Extreme Rainfall is crucial for Crop production and food security in Eastern Africa. This paper seeks to investigate the changes and variability in wet days and dry spells over the IGAD region. Data used are Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS). Several statistical methods and wet days and dry spells thresholds at ≥ 1mm applied on decadal (10 years), 20, 30 and 41 years. The results show that decrease in the number of wet days lead to longer dry spells. The majority of districts in Uganda, southwestern South Sudan, southwestern zones in Ethiopia, highlands of western and Nyanza counties in Kenya observed the highest number of wet days (50–70 days) and lowest consecutive dry spells (0–1 spells). Uganda and South Sudan are the two countries with lowest variability on wet days (highest variability in dry spells). Again, South Sudan and Uganda, most parts of Ethiopia, highlands of western Kenya observed 90–100% probability of exceeding 7 and 14 days (1 and 2 spells) during March-May (MAM), June-August (JJA) and September-November (SON). Northeastern Kenya and Somalia, southeastern Ethiopia, most parts of Eritrea and Djibouti observed less than 5% of probability under 7, 14,21,28 days (1,2,3,4 spells). In addition, most parts of the region observed decreased number of wet days in the 1980s and 1990s, while the last decade (2011–2020) experienced an increase during MAM and JJA. These findings are important for rain-fed agriculture, supplementary irrigation planning and food security in the IGAD region.
Research Article
Observed Changes and Variability in wet days and Dry Spells over IGAD region of Eastern Africa
https://doi.org/10.21203/rs.3.rs-2493554/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Wet days
dry spells
Changes and Variability
IGAD region
Eastern Africa
Africa is one of the most vulnerable continents to climate change (IPCC, 2022) due to dependability on climate sensitive based livelihoods and low adaptive capacity( Ogallo 1989, Ntale et al 2003, Omondi et al 2009, 2012, Osima et al., 2018). The climate change affects the variability of the monthly precipitation in terms of frequency, duration, and magnitude of wet/dry days, wet/dry spells over many regions in Greater Horn of Africa (GHA). The Gross domestic product (GDP) of most countries in Equatorial East Africa are sensitive to changes and variability in extreme events (Philippon et al., 2015a). Agriculture plays a vital role in the lives of the people of East Africa, with about around 80% of the population being engaged in agriculture activities especially smallholder farming (Recha et al., 2012).
Rainfall is the most important meteorological variable for rain-fed agriculture production over East Africa (Shukla et al., 2014). In the arid and semi-arid climate of the African continent, the more the counts of dry days, the longer the length of the dry spell and the lesser the number of wet days, the shorter the duration of wet the spell as well as the total amount of rainfall (Bouagila & Sushama, 2013). Rainy season over East Africa is highly variable in both space and time (Camberlin et al., 2009) it begins in February over southern parts of Uganda to central and northern parts of Sudan in July and August. However, long rains which occur between March and May (MAM) are crucial for sub-regions within equatorial East Africa ( Williams and Funk 2011; Lyon and Dewitt 2012; Liebmann et al. 2014, Endris et al., 2019), while JJAS season is usually referred to as summer monsoon rains, which is called Kiremt in Ethiopia, Kareef in both Sudan and South Sudan, which account for 65–95% over Ethiopia (Segele & Lamb, 2005), 80–95% over South Sudan(Omoj et al, 2016) and 95–99% of total annual rainfall over Sudan(Lotfie et al., 2018).
The day-to-day characteristics of total rainfall, intensities and variability determined by wet and dry episodes (Ratan & Venugopal, 2013). Wet days are sequences of days with rain, conversely, the dry spells are sequences of days without rain at 1mm minimum threshold (Bärring et al., 2006, Apelgren et al., 2019).The intensity and duration of droughts and floods, extra-ordinary heavy rainfall is directly proportional to the number of days with/without rainfall (Korbinian Breinl, et al., 2020). In most cases, the extreme wet spells/dry spells could be referred to floods and drought (Ndehedehe et al., 2023). Long wet days/wet spells, dry days/dry spells are one of drought impacts indicator and severe risks to socioeconomic activities (Mwangi et al., 2014). Putting into account climatic conditions of any region, the extreme wet spells can lead to saturated soils and thus influence the flood hazard (Korbinian Breinl, et al., 2020), water- related vector-borne diseases (Thomson et al., 2018), water quality (Schilling et al., 2020), rivers flows (Rameshwaran et al., 2021) and urban water management.
Regional, national and local information on wet days and dry spells are critical for monitoring and managing the current and future impacts and developing sustainable food security interventions (Haile et al., 2020). Many studies in eastern Africa have indicated the important of assessing the observed changes in wet and dry spells, heatwaves, droughts, flood patterns and other unexpected extra-ordinary extreme weather events during main rainy seasons and rainfall peaks(Funk et al 2008, Lyon and Dewitt 2012, Mwangi et al 2014,Haile et al.,2020a,Gudoshava et al.,2020a), because these extremes combined with other socio-economic factors are responsible for crop yield reduction, food insecurity, and water scarcity, which impacted farmers and pastoralists communities who migrate to other region in search of water and green pasture for their livestock.
The observed trends in frequencies and intensities of extremes climate events post a big threat to socio-economic sectors activities (Hutchinson et al., 2013), countries development agenda (Kumar et al., 2012), agriculture planning and main staple food production stability (Kumar et al., 2012), crop growth and food security(Chemura et al., 2020). For proper future planning and managing water resources for rain-fed agriculture (Thornton et al., 2009) and food security (Shukla et al., 2014), the information on mean state of wet/dry days or wet/dry spells period is not enough, it is also necessary to determine the characteristics in terms of frequency, duration and intensity of the rainfall variability (Yapo et al., 2020). Duration of wet/dry days or wet/dry spells determines supply and demand levels of the water resources (Mandapaka, et al2016). Prolonged wet/dry days and wet/dry spells appear to be a significant factor for rain-fed food crops over Africa, Sahel(Zhang et al., 2018)),Eastern Arica (Barron et al., 2003, Nicholson, 2017, Gitau et al., 2018).
Observed change and variability in wet days and dry spells may drive the types of cash and food crops, where and when to plan, how to best prepare the agriculture land, suitability of soil for crops as the soil moisture and soil water contents determines the crop growth and productivity (Shaxson, et al., 2003, Shongwe et al., 2011) where, uncertain climate conditions such as present and future extreme rainfall events and extra-ordinary rainfall amounts results in the floods or drought in Africa (Verschuren, et al., 2000). The twice-annual passage of the intertropical convergence zone (ITCZ) report by Nicholson, (2017b) could explain the highest and lowest number of wet days and dry spells. The characteristics of wet/dry spells in terms of frequency, duration and magnitude over equatorial East Africa revealed more wet spells with precipitation of above 14 mm per day (Philippon et al., 2015). A research by Gitau, et al (2008) found that 1-day wet and dry spells were dominant in western Kenya. The duration and magnitude of wet/dry spells varies with climatic zones and that there was no clear monotonic trend in wet/dry spells over Akobo Basin of Ethiopia (Kebede, et al, 2017). Although the underlying mechanisms can be complex, Changes in the timing of wet days and dry spells can for instance pose a threat to food production, water deficit, crops, sowing and developmental stages(Troy et al., 2015, Korbinian Breinl et al, 2020). The extra-ordinary wet days increase risks of flooding and extreme dry spells contribute to droughts and their various impacts.
The goal of this paper is to determine the observed changes and variability of wet days and dry spells for three main agricultural seasons (MAM, JJA, SON) and dry season (DJF) over the IGAD region of Eastern Africa for successful rain-fed agriculture production planning, selection of the crop type and variety. The paper is structured as follows: Section 2 describes the CHIRPS daily gridded satellite estimates, area of study, statistical methods and thresholds used to define wet days and dry spells. The results and discussions are described in Section 3 and 4 which contains spatial and inter-annual variability patterns of means, change and variability, probability of exceeding defined days and dry spells thresholds using different techniques and thresholds, while conclusions presented in Section 5 together with recommendation statements.
