Characterize of Drought Events via the Decile, SPI and RDI in Southeast Ethiopia

doi:10.21203/rs.3.rs-5111418/v1

Download PDF

Research Article

Characterize of Drought Events via the Decile, SPI and RDI in Southeast Ethiopia

https://doi.org/10.21203/rs.3.rs-5111418/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Drought characterization can be used to devise an effective early warning system as a key component in the development of drought preparedness plans that manage drought impacts. The aim of this study was to characterize the drought events, particularly those involving long-term of meteorological and agricultural events, which are the predominant types affecting communities. Conventionally, regional drought analysis is performed on the basis of drought indices for the identification of drought intensity, frequency, duration and areal extent across southeast Ethiopia. The indices used for the study were the Precipitation Deciles (PD), Standardized Precipitation Index (SPI) and Reconnaissance Drought Index (RDI) for the period of 1987–2018. The spatial extent of drought in the study area was interpolated via the inverse distance weighted method via the spatial analyst tool of ArcGIS. Most of the studied stations experienced drought events in 1991/1992, 1998/1999, 2001/2002, 2010/2011 and 2011/2012. The results indicate that 2010/2011 was a year of exceptionally widespread drought. To this end, having information on the intensity, frequency, duration and areal extent of drought needed in water conservation policies can help bridge the gap between water availability, supply, and demand in drought-affected areas.

Climatology

Precipitation deficiency

Drought intensity

Drought frequency

Drought duration and Areal coverage of drought

Drought is a recurrent climate phenomenon that occurs in most parts of the world, with varying magnitudes, intensities, frequencies, and durations (Wilhite, 1993; Shatanawi et al., 2013). Recent instances of drought have underscored the vulnerability of both communities and ecosystems to prolonged periods of precipitation deficiency (Smakhtin & Hughes, 2004). Described as a normal feature of any climate, drought is a temporary but recurrent natural disaster stemming from inadequate precipitation, leading to substantial economic losses (Smakhtin & Hughes, 2004). While it is impossible to prevent droughts entirely, developing a drought preparedness plan and managing its impacts are viable strategies. The effectiveness of such measures relies, in part, on accurately defining droughts and quantifying their characteristics (Smith & McKeon, 1998).

Droughts lack a universally accepted definition that applies to all situations. Most definitions, however, focus on rainfall deficiency resulting in water scarcity for various water-dependent activities (Wilhite, 1993; Wilhite & Glantz, 1985). According to Morid et al. (2006), Karavitis (1999), and Szinell et al. (1998), drought is characterized as "the state of adverse and widespread hydrological, environmental, social, and economic impacts due to less than expected water quantities."

Globally, droughts rank among the most significant natural disasters, impacting environmental, economic, and social conditions (Paulo et al., 2012; Morid et al., 2006). They affect various aspects of the hydrological cycle, from precipitation to streamflow in surface water systems or groundwater aquifer recharge and storage. Assessing drought magnitude is crucial for evaluating its consequences and implementing mitigation strategies. The magnitude refers to the amount of rainfall or water storage deficit at a specific location and time. Droughts typically manifest slowly, and their impacts may persist for years after the event ends (Morid et al., 2006).

While drought is a naturally recurring extreme event (Wilhite, 1993; Shatanawi et al., 2013), empirical and modeling studies indicate that climate change is likely to amplify drought intensity, frequency, and duration in certain regions over the coming decades (IPCC II, 2014; Degefu and Bewket, 2013). The impact of drought depends on factors such as the magnitude, intensity, frequency, duration, and spatial extent of the rainfall deficit (Degefu and Bewket, 2013; Zargar et al., 2011).

Droughts are categorized as meteorological, agricultural or hydrological. Meteorological drought indicates a deficiency of rainfall compared with normal rainfall in a given region. Agricultural drought occurs when rainfall and soil moisture are inadequate to meet the water requirements of crops. Hydrological drought indicates the scarcity of water in surface and underground resources (Wilhite & Glantz, 1985).

Indices play crucial roles in characterizing and monitoring drought, simplifying complex climatic functions and quantifying anomalies in terms of severity, frequency, and duration (Tsakiris et al., 2007). They facilitate the communication of drought magnitude to a wide audience and find utility across academic, strategic, and operational domains (Morid et al., 2006). Since no single index universally fits all situations, employing multiple indices is often necessary (Morid et al., 2006). Indices furnish quantitative information about drought magnitude, severity, frequency, and duration aiding in historical analysis and enabling the forecasting of future probabilities (Keyantash and Dracup, 2002).

Although drought is becoming the most common and damaging natural hazard in Ethiopia, no detailed studies have been conducted on its magnitude, intensity, frequency, duration and spatial extent in different regions of Ethiopia (Degefu and Bewket, 2013). This study focuses on meteorological and agricultural drought, using precipitation for precipitation deciles (PD) and the standardized precipitation index (SPI) (and, in the case of the reconnaissance drought indexes (RDIs), temperature to generate potential evapotranspiration) as the primary determinants. Drought indices respond to abnormally dry weather conditions but do not account for long-term hydrological impacts or human interventions such as irrigation.

