Next Article in Journal
Significance of Standardizing Carbon Dioxide Measurement Time Within Lake Systems to Constrain CO2 Outgassing Estimation Uncertainties
Next Article in Special Issue
Making Different Decisions: Demonstrating the Influence of Climate Model Uncertainty on Adaptation Pathways
Previous Article in Journal
Evaluation of Layered Structures Impregnated with Fe or Cu as Catalysts in a Fenton-like Process for the Removal of 17α-Ethinylestradiol in Aqueous Solution: Operational Parameters and Ecotoxicity
Previous Article in Special Issue
Multiscale Factors Driving Extreme Flooding in China’s Pearl River Basin During the 2022 Dragon Boat Precipitation Season
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 7
10.3390/w17071044
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evaluation of Historical Dry and Wet Periods over Lake Kyoga Basin in Uganda

by
Hassen Babaousmail
1 and
Moses A. Ojara
2,3,*
1
School of Atmospheric Science and Remote Sensing, Wuxi University, Wuxi 214105, China
2
Division of Station Network and Observations, Department of Meteorological Services, Plot 21, 28 Port Bell Rd, Kampala P.O. Box 7025, Uganda
3
Green Life Research Initiative Uganda Limited, Namulonge-Nabalanga, Wakiso P.O. Box 1179, Uganda
*
Author to whom correspondence should be addressed.
Water 2025, 17(7), 1044; https://doi.org/10.3390/w17071044
Submission received: 28 February 2025 / Revised: 23 March 2025 / Accepted: 31 March 2025 / Published: 2 April 2025
(This article belongs to the Special Issue Climate Change and Hydrological Processes)
Browse Figures
Versions Notes

Abstract

:
Rainfall datasets from the Uganda National Meteorological Authority (UNMA) for 1981–2017 and two reanalysis datasets (Climate Hazards Group Infrared Precipitation with Stations data (CHIRPS) and Tropical Applications of Meteorology using Satellite data (TAMSAT) were used to compute drought and flood tendencies from 1981 to 2017. The cumulative departure index (CDI) and rainfall anomaly index (RAI) were computed to show drought and flood tendencies in the region. Meanwhile, dry days (DD) and wet days (WD) were computed based on the definition as a day of the season with rainfall amounts less than 1.0 mm and greater than 1.0 mm, respectively. The CDI graphics indicated below-average rainfall during 1981–1987 and relatively wetter conditions during 1989–1995 for all stations in the region. Generally, seasonal rainfall declined over the first 27 years but an increasing trend in both MAM (March–April–May) and SOND (September–October–November–December) was observed in most stations during 2006–2017. The highly variable seasonal rainfall in the region is expected to impact the livelihoods of the communities. This study recommends that the use of tailor-made weather and climate information for planning economic development programs such as agriculture will play a critical role in improving the livelihood of the communities in the region.

1. Introduction

African countries are expected to be the most affected by climate change and variability yet are the least prepared to deal with the effects [1,2,3,4,5,6]. Climate change is also emerging as one of the main challenges humankind will have to face for many years to come. It could become a major threat to world food security, as it has a strong impact on food production, access, and distribution [7]. Across East Africa (EA), rainfall is predicted to increase by 5–20% during the December to February rainy season and by 5–10% between the June and August rainy seasons [8]. In Uganda, it is stated that current average temperatures are expected to increase by between 0.7 °C and 1.5 °C by the 2020s [9]; however, this will be interrupted by the naturally variable climate by high spatial and temporal variability of rainfall largely explained by natural phenomena such as the El Niño and Southern Oscillation (ENSO) [10,11].
Climate variability and recent changes have exposed most parts of the East African region to poverty and hunger [12,13], starvation, and death, as seen recently in northeastern Uganda. This is exacerbated by the impact of unreliable weather patterns, usually resulting in worse food insecurity in the region [14]. Precipitation variability is expected to intensify the magnitude and frequency of flood and drought events that negatively affect agricultural sectors in most countries of EA [4,15], yet the increasing human population continues to exert pressure on food security, worsening the already fragile situation [16].
Climate variability impacts have the potential to undermine and even undo the progress made in improving the socio-economic well-being of countries, especially those that rely on natural weather systems for agriculture, like Uganda [17]. Furthermore, given an estimate that 80% of farmers produce food primarily to meet their family needs, rural poverty in Uganda could worsen with climate change, and due to their low income and lower technological and capital stocks, households are predicted to have limited options to adapt to climate change [6,18,19].
In the recent study to determine climate change trends in the Lake Kyoga Basin using standardized precipitation and anomaly indexes, significant changes in precipitation were revealed at some stations during the period 1981–2020, and the frequency of severe wet weather events was more than for dry weather events in many stations, further indicating that precipitation was increasing over the Lake Kyoga Basin [20].
Similarly, studies using long-term (1948–2016) changes in gridded (0.25° × 0.25°) Princeton Global Forcing (PGF) precipitation and potential evapotranspiration (PET) data over two sub-catchments in the Lake Kyoga Basin revealed negative trends (p < 0.05) in annual precipitation and positive anomalies in the early 2000s through 2016, which forms part of this current study period of 1981–2017. Negative anomalies existed between 1960 and 2000. Both were seasonal [18]. Additionally, analysis of extreme precipitation events over the Lake Kyoga Basin using the same gridded (0.25° × 0.25°) PGF precipitation dataset and period (1948–2016 from Princeton Global Forcing (PGF)) showed that the number of wet days decreased, although insignificantly (p > 0.05), and less frequent rains were received over the Lake Kyoga Basin, although some events had high intensity [17].
Using CenTrends gridded precipitation (1961–2015), the March to May (MAM) precipitation was shown to have insignificantly decreased (p > 0.05) in most parts of the Lake Kyoga Basin, while in the second season of September to November (SON), precipitation revealed a positive trend [17]. The variability of seasonal and annual precipitation over the Lake Kyoga Basin also showed an abrupt change in rainfall in the MAM season in 1982; the SON season did not show a significant abrupt change but a significant (p < 0.05) increase in rainfall above the upper limit from 1994 to date [19].
The impacts of highly uncertain rainfall patterns in the Lake Kyoga Basin provide major challenges to rural communities that consist of subsistence and nomadic farmers, as issues of water shortages for agriculture are common. This is usually due to drier conditions experienced in the areas with occasionally wet periods due to erratic rainfall. Economic losses have not been reliably quantified, but physical damages include crop failures due to droughts or floods, destruction to crops by hailstones during wet periods, livestock death due to lack of water or lightning strikes, and damages to physical infrastructures such as roads and buildings, among others. In order to minimize such losses and destruction reliably, location-specific weather and climate information need to be available.
There have been general studies that generated climate change and variability information in Uganda [17,19,20,21], but these cannot be used for adaptation at the local level. Further, recent trends in rainfall, weather, and climate information need to be frequently generated to aid in making both strategic and tactical decisions. There is a need for proper planning of agricultural activities around the Lake Kyoga Basin based on reliable weather and climate information to adapt to climate variability with its associated implications on food security, poverty, and health-related challenges.
The remaining sections of this study are structured as follows. Section 2 describes the study area. Section 3 is reserved for data and methods, while Section 4 presents the main results. Section 5 is for discussion. The conclusions and recommendations are highlighted in Section 6.

2. Study Area

The study area lies along geographical latitude and longitude of 0°13′ N, 3°41 N and longitude 32°02′ E, 34°52′ E. The Lake Kyoga Basin is among the largest basins in Uganda (Figure 1).
Generally, the Lake Kyoga Basin is a low laying area with an elevation ranging between 450 m and 867 m as per the Shuttle Radar Topography Mission digital model (SRTM) released in 2004. The climate of the basin is modified by the large swampy surrounding area. The basin has a rainy season from March to November, with a marked minimum in June and marked peaks in April to May and August to October, while December and January are the driest months.

