alamos National Laboratory sea ice model adopted by the         References:
UK Meteorological Office. The Flocco et al. model produces maps         Flocco, D. and D. L. Feltham, (2007), A continuum
of melt pond area fraction (and pond depth) and shows that the            model of melt pond evolution on Arctic sea
presence of melt ponds, through lowering the average albedo of            ice, J. Geophysical Research, 112, C08016,
                                                                          doi:10.1029/2006JC003836
sea ice, significantly affects the heat and mass budget of the ice,
                                                                        Flocco, D., D. L. Feltham, A. K. Turner, (2010),
with a reduction in the basin-averaged ice thickness of over a            Incorporation of a physically-based melt pond
metre (Figure 2).                                                         scheme into the sea ice component of a climate
                                                                          model, J. Geophysical Research, in press
The development of a GCM-optimised melt pond routine, and the           Laxon S., N. Peacock, and D. Smith, (2003) High
demonstrated impact of ponds on Arctic sea ice predictions, makes        interannual variability of sea ice thickness in the
it practical and desirable to incorporate the new melt pond physics      Arctic region, Nature, 425, 947–950
into Intergovernmental Panel Climate Change climate models.             Perovich, D. K., T. C. Grenfell, B. Light, J. A. Richter-
                                                                          Menge, M. Sturm, W. B. Tucker III, H. Eicken, G. A.
                                                                          Maykut, and B. Elder, (1999), SHEBA: Snow and ice
                                                                          studies CD-ROM, version 1.0
     DATA A S SIMIL AT ION

     DATA ASSIMILATION FOR MODELLING THE CARBON CYCLE
     Sylvain B. Delahaies and Ian Roulstone, University of Surrey

     Throughout the NCEO, data assimilation (DA) is used
     to combine new Earth observation measurements with
     different models of the Earth system components.
     At the heart of our work is the development of
     methods to understand and represent uncertainty in
     the observations and the models. In many cases the
     observations are only indirectly related to the physical
     quantities of interest. Furthermore, the observations
     may be irregularly distributed in space and time and
     will contain measurement errors. In order to extract
     the maximum understanding from the observations it
     is necessary to combine them with our understanding
     of the physical systems, as embodied in prediction
     models, taking into account uncertainties in both
     the data and the models.
     Many models of the Earth system are very highly parameterised.
     These parameters appear in the models themselves and in the
     formulation of the mathematical operations that provide the link
     between model and observation. The data linked ecosystem
                                                                             Figure 1: Result of a 4DVar assimilation: (top) NEE, (bottom) LAI.
     carbon (DALEC) model is a simple model to predict the net
     ecosystem exchange (NEE) of CO2 and the leaf area index
     (LAI) for evergreen and deciduous forests. DALEC contains 17
     controlling parameters which need to be carefully adjusted for
     each forest under study. Eddy covariance (EC) towers measure
     the NEE of CO2 at different sites; these measurements provide
     an opportunity to estimate the parameters. The Regional Fluxes
     Estimation Experiment (REFLEX) is a DA inter-comparison project:
     nine participants combined DALEC and EC towers observations
     together with local measurements of LAI using DA techniques
     (Monte-Carlo and EnKF), they compared the strengths and
     weaknesses of the different methods to estimate parameters and
     predict carbon fluxes with confidence intervals.
     Following REFLEX, the work carried out at the University of
     Surrey focuses on variational assimilation algorithms to combine
     DALEC with EO (Figure 1) and, using adjoint techniques, to              Figure 2: Simulation of the dispersion of Co2 emanating from 3 sites.
     estimate the sensitivity of the analysis with respect to parameters
     and observations. Quantifying such sensitivities is a crucial part of
                                                                               References:
     our science. Our next goal is to couple a chemical transport model
                                                                               Fox, A., M.Williams, A. D. Richardson, D.Cameron,
     (Figure 2) with DALEC at regional to continental scale and to apply
                                                                                 J. H.Gove, T. Quaife, D. Ricciuto, M.Reichstein,
     inverse modelling techniques to retrieve the sources and strengths          E.Tomelleri, C.Trudinger, M.T. Van Wijk, (2009), The
     of CO2 emissions, and to improve the parameterisation of DALEC.             REFLEX project: comparing different algorithms and
                                                                                 implementations for the inversion of a terrestrial
                                                                                 ecosystem model against eddy covariance data,
                                                                                 Agricultural and Forest Meteorology 149, 1597–1615




18
                                                                                                                                                                 19




DEVELOPMENT OF A CONVECTIVE-SCALE ENSEMBLE PREDICTION SYSTEM
Stefano Migliorini and Ross Bannister, University of Reading




Figure 1. A simple deterministic forecast of precipitation rates over southern UK in summer 2007 (left panel) determined from the best knowledge available
an hour before; predictions of probability of precipitation rates valid at the same time (right panels) for three different precipitation intensity thresholds
(0.125mm/h, 1.0mm/h, 5.0mm/h) derived from an ensemble of forecasts. Radar measurements of precipitation valid at the same time (not shown) show
occurrence of rainfall where the deterministic forecast does not predict any precipitation, but where the ensemble forecast shows instead a non-zero
probability of precipitation. There are, however, still problems with false alarm predictions and with the correct predictions of high-intensity rainfall.


A high priority in weather forecasting is to improve                                at one hour lead time. While this progress is encouraging, we think
predictions of severe weather, such as severe                                       that further improvements will require the use of a larger number
thunderstorms and flash floods. Advances are                                        of forecasts in the ensemble to make better use of radar and
needed both in the way numerical weather prediction                                 satellite measurements.
models represent moist convection and in the way
initial conditions for forecasts are derived from
measurements by ground-based radar and Earth-                                                                                   Stefano Migliorini and
                                                                                                                                Ross Bannister
orbiting satellites. In collaboration with the Met Office
Joint Centre for Mesoscale Meteorology, we have                                                                                 ‘Over the last few years, the
made progress in developing a more advanced, high-                                                                              meteorological community
                                                                                                                                has been trying to improve
resolution weather forecasting system.
                                                                                                                                the skill of quantitative
The numerical model is based on a 1.5 km version of the                                                                         predictions of precipitation,
Met Office’s Unified Model, a scale fine enough to permit the                                                                   with a particular focus on
development of atmospheric convection. Multiple high-resolution                                                                 hazardous weather such as
weather forecasts are made from similar starting conditions (the                                                                severe thunderstorms and
so-called ensemble approach). This approach allows predictions of                                                               flash floods.’
weather events, such as precipitation, that assign a probability of
occurrence and intensity. Given the chaotic and turbulent nature
of the atmosphere, it is most meaningful to provide forecasts of
probabilities of events. The current focus of research is on very
short-range forecasts up to two hours ahead, but the intention is
                                                                                       References:
to develop the system to fit seamlessly into longer range forecasts,
                                                                                       Migliorini, S. M. Dixon, R. Bannister, S. Ballard,
which will require further research on the combined use of                              (2010), Ensemble prediction for nowcasting with a
satellite and radar data.                                                               convection-permitting model. Part I: Description of
                                                                                        the system and the impact of radar-derived surface
Our results using a 24-member ensemble of forecasts show that                           precipitation rates, Tellus A, in review
the new system produces fairly reliable forecasts of precipitation
     DATA A S SIMIL AT ION

     EFFICIENT NONLINEAR DATA ASSIMILATION
     Peter Jan van Leeuwen, Department of Meteorology, University of Reading

