Figures
Abstract
Distribution and trends of rainfall reveal spatial and temporal variability that have a paramount effect on the life and livelihood of small-holder farmers. This study aimed to analyze spatial variability and temporal trends of rainfall distribution across the three Agro-Ecological Zones (AEZs) of East Guji. Time series gridded daily rainfall data (1990–2020) were collected from the Ethiopian Meteorological Institution. Different descriptive statistics, trend tests: Man Kendal and Sen’s slope estimator, Inverse Distance Weighted Index and Precipitation Concentration Index (PCI) was used in the study. The finding demonstrated that altitude and rainfall decrease as one advances from the western (highland) to the eastern (lowland) direction in the study area where the highest rainfall was recorded in Solemo (highland) and the least in Negele (lowland).The study showed that as altitude increases annual rainfall also increases and rainfall variability decreases. Similarly the mean length of the growing season declines as one advance from the highlands to the lowlands. The PCI of the lowlands, midlands, and highlands AEZs was 19%, 17%, and 12% respectively. The PCI showed that those highlands had moderately concentrated rainfall but both lowlands, and midlands, had an irregular distribution of rainfall. The Coefficient of Variation (CV) indicated that highland areas had moderate variability in rainfall in all seasons except winter. In contrast, the low and midlands had shown high variability of rainfall (>30%) in all seasons. From a seasonal perspective, both CV and PCI revealed that the winter season showed more variability than others. Moreover, a significant increasing trend of annual rainfall was observed in the highlands AEZs (Bore 15.3mm/year and Solemo14.6mm/year), lowland AEZs (Chembe 10.9mm/year, Dawa 8mm/year and Bitata 7.8mm/year) as well as midland AEZs (Kercha 14.5mm/year) at a significant level of 5%. Therefore, strategies should be designed to use additional water resources for irrigation; and provide short-cycle grown and drought-resistant crops in the rest of the midlands and lowlands AEZs.
Citation: Sahilu M, Tekalign S, Mohammed Y, Sishaw T, Kedir H, Asfaw S (2024) Spatiotemporal trends and variability of rainfall across agro-ecologies in East Guji Zone, Southeast Ethiopia. PLOS Clim 3(3): e0000361. https://doi.org/10.1371/journal.pclm.0000361
Editor: Ahmed Kenawy, Mansoura University, EGYPT
Received: March 18, 2023; Accepted: January 14, 2024; Published: March 21, 2024
Copyright: © 2024 Sahilu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Since the data for the study was obtained from the third body the Ethiopian Metrological Institute with an agreement for using the data for the research purpose. Therefore, you can obtain the data from the organization using their e-mail address nma.datausers@mail.com.
Funding: We would like to acknowledge the Hossana College of Education and Haramaya University for funding the study. With collaboration both institution provided financial support for the corresponding author MS for the her completion of PHD work. The funders had no role in study design, data collection, analysis and interpretation, the writing of this manuscript, and in the decision to submit it for publication.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Global climate change and climate variability increased erratic rainfall in many parts of the world [1]. Africa is a continent highly pretentious about the problem. Climate change caused increased rainfall variability in much of Sub-Saharan Africa (SSA) [2]. Rainfall variability has a significant effect on the welfare of the household and national production of agricultural-based economies [3]. The future social-economic development of the African community is determined by water acquired from highly variable rainfall [4]. In most parts of SSA, the amount of precipitation will decline by about 20% [5] while its variability will increase [6].
The remarkable spatial variability of rainfall will continue in the future [7–9]. Several factors from local to global level influence the distributions of rainfall in Ethiopia. In Ethiopia, a spatial variation of rainfall is observed across the different AEZs.The local variability of rainfall in Ethiopia is articulated by multifaceted topography which ranges from 4620m above sea level at Ras Dashen Mountain to the lowest at Afar Depression (Kobar Sink) 120m below sea level [10]. There is also temporal variability of rainfall from days to seasons and decades. Particularly, rainfall variability has great influences on the life and livelihood of smallholder farmers in Ethiopia. Ethiopia is the most vulnerable to the effects of rainfall variability. Since agriculture is the backbone of the Ethiopian economy it contributes about 40% of the GDP, 80% of total employment, and 90% of exports. Besides, 79% of the population depends on rain-fed agriculture for their income [11,12].
Besides global factors: such as the El Nino–Southern Oscillation (ENSO), Indian Ocean Dipole (IOD) and the Inter Tropical Convergence Zone (ITCZ) influenced the seasonal rainfall variability over Ethiopia [13,14]. El Nino weakens the high-pressure air mass in the South Indian and South Pacific Oceans that causes rainfall in Ethiopia [15]. Complete models on the impact of El Nino on Ethiopian climate variability are lacking because of the existence of different AEZs which resulted in many microclimate and a lack of meteorological data [16].
So far several studies have found that-complex trend of rainfall distribution in various parts of the country. [17–19] reported significant declines in the annual and summer (kiremt) rainfall totals in the eastern, southern, and southwestern, northern, north-central, north-western, and western parts of the country. However the study by Rosell & Holmer, [19] also indicated an increasing trend of annual rainfall at some locations in eastern Ethiopia as well as an increase in annual and summer (kiremt); a decrease in spring (belg) rainfall in central Ethiopia. Bayable et al. [20] reported the annual rainfall showed a non-significant decreasing trend over Western Harerge. Habte et al. [21] reported a declining trend of annual and March-May rainfall (long rain) or spring in the Guji Zone.
Further, the national-level study by Wagesho et al. [22] showed varied rainfall patterns in different parts of the country. Likewise, Mohammed et al. [23] reported a significantly increasing trend of summer rainfall in some of the stations and a declining tendency in spring rain in all studied stations in Southern Wollo. Abegaz & Mekoya, [24] also reported a significant decreasing trend in spring (belg) and winter (bega) rainfall and increasing annual and summer (Kiremt) rainfall in central Ethiopia. Likewise, Alemayehu et al. [25] showed significantly increasing trends of annual and seasonal rainfall totals in the western part of Ethiopia. Similarly, Belay et al. [26] reported an increasing trend of annual, summer (Kiremt) and winter (Bega) rainfall whereas the spring (belg) season rainfall showed a significant decreasing trend in southern Ethiopia. A recent study conducted by Worku et al. [27] demonstrates rainfall showed a significant increasing trend during August, October, and November and was extremely variable during December, January, and February in Borena, Southern Ethiopia.
Although topography is prominent in determining the distribution of rainfall in Ethiopia most of the existing studies were not AEZs centered except the study in Welayita [28] in North Western Ethiopia [29] in Northern Shewa [30]. Since large-scale studies highly mask the spatial and temporal variability of rainfall, further AEZ based local-level investigation of variability and trends of rainfall events were suggested by many authors, [27,31–33]. Thus, the ongoing issue of rainfall variability requires current investigation across the AEZs due to the existence of diverse micro climates in the country. This study gives prior emphasis on the influences of AEZs in the spatial variability and temporal trends of rainfall distributions in Guji Zone.