2.1. Study area
The study focuses on Intergovernmental Authority on Development (IGAD) member states of Sudan, Eritrea, Djibouti, South Sudan, Ethiopia, Kenya, Somalia and Uganda (Fig. 1). The geographical coordinates of the region are latitude 21.4°-51.2°E and Longitude 5°-23.2°N. The region is characterized by complex topography, armed conflicts and food insecurity. The region elevation varies from an area below sea level over Sudan to highest points of Mount Kenya at 5,199 m as the second highest mountain in Africa after Mount Kilimanjaro (5,895 m) in Tanzania, Rift Valley extend from Kenya to Ethiopia.
The IGAD region climate is affected severely by these high elevation landmarks, seasonal movement of intertropical convergence zone (ITCZ) north and southward which is the one of factors that determine the variation in four different rainfall seasons such as December, January, February (DJF), March, April, May (MAM), June, July, August, September and October, November, December (OND). Also, many studies show the climate of region is influenced by El Nino/Southern Oscillation ( Ogallo, 1988, Indeje et al., 2000, Anyah and Semazzi, 2006,Anyah & Qiu, 2012) as well as variability of sea-surface temperature over the Indian Ocean (Williams & Funk, 2011). The different ENSO phases (El Niño and La Niña or neutral) have different impacts over different parts of the region (Clark et al., 2003, Anyah & Qiu, 2012). The variation in climatic zones whether warm deserts or humid highland climate are mainly driven by orography, geography and micro- synoptic systems(Peel et al., 2007). These situation offers an opportunity or could be a challenge affecting the accuracy of Satellite rainfall estimates over the region.
The areas labelled with sky blue are five potential agricultural production areas (El Gadaref state in Sudan, Upper Nile state in South Sudan, Arsi zone in Ethiopia, Trans Nzoia county in Kenya, Arua districts in Uganda) used to access the temporal patterns of wet and dry spells over the region.
2.2. Datasets
High resolution Satellite Rainfall Estimates (SRE) products selected for this study are Climate Hazards Group (CHG) Infrared Precipitation with in-situ station (CHIRPS) datasets from the University of California at Santa Barbara (UCSB). CHIRPS product is developed at 0.05° spatial resolution and daily, pentadal, dekadal, and monthly temporal resolution available from 1981 to near present (Funk et al., 2015.The multiple steps detailed major input datasets and processes, merging with station data, weighted bias ratios, homogeneity of the time series and algorithm used to generate CHIRPS products provided in Funk et al., (2014; 2015a).
2.2 Methods
2.2.1 Criterion and threshold for Wet dry days and dry spells
There are a considerable number of definitions and thresholds for calculating wet days and dry spells patterns in the Literature. In most cases, these definitions and thresholds produce different wet/dry days and spells even when applied to the same observation or gridded dataset (e.g. Laux et al. 2008). The first and most used methods in literature are those apply threshold values on Total rainfall amount (e.g. (Stern et al 1981, Sivakumar 1988, Segele and Lamb 2005, Marteau et al 2009, Marteau et al 2011), Fraction of evapotranspiration, Walter method(Olatunde & Love, 2018)), Number of rainy days and spell lengths(Gudoshava et al., 2020a). All these methods observed limitation regarding number of rainy days within threshold of total rainfall amount, seasonal extremes indices and probability of exceeding defined wet days and dry spells. Therefore, in this study wet days and dry spells criterion and thresholds were determined by adopting a threshold of 1mm per day used to define a wet/dry day, 7 consecutive wet days (1 wet spells), 7 consecutive dry spells(1 dry spells), probability of exceeding 7,14, 21 and 28 wet days(1,2,3,4 spells), probability of exceeding 1,2,3,4 dry spells(7,14,21 and 28 dry days). The Climdex indices use a threshold of 1mm per day to define wet/dry days for the calculation of maximum number of Consecutive Wet Days (CWD) and maximum number of Consecutive Dry Days (CDD). The indices used by many researchers around the world(Seneviratne et al., 2012, Polade et al., 2014, Funk et al., 2019, Giorgi et al., 2019, Wainwright et al., 2021). The adoption of these criteria informed by nature of rainfall required for food crops production such as maize and sorghum and humid climate in potential agricultural areas. In addition, to avoid unreasonable wet days and dry spells.
| Wet/dry days and spells criterion and threshold will be based on rainy days and spell lengths |
|---|
| Threshold for wet days: ≥ 1mm |
| Threshold for wet days: ≤ 1mm |
| Threshold for wet spells:7 consecutive days at least with 1mm rain |
| Threshold for dry spells: 7 consecutive days at least without 1mm rain |
| Probability of exceeding for wet days: 7,14,21 and 28 wet days |
| Probability of exceeding for dry spells: 1,2,3 and 4 dry spells |
| For MAM season: 92 days are maximum number of wet days or 13.1 dry spells |
| For JJA season: 92 days are maximum number of wet days or 13.1 dry spells |
| For SON season: 91 days are maximum number of wet days or 13 dry spells |
| For DJF season: 90 days are maximum number of wet days or 12.8 dry spells |
2.2.2 Statistical Methods for Wet days and Dry spells
The main statistical methods used for climatology and baseline of variability in wet days and dry spells are arithmetic mean of decadal (10 years), 20, 30 and 41 years, changes in means, Coefficient of variation (CV), Anomaly, Probability of exceeding. The arithmetic means of wet and dry spells, which is the sum of number of days with at least ≥ 1mm or ≤ 1mm threshold divided by number of years considered in analysis as described in Eq. (1).
\(\stackrel{-}{X}= \frac{1}{{N}} \sum _{{i}=1}^{{n}}{{X}}_{{i}} \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .\left(1\right)\)
Where \(\stackrel{-}{X}\) = average (or arithmetic mean), n is the sample size, which are 10, 20, 30 and 41 years, \({X}_{i}\) = the value of each individual seasons and yearly number of wet or dry spells being averaged. For purpose of this study, the means computed on decadal are 1981–1990(1980s), 1991–2000(1990s), 2001–2010(2000s), 2011–2020(2010s), 20 years (1981–2000 and 2001–2020), 30 year or climatology (1981–2010) and 41 years (1981–2021) to assess the mean state of wet days and dry spells.
The CV, which is standard deviation ratio (\({\sigma }_{x}\)) divided by mean \(\overline{X}\) used to measure variability in wet days and dry spells as described in Eq. (2).
$$\text{C}\text{V}\left(\text{%}\right)=\left( \frac{{\sigma }_{x}}{\stackrel{-}{X}}\right)*100\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(2\right)$$
Where \({\sigma }_{x}\) is Standard Division (SD) computed from Eq. 3 below.
\({\sigma }_{x}{}= \sqrt{\frac{1}{\mathbf{n}}\sum _{\mathbf{i}=1}^{\mathbf{n}}{({\mathbf{x}}_{\mathbf{i}}- \stackrel{-}{\mathbf{x}} )}^{2}}\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .\left(3\right)\)
The statistical methods used to compute climatology or mean state, mean of variability is average of 30 years (1981–2010) for mean. The departure from the mean(anomaly) and standardized anomaly are used to determined areas with increase/decrease in wet days and dry spells. This done by subtracting each year,10 years (1981–1990, 1991–2000, 2001–2010, 2011–2020) from 30 year or climatology (1981–2010)
The difference in two means of current 20 years (2001–2020) compared with previous 20 years (1981–2000) used to assess the whether there is shift in means state and mean of variability of wet days (increased or decreased trends) and dry spells (prolonged dry spells). It can be written as in Eq. 4 bellow.