     Data assimilation combines observations of the Earth                                  15
     system with numerical models to obtain a better
     description of the processes at work, and allow for
                                                                                           10
     better predictions. With ever increasing resolution in
     the numerical models and more advanced observations
     the data assimilation problem is becoming more and                                     5

     more nonlinear. The present-day data assimilation
     methods like the Ensemble Kalman Filter or 4DVar are                                   0
     based on linearisations.
     A fully nonlinear data assimilation method has been developed                         −5
     based on a so-called Particle Filter. In this method ensembles of
     models are propagated forward in time, and combined with the
                                                                                          −10
     observations in a fully nonlinear way. Traditional Particle Filters                        0   100   200   300   400   500   600   700   800   900   1000
     are highly inefficient and millions or more model runs are needed
     for real-sized geophysical problems. By exploring the so-called
                                                                                       Figure 2. The new Particle Filter tested on the Lorenz 1996 model: a highly
     proposal density idea we have been able to generate an extremely
                                                                                       nonlinear 1000 dimensional model that captures some basic features of
     efficient Particle Filter (see Figure 1).                                         the mid-latitude atmosphere. The black line is the truth we want to recover
     The new method has been tested in up to 1000 dimensional highly                   with the Particle Filter, the red crosses and bars are noisy observations
                                                                                       from this truth with their error bars, and the green lines are the 20
     nonlinear models and shows perfect scaling: the ensemble size
                                                                                       ensemble member runs. The truth is reconstructed very well with these 20
     does not grow with the dimension of the system (see Figure                        model runs, where traditionally millions of model runs were needed.
     2). This suggests that the method is applicable to real-size
     geophysical systems too. We are testing the new method now on
     systems with about 100,000 dimensions, and, if successful, will                                                              Peter Jan van Leeuwen
     apply it to real systems like numerical weather prediction models                                                            ‘With ever increasing
     and climate models.                                                                                                          resolution in numerical
                                                                                                                                  models and more advanced
                                                                                                                                  observations the data
                                                                                                                                  assimilation problem is
                                                                                                                                  becoming more and more
                                                                                                                                  nonlinear.’




     Figure 1. Comparison of traditional and new Particle Filter. The light blue
     line denotes the probability density function (pdf) at previous analysis. The        References:
     blue bars denote the ensemble members. These are propagated forward                  Van Leeuwen, P. J., (2009), Particle filtering in
     in time as the brown arrows, and end up as the blue bars at the time level             geophysical systems, Monthly Weather Rev. 137,
     where new observations are present. The green line denotes the pdf of                  4089–4114
     the observations. In the traditional Particle Filter (left) the majority of the      Van Leeuwen, P. J., (2010), Nonlinear Data
     ensemble members are very different from the observations, while using                 Assimilation in geosciences: an extremely efficient
     the new method (right) the ensemble members are all very close to the                  particle filter, QJR Met Soc, submitted
     observations, making the method much more efficient.




20
                                                                                                                                                     21


 Assimilating Atlantic MOC changes, upper ocean temperatures and
 Assimilating Atlantic MOC changes, upper ocean temperatures and
                             sea level
                             sea level
                              Keith Haines, Vladimir Stepanov and Maria Valdivieso, University of Reading
                              Keith Haines, Vladimir Stepanov and Maria Valdivieso, University of Reading

 ASSIMILATING ATLANTIC MOC CHANGES, UPPER OCEAN
Ocean circulation plays a key role in climate. It observations. Figure 1b showstheimprovement in the
Ocean circulation plays a key role in climate. It observations. Figure 1b showstheimprovement in the
transports heat and so influences sea surface MOC obtained by assimilating Argo data to obtain a
 TEMPERATURES AND SEA LEVEL
transports heat and so influences sea surface MOC obtained by assimilating Argo data to obtain a
temperatures, and therefore improving circulation in good 2004 initial state, and then assimilating the
temperatures, and therefore improving circulation in good 2004 initial state, and then assimilating the
ocean and coupled models is a key step to making Rapid array data to further enhance the MOC.
  Keith Haines, Vladimir Stepanov and step to making
ocean and coupled models is a key Maria Valdivieso, University of Reading further enhance the MOC.
                                                                  Rapid array data to
longer range climate predictions. We have developed
longer range climate predictions. We have developed
a method that successfully controls the Atlantic Figures 2a,b show the change in upper ocean heat
aOcean circulation successfully controls the Atlantic Figures 2a,b show MOC obtained byin upper ocean heat
    method that plays key role in                                  the improvement in the the change assimilating Argo (SSH)
Meridional Overturninga Circulationclimate. Itin ocean content (top 300m T) and sea surface height data
                                               (MOC)
Meridional heat and so influences sea (MOC) in ocean content a good300m T) and sea then assimilating the (SSH)
                Overturning Circulation surface                                                             surface
                                                                   to obtain (top 2004 initial state, Rapid arrayheight Rapid
models by assimilating ocean densities from the that result after 3-4 years ofand
  transports                                                                                                         assimilation.
models by assimilating ocean densities from the that result after 3-4 years of Rapid array assimilation.
  temperatures, and Observations circulation 26N, Sea data to further enhance the MOC.
Rapid Monitoringtherefore improvingArray at in                     array surface temperatures (SST) cannot change
Rapid Monitoring Observations Array at 26N, Sea surface temperatures (SST) cannot change
                                                                   Figures in show ocean only model due to (top
particularly the western measurements around much 2 a,bthis the change in upper ocean heat contentimposed
  ocean and the models measurements around much in this ocean only model due to imposed
particularly coupledwesternis a key step to making
            and the Bahamas where the strong Gulf atmospheric surface height (SSH) that resultcoupledyears of
Florida range climate predictions. We have developed
  longer                                                           300m T) and sea conditions but in a after 3-4 climate
Florida and the Bahamas where the strong Gulf atmospheric conditions but in a coupled climate
Stream and deep western boundary Atlantic flow, model SST wouldSea surface temperatures changes with
  a method that successfully controls the          currents        Rapid array assimilation. also show big (SST) cannot
Stream and deep western boundary currents flow, model SST would also show big changes with
Figure 1a.                                                        feedback on the atmosphere. The next step is to
  Meridional
Figure 1a. Overturning Circulation (MOC) in ocean                  change much in the atmosphere. to imposed step is
                                                                  feedback on this ocean only model dueThe nextatmospheric to
                                                                   conditions but in a coupled climate model SST higher show
                                                                  demonstrate these improvements atwould alsoresolution
  models by assimilating ocean densities from the Rapid           demonstrate these improvements at higher resolution
                                                                   big in coupled models the atmosphere. The next step is
                                                                  andchanges with feedback on where climate predictions can
  Monitoring Observations Array at 26N, particularly              and in coupled models where climate predictionstocan
                                                                   demonstrate these improvements at of ocean heat content,
                                                                  be launched. The patterns higher resolution and in coupled
  the western measurements around Florida and the                 be launched. The patterns of ocean heat content,
                                                                   models where climate well monitored by EO data and we
                                                                  SST and SSH arepredictions can be launched. The patterns of
  Bahamas where the strong Gulf Stream and deep                   SST and SSH are well monitored by EO data and we
                                                                   ocean heat content, to define and look for by EO data and
                                                                  will now be ableSST and SSH are well monitoredcharacteristic
  western boundary currents flow, Figure 1a.                      will now be able to define and look for characteristic
                                                                   we will now able to define with for characteristic changes.
                                                                  signaturesbeassociated and lookAtlantic MOCsignatures
                                                                  signatures associated with Atlantic MOC changes.
                                                                                    supported
                                                                  This work is Atlantic MOC by the NCEO and the Rapid-
  Assimilation increases the MOC strength at a range of latitudes  associated is                  by the NCEO is supported by
                                                                  This workwith supported changes. This work and the Rapid-
                                                                  Watch Valor project.
  as well as increasing the northward heat transports. The free   Watch Valor project.
                                                                   the NCEO and the Rapid-Watch Valor project.
 running model has large errors at the deep western boundary and
 the Rapid assimilation uses regression coefficients derived along                           (a)
 the boundaries to spread the array observations. Figure 1b shows                            (a)