3. Materials and Methods

3.1. Materials

3.1.1. Observed Station Climate Data

The in situ daily and monthly rainfall climate data were obtained from the Uganda National Meteorological Authority for Namulonge, Tororo, Soroti Jinja, Lira, Serere, Kiige, Buginyanya, Kotido, and Namalu stations. The duration for all datasets was taken in the range of 1981 to 2017 and collected from manual instruments. The stations are presented in Figure 1 and Table 1. The quality control measures for the observed rainfall dataset included checks control to identify the negative precipitation values, typing errors to identify gaps in the dataset, and false zeros [22]. All the rainfall data were processed through an extensive series of quality control procedures to ensure erroneous values, such as errors in manual keying. Negative daily precipitation amounts were removed and/or identified. Some stations, such as Namulonge, Tororo, Soroti, Jinja, and Lira, had daily rainfall datasets, which are required for the determination of critical phenomena like the onset and cessation of rainfall. The remaining stations had only monthly rainfall totals; these included Jinja, Nakasongola, Serere, Kiige, Buginyanya, Kotido, and Namalu. Thus, we are unable to determine the onset and cessation in these rainfall stations and zones. Almost none of the stations had complete temperature datasets (except for Soroti and Jinja, both located at the airfield).

3.1.2. Climate Hazards Group Infrared Precipitation with Stations Data (CHIRPS)

The gridded precipitation of the spatial resolution of 0.05° × 0.05° for 1981 to 2017 was sourced from the Climate Hazards Group Infrared Precipitation with Stations data (CHIRPS). CHIRPS is a product developed by merging three types of high-resolution data, including global climatology, satellite estimates, and in situ observations, employing a combination of interpolation techniques that improves rainfall products [23]. Funk et al. [23] provide an elaborated detail of CHIRPS data. Over EA, the lack of well-distributed and managed rain gauges is a challenge to the spatial analysis of climate data [24] that can only be overcome by the use of satellite or reanalysis datasets. Data can be accessed through the webpage https://data.chc.ucsb.edu/products/CHIRPS-2.0/global_daily/netcdf/p25/ (accessed on 20 October 2024).

3.1.3. TAMSAT Precipitation Data

The TAMSAT (Tropical Applications of Meteorology using Satellite data) precipitation dataset is widely used for monitoring and studying rainfall patterns in Africa. The data have a spatial resolution of 0.0375° × 0.0375° (approximately 4 km × 4 km) and temporal resolution of daily, 10-day, and monthly [25]. TAMSAT precipitation data are available from 1983 to the present and can be accessed through the links:
https://data.tamsat.org.uk/data-download/rainfall/ (accessed on 25 October 2024) and https://research.reading.ac.uk/tamsat/regridded-files/ (accessed on 25 October 2024).
TAMSAT precipitation data are derived from a combination of satellite and ground-based observations, such as geostationary satellite images from the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT), rain gauge data from ground-based rainfall measurements from national meteorological agencies and research institutions, and finally, Tropical Rainfall Measuring Mission (TRMM) satellite data (to 2015). The TAMSAT algorithm uses a combination of techniques to estimate rainfall from satellite data, including Cold Cloud Duration (CCD) and Rainfall Estimation Algorithm (REA) [25].
A Taylor diagram was used to evaluate these datasets, CHIRPS and TAMSAT, against the observed monthly rainfall datasets based on correlation (r), the RMSD, and the fraction of their variances [26].

3.2. Methods

3.2.1. Descriptive Statistics and Mann–Kendall Trend Analysis

For all the stations, monthly observed rainfall data were computed into annual rainfall and the two main rainfall seasons of March–May (MAM) and September–December (SOND) following the well-established seasons in the region [20]. Descriptive statistics were computed to mean, minimum, and maximum rainfall over the annual and seasonal basis in the region.
The trend analysis was performed using a rank-based non-parametric Mann–Kendall test (MK) [27,28]. The MK test was selected because it allows the existence of a trend in any data to confirm the null hypothesis of no trend. It does not require the sample to conform to any specific probability distribution since it works well even with insufficient or abnormal values.
The advantage of this technique is that it does not require any hypothesis about the variables and is more powerful than a parametric test. The Mann–Kendall (MK) test statistic was computed at a 5% significance level. In recent years, the MK statistical approach for trend analysis in climate studies has been employed [8,29]. Further discussion and the mathematical expression used to calculate MK trend statistics can be referred to in several studies [8,29].

3.2.2. Sen’s Slope Methods

To show the rate of change within the annual and seasonal rainfall, Sen’s slope test [30,31] was employed. The test gives a more robust estimation of the direction, especially when the trend cannot be estimated by other statistical approaches like Kendall’s test statistics or regression. This method is considered adequate due to its robust handling of outliers in the datasets. It is not affected by any extreme distributions and does not entail any normal distribution of the residuals. It has been generally employed in various studies to examine the linear tendencies of hydro/climatic variables across multiple domains, e.g., [8,32].
The Sen’s slope test is illustrated by Equations (1) and (2) below:
t i = Y j − K k j − k
where Y j and K k represent data values at the time, respectively. However, considering the expression of S in the Equation (2) below:
S = T M 2 + T M + 2 2 T M + 1 / 2 M   =   e v e n M   =   o d d
The median of these M values of S is a Sen’s estimator of the slope. If M is odd, then Sen’s estimator is computed by the top part of Equation (2), and if M is even, Sen’s estimator is calculated by the lower part of Equation (1). Finally, S is tested by a two-sided test at a 100% (1-α) confidence interval, and the actual slope is obtained.

3.2.3. Number of WET/DRY Days over MAM and SOND Rainfall Seasons

For the two well-known rainfall seasons over Uganda and EA (MAM and SOND), wet and dry days are generated by the common definition that a rainy day must experience at least 1.0 mm of rainfall, while a dry day is a day with less than 1.0 mm of rainfall [33,34]. When recorded over a concessive period, the two variables are referred to as wet and dry spells, respectively. Analyzing the number of wet and dry days is crucial for understanding rainfall variability, managing water resources, planning agricultural activities, studying climate change impacts, and maintaining ecosystem health.
The onset (start) and the cessation (end) of the rainy season were calculated based on the condition that a place must receive 20 mm of rainfall over 2 successive days and should not be followed by a dry spell of 9 days in the next 30 days at the time of the overhead appearance of the sun (March and September). The end of the season was described as being the first 9 days of a dry spell after May 1st and November 1st for the cessation of the first and second seasons, respectively. Consideration for the water balance failing to zero is observed as well. This helped to identify the season length or growing period and amount of rainfall each year in consideration.

3.2.4. Cumulative Departure Index (CDI) for 1981–2017

The MAM and SOND seasonal rainfall variability was further assessed by the cumulative departure index (CDI). The CDI involves the estimation of the arithmetic mean of seasonal rainfall for the period of record. The means of seasonal rainfall were then normalized by Equation (3) below:
C D I = r 1 − R 1 / S
where r 1 represent the actual seasonal rainfall of a given year. R 1 is the mean rainfall of the total length of the period and S the standard deviation of the total length of the period. Results of the values were cumulatively added to each other for the period of record and plotted to achieve long-term trends for seasonal as used [33,34].

3.2.5. Rainfall Anomaly Index (RAI) for 1981–2017

Further, the rainfall anomaly index (RAI) was analyzed to show the variability of annual rainfall (frequency and intensity of dry and wet years) and presented through graphical plots for visual interpretation. The RAI is suggested to be very effective in detecting the persistence of drought periods [33]. Rainfall variability indices were used to establish drought periods and to show some haphazard values for drought indication [33]. These simple indices, which use rainfall as the only input, perform comparatively well compared with more complicated indices in depicting periods and intensity of droughts [33,35]. RAI was used to describe annual rainfall variability and it is currently used by many authors [36].
The RAI is calculated from Equation (4) for positive anomalies and Equation (5) for negative anomalies.
R A I = + 3 R f − M R f M H 10 − M R f
R A I = − 3 R f − M R f M L 10 − M R f
where RAI represents the annual rainfall anomaly index, R f the actual rainfall for a given year, M R f the mean of the total length of the record, M H 10 the mean of the ten highest values of rainfall on record, and M L 10 the lowest values of rainfall on record.