\(\nabla\stackrel{-}{X}\ = \frac{1}{{N}} \sum _{{j}=1}^{{n}}{{X}}_{{j}} -\sum _{{i}=1}^{{n}}{{X}}_{{i}} \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(4\right)\)
Where, (\({{X}}_{{j}}\)) represent current 20 years (2001–2020) and (\({{X}}_{{i}}\)) represented previous 20 years (1981–2000). The Coefficient Variation (CV) which is standard deviation ratio (\({\sigma }_{x}\)) divided by mean \(\stackrel{-}{X}\) as described in Eq. 3.\(\)
Normalized departures of year and season from climatology used to rank the wet days and dry spells anomalies objectively(Service, 2001). This method widely used to objectively rank synoptic-scale events (e.g. (Hart and Grumm 2001, Junker et al 2009, Graham and Grumm 2010). In this study, in season of each year, the standardized anomaly, SAI is derived by subtracting the local mean, \({x}^{-}\), from the area-average seasonal mean of the available wet and dry data within the study region and divided by the corresponding standard deviation of inter-annual variability estimated from the reference period (1981–2010) as described in Eq. 5:
\(\text{S}\text{A}\text{I}=\frac{x-{x}^{-})}{{\sigma }_{x}\ }\dots \dots \dots \dots \dots \dots \dots \dots \dots ..\left(5\right)\)
SAI is used to rank events based on departures from the local climatology in units of standard deviation (𝜎).
Linux based Climate Data Operators (CDO) commands lines, R- statistical Package Climate Data Tool (CDT) and ArcGIS 10.4 used in plotting and mapping spatial maps.
3.1 Mean and variability in Wet Days and Dry Spells
Figure 2 shows the average number of wet days and dry spells during MAM, JJA, SON and OND seasons based on an average of 30 years mean (1981–2010) over the IGAD region of Eastern Africa. Generally, the results show the areas with the highest number of wet days vary from season to season and within countries. On seasonal timescale, the MAM season (Fig. 2a), the highest number of wet days (55-65days) are recorded over southwestern parts of the region, mainly over southwestern South Sudan, southern and south-Eastern parts of Uganda, Lake Victoria basin and the highland of western and Nyanza region of Kenyan, while everywhere in Sudan, Eritrea, Djibouti, northern Somalia, Upper Nile state in South Sudan, southeastern, northern and northeastern Ethiopia, northeastern Kenya recorded lowest number of wet days, which is less than 15 days per three months. Similarly, the JJA season (Fig. 2b), the highest number of wet days (65–75) were recorded over western South Sudan, highlands of central and western Ethiopia, whereas northern Sudan, southeastern Ethiopia zones, majority of counties in Kenya, most parts of districts in Somalia and coastal Djibouti recorded lowest number of wet days. In SON season, the highest number of wet days were recorded over most parts of Uganda, south-western South Sudan and Ethiopia, while most parts of Sudan, eastern and Southeastern Ethiopia, Kenya, Somalia, Djibouti, Eritrea recorded the lowest number of wet days ranged between 45–60 dry days (Fig. 2c). During the DJF season, except southern and central Uganda, Lake Victoria basin, the rest of IGAD regions recorded lowest wet days ranging between 0–10 dry days (Fig. 2d). The wet days patterns proved the potential of the highland of western Kenya, most parts of Uganda, southwestern Ethiopia during MAM, JJA and SON, most parts of South Sudan during MAM and JJA for sustainable food crops production and food security. The south-northwards and north-southward movement of ITCZ around the Equator suggests the highest (lowest) number of wet and dry days over the region. Across all four seasons, out of 92 possible wet days, the sub-regions in Northern and Red Sea states in extreme northern Sudan, northeastern consistently recorded less than 5 wet days. Generally, the highest number of wet days (lowest wet days) are recorded in areas where the MAM, JJA, SON and DJF season are considered as the main rainfall season (dry climatology). The highest number of wet days explained the importance of JJA for South Sudan, Ethiopia, while MAM and OND for Kenya, MAM, JJA, SON for the Uganda agriculture sector.
Figure 3 presents the mean dry spells(climatology) patterns during four seasons. Conversely, the shortest and longest continued dry-spells recorded over areas with highest and lowest wet days as described in Fig. 2. Most parts of Kenya, Somalia and Ethiopia recorded 3–5 consecutives dry spells (21–35 consecutive dry days) during MAM (Fig. 3a), whereas southwestern and northern parts of South Sudan, western Ethiopia, rain belts in southern parts of Sudan recorded the less than 1 spells as the shortest spells in IGAD region during JJA season (Fig. 3b). Nevertheless, the OND shows 3–5 spells (21–35 consecutive dry days) recorded over most parts of Kenya, Somalia and southeastern Ethiopia, while southwestern Uganda recorded 1 spell as longest (Fig. 3c). Similarly, central and northern Uganda, most parts of southern, central and northern Kenya, northeastern, central and southwestern Ethiopia have the longest spells during DJF season(Fig. 3d). Lake Victoria basin consistently observed lowest spells across all four seasons. In addition, these results show, the majority of lands in the IGAD region are dry and susceptible to agriculture failure due to 2–4 consecutive dry spells, therefore, the prolonged number of dry spells over most parts of ASALs in the IGAD region explained why they experienced chronic food insecurity. The mean wet days and dry spells patterns explained south-north timing of rainfall onset progression and retreat(cessation) from March to November.
Figure 4 presents the seasonal mean state of the variability of the wet days (first raw) and dry spells (second raw) at the 1mm threshold for the 1981–2010 reference period. The results show areas and seasons with the highest amount of total rainfall as noted in previous study by Cattani et al., (2018), wet days and less dry spells recorded less variability compared to dry regions. Thus, most parts of southwestern South Sudan, most parts of Uganda, highlands of western Kenya and Ethiopia have lowest variability (less 10%), whereas ASALs in southern parts of Sudan, northeastern Kenya, eastern and southeastern Ethiopia especially Eastern Hararghe and northern Somalia recorded highest variability in wet days (Fig. 4a), because these areas experience low mean rainfall through the year(Schreck & Semazzi, 2004). According to Muluneh Sorecha & Bayissa, (2017), the CV values of, 41, 34 and 31% recorded Babile, Haramaya and Kersa districts in Eastern Hararghe, Ethiopia. Most parts of South Sudan, central and western Ethiopia observed lowest variability during JJA (Fig. 4b), and the same patterns of low variability continued to SON seasons, however, the highest variability was observed over northeastern Kenya, southeastern Ethiopia and northern Somalia (Fig. 4c). Due to widespread dry conditions during DJF season over the IGAD region, the highest variability (more than 100%) is observed everywhere in Sudan, northern parts of South Sudan, south-eastern Ethiopia, most parts of Somalia and northeastern Kenya (Fig. 4d). On the contrary, the dry spell patterns (Fig. 4e-h) or second row shows highest variability of dry spells over the same regions and areas with lowest variability on wet days. Therefore, most parts of Uganda during MAM and SON (Fig. 4e and 6g), most parts of South Sudan during JJA season (Fig. 4a) observed highest variability of wet days (more than 100%) compared to other countries in the IGAD region. The patterns of variability in wet and dry spells revealed that areas with lowest variability in wet days (highest variability of dry spells) are most conducive environments for rain-fed food crops production and provided food basket and food security for the IGAD region.
