Diapositiva 1

1
EGU2012-8848
Regional and sub-regional climate simulations
over Sub-Saharian African regions and the
influence on the heat waves hazard
E. Bucchignani (1), P. Mercogliano (1), M.
Montesarchio (1), P. Capuano (2) (3), M.
Sellerino (2), A. Di Ruocco (2), and W. Kombe (4)
(1) C.M.C.C., Impacts on Soil and Coast, Italy
(e.bucchignani_at_cira.it), (2) AMRA S.c. a r.l.,
Napoli, Italy, (3) University of Salerno,
Fisciano, Italy, (4) ARDHI University, Dar Es
Salaam, Tanzania
5. Heat Waves
1. Introduction
3. Dar Es Salaam case
The number of hot days per year (Fig. 7a) shows
an increase of heat wave recurrence from 1961 to
1995. The successive period 1996-2000 shows a
minimum in the recurrence of the phenomenon, but
the last period 2001-2011 has experienced a
strong increase in the Heat Wave occurrence. Hot
days number evaluated by observed and modeled
data are also compared. The forecasted period
2011-2050 shows a strong increase in the heat
waves occurrence
Africa is considered a continent particularly
vulnerable to the effects of climate change but
the development of reliable predictions of future
climate change is complicated by the lack of
accurate baseline data on current climate and the
complexity of drivers of African climate
variability (i.e., El Niño/Southern Oscillation,
land cover change). In the framework of the
EU-FP7 CLUVA project (CLimate change and Urban
Vulnerability in Africa), regional projections of
climate change at high spatial resolution, based
on the IPCC RCP4.5 and RCP8.5 scenarios have been
performed for selected areas surrounding five
African cities of interest (i. e. the project
test cities). Climate simulations, spanning
over the period 1950-2050, have been compared
with large scale observed data (CRU) and with
local data provided by the municipalities of the
test cities, in terms of two-metre temperature
and precipitation. The variations of the
fundamental climate parameters will constrain
different hazard at different time-scales. In the
framework of the assessment of the environmental
impacts and the risks of climate changed induced
hazards likely to affect urban areas at various
time frames, we analyzed a dataset of daily
maximum temperature from Dar Es Salaam (Tanzania)
in the period 1961-2000 to evaluate the
generation of heat waves, defined following a
peak over threshold approach, and the evolution
in the extreme warm conditions up to 2050.
(a)
Fig. 3 shows the map of the bias of 2-meter
temperature, (averaged over the time period
1971-2000) for DJF (a) and JJA (b) obtained by
comparing the CLM solution with CRU dataset.
Fig. 4 shows the seasonal cycle of temperature,
respectively for CLM solution (with and without
bias correction) and for observed data.
Heat wave duration (Fig. 7b) is a key element to
evaluate the societal heat wave impacts, reducing
the capacity of adaptation. The maximum length
(days) of heat wave episodes per year shows a
mean value of 4-5 days with some peak in 1974,
1987, 1994 and 2009. Successively, there is a
clear increase in the average duration of heat
wave with peaks of 18-19 days.
(b)
Fig. 4
Fig. 7a,b
Fig. 3 (a) (b)
Heat wave duration and hot days number are
strictly correlated (Fig. 8) showing that the
temperature rise could generate not only an
increase of heat waves number but mainly a longer
average duration, that can strongly affect the
resilience capacity of the population,
particularly the elder people.
2. Regional and sub-regional climate simulations
over Sub-Saharian African regions
Climate projections are generally conducted using
global models, which provide information around
the whole Earth surface. However, they are
generally unsuitable to simulate the climate of
very limited areas, since they are characterized
by resolutions generally around or coarser than
100 km, which is too poor for impact studies on
areas where many important phenomena occur at
spatial scales lower than 10 km. The usage of a
Regional Climate Model (RCM) with a horizontal
resolution of about 10 km can be a useful tool
for the description of the climate variability on
local scale. The CLM is the climate version of
the COSMO model, which is the operational
non-hydrostatic mesoscale weather forecast model
developed by the German Weather Service.