Given the high dependence on rain-fed agriculture, it is important to precisely characterize the spatial variability, temporal trends, onset and secession, and distribution of rainfall in each AEZ. Hence, there is no previous investigation that has examined the spatial variability and temporal trends of spring and autumn rainfall in the three AEZs of the study area. The study fills gaps in knowledge of the spring (belg) and autumn (meher) seasons, spatial variability and temporal trends of rainfall in the different AEZs of the Guji zone. Thus, identifying rainfall variability at this micro-scale level provides relevant information which enables stakeholders to design context-specific adaptation strategies in different AEZs. Therefore, the objective of this study is to analyze spatial variability, temporal trends, onset and cessation, and distribution of spring and autumn rainfall across the three AEZs using long-time series rainfall data.
2. Descriptions of the study area
2.1 Location of the study area
This study was conducted in the East Guji Zone, in the Southern part of Oromia Regional State, Ethiopia. Geographically, the East Guji Zone is located between 4° 30′ N—6° 30′ N latitude and 38° 15′ E- 40° 5′ E longitudes (Fig 1), and covers a total area of about 18,577 Km2. It shares a border with Gedeo and Sidama on the North, Somali Regional State in the South, Bale Mountain on the East and Borena Zone on the West. The northern and eastern neighbors of the Eastern Guji zone mainly have highland AEZs that experience wetter microclimate [21]. The southern and the western adjacent land of the study area holds the lowland AEZs that experience relatively drier climate conditions [27]. Borena is mainly lowland which embraces a semiarid and arid climate and recently has been highly pretension by drought [34].
Source: Projection WGS_1984_ UTM _Zone_37°. The map is produced using ArcGIS 10.8: Ethiopia GeoPortal -Free access to Shape file. (https://ethiopia.africageoportal.com).
The total population of the area as projected by [35] was 1,499,013 in 2017. The Guji Oromo are the dominant Ethnic group which represented 95.57%, the Amhara https://en.wikipedia.org/wiki/Amhara_people 2.43%, and the Somali 2% [36]. They are predominantly pastoralists, agro-pastoralists and farmers [37]. Coffee is an important cash crop in the Guji zone [38].
2.2. The biophysical characteristics of East Guji
The study area is situated in the southern part of Ethiopia, which experienced bimodal rainfall cycles in spring (March–May) with a peak in April/May and autumn (October–November) [14].The area is characterized by a bimodal rainfall pattern where the long rainy season (Arfansa) occurs from March to May and covers about 48% of the annual rainfall whereas the short rainy season (Hagaya) extends from September to November which shares about 36% of the annual rainfall. Thus, together spring (belg or arfansa) and autumn (meher or Hagaya) cover around 84% of rainfall in lowlands and midlands AEZs of East Guji (Fig 4). Besides, the mean rainfall in the midlands and lowlands for winter and summer was 6% and 10% respectively.
Therefore, four seasons have been identified in the study area namely; the major dry season winter or bega (Dec- Feb), the long rainy period named spring or belg (Mar-May), summer or kiremt (June-August) and the short rainy period is autumn or meher (Sept-Nov).The winter (Bega) and summer (Kiremt) are relatively dry seasons except for the highland area which receives better rainfall in the summer season. Thus, the highlands part of Eastern Guji receives almost the corresponding amount of rainfall in the three seasons: spring (34%), summer (30%), and autumn (29%).
The elevation of the study area ranges from 632m to 2960m above sea level, and is categorized under the three AEZs. They are: lowlands (Kola) 500m-1500m, midland (Woina Dega) 1500m-2300m, and highland (Dega) 2300m-3200m based on [39] which; covers about 27.3%, 51.5%, and 21.2% of the total area, correspondingly. The mean annual rainfall in the Guji Zone for lowlands, midlands, and highlands is 747mm, 932mm, and 1232mm respectively (Table 6). The mean temperature records were 22.1°c, 19.4°c, and 15.8°c in the lowlands, midlands and highlands AEZs correspondingly.
As a result of a multiplicity of AEZs and soil set up in the area; a variety of vegetation cover: dense natural forests, planted trees, acacia, and scrubs are major land cover in Guji. The dense forest resource of the Eastern Guji Zone comprises the Bore-Anferara National Forest Priority Area and the Anferara-Wadera National Forest Priority area. These areas are among the 58 Priority Forest Areas (PFAs), which have been so designated to protect biodiversity and conserve forest resources in Ethiopia.
3. Materials and methods
3.1. Research design
The study followed a quantitative research design. Quantitative research is selected because the study is grounded on observed (recorded) statistical data. A standardized procedure was followed to analyze the data. Thus, the output of the study can be replicated and generalized to a similar study. It is a non-experimental, descriptive and observational research. Non-experimental quantitative research design refers to there is no manipulation of variables in the study. The research described the spatial variability and temporal trends of rainfall across the three AEZs in the Eastern Guji. Gridded rainfall data (0.04° by 0.04°) of the sample stations were taken from Ethiopian Meteorological Institutions (EMI) to detect, quantify and generalize variability and trends of rainfall over the studied period (1990–2020). The gridded rainfall data was chosen over meteorological station data since the latter has more missing values and short time series data. Gridded time series rainfall data of eleven stations were assessed quantitatively to analyze spatial variability, temporal trends, onset and cessation and distribution of rainfall. Some descriptive statistics such as, statistical, and spatial analysis tools, were used in the study. Finally, results were systematized on tables and figures followed by proper discussions.
3.2. Sampling procedure
Probability and non-probability sampling procedure was used to select an appropriate sample for the study. Guji Zone was selected through purposive sampling as it represents three AEZs such as lowlands, midlands and highlands which can be used to show the spatial dimensions of rainfall variability in the area. A stratified random sample procedure was applied to select the required groups or strata for the study. AEZs are considered as strata for the study. A random sample is selected from each stratum based on the percentage of each AEZs embraced. Lowland, midland and highland AEZs cover about 27.5%, 51.2% and 21.3% of the total area, respectively. By considering the distributions of AEZ in the study area, five, four, and two stations from midland, lowland, and highland AEZs respectively were randomly assigned. Accordingly, the randomly chosen stations (grid cells) include: Chembe, Dawa, Negele, Bitata, Oddo Shakiso, Adola (Kibre Mengist), Harekello (Goro Dola), Wadera, Kercha, Solemo (Uraga), and Bore (Table 1).
