Climate Exchange

[

] 273

apacity

evelopment

Improving the application of science and climate models

for early warning systems and adaptation planning

The CSRP is developing new, experimental, real-time

climate forecasts for users in Africa. Products have

been designed to meet the needs expressed in the initial

consultation of user needs and are currently being

tested with African regional organisations and National

Meteorological Services. The timing of the onset of rains

is crucial to agricultural production and food security. If

planting is done too early, seeds rot in the ground; if too

late, the early growing season is missed. Experimental

seasonal forecasts of onset timing have been developed,

using the Met Office’s seasonal forecasting system, and

evaluated over past seasons. The performance is encour-

aging – in particular, good guidance for the early onset of

the 2011 short rains season over the Greater Horn of Africa

was achieved (above). The forecasts are currently being

trialled by climate organisations in west, east and southern

Africa and at Regional Climate Outlook Forums (above

right). These dynamical forecasts of onset are the first of

their kind and are additional tools to help African regional

centres and NMHSs improve their predictions of onset and

thereby increase agricultural output and food security.

A new monthly-to-decadal prediction system has been

developed based on the newHadGEM3 climate model and

has improved our abilities to predict multi-annual (next

five years) rainfall and temperature averages over Africa.

The key impact of this is enhanced potential to provide

Key results of the consultation were as follows:

• There was wide recognition of an urgent need to improve the

understanding and modelling of African climate in order to

provide reliable climate early warning systems and information

for adaptation

• Improvement of seasonal-range predictions (to six months ahead)

was seen as a higher priority than multi-annual to decadal-range

predictions, though the need for decadal-range information for

adaptation was seen as urgent

• There was a very strong signal that development of capability to

predict temporal distributions of rainfall (such as season onset,

duration and dry spell frequency) was the highest priority need.

The consultation also addressed training needs and possible ways

to respond. The results of both the research and training prioritiza-

tion exercises were then used to make adjustments to the project’s

science and training components.

CSRP outputs

Improved understanding and modelling of African climate and its drivers

The first output is advancing understanding of African climate proc-

esses in order to improve their representation in climate models.

Work is focusing on the new Met Office Hadley Centre climate model,

HadGEM3. Key results include improved representations of rainfall in

all important sub-Saharan rainy seasons, enhanced understanding of the

drivers of variability in theWest AfricanMonsoon and a comprehensive

assessment of the ability of present-day climate models to represent the

important influences of the global oceans on African rainfall.

Image: UK Met Office

Forecast probabilities for early (left) and late (right) onset of the 2011 short rains season over the Greater Horn of Africa. The predictions give a

substantially raised probability (>60%) of an early onset over much of the eastern side of the Greater Horn of Africa and early onset was observed to

occur (with definitions used here, early onset has a baseline chance of occurring of 33%).