 |
| | | | | Rainfall Variability | |
| |
|
|
| Tanzania
|
 |
|
|
|
|
|
|
|
|
|
|
Tanzania
|
|
|
|
|
|
|
|
Data
In order to apply the methodology used on the global scale with higher spatial
and temporal resolution in the period 1983 to 2002 dekad rainfall grids at a
spatial resolution of approx 0.1 degree are computed. Concerning Burkina Faso,
133 rainfall stations have been available for the selected time horizon. In this
case interpolation techniques can be applied directly on rainfall station data.
The interpolation has been performed with inverse distance and regression methods
provided within SEDI (AMS).
For Tanzania, Cambodia and Nepal available stata data was insufficient. So, the
generation of the rainfall grids has been achieved through a "downscaling" procedure
of global NOAA rainfall grids having a monthly resolution and a spatial resolution
of 0.5 degrees. Within this procedure for each of the three countries a "virtual"
grid of rainfall stations is constructed with a densitiy and distribution of grid
points comparable to the station data in Burkina Faso. After the extraction of the
rainfall data for each grid point from the global NOAA rainfall grids, inverse
distance and regression methods (SEDI) are used again in order to generate the
desired spatial resolution. In coherence with Burkina Faso, most interpolation
has been carried-out using the inverse distance technique. Finally, the monthly
rainfall grids are converted to dekadal grids using a utility by R. Gommes.
The rainfall maps and the maps presenting the rainfall index for agroecological
zones and the first administrative level concerning Burkina Faso and Tanzania are
in Hammer-Aitoff projection and all maps concerning Cambodia and Nepal are in Goodes
Homolsine projection.
Climatological facts about Tanzania
Like most of Southern Africa, the southern and western regions of Tanzania
have only a single wet season. One rainy season is observed in austral
summer (October-April). But in the north-east of Tanzania bi-modal rains are
found. Similar to Kenya, the latter tends to receive rain during October to
December and during March to May. Furthermore, Tanzania has high elevation
climates. Consequently, connsidering the whole country, planting and harvesting
takes place throughout the year (Gommes and Houssian, 1982).
Several papers (Ogallo, 1988, 1989; Hastenrath et al., 1993; Reason et al., 2002)
confirm an influence of the El Nino-Southern Oscillation phenomenon on Tanzanian
climate. But the western part of Tanzania seems to be in the transition region
between the areas of strong ENSO impact with above average rainfall over East
Africa and below average rainfall over southern Africa during an El Nino event.
Additionaly, the dipol-mode of the Indian Ocean (Saji et al., 1999), which is
an ENSO-like ocean-atmospheric system acting over the Indian Ocean, influence
Tanzanian rainfall amounts. A positive (negative) Indian Ocean Dipol (IOD) is
associated with a lower (higher) SST in the East Indic and a higher (lower)
SST in the West Indic. A positive IOD produces weaker equatorial westerlies,
an increase of SST in the Western Indic and higher rainfall amounts in East
Africa. Worth mentioning, that this goes along with lower temperature
differences between the Indian Ocean and the Indian Subcontinent. The result
is a weaker southwest summer monsoon in India.
Besides these large scale modes, considerable intraseasonal variability
during the northeast monsoon and the transitions to and from the southwest
monsoon near the beginning and end of the wet season can produce wet and
dry spells during the dry season.
|
|
|