3.3. Methods of data analysis
Several methods exist to estimate the variability of the date of rain onset and cessation of the growing season. For the study onset and cessation date was determined using Walter’s formulation as modified by [40]. The method was selected because it used a threshold value of 51mm accumulated rainfall to determine the onset. Since about 80% of the locality comprises midland and lowland. Relatively, low rainfall coupled with high temperature and evapotranspiration distinguishes the locality. Consequently, ample precipitation is required for planting in the area. Thus, the method is minimized considering false onset days which result in early departure of the rainy period. By this method, the onset date of the rain is the time a place receives an accumulated amount of rainfall over 51mm and not the first day the rain falls. Cessation date: is the date after which not 51mm of rain is expected. The method is expressed as: [41].
Eq (1)
Where, DM = number of days in the month containing the onset; TM = total rainfall for the month in which accumulated rainfall exceeds 51mm; AP = accumulated rainfall of previous months just before the month in reference; 51mm = the threshold of rainfall for both Onset/End months. Where such an onset date was followed by rainfall amount less than 51 mm, the next rain date that is up to 51mm or more will be chosen [42]. The length of the growing season, under rain-fed conditions, is defined as the period from the date of the onset of the rainy season to its cessation.
The study employed Mann-Kendal’s test and Sen’s slope estimators to identify long-term rainfall trends and the rate of change in monthly, seasonal and annual time steps in three AEZs of the study area respectively. The Man Kendall test is highly recommended for general use by the WMO [43]. It is chosen for trends test in a time series data since it doesn’t request for normality or linearity, is less sensitive to outliers or is robust against the influence of extremes [44,45].
When the Z value exceeds either of the confidence limit lines, it shows a significant trend at a given significance level (< 0.05). Hence, Ho is rejected and in place, H1 is accepted. Where, n is the number of data points; xj and xi are the time series observations in year j and i, j>1.The Mann-Kendall statistic S of the series x is obtained by the following Equation:
Eq (2)
Eq (3)
Sign (Xj–Xi) means the individual sign capability that takes on the values [1, 0, or -1]. A positive S value indicates an ever-increasing trend, and a negative value indicates a downward trend. Compute the variance of S as follows:
Eq (4)
Where: n is the data point’s number, g is the zero difference between compared values number, tp is the number of data points in the pth group. A standardized measure of test statistics (Zmk), determined using the following equation:
Eq (5)
The magnitude of the change in the time series was detected by a simple non-parametric procedure developed by [46]. This test computes the linear rate of change (slope) and the intercept as shown in Sen’s method [46]. The magnitude of the trend is calculated by using Sen’s slope estimator in the following equation:
Eq (6)
Where β is Sen’s slope estimate β > 0 indicates an upward trend in a time series. Otherwise the data series presents a downward trend during the period.
The precipitation concentration index (PCI) defined by [47] is a powerful indicator of temporal precipitation distribution. PCI is generally used for evaluating seasonal precipitation changes to investigate the heterogeneity of monthly rainfall data. In the study, PCI was used to analyze the monthly and annual variability of precipitation. The calculation is described as follows:
Eq (7)
Where: Pi = the total rainfall of ith month
As indicated by [47], PCI values are categorized as uniform (<10) presents a uniform distribution of rainfall, (11–15) indicates moderate, (16–20) shows irregular, and (> 21) shows a strong irregular monthly rainfall distribution.
Variability of rainfall is computed using the coefficient of variation (CV) [48], Coefficient of variation (CV) provides a measure of year-to-year variation in the data series. As documented by Hare [48], the degree of rainfall variability is classified as high (CV > 30), moderate (20 < CV > 30) and low (CV < 20).
Eq (8)
Where: CV = coefficient of variation σ = standard deviation μ = Mean rainfall (mm)
Inverse distance weighted (IDW) is a spatial analysis tool used to illustrate the spatial trends of observed rainfall. The IDW is a measure between neighboring stations for time series. An inverse distance interpolation is one of the simplest and most popular interpolation techniques. It combines the proximity concept with the gradual change of the trend surface. An inverse distance (ID) weighted interpolation is defined as a spatially weighted average of the sample values within the search neighborhood [49]. The spatial distributions of annual and seasonal rainfall (spring and autumn) were mapped using IDW interpolation in arc GIS.
4. Results and discussions
4.1. Agro ecology-based spatial distribution of rainfall in the study area
The section clarifies the spatial distribution of rainfall across the three AEZs of the Eastern Guji. Accordingly, the highlands (Dega), midlands (Waina Dega), and lowlands (Kola) AEZs account for 27%, 51.5% and 21.2% of the total area respectively (Fig 2). AEZs clearly articulated the spatial distribution of rainfall in the study area. The western part of the study area is the highland (Dega AEZ) where the highest intensity of annual rainfall (1273mm) was recorded in Solemo (Uraga) (Fig 3). The central and eastern part of the study area is mainly midland (Waina Dega) AEZ. The south eastern part of the study area is predominantly lowland (Kola) where the lowest rainfall (624mm) was recorded in Negele (Liben) (Fig 3). As one advances from the western (highlands) to eastward (lowlands) direction in the study area both altitude and rainfall decline (Fig 2). The spatial distribution map of rainfall also denotes that maximum rainfall was recorded in the spring (belg) season followed by autumn (meher). The peak of spring rainfall was recorded around Adola and Shakiso (midlands), autumn rainfall was highest around Chembe (lowland); summer and annual rainfall was greatest in Bore and Solemo in the highland AEZ (Fig 3).
The map is produced using ArcGIS 10.8: Ethiopia GeoPortal—Free access to Shape file. (https://ethiopia.africageoportal.com).
The map is produced using ArcGIS 10.8: Ethiopia GeoPortal- Free access to Shape file. (https://ethiopia.africageoportal.com).
Finally, the spatial variability of rainfall in the study area was expressed through contrasting reality observed across the highlands, midlands, and lowlands AEZs in the study area. Primarily, the onset and cessation also resulted in LGP diminishing from the highlands to the lowlands and midlands AEZs. Single and long LGP was witnessed in the highland AEZ whereas a dual growing period with short LGP was observed in the lowlands and midland AEZs of the study area. Besides, the seasonal distribution of rainfall in the three AEZs of the study area revealed that the highlands received almost the same amount of rainfall in three seasons except winter (Fig 3). In contrast, the lowlands and midlands gain rainfall maxima in the spring and autumn seasons whereas summer and winter are relatively dry seasons (Fig 3). In addition, a significant increasing trend of the annual and autumn rainfall was observed mainly in the highlands, most of the lowlands and in some of the midlands AEZ.
On the other hand, the rest stations in the mid and lowlands AEZs have an insignificant increasing trend of the annual rainfall. Furthermore, relatively higher PCI & CV values were observed in the lowlands (Kola) and midlands (Weynadega). In contrast, it was lower for the highlands (Dega) AEZ. To conclude, 78.8% of the study area is enclosed with the midlands and lowlands which received scarce rainfall. The implication is that midlands and lowlands faced severe problems of variability in the onset and end, short LGP and lack of significant trend on monthly and seasonal rainfall. Consequently, scarcity of rainwater for agriculture prevailed in the mid and low lands AEZ of the study area.