Successively, the model has been updated by the
CLM-Community, in order to develop also climatic
applications. A huge variety of applications of
the model system exists, covering high resolution
simulations of mega-cities to medium resolution
simulations of continents, tropical to arctic
latitudes, paleo studies, the recent past and
climate scenarios for the 21st century. This
makes it highly relevant for climate science and
for climate mitigation and adaptation politics.
Fig. 5 shows the time series of 2-meter
temperature averaged over a square area (length
40 km), centred on Dar Es Salaam and obtained
with both the considered scenarios. An increase
of temperature is projected, especially employing
RCP8.5 scenarios.
Fig. 5
Fig.8
Fig. 9
The frequency distribution plot of hot days
duration (Fig. 9) for five separate bi-decadal
periods, according to the legend of top right
corner, shows the temporal change of heat wave
characteristics. This distribution have became
longer tailed with time. The number of events of
the maximum length lasting 5 days could increase
from 3 to 24 over 100 years (from 1950-70 to
2030-2050).
4. Heat Waves
Heat wave generally refers to periods of
exceptionally warm temperatures, but there is no
universal technical definition (Robinson, 2001).
The impacts of heat waves on the society are
determined also by temporal duration (Stephenson,
2008), in addition to their frequency, in fact
the capacity of adaptation can be reduced with
prolonged exposure to high temperature and
humidity.
For this specific activity, the utility of
COSMO-CLM is due to the fact that the model
contains physical parameterizations (cloud,
precipitations, moist convection radiation)
suitable for African areas. In the present work
the climate predictions have been obtained
forcing COSMO-CLM with the CMCC-MED global model
output. The IPCC emission scenarios considered
are the RCP4.5 and RCP8.5.
A heat-wave is defined, here, as a period in
which the maximum temperatures are over the 90th
percentile of the monthly distribution for at
least three days, fixing a minimum threshold for
Tmax to exclude months with less hot temperature
(similar to winter time), that for Dar Es Salaam
was fixed to 32C. The 90th percentile has been
evaluated over a climatological base period
(1961-1990).
6. Conclusions
Regional Climate Model with a horizontal
resolution lesser than 10x10 km can be a useful
tool for the description of the climate
variability on local scale, suitable for impact
studies. We evaluate regional climate projection
for some African areas, particularly for Dar Es
Salaam. We use the climate simulations to
evaluate the changes in the heat wave occurrence.
The combined change in the mean and variance of
CRFD has contributed to an increase in the
frequency of hot days and in their duration.
Projected future warming in the Dar Es Salaam
area shows a further increase in the heat waves
parameters. The expected persistence of
long-lived heat waves lasting approximately 1.5-2
weeks is clearly longer with respect to the
climatological period (1961-1990). During 100
years, short lived but more intense waves are
more than doubled in duration. It is evident the
needs for the national health services to develop
strategies for the mitigation of the heat wave
effects, to enhance the resilience of the
population, particularly the elder people.
A rectangular limited domain centred over the
city of Dar Es Salaam has been considered
(Fig.1). It has been discretized using a spatial
resolution of 8 km, employing a computational
grid with 95 x 135 grid points and 40 vertical
levels. The time resolution is 50 sec. Fig. 2
shows the Carbon Dioxide concentration time
series used to force the model, according to the
different scenarios.
Fig. 6a
Fig. 6a shows the comparison between the observed
and bias-corrected modelled cumulative relative
frequency distribution (CRFD) for daily Tmax. The
differences are partly due to an effect of the
incompleteness of observed dataset. Fig. 6b shows
CRFD change over time, for each of ten
independent periods, according to the legend in
the top right-hand corner. Taking into account
for the data coverage, it is clear that the Tmax
peak increases from 30.4C (1961-70) to 31.5
(2001-2011). In addition all the distribution is
moving towards higher temperature in the last
decades.
Fig. 1
References
Stephenson, D. B. (2008). Definition, diagnosis,
and origin of extreme weather and climate events.
In Diaz, H. and Murnane, R., editors, Climate
extremes and society, chapter 1, pages 1123.
Cambridge University Press Robinson P. J.
(2001), On the Definition of a Heat Wave, J.
Appl. Meteor., 40, 762-775 Rockel B., Will A.,
Hense A., 2008. The regional Climate Model
COSMO-CLM (CCLM), Meteorologische Zeitschrift, 17
(4) 347-348.
Fig. 6b
Fig. 2