| Summary: | Activity in several fields such as hydrology, climatology and agriculture
related and determined by variation of rainfall. Variability of rainfall distribution is
influenced by many factors, from global to local scale. In this research, analysis of
rainfall variation with global and local scale factors were done. Global scale are in
the forms of global climate phenomena, they are the teleconnection sea surface
temperature (SST) anomalies of the Pacific Ocean (Nino 3.4 Index) and the
difference of SST anomalies between eastern (IODE) and western (IODW) of
Indian Ocean (DMI). The regression analysis showed that SST variations occured in
both of the ocean significantly influenced rainfall variations in some locations in
Central Java and DIY during SON, DJF and JJA periods. Impacts of Nino 3.4 index
more apparent in Yogyakarta and eastern part Central Java. IODE and DMI were
significantly more influential in western part of Central Java. The influence of these
three variables to rainfall variations in significant locations are about 40 % -72 %.
SST anomalies in the Pacific Ocean could generated ENSO phenomenon (El Nino
or La Nina). Meanwhile, the difference of SST anomalies between western and
eastern could generated IOD phenomenon. El Nino event caused average rainfall
decrease and the difference is significant compared to neutral conditions during
SON period. During DJF period, the reduction of rainfall is more significant in the
lowlands (elevation <100 m above sea level/asl) during El Nino and IOD (+) events
simultaneously. La Nina event caused rainfall increased during DJF period, and
extreme rainfall became more frequent. La Nina impact is more significant in
locations that are below 500 m asl .
In the local scale, geographical position and topography are factors that
influenced rainfall variations. Based on the geographical position, seasonal rainfall
(SON, DJF, MAM and JJA) are generally higher in the western than eastern part of
the study area and positively correlated with altitude. Spatial rainfall can be
obtained by interpolation. To estimate spatial rainfall is used non-linear regression
multivariate approach with topography as independent variables. Topographic
variables considered are latitude, longitude, altitude, slope and aspect. The rainfall
estimation model formed from the combination of topographic variables. The best
model, that have performance or better accuracy rate was constructed from latitude,
longitude and altitude. This best model is more consistent than IDW spatial model.
|
|---|