4.2. Variability of onset and cessation of rainfall in the study area
The distribution of variability of the onset and cessation of rainfall and its temporal trends in East Guji from 1990–2020 was discussed in this section. Accordingly, the lowlands and midlands of the study area have two distinct onset and cessation periods in the autumn and spring seasons (Table 2). Thus, they are familiar with dual growing seasons. The average onset date of spring rainfall in the highland AEZs was April 18 and ends on November 2. Hence, the highland AEZ of the study area has a single growing period as well as an onset and cessation period. Presently, the smallholders farmers in the highlands started using the spring season (April-May) for short season growing vegetables such as Potato, Tomato and Onion.They cultivate cereal crops such as Barley, Wheat, Maize, Peas and Bea using the summer and autumn rainfall (June–November).
In the midland AEZ the spring rainfall begins on April 20 and terminates on 24 May, the autumn rainfall starts on October 13 and ceases on November 10.Consequently, the mean length of growing period (LGP) in the midlands for the spring and autumn seasons was 35 and 27 days respectively. Further, spring rainfall arrives on 25 April and halts on May 18 in the lowlands. Besides, autumn rainfall starts on October 15 and ends on November 7 in the lowlands. Moreover, in the lowland LGP were 23 and 16 days for the spring and autumn seasons respectively. Thus, as one moves from highlands AEZs to lowlands and midlands in the study the LGP declines.
CV is a useful measure of variability in the length of the growing season; because particularly in short LGP, relatively small changes in LGP have important consequences. High CV is dominant in areas with relatively short growing seasons, while low CV value corresponds to long growing periods. Similarly, the highland AEZs had long LGP which is 198 days and witnessed low CVs 16% and 6% for the onset and cessation periods respectively. The finding implies the highland of the study area relishes long wet spells from April to November. Hence rainfall distribution in the highlands was regular. However, in the midlands and lowlands practicing rain-fed agriculture with such a short LGP has been puzzling.
The CV for the onset of spring rainfall in lowland and midland was 21% and 19% whereas; for the autumn season was 24% and 20% respectively. The CV for the cessation of spring and autumn rainfall in both low and midland is 33% and 36% respectively. Thus, very high CV (>30) in the cessation of the spring and autumn rainfall together with short LGP was detected in the lowlands and midlands AEZs of the study area. As demonstrated in Table 2, more variability of rainfall was observed in lowlands and midlands in the cessation period than in the onset period. Therefore the ending date for both spring and autumn rainfall is erratic. The contrasting result was obtained by [50] who found the presence of more variability in the onset of the spring (Belg) rains than the cessation period.
Table 3 demonstrates the spatial-temporal trends of the onset and cessation of seasonal rainfall in the Guji Zone. There was no statistically significant trend in the onset of spring season rain in all AEZs of the study area. Yet, an insignificant declining trend was detected at the onset of spring rainfall in most parts of the study area. Besides, an insignificant trend was detected in the cessation date of spring rainfall in all AEZs except the midland (Wadera) where the ending date of the spring rainfall was significantly increased. Nevertheless, a significant declining trend (delay) in the onset of autumn rainfall was observed in the midlands (Adola). Moreover, an insignificant decline in the onset of autumn rainfall was observed in all except in the lowlands (Negele and Bitata). Whereas, a significant increase in the end date of autumn rainfall was observed in the midland (Harekello and Kercha). Except for the above-mentioned stations there was no statistically significant trend in the onset as well as the termination of autumn rainfall in all AEZs of the study area.
The implication is that both spring and autumn seasons’ rainfall began and terminated with divergent dates. Therefore, as Tables 2 and 3 illustrate inconsistency in the beginning and termination of both spring and autumn season rainfall was common in lowland and midland AEZs of the study area.
In a nutshell, as one moves from highlands to lowlands in the study area the LGP declined, the reliability of rainfall weakened and rainfall variability increased (Table 2). Therefore, too short LGP, and erratic rainfalls are great challenges for smallholder farmers in the lowlands and midlands part of the study areas. As a result, crops will suffer from a lack of moisture which leads to insufficient crop production. Since delay in planting due to late onset and early departure of rain may result in reduced yield, planting following a “false” onset of the growing season may lead to failure and the need for expensive replanting [51,52].
4.3. Agro ecology-based distribution of rainfall in the study area
The section explains monthly, seasonal and annual spatial and temporal distribution of rainfall in the three AEZs of the East Guji Zone. The distribution of mean monthly rainfall (MMR) in East Guji is summarized in Fig 4 and mean seasonal at Fig 5 and the mean annual distribution at Fig 7. Accordingly, considerable spatial and temporal variations were observed in the three AEZs. Hypothetically, March, April, May, and September, October, and November are expected wet months of the year in the southern and south eastern part of Ethiopia. However, as a result of the perceived variability of rainfall, disparate realities were discovered in the study area. March and September remained as dry months. It was also affirmed in Fig 4 that April, May, and October were the three wettest months of the year in the lowlands and midlands AEZs. It is also noted that April to November was wet and December to March were dry months in the highlands AEZs.
Data Source: EMI.
Data Source: EMI.
The output of PCI also confirmed the variability of monthly rainfall distribution (Table 4). Moderately concentrated precipitation was observed in the highland and irregular distribution of rainfall was distinguished in the lowland and midland part of the study area.
In the lowlands and midlands AEZs rainfall occurs in two seasons which are after and before the beginning of the dry seasons. Particularly, lowlands and midlands AEZs receive rainfall during spring (belg) and autumn (meher) seasons with spring maxima (Fig 5). Moreover, little amount of precipitation was obtained in the summer and winter seasons in the study areas except in the highlands. In the southern and southeastern parts of Ethiopia, the ITCZ passes two times which results in a bi-modal pattern of rainfall. This bi-modal pattern results in a rainy season in March-May when the convergence zone travels north, and another rainy season when the zone migrates south in September-November. However, summer is a dry season [12,14,53].
Besides the spring and autumn precipitation, the highlands in the study area gain additional rainfall in the summer season; the same is true for most parts of the country. Hence, summer, spring and autumn are wet seasons in the highlands. They experienced long wet spells for most months of the year. As to EPCC [7] topographic highs play a major role in releasing the conditional thermodynamic instabilities of the moist incoming air into the country strengthening convective developments.
Consequently, the contributions of seasonal rainfall to total annual rainfall varied among the highland, midlands and lowland AEZs. The midlands and lowlands receive (84%) of precipitation in the spring and autumn seasons (Fig 5). Moreover, almost all areas found in mid and lowland AEZs except Chembe received their rainfall peak in the spring (belg) season (Fig 5). Chembe received the greatest rain of the autumn season. As a result, the three major AEZs in the study area reveal a substantial difference in the distribution of annual rainfall (Fig 6). Accordingly, the highest mean annual rainfall was observed in the highland Solemo (1262mm) and the smallest in the lowland Negele (624mm) during the studied period. Therefore, the highland AEZ acquires the highest whereas the lowlands obtain the least amount of rainfall (Fig 7). There was spatial and temporal variability in the distribution of monthly, seasonal and annual rainfall in the study area (Figs 4,5,7 and 8).
