| dc.description.abstract | The changes in climate manifest as prolonged hotter and drier seasons, changes in the length
of seasons, and more frequent to severe droughts. In order to understand these changes, this
study was conducted with the main objective as examining the spatiotemporal patterns of
observed and projected rainfall and temperature. Lake Kyoga Basin was used as the case study.
The study analyzed rainfall and air temperature observations for the period 1990 to 2020 and
projections for the period 2021 to 2060. It analyzed the rainfall and sea surface temperatures
teleconnection signals over the period 1990 to 2020. Various methods including homogeneity
test, interpolation, Mann-Kendall trend test (MK-test), Standard Precipitation Index (SPI)1
,
derivation of climate indices, spatial and statistical correlation, and simple linear regression
were used.
The study found that the Lake Kyoga Basin has a bimodal rainfall regime with peaks in April
(average rainfall of 184.9mm) and October (average rainfall of 157.6mm). SPI showed that
Buginyanya and Namulonge had the highest wet periods while Kotido, Lira and Apac had the
highest incidences of dry periods. The MK-tests presented an insignificant reduction of the
March to May (MAM) and December to February (DJF) rainfall at 95% confidence interval
with z= -0.63, z=-0.50 and p-values of 0.53 and 0.62 respectively. The study further noted an
insignificant increase in rainfall during the June to August (JJA) and September to November
(SON) season, according to the positive z and the Sen’s slope values; although, the p-values
for JJA (0.09) and for SON (0.48) are not significant at 95% confidence interval.
The models, CNRM-CM6-1 and CanESM5 indicated a significant rise in both temperature and
rainfall across the Lake Kyoga Basin. Further analysis of Consecutive Dry Days (CDD) and
Consecutive Wet Days (CWD) using historical rainfall from 1990 to 2020 showed that Apac
had comparatively more CDD (p-value: 0.05) whereas Buginyanya had comparatively more
CWD (p-value: 0.02). Analysis of projected rainfall (2021 to 2061) revealed that Karamoja
region, mostly Kotido is projected to have more CDD reaching up to 76 days per year compared
to Kituuza, Jinja and Namulonge with about 65 CDD days per year. Tororo, and Buginyanya The MAM and JJA seasons are projected to have a reduction in the rainfall percentage changes
of CNRM_CM6_1 model for both SSP4342
(26.2% and 46.5%), SSP585 (12.1% and 12.1%)
respectively. The SON and DJF seasons exhibits a positive percentage change of (22.9% and
10.7%) for CanESM5 SSP434. The DJF season exhibits an increment of 10.7% and 31.6% for
both SSP434 in both models and a negative change of 65.4% and 36.6% for the SSP585 in both
models. The study further found a significant correlation for annual average rainfall (r = 0.44,
p-value 0.01) and SON seasonal rainfall (r=0.51, p-value 0.003) for the and over the Lake
Kyoga Basin with Western Indian Ocean Dipole (WIOD) at 95% confidence intervals.
Additional analysis using simple linear regression to examine the predictability of rainfall using
WIOD showed that annual average rainfall and SON regression models were considered basing
on the model goodness of fit R-square values. (r
2 = 0.35, and r=0.48). The Model output showed
that given the slope and X-intercept, Y-intercept (average SON, DJF rainfall) can be
successfully predicted. The study recommends that wetter areas like Buginyanya, Tororo,
Kituuza, Jinja, and Namulonge always to clear drainage channels, establish early warning
systems and pay attention to the timely weather and climate information disseminated via
different channels in order to initiate early action and reduce vulnerability. Drier areas like
Kotido, Apac, Soroti should advocate for water harvesting tools like tarpaulin, and irrigation
schemes. Practice tree planting and introduction of new resistant crop varieties to cope with
changing rainfall and temperatures. | en_US |