Given the urgent need for comprehensive regional drought management policies, this study adopts a proactive approach. Characterizing drought events via various indices is crucial for developing alternative early warning mechanisms, drafting effective drought preparedness plans, managing drought impacts, and implementing adaptation and mitigation strategies (McKee et al., 1993).

2.1 Description of the study area

The south-eastern part of Ethiopia is located between 4° and 11° N latitude and 40° and 48° E longitude (Fig. 1). The total area of the region is estimated to be 340,000km² (Abdulahi et al., 2020). According to the 2007 census conducted by the Central Statistical Agency of Ethiopia (CSA), the region has a total population of 4,445,219 of which 2,472,490 and 1,972,729 are male and female respectively. Furthermore, 682,857 households have resided in the entire region. Nearly 80% of the population are rural pastoralists and agro-pastoralists(UNDP, 1999).

The altitudinal coverage ranges from 400 m in the southeast to approximately 1600 m in the north. The climate of the region is categorized as arid or semiarid. Rainfall in the region is low, erratic and unreliable. The northern part receives summar rains, which range from July to September and the southern wider part of the study area has spring and autumn rainy seasons which range from March to April and from March to April respectively (Mekonnnen et al., 2021). The temperature ranges from 20–45°C and the average annual rainfall is 300–500 mm (Region & Gebre-mariam, 2007).

2.2 Data source

Reconstructed/merged monthly rainfall and temperature (maximum and minimum) data were collected from the Ethiopia Meteorological Institute for time series of 1987 to 2018. The data were extracted as point data for ten selected meteorological stations (Fig. 1). The reconstructed dataset merging two datasets includes station gauge data from the Ethiopia Meteorological Institute and satellite data from the European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) and the US National Aeronautics and Space Administration (NASA).

2.3 Methods

The drought analysis was performed via DrinC software (Drought Indices Calculator), which was developed to provide a simple interface for the calculation of drought indices. The precipitation deciles indices (PDs), standardized precipitation indices (SPIs), and reconnaissance drought indices (RDIs) were used to designate the drought intensity. The SPI and RDI indices were computed for the 12-month time series for the characterization of drought. The common characteristic of the selected indices is that they require a relatively small number of data for their calculation and the results can be easily interpreted and used in strategic planning and operational applications (Tigkas et al., 2013).

2.3.1 Precipitation Deciles (PD)

One of the simplest drought indices is precipitation index (PD), which was introduced by Gibbs and Maher (1967). The precipitation totals for the preceding three months are ranked against climatologic records. If the sum falls within the lowest decile of the historical distribution of 3-month totals, then the region is considered to be under drought conditions (Kininmonth et al. 2000). The drought ends when: (i) the precipitation measured during the past month already places the 3-month total in or above the fourth decile or (ii) the precipitation total for the past three months is in or above the eighth decile.

The first decile is the precipitation amount not exceeded by the lowest 10% of the precipitation occurrences. The second decile is the precipitation amount not exceeded by the lowest 20% of occurrences. These deciles continue until the rainfall amount identified by the tenth decile is the largest precipitation amount within the long-term record. By definition, the fifth decile is the median, and it is the precipitation amount not exceeded by 50% of the occurrences over the period of record. The deciles are grouped into five classifications (Table 1).

Table 1

Classification of the decile index
Decile Class	Description
deciles 1–2: lowest 20%	much below normal
deciles 3–4: next lowest 20%	below normal
deciles 5–6: middle 20%	near normal
deciles 7–8: next highest 20%	above normal
deciles 9–10: highest 20%	much above normal

2.3.2 Standardized Precipitation Index (SPI)

The SPI is particularly suitable for comparing meteorological drought among different time periods and regions with different climatic conditions. This is because it is used at a variety of time scales and only requires precipitation data. For the calculation, the long-term precipitation record for a desired period was fitted to a probability distribution, and then transformed into a normal distribution so that the mean SPI for the location and desired period was zero (McKee et al. 1993; Edwards and McKee 1997). The gamma distribution is defined by its frequency or probability density function:

$\:g\left(x\right)=\:\frac{1}{{\beta\:}^{a}\varGamma\:\left(a\right)}{x}^{a-1}{e}^{\raisebox{1ex}{$-x$}\!\left/\:\!\raisebox{-1ex}{$\beta\:$}\right.}$ , for x > 0…………………………………………..1

in which α and β are the shape and scale parameters respectively, x is the precipitation amount and $\:\varGamma\:\left(a\right)$ is the gamma function. Parameters α and β of the gamma pdf are estimated for each station and for the time scale of interest (12 months). The maximum likelihood estimations of α and β are as follows:

$\:a=\frac{1}{4A}$ (1+ $\:\sqrt{1+\:\frac{4A}{3}})$ , β=$\:\frac{\stackrel{-}{x}}{a}$, where 𝑨=$\:\text{ln}(\stackrel{-}{x})$ - $\:\frac{?\text{ln}\left(x\right)}{n}$………………….2

and n is the number of observations.