4. Results

4.1. Relationship Between CHIRPS and TAMSAT and Observed Station Rainfall Data

Figure 2 shows Taylor diagrams for 4 main rainfall stations in the Lake Kyoga Basin, showing the relationship between the two reanalysis datasets (CHIRPS and TAMSAT) with observed datasets. Results showed a strong positive correlation coefficient between the observed station datasets and reanalysis ranging from +0.5 to 0.8 across the four stations, with TAMSAT being a better performer than CHIRPS. TAMSAT and CHIRPS are high-resolution reanalysis datasets that have been blended with station-observed datasets; their quality in the African region has been highly discussed [23]. This is represented by a strong positive correlation that provides confidence in using these datasets to present the ground-based observations in the region with scarcely populated observation stations.

4.2. Mean Monthly Annual Rainfall Cycles over Different Stations

Figure 3 presents the mean monthly annual rainfall for 9 stations in the Lake Kyoga Basin during the period 1981–2017. Results reveal unimodal rainfall at all the stations within the Lake Kyoga Basin but show a slight decline in June. This result is in agreement with previous studies on the region [37]. Several factors usually contribute to rainfall in the Lake Kyoga Basin, including the north/south movement of ITCZ and the Congo air mass, and rainfall is modified by topographic features [36,37,38].
The results showed high values of seasonal rainfall recorded during the months of March–April–May (MAM) and September–October–November–December (SOND), coinciding with two main modal rainfall regimes (MAM) and (SOND) reported by other studies over the regions [39,40]. In this study, these two main rainy periods, MAM and SOND, are considered the major seasons for subsequent use in the seasonal analysis. The main rainfall season from March to May (MAM) is usually referred to as the “long rains” and September to December as the “short rains” in EA [39,40,41].

4.3. Descriptive Statistics and Trends of Seasonal Annual Rainfall

A descriptive statistical analysis of seasonal annual rainfall is presented in Table 2. The results show that the highest annual rainfall of 3179.4 mm was recorded at the Buginyanya station in the Mount Elgon area, followed by 2015.7 mm recorded at the Namulonge station located near Lake Victoria. The annual rainfall over the region showed positive trends at all the stations, but statistically significant trends were observed at some stations, including the Namalu, Tororo, and Jinja stations. At these stations, the rate of change is 87.8, 8.6, and 50.4 (mm/decade), respectively.
Further, results also revealed that MAM seasonal rainfall had both decreasing and increasing trends, although both are statistically insignificant at all the stations. In terms of rainfall amount, MAM revealed between 236.5 mm at Namalu station and 773.9 mm at Tororo station. SOND seasonal rainfall over the region ranged from 184.7 to 707.4 mm at Namalu and Tororo stations. Overall, there was an increasing trend in the SOND seasonal rainfall amount, with statistically significant trends at some stations, including the Namalu, Buginyanya, and Tororo stations. For SOND, Sen’s lope analysis showed that the rate of change for these stations is 5.9, 4.9, and 68.7 mm/decade at Namalu, Buginyanya, and Tororo stations, respectively.
The annual and seasonal rainfall revealed that there is a high spatial variation of rainfall within the region as some stations reported an amount about thrice the other stations within the regions, and others showed statistically significant trends compared to the others. The high spatial variability of rainfall in the region has been reported [17,37,41]. This high spatial variability, therefore, calls for location-specific approaches in managing the impacts of rainfall and climate events in the region.

4.4. Onset and Cessation of Rainfall for Lake Kyoga Basin

Table 3 shows the summary statistics for onset, cessation date, and seasonal length for 5 major rainfall stations in the Lake Kyoga Basin. For Soroti station, the analysis results showed that over the recent years 1981–2017, the mean start of rains, or season onset, is 15 March of the year, the mean cessation date is 27 June, and the mean season’s length is approximately 3.5 months. The earliest start of the rains could be as early as 29 February. The latest possible date for the start of the rains is as late as 15 April.
Similarly, at Namulonge station, results revealed that 19 March is the mean date for the onset of seasonal rainfall, with 22 June as the mean cessation date. It was revealed that the earliest possible date is 3 March, while the latest onset of the first rainfall period could be as late as 14 May. In regard to the cessation date, the mean cessation date is 7 July, while the mean seasonal length is 3.3 months.
For Tororo station, results revealed that the mean onset of the rainy season is on 19 March, and the mean cessation date is 7 July, while the mean seasonal length is about 3.6 months. It was also found that the first season can start as early as 1 March and end by 1 June. Additionally, the second season presented a mean onset date of 26 August and a cessation date of 7 December, with about 3.3 months as the mean seasonal length. Conversely, the second season revealed 1 Aug as the earliest onset date and 17 November as the earliest cessation date; this also indicated 1.5 months as the shortest seasonal length that could be experienced. The second rainfall season could also start on 7 October and cease as late as 31 December, presenting the longest seasonal length of 4.7 months.
For the Lira station, the mean onset and cessation of the first rainy seasons are revealed as 29 March and 6 August, respectively, with the mean seasonal length being approximately 2.9 months. Similarly, the mean onset and cessation dates for the second rainy season are approximated to be 14 August and 27 November, with a mean seasonal length of 3.4 months revealed for this station. At Lira station, the first rainy season of MAM could start as early as 2 March and end as early as 3 June, showing the shortest season of approximately 2 months. Moreover, the second rainfall season could also start (onset) by 1 August and end by 1 November, with the shortest seasonal length of 3 months. For the second season, the latest onset and cessation dates are 9 October and 14 December, respectively. This also shows the longest seasonal length, 4.5 months. In addition to these, the rainfall station showed 9 October and 30 December as the latest onset and cessation of the second rainy season, respectively. The longest seasonal length in this rainfall zone is approximately 3.6 months.
Finally, at Buginyanya station, results revealed that 13 March is the mean date for the onset of seasonal rainfall, with 13 July as the mean cessation date. It was revealed that the earliest possible date is 29 February, while the latest onset of the first rainfall period could be as late as 26 April. In regard to the cessation date, the mean cessation date is 13 July, while the mean seasonal length is 3.3 months. For the second season, the mean onset date is 4 September, and the cessation date is 24 November, with a seasonal length of 2.7 months. The earliest possible start of the rains is 19 August, and the cessation date is 20 November, with a seasonal length of 3.1 months. The latest possible start of the second season is 4 November, and the cessation date is 24 December, with a seasonal length of 3.3 months.

4.5. Temporal Variation in Annual and Seasonal Rainfall in Lake Kyoga Basin

Figure 4 shows the temporal variation in the annual rainfall over the Lake Kyoga Basin during 1981–2017. Results indicate a decline in the annual rainfall during the first period of the study, 1981–1985, for most stations, including Kiige, Namulonge, Buginyanya, Serere, Lira, and Namalu; this coincides with the reported drought episodes of 1983–1985 over EA which caused severe impacts on socio-economic activities in the region [40].
The mean annual rainfall in the Lake Kyoga Basin gradually increased from 1985 to 2005 except for the two adjacent stations, Soroti (e) and Serere (g), which indicated a declining trend over the same period. Conversely, after 2005, most rainfall stations in the Lake Kyoga Basin revealed a drop in annual precipitation. The declining rainfall in Uganda was reported in a recent analysis [42].
Contrarily, parts of the Lake Kyoga Basin experienced a lot of rainfall over the years, which caused mudslides on several occasions, such as in 2015, 2018, and 2019, to the extent of causing catastrophic loss of human lives and destruction of property and infrastructure [20].