Data Source: EMI.
Data Source: EMI.
Data Source: EMI.
4.4. Agro ecology-based rainfall variability in the study area
Analysis of the variability of rainfall has been done using PCI and CV.PCI is a prominent indicator of rainfall variability. It clearly shows the level of uniformity or concentration in the distribution of rainfall across the months of the year. PCI value revealed the presence of inconsistent monthly rainfall distribution in the East Guji zone. Accordingly, there was no place in the study area with a uniform distribution of rainfall throughout months of the year (< 10% PCI).Thus, the mean PCI value of the lowlands, midlands, and highlands AEZs was 19%, 17%,and 12% respectively. As a result, lowlands and midlands acquired irregular rainfall distribution but highlands have moderately concentrated precipitation. Likewise, the year-to-year variability of rainfall denoted 27%, 21%, and 19% CV in the lowlands, midlands, and highlands correspondingly. Thus, the highland part of the study area holds low; whereas, moderate year-to-year variability of rainfall was detected in the low and midlands (Fig 9). Therefore, relatively higher PCI & CV values were observed in the lowlands (Kola) and midlands (Weynadega). In contrast, it was lower for the highlands (Dega) AEZ. Hence, based on PCI & CV there was a significant difference in the level of rainfall variability across the AEZs. As one advances from highlands to lowlands in the study area; altitude and rainfall decrease and inter-seasonal and annual variability of precipitation increases. The finding of the study corresponds with the studies of [31,54] which stated that as annual and seasonal rainfall increases; PCI values decrease.
Data Source: EMI.
According to Fig 10, the CV in the lowlands and midlands AEZs in the spring and autumn seasons was high (CV>30). Both PCI and CV have assured the existence of high seasonal variability of rainfall distribution in the mid and lowland AEZs. As NMA (55) [55] rainfall variability greater than 30% CV is risky for farmers who rely on rain-fed agriculture. Therefore, the detected higher CV resulted in lesser dependability of the seasonal rainfall for agricultural activities in the lowland and midland AEZs.
Data Source: EMI.
In contrast, the CV in the highland implies moderate variability of rainfall (CV <30%) which was 28%, 30%, and 29% in the spring, summer and autumn seasons respectively. The trend and magnitude showed a non-significant declining trend in winter rainfall in all AEZs. Besides, very high CV was observed in the winter season 102%, 99%, and 68% in lowlands, midlands, and highlands AEZs respectively. The observed extremely high CV was an expression of extremely high rainfall variability in the winter season in the study area. Likewise, both PCI and CV are highest in the winter season which reveals greater rainfall variability in the study area. Similar results were reported by [21,27] whereby extremely high variability (CV>90%) of winter season rainfall was noticed in Borena and Guji Zones respectively. Winter rain is the smallest in amount but contributes to humidity and pasture to grazing land. According to [56] significant dry periods during the winter season may have impacts if the deficiency continues into the growing season, resulting in low soil moisture recharge and deficient soil moisture at spring planting.
4.5. Agro ecology-based trends of rainfall in the study area
The trends and magnitudes of changes in mean monthly rainfall (MMR) across the three AEZs of the study area were summarized in Table 4. The magnitude of the change in rainfall across different months was disparate. Accordingly, Sen’s slope denotes positive and negative magnitudes in the distribution MMR. It revealed that a negative slope or declining magnitude of MMR distribution covered 38% of the studied months or periods. However, 62% of the studied months hold a positive magnitude of MMR distribution. The implication is an escalation in the magnitude of rainfall in the study area.
As shown in Table 4 the monthly rainfall lacked statistically significant trends in almost all months of the year. Exceptionally, a statistically significant increasing trend of rainfall was observed in Solemo (highland AEZ) and Bitata (lowland AEZ) in September at a significant level of (P< 5%). Similarly, in Kercha (midland AEZ) October rainfall showed statistically significant positive trends. As shown in Table 4, except for September and October, there were no significant trends in monthly rainfall distribution across all AEZs.
Likewise, there were no statistically significant trends of seasonal rainfall distribution except in the autumn season. A non-significant declining trend of spring season rainfall was observed in: Adola, Shakiso, Wadera (midlands) and Negele (lowland) during the study period. Therefore, spring (long rainy period) rainfall showed an insignificant decreasing trend at Adola, Shakiso, Wadera and Negele. The findings of decreasing trend of spring rainfall agree with the result of [21,23,24,26] who reported decreasing trend of spring or belg (March-May) season rainfall in different parts of the country. Similar results were also reported [57–61] in Eastern Africa in the ‘‘long rains” season in (March–May) which denoted a long-term decline. Consequently, an insignificant increase of annual rainfall was observed at the above-mentioned specific localities. La Niña affects MAM precipitation and leads to deficit rainfall in Southern Ethiopia [62].
Nevertheless, a non-significant increasing trend was detected in the rest of a great number of stations in the spring or belg rainfall. Thus, the highlands, most Lowlands: Chembe, Dawa and Bitata (lowland) and Harekello (midland) had shown a non-significant increasing trend of spring rainfall (Table 5). Besides, the value of Sen’s slope denoted a non-significant decline in the magnitude of spring rainfall in the Kercha (midland). Likewise, an insignificant decline in summer rainfall was detected in the lowlands (Bitata, Negele) and midlands (Harekello) stations.
The highland AEZs experienced a non-significant increasing trend in summer and spring but a statistically significant increase in autumn rainfall. Remarkably, autumn rainfall revealed a significant increasing trend in Bore and Solemo, (highlands), Chembe (lowland) as well as in Kercha (midland) stations at a significant level of 5%. This is in line with the studies of [21,27] who reported a significantly increasing trend of rainfall during the autumn season (short rainy period) in Borana and Guji zones correspondingly. Studies also confirmed that in S-Ethiopia there has been a rainfall deficit for the main rainfall season (MAM) and excessive rainfall for the Small Rainfall season (ON) [13,63,64].
As with the long rains, these regional-mean long-term linear trends are punctuated with periods of pronounced anomalous rainfall, including 1997–1998 and 2019–2020 when short rains totals were 2–3 times higher than climatological values. Indeed, rainfall anomalies of 100–250 mm year−1 are observed in 1997, 2006, 2012, 2015 and 2019, linked to variability associated with ENSO and corresponding interactions with the IOD [13,61,65–67]. The short-term variability, driven by changes in ENSO and the IOD, and unequal warming across the Indian Ocean [68]. Likewise, in the study areas also these were wet years: 1993, 1996, 2001, 2005, 2006, 2012, 2013, 2014, 2015, 2018 and 2020. The short rain has shown an increase in the same years in the study area.