  (a)

                                       Atlantic MOC:26.5N
                         30



                         25
                                                            (b)
                                                            (b)
                         20
        Transport (Sv)




                                                                                             (b)
                         15
Figure 1 (a) shows the locations of the Rapid                                                (b)
Figure 1 (a) shows the locations of the Rapid
Mooring Array across the N Atlantic at 26N. (b)
Mooring Array across the N Atlantic at 26N. (b)
        10
shows the MOC as calculated from the Rapid array
shows the MOC as calculated from the Rapid array
data (blue, mean MOC 18.7Sv), and the NEMO 1
data (blue, mean MOC 18.7Sv), and the NEMO 1
         5
model MOC for a run assimilating only Rapid
Mooring data for a run (red 18.5Sv), and a run
model MOC below 900m assimilating only Rapid
                    below 900m 2006.5 2007 2007.5 a
Mooring data 2005 2005.5 2006 (red 18.5Sv), and2008 run
         0
            any
without 2004.5 assimilation in this period (black 17.2Sv).
  (b)
without any assimilation in this period (black 17.2Sv).
Both runs start from a conventional reanalysis which
has runs Shows the locations of other Mooring Array to January
  Figure 1(a). start Argo conventional prior across the
Bothassimilatedfrom aandthe Rapid datareanalysis which
has assimilated Argo and other data prior to January
  N Atlantic 26N.
2004. Aat second control (purple 13.3Sv) has run
2004. A second control (purple 13.3Sv) has run
freely from 1988 without any assimilation.array data (blue,
  Figure 1(b). Shows the MOC as calculated from the Rapid
        from 1988 without any assimilation.
freely MOC 18.7Sv), and the NEMO 1° model MOC for a run assimilating
  mean
Assimilation increases 900m (red 18.5Sv), and a runat a range
  only Rapid Mooring data below the MOC strength without                       Figure 2 Shows (a) the upper ocean heat content
Assimilation in this period (black 17.2Sv). Both runs start at aa range
  any assimilation increases the MOC strength from
                                                                                            2 Shows (a) the upper ocean heat runs in
                                                                               Figure Average difference for 2007 between the red and blackcontent
of latitudes as well as increasing the northward heat                          (top 300m T inC) and (b) the sea level (in cm)
                                                                                Figure 2.
    latitudes as well as increasing the northward heat
ofconventional reanalysis which has assimilated Argo and other data prior to   (top 300mNEMOinC) and 3-4 years of assimilation of the Rapid
                                                                                Fig 1b), in responsemodel after (b) the difference between
                                                                                                T 1°                       sea level (in cm)
transports. The free running model has large errors                            change the              (2007 average
transports. The free running13.3Sv) has has large errors
  January 2004. A second control (purple model run freely from 1988             array data below 900m. (2007 average difference between
                                                                               change response (a) top 300m T (in°C), (b) associated change in sea
at the deep western boundary and the Rapid                                     red and black runs in Fig 1b), in the NEMO 1 model
atwithout any assimilation.
     the deep western boundary and the Rapid                                   red and black runs in Fig 1b), in the NEMO 1 model
                                                                                level (in cm).
assimilation uses regression coefficients derived                              after 3-4 years of assimilation of Rapid array data
assimilation uses regression coefficients derived                              after 3-4 years of assimilation of Rapid array data
along theboundaries tospread the array                                         below 900m only.
along theboundaries tospread the array                                         below 900m only.
     DATA A ND MODEL
     IN T ERC OMPA RISONS
     SPARC CCMVAL REPORT
     Martyn Chipperfield, University of Leeds

     The CCMVal project is a major international                                    WMO/UNEP assessments of the future of the ozone layer. While
                                                                                      Chapter 6: Stratospheric Chemsitry        215
     collaboration aimed at a process-based evaluation                              these assessments show ozone predictions, there is no supporting
     of chemistry-climate models (CCMs). It is organised N O is analysis to investigate the cause ofand in
           of a model to reproduce this slope of 2 indicates a failing               the main source of stratospheric NOy
                                                                                                                             differences in predictions
                                                                                                                                 2
              the framework of the World Climate Research CCMValbetween models, and therefore there is uncertainty over the
           ofthechemistry. The CNRM-ACM model appears to have           the          runs the only source considered. Overall, stra-
     within slope slightly larger than 2 and a stratosphere that is too tospheric N O has 3 destruction channels:
           a                                                                        robustness of the model predictions. The CCMVal project provides
                                                                                                                                         2
           moist. Other models reproduce the stratospheric slope and
     Programme (WCRP) Stratospheric Processes of 2 their                            this detailed analysis O the CCMs.
           but have lower stratospheric H O overall due presumably                        N O + hν → N + of             (6.2a)
     Role in Climate (SPARC) project.                        2
                                                                                                                                                      2                                  2

            to different input at the tropical tropopause. This is not a                                                                         N2O + O(1D) → (Eyring       (6.2b)
                                                                                                                                             The CCMVal Report2NO et al., 2010) has separate chapters
            failing behind CCMVal is a thorough being evaluated
     The rationaleof the chemistry scheme, which isevaluation of the 15                                                                                for 1D) → N radiation,(6.3c)
                                                                                                                                             covering, + O(example, + O       dynamics, transport and
            here, but these low H O mixing ratios will have an impact                                                                            N 2O
     or so different worldwide 2CCMs which are used in the regular                                                                                                2     2
                                                                                                                                             chemistry. NCEO scientists co-led the chemistry chapter and were
            on calculated model HOx, for example.
                                                                                                                              Section 6.3.2 examined the NOy:N2O correlation for
                          2.0                                          2.0                                             2.0                                                         2.0


                          1.5                                          1.5                                             1.5                                                         1.5
             CH4 (ppmv)




                                                          CH4 (ppmv)




                                                                                                          CH4 (ppmv)




                                                                                                                                                                      CH4 (ppmv)
                          1.0                                          1.0                                             1.0                                                         1.0


                          0.5                                          0.5                                             0.5                                                         0.5

                                          MIPAS                                           AMTRAC3                                              CAM3.5                                                        CCSRNIES
                          0.0                                          0.0                                             0.0                                                         0.0
                             0   2       4        6   8                      0   2       4        6   8                   0          2       4         6       8                         0         2        4        6     8
                                     H2O (ppmv)                                      H2O (ppmv)                                          H2O (ppmv)                                                     H2O (ppmv)

                          2.0                                          2.0                                             2.0                                                         2.0


                          1.5                                          1.5                                             1.5                                                         1.5
             CH4 (ppmv)




                                                          CH4 (ppmv)




                                                                                                          CH4 (ppmv)




                                                                                                                                                                      CH4 (ppmv)




                          1.0                                          1.0                                             1.0                                                         1.0


                          0.5                                          0.5                                             0.5                                                         0.5

                                           CMAM                                            CNRM-ACM                                            E39CA                                                           EMAC
                          0.0                                          0.0                                             0.0                                                         0.0
                             0   2       4        6   8                   0      2       4        6   8                   0          2       4         6       8                         0         2        4        6     8
                                     H2O (ppmv)                                      H2O (ppmv)                                          H2O (ppmv)                                                     H2O (ppmv)