4.6. Number of Wet/Dry Days During MAM and SOND Rainfall Seasons

The Lake Kyoga Basin received an average of 30–70 wet days during MAM rainfall seasons over most parts of the region, with a few areas receiving less than 20 wet days during MAM (Figure 5a,b). The average number of dry days from two gridded satellite datasets, CHIRPS 2.0 and TAMSAT v3.1, are provided in Figure 5c,d. The highest number of dry days (80) occurred in the northeastern part of the Lake Kyoga Basin, while the lowest number of dry days were experienced in the southern part of the basin, bordering the Lake Victoria region, which usually receives the highest rainfall amount in Uganda. As observed from Figure 5, the number of dry days during MAM rainfall seasons outstripped their corresponding wet days during the same rainfall seasons.
For the second rainfall season, during SOND, a similar pattern and range for the number of wet and dry days are observed (Figure 6). For instance, the southern and western parts of the Lake Kyoga Basin experienced more wet days compared to the northeastern part of the basin. In this case, the number of wet days ranged from 40 to 70 days. For the northeastern parts of the Lake Kyoga Basin, the number of wet days ranged from 20 to 50 days (Figure 6a,b).
Concerning the dry days, the southern and western parts of the Lake Kyoga Basin experienced few wet days compared to the northeastern part of the basin, which ranged from 20 to 40 days mainly (Figure 6c). In both reanalysis datasets, the northeastern part of the Lake Kyoga Basin revealed the highest number of dry days, ranging from 50 to 80 days.
Northeastern Uganda is part of Uganda (known as the Karamoja region) and receives the lowest number of wet days, which is translated to the lowest amount of rainfall received in Uganda. The region is typically semi-arid in nature and usually experiences harsh weather conditions with high temperatures and poor rainfall distribution associated with very short rainy periods, which causes flash floods [37,41,43]. Poor rainfall distribution in the northeastern region frequently results in droughts, crop failures, death of livestock, famine, loss of livelihood, and even human death [43].

4.7. Cumulative Departure Index (CDI) over Lake Kyoga Basin

Figure 7a–g shows a graphical representation of the seasonal CDI for selected stations in the Lake Kyoga Basin over the period 1981–2017. The variability of seasonal rainfall is shown by an upward and downward movement of the CDI graphs, which correspond to surplus and deficit rainfall, respectively.
For Namulonge station (Figure 6a), the “long rains” (MAM) season is observed to show below-average rainfall for most of the years during the period 1983–2017, with only 3 years showing positive rainfall, while SOND showed either near normal rainfall or below average. The years that have anomalously high MAM seasonal rainfall include 1982, 1997–1998, and 2016. The years that showed anomalously low MAM rainfall at this station include 1996, 2000, and 2007–2008. Meanwhile, analysis of SOND seasonal rainfall revealed anomalously high and low rainfall in 1995 and 1982–1983, 1986, 1989, and 2016, respectively.
At Tororo station, the variability in seasonal rainfall is also evidenced as the CDI graphs presented both downward and upward movement in both seasons (Figure 6b), but both seasons recorded near normal long-term mean rainfall. The CDI curve for MAM indicates extremely high rainfall in 2013, 2012, and 2006 in the last 10 years and 1990–1991 in the previous decade. Contrarily, 2004 and 2016 are two years that present below-average rainfall for MAM seasons. There is no clear pattern for the SOND rainfall season as both downward and upward trends are observed randomly, showing high variability in seasonal rainfall. The SOND season revealed that the highest upward trends in rainfall were recorded in 1997, 2006, and 2014, while 1986, 1990, and 2010 showed highly marked downward movement of the cumulative departure index.
As in the first two stations above, the seasonal variability in rainfall at Soroti station (Figure 7c) is visibly clear, as the MAM season records below-normal rainfall in the first 2 years. An improvement in MAM rainfall is attained during the period 1985–1998, as indicated by the upward movement of CDI graphs. This is subsequently followed by a reduction in MAM rainfall in the next 4 years, as the CDI graphs move into the negative phase from 1999 to 2002. From 2003 to 2017, 5 years showed positive CDI values and upward trends in rainfall, while the remaining years indicate below-normal rainfall. The station showed extremely low rainfall during MAM in 1999 and over 2009–2010, while extremely high rainfall was experienced in 1991–1992 and 1996. The results revealed negative CDI graphs for SOND in 22 years during the period 1981–2005 and only 3 years of positive CDI graphs. Toward the last decade (2006–2017), SOND seasonal rainfall expressed improvement, as all the years registered positive CDI values and an upward movement of CDI graphs.
At Lira station (Figure 7d), both the MAM and SOND seasons are observed to show below-average rainfall for most of the years; however, a few years recorded extremely high rainfall (above average) during the MAM rainy season, including 1982, 1997–1998, 2015–2016, and extremely below-average rainfall during 1983–1984, 1999–2000, 2007–2008, 2010, 2012, and 2014. The worst-case scenario is seen during SOND, where only 3 years show above-average rainfall, namely the years 1994–1995 and 2002. Extremely dry conditions and below-normal rainfall are observed in all years, most notably in 1982–1983,1986, 1989, 2004–2006, 2009–2010, and 2015.
Further, Jinja station (Figure 7e) showed clear seasonal variability in rainfall from 1981 to 1993; most years displayed below-average rainfall in both the MAM and SOND seasonal rainfall except in a few cases, such as in 1987–1988 and previously in 1982–1983 with above-average seasonal rainfall. From 1995 to 2002, MAM received two years of extremely high rainfall (1996 and 1999), as SOND recorded above-average rainfall in six years. Similarly, the years 2003–2011 showed below-normal rainfall in both seasons except on a few occasions, such as in 2003 and 2006. The last 6 years (2011–2017) showed more upward movement in SOND seasonal and downward movement in MAM rainfall. The MAM season received substantially high rainfall during 1995–1997 and 2000, and a drastic reduction in MAM seasonal rainfall in 1985–1986, 1990–1991, 2005, and 2013. The CDI bars showed a negative trend in rainfall in the second season of SOND from 1981–1987 and 1991–1995. From 1996 onward, SOND rainfall indicated a positive trend punctuated with years of reduced rainfall, most predominantly in 2006 and 2008–2009.
Buginyanya station (Figure 7f), in the montane region, shows the downward movement of MAM seasonal rainfall from 1981 to 1996, with the exception of 1984 and 1994 with near-normal rainfall. There is no difference in the observed trends for SOND seasonal rainfall at Buginyanya station, as most years (1981–1999) show decreased rainfall (below-average rainfall) except on a few occasions in 1985–1987 and 1994–1995. Proceeding from 2000–2002, a positive trend indicated by upward bars in rainfall is recorded, followed by below-average rainfall over the period 2003–2004, which is followed by near-normal rainfall from 2005 to 2011. CDI graphs showed that the last 6 years registered the wettest years for SOND rainfall, including 2001–2002 and 2013–2014, while 1982–1984,1988–1993, and 2007 were shown to be the driest years in terms of SOND rainfall for this period of analysis.
At Namalu station (Figure 7g), the years 1983–1987, 1992–1995, 1999–2005, 2008–2009, 2014–2015, and 2017 were significantly dry (below average) in terms of MAM seasonal rainfall. The dry MAM seasons were punctuated with occasionally wet seasons (above-average rainfall), such as during the periods 2006–2007, 1996–1999, and 2012–2013. The second season (SOND) showed below-average rainfall over the period 1981–1998 and above-average rainfall in the following 4 years (1999–2002), followed by a further 3 years of below-average rainfall (2003–2005). Over the last 12 years (2006–2017), SOND received improved rainfall performance, with the only drop in 2009. Extremely high and low rainfall during SOND was experienced in 2012 and 1986, respectively.