$\:H\left(x\right)=\:q+\left(1-q\right)G\left(x\right)$ …………………………………………………3

in which $\:q$ is the probability of zero precipitation and G(x) is the cumulative probability of the incomplete gamma function. If m is the number of zeros in a precipitation time series, then q can be estimated by m/n. The cumulative probability H(x), is then transformed to the standard normal random variable z with a mean of zero and variance of one, which is the value of the SPI.

Generally, monthly precipitation was not normally distributed so a transformation was performed such that the derived SPI values followed a normal distribution. The SPI is the number of standard deviations for which the observed value deviates from the long-term mean for a normally distributed random variable. One interpretation of the resulting values is presented in Table 2 (Tsakiris and Vangelis 2004).

Table 2

Classification of Standardized Precipitation Index
SPI values	Classification
2.0 or more	Extremely Wet
1.5 to 1.99	Very Wet
1.0 to 1.49	Moderately Wet
− .99 to .99	Near Normal
-1.0 to -1.49	Moderately Dry
-1.5 to -1.99	Severely Dry
-2 or less	Extremely Dry

2.3.3 Reconnaissance Drought Index (RDI)

The RDI is expected to be a more sensitive index than the SPI since it is related not only to precipitation. The RDI is as simple as the SPI and incorporates evapotranspiration. A variety of applications have shown that the RDI behaves in a similar manner as the SPI. However, the results differed when comparing drought severity in different climatic regions.

Drought severity can be assessed through the computation of the RDI and more precisely through its standardized form (RDIst). The RDI was developed to approach the water deficit in a more accurate way, as a sort of balance between input and output in a water system (Tsakiris et al., 2007; Tsakiris & Vangelis, 2005). The RDI is based on both cumulative precipitation (P) and potential evapotranspiration (PET), which are measured (P) and calculated (PET) determinants. The calculation method of PET, however, does not seem to affect the results of RDI in any way (Vangelis et al., 2013). The initial value ($\:{a}_{k}$) of the RDI is calculated for the i-th year on a time basis of k (months) as follows:

$\:{a}_{k}^{\left(i\right)}=\:\frac{\underset{j=1}{\overset{k}{?}}{P}_{ij}}{\underset{j=1}{\overset{k}{?}}{PET}_{ij}}$ , i = 1(1)N and j = 1(1)k……………………….4

where $\:{P}_{ij}$ and $\:{PET}_{ij}$ are the precipitation and potential evapotranspiration of the j-th month of the i-th year, respectively, and N is the total number of years of available data. The values of ($\:{a}_{k}$) satisfactorily follow both the lognormal and the gamma distributions in a wide range of locations and different time scales, in which they were tested (Tigkas, 2008; Tsakiris et al., 2008). By assuming that the lognormal distribution is applied, the following equation can be used for the calculation of RDI_st:

$\:{RDI}_{st}^{\left(i\right)}=\:\frac{{y}^{\left(i\right)}\:-\:\overline{y}\:}{{\widehat{s}}_{y}}$ ……………………………………………………5

where $\:{y}^{i}$ is ln($\:{a}_{k}^{\left(i\right)}$), $\:\overline{y}$ is its arithmetic mean and $\:{\widehat{\sigma\:}}_{y}$ is its standard deviation.

If the gamma distribution is applied, RDI_st can be calculated by fitting the gamma probability density function (GPDF) to the given frequency distribution of $\:{a}_{k}$ (Tigkas, 2008; Tsakiris et al., 2008). For short reference periods (e.g. monthly or 3-months) which may include zero values for the cumulative precipitation of the period, the RDI_st can be calculated on the basis of a composite cumulative distribution function including the probability of zero precipitation, and the gamma cumulative probability.

Positive values of RDI_st indicate wet periods, whereas negative values indicate dry periods compared with the normal conditions of the area. The severity of drought events increases when RDI_st values are getting highly negative. Drought severity can be categorized into mild, moderate, severe and extreme classes, with corresponding boundary values of RDI_st (-0.5 to -1.0), (-1.0 to -1.5), (-1.5 to -2.0) and (< -2.0), respectively. The RDI is calculated for a hydrological year in 3, 6, 9 and 12 month reference periods.

This study investigated the historical precipitation deficiency, frequency, intensity, duration and areal extent of droughts in the southeastern region of Ethiopia from 1987–2018, utilizing three key annual drought indices: the precipitation decile (PD), standardized precipitation index (SPI), and reconnaissance drought index (RDI). Data were collected from ten meteorological stations across the region to analyze the drought patterns and their variability comprehensively over the study period. The findings of this study shed light on the dynamics of drought providing valuable insights for drought assessment, management and planning.

3.1 Precipitation Deficiency Years

Precipitation deficiency years were identified by an examination of the Decile Precipitation Index, which was conducted across ten selected meteorological stations and indicates much below normal and below normal rainfall conditions. Within the study period spanning from 1991/1992 to 2016/2017, instances of rainfall deficiency were prominent, notably occurring in the years 2001/2002, 2003/2004, 2007/2008, 2008/2009, 2010/2011, and 2016/2017. In particular, in 2010/2011 and 2016/2017 much less than normal rainfall affected nearly the entire region (Tables 3 & 4).