                          2.0                                          2.0                                             2.0                                                         2.0


                          1.5                                          1.5                                             1.5                                                         1.5
                                                          CH4 (ppmv)




                                                                                                          CH4 (ppmv)




                                                                                                                                                                      CH4 (ppmv)
             CH4 (ppmv)




                          1.0                                          1.0                                             1.0                                                         1.0


                          0.5                                          0.5                                             0.5                                                         0.5

                                          GEOSCCM                                        LMDZrepro                                             MRI                                                           NiwaSOCOL
                          0.0                                          0.0                                             0.0                                                         0.0
                             0   2       4        6   8                   0      2       4        6   8                   0          2       4         6       8                      0            2        4        6     8
                                     H2O (ppmv)                                      H2O (ppmv)                                          H2O (ppmv)                                                     H2O (ppmv)

                          2.0                                          2.0                                             2.0                                                         2.0


                          1.5                                          1.5                                             1.5                                                         1.5
             CH4 (ppmv)




                                                          CH4 (ppmv)




                                                                                                          CH4 (ppmv)




                                                                                                                                                                      CH4 (ppmv)




                          1.0                                          1.0                                             1.0                                                         1.0


                          0.5                                          0.5                                             0.5                                                         0.5

                                           SOCOL                                            ULAQ                                               UMSLIMCAT                                                     UMUKCA-METO
                          0.0                                          0.0                                             0.0                                                         0.0
                             0   2       4        6   8                   0      2       4        6   8                   0          2       4         6       8                         0         2        4        6     8
                                     H2O (ppmv)                                      H2O (ppmv)                                          H2O (ppmv)                                                     H2O (ppmv)


                          2.0                                          2.0
                                                                                                                              1000-850 hPa            850-500 hPa                            500-250 hPa
                                                                                                                               250-150 hPa            150-100 hPa                             100-50. hPa
                          1.5                                          1.5                                                      50-30.0 hPa            30-10.0 hPa                             10.-5.0 hPa
             CH4 (ppmv)




                                                          CH4 (ppmv)




                                                                                                                                 5.0-1.0 hPa            1.0-0.5 hPa                            0.5-0.1 hPa
                          1.0                                          1.0


                          0.5                                          0.5

                                        UMUKCA-UCAM                                        WACCM
                          0.0                                          0.0
                             0   2       4        6   8                      0   2       4        6   8
                                     H2O (ppmv)                                      H2O (ppmv)

             Figure 6.13: Correlation of CH4 (ppmv) vs. H2O (ppmv) for zonal-mean monthly-mean output from the final 10
             years of REF-B1 runs from 17 CCMs and MIPAS data. The solid line is the best fit to the model/satellite data
     Figure 1. Correlation of CH4 (ppmv) vs. H2O (ppmv) hPa. The dashed line shows the equation H O + 2CH = 7models and Oxford ENVISAT MIPAS data.
             sampled between 60°N-60°S, 70-0.5 for zonal-mean monthly-mean output from 10 years of 17 CCMVal ppmv.
                                                                                                 2       4
     The solid line is the best fit to the model/satellite data sampled between 60°N-60°S, 70-0.5 hPa. The dashed line shows the equation H2O + 2CH4 = 7 ppmv. The
     observations show high CH4 /high H2O values in the troposphere. H2O is a minimum around the tropopause and then increases in the stratosphere as CH4 is oxidized.
     One CH4 molecule yields up to 2 H2O molecules. The CCMs generally reproduce this behaviour. The two UKCA runs are slightly different – the Met Office run has a
     drier stratosphere and a lower stratospheric production rate of H2O than the more realistic UCAM run. Figure taken from Chapter 6 of Eyring et al. (2010).



22
                                                                                                                                                       23




co-authors or contributors on many other chapters. The chemistry              provided (from the Met Office and Univ. Cambridge), along with
chapter (Lead Authors M. Chipperfield (U. Leeds) and D. Kinnison              results from an older version of the UM (from Univ. Leeds) for
(NCAR)) made extensive use of satellite EO data, including results            reference with previous assessments. Results were submitted to
from MIPAS, SCIAMACHY, AURA MLS, ODIN, ACE and HALOE.                         a data archive at the BADC, after which many tens of scientists
The chapter compared radicals, reservoir species and source                   performed analysis for different diagnostics. Participation in
gases globally and also with a particular focus on the winter/                CCMVal has led to a much better understanding of the strengths
spring polar region. A significant advance of the CCMVal project              and weaknesses of the UKCA model.
was the attempt at quantitative grading of the models for different
processes, an approach which will probably set a framework for                   References:
similar model-model studies in the future. Overall results from 18               Eyring, V., T. Shepherd, D. Waugh, et al., (2010) SPARC
worldwide CCMs were contributed, although some of these runs                      CCMVal Report, in press
                                                                                             Chapter 6: Stratospheric Chemsitry              235
were based on very similar models. Two UKCA simulations were

                                      Equivalent Latitude 89S-79S and Θ Range 350K-1000K
            1000
                    AMLS                       MMM                           CAM3.5                        CCSRNIES
             900
             800
                                                                                                                                            OCT
     Θ/ K




             700
             600
             500
             400
             300
            1000                                                                                                                            SEP
             900    CMAM                       CNRM-ACM                      EMAC                          GEOSCCM
             800
     Θ/ K




             700
             600
             500                                                                                                                            AUG
             400
             300
            1000
                    LMDZrepro                  MRI                           NiwaSOCOL                     SOCOL
             900
             800                                                                                                                            JUL
     Θ/K




             700
             600
             500
             400
             300                                                                                                                            JUN
            1000
             900    ULAQ                       UMSLIMCAT                     UMUKCA-METO                   WACCM
             800
     Θ/ K




             700
             600
                                                                                                                                            MAY
             500
             400
             300
                0       1     2      3      40   1     2      3      40   1     2      3      40   1     2      3      4
                    Hydrogen Chloride (ppbv) Hydrogen Chloride (ppbv) Hydrogen Chloride (ppbv) Hydrogen Chloride (ppbv)
    Figure 6.32: Climatological profiles of HCl from mid-May through mid-October for Aura MLS, 14 CCMs, and
       the multi-model mean.
Figure 2. Climatological profiles of HCl from mid-May through mid-October within the Antarctic polar vortex for NASA AURA MLS, 14 CCMs, and the
multi-model mean (MMM). The observations show how HCl decreases in the lower stratosphere through winter and early spring due to chlorine activation
       tracer correlation method destroy O3). The good MLS profiles
(i.e. conversion to active forms which(TRAC) are inOctober agreement showare low high HCl which indicates deactivation and an end to O3dur-
                                                                                   a peak of enough to allow the activation of chlorine loss.
            results show other established methods (Tilmes et of
       with generallyfrom this behaviour but with important differencesal.,detail.ing winter and spring. Thisactivation depends on many factors in
The CCMs                                                                           Accurate simulation of chlorine measure is called the potential
                                         are clearly uncertainties in all
       2004; WMO, 2007). Therewell as chemistry. The UKCA run gives strong, almost complete activation in the lowerdetails can but found innot
the model: Temperature, dynamics as                                               for chlorine activation (PACl) and stratosphere be this does
       ozone depletion approaches, taken from Chapter 6 of method
extend as high as in the MLS data. Figurehowever, the TRAC Eyring et al. (2010).  Tilmes et al. (2008). PACl is a measure that quantifies to
    has been shown to result in an under-estimation of chemi-                what amount meteorological conditions allow chlorine to
    cal ozone depletion rather than in overestimation in cases               be activated, and therefore ozone depletion to occur. This
    of a less isolated polar vortex, as summarized in Müller et              measure however does not necessarily imply that the mod-
    al. (2005, 2008).                                                        el vortex size and temperature distribution are simulated
          DATA A ND MODEL
          IN T ERC OMPA RISONS
          PREDICTION OF STORMS BY ENSEMBLE PREDICTION SYSTEMS
          Lizzie Froude, University of Reading