4.8. RAI over Lake Kyoga Basin for 1981–2017

The results of the annual rainfall anomaly index (RAI) are presented in Figure 8. The results show that annual rainfall varied a lot in the long-term mean, as different stations revealed different ranges of RAI values throughout the analyzed years. For example, Namulonge station recorded an extreme range of −5.4 to 6.3 in 1983 and 1997, respectively. For the case of Tororo station, the two extreme range values of −5.2 to 7.5 were reported in more recent years, 2001 and 2004. The range at Soroti is from −3.9 to 5.6, which occurred in the earlier years of 1984 and 1988. Meanwhile, Serere station recorded a range of −4.8 to 6.7 in 2012 and 2000, respectively. The negative indices represent dry years, and the positive indices show wet or rainy years with varying intensities in each case.
For the first 20 years (1981–2000), 12 cases of extremely dry conditions were revealed within the Lake Kyoga Basin. In such circumstances, Soroti station reported 3 cases of extremely dry years in 1982, 1984, and 1989, while Namulonge station also reported 3 cases in 1983, 1993, and 2000. Meanwhile, the remaining stations showed one case of an extremely dry year. RAI values indicate that 1981 and 1999 were the wettest years when all the stations reported extremely wet conditions (positive RAI). Moreover, during the period (2000–2017), 34 cases of extremely dry years were experienced in 9 locations examined in the Lake Kyoga Basin. The distribution of extremely dry conditions was uniform at some stations. For example, Soroti, Serere, and Buginyanya stations each recorded 5 years of extremely dry conditions, while Jinja registered 4 years of extremely dry years. The remaining stations recorded at least 3 years of extremely dry years.

5. Discussion

Results from the cumulative departure index show several years of below-average rainfall, showing drying or drought tendencies over the region during the first rainy period of March–May (MAM). The March–May long rains are critical for the East Africa region and have been quite poor in recent years [24]. The decline in MAM rainfall in the EA region has been prominently analyzed and discussed [44]. The last 10 years of the study showed rainfall increasing in both the MAM and SOND seasons, contributing to an increase in annual rainfall over most stations in the basin. However, the increase in rainfall over the basin is punctuated with extreme or severe drought years, as witnessed in results in 2009–2010, 2014, 2016, and 2017, further indicating a drought condition in the region. Ongoma et al. [8] also showed a positive trend in March–April–May (MAM) and October–November–December (OND) seasonal rainfall over the East Africa region. The months of November–December have seen an increase, probably reflecting increased November rainfall in East Africa [45].
These results showed high positive RAI values for Namulonge station in the same region in the years 1989,1997, and 2001, indicating that they were extremely wet, which is in agreement with recent studies that analyzed rainfall variability for the Lake Victoria Basin adjacent to the Lake Kyoga Basin [45] and found that the years 1989, 1997, and 2001 had anomalously high total annual rainfall in the region. The most notable extremely wet years are 1981 for most stations except Namulonge station and 1999, 2011, and 2015 for most stations. However, using rainfall data for 1949–2009, ref. [46] found that rainfall at the Namulonge station decreased.
The main controlling mechanism of these extremely high rainfall events has been linked to the reversal of dipole in atmospheric circulation and Indian Ocean sea surface temperatures [47]. The rains are moderated by weather phenomena such as the El Niño Southern Oscillation (ENSO) [38] and the Indian Ocean Dipole (IOD) [48]. The El Niño Southern Oscillation (ENSO) phenomena are strongly associated with the inter-annual variability of rainfall in this region [38]. This agrees with our present study, as years with El Niño events such as 1981, 1997–1998, 2006–2007, and 2015 were recorded as wet years.
The result showed high spatial variability, as stations indicate different results when similar weather phenomena occur. For example, in 1996 and 2015, most stations showed extremely high rainfall, but a few stations still showed extremely dry conditions over the same period. Such differences in weather conditions are closely related to the contribution of the local circulation systems resulting from land surface heterogeneity induced by biophysical factors such as vegetation, large open water bodies, and topography [49,50]. Drought conditions exhibited in some years in most stations severely affect agricultural, water, economic, and health sectors and human death has been reported in the Lake Kyoga Basin [43].
In the East African tropical climate, rainfall remains one of the most important drivers of many socio-economic activities like agriculture and electricity generation [51], but variability in rainfall imposes greater risks to such socio-economic sectors. Extreme drought and famine occurred in other parts of Africa at the beginning of 2006, spreading to the East African region, especially the northern part of Uganda, Kenya, and Somalia, more severe than those of 1984, 1999, and 2000 in the East African region [44,46].
Globally, an estimated eleven million people have died of drought-related causes, while two billion people have been affected by drought since 1900 [52]. The frequencies and intensity of drought are predicted to increase [2], which confirms the results that the current years (2001–2017) had the highest frequencies and more intense droughts than the period 1981–2000.
Generally, 2010 was extremely dry for Soroti, Jinja, and Buginyanya and severely dry for most stations, including Tororo and Namalu, possibly caused by the failure of the “short rains” of 2010 linked to the La Nina condition. This was followed by a further decline in the “long rains” in MAM in 2011 in EA [53]. The year 2011 was one of the wettest years in the Lake Kyoga Basin, as most of the RAI values were greater than 0.3. The results also showed that in the period 1997–1998, some stations, such as Namulonge, Tororo, Jinja, and Kiige, had positive RAI, representing extremely wet years. This could be associated with the strong El Niño episode of that year, which is referred to as the “climate event” of the century [54].
Similarly, the extreme El Niño events in 1982/1983 could have boosted the rainfall of other stations in the region; however, extremely dry conditions were recorded in stations such as Namulonge, Tororo, and Soroti. This also shows the variability of rainfall in the region occurring within short distances [55]. The recent wet year (2015) is possibly another effect of El Niño events experienced in the region. This increases the chances of most locations recording extremely wet conditions [56], but this is usually followed by the opposite phase, the La Nina event, which is usually dry and delivers less rainfall in some areas [57], consistent with the results of 2016 in this study. However, this is not exclusively true, as some patches of dry conditions persisted at some stations within the basin in the same year (2016).
According to these results, the region registered five incidences of extremely dry years and seven cases of severely dry years at different stations. The results agree with seven drought cases reported between 1991 and 2000 [58]. The Greater Horn of Africa has been affected by drought almost every year for the past two decades, and the most severe drought occurred in 2009 and 2011 in Kenya [51], which is true to the test results for RAI in 2009.

6. Conclusions

This study analyzed the rainfall data for the Lake Kyoga Basin using the CDI and RAI. The results show a declining trend in March–May (MAM) rainfall, indicating drying or drought tendencies over the region. However, the last decade of the study period showed an increase in rainfall in both MAM and October–December (SOND) seasons. The study also identified extreme wet and dry years, with El Niño events contributing to the wet years. The results highlight the high spatial variability in rainfall in the region, with local circulation systems playing a significant role. The study’s findings have important implications for agricultural, water, economic, and health sectors in the region, which are vulnerable to rainfall variability and droughts. Overall, this study contributes to the understanding of rainfall variability in the Lake Kyoga Basin and highlights the need for effective climate risk management strategies to mitigate the impacts of droughts and floods in the region.
The study’s findings highlight the need for effective climate risk management strategies to mitigate the impacts of droughts and floods in the Lake Kyoga Basin, which can be achieved through the development of early warning systems, climate-resilient agriculture practices, and water harvesting and storage infrastructure. Finally, the study’s findings emphasize the importance of understanding rainfall variability and its impacts on socio-economic sectors. To improve climate monitoring and prediction capabilities, investing in climate observation networks, enhancing climate modeling and forecasting capabilities, and promoting climate information dissemination and applications are essential. This will enable decision-makers to make informed decisions and develop effective adaptation strategies.

Author Contributions

Conceptualization, M.A.O. and H.B.; methodology, M.A.O.; software, M.A.O. and H.B.; validation, M.A.O.; formal analysis, M.A.O. and H.B.; investigation, M.O; resources, M.A.O. and H.B.; data curation, M.A.O. and H.B.; writing—original draft preparation, M.A.O.; writing—review and editing, M.A.O.; visualization, M.A.O. and H.B.; supervision, M.O; project administration, M.A.O.; funding acquisition, M.A.O. and H.B.; All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data are available upon request.