The observed increase in the autumn rainfall in parts of the study area resulted in a subsequent increase in the trend in annual rainfall for highlands, midlands and lowlands except Wadera (Fig 8). Significant positive change in annual rainfall was observed in the highlands (Bore 15.3mm/year and Solemo14.6mm/year),lowlands (Chembe 10.9mm/year, Dawa 8mm/year and Bitata 7.8mm/year), as well as midland (Kercha 14.5mm/year) stations at a significant level of 5% (Table 6). Besides, a non-significant increasing trend of annual rainfall was observed in the remaining midlands (Adola, Shakiso and Harekello) and lowlands (Negele) except for Wadera (midland) stations which holds an insignificant declining trend of annual rainfall.
To conclude, around half of the stations in the study area were experiencing a significantly increasing trend of annual rainfall (Table 5). However, the rest of them except Wadera had shown a non-significant increasing trend in annual rainfall. In general, portions of the study areas experienced a significant increase in the trend of annual rainfall. The finding coincides with studies of [19,20]; [24–26] who reported an increasing trend of annual rainfall in different parts of the country. The study also coincides with [63–64,69,70] who reported a notable increase in East African rainfall. Similar findings also explained that the increases in annual rainfall in the study area are largely a result of an increasing in the ‘short’ rainfall (OND) season in the area [63]. Therefore, at the mentioned locality most of the statistically significant increases in annual rainfall are directly related to the observed significant increase in trend in autumn (meher) rainfall.
However, the finding of increasing annual rainfall contradicts the studies made [19–21,43] that found a decreasing trend of annual rainfall in a different part of the country. Hence, the diverse trend was observed in different AEZs of the study area, ensuring varied spatial-temporal trends in rainfall distributions in Ethiopia. Moreover, [14] confirmed that a cold Gulf of Guinea leads to a strengthening of the TEJ and a suppressed ITCZ, in turn giving rise to increased rainfall in southern Ethiopia with a reduction in the north and central regions. This disparity has sometimes been called the East African climate paradox [57].
Besides, the choice of study period strongly influences the results of trend analysis of rainfall seriously due to the effects of temporal variability [31]. Previous studies made by Habte et al. [21] in the Guji Zone from 1986–2017 showed a declining trend of annual rainfall in the study area. A Similar finding was reported by [61] in Eastern Africa the long rains exhibited consistent negative trends from 1985 to 2017. Notable declines occurred in 1999 and 2010–2011 in Eastern Africa the long rains.
In contrast, as observed in Figs 5 & 8 the wetness of the study area was enhanced during the ending period of the present study (2019 & 2020). In other words, the perceived increase in rainfall was recorded after the earlier study was already conducted in the study area. The study of [61] supported the finding of the current study by explaining the recovery of very wet long rains in 2018 and 2020 in Eastern Africa. Therefore, a comparable declining trend of long rainfall was observed in East Africa from 1985–2017 but the wetness was restored in 2018–2020.Thus, contrasting results obtained in the current study might be due to the chosen study periods which already determine the output of the trend analysis. Therefore, the temporal variability of rainfall in the study area is articulated by the period during which the study was conducted. Whereas, the observed spatial variability of rainfall in Guji zone was due to the diverse AEZ ranging from 632m -2960m (Fig 2).
5. Conclusions
Identifying the variability of rainfall in space and time, onset and cessation, and distributions across AEZs is essential since rain-fed agriculture is the mainstay of smallholder farmers in the study area. The finding denotes the presence of spatial and temporal variation in the distribution of rainfall in the three AEZs of the study areas. Midlands and lowlands experience bimodal rainfall, two distinct onset and cessations and dual LGP in the autumn and spring seasons. Whereas, highlands obtain a considerable amount of rainfall in spring, autumn and summer. Furthermore, variability in the days of onset and departure of rainfall was observed in the three major AEZs of the study area. Complex trends in the distribution of rainfall had been found in the area. Insignificant, increasing trends of spring rainfall had been discovered in most parts of the study area. Partially, there was a significant increase in autumn rainfall in the study area. Likewise, the annual rainfall had shown an insignificant increasing trend in some parts of the study area, a significant increasing trend was observed in the highlands (Bore and Solemo); lowlands (Chembe, Dawa and Bitata) as well as in the midland (Kercha) stations.
Thus, the PCI value decreases with increasing altitude and rainfall. As one moves from the highlands to the lowlands in the study area; rainfall decreases and the inter-annual variability of precipitation will increase. Therefore, moderate year-to-year variability was detected in the low and midlands whereas the highlands exhibited lesser year-to-year variability of rainfall. High monthly and seasonal variability together with the late onset and early termination of the spring and autumn rainfall in the midlands and lowlands parts of the study was a challenge for the small-holder farmers. Although rainfall variability was discovered as a problem its magnitude varies across the different AEZs in the study area. The erratic nature of rainfall resulted in shortage of water for rain fed agriculture in the mid and lowlands AEZ. Thus, more consideration should be given to the lowlands and midland AEZs of the study areas.
Therefore, selecting and providing appropriate short-cycle grown and drought-resistant crops to the midlands and lowlands AEZs of the study area is critical. Furthermore, additional support to strengthen the use of water resources for (irrigation/water harvesting) agricultural activities should be done for smallholder farmers in the midlands and lowlands AEZs. Accurate, reliable, and timely information about the onset and end of rainfall should also be provided for the community to avoid planting crops on the false onset date. Due to limitations of time and resources the study was done at the scope of “Spatiotemporal Trends and Variability of Rainfall”. Further investigation of the spatial distribution and temporal trends of temperature together with an assessment of the rate of evaporation across the three AEZ in the study area is vital to understanding the micro climate of the area. So that the result will alleviate the issues related to the livelihood of smallholders.
References
- 1. Merabtene T, Siddique M, Shanableh A. Assessment of seasonal and annual rainfall trends and variability in Sharjah City, UAE. Advances in Meteorology. 2016;2016:1–3.
- 2. Badesso BB. Spatial and temporal rainfall trend analysis: A case study of south western, Ethiopia. Civil and Environmental Research. 2017;9(8).
- 3. Shiferaw B, Tesfaye K, Kassie M, Abate T, Prasanna BM, Menkir A. Managing vulnerability to drought and enhancing livelihood resilience in sub-Saharan Africa: Technological, institutional and policy options. Weather and climate extremes. 2014 Jun 1;3:67–79.
- 4. Washington R, Preston A. Extreme wet years over southern Africa: Role of Indian Ocean sea surface temperatures. Journal of Geophysical Research: Atmospheres. 2006 Aug 16;111(D15).