Furthermore, the study analyses the means values at the countries or national level (Djibout, Eritrea, Ethiopia, Kenya, Somalia, Sudan, South Sudan and Uganda) and region levels (average of all pixels over IGAD region) for wet days and dry spells during 1980s, 1990s,2000s, 20, 30- and 41-years average presented in Fig. 5. Generally, Uganda had the highest number of wet days (lowest consecutive dry spells) in both MAM SON seasons, while South Sudan recorded the highest during JJA. On the other hand, Somalia and Djibouti recorded the highest number of dry spells over the season of MAM and OND. In rankings countries based on number of wet days(dry spells), Uganda recorded the highest number of wet days (lowest dry spells), followed by South Sudan, Ethiopia, Kenya, Somalia, Eritrea, Djibouti and Sudan during MAM season(Fig. 5a,b), while South Sudan on top of countries based on number of wet days(dry spells), followed by Uganda, Ethiopia, Eritrea, Sudan, Djibouti, Kenya and Somalia (Fig. 5c,d). for OND, again Uganda ranked on top, followed by South Sudan, Ethiopia, Kenya, Eritrea, Somalia, Sudan and Djibouti (Fig. 5e, f). Due to a big portion of ASALs in northern Sudan, northeastern Kenya, southeastern Ethiopia, and most parts of central Somalia, the IGAD region recorded less than 10 wet days (more than 8 consecutive dry spells) during MAM. The recent years (2011–2020) recorded the highest mean wet days in South Sudan, Uganda, Ethiopia, Kenya and average of the entire IGAD region.
Figure 6 below shows the average number of wet days(first column) and dry spells(second column) during 1980s, 1990s, 2000s,2010s (1981–1990, 1991–2000, 2001–2010,2011–2020), 30 years average (1981–2010), 41 years average (1981–2021) and 20 years average (1981–2000,2001–2020) in Arua, AlQadarif, Upper Nile, Arsi, and Trans Nzoia as the most five agricultural districts, state, zone and county in Uganda, Sudan, South Sudan and Ethiopia. The results revealed that the mean values for four decades (1980s, 1990s, 2000s,2010s) and 41 years average (1981–2021) for MAM, JJA and SON Seasons recorded very close mean values across all five regions. The different timescale means values for MAM season showed that Arua followed by Trans Nzoia recorded the highest mean of wet days (lowest mean consecutive dry spells), and AlQadarif recorded the lowest mean (highest mean dry spells) during the season (Fig. 6a, b). The highest wet days (lowest dry spells) during JJA were recorded over Arua and Upper Nile (Fig. 6c, d). The patterns of JJA continued, where the Arua recorded the highest mean of wet days (lowest mean consecutive dry spells) during up to SON (Fig. 6e, f). in Addition, the Arsi zone in Ethiopia recorded the lowest mean values (highest mean value of consecutive dry spells) of wet days during OND.
3.2 Probability of Wet Days and Dry Spells exceeding defined thresholds
Figure 7 and 8 shows the probability of wet days and dry spells exceeds 7, 14,21,28 consecutive wet days and 1,2,3,4 consecutive dry spells during MAM, JJA, SON and DJF season reference to 1981–2010 climatology. The results for MAM season (Fig. 7a-d) show everywhere in Uganda, South Sudan and Ethiopia, Kenya, Somalia, Djibouti and most parts of Eritrea observed more than 90% probability of wet days to exceed 7 consecutive wet days (Fig. 7a). However, these areas started shrinking continuously with an increase in probability of consecutive wet days from 7 days to 14,21 and 28 days. Under probability of 14 consecutive wet days (Fig. 7b), the highest probability recorded in the whole country of Uganda, most parts of South Sudan, central and southwestern Ethiopia, highlands of western and Nyanza in Kenya with probability more than 80% southern, while the lowest probability was recorded in the central and northern part of Sudan, northwestern Eritrea with probability less than 1%. The same patterns observed under probability of exceeding 21 and 28 consecutive wet days (Fig. 7c, d), where the whole country of Uganda, southwestern South Sudan and Ethiopia recorded highest probability (80–100%) and lowest probability (1–5%) extended to northeastern Kenya, northeastern and southeastern Ethiopia, northern Somalia, everywhere in Eritrea and Djibouti. For JJA, more than 90% probability of exceeding 7 and 14 consecutive wet days which cover whole parts of Uganda, South Sudan, western, central and northern parts Ethiopia, western and costal Kenya (Fig. 7e) has continuously shrink to cover lake Victoria basin, central and northern Uganda, and southwestern Ethiopia for probability of exceeding 21 and 21 consecutive wet days, whereas lowest probability extended to northern Sudan, most parts of Kenya, everywhere in Somalia, southeastern South Sudan and Ethiopia(Fig. 7g,h). The highest probability(more than 80%) during SON observed over whole Uganda and South Sudan, most parts of Ethiopia, western and coastal counties in Kenya(Fig. 7i), most parts of Uganda and South Sudan, western and northwestern Ethiopia, western and Nyanza in Kenya (Fig. 7j,k) and patterns reduced to cover most parts of Uganda, western South Sudan and Ethiopia, western and Nyanza in Kenya(Fig. 7l) for probability of exceeding 7,14,21 and 28 consecutive wet days respectively. The patterns of probability for DJF (figure m-q), show the highest probability was recorded over southern Uganda, western and southern parts of Kenya for probability of exceeding 7days (Fig. 7m), which reduced to southern Uganda only under for probability of exceeding 21 and 28 days (Fig. 7o, q). In the IGAD region, Uganda and South Sudan, western and Nyanza counties in Kenya continuously recorded more than 80% probability of exceeding 7–28 days during MAM, JJA and SON season. No any changes in pattern of probability of exceeding 7–28 days over Sudan across all four seasons (MAM, JJA, SON and DJF) because these areas are dry climatology.
On other hand, the areas with lowest probability (less than 1%) under consecutive wet days recorded the highest probability (100%) exceeding 1,2,3 and 4 consecutive dry spells during MAM, JJA, SON and DJF seasons. Sudan, Eritrea, Ethiopia, Kenya, Djibouti and Somalia recorded 100% probability of exceeding 1 consecutive dry spell during MAM, SON and DJF seasons (Fig. 8a, I, m), whereas Western South Sudan, southern parts of Sudan, highlands of western Ethiopia are only areas recovered lowest probability (less than 5%) of probability of exceeding1 consecutive dry spell during JJA season(Fig. 8e). No any changes in pattern of 1% probability of exceeding 2,3 and 4 consecutive dry spells over northern parts of Sudan during MAM season (Fig. 8b, c,d) because these areas are dry climatology and ICTZ never crossed Sudan border this time of year. Most parts of Kenya and Ethiopia during SON and DJF recoded 60–100% probability of exceeding 2,3 and 4 consecutive dry spells because it is not possible for 7–28 days to pass without 1mm rainfall thresholds. Since JJA is main season over South Sudan, northern Uganda, highlands of western Ethiopia, this explain why the probability of exceeding 2,3 and 4 consecutive dry spells are less than 1% (Fig. 8f, g,h). Similarly, the patterns of 1% of 1% probability of exceeding 2,3 and 4 consecutive dry spells covered all Sudan during DJF season (Fig. 8n, o, q).