Conflicts of Interest

Author Moses A. Ojara was employed by the company Green Life Research Initiative Uganda Limited. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Min, S.; Zhang, X.; Zwiers, F.W.; Hegerl, G.C. Human Contribution to More-Intense Precipitation Extremes. Nature 2011, 470, 378–381. [Google Scholar] [CrossRef] [PubMed]
  2. IPCC. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation; IPCC: Geneva, Switzerland, 2012. [Google Scholar]
  3. IPCC. Climate Change 2014: Synthesis Report; Pachauri, R.K., Ed.; IPCC: Geneva, Switzerland, 2014. [Google Scholar]
  4. Adhikari, U.; Nejadhashemi, A.P.; Woznicki, S.A. Climate Change and Eastern Africa: A Review of Impact on Major Crops. Food Energy Secur. 2015, 4, 110–132. [Google Scholar] [CrossRef]
  5. Almazroui, M.; Saeed, F.; Saeed, S.; Islam, M.N.; Ismail, M. Projected Change in Temperature and Precipitation Over Africa from CMIP6. Earth Syst. Environ. 2020, 4, 455–475. [Google Scholar] [CrossRef]
  6. Nuwagira, U.; Yasin, I. Review of the Past, Current, and the Future Trend of the Climate Change and Its Impact in Uganda. East Afr. J. Environ. Nat. Resour. 2022, 5, 115–126. [Google Scholar] [CrossRef]
  7. Lobell, D.B.; Burke, M.B.; Tebaldi, C.; Mastrandrea, M.D.; Falcon, W.P.; Naylor, R.L. Prioritizing Climate Change Adaptation Needs for Food Security in 2030. Science 2008, 319, 607–610. [Google Scholar]
  8. Ongoma, V.; Chen, H.; Omony, G.W. Variability of Extreme Weather Events over the Equatorial East Africa, a Case Study of Rainfall in Kenya and Uganda. Theor. Appl. Climatol. 2018, 131, 295–308. [Google Scholar] [CrossRef]
  9. Government of Uganda; UNFPA. The State of Uganda Population Report 2009: Addressing the Effects of Climate Change on Migration; United Nations Population Fund: New York, NY, USA, 2009.
  10. MacLeod, D.; Caminade, C. The Moderate Impact of the 2015 El Niño over East Africa and Its Representation in Seasonal Reforecasts. J. Clim. 2019, 32, 7989–8001. [Google Scholar] [CrossRef]
  11. Palmer, P.I.; Wainwright, C.M.; Dong, B.; Maidment, R.I.; Wheeler, K.G.; Gedney, N.; Hickman, J.E.; Madani, N.; Folwell, S.S.; Abdo, G.; et al. Drivers and Impacts of Eastern African Rainfall Variability. Nat. Rev. Earth Environ. 2023, 4, 254–270. [Google Scholar] [CrossRef]
  12. Thornton, P.K.; Owiyo, T.; Jones, P.G.; Owiyo, T.; Kruska, R.L.; Herrero, M.; Orindi, V.; Bhadwal, S.; Kristjanson, P.; Notenbaert, A.; et al. Climate Change and Poverty in Africa: Mapping Hotspots of Vulnerability. Afr. J. Agric. Resour. Econ. 2008, 2, 24–44. [Google Scholar] [CrossRef]
  13. Shikuku, K.M.; Winowiecki, L.; Twyman, J.; Eitzinger, A.; Perez, J.G.; Mwongera, C.; Läderach, P. Climate Risk Management Smallholder Farmers’ Attitudes and Determinants of Adaptation to Climate Risks in East Africa. Clim. Risk Manag. 2017, 16, 234–245. [Google Scholar] [CrossRef]
  14. Schlenker, W.; Lobell, D.B. Robust Negative Impacts of Climate Change. Environ. Res. Lett. 2010, 5, 014010. [Google Scholar] [CrossRef]
  15. Pachauri, R.K.; Reisinger, A. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; IPCC: Geneva, Switzerland, 2007. [Google Scholar]
  16. African Union. State of Africa’s Population 2017: Keeping Rights of Girls, Adolescents and Young Women at the Centre of Africa’s Demographic Dividend; African Union: Ababa, Ethiopia, 2017. [Google Scholar]
  17. Onyutha, C.; Acayo, G.; Nyende, J. Analyses of Precipitation and Evapotranspiration Changes across the Lake Kyoga Basin in East Africa. Water 2020, 12, 1134. [Google Scholar] [CrossRef]
  18. Mubialiwo, A.; Chelangat, C.; Onyutha, C. Changes in Precipitation and Evapotranspiration over Lokok and Lokere Catchments in Uganda. Bull. Atmos. Sci. Technol. 2021, 2, 2. [Google Scholar] [CrossRef]
  19. Atube, F.; Malinga, G.M.; Nyeko, M.; Okello, D.M.; Mugonola, B.; Omony, G.W.; Uma, I.O. Farmers’ Perceptions of Climate Change, Long-Term Variability and Trends in Rainfall in Apac District, Northern Uganda. CABI Agric. Biosci. 2022, 4, 1–16. [Google Scholar] [CrossRef]
  20. Obubu, J.P.; Mengistou, S.; Fetahi, T.; Alamirew, T.; Odong, R.; Ekwacu, S. Recent Climate Change in the Lake Kyoga Basin, Uganda: An Analysis Using Short-Term and Long-Term Data with Standardized Precipitation and Anomaly Indexes. Climate 2021, 9, 179. [Google Scholar] [CrossRef]
  21. Onyutha, C. Trends and Variability of Temperature and Evaporation over the African Continent: Relationships with Precipitation. Atmosfera 2021, 34, 267–287. [Google Scholar]
  22. Filahi, S.; Tanarhte, M.; Mouhir, L.; El Morhit, M.; Tramblay, Y. Trends in Indices of Daily Temperature and Precipitations Extremes in Morocco. Theor. Appl. Climatol. 2016, 124, 959–972. [Google Scholar] [CrossRef]
  23. Funk, C.; Peterson, P.; Landsfeld, M.; Pedreros, D.; Verdin, J.; Shukla, S.; Husak, G.; Rowland, J.; Harrison, L.; Hoell, A.; et al. The Climate Hazards Infrared Precipitation with Stations—A New Environmental Record for Monitoring Extremes. Sci. Data 2015, 2, 1–21. [Google Scholar] [CrossRef]
  24. Wainwright, C.M.; Marsham, J.H.; Keane, R.J.; Rowell, D.P.; Finney, D.L.; Black, E.; Allan, R.P. ‘Eastern African Paradox’ Rainfall Decline Due to Shorter Not Less Intense Long Rains. Npj Clim. Atmos. Sci. 2019, 2, 1–9. [Google Scholar] [CrossRef]
  25. Maidment, R.I.; Grimes, D.; Black, E.; Tarnavsky, E.; Young, M.; Greatrex, H.; Allan, R.P.; Stein, T.; Nkonde, E.; Senkunda, S.; et al. Data Descriptor: A New, Long-Term Daily Satellite-Based Rainfall Dataset for Operational Monitoring in Africa. Sci. Data 2017, 4, 170063. [Google Scholar] [CrossRef]
  26. Taylor, K.E. Taylor Diagram Primer. 2005. Available online: https://pcmdi.llnl.gov/staff/taylor/CV/Taylor_diagram_primer.pdf (accessed on 20 January 2025).
  27. Mann, H.B. Non-Parametric Test Against Trend. Econometrica 1945, 13, 245–259. [Google Scholar]
  28. Kendall, M.G. Appendix: Mann-Kendall Trend Tests; Oxford University Press: Oxford, UK, 1975; p. 202. [Google Scholar]
  29. Ayugi, B.O.; Tan, G. Recent Trends of Surface Air Temperatures over Kenya from 1971 to 2010. Meteorol. Atmos. Phys. 2019, 131, 1401–1413. [Google Scholar] [CrossRef]
  30. Theil, H. A Rank-Invariant Method of Linear and Polynomial Regression Analysis. In Henri Theil’s Contributions to Economics and Econometrics; Springer: Dordrecht, The Netherlands, 1950; pp. 345–381. [Google Scholar] [CrossRef]
  31. Sen, P.K. Estimates of the Regression Coefficient Based on Kendall’s Tau. Am. Stat. Assoc. 1968, 63, 1379–1389. [Google Scholar] [CrossRef]
