VACS Deliverables since SSG 14

1
VACS Deliverables since SSG 14

• Climate Prediction workshop, Dar es Salaam, July
  06 trained 30 operational scientists from NMSs
  and ocean agencies from 20 African countries in
  the Climate Predictability Tool (CPT) software
• Workshop funded by Royal Society / S A govt (NRF)
  on southern African programme last week
  GIRAFFE
• Ongoing development of WCRP-CLIVAR African
  Climate Atlas
• VACS East African programme
• AMMA (former Co-Chair of VACS also Co-Chair of
  AMMA) special issue of CLIVAR Exchanges
• Africa breakout session organised at the WCRP
  Seasonal Prediction workshop, Barcelona, June 07
• Young African Scientists Day at JSC Zanzibar
  organised by VACS outcomes include special
  issue of Int. J. Climatol. (10-12 papers) and
  entrainment of these young scientists in GIRAFFE
  and East African programmes
• 2006 BAMS article African climate change
  impacts need to be dealt with on S2D timescales
• Africa sessions at conferences organised by VACS
  members eg AMS Jan 08 New Orleans (Kerry Cook)
• Hosted IOP4 at SA Weather Service to develop
  interactions with IOP

2
VACS Southern and Eastern African Climate
Predictability Workshop Tanzania Meteorological
Agency, Dar es Salaam 10-13th July 2006
Seasonal Climate Prediction Network for Africa

• Attendees from NMSs of each southern African
  country and 5 East African countries as well as
  from 6 operational ocean agencies
• Climate Predictability Tool -IRI

The Climate Predictability Tool (CPT) provides a
Windows package for constructing a seasonal
climate forecast model, performing model
validation, and producing forecasts given updated
data. Its design has been tailored for producing
seasonal climate forecasts using model output
statistic (MOS) corrections to climate
predictions from general circulation model (GCM),
or for producing forecasts using fields of
sea-surface temperatures. Although the software
is specifically tailored for these applications,
it can be used in more general settings to
perform canonical correlation analysis (CCA) or
principal components regression (PCR) on any
data, and for any application.
3
GIRAFFE
reGionally IntegRated southern AFrican climate
and Forecasting for sociEty
Workshop sponsors UK Royal Society / SA govt
National Research Foundation
4
GIRAFFE
reGionally IntegRated southern AFrican climate
and Forecasting for sociEty
5
GIRAFFE DESIGN QUESTIONS

• Research Themes
• Basic state
• Variability
• Modelling
• S2D Prediction
• Climate Change
• Observing System
• Training
• User Interface

6
BASIC STATE

• Justifications
• Forcing of global circulation which originates
  from southern Africa is unknown (e.g. MJO,
  convectively forced waves)
• Southern Africa is expected to be subjected to
  major desertification as a result of ACC but
  model simulation of the basic state and
  teleconnections are problematic

7
BASIC STATE

• Justifications
• Examination of forcing of global circulation from
  African convection (e.g. MJO)
• Detailed assessment of performance of climate
  models, individual and ensembles, for seasonal
  and climate change uses
• Examination of forcing modes in atmosphere vs.
  models modes produce higher-order statistics
  required by users
• Enhanced information on potential climate
  predictability in the region through the year
• Improved post-processing of model information for
  users
• Major SADC input to work

8
BASIC STATE

• Questions
• Dynamics of rain-bearing and other systems
• Mechanisms of cloud band generation
• Mechanisms of other convective systems
• Dynamics of the flow at 5-20S over Africa
  (Angola Low, easterly waves, ITCZ dynamics and
  uniqueness of meridional arm, why a standing wave
  with no propagation into Atlantic Brazilian
  hurricane? ?)
• MJO initiation (NB effects on global
  circulation)
• Wave generation by African convection effects on
  global circulation
• Intraseasonal dynamics (semi-annual cycle)
• Causes of winter dryness
• Controls on surface temperatures
• Affect of aerosols on dynamics
• Forcing mechanisms links to global circulation
• Mechanisms by which ENSO affects African rainfall
• Role and mechanisms of surrounding ocean basins
  on African rainfall
• Mid to high latitude mechanisms on African
  rainfall variability (SAM, wave 3-4, sea ice)
• Land surface mechanisms
• Predictability
• Of the SW Indian Ocean upper heat content and
  SSTs
• Effect on land surface feedback relevant space
  and time scales