        Extra-tropical cyclones are the main natural             of the probability of the future weather. These initial stages are
        hazard of north-west Europe causing large amounts        obtained by applying small perturbations to the analysis (obtained
        of damage via strong winds and flooding. Whilst they     by assimilating observational data, including satellite, with a short
        can be damaging they can also be beneficial, providing   range forecast).
        the majority of the precipitation received in the mid-   A storm tracking methodology (Hodges 1996, 1999) has been used
        latitudes and are therefore vital to activities such as  to analyse the prediction of cyclones by EPS. Figure 1 shows an

                                           Prediction of                  by tracks and intensities predicted by the European
        agriculture. The accurate prediction of these weather storms of the ensemble prediction systems
                                                                 example
                                                                 Centre for Medium Range Weather Forecasts (ECMWF) EPS for a
        systems is
     Prediction therefore of key importance.                      systems
                    of storms by ensemble prediction Pacific storm.
                             
          An Ensemble Prediction System (EPS) is a type of forecasting               Froude, University of Reading
                                                                             Lizzierecent programme called TIGGE (THORPEX Interactive Grand
                                                                                  A
                            a set of Froude, University of members)
          system, in which Lizzie multiple forecasts (ensemble Reading
          are integrated from different initial states to obtain an estimate      Global Ensemble, http://tigge.ecmwf.int/) provides an archive of



               0                   0
                                                             0                 0       8                                                    8
               0                                             0                     8                                                8
                                                                                           7                                                    7
                                                              7                                                     7
              1                         2                   1 6            27                                                                           7
              1                                          4
                                                            1 5               6                           4             6
                  1                                                                                                                                         6
                       2                                     1                6                                             5
                         2                                      2                                                                                           6
                      Analysis                                    2
                      Mean Track            3 33               Analysis
                      Perturbed                          44    Mean Track 3 3 3
                                                                         5
                      Control                                      5
                                                               Perturbed                                  44
                                                                                                                                    5           5
                                   17
                                            Analysis
                                                               Control
                                   16
                                   15       Mean Intensity                   17
                                   14       Perturbed                        16         Analysis
                                   13       Control                          15         Mean Intensity
                                   12                                        14         Perturbed
                                   11
                                                                             13         Control
                                   10
                       Intensity




                                    9
                                                                             12
                                    8                                        11
                                    7                                        10
                                                                   Intensity




                                    6                                         9
                                    5                                         8
                                    4                                         7
                                    3
                                                                              6
                                    2
                                    1                                         5
                                    0                                         4
                           0      1       2      3      4       5      6     73   8   9
                                                   Forecast Lead Time (days)
                                                                              2
                                                                              1
1.Storm tracks 1. Storm tracks and intensities predicted byECMWF ensemble prediction system for a Pacific
         Figure
                  and intensities predicted by the the                        0
pical cyclone occurring in February 2005.                                       0   1   2     3      4       5      6           7       8           9
         ECMWF ensemble prediction system for a Pacific extra-
                                                                                                Forecast Lead Time (days)
         tropical cyclone occurring in February 2005.
                                         
 pical cyclones are the main natural hazard A storm tracking methodology (Hodges 1996, 1999)
                                 Figure 1.Storm tracks used to analyse predicted by cyclones
h-west Europe causing large amounts of has been and intensities the prediction of the ECMWF ensemble prediction system fo
                                 extra-tropical by EPS. Figure 1 shows an example
  via strong winds and flooding. Whilst they cyclone occurring in February 2005. of the tracks
   damaging they can also be beneficial, and intensities predicted by the European Centre for
g the majority of the precipitation received in Medium Range Weather Forecasts (ECMWF) EPS
        24
 latitudes and are therefore vital to activities for a Pacific storm.
                                                                                                                                                                              25




EPS data from different operational weather centres around the
world. Recently the storm tracking methodology has been used                          References:

to compare 9 different EPS archived in this programme. Forecast                       Froude, L. S. R., (2010), TIGGE: Comparison of the
                                                                                        Prediction of Northern Hemisphere Extratropical
verification statistics have been produced for cyclone position,                        Cyclones by Different Ensemble Prediction Systems.
intensity and propagation speed, showing large differences                              Weather and Forecasting, 25, 819–836
between the different EPS.                                                            Froude, L. S. R. (2009), Regional Differences in the
                                                                                        Prediction of Extra-tropical Cyclones by the ECMWF
The results show the ECMWF EPS has the highest level of skill                           Ensemble Prediction System. Mon. Wea. Rev., 137,
for all cyclone properties. Several of the EPS significantly under-                     893–911
predict intensity and interestingly all EPS under-predict the                         Froude, L. S. R., L. Bengtsson, and K. I. Hodges, (2007),
                                                                                        The Prediction of Extra-tropical Storm Tracks by the
propagation speed, i.e. the cyclones move too slowly on average
                                                                                        ECMWF and NCEP Ensemble Prediction Systems.
in all EPS (see Figure 2). For further details of this work please see                  Mon. Wea. Rev., 135, 2545–2567
Froude et al. (2007) and Froude (2009, 2010).                                         Hodges, K. I., (1995), Feature tracking on the unit
                                                                                       sphere. Mon. Wea. Rev., 123, 3458–3465
This work was partially funded by oil and gas consultancy
                                                                                      Hodges, K. I., (1999), Spherical nonparametric
Schlumberger (www.slb.com). Forecast information about severe                          estimators applied to the UGAMP model integration
weather is vital to the oil and gas industry for the management of                     for AMIP. Mon. Wea. Rev., 124, 2914–2932
oil and gas operations both onshore and offshore.




                                                                                                                         (b)
                                       positive bias: storms strength overpredicted
                            (a)
                                                                                       Propagation Speed Bias (kmh -1)
 Intensity Bias 10 -5s -1




                                                                                                                                                  Forecast Lead Time (days)
                                                Forecast Lead Time (days)

                            negative bias: storms strength underpredicted                                                      negative bias: storms propagate to slowly



Figure 2. Bias in prediction of extra-tropical cyclone (a) intensity and (b) propagation speed by different ensemble prediction systems.
     LEG AC Y DATA SE T S A ND
     QUA LI T Y A S SUR A NCE
     NCEO’S LASTING LEGACY DATASETS
     Victoria Bennett, NERC Earth Observation Data Centre/British Atmospheric Data Centre,
     Rutherford Appleton Laboratory

     The data generated from the activities within NCEO                    volcanic emissions and aerosol data. Model output from climate,
     are a valuable resource, and as such require effective                numerical weather prediction and chemical transport models are
     management so they remain an asset in years to come.                  also stored and disseminated to users.
     In particular, the long term global data sets produced                The data and documentation can be discovered, searched and
     using Earth Observation are key to understanding and                  browsed via the data catalogue and, given appropriate data types
     monitoring global change. The professional curation of                and formats, can feed into visualisation and analysis tools.
     these data ensures that the impact of NCEO’s activities
     reaches a wider audience, enabling knowledge                                                              Victoria Bennett
     exchange between different disciplines, sectors and
                                                                                                               ‘The data generated
     organisations.                                                                                            from the activities within
     The NERC data centres hosted by the Centre for Environmental                                              NCEO are a valuable re-
     Data Archival (CEDA) at RAL are the NERC Earth Observation                                                source, and as such require
     Data Centre (NEODC) and the British Atmospheric Data Centre                                               effective management so
                                                                                                               they remain an asset in
     (BADC). Together they hold, and make available, over 500 TB of
                                                                                                               years to come.’
     environmental data. In the last year over 2,700 users downloaded
     almost 90TB of data in approximately 15 million files. The
     helpdesks responded to nearly 4000 queries providing information
     to users on how to access, use and understand the data.
     Datasets already available through the data centres from NCEO
     scientists include products relating to global plankton and primary
     productivity, air-sea gas exchange, atmospheric ozone profiles,
     clouds and fire radiative power. More are expected in the coming
     years, including ice extent, sea surface temperature, methane,




26
                                                                                                                                                        27




SUCCESSFUL AIRBORNE CAMPAIGN IN THE ARCTIC
Daniel Gerber, Richard Siddans, Jolyon Reburn, Brian Kerridge (NCEO), Brian Moyna, Matthew Oldfield,
Simon Rea and David Matheson (MMT), Rutherford Appleton Laboratory




                                                                                                                  Figure 1. Impressions from
                                                                                                                  the PremierEx Campaign at the
                                                                                                                  Arena Arctica, Kiruna, in March
                                                                                                                  2010 showing the Geophysika
                                                                                                                  high-altitude aircraft with the
                                                                                                                  MARSCHALS mm-wave limb-
                                                                                                                  sounder in the front instrument
                                                                                                                  bay.

Atmospheric gases of importance to climate have been
measured in a new wavelength region over the Arctic
for the first time by a team involving NCEO scientists
from Rutherford Appleton Laboratory (RAL).
These measurements were made with an airborne millimetre-wave
atmospheric limb-sounder, built for ESA by RAL’s Millimetre-
wave Technology Group to demonstrate observing capabilities
of an advanced future satellite mission called PREMIER. In the
framework of the ESA funded PremierEx Campaign (the PREMIER
Experiment) the Russian high-altitude aircraft ‘Geophysika’ was
deployed from Kiruna, Sweden (68degN, 20degE). On a successful           Figure 2. Top Row: Atmospheric radiances measured by MARSCHALS in
                                                                         three frequency bands targeting H2O, O3 and CO; as well as other minor
flight on 10th March 2010 at 19km altitude, MARSCHALS
                                                                         species. Bottom Row: Cloud profile measured by the integrated Optical
demonstrated unique properties of the millimetre-wave region to          Cloud Monitor. Spectral features are visible well below cloud top altitudes.
observe the upper troposphere and lower stratosphere; the height-
range to which surface climate is most sensitive.                                                                  Brian Kerridge
From first inspection, the measured spectra have already shown                                                     ‘The NCEO team at
that vertical profiles of water vapour, ozone and other key gases                                                  RAL now look forward to
can be measured in the Arctic down to 5km altitude, and also in                                                    analysing the potential of
the presence of cirrus clouds which obscure infrared and shorter-                                                  the PREMIER mission to
wavelengths. This is a significant milestone for the proposed                                                      observe such processes on
PREMIER satellite mission.                                                                                         a global scale from space
                                                                                                                   in the future.’
The NCEO team at RAL with colleagues in UK universities and
around Europe now look forward to analysing the campaign data
in detail, to investigate exchange of mid-latitude and Arctic air, the
                                                                           References:
transport of polluted air from USA to Europe and the potential of
                                                                           ‘The PREMIER Experiment (PremierEx)’,
the PREMIER mission to observe such processes on a global scale              D. Gerber, R. Siddans, W. J. Reburn, B. J. Kerridge,
from space in the future.                                                    B. P. Moyna, M. L. Oldfield, S. Rea, D. N. Matheson,
                                                                             Rutherford Appleton Laboratory, CEOI Conference
                                                                             2010, Warwick, UK
     MIS SION SUPP OR T

      OBJECTIVES

     As part of its award from NERC, the NCEO received funding to support a portfolio of activities entitled ‘Mission
     Support’, primarily concerned with the development and early use of new satellite based Earth observation
     missions. Activities include:
        •	   development of early concepts and science drivers for new missions
        •	   provision of travel funds for science team members allowing them to be closely involved with mission
             development
        •	   support for scientists to develop algorithms in anticipation of launch of planned missions
        •	   continued development of algorithms and retrieval methods in the light of experience after launch
     Over the past year, 35 awards have been made, contributing to some 27 different missions or mission concepts.



      THE MAIN AWARDS ARE DESCRIBED BELOW


                                BIOMASS/IONOSAR – PI: Shaun Quegan, Sheffield
                                The project will develop strategies and methods to ensure that ionospheric effects do not hinder successful completion
                                of BIOMASS science objectives. The objective of ESA’s candidate Earth Explorer BIOMASS mission is to acquire global
                                measurements of forest biomass to assess terrestrial carbon stocks and fluxes. The mission is envisaged as a novel
                                spaceborne P-band synthetic aperture polarimetric radar.
                   Credit ESA




                                PREMIER Mission Support – PI: Brian Kerridge, RAL
                                The project will analyse data from the airborne PREMIER precursor instrument, compare with SLIMCAT/TOMCAT
                                model runs and hence deliver inputs to the scientific and technical case for PREMIER. The PREMIER mission aims to
                                advance our understanding of the processes that link trace gases, radiation and chemistry in the upper troposphere and
                                lower stratosphere. The instrumentation will consist of an infrared limb-imaging spectrometer and a millimetre-wave
                                limb-sounder.
                   Credit ESA




                                SWARM – PI: Richard Holme, Liverpool; Malcolm Dunlop, RAL
                                The objective of the SWARM mission, to be launched soon by ESA, is to provide the best ever survey of the geomagnetic
                                field and its temporal evolution, and gain new insights into improving our knowledge of the Earth’s interior and climate.
                                NCEO is funding two projects. One will develop methodologies to account for external field “noise” from sources above
                                the Earth. The other is concerned with calibration and validation activities for SWARM.
                   Credit ESA




                                GOCE/GRACE – PI: Philip Moore, Newcastle
                                GOCE (ESA) and GRACE (NASA) are complementary missions, currently in orbit, supplying measurements of the Earth’s
                                gravity field to high precision. The objectives of the project are to recover regional, high accuracy and high resolution
                                static gravity fields from a combination of the two instruments for scientific applications in oceanography, polar studies
                                and geodesy.
                   Credit ESA




28
                                                                                                                                                                             29




                                      EarthCARE – PI: Robin Hogan, Reading
                                      The project will develop and make available a new retrieval scheme for deriving properties of clouds, precipitation and
                                      aerosols from the suite of instruments on the EarthCARE satellite being developed jointly by the European (ESA) and
                                      Japanese (JAXA) space agencies.


                        Credit ESA



                                      GOSAT – PI: Paul Palmer, Edinburgh
                                      Satellite observation of CO2 from the Japanese GOSAT satellite are providing a step change in our current understanding
                                      of the carbon cycle, but using them presents significant challenges. The Mission Support project is helping to evaluate the
                                      official GOSAT XCO2 product that can subsequently be used to test current understanding of land-based fluxes of CO2.

                        Credit ESA


                                      Spaceborne Multispectral Canopy Lidar – PI: Iain Woodhouse, Edinburgh
                                      Mission Support funding aims to develop a new mission concept for a multispectral canopy lidar called SpeCL. The
                                      mission will measure the vertical profile of a forest and simultaneously determine the spectral characteristics of that
                                      profile to determine the global distribution of above ground biomass in the world’s forests.

     Credit University of Edinburgh




                                                                          Credit NASA                                                                           Credit ESA


CLARREO – PI: John Harries, Imperial College                                            GMES Sentinel 3 – PIs: Chris Merchant, Edinburgh;
                                                                                        Martin Wooster, King’s London
NASA’s Climate Absolute Radiance and Refractivity Observatory (CLARREO)
mission, will measure the Earth’s outgoing infrared spectrum using                      The future series of Sea and Land Surface Temperature Radiometers
advanced Fourier spectrometers to provide detailed information about the                (SLSTRs) on board the European GMES Sentinel 3 missions will have unique
Earth’s energy balance. The Mission Support project is aimed at a series                capabilities for long term observation of sea and land surface temperature,
of studies, working with colleagues in the USA, to determine the minimum                active fires, surface reflectance and atmospheric aerosols, building on
requirements to achieve the goal of detecting decadal scale climate change              the current series of (A)ATSR instruments. Mission Support has funded 2
and attributing the changes seen to specific causes.                                    projects aimed at land surface temperature, albedo and aerosol retrievals;
                                                                                        and fire power estimation.


 SMALL AWARDS

Travel funds and exploratory awards have been made to help UK scientists participate in and exploit a wide range
of key EO missions, including:
   •	 Indian, Korean and European instruments to measure ocean colour
   •	 Future NASA missions including Global Precipitation Measurement (GPM), Surface Water Ocean Topography
      (SWOT), Soil Moisture Active and Passive (SMAP)
   •	 ESA’s recently launched SMOS mission, focusing of improved measurements of ocean salinity.
Awards are also being used to explore the value of space data to monitor and understand volcanic processes, river
discharge, flooding, cloud top heights, wind fields and water vapour.
             Centre for Earth Observation Instrumentation -
               objectives and achievements, 2009-2010
    CEN T RE F OR E A R T H OB SERVAT ION
    INS T RUMEN TAT ION
   Centre for Earth Observation Instrumentation -
            objectives and achievements, 2009-2010




   OBJECTIVES
     OBJECTIVES
  The Centre for
OBJECTIVES              Earth Observation Instrumentation was created in 2007 as a result of joint support from th
     CentreCentre for Earth Observation Instrumentation was created in 2007 as a Strategy Board (TSB) and industries. Fu
       The for Earth Observation Instrumentation was (NERC), the Technology support joint support from the
TheNatural Environment Research Councilcreated in 2007 as a result of jointresult of from the
    byNatural Environment Research (NERC), the Technology Strategy Board (TSB) and industries. Fundedthe University of Leices
         Environment Research Council Council (NERC), the Technology Strategy Board together with
Natural the UK Space Agency, the CEOI is a partnership led by Astrium (TSB) and industries. Now funded
    STFC/ Rutherford Appleton partnership led by Astrium together with the University of University of Leicester,
by the UK Space Agency, the CEOI is a Laboratory and QinetiQ. The CEOI aim, andLeicester, success, is to bring togeth
       by the UK Space Agency, the CEOI is a partnership led by Astrium together with the key to
STFC/ Rutherford Appleton Laboratory and QinetiQ. The CEOI aim, and key to success, is to bring together
       STFC/ of UK industry industry and academia, encouraging the participation of
    the strengths of UKand academia, encouraging the participation of the full Earth observation the full Earth observation
the strengths Rutherford Appleton Laboratory and QinetiQ.
    community. Adopting this approach, the CEOI vision is to develop and strengthen UK expertise and
community. Adopting this approach, the CEOI vision is to develop and strengthen UK expertise and
    capabilitiesinstruments and positioning the UKtogether the strengths of UK leadingand academia, encouraging
       The in EO in EO instruments to bring to win leading UK to win industry roles in
capabilitiesCEOI aim, and key to success, isand positioning theroles in future international space future international space
programmes. The unified objectiveEarth observation community. Adopting this approach, the CEOI vision is to develop and
       the participation of unified these space programmes is to:
    programmes. Thethe full of objective of these space programmes is to:
       strengthen UK expertise and capabilities in EO instruments and positioning the UK to win leading roles in future
      improve our understanding of the processes driving the Earth system
           and space changes in the climate system
       internationalmonitorprogrammes. The unified objectives of these space programmes are to:
      detectimprove our understanding of the processes driving the Earth system
           monitor the environment in which we live, including the weather, land, atmosphere and oceans and
              detect and monitor changes in the climate system
            natural and human-induced hazardsthe processes driving
             •	 improve our understanding of                               •	 monitor the environment in which we live, including
             monitor the environment in which we live, including the weather, land, atmosphere and ocean
             the Earth system                                           the weather, land, atmosphere and oceans and
              natural and human-induced hazards
NCEO is pleased to collaborate with CEOI to ensure that the science drivers for new technology and new
missions are detect and to the UKchanges in the climate system
         •	 articulated monitor technology community.                   natural and human-induced hazards
  NCEO is pleased
ACHIEVEMENTS               to collaborate with CEOI to ensure that the science drivers for new technology and
     NCEO is are articulated to the CEOI to ensure that the science drivers for new technology and new missions are
   missions pleased to collaborate with UK technology community.
     articulated to the UK technology community.
Development of technologies for future EO missions
   ACHIEVEMENTS
                   Passive Microwave developments (STFC-RAL with Astrium)
     ACHIEVEMENTS Key technologies under development include a novel single sideband separating sub-
   Development of technologies for future EO missions a novel substrate-less optical
                   millimeter mixer, local oscillator source technology and
           DEVELOPMENT OF TECHNOLOGIES FOR FUTURE EO MISSIONS developments in microwave instruments, such
                             filter. Future missions that will use these
                              as Premier and Post-EPS for atmospheric composition measurement, are required for
                                           Passive order to understand the climate.
                              weather forecasting and in Microwave developments (STFC-RAL with                                         Astrium)
                                 Passive Microwave developments (STFC-RAL with Astrium)
                                              Key technologies under development include a novel single sideband separating s
                                  Key technologies under development include a oscillator source technology and a local oscillator
                                              millimeter mixer, local novel single sideband separating with SSTL)
                              CompAQS UVN compact spectrometer (University of Leicester sub-millimeter mixer, novel substrate-less op
                                  source technology and a novel substrate-less optical filter. Future missions that will use these developments in microwave
                              A demonstrator of a novel and missions that spectrometer to measure air quality hasin microwave instruments, s
                                              filter. Future compact UV/VIS will use these developments
                                  instruments, such as demonstrator has for atmospheric composition measurement, are required for weather
                              been developed. The PREMIER and Post-EPSbeen tested at the University of Leicester's
                                  forecasting as Premier and Post-EPS for atmospheric composition measurement, are required
                                              and Centre understand exceptionally compact instrument for differential
                              Space Research in order toand is an the climate.
                                              weather forecasting and in order to understandhas led to the
                              optical absorption spectroscopy (DOAS) applications. Further development the climate.
                              CityScan concept which is a ground-based instrument to provide 3D gas concentration
                                  CompAQS UVN compact spectrometer (University of Leicester with SSTL)
                              measurements and aerosol information across urban areas.
                                            CompAQS UVN compact spectrometer (University of Leicester with SST
                                 A demonstrator of a novel and compact UV/VIS spectrometer to measure air quality has been developed. The
                                 demonstrator has been tested at the Heterodyne Radiometer (STFC-RAL and is an exceptionally measure air quality
                                            A demonstrator University of and compact UV/VIS spectrometer to compact
                              Hollow wave guides for Laserof a novelLeicester’s Space Research Centre with
                                 instrument been developed. The spectroscopy (DOAS) applications. Further development has led to the
                              QinetiQ) for differential optical absorption demonstrator has been tested at the University of Leicest
                                 CityScan concept which is a ground-based new way to manufacturing compact, measurements and aerosol
                              QinetiQ has developed a fundamentally instrument of provide 3D gas concentrationlow mass
                                                       Research Centre and is an exceptionally compact instrument for differen
                                 informationSpacesystems.
                              and low cost across urban areas. These systems can maintain optical alignment in harsh
                                             optical
                                            thermal absorption spectroscopy (DOAS) applications. Further development
                              vibration and optical environments and the technology has been applied to the Laser                                           has led to
                                  Hollow wave guides for has revealed great potential to reduceinstrumentthe provide 3D gas
                              Heterodyne Radiometer. This Laser Heterodyne Radiometer (STFC-RAL with QinetiQ)
                                            CityScan concept which is a ground-based the size of to                                                         concentra
                              instrument bymeasurements and aerosol information across urban areas.
                                             an order of magnitude.
                                 QinetiQ has developed a fundamentally new way of manufacturing compact, low mass and low cost optical systems.
                                 These systems can maintain optical alignment in harsh vibration and thermal environments and the technology has been
                                 applied to the Laser Heterodyne Radiometer. This has revealed great potential to reduce the size of the instrument by an
                                 order of magnitude.
                                            Hollow wave guides for Laser Heterodyne Radiometer (STFC-RAL with
                                            QinetiQ)
                                            QinetiQ has developed a fundamentally new way of manufacturing compact, low m
  30                                        and low cost optical systems. These systems can maintain optical alignment in ha
                                            vibration and thermal environments and the technology has been applied to the La
                                                                                                                                                          31




                            GNSS-Reflectometry for measuring sea-surface state (SSTL with NOC,
                            Universities of Surrey and Bath)
                            This project is developing a flexible multi-channel receiver of reflected GNSS signals for surface measurements.
                            This development is important for deriving scientific data on the nature of the reflecting surface, such as the sea surface
                            roughness or soil moisture content. The measurement of ocean roughness is important for operational ocean and
                            weather forecasting.



                            Thermal IR detectors and on-board processing (Astrium with Selex Galileo,
                            STFC-RAL and University of Leicester)
                            The project team are using NERC’s Molecular Spectroscopy Facility to conduct experimental tests on a 2D thermal
                            infrared detector array system and investigate the design and operational issues in using 2D detector arrays to improve
                            the spatial resolution and coverage of Fourier Transform Spectroscopy (FTS). Future space-borne FTS instruments will
                            require high performance on-board digital signal processing to organise the large quantity of data produced by these high
                            resolution sensors. The latest technology is able to achieve this within tight mass, power and volume constraints.


 SEEDCORN PROJECTS

Seedcorn projects are smaller, more speculative projects which may have strong enabling potential for future earth
observation applications. Those funded in the 2nd CEOI Open Call which commenced in early 2009 include: Multiangular
IR Stereo Radiometer (MISRlite), Air quality monitoring from High Altitude Platforms and Frequency Selective Surface
(FSS) Filters. In the 3rd Open Call, three mainstream and five seedcorn projects were selected for funding by an
independent review panel, these projects commenced in 2010. Further information about these projects and the CEOI is
available at www.ceoi.ac.uk.

 INVESTING IN THE FUTURE


                   Horizon                                                                                             Business
                  Scanning                                                                                            Development
         Identify the UK’s highest priority                                                                         The CEOI carries out a
          future EO missions through the                                                                     comprehensive knowledge exchange
         Challenge Workshops. The 2009                                                                        programme focussed on identifying
      workshops covered surface/atmosphere                                                                   potential non-space applications for
       interactions, technologies for future                                                                        the technologies under
          Lidar missions and operational                                                                                 development.
                    EO missions.




                Learning                                                                                                 Publicity
            and Development                                                                                         The CEOI publicises its
         A long-term objective is to develop                                                                    technologies and achievements
          the highly skilled workforce and                                                                       through articles, conferences
      leadership necessary to maintain the UK                                                                 and seminars. There was significant
        at the forefront of the worldwide EO                                                                  recognition of CEOI activities in the
        community. This is achieved through                                                                      Space Innovation and Growth
            the training and development                                                                               Strategy reports.
                      programme.
     IN T ERN AT ION A L SPACE
     INNOVAT ION CEN T RE
     VISION
     The International Space Innovation Centre at Harwell                            as a UK shop window across the world. NCEO has
     will be a major focal point for UK space activities,                            been driving the vision for ISIC from the outset,
     promoting the best of UK science, technology and                                particularly focusing on the Earth observation
     innovation in space related applications and acting                             elements.

                                          UK Strategic Hub for End-to-End Operations in Earth Observation

     Observing                           Satellite                    Data                       Science and                          Applications and
                                                                                                                              
     technology                          operations                   management                 research                             services

     • Design and test facilities        • Satellite control          • State of the art         • Calibration and validation         • Application centres
     • Concurrent design                 • Mission payload              data management            of climate data                    • Business incubation
       facility                            planning                     facilities               • Synthesis and data
                                         •Ground segment              • UK ESA PAC/PAF             assimilation
                                          co-ordination               • UK EO Data Centre        • Data visualisation
                                                                                                 • Future observing systems


                                                                Outreach, education and training


                                        World-class platform for innovation, wealth creation and knowledge



                                              Government                    Business                      Public
                                             Informing and             Creating commercial        Inspiring, informing,
                                            delivering policy              opportunity                 influencing


     CURRENT DEVELOPMENTS
     The first phase of development is underway thanks
                                                                                                                                Andy Shaw, NCEO
     to a grant of £12M from the Strategic Investment
                                                                                                                                Knowledge Exchange
     Fund awarded to a consortium led by the Science and                                                                        Director
     Technology Facilities Council (STFC). NCEO is leading
                                                                                                                                ‘Space is entering a new
     the design and development of the ISIC Visualisation                                                                       phase of technical and
     Centre which will:                                                                                                         market development.
        •	 provide tools and facilities for data exploration and                                                                Exciting science and
                                                                                                                                commercial opportunities
           visualisation
                                                                                                                                exist for the UK that
        •	 enable access to large archives of scientific data                                                                   demand greater levels
        •	 generate content for outreach and media purposes                                                                     of collaboration and
                                                                                                                                cooperation, nationally and
        •	 enable innovation in market-focused applications                                                                     internationally than ever
        •	 provide a flexible environment for collaborative                                                                     before.’
           problem-solving by bringing business and science
           together.



32
                          4




NCEO MEETINGS 2009–2010
NCEO Annual Science Meeting, September 2009




National Centre for Earth Observation
The University of Reading,
Earley Gate Bldg 58, Reading, RG6 6BB, UK
Tel:   +44 (0)118 378 6728/8317
Fax: +44 (0)118 378 5576
Email: info@nceo.ac.uk
NCEO Director: Professor Alan O’Neill
Email: alan.oneill@nceo.ac.uk

NCEO Administrator: Jan Fillingham
Email: jan.fillingham@nceo.ac.uk
Web: www.nceo.ac.uk

								
To top