  32. Mumo, L.; Yu, J.; Ayugi, B. Evaluation of Spatiotemporal Variability of Rainfall over Kenya from 1979 to 2017. J. Atmos. Sol. Terr. Phys. 2019, 194, 105097. [Google Scholar] [CrossRef]
  33. Caloiero, T.; Coscarelli, R. Analysis of the Characteristics of Dry and Wet Spells in a Mediterranean Region. Environ. Process. 2020, 7, 691–701. [Google Scholar]
  34. Kendon, E.J.; Stratton, R.A.; Tucker, S.; Marsham, J.H.; Berthou, S.; Rowell, D.P.; Senior, C.A. Enhanced Future Changes in Wet and Dry Extremes over Africa at Convection-Permitting Scale. Nat. Commun. 2019, 10, 1794. [Google Scholar] [CrossRef]
  35. Tilahun, K. Analysis of Rainfall Climate and Evapo-Transpiration in Arid and Semi-Arid Regions of Ethiopia Using Data over the Last Half a Century. J. Arid Environ. 2006, 64, 474–487. [Google Scholar] [CrossRef]
  36. Hänsel, S.; Schucknecht, A.; Matschullat, J. The Modified Rainfall Anomaly Index (MRAI)—Is This an Alternative to the Standardised Precipitation Index (SPI) in Evaluating Future Extreme Precipitation Characteristics? Theor. Appl. Climatol. 2015, 123, 827–844. [Google Scholar] [CrossRef]
  37. Kansiime, M.K.; Wambugu, S.K.; Shisanya, C.A. Perceived and Actual Rainfall Trends and Variability in Eastern Uganda: Implications for Community Preparedness and Response. J. Nat. Sci. Res. 2013, 3, 179–195. [Google Scholar]
  38. Indeje, M.; Semazzi, F.H.M.; Ogallo, L.J. ENSO Signals in East African Rainfall Seasons. Int. J. Climatol. 2000, 46, 19–46. [Google Scholar] [CrossRef]
  39. Brown, E.; Sutcliffe, J.V.; Brown, E.; John, V. The Water Balance of Lake Kyoga, Uganda the Water Balance of Lake Kyoga, Uganda. Hydrol. Sci. J. 2013, 58, 341–353. [Google Scholar] [CrossRef]
  40. Ogwang, B.; Nimusiima, A.; Tindamanyire, T.; Serwanga, M.; Ayesiga, G.; Ojara, M.; Ssebabi, F.; Gugwa, G.; Nsubuga, Y.; Atim, R.; et al. Characteristics and Changes in SON Rainfall Over. J. Environ. Agric. Sci. 2016, 8, 45–53. [Google Scholar]
  41. Mubiru, D.N.; Komutunga, E.; Agona, A.; Apok, A.; Ngara, T. Characterising Agrometeorological Climate Risks and Uncertainties: Crop Production in Uganda. S. Afr. J. Sci. 2012, 108, 108–118. [Google Scholar] [CrossRef]
  42. Nsubuga, F.W.; Rautenbach, H. Climate Change and Variability: A Review of What Is Known and Ought to Be Known For. Int. J. Clim. Change Strateg. Manag. 2018, 10, 752–771. [Google Scholar] [CrossRef]
  43. NAPA. Climate Change: Uganda National Adaptation Programmes of Action (NAPA); Ministry of Water and Environment (MWE): Kampala, Uganda, 2007.
  44. Rowell, D.P.; Booth, B.B.; Nicholson, S.E.; Good, P. Reconciling Past and Future Rainfall Trends over East Africa. J. Clim. 2015, 28, 9768–9788. [Google Scholar] [CrossRef]
  45. Clive, A.S. The Changing Climate of Africa Part I: Introduction and Eastern Africa. In African Ecology—Benchmarks and Historical Perspectives; Springer: Berlin/Heidelberg, Germany, 2012; pp. 57–140. ISBN 9783642228711. [Google Scholar]
  46. Kizza, M.; Rodhe, A.; Xu, C.-Y.; Ntale, H.K.; Halldin, S. Temporal Rainfall Variability in the Lake Victoria Basin in East Africa during the Twentieth Century. Theor. Appl. Climatol. 2009, 98, 119–135. [Google Scholar] [CrossRef]
  47. Nsubuga, F.W.N.; Olwoch, J.M.; Rautenbach, C.J.D.W. Climatic Trends at Namulonge in Uganda: 1947–2009. J. Geogr. Geol. 2011, 3, 119–131. [Google Scholar] [CrossRef]
  48. Saji, N.H.; Vinayachandran, P.N. A Dipole Mode in the Tropical Indian Ocean. Nature 1999, 401, 360–363. [Google Scholar]
  49. Behera, S.K.; Luo, J.J.; Masson, S.; Delecluse, P.; Gualdi, S.; Navarra, A.; Yamagata, T. Paramount Impact of the Indian Ocean Dipole on the East African Short Rains: A CGCM Study. J. Clim. 2005, 18, 4514–4530. [Google Scholar] [CrossRef]
  50. Ogwang, B.A.; Chen, H.; Li, X.; Gao, C. The Influence of Topography on East African October to December Climate: Sensitivity Experiments with RegCM4. Adv. Meteorol. 2014, 2014, 14. [Google Scholar] [CrossRef]
  51. Ayugi, B.; Tan, G.; Rouyun, N.; Zeyao, D.; Ojara, M.; Mumo, L.; Babaousmail, H.; Ongoma, V. Evaluation of Meteorological Drought and Flood Scenarios over Kenya, East Africa. Water 2020, 11, 307. [Google Scholar] [CrossRef]
  52. Ghaleb, F.; Mario, M.; Sandra, A. Regional Landsat-Based Drought Monitoring from 1982 to 2014. Climate 2015, 3, 563–577. [Google Scholar] [CrossRef]
  53. Lyon, B.; Dewitt, D.G. A Recent and Abrupt Decline in the East African Long Rains. Geophys. Res. Lett. 2012, 39, 1–5. [Google Scholar] [CrossRef]
  54. Cai, W.; Borlace, S.; Lengaigne, M.; Van Rensch, P.; Collins, M. Increasing Frequency of Extreme El Niño Events Due to Greenhouse Warming. Nat. Clim. Change 2014, 4, 111–116. [Google Scholar] [CrossRef]
  55. Nicholson, S.E. Climate and Climatic Variability of Rainfall over Eastern Africa. Rev. Geophys. 2017, 55, 590–635. [Google Scholar] [CrossRef]
  56. ACF. 2015/16 El Nino Event Global Report; ACF: London, UK, 2015. [Google Scholar]
  57. Ntale, H.K.; Gan, T.Y. East African Rainfall Anomaly Patterns in Association with El Niño/Southern Oscillation. J. Clim. 2004, 9, 257–268. [Google Scholar] [CrossRef]
  58. UNISDR. Progress and Challenges in Disaster Risk Reduction: A Contribution Towards the Development of Policy Indicators for the Post-2015 Framework for Disaster Risk Reduction; ISDR: Geneva, Switzerland, 2014. [Google Scholar]
Water 17 01044 g001
Figure 1. Map of the Lake Kyoga Basin in Uganda (inset) showing the elevation of the Lake Kyoga Basin and the weather stations (black dots).
Figure 1. Map of the Lake Kyoga Basin in Uganda (inset) showing the elevation of the Lake Kyoga Basin and the weather stations (black dots).
Water 17 01044 g001
Water 17 01044 g002
Figure 2. Taylor diagram showing the performance of CHIPRS and TAMSAT datasets against station observed rainfall values.
Figure 2. Taylor diagram showing the performance of CHIPRS and TAMSAT datasets against station observed rainfall values.
Water 17 01044 g002
Water 17 01044 g003
Figure 3. Mean monthly rainfall cycle for different stations for the period 1981–2017.
Figure 3. Mean monthly rainfall cycle for different stations for the period 1981–2017.
Water 17 01044 g003
Water 17 01044 g004
Figure 4. Temporal variation in total annual rainfall for selected stations during 1983–2017. The blue lines in the graphs shows the variation of the annual rainfall over the years, while red lines indicate the simple regression lines showing increasing trends for upward lines and decreasing trends for downward lines. Significance of the trends are tested using MK test and results indicated in Table 2.