- 5. Parry ML, Rosenzweig C, Iglesias A, Livermore M, Fischer G. Effects of climate change on global food production under SRES emissions and socio-economic scenarios. Global environmental change. 2004 Apr 1;14(1):53–67.
- 6. Change IC. Impacts, adaptation and vulnerability. Part A: global and sectoral aspects. Contribution of working group II to the fifth assessment report of the intergovernmental Panel on Climate Change. 2014 Mar;1132.
- 7.
CHANGE EP. FIRST ASSESSMENT REPORT WORKING GROUP II-CLIMATE CHANGE IMPACT, VULNERABILITY, ADAPTATION AND MITIGATION V.
- 8. Zerga B, Gebeyehu G. Climate change in Ethiopia variability, impact, mitigation, and adaptation. Journal of Social Science and Humanities Research. 2016 Apr;2(4):66–84.
- 9. Simane B, Beyene H, Deressa W, Kumie A, Berhane K, Samet J. Review of climate change and health in Ethiopia: status and gap analysis. Ethiopian Journal of Health Development. 2016;30(1):28–41. pmid:28867919
- 10.
FAS, 2019. (2019). Geography of Ethiopia and the Horn. Manual.
- 11. Young HR, Klingaman NP. Skill of seasonal rainfall and temperature forecasts for East Africa. Weather and Forecasting. 2020 Oct 1;35(5):1783–800.
- 12. Matewos T. Climate change-induced impacts on smallholder farmers in selected districts of Sidama, Southern Ethiopia. Climate. 2019 May 22;7(5):70.
- 13. Nicholson SE. Climate and climatic variability of rainfall over eastern Africa. Reviews of Geophysics. 2017 Sep;55(3):590–635.
- 14. Diro GT, Toniazzo T, Shaffrey L. Ethiopian rainfall in climate models. African climate and climate change: physical, social and political perspectives. 2011:51–69.
- 15. Kassahun B. Ye’ayer Mezabat’na tinbi’ya k’Itiopia Antsar (Climate Change and Forecast in Ethiopia, in Amharic). Ina Meeting organized by the DPPC on Nehase 1991.
- 16.
Wolde-Georgis T, Aweke D, Hagos Y. The Case of Ethiopia.
- 17. Conway D, Mould C, Bewket W. Over one century of rainfall and temperature observations in Addis Ababa, Ethiopia. International Journal of Climatology: A Journal of the Royal Meteorological Society. 2004 Jan;24(1):77–91.
- 18. Seleshi Y, Zanke U. Recent changes in rainfall and rainy days in Ethiopia. International Journal of Climatology: A Journal of the Royal Meteorological Society. 2004 Jun 30;24(8):973–83.
- 19. Rosell S, Holmer B. Rainfall change and its implications for Belg harvest in South Wollo, Ethiopia. Geografiska Annaler: Series A, Physical Geography. 2007 Dec;89(4):287–99.
- 20. Bayable G, Amare G, Alemu G, Gashaw T. Spatiotemporal variability and trends of rainfall and its association with Pacific Ocean Sea surface temperature in West Harerge Zone, Eastern Ethiopia. Environmental Systems Research. 2021 Dec;10(1):1–21.
- 21. Habte M, Eshetu M, Maryo M, Andualem D, Legesse A. Effects of climate variability on livestock productivity and pastoralists perception: The case of drought resilience in Southeastern Ethiopia. Veterinary and Animal Science. 2022 Jun 1;16:100240. pmid:35257034
- 22. Wagesho N, Goel NK, Jain MK. Temporal and spatial variability of annual and seasonal rainfall over Ethiopia. Hydrological Sciences Journal. 2013 Feb 1;58(2):354–73.
- 23. Mohammed Y, Yimer F, Tadesse M, Tesfaye K. Variability and trends of rainfall extreme events in north east highlands of Ethiopia. Int. J. Hydrol. 2018;2(5):594–605.
- 24. Abegaz WB, Mekoya A. Rainfall variability and trends over Central Ethiopia. Rema. 2020;10(39.58):2054.
- 25. Alemayehu A, Maru M, Bewket W, Assen M. Spatiotemporal variability and trends in rainfall and temperature in Alwero watershed, western Ethiopia. Environmental Systems Research. 2020 Dec;9(1):1–5.
- 26. Belay A, Demissie T, Recha JW, Oludhe C, Osano PM, Olaka LA, et al. Analysis of climate variability and trends in Southern Ethiopia. Climate. 2021 Jun 15;9(6):96.
- 27. Worku MA, Feyisa GL, Beketie KT. Climate trend analysis for a semi-arid Borana zone in southern Ethiopia during 1981–2018. Environmental Systems Research. 2022 Mar 4;11(1):2.
- 28. Esayas B, Simane B, Teferi E, Ongoma V, Tefera N. Trends in extreme climate events over three agroecological zones of southern Ethiopia. Advances in Meteorology. 2018 Oct 16;2018:1–7.
- 29. Taye M, Simane B, Zaitchik BF, Selassie YG, Setegn S. Rainfall variability across the agro-climatic zones of a tropical highland: the case of the jema watershed, northwestern Ethiopia. Environments. 2019 Nov 12;6(11):118.
- 30. Shekuru AH, Berlie AB, Bizuneh YK. Variability and trends of temperature and rainfall over three agro-ecological zones in North Shewa, Central Ethiopia. Arabian Journal of Geosciences. 2022 Sep;15(18):1495.
- 31. Bewket W, Conway D. A note on the temporal and spatial variability of rainfall in the drought‐prone Amhara region of Ethiopia. International Journal of Climatology: A Journal of the Royal Meteorological Society. 2007 Sep;27(11):1467–77.
- 32. Degefu MA, Bewket W. Trends and spatial patterns of drought incidence in the omo‐ghibe river basin, ethiopia. Geografiska Annaler: Series A, Physical Geography. 2015 Jun 1;97(2):395–414.
- 33. Gamachu D. Some patterns of altitudinal variation of climatic elements in the mountainous regions of Ethiopia. Mountain Research and Development. 1988 May 1:131–8.
- 34. Dejene IN, Moisa MB, Gemeda DO. Spatiotemporal monitoring of drought using satellite precipitation products: The case of Borena agro-pastoralists and pastoralists regions, South Ethiopia. Heliyon. 2023 Mar 1;9(3). pmid:36895373
- 35.
https://www.statsethiopia.gov.et/population-projection/Ethiopia CS. Population Projection of Ethiopia for All Regions at Wereda Level from 2014–2017. Agency CS, editor. Addis Ababa. 2013.
- 36.
Authority CS. Population and housing census of Ethiopia. Addis Ababa: Central Statistics Authority. 2007.
- 37. Abate T. Contribution of indigenous knowledge to climate change and adaptation response in Southern Ethiopia. J. Earth Sci. Clim. Chang. 2016;7:377.