3.3 Change in Mean of Wet Days and Dry Spells
Figure 9 present the spatial patterns of changes in Seasonal mean state of Number of wet days at 1 mm thresholds for the four decades 1980s,1990s,2000s and 2010s (1981–1990,1991–2000,2001–2010 and 2011–2020), average of 40 years (1981–2020) and differences between current (2001–2020) and previous 20 years (1981–2000) relative to 1981–2010 reference period. This result shows high decadal (10 years) variability of wet days over IGAD region. The wet days have deceased over South Sudan, western Ethiopia in1980s (Fig. 8a), southeaster Ethiopia and everywhere in Uganda, Kenya and Somalia in 1990s (Fig. 9b), costal and northeastern Kenya, southern and central Somalia and most parts of Ethiopia in 2000s, while increase in wet days was observed over Kenya, central and southeastern Ethiopia in 1980s, southern parts of Sudan, western Ethiopia in 1990s, most parts of South Sudan and Uganda in 2000s(figure c) and a large portion of IGAD regions in recent years 2011–2020 (Fig. 9d). The extra-ordinary increase in wet days signals in recent years (2011–2020) maintained the increase signals over Uganda, South Sudan and highlands of western and Nyanza counties in Kenya when current 20 years (2001–2020) compared to previous 20 years (1981–2000) during MAM season (Fig. 9e). For JJA, the decreased wet days over Sudan, South Sudan and Uganda and increased over highlands of western Ethiopia, western and Nyanza in Kenya in 1980s (Fig. 9f) has recovered in 1990s (Fig. 9g, h), while increase continued in 2010s (Fig. 9i) and clearly increased wet days were captured when recent 20 years compared with the previous (Fig. 9j).The decreased wet days in 1998s during MAM and JJA continued in OND season over Uganda, South Sudan and Sudan (Fig. 9k). The decreased wet days signals have recovered over western Ethiopia, Sudan and northern parts of South Sudan in 1990s (Fig. 9l), while most parts of Ethiopia observed decrease again in 2000s (Fig. 9m) and northern parts of South Sudan, northwestern Ethiopia and almost everywhere in Sudan in 2010s (Fig. 9n). The decreased wet days over Sudan and northern parts of South Sudan and incresed patterns over lake victoria basin, eastern Uganda, most parts of Kenya, central and southeastern Ethiopia, everwhere in Soamila during OND decade of 2011–2020 is the most higest incresed(increase) in wet days patterns in the record of last 40 years. The patterns of DJF season(Fig. 9q-u) revelead the decresed wet days signals over Uganda in 1998s and 1990s(Fig. 9s,t) recovered in 2000s and decresed again in 2010s(Fig. 9u).The increase in wet days over most parts of the region in current 20 year (2001–2020) compared to previous 20years (1981–2000) contributed largely by extra-ordinary wet days in 2018, 2019 and 2020.
Figure 10 present the spatial patterns of changes in Seasonal mean state of number of dry spells at 1 mm thresholds for the four decades 1980s,1990s,2000s and 2010s (1981–1990,1991–2000,2001–2010 and 2011–2020), average of 40 years (1981–2020) and differences between current (2001–2020) and previous 20 years (1981–2000) relative to 1981–2010 reference period. This result shows high decadal (10 years) variability of dry spells over the IGAD region as observed in wet days analysis already discussed in Fig. 9. The dry spells has increased over South Sudan, western Ethiopia in1980s (Fig. 8a), southeastern Ethiopia and everywhere in Uganda, Kenya and Somalia in 1990s (Fig. 10b), coastal and northeastern Kenya, southern and central Somalia and most parts of Ethiopia in 2000s, while decrease observed over Kenya, central and southeastern Ethiopia in 1980, southern parts of Sudan, western Ethiopia in 1990s, most parts of South Sudan and Uganda in 2000s(figure c) and most parts of region in recent years 2011–2020(Fig. 10d). The strong of decrease (increase) in dry spells signals in 2010s (1990s and 2000s) maintained the decrease (increase)signals over Uganda, South Sudan and highlands of western and Nyanza counties in Kenya when current 20 years (2001–2020) compared to previous 20 years (1981–2000) during MAM season (Fig. 9e). for JJA, the increased dry spells over Sudan, South Sudan and Uganda and increased over highlands of western Ethiopia, western and Nyanza in Kenya in 1980s (Fig. 10f) has recovered in 1990s (Fig. 10g, h), while decrease continued in 2010s (Fig. 10i) and decreased dry spells capture well when recent 20 years compared with the previous (Fig. 10j).The increased dry spells in 1998s during MAM and JJA also continued in OND season over Uganda, South Sudan and Sudan (Fig. 10k). The increased dry spells signals have recovered over western Ethiopia, Sudan and northern parts of South Sudan in 1990s (Fig. 10l), while most parts of Ethiopia exhibit significant increase again in 2000s (Fig. 10m) and northern parts of South Sudan, northwestern Ethiopia and almost everywhere in Sudan in 2010s (Fig. 10n). The increased dry spells patterns over Sudan and northern parts of South Sudan and increased patterns over lake victoria basin, eastern Uganda, most parts of Kenya, central and southeastern Ethiopia, everywhere in Somalia during OND decade of 2011–2020 is the highest increase in dry spells patterns in the last 40 years due to extraordinary reduction of dry spells due heavy rain observed in last 6 years in the region. The patterns of DJF season (Fig. 10q-u) revealed the decreased dry spell signals over Uganda in 1998s and 1990s (Fig. 9s,t) recovered in 2000s and decreased again in 2010s(Fig. 10u). Increases in dry season rainfall over East Africa is also found by Wainwright et al., (2021).
The raw figures of changes in wet days and dry spells illustrated in table 2. The comparison for means in last 41 years (1981–2020), four decades of 1980s,1990,2000s and 2010s (1981–1990,1991–2000,2001–2010 and 2011–2020 respectively), current 20 year (2001–2020) and previous 20 years (1981–2000) from climatology(1981–2010 )over five potential agricultural production areas (El Gadaref in Sudan, Upper Nile in South Sudan, Arsi in Ethiopia, Tranz Nzoia in Kenya, Arua in Uganda. The MAM season results revealed that Tranz Nzoia in western Kenya recorded 5 wet days increase during 2011s, followed by 3 wet days over Trans Nzoia and Arsi in Ethiopia during 1981s, while 3 wet days decreased over Trans Nzoia during 1990s. During the last decade (2011–2020), the wet days over Trans Nzoia have increased by 8 and 11 days during JJA and OND respectively. As evidence of a slight increased MAM rainfall trend in the last decade 2011s, the Arsi zone, Upper Nile state, Trans Nzoia county and Arua district observed decrease in dry spells. In addition, the JJA observed − 0.8 decrease in dry spells over Trans Nzoia as the highest changes in dry spells across all MAM, JJA and SON seasons.
Table 2:illustrated the changes in wet days and dry spells over five potential agricultural production areas (El Gadaref in Sudan, Upper Nile in South Sudan, Arsi in Ethiopia, Tranz Nzoia in Kenya, Arua in Uganda
3.4 Inter-annual variability in Wet Days and Dry Spells
The standardized Anomaly Index (SAI) for wet days and dry spells over entire pixels over the IGAD region presented in 11. The results show an increase in the trend of wet days over the IGAD region during MAM, JJA and SON (Fig. 10a). This increase is attributed to the increase in intensity and frequency of wet days during the last decade (2011–2020) across all three main rainy seasons in the region. This finding is not consistence with studies by Nicholson, (2017), who reported decline in rainfall over East during MAM season in recent decades. In the last 40 years, 2018, 2019 and 2019 are the years with the highest increase in wet days. The MAM 2018, JJA 2020 and OND 2019 are the wettest seasons in the last 41 years with the highest increase in number of wet days anomalies. The MAM 1981, JJA 2018 and OND 1997 are the second wettest years in the ranking. On other hand, the dry spell anomalies show decreasing in trend across all three seasons (Fig. 11b). The MAM 1992, JJA 1982 and OND 1994 are the most years observed increase in number of dry spells anomalies in the last 41 years, while MAM 1987 and 1990, MAM2001,2008 and 2012, OND2019 are the most years with decrease in dry spells anomalies over IGAD region.
Figure 12 presents the Inter-annual variability of wet days anomaly over countries member states and entire pixels over the IGAD region. Generally, the results show, the IGAD region observed less than 10 days increase/decrease in wet days. Uganda and South Sudan as counties with the highest amount of total rainfall observed the highest number of wet days and highest(lowest) changes in wet days compared to other countries. MAM 2018 recorded extra-ordinary increased number of wet days over Kenya, Uganda, Somalia and South Sudan (Fig. 12a). MAM 1981 is the wettest (highest increased number of wet days) in the last 40 years over Kenya. In recent years (2011–2020), the JJA over Uganda recorded yearly variation in wet days patterns with 2021 recorded the highest decrease in wet days. the decreased rainfall patterns in 1980 and 1990s over Uganda, South Sudan, Ethiopia, Eritrea and Somalia due to decreased JJA wet days (Fig. 12b). The OND 1997 and 2019 are the most recorded extra-ordinary increased wet days in the last 40 years over Somalia, Ethiopia, Kenya and Eritrea (Fig. 12c). The year 1993 over Uganda recorded the highest decrease in wet days in the records of the last 40 years. The wet days patterns played a significant role in increased/decreased total rainfall and changes in rainfall trends.