9
VARIABILITY

• Modulation of ENSO modes, and their interactions
  with southern African climate (rainfall and
  temperature), on decadal and multi-decadal
  timescales
• Stability of teleconnections (ENSO plus others)
• Combination effects of teleconnections on
  southern African climate
• What is the influence of the PSA on southern
  African climate

10
MODELLING

• Quality of simulations of the basic state
  (rainfall and temperature) (q.v.)
• Quality of simulations of the modes of
  variability (q.v.)
• What are the limits to our ability to model
  regional climate systems?
• Mechanistic examination of regional climate
  dynamics
• What are the benefits of high resolution over low
  resolution modelling in simulating regional
  climate?
• What is the sensitivity of the simulation of
  regional climate to various parameterisation
  schemes?
• Does high resolution modelling of the
  neighbouring oceans assist in simulating regional
  climate?
• What is the sensitivity in models of regional
  rainfall systems to variations in aerosol
  concentrations?

11
PREDICTION

• What are the implications and limitations
  introduced for prediction of the ability of the
  current models to simulate the basic state
  (rainfall and temperature) and variability, as
  determined in earlier sections
• Do coupled models add value/quality/operational
  performance over and above other modelling
  approaches
• What is the predictability of predictability
  forecasting uncertainty?
• What is needed in order to make land surface
  assimilations over southern Africa?
• What is the predictability of variations in
  intra-seasonal rainfall?
• What is the optimal ensemble prediction system
  for the region (resolution, number of models,
  number of members, weighting, post-processing,
  etc.)?
• What is the optimal approach for generating
  forecast variables for inclusion in application
  models?

12
CLIMATE CHANGE

• What are the implications and limitations
  introduced for climate change prediction of the
  ability of the current models to simulate the
  basic state (rainfall and temperature) and
  variability, as determined in earlier sections
• How will the basic state (rainfall and
  temperature) (q.v.) and variability (q.v.) of
  southern African climate change in future
  climates?
• How well are high impact events simulated in
  climate models, how will these change, and how
  well are these likely to be simulated for future
  climates?
• What will be the future dynamic structures of
  southern African rain bearing systems?
• How will teleconnections to southern Africa,
  including ENSO, change under future climates?
• What are the potential impacts of future
  predicted climates on society, what uncertainties
  are involved, and what are the limitations to
  this knowledge?
• How do the uncertainties of climate change
  predictions affect planning?

13
OBSERVING SYSTEMS

• Climate systems
• Target Angola Benguela Frontal Zone/south-east
  Atlantic and Indian Ocean thermocline ridge
  AIP/IOP
• Land surface lack of data
• How do we link this to?
• ClimDev
• INDOOS
• PIRATA
• GCOS
• TIGER, etc.
• Observations for applications
• How do we access existing data?
• How do we encourage creation of quality data
  sets?
• How do we encourage use of accompanying data and
  metadata?
• How do we promote QA?
• How do we link to training activities?
• Database management (climate and
  applications-relevant)
• How to connect to ClimDev (see Roger Stern
  contribution)?
• How might it be developed across SADC?
• How best to handle in SAGRADEX?

14
IMPROVED SOCIETAL USES FOR CLIMATE INFORMATION

• What is the current usage of climate information
  and prediction in decision making?
• List of sectors/institutes/levels
• List of information currently used
• What empirical underpinning is required to
  further develop application models?
• What are the societal consequences of rainfall
  variability in the region
• What defines a high impact event?
• What climate information is required for societal
  use?
• Role of near-real time observation products
• Level of model skill required
• Timing of delivery of information
• How might society be conditioned to expecting
  longer-range predictions only at times of
  relatively high predictability?
• How should uncertainty be conveyed to society?
• What is required to action the use of climate
  information and forecasting (including ensemble
  prediction) systems for society?
• Data availability
• Linking information to decision processes
  understanding decision processes
• Post processing (downscaling bias correction)
• Verification of probabilistic output
• Is there a need for selected pilot studies?
• What are the main issues preventing data use

15
TRAINING

• How to incorporate pilot studies and RCOFs for
  training?
• How to produce training materials, including
  workshops, web sites, wikis, blogs?
• Centrally controlled?
• Overall management?
• Brand image?
• How do we train climate scientists?
• How to attract from other disciplines (ACCESS)?
• How to include relevant climate science in other
  curricula?
• What is the best strategy to include
  governmental, NGO and social sector officers in
  forecasting systems?
• What level of involvement works best, and at what
  stage of the process?
• How do advise the general public of climate risk
  and the use of climate information (ACCESS)?
• How to involve the media?