- 38. Prates I, Lazzari E, Anker R, Anker M. Living Income For Guji Zone (Oromia Region), Ethiopia 2021. Universidad Privada Boliviana; 2021 Jun.
- 39. Hurni H (1998). Agro ecological belts of Ethiopia: Explanatory notes on three maps at a scale of 1:1,000,000. Research Report, Soil Conservation Research Program, Addis Ababa, 43.
- 40. Olaniran OJ. The onset of the rains and the start of the growing season in Nigeria. Nigerian Geographical Journal. 1983;26(1):81–8.
- 41. Patrick O. A., Emmanuel N., & Obadiah A. A. (2019). An Analysis of the Impact of Rainfall Onset, Cessation and Length of Growing Season Variability on Crop Yields in Benue State, Nigeria. East African Scholars J Agri Life Sci, 2(9), 439–442.
- 42. Mosunmola IA, Samaila IK, Emmanuel B, Adolphus I. Evaluation of Onset and Cessation of Rainfall and Temperature on Maize Yield in Akure, Ondo State, Nigeria. Atmospheric and Climate Sciences. 2020 Feb 6;10(2):125–45.
- 43. Mekonen AA, Berlie AB. Spatiotemporal variability and trends of rainfall and temperature in the Northeastern Highlands of Ethiopia. Modeling Earth Systems and Environment. 2020 Mar;6:285–300.
- 44. Wang F, Shao W, Yu H, Kan G, He X, Zhang D, et al, Wang G. Re-evaluation of the power of the mann-kendall test for detecting monotonic trends in hydrometeorological time series. Frontiers in Earth Science. 2020 Feb 6;8:14.
- 45. Poudel S, Shaw R. The relationships between climate variability and crop yield in a mountainous environment: A case study in Lamjung District, Nepal. Climate. 2016 Mar 2;4(1):13.
- 46. Sen PK. Estimates of the regression coefficient based on Kendall’s tau. Journal of the American statistical association. 1968 Dec 1;63(324):1379–89.
- 47. Oliver JE. Monthly precipitation distribution: a comparative index. The Professional Geographer. 1980 Aug 1;32(3):300–9.
- 48.
Hare FK. Climate and desertification: a revised analysis. World Meteorological Organization; 1983.
- 49. Diodato N, Ceccarelli M. Interpolation processes using multivariate geostatistics for mapping of climatological precipitation mean in the Sannio Mountains (southern Italy). Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group. 2005 Mar;30(3):259–68.
- 50.
Abebe M. The onset, cessation and dry spells of the small rainy season (Belg) of Ethiopia. National Meteorological Agency, Addis Ababa, Ethiopia. 2006 Sep.
- 51.
Dunning C, Black E, Allan RP. Diagnosing observed characteristics of the wet season across Africa to identify deficiencies in climate model simulations. InAGU Fall Meeting Abstracts 2017 Dec (Vol. 2017, pp. GC33B-1072).
- 52. Alemayehu A, Bewket W. Vulnerability of smallholder farmers’ to climate change and variability in the central highlands of Ethiopia. Ethiopian Journal of the Social Sciences and Humanities. 2016;12(2):1–24.
- 53.
Knudsen C. Factors influencing interannual variability of Belg rain in Ethiopia (Master’s thesis, The University of Bergen).
- 54. Hessebo MT, Woldeamanuel T, Tadesse M. Spatial and temporal climate variability and change in the bilate catchment, central Rift Valley lakes region, Ethiopia. Physical Geography. 2021 May 4;42(3):199–225.
- 55.
Ababa A. Climate change national adaptation programme of action (Napa) of Ethiopia. National Meteorological Services Agency, Ministry of Water Resources, Federal Democratic Republic of Ethiopia, Addis Ababa. 2007 Jun.
- 56.
Wilhite DA. Drought as a natural hazard: concepts and definitions.
- 57. Rowell DP, Booth BB, Nicholson SE, Good P. Reconciling past and future rainfall trends over East Africa. Journal of Climate. 2015 Dec 15;28(24):9768–88.
- 58. Lyon B, DeWitt DG. A recent and abrupt decline in the East African long rains. Geophysical Research Letters. 2012 Jan;39(2).
- 59. Viste E, Korecha D, Sorteberg A. Recent drought and precipitation tendencies in Ethiopia. Theoretical and Applied Climatology. 2013 May;112:535–51.
- 60. Funk C, Hoell A, Shukla S, Blade I, Liebmann B, Roberts JB, et al. Predicting East African spring droughts using Pacific and Indian Ocean sea surface temperature indices. Hydrology and Earth System Sciences. 2014 Dec 10;18(12):4965–78.
- 61. Palmer PI, Wainwright CM, Dong B, Maidment RI, Wheeler KG, Gedney N, et al. Drivers and impacts of Eastern African rainfall variability. Nature Reviews Earth & Environment. 2023 Apr;4(4):254–70.
- 62. Mcsweeney C, New M, Lizcano G, Lu X. The UNDP Climate Change Country Profiles: Improving the accessibility of observed and projected climate information for studies of climate change in developing countries. Bulletin of the American Meteorological society. 2010 Feb 1;91(2):157–66.
- 63. Asfaw A, Simane B, Hassen A, Bantider A. Variability and time series trend analysis of rainfall and temperature in northcentral Ethiopia: A case study in Woleka sub-basin. Weather and climate extremes. 2018 Mar 1;19:29–41.
- 64. Degefu MA, Rowell DP, Bewket W. Teleconnections between Ethiopian rainfall variability and global SSTs: observations and methods for model evaluation. Meteorology and Atmospheric Physics. 2017 Apr;129:173–86.
- 65. Saji NH, Goswami BN, Vinayachandran PN, Yamagata T. A dipole mode in the tropical Indian Ocean. Nature. 1999 Sep 23;401(6751):360–3. pmid:16862108
- 66. Webster PJ, Moore AM, Loschnigg JP, Leben RR. Coupled ocean–atmosphere dynamics in the Indian Ocean during 1997–98. Nature. 1999 Sep 23;401(6751):356–60. pmid:16862107
- 67. Black E. The relationship between Indian Ocean sea–surface temperature and East African rainfall. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2005 Jan 15;363(1826):43–7. pmid:15598619
- 68. Liebmann B, Hoerling MP, Funk C, Bladé I, Dole RM, Allured D, et al. Understanding recent eastern Horn of Africa rainfall variability and change. Journal of Climate. 2014 Dec 1;27(23):8630–45.
- 69. Shongwe ME, Van Oldenborgh GJ, Van den Hurk B, van Aalst M. Projected changes in mean and extreme precipitation in Africa under global warming. Part II: East Africa. Journal of climate. 2011 Jul 15;24(14):3718–33.
- 70. Otieno VO, Anyah RO. CMIP5 simulated climate conditions of the Greater Horn of Africa (GHA). Part II: Projected climate. Climate dynamics. 2013 Oct;41:2099–113.