In this paper, we have used criterion and thresholds in identifying wet days and dry spells characteristics. Change and variability in wet days and dry spells was observed over many parts of IGAD region. The decrease in the number of wet days lead to longer dry spells and shorter wet spells. Out of 90–92 possible number of wet days and 12–13 consecutive number of dry spells during MAM, JJA, SON and DJF seasons, the majority of districts in Uganda, southwestern sub-counties in South Sudan, southwestern zones in Ethiopia and highlands of western and Nyanza counties in Kenya observed the highest number of wet days (50–70 days) and lowest consecutive dry spells (0–1 spells). The highest mean value of wet days (dry spells) of all pixels in the IGAD region are less than 10 days (more than 1 dry spell) with 8 dry spells as highest consecutive spells during MAM. These results are consistent with the earlier findings (Gitau et al., 2013) while studying wet and dry spells over Equatorial Eastern Africa. Uganda (48,31,43 days) followed by South Sudan (33,47,35 days) are countries with the highest number of wet days and less dry spells during MAM, JJA and SON respectively. Whereas, MAM in Sudan, JJA in Somalia and SON in Djibouti observed lowest wet days. The last decade (2011 − 2010) observed the highest number of wet days (lowest dry spells) compared to 1980s,1990s and 2000s decades. These results are in close agreement with finding by Ayugi et al., (2021) that focused on comparison of CMIP6 and CMIP5 models in simulating extreme precipitation over East Africa. The analysis of all pixels over most five potential agricultural production areas of El Gadaref state in Sudan, Upper Nile state in South Sudan, Arsi zone in Ethiopia, Trans Nzoia County in Kenya, Arua district in Uganda shows Arua and Trans Nzoia observed highest number of wet days (lowest dry spells) across MAM and SON, while El Gadaref, Upper Nile and Arsi are the most areas with highest number of wet days (lowest dry spells). This show JJA is main rainy and food crops production season over Sudan, Ethiopia and South Sudan, While, MOM and SON are the most important food crops production season over Kenya and Uganda. These findings are closely in agreement with earlier findings (Kim et al., 2020) (Maidment et al. 2015) while evaluating the CMIP6 multi-model ensemble for climate extreme indices. The mean state of variability in wet and dry spells shows all areas and countries with highest number of wet days (lowest dry spells) observed the lowest variability in wet days (highest variability in dry spells). Uganda and South Sudan are the two countries observed with lowest variability on wet days (highest variability in dry spells). This may be consistence with observed variability in total rainfall over East Africa (Mutai & Ward, 2000). Arid and desert climatic zones in the region observed less than 1% variability on wet days (100% variability in dry spells). The variability in East African rainfall patterns historically has been linked to the influences of the atmospheric phenomena over Indian Ocean ( Saji and Yamagata, 2003), ( Chang'a et al., 2020), the Pacific Ocean( Ogallo, 1988), ( Indeje et al., 2000) and the high variance of westerly winds over the central equatorial Indian Ocean especially during October–November( Hastenrath, 2007, Nicholson, 2017). Probability of wet/dry days and spells exceeding 7, 14,21,28 consecutive wet days and 1,2,3 and 4 consecutive dry spells revealed the region and countries with more than 35 wet days and 5 dry spells observed more than 80% probability of exceeding of wet days (1–5% probability of exceeding of dry spells). Hence, everywhere in South Sudan and Uganda, most parts of Ethiopia, Lake Victoria basin and highlands of western Kenya observed 90–100% of probability of exceeding 7 and 14 days (1 and 2 spells) during MAM, JJA and SON. A study by Muluneh Sorecha & Bayissa, (2017) found about 99 and 80% probability of dry spell lengths of 5 and 7 days in Babile and Haramaya districts of Eastern Hararghe, Ethiopia. The patterns of probability continued in shrinking when examined under 21 and 28 days, 3 and 4 spell thresholds. Arid and desert regions in Sudan, ASALs in Kenya and Somalia, southeastern Ethiopia, most parts of Eritrea and Djibouti observed less than 5% of probability under 7, 14,21,28 consecutive days (1,2,3,4 spells) thresholds. Most parts of South Sudan, Sudan and Uganda at 1 mm thresholds experienced a decreased number of wet days in the 1980s, while last decade (2011–2020) experienced an increase in wet days everywhere in the region during MAM and JJA. This patterns in line with the study by Wainwright et al., (2019), which reported that decline in the Eastern African Long Rains from the 1980s to late 2000s. However, northern parts of South Sudan, almost everywhere in Sudan experienced more than 10 days of decreased wet days during OND. Most parts of Uganda experienced decreased wet days specially in the last 10 years(decade). On the other hand, most parts of Uganda, South Sudan, southern and central Sudan, central and northern Ethiopia experienced decreased dry spells during MAM and JJA seasons, whereas northern parts of Ethiopia and South Sudan, southern parts of Sudan and northeastern Kenya experienced increased dry spells. The 1981–1990 decade experienced increased dry spells, while decreased during 2011–2020. This in agreement with studies by Nicholson, (2017), who reported decline in rainfall over East during MAM season in recent decades
At regional level, the trend of wet days and dry spells of all pixels over the IGAD region observed increased in wet days and decreased in dry spells across all MAM, JJA, SON and DJF seasons. Similarly, at national or IGAD countries member states, the trend of wet days and dry spells experienced slight increase in wet days (decreased dry spells) during JJA and SON, increased (decreased dry spells) over Uganda and South Sudan during MAM, while signal have no clear changes over Sudan, Eritrea, Djibouti and Somalia during MAM season. At sub-national level, the trend of wet days and dry spells over El Gadaref state, Upper Nile state, Arsi zone, Tranz Nzoia county and Arua district confirmed the signals at countries levels with increased wet days and decreased dry spells especially during 2011–2020 decade. The findings may support Eastern African Paradox’ rainfall decline due to shorter not less intense long Rains(Wainwright et al., 2019).The patterns of wet days and dry spells in this study explain the decreased trend of total rainfall in 1980s and 1990s which recovered in recent years (2011–2020), due to increase in number of wet days and decreased dry spells at regional, national and subnational levels.
Acknowledgements: This paper is part of a PhD work at the University of Nairobi, Kenya, Faculty of Science & Technology, Department of Earth & Climate Sciences. Also, author wish to acknowledge Climate Hazard Center, UC Santa Barbara for open access Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS)for archiving and open access to the CHIRPS data
Funding: Not applicable, however the authors wish to acknowledge the support by Intra-ACP Climate Services and Related Applications (ClimSA) support to IGAD climate Prediction and Application Centre (ICPAC) for the funds promised for the open access publication of this research.
Conflict of interest: All authors declare no competing interests
Consent to participate: All authors consent to participate
Consent for publication: All authors consent to publish this work
Data Availability: The secondary datasets generated during analysis are available via request. The primary data are open access from Index of /products/CHIRPS-2.0/global_daily/netcdf/p05 (ucsb.edu)
Code availability: We used R-Package Climate Data Tool (CDT) codes
Author contribution: All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Paulino Omoj. Omay. The first draft of the manuscript was written by Paulino Omoj. Omay supervised by Nzioka J. Muthama , Oludhe Christopher , Josiah M. Kinama . Final manuscript version reviewed by Guleid Artan and Zachary Atheru . All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Preprint server: I have not submitted this manuscript to any preprint server before submitting to Environmental Science and Pollution Research Journal.