Table 3

Years with much less than normal rainfall
Station	Much Below Normal Rainfall
Adigala	2007/2008	2008/2009	2010/2011	2011/2012	2014/2015	2016/2017
Gursum	1996/1997	2001/2002	2003/2004	2010/2011	2011/2012	2014/2015
Jijjiga	1998/1999	1999/2000	2007/2008	2010/2011	2011/2012	2012/2013
Kebribeya	1989/1990	1990/1991	2003/2004	2007/2008	2014/2015	2016/2017
Shinile	1990/1991	1996/1997	2001/2002	2008/2009	2010/2011	2014/2015
Darore	1991/1992	1998/1999	2003/2004	2005/2006	2010/2011	2016/2017
DolloDdo	1987/1988	1991/1992	1992/1993	1993/1994	2010/2011	2016/2017
Gode	1991/1992	1993/1994	2001/2002	2008/2009	2010/2011	2016/2017
Ime	1991/1992	1998/1999	2001/2002	2003/2004	2010/2011	2016/2017
Kebridehar	1987/1988	1991/1992	1998/1999	2010/2011	2011/2012	2016/2017

Table 4

Below normal rainfall years
Station	Below Normal Rainfall
Adigala	1988/1989	1990/1991	1991/1992	1996/1997	2001/2002	2004/2005
Gursum	1987/1988	2000/2001	2004/2005	2008/2009	2012/2013	2015/2016
Jijjiga	1988/1989	1996/1997	2000/2001	2003/2004	2008/2009	2014/2015
Kebribeya	1991/1992	1992/1993	1999/2000	2010/2011	2012/2013	2015/2016
Shinile	1987/1988	1991/1992	2003/2004	2004/2005	2011/2012	2012/2013
Darore	1994/1995	1996/1997	1999/2000	2002/2003	2007/2008	2008/2009
DolloDdo	1995/1996	1998/1999	2000/2001	2001/2002	2005/2006	2015/2016
Gode	1989/1990	1992/1993	1996/1997	1998/1999	1999/2000	2007/2008
Ime	1990/1991	1996/1997	1999/2000	2005/2006	2007/2008	2008/2009
Kebridehar	2001/2002	2002/2003	2003/2004	2005/2006	2007/2008	2014/2015

The findings also reveal an increase in the frequency and duration of rainfall deficiency years since 2007/2008, especially in Adigala, Jigjiga, and Shinile. During the middle decade, the Jigjiga and Kebridehar districts experienced three consecutive years of rainfall deficiency, and Jigjiga and Shinile also experienced rainfall deficiency during the last decade. Moreover, much below normal and below normal rainfall has persisted for two to three years during the last decade in proximity to these districts (Fig. 2).

The majority of stations experienced much below normal and below normal rainfall in all the specified years. This consistent pattern suggests that most districts faced prolonged periods of precipitation deficiency, which likely had significant impacts on agriculture, water resources, and livelihoods in the region, potentially leading to reduced crop yields, water scarcity, and socioeconomic challenges for the local population. This also suggests that a year with recurrent rainfall deficiency led to drought events and a variable pattern of drought occurrence which could pose challenges for agricultural productivity and water availability in the region. The results also suggest intermittent rainfall deficiency interspersed with relatively normal rainfall years, which could create challenges for agricultural planning and water management strategies.

3.1.1 Precipitation Threshold Value

The precipitation thresholds exhibit a range of precipitation values at the selected station within the study period where the precipitation amount was not exceeded by the lowest 40% of occurrences which indicates precipitation deficiency. The values vary from 105.4 mm to 611.6 mm for much below normal rainfall and from 174.4 mm to 680.4 mm for below normal rainfall. The highest thresholds were recorded in the Gursum district, ranging from 523.0 mm to 680.4 mm, whereas the lowest thresholds were observed in the Kebridehar and Gode districts (Table 5). Other stations located in similar climatic zones with the selected stations generally experienced comparable challenges during rainfall deficiency years and the nearest designated rainfall threshold value. Reduced precipitation can lead to water stress in trees, increasing their susceptibility to diseases and insect infestations. This can result in widespread dieback and decline in forest health.

Table 5

Threshold values for the selected stations
Thresholds	Adigala	Gursum	Jigjiga	Kebribeya	Shinile	Daror	Dolloado	Ime	Gode	Kebridehar
10%	174.8	523.0	403.0	341.0	320.4	191.9	122.2	165.5	109.6	105.4
20%	197.0	611.6	454.0	383.8	347.2	217.4	149.5	220.6	147.8	159.0
30%	211.0	673.0	487.3	403.6	376.2	231.7	181.0	274.7	174.4	183.5
40%	243.0	680.4	503.0	413.8	398.0	269.8	196.2	292.8	186.0	203.2

3.2 Drought characterization

The primary focus of drought characterization was to analyze historical drought occurrences from 1987–2018. Drought events were characterized via the SPI-12 and RDI-12 to gain an insight into their frequency and intensity. The results revealed that 16% (47) of drought events were recorded via the SPI-12, whereas 28% (85) were identified via the RDI-12 across the ten selected stations, indicating varying degrees of severity. The remaining 84% (253) and 72% (215) of the observed periods were classified as normal and wet, respectively, on the basis of SPI-12 and RDI-12 in the southeastern region of Ethiopia (Fig. 3).