Figure 4. Temporal variation in total annual rainfall for selected stations during 1983–2017. The blue lines in the graphs shows the variation of the annual rainfall over the years, while red lines indicate the simple regression lines showing increasing trends for upward lines and decreasing trends for downward lines. Significance of the trends are tested using MK test and results indicated in Table 2.
Water 17 01044 g004
Water 17 01044 g005
Figure 5. Comparison of wet/dry days during the MAM season using two gridded satellite datasets, CHIRPS-2.0 and TAMSAT v3.1, for the historical period 1983–2017.
Figure 5. Comparison of wet/dry days during the MAM season using two gridded satellite datasets, CHIRPS-2.0 and TAMSAT v3.1, for the historical period 1983–2017.
Water 17 01044 g005
Water 17 01044 g006
Figure 6. Comparison of wet/dry days during the SOND season using two gridded satellite datasets, CHIRPS-2.0 and TAMSAT v3.1, for the historical period 1983–2017.
Figure 6. Comparison of wet/dry days during the SOND season using two gridded satellite datasets, CHIRPS-2.0 and TAMSAT v3.1, for the historical period 1983–2017.
Water 17 01044 g006
Water 17 01044 g007
Figure 7. (a–g) Graphical representation of the time series of seasonal cumulative departure index (CDI) for selected stations in the Lake Kyoga Basin. The light grey color bar chart represents MAM and the black color chart shows SOND.
Figure 7. (a–g) Graphical representation of the time series of seasonal cumulative departure index (CDI) for selected stations in the Lake Kyoga Basin. The light grey color bar chart represents MAM and the black color chart shows SOND.
Water 17 01044 g007
Water 17 01044 g008
Figure 8. The annual rainfall anomaly index (RAI) for selected stations in the Lake Kyoga Basin in Uganda.
Figure 8. The annual rainfall anomaly index (RAI) for selected stations in the Lake Kyoga Basin in Uganda.
Water 17 01044 g008
Table 1. Detailed geographical information of meteorological stations in the Lake Kyoga Basin.
Table 1. Detailed geographical information of meteorological stations in the Lake Kyoga Basin.
StationsLongitude
(Degree)		Latitude
(Degree)		Elevation
(Meters)		Data Period
(Years)		Rainfall
Zones
Namulonge32.60.51128.31981–2017B
Tororo34.20.711761981–2017D
Soroti33.61.71115.81981–2017E
Jinja33.10.51175.11981–2017A2
Lira32.32.21120.21981–2017I
Serere33.51.510981961–2017E
Kiige331.11089.41981–2017B
Buginyanya34.41.318751981–2017F
Kotido34.131219.51981–2017G
Namalu34.821274.41981–2017L
Table 2. Shows a statistical description of annual and seasonal rainfall and MK trend and Sen’s slope for four major stations in the Lake Kyoga Basin in Uganda. The stations with an asterisk (*) mark show a statistically significant trend; Min and Max show minimum and maximum values in millimeters (mm). Negative Sen’s slope values show a decreasing trend, while a positive Sen’s slope indicates a positive trend.
Table 2. Shows a statistical description of annual and seasonal rainfall and MK trend and Sen’s slope for four major stations in the Lake Kyoga Basin in Uganda. The stations with an asterisk (*) mark show a statistically significant trend; Min and Max show minimum and maximum values in millimeters (mm). Negative Sen’s slope values show a decreasing trend, while a positive Sen’s slope indicates a positive trend.
AnnualMAMSOND
StationMean
(mm)		Max
(mm)		Min
(mm)		p-v (MK)Sen’s Slope
(mm/Decade)		Mean
(mm)		Max
(mm)		Min
(mm)		p-v (MK)Sen’s Slope
(mm/Decade)		Mean
(mm)		Min
(mm)		Max
(mm)		p-v
(MK)		Sen’s Slope
(mm/Decade)
Namalu1180.61633.5765.00.01 *87.8385.0534.3236.50.53911.8346.5184.7673.10.005.9
Buginyanya1719.63179.420640.1163.9526.7681.7337.00.5399.9560.1224.4861.50.044.9
Kiige1272.41530.7958.50.1632.6467.1645.1364.70.685−4.7460.6180.3723.80.093.4
Namulonge1719.62015.71240.50.2752.9470.7615.4326.30.6474.3452.8140.9772.60.7212.3
Soroti1411.91813.91148.00.2530.3484.8660.0366.50.969−7.0455.0244.0740.20.3432
Tororo1573.32203.51191.80.01 *8.6578.1773.9411.20.4107.8532.9153.8955.10.0168.7
Jinja1331.91739.11073.50.02 *50.4476.8612.5318.90.24412.7489.7167.0863.20.4129.1
Lira1405.61698.0997.30.3819.6425.8534.1296.60.302−10.5484.8202.0707.40.2330.1
Table 3. Summary of onset, cessation dates, and seasonal length for different rainfall stations in the Lake Kyoga Basin 1981–2017. CV is the coefficient of variation of the different seasonal characteristics.
Table 3. Summary of onset, cessation dates, and seasonal length for different rainfall stations in the Lake Kyoga Basin 1981–2017. CV is the coefficient of variation of the different seasonal characteristics.
MAMSOND
StationStatisticsOnset (Dates)Cessation
(Dates)		Seasonal Length MonthsOnset (Date)Cessation
(Dates)		Seasonal
Length Days
SorotiMean15 March27 June3.521 July15 November3.7
Minimum29 February30 May2.01 July1 November1.8
Maximum15 April29 September6.723 October31 December5.3
CV11%11%33%11%5%5%
NamulongeMean30 March22 June2.721 August24 November3.1
Minimum3 March1 June1.01 August1 November1.6
Maximum14 May10 August4.631 October26 December4.4
CV22%9%30%9%6%22%
TororoMean19 March7 July3.626 August7 December3.3
Minimum1 March1 June2.21 August17 November1.5
Maximum27 April15 October6.27 October31 December4.7
CV19%16%27%8%5%23%
BuginyanyaMean16 March13 July6.54 September24 November2.7
Minimum29 February29 June6.019 August20 November3.1
Maximum26 April29 November11.14 November24 December3.3
CV21%16%23%7%3%20%
LiraMean29 March6 August2.914 August27 November3.4
Minimum2 March3 June2.01 August1 November3.0
Maximum17 May14 December4.59-Oct30 December3.6
CV19%31%1%7%5%1%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Babaousmail, H.; Ojara, M.A. Evaluation of Historical Dry and Wet Periods over Lake Kyoga Basin in Uganda. Water 2025, 17, 1044. https://doi.org/10.3390/w17071044

AMA Style

Babaousmail H, Ojara MA. Evaluation of Historical Dry and Wet Periods over Lake Kyoga Basin in Uganda. Water. 2025; 17(7):1044. https://doi.org/10.3390/w17071044

Chicago/Turabian Style

Babaousmail, Hassen, and Moses A. Ojara. 2025. "Evaluation of Historical Dry and Wet Periods over Lake Kyoga Basin in Uganda" Water 17, no. 7: 1044. https://doi.org/10.3390/w17071044

APA Style

Babaousmail, H., & Ojara, M. A. (2025). Evaluation of Historical Dry and Wet Periods over Lake Kyoga Basin in Uganda. Water, 17(7), 1044. https://doi.org/10.3390/w17071044

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop