Climate Change Implications on the Rangelands Condition in West Guji and Borana Zones, Southern Ethiopia
Tesfaye Dejene
1
(
Wondo Genet College of Forestry and Natural Resources, Hawassa University, Shashemene, Ethiopia
)
Gemedo Dalle
2
(
Center for Environmental Science, Addis Ababa University, Addis Ababa, Ethiopia
)
Teshale Woldeamanuel
3
(
Wondo Genet College of Forestry and Natural Resources, Hawassa University, Shashemene, Ethiopia
)
Muluken Mekuyie
4
(
Wondo Genet College of Forestry and Natural Resources, Hawassa University, Shashemene, Ethiopia
)
Keywords:
Abstract :
Climate Change Implications on the Rangelands Condition in West Guji and Borana Zones, Southern Ethiopia
Tesfaye Dejene A*, Gemedo Dalle B, Teshale Woldeamanuel C, Muluken Mekuyie C
A PhD student, Wondo Genet College of Forestry and Natural Resources, Hawassa University, Shashemene, Ethiopia *(Corresponding author), E-mail: tesfayedejene2011@gmail.com
B Associate Prof., Center for Environmental Science, Addis Ababa University, Addis Ababa, Ethiopia.
C Assistant Prof., Wondo Genet College of Forestry and Natural Resources, Hawassa University, Shashemene, Ethiopia.
Abstract. Climate change is one of the most serious threats to rangeland. The implications of climate change on the rangelands conditions of the Duda and Gomole in Southern Ethiopia were examined. Measurements of rangeland were made using field survey and focus group discussions. Descriptive statistics, one way ANOVA was employed to detect whether significant difference exists between range conditions, while Pearson bivariate correlation used to test the relationships between climate factors and herbaceous plants. The Mann-Kendall test was used to determine the trends of rainfall and temperatures over times; meanwhile Sen's slope estimator was used to test the magnitude of changes. The results show that over 38 years (1981-2018) no significant annual rainfall trend was in both Duda and Gomole rangelands, despite a slight increase in rainfall. However, the annual minimum temperature significantly increases in Duda rangeland, while the annual maximum temperature significantly decreases in Gomole rangeland. The major herbaceous species in Duda rangeland (Cenchrus ciliaris, Cynodon dactylon, Digitaria naghellensi, Digitaria milanjiana and Panicum maximum) had positive correlation with annual rainfall and negative correlation with minimum temperature between 2010 and 2017 years. In the Gomole rangeland, Chrysopogon aucheri positively correlated with annual rainfall, whereas Cenchrus ciliaris and Bothriochloa insculpta negatively correlated with minimum temperature between 2009 and 2018 years. The Duda rangeland was in poor condition where that of Gomole was in fair condition. This difference was mainly due to slight difference between climate of Duda (arid and semi-arid) and Gomole (arid) of the rangelands. The multiple linear regression model result showed that changes in rainfall and temperature accounted for 46% of variability on Duda rangeland condition, whereas rainfall changes accounted for 32% of variability on Gomole rangeland condition. Therefore, the present study suggests rangeland rehabilitation in the short run through soil and water conservation, bush clearing and burning, and the planting of desirable grass species are critical to reduce the effects of climate change on rangeland.
Key words: Climate Change; Rangeland Condition; Herbaceous; Southern Ethiopia
Introduction
Rangelands are lands characterized by primarily grasses, grass-like plants, forbs, and shrubs as native flora (Joshi et al., 2004). Rangelands cover more than 40% of the earth’s geographical area, including 43% of Africa (Kassahun et al., 2008) and 62% of Ethiopia (Dalle et al., 2006; Hoffman & Vogel, 2008). Millions of pastoral and agro-pastoral communities rely on rangelands for food and economic survival (Godde et al., 2019). Rangelands provide livelihoods for approximately 27% of Ethiopians (Kassahun et al., 2008). However, climate change, fire, and land-use conversion all have global impacts on rangeland (Naidoo et al., 2013). As a result, rangeland ecosystems are changing at an unprecedented rate, as evidenced by the propagation of woody plants and the modification of herbaceous species (Naidoo et al., 2013; Zerga, 2015). Furthermore, climate change is visible in the rangeland environment as rising temperatures and changes in precipitation, pose challenges to biodiversity and people's livelihoods (Ericksen et al., 2011). Rising temperatures, for example, appear to be reducing rangeland ecosystems' potential to retain water for plant growth (Kariuki et al., 2009). As a result, changes in vegetation cover have undoubtedly been observed. It accelerates the replacement of herbaceous vegetation with woodlands in the world's warm rangelands (Mckeon et al., 2009; Herrero et al., 2016; Briske, 2017; Hudson et al., 2019).
The sub-Saharan African rangelands, which include the East African rangelands, are really hard hit by the increasing water scarcity as a result of climate change (Ericksen et al., 2011). Moreover, changes in vegetation distribution in rangelands may be a cause of further change and have primarily become a driver for runoff in grazing lands, soil condition decline, and thus range condition deterioration (Ericksen et al., 2011; Briske, 2017). Furthermore, in arid and semiarid rangelands, rising human and livestock populations have significantly exacerbated rangeland condition deterioration through land use conversion caused by settlement expansions and overgrazing of grazing lands, resulting in a lack of fodder availability for livestock production and, ultimately, the pastoral livelihood failure (Yilma et al., 2009Briske, 2017; Getachew, 2017).
Although numerous studies have been conducted in southern Ethiopian rangelands to assess rangeland conditions, vegetation distribution, and the effects of rangeland management strategies and physical environment on rangeland conditions (Oba et al., 2000; Angassa et al., 2006; Dalle et al., 2006b). There has not yet been a study that specifically addresses the implications of climate change on the rangeland condition of Southern Ethiopia. This study was conducted to better examine the implication of climate change on the Duda and Gomole rangeland conditions in the West Guji and Borana zones of southern Ethiopia. Specifically, it was attempted to analyze the association between the rangeland climate and herbaceous plant, and the implications of climate change on rangeland conditions.
Materials and Methods
Study area
The research was conducted in the rangelands of the West Guji and Borana zones in southern Ethiopia. The geographic locations of the West Guji zonal areas are found between 5o 05' N to 5o 74' N and 37o 8′ E to 38o 8′ E. The west Guji zone has a total land area of 18,577 km2 with altitude varies between 700 and 3500m. The Borana zonal stretches from 3°51′ N to 5°32′ N and 36°6′ E to 39°74′ E, with a total land area of 95,000 km2 and altitude varies between 750 and 2000m (Coppock, 1994; Dalle et al., 2006a). The first sample rangeland is the Duda rangeland of West Guji zone, is found in 5o05′ and 5o29′ N and 37o90′ and 38o50′ E with its altitude varies between 936 and 2017m. The Gomole rangeland of the Borana zone, is the second sample rangeland, and is situated between 3°97′ and 5°08′ N and 37°80′ and 39°90′ E with altitude varies between 801 and 2251m (Fig 1).
Fig. 1. Location map of the study rangelands
The topographies of the study rangelands are plain land escapes with a volcanic cones, and depressions (Dalle et al., 2006a). The study rangelands have four major soil types: red soil (Biyye Diimma/Wayama), black soil (Koticha), white or gray soil (Biyye Adii), and sandy soil (Mansaa), which account for the majority of the soil types in Southern Ethiopia (Coppock, 1994; Angassa & Baars, 2000; Tolera & Abebe, 2011). The rangelands of Southern Ethiopia are primarily covered by grasslands, perennial herbaceous vegetation, and woody vegetation (Homann, 2004). In general, four major vegetation types characterize the vegetation of rangelands