- Anyah, R. O., & Qiu, W. (2012). Characteristic 20th and 21st century precipitation and temperature patterns and changes over the Greater Horn of Africa. International Journal of Climatology. https://doi.org/10.1002/joc.2270
- Ayugi, B., Zhihong, J., Zhu, H., Ngoma, H., Babaousmail, H., Rizwan, K., & Dike, V. (2021). Comparison of CMIP6 and CMIP5 models in simulating mean and extreme precipitation over East Africa. International Journal of Climatology, 41(15), 6474–6496. https://doi.org/10.1002/joc.7207
- Bärring, L., Holt, T., Linderson, M. L., Radziejewski, M., Moriondo, M., & Palutikof, J. P. (2006). Defining dry/wet spells for point observations, observed area averages, and regional climate model gridboxes in Europe. Climate Research, 31(1), 35–49. https://doi.org/10.3354/cr031035
- Barron, J., Rockström, J., Gichuki, F., & Hatibu, N. (2003). Dry spell analysis and maize yields for two semi-arid locations in east Africa. Agricultural and Forest Meteorology. https://doi.org/10.1016/S0168-1923(03)00037-6
- Bouagila, B., & Sushama, L. (2013). On the current and future dry spell characteristics over Africa. Atmosphere. https://doi.org/10.3390/atmos4030272
- Camberlin, P., Moron, V., Okoola, R., Philippon, N., & Gitau, W. (2009). Components of rainy seasons’ variability in Equatorial East Africa: Onset, cessation, rainfall frequency and intensity. Theoretical and Applied Climatology, 98(3–4), 237–249. https://doi.org/10.1007/s00704-009-0113-1
- Cattani, E., Merino, A., Guijarro, J. A., & Levizzani, V. (2018). East Africa Rainfall trends and variability 1983-2015 using three long-term satellite products. Remote Sensing, 10(6). https://doi.org/10.3390/rs10060931
- Chemura, A., Schauberger, B., & Gornott, C. (2020). Impacts of climate change on agro-climatic suitability of major food crops in Ghana. PLoS ONE, 15(6), 1–21. https://doi.org/10.1371/journal.pone.0229881
- Clark, C. O., Webster, P. J., & Cole, J. E. (2003). Interdecadal variability of the relationship between the Indian Ocean zonal mode and East African coastal rainfall anomalies. Journal of Climate, 16(3), 548–554. https://doi.org/10.1175/1520-0442(2003)016<0548:IVOTRB>2.0.CO;2
- Endris, H. S., Lennard, C., Hewitson, B., Dosio, A., Nikulin, G., & Artan, G. A. (2019). Future changes in rainfall associated with ENSO, IOD and changes in the mean state over Eastern Africa. Climate Dynamics, 52(3–4), 2029–2053. https://doi.org/10.1007/s00382-018-4239-7
- Gitau, W., Ogallo, L. A., & Mutemi, J. N. (2008). Intraseasonal characteristics of wet and dry spells over Kenya. J.Kenya Meteorol. Soc.
- Gitau, W., Ogallo, L., Camberlin, P., & Okoola, R. (2013). Spatial coherence and potential predictability assessment of intraseasonal statistics of wet and dry spells over Equatorial Eastern Africa. International Journal of Climatology, 33(12), 2690–2705. https://doi.org/10.1002/joc.3620
- Gudoshava, M., Misiani, H. O., Segele, Z. T., Jain, S., Ouma, J. O., Otieno, G., Anyah, R., Indasi, V. S., Endris, H. S., Osima, S., Lennard, C., Zaroug, M., Mwangi, E., Nimusiima, A., Kondowe, A., Ogwang, B., Artan, G., & Atheru, Z. (2020a). Projected effects of 1.5 °C and 2 °C global warming levels on the intra-seasonal rainfall characteristics over the Greater Horn of Africa. Environmental Research Letters. https://doi.org/10.1088/1748-9326/ab6b33
- Gudoshava, M., Misiani, H. O., Segele, Z. T., Jain, S., Ouma, J. O., Otieno, G., Anyah, R., Indasi, V. S., Endris, H. S., Osima, S., Lennard, C., Zaroug, M., Mwangi, E., Nimusiima, A., Kondowe, A., Ogwang, B., Artan, G., & Atheru, Z. (2020b). Projected effects of 1.5 °C and 2 °C global warming levels on the intra-seasonal rainfall characteristics over the Greater Horn of Africa. Environmental Research Letters, 15(3), 34037. https://doi.org/10.1088/1748-9326/ab6b33
- Haile, G. G., Tang, Q., Hosseini-Moghari, S. M., Liu, X., Gebremicael, T. G., Leng, G., Kebede, A., Xu, X., & Yun, X. (2020). Projected Impacts of Climate Change on Drought Patterns Over East Africa. In Earth’s Future (Vol. 8, Issue 7). https://doi.org/10.1029/2020EF001502
- Hutchinson, S., Gomes, C., Alleyne, D., & Phillips, W. (2013). An assessment of the economic and social impacts of climate change on the tourism sector in the Caribbean. 41.
- IPCC. (2022). Climate Change 2022 - Mitigation of Climate Change - Working Group III. Cambridge University Press, 1454. https://www.ipcc.ch/site/assets/uploads/2018/05/uncertainty-guidance-note.pdf.%0Awww.cambridge.org
- Kebede, A., Diekkrüger, B., & Edossa, D. C. (2017). Dry spell, onset and cessation of the wet season rainfall in the Upper Baro-Akobo Basin, Ethiopia. Theoretical and Applied Climatology. https://doi.org/10.1007/s00704-016-1813-y
- Kim, Y. H., Min, S. K., Zhang, X., Sillmann, J., & Sandstad, M. (2020). Evaluation of the CMIP6 multi-model ensemble for climate extreme indices. Weather and Climate Extremes, 29, 100269. https://doi.org/10.1016/j.wace.2020.100269
- Korbinian Breinl, Giuliano DiBaldassarre, MaurizioMazzoleni, D. L. (2020). Extreme dry and wet spells face changes in their duration and timing Environmental Research Letters Extreme dry and wet spells face changes in their duration and timing. Enviroment Research Letter, 15. https://doi.org/10.1088/1748-9326/ab7d05
- Kumar, U., Singh, P., & Boote, K. J. (2012). Effect of Climate Change Factors on Processes of Crop Growth and Development and Yield of Groundnut (Arachis hypogaea L.). In Advances in Agronomy (Vol. 116). https://doi.org/10.1016/B978-0-12-394277-7.00002-6
- Lotfie, A. Y., Abdelrahman, A. K., Faisal, M. E.-H., Ahmed, M. A., Hussain, S. A., Abdelhadi, A. W., Yasunori, K., & Imad-eldin, A. A.-B. (2018). Rainfall variability and its implications for agricultural production in Gedarif State, Eastern Sudan. African Journal of Agricultural Research, 13(31), 1577–1590. https://doi.org/10.5897/ajar2018.13365
- Mandapaka, P. V., Qin, X., & Lo, E. Y. M. (2016). Seasonal and interannual variability of wet and dry spells over two urban regions in the western Maritime Continent. Journal of Hydrometeorology. https://doi.org/10.1175/JHM-D-15-0100.1
- Muluneh Sorecha, E., & Bayissa, B. (2017). Dry spell length analysis for crop production using Markov-Chain model in Eastern Hararghe, Ethiopia. Journal of Degraded and Mining Lands Management, 04(04), 891–897. https://doi.org/10.15243/jdmlm.2017.044.891
- Mutai, C. C., & Ward, M. N. (2000). East African rainfall and the tropical circulation/convection on intraseasonal to interannual timescales. Journal of Climate, 13(22), 3915–3939. https://doi.org/10.1175/1520-0442(2000)013<3915:EARATT>2.0.CO;2
- Mwangi, E., Wetterhall, F., Dutra, E., Di Giuseppe, F., & Pappenberger, F. (2014). Forecasting droughts in East Africa. Hydrology and Earth System Sciences, 18(2), 611–620. https://doi.org/10.5194/hess-18-611-2014
- Ndehedehe, C. E., Ferreira, V. G., Adeyeri, O. E., & Correa, F. M. (2023). Global assessment of drought characteristics in the Anthropocene Resources , Environment and Sustainability Global assessment of drought characteristics in the Anthropocene. Resources, Environment and Sustainability, 12(January), 100105. https://doi.org/10.1016/j.resenv.2022.100105
- Nicholson, S. E. (2017a). Climate and climatic variability of rainfall over eastern Africa. Reviews of Geophysics. https://doi.org/10.1002/2016RG000544
- Nicholson, S. E. (2017b). Climate and climatic variability of rainfall over eastern Africa. Reviews of Geophysics, 55(3), 590–635. https://doi.org/10.1002/2016RG000544
- Olatunde, A. F., & Love, J. O. (2018). Recent Changes in Onset and Cessation Dates of Rainfall and their Effects on Farming Activities in Sub-Urban Areas of Lokoja. International Journal of Social Sciences., 12(2), 34–54.
- Omay, P. O., Christopher, O., Atheru, Z., & Atheru, Z. (2022). Changes and Variability in rainfall onset , cessation and length of rainy season in the IGAD region of Eastern Africa.
- Omoj, P., Ogallo, L., Oludhe, C., & Gitau, W. (2016). Temporal and Spatial Characteristics of the June-August Seasonal Rainfall and Temperature over South Sudan. Journal of Meteorology and Related Sciences. https://doi.org/10.20987/jmrs.5.08.2016
- Osima, S., Indasi, V. S., Zaroug, M., Endris, H. S., Gudoshava, M., Misiani, H. O., Nimusiima, A., Anyah, R. O., Otieno, G., Ogwang, B. A., Jain, S., Kondowe, A. L., Mwangi, E., Lennard, C., Nikulin, G., & Dosio, A. (2018). Projected climate over the Greater Horn of Africa under 1.5 °c and 2 °c global warming. Environmental Research Letters, 13(6). https://doi.org/10.1088/1748-9326/aaba1b
- Peel, M. C., Finlayson, B. L., & McMahon, T. A. (2007). Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth System Sciences, 11(5), 1633–1644. https://doi.org/10.5194/hess-11-1633-2007
- Philippon, N., Camberlin, P., Moron, V., & Boyard-Micheau, J. (2015a). Anomalously wet and dry rainy seasons in Equatorial East Africa and associated differences in intra-seasonal characteristics. Climate Dynamics, 45(7–8), 1819–1840. https://doi.org/10.1007/s00382-014-2436-6
- Philippon, N., Camberlin, P., Moron, V., & Boyard-Micheau, J. (2015b). Anomalously wet and dry rainy seasons in Equatorial East Africa and associated differences in intra-seasonal characteristics. Climate Dynamics. https://doi.org/10.1007/s00382-014-2460-6
- Rameshwaran, P., Bell, V. A., Davies, H. N., & Kay, A. L. (2021). How might climate change affect river flows across West Africa? Climatic Change, 169(3–4), 1–28. https://doi.org/10.1007/s10584-021-03256-0
- Ratan, R., & Venugopal, V. (2013). Wet and dry spell characteristics of global tropical rainfall. Water Resources Research, 49(6), 3830–3841. https://doi.org/10.1002/wrcr.20275
- Recha, C. W., Makokha, G. L., Traore, P. S., Shisanya, C., Lodoun, T., & Sako, A. (2012). Determination of seasonal rainfall variability, onset and cessation in semi-arid Tharaka district, Kenya. Theoretical and Applied Climatology. https://doi.org/10.1007/s00704-011-0544-3
- Schilling, J., Hertig, E., Tramblay, Y., & Scheffran, J. (2020). Climate change vulnerability, water resources and social implications in North Africa. Regional Environmental Change, 20(1). https://doi.org/10.1007/s10113-020-01597-7
- Schreck, C. J., & Semazzi, F. H. M. (2004). Variability of the recent climate of eastern Africa. International Journal of Climatology. https://doi.org/10.1002/joc.1019
- Segele, Z. T., & Lamb, P. J. (2005). Characterization and variability of Kiremt rainy season over Ethiopia. Meteorology and Atmospheric Physics, 89(1–4), 153–180. https://doi.org/10.1007/s00703-005-0127-x
- Seneviratne, S. I., Nicholls, N., Easterling, D., Goodess, C. M., Kanae, S., Kossin, J., Luo, Y., Marengo, J., Mc Innes, K., Rahimi, M., Reichstein, M., Sorteberg, A., Vera, C., Zhang, X., Rusticucci, M., Semenov, V., Alexander, L. V., Allen, S., Benito, G., … Zwiers, F. W. (2012). Changes in climate extremes and their impacts on the natural physical environment. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the Intergovernmental Panel on Climate Change, 9781107025, 109–230. https://doi.org/10.1017/CBO9781139177245.006
- Service, N. W. (2001). Using Normalized Climatological Anomalies to Rank Synoptic-Scale Events Objectively. 2426–2442.
- Shaxson, T. F., Barber, R. G., & Food and Agriculture Organization of the United Nations. (2003). Optimizing soil moisture for plant production : the significance of soil porosity. In FAO soils bulletin,.
- Shongwe, M. E., van Oldenborgh, G. J., van den Hurk, B., & van Aalst, M. (2011). Projected changes in mean and extreme precipitation in Africa under global warming. Part II: East Africa. Journal of Climate. https://doi.org/10.1175/2010JCLI2883.1
- Shukla, S., McNally, A., Husak, G., & Funk, C. (2014). A seasonal agricultural drought forecast system for food-insecure regions of East Africa. Hydrology and Earth System Sciences. https://doi.org/10.5194/hess-18-3907-2014
- Thomson, M. C., Muñoz, Á. G., Cousin, R., & Shumake-Guillemot, J. (2018). Climate drivers of vector-borne diseases in Africa and their relevance to control programmes. Infectious Diseases of Poverty, 7(1), 81. https://doi.org/10.1186/s40249-018-0460-1
- Thornton, P. K., Jones, P. G., Alagarswamy, G., & Andresen, J. (2009). Spatial variation of crop yield response to climate change in East Africa. Global Environmental Change. https://doi.org/10.1016/j.gloenvcha.2008.08.005
- Verschuren, D., Lalrd, K. R., & Cumming, B. F. (2000). Rainfall and drought in equatorial east Africa during the past 1,100 years. Nature. https://doi.org/10.1038/35000179
- Wainwright, C. M., Black, E., & Allan, R. P. (2021). Future Changes in Wet and Dry Season Characteristics in CMIP5 and CMIP6 simulations. Journal of Hydrometeorology, September. https://doi.org/10.1175/jhm-d-21-0017.1
- Wainwright, C. M., Marsham, J. H., Keane, R. J., Rowell, D. P., Finney, D. L., Black, E., & Allan, R. P. (2019). ‘Eastern African Paradox’ rainfall decline due to shorter not less intense Long Rains. Npj Climate and Atmospheric Science. https://doi.org/10.1038/s41612-019-0091-7
- Williams, A. P., & Funk, C. (2011). A westward extension of the warm pool leads to a westward extension of the Walker circulation, drying eastern Africa. Climate Dynamics, 37(11–12), 2417–2435. https://doi.org/10.1007/s00382-010-0984-y
- Yapo, A. L. M., Diawara, A., Kouassi, B. K., Yoroba, F., Sylla, M. B., Kouadio, K., Tiémoko, D. T., Koné, D. I., Akobé, E. Y., & Yao, K. P. A. T. (2020). Projected changes in extreme precipitation intensity and dry spell length in Côte d’Ivoire under future climates. Theoretical and Applied Climatology. https://doi.org/10.1007/s00704-020-03124-4
- Zhang, W., Brandt, M., Tong, X., Tian, Q., & Fensholt, R. (2018). Impacts of the seasonal distribution of rainfall on vegetation productivity across the Sahel. Biogeosciences. https://doi.org/10.5194/bg-15-319-2018
No competing interests reported.