16
NEXT STEPS/SCHEDULE

• Meeting Report (2 pages) Richard/Chris end
  09/07 send to RS, NRF
• Andy to set up a blog or wiki end 09/07
• Core Group to review consistency of design
  (spreadsheet Andy to set up on Google) end
  first week 10/07
• Group at SAGRADEX meeting to review
  coordination by Chris mid 11/07
• Preparations for Flier and White Paper end
  11/07
• Programme logo Andy
• Budget estimate Chris/Richard/Andy
• Organisational diagram Andy
• Create extended list of potential collaborators
  Chris/Willem for Africa Richard/Andy/Mike for
  elsewhere
• Time line (for programme spin-up) Core Group
• Interrelationship in time for project components
  Mike
• Brief status review Willem/Chris
• A4/A5 Flier and stock powerpoint slides Andy
  end 11/07
• Web site initialisation Willem/Chris mid
  11/07
• White Paper (10-12 pages with executive summary)
  - Richard/Chris lead end 11/07
• gain endorsement (CLIVAR, ESSP) Richard/Chris
• approach potential collaborators (see list)
  Core Group
• approach funders (RS, NRF, DST, NERC, funders
  slide, plus others check UK/David King) Core
  Group
• identify champions primarily, but not necessarily
  uniquely, for societal link Core Group

17
GIRAFFE reGionally IntegRated southern AFrican
climate and Forecasting for sociEty a 5-year
programme with vision!
18
WCRP-CLIVAR African Climate Atlas

• Part I - Climatology
• Part II - Anomalies
• Part III - TOMS Absorbing Aerosol Index
  (interactive visualization)
• Part IV ERA-40 interactive plotting tool
• Part V African Climate FAQs
• Part VI IPCC AR4 data portal

Part VI Thanks to WCRP secured funds from
GEO IPCC AR4 coupled simulation data and plotting
tool Interactive Data and maps for download
19
Young African Scientists Day JSC Zanzibar
March 2007
Young African Scientists Day JSC Zanzibar
March 2007 Kenya, Tanzania, Uganda, Rwanda,
Senegal, Botswana, Mozambique, South Africa,
Zambia Int. J. Climatol. Special Issue
submissions now
Sunset at Stone Town beach, Zanzibar, March 30,
2007
20
VACS Lake Victoria project
97-98 El Nino
1961-62 IO Warming
no trend pre-1961 lake levels
Observed lake level drop due to hydroelectric
dam water over release at source of the Nile?
AND/OR
Is lake level drop due to 2005-6 record
drought/temperatures of eastern Africa?
Goal Verification Prediction
Satellite
NCEP reanalysis
Nile Outflow Lake Level
RegCM3 RCM
RegCM3 RCM
VACS Modeling Research (surface temperature and
winds)
21
Prediction of Lake Victoria Levels
Lake Victoria levels validation based on
preliminary results using our version of the
modified Tate et al (2004) modified water balance
model (diamond - blue), and observed lake levels
(square - pink). Input is six rain-gauge station
data around lake. Next step is to use input from
the RegCM3 regional climate model.
22
Greater Horn of Africa Regional Model
Inter-comparison Project(ARMIP)

• Objectives
• Inter-compare regional climate simulations of the
  GHA climate based on different state-of-the-art
  RCMs to explore and evaluate uncertainties in the
  present skill of these models over the region.
  Participating Models (so far NCEP-RSM PRECIS,
  RAMS, RegCM3, WRF/MM5)
• Develop criteria for modifying and improving the
  parameterizations of various physical processes
  (e.g., convection, boundary layer, radiation) as
  appropriate for improving the performance of
  regional climate models over the GHA sub-region
• Develop a regional seasonal climate dynamical
  prediction system
• Develop high-resolution regional climate change
  scenarios to assess the impacts of climate change
  on regional hydro-climates, including
  fluctuations of Lake Victoria surface elevation
  and the hydrology of the entire Nile River
  Basin
• NOTE The ARMIP Project is an initiative of
  Richard Anyah, currently funded (limited) by
  NSF2007-2009. It would be great if CLIVAR-VACS
  panel members who are currently running Regional
  Models would come forward to participate by
  donating their models, time and computing
  facilities.