The outcomes of the annual SPI and RDI indicate that the majority of observations fall within the "Normal & Above" category, indicating that most of the time there is no significant rainfall deficit. However, only a small percentage of the RDI category is in the "moderate", and "mild" category, suggesting periods of slightly below-average precipitation. The percentages for the extreme and severe categories are provided, suggesting that in both cases rainfall deficits occur but conditions are relatively rare.

This condition can disrupt the delicate balance of forest ecosystems, affecting not only tree health but also soil composition, wildlife habitat, and overall biodiversity. Understanding the frequency distributions of these indices can provide valuable insights into drought patterns and their potential impacts on forestry. These interpretations of them can assist in drought monitoring, mitigation, and planning for resilience measures.

3.3 Drought intensity

The data indicate the severity of drought conditions categorized into extreme, severe, moderate, and possibly mild levels on the basis of the annual SPI and RDI indices. These categories represent different degrees of water deficiency and stress on vegetation, including forests. The findings indicate that within the span of 30 years in southeast Ethiopia, 15% of extreme, 28% of severe, and 57% of moderate drought events were based on the SPI-12, and 10% of extreme, 15% of severe, 29% of moderate and 46% of mild drought events were based on the RDI-12 (Fig. 4). These percentages likely represent the frequency or occurrence of each intensity category within the dataset for all study areas.

The SPI shows a greater percentage of moderate drought conditions than severe or extreme droughts conditions. Conversely, the RDI indicates a greater percentage of mild drought conditions, followed by moderate and severe droughts (Fig. 4). The highest intensity droughts were observed at Kebridehar, with the value of -3.90 for SPI-12 and − 3.89 for RDI-12 during the 2010/2011 period, among the total ten selected stations.

Extreme and severe drought categories, as indicated by lower SPI and RDI values, can have severe impacts on forest health. While moderate droughts may not cause immediate catastrophic damage, they can still impact vegetation health and productivity. Even mild drought conditions can affect vegetation, although the impact may be less severe than that of more intense drought levels. Owing to these different intensities, reduced growth rates, increased susceptibility to pests and diseases, and changes in species composition may occur. Stress-induced physiological changes in vegetables and decreased resilience to other environmental stressors can also be observed. In the context of drought assessment and management, the indices help in quantifying and monitoring drought conditions.

3.4 Drought frequency

Temporal analysis of historical data revealed fluctuations in drought occurrence across the years. The SPI and RDI provide quantitative measures of drought severity and frequency on the basis of precipitation and temperature data. The analysis reveals a total of 47 instances of annual drought events based on SPI-12, spanning from mild to extreme intensity, and 85 occurrences based on RDI-12, ranging from moderate to extreme intensity, with irregular cycles observed across all stations. Within the study period, individual stations experienced drought events ranging from 4–6 times on the basis of the SPI-12 and from 6–11 times on the basis of the RDI-12 (Table 6). The investigation of the reoccurrence of drought events at each selected station confirms the evaluation of the future occurrence of drought depending on the last drought year.

For example, at the Adigala, Kebribeya, and Dolloado stations, drought events were recorded 10 times according to RDI-12 and 5 times at the Adigala, Kebribeya, Shinile, and Ime stations, with 6 occurrences at the Dollo Ado and Daror districts based on SPI-12 (Table 6). In the Siti zone, drought events are the main hazard affecting the zone and the area has already experienced extended drought conditions from 2011–2013 (F et al., 2016) which confirms the drought frequency of Shinile which is the capital city of the Siti zone.

Table 6

Frequency of drought intensity
Stations	Drought Intensity
Stations	Extreme		Severe		Moderate		Mild	Total
	SPI	RDI	SPI	RDI	SPI	RDI	RDI	SPI	RDI
Adigala	1	1	1	1	3	3	5	5	10
Gursum	-	-	2	3	2	2	2	4	7
Jijjiga	1	1	2	2	1	1	4	4	8
Kebribeya	1	1	-	1	4	3	5	5	10
Shinile	1	1	-	-	4	4	6	5	11
Dolo Ado	-	-	1	1	5	5	4	6	10
Ime	1	2	4	2	-	1	2	5	7
Daror	1	1	-	-	5	3	4	6	8
Kebridehar	1	1	1	1	1	1	3	3	6
Gode	-	-	2	2	2	2	4	4	8
Totals	7	8	13	13	27	25	39	47	85

Both indices provide valuable information for understanding the frequency of drought events in a given area, which can assist in planning and implementing drought mitigation strategies, water resource management, and agricultural planning. Importantly, the interpretation may vary on the basis of the specific context, study area, and methodology used to calculate these indices.

Understanding the frequency distribution of drought events, as represented by SPI and RDI categories, is crucial for forestry management. This information can help forestry professionals anticipate recurring drought patterns and implement proactive measures to mitigate their effects. For example, if extreme drought events occur infrequently but with severe consequences, forestry management strategies may focus on enhancing forest resilience through measures such as species diversification, improving soil water retention, and implementing drought-tolerant forest management practices.

3.5 Comparison of Drought Indices

Comparative analysis of SPI, RDI, and PD provided insights into their effectiveness in assessing the relationships among drought indices in the study area during the spinning period. SPI shows a strong correlation with RDI and PD in identifying drought events. RDI also had a strong correlation with PD (Table 7). The correlation values of these indices confirm the occurrence of drought events at the selected stations. Understanding the frequency distributions of these indices can provide valuable insights into drought events and their potential impacts on forestry.

Complement SPI and RDI by highlighting extreme events, which are critical in understanding the strength of combined drought dynamics. Integrating multiple indices enhances the robustness of drought assessment and enables stakeholders to make informed decisions regarding drought mitigation and adaptation measures. Understanding the relationships among drought indices is crucial for informing forest management decisions. For example, forest managers may implement adaptive strategies such as thinning, prescribed burning, and selective tree species planting to increase resilience to drought stress and minimize negative impacts on forest ecosystems.

Table 7

Relationships among drought indices
Correlation Value
Station	SPI & RDI	RDI & Decile	SPI & Decile
Adigala	0.988	0.946	0.970
Gursum	0.986	0.922	0.930
Jigjiga	0.992	0.962	0.973
Kebribeya	0.990	0.910	0.926
Shinile	0.953	0.885	0.946
Dorer	0.998	0.904	0.915
Dolloado	0.999	0.977	0.978
Gode	0.999	0.909	0.913
Ime	0.994	0.942	0.943
Kebridehar	0.999	0.848	0.847

3.6 Drought Duration

The duration of drought was assessed by examining the period from its onset to its end, which was determined through monthly negative SPI and RDI values. Across most of the selected stations, the drought duration ranged from 2–4 months for both the SPI and RDI cases. The longest recorded drought duration was seven months, which was observed at Adigala station during the 2008/2009 period. Additionally, a five-month drought duration was noted at the Shinile station during the years 2001/2002 and 2008/2009.

In response to the Somali Region, the Drought Recovery and Preparedness Project 2008/09 was severe drought year conformed that prolonged drought had occurred in Adigala and Shinile Districts (OXFAM GB, 2013). Prolonged periods of these drought intensities can lead to reduced soil moisture, affecting tree growth and survival. In addition to increased tree mortality, reduced productivity, and vulnerability to pests and diseases are common consequences.

3.7 Drought Areal Extent

Considering the areal coverage of drought severity over different periods, a classification system is likely used to categorize the severity of drought events. Areal coverage suggests how much land area was affected by different levels of drought severity during a specific time period. The map indicates different levels of drought severity, ranging from normal to severe drought and possibly other classifications such as extreme drought and moderate drought. The examination of drought intensity across ten selected stations in the south eastern region of Ethiopia reveals spatial disparities (Figs. 5a-e). This is essential for considering the geographic distribution and impact of drought events. These classifications are crucial for understanding the intensity of drought events during each time period.

Specifically, in the study area, the 2011/2012 drought year predominantly affected the northern regions, whereas the 1991/1992 and 1998/1999 drought impacts were concentrated in the southern parts. The years 2010/2011 and 2001/2002 experienced widespread droughts across the entire region, albeit with varying intensities (Figs. 5a-e). The lack of rainfall in 2001/2002 led to severe challenges in the Shinile district, significantly affecting pasture and water availability, and resulting in substantial livestock losses, particularly among cattle (Paper et al., 2003).

The spatial variations in drought intensity underscore the diverse climatic conditions prevalent across the regions that assess the spatial extent of drought-affected areas. Areas characterized by lower elevations or diminished precipitation levels are more susceptible to frequent drought occurrences, posing significant challenges to agricultural and water resource management efforts. The provided data and images offer insights into the intensity of drought conditions and the strength of potential impacts on specific regions.

Visual representations of the data by map illustrating the distribution of drought intensity categories over time across specific regions. These images can help visualize trends, patterns, and fluctuations in drought severity, aiding forestry professionals in monitoring and managing forest resources. By analyzing these images, forestry experts can identify periods of heightened drought stress, assess the spatial extent of drought-affected areas, and prioritize management interventions, such as implementing forest conservation measures, adjusting harvesting practices, or initiating reforestation efforts in severely impacted regions.

3.8) Implications for Water Resource Management and Agriculture

The findings of this study hold significant implications for water resource management and agricultural practices in the southeastern region of Ethiopia. Understanding the frequency and severity of drought events is crucial for devising effective mitigation strategies, such as water conservation measures, crop diversification, and improved irrigation techniques. Additionally, the identification of vulnerable areas with high drought frequencies can aid in targeted interventions and resource allocation to support communities reliant on agriculture and livestock.

By identifying areas prone to frequent drought events and understanding the underlying climatic drivers, policymakers and stakeholders can implement targeted interventions to increase resilience and mitigate the impacts of drought. This may include the development of drought-resistant crop varieties, the expansion of water harvesting and storage infrastructure, and the promotion of sustainable land management practices.

3.9) Limitations and Future Research Directions

It is essential to acknowledge the limitations of this study, including the reliance on historical meteorological data and the inherent uncertainties associated with drought indices. Future research efforts could focus on integrating additional datasets, such as soil moisture and vegetation indices, to improve the accuracy of drought assessment. These methods could also include remote sensing data and advanced modelling techniques to increase the accuracy of drought assessment and prediction. Moreover, conducting on-the-ground assessments and engaging local communities and stakeholders in the research process would provide valuable insights into the socioeconomic impacts of drought and inform adaptive strategies at the grassroots level.

In conclusion, this study provides a comprehensive analysis of drought events in the south eastern region of Ethiopia from 1987–2018, utilizing the PD, SPI and RDI indices. The findings underscore the temporal and spatial variability of drought events and highlight the importance of multi-index analysis for robust drought assessment. The consistent occurrence of rainfall deficiency across the representative stations in the specified years underscores the region's susceptibility to drought. Understanding the severity and frequency of drought events through indices is critical for assessing their impact on vegetation and implementing adaptive management practices to increase forest resilience.

Decision-making on the basis of quantitative index-based values is essential for the appropriate and accurate assessment of drought severity and as input into an operational comprehensive drought plan. Analyzing drought indices and associated data can help researchers and forest managers assess the vulnerability of forest ecosystems to drought stress, identify areas at risk, and develop strategies for mitigating drought impacts through forest management practices such as prescribed burning, thinning, and watershed management. Additionally, monitoring drought conditions over time is essential for adaptive management and planning for climate change resilience in forestry.

By elucidating the dynamics of drought patterns, this research offers valuable insights for policymakers, researchers, and practitioners, who contribute to informed decision-making processes aimed at building resilience to climate extremes and safeguarding livelihoods in drought-prone regions. On the basis of this evidence forestry management strategies may include promoting climate-resilient tree species, implementing sustainable harvesting practices, and improving water management techniques. This highlights the urgent need for effective drought mitigation and adaptation strategies to protect vulnerable communities from the impacts of drought. Furthermore, drought can exacerbate other environmental stressors, such as wildfire risk, which can have devastating effects on forest ecosystems.

Understanding these dynamics is crucial for effective forest management and conservation, especially in mitigating the adverse effects of drought on forest ecosystems and ensuring their long-term resilience and sustainability. Collaboration among forest managers, researchers, policymakers, and local communities is essential for developing holistic approaches that address the complex challenges posed by these drought intensities around nominated representative districts.

Overall, this study underscores the importance of drought characterization as a foundational step toward establishing effective early warning systems and building resilience against the impacts of drought. By bridging the gap between water availability, supply, and demand in drought-affected areas, informed decision-making can mitigate the adverse effects of drought, safeguarding livelihoods and promoting sustainable development in southeast Ethiopia.

Competing Interests

All the authors declare that they have no financial interests. The authors have no relevant financial or nonfinancial interests to disclose. The authors have no conflicts of interest to declare that are relevant to the content of this article. All the authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or nonfinancial interest in the subject matter or materials discussed in this manuscript. The authors have no financial or proprietary interest in any material discussed in this article. The authors are responsible for the correctness of the statements provided in the manuscript.

Funding

This study was funded by Ethiopia Forestry Development. All the authors have the relationship with Ethiopian forestry development as employment. The organization does not gain or lose financially through publication of this manuscript.

Author Contributions

Data collection and analysis, interpretation of the data and the first draft of the manuscript were written by the corresponding author, Mr. Kedir Kemal. All the authors commented on previous versions of the manuscript. Mr. Moges Molla contributed to the study conception and design. Dr. Zenebe Mekonin and Dr. Abebaw Shimelis revised the work critically for important intellectual content. All the authors read and approved the final manuscript. They agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Mr. Moges Molla and Dr. Zenebe Mekonin approved the version to be published.

Data availability

The datasets analzsed during the current study were generated and collected from the Ethiopia Meteorological Institute. The datasets are not publicly available due to the data policies of the institute. But the generated data are available from the corresponding author upon reasonable request.