23
Predictability of East African short rains based
on regional model simulations
NEK
NUG
NEK
NUG
CEK
CTZ
CTZ
CEK
Figure 3 Regional Model (RegCM3) simulated
interannual rainfall variability in different
homogeneous climate zones over East Africa
(Anyah and Semazzi, 2007)
24
AMMA - WG1Scientific objectives (1)

• Integrative view of the variability/impact/predict
  ability of the WAM
• Consider processes operating at global and
  regional scales
• Variability and Predictability of the WAM
• Key-SST anomaly patterns linked to WAM
  variability and associated teleconnection
  mechanisms Impact of other monsoon regions and
  the rest of African continent To what extent
  this is predictable ?
• What mechanisms determine the observed
  intraseasonal variability over West Africa ?
  Roles of intra-continental continental
  teleconnections and equatorial waves
• Mechanisms that control the annual cycle of the
  WAM including monsoon onset
• To what extent the key modes of variability of
  the WAM are predictable ? To highlight the
  predictable aspects of the WAM

25
SST-WAM coupled modes at interannual scale (obs)
Joly et al. 2006
26
SST-WAM coupled modes at interannual scale
(IPCC4)
Coupled models do not simulate accurately the
teleconnections at interannual scale
Joly et al. 2006
27
AMMA - WG1Scientific objectives (2)

• Integrative view of the variability/impact/predict
  ability of the WAM
• Consider processes operating at global and
  regional scales
• Impacts of the WAM on the global scale
• Variability and Predictability of Atlantic
  tropical cyclone activity role of WAM heating
  and teleconnections that impact the environment
  where the cyclones develop Variability of the
  weather systems that trigger many of the tropical
  cyclones
• Aerosol variability Improve our knowledge of
  the aerosol physical, chemical and radiative
  properties (Sahara dust, biomass burning,
  sulfates) sign of (in-)direct radiative effect
  modification of aerosol properties during
  long-range transport effect on cloud properties
• Atmospheric chemistry trace constituents
  uplifted by convection in the free atmosphere,
  then transported over large distances What is
  the extent of this long-range transport and their
  impact on the global oxydizing capacity and
  global radiative forcing ?

28
Monitoring of African easterly waves

• Hovmoller space-time diagram of 700 hPa curvature
  vorticity averaged between 5N and 15N based on
  GFS analysis. The numbers refer to synoptic
  systems that meet the tracking criteria used by
  Berry et al. (2007). Those in red became tropical
  cyclones in the Atlantic basin. These were Chris
  (6), Ernesto (13), Debby (14), Florence (18),
  Gordon (19), Helene (21), Isaac (23).On the
  vertical axis time is expressed as month-day/hour.

Berry et al. 2007
29
Issues for SSG

• Africa tends to be just viewed as a monsoon
  system rather than a vast continent with distinct
  regional climate processes and impacts that
  include monsoons
• Despite very limited resources, VACS has made
  significant progress
• For Africa, adaptation to climate change and
  variability is vital info gap on decadal
  timescales
• Progress largely depends on the enthusiasm of a
  small group of individuals, crucial help from
  WCRP and external funding
• The challenge is to extend this enthusiastic
  small group to a much larger network throughout
  Africa and worldwide

30
SUMMARY

• Despite very limited resources, VACS has made
  significant progress
• For Africa, adaptation to climate change and
  variability is vital
• Progress largely depends on the enthusiasm of a
  small group of individuals, crucial help from
  WCRP and external funding
• The challenge is to extend this enthusiastic
  small group to a much larger network throughout
  Africa and worldwide