De Jong R, de Bruin S, de Wit A, Schaepman ME, Dent DL (2011) Analysis of monotonic greening and browning trends from global NDVI time-series. Remote Sens Environ 115(2):692–702
Degefu MA, Bewket W (2013) Trends and spatial patterns of drought incidence in the omoghibe river basin. Ethiopia Geogr Annaler: Ser Phys Geogr 10.1111// geoa.12080
Degefu MA, Bewket W, Amha Y (2022) Evaluating performance of 20 global and quasi-global precipitation products in representing drought events in Ethiopia I: Visual and correlation analysis. Weather Clim Extremes 35:100416
F A, W KSLGTS, A., M, A., Z, Y.,A, S (2016) Child Malnutrition and Associated Factors among Drought Affected Pastoralists in the Somali Region, Eastern Ethiopia. J Nutr Health Sci 3(4). https://doi.org/10.15744/2393-9060.3.402
Julien Y, Sobrino JA, Mattar C, Ruescas AB, Jimenez-Munoz JC, Soria G, Hidalgo V, Atitar M, Franch B, Cuenca J (2011) Temporal analysis of normalized difference vegetation index (NDVI) and land surface temperature (LST) parameters to detect changes in the Iberian land cover between 1981 and 2001. Int J Remote Sens 32(7):2057–2068
IPCC II (2014) Climate change 2014; Impacts, adaptations and vulnerability, Working Group II Contribution to the Fifth Assessment Report of Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, MA
Karavitis CA (1999) Decision support systems for drought management strategies in metropolitan Athens. Water Int 24(1):10–21
Keyantash J, Dracup J (2002) The quantification of drought: an evaluation of drought indices. Bull Am Meteorol Soc 83(8):1167–1180
Mehta D (2022) Temporal analysis of rainfall and drought characteristics over Jalore District of S-W Rajasthan. 17(1), 254–267. https://doi.org/10.2166/wpt.2021.114
Morid S, Smakhtin V, Moghaddasi M (2006) Comparison of seven meteorological indices for drought monitoring in Iran. Int J Climatology: J Royal Meteorological Soc 26(7):971–985
Paulo A, Paulo AA, Pereira LS (2006) Drought Concepts and Characterization Drought Concepts and Characterization. March. https://doi.org/10.1080/02508060608691913
Paulo AA, Rosa RD, Pereira LS (2012) Climate trends and behavior of drought indices based on rainfall and evapotranspiration in Portugal. Nat Hazards Earth Syst Science) 12(5):1481–1491. 10.5194/nhess-12-1481-2012
Sen PK (1968) Estimates of the regression coefficient based on Kendall’s tau. J Am Stat Assoc 63(324):1379–1389
Shatanawi K, Rahbeh M, Shatanawi M (2013) Characterizing, monitoring and forecasting of drought in Jordan river basin. J Water Resour Prot 5(12):1192–1202
Smith DMS, McKeon GM (1998) Assessing the historical frequency of drought events on grazing properties in Australian rangelands. Agric Syst 57(3):271–299
Sobrino JA, Julien Y (2011) Global trends in NDVI-derived parameters obtained from GIMMS data. Int J Remote Sens 32(15):4267–4279
Szinell CS, Bussay A, Szentimrey T (1998) Drought tendencies in Hungary. Int J Climatology: J Royal Meteorological Soc 18(13):1479–1491
Tian F, Wang Y, Fensholt R, Wang K, Zhang L, Huang Y (2013) Mapping and evaluation of NDVI trends from synthetic time series obtained by blending Landsat and MODIS data around a coalfield on the Loess Plateau. Remote Sens 5(9):4255–4279
Tsakiris G, Nalbantis I, Vangelis H, Verbeiren B, Huysmans M, Tychon B, Jacquemin I, Canters F, Vanderhaegen S, Engelen G, Poelmans L, De Becker P, Batelaan O (2013) A System-based Paradigm of Drought Analysis for Operational Management. Water Resour Manage. https://doi.org/10.1007/s11269-013-0471-4
Tsakiris G, Pangalou D, Vangelis H (2007) Regional drought assessment based on the Reconnaissance Drought Index (RDI). Water Resources Management. https://doi.org/10.1007/s11269-006-9105-4
Vangelis H, Tigkas D, Tsakiris G (2013) The effect of PET method on reconnaissance drought index (RDI) calculation. J Arid Environ 88:130–140
Wilhite DA (1993) The enigma of drought. Drought assessment, management, and planning: Theory and case studies. Springer, pp 3–15
Wilhite DA (1993) Drought assessment, management, and planning: theory and case studies. Natural Resource Management and Policy Series, vol 2. Kluwer
Wilhite DA, Glantz MH (1985) Understanding: the drought phenomenon: the role of definitions. Water Int 10(3):111–120
Zargar A, Sadiq R, Naser B, Khan FI (2011) A review of drought indices. Environ Reviews 19:333–349

The authors declare no competing interests.

Download PDF

Version 1

posted

You are reading this latest preprint version

Characterize of Drought Events via the Decile, SPI and RDI in Southeast Ethiopia

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods

2.1 Description of the study area

2.2 Data source

2.3 Methods

2.3.1 Precipitation Deciles (PD)

2.3.2 Standardized Precipitation Index (SPI)

2.3.3 Reconnaissance Drought Index (RDI)

3. Results and Discussions

3.1 Precipitation Deficiency Years

3.1.1 Precipitation Threshold Value

3.2 Drought characterization

3.3 Drought intensity

3.4 Drought frequency

3.5 Comparison of Drought Indices

3.6 Drought Duration

3.7 Drought Areal Extent

3.8) Implications for Water Resource Management and Agriculture

3.9) Limitations and Future Research Directions

4. Conclusion

Declarations

Competing Interests

Funding

Author Contributions

Data availability

References

Additional Declarations

Status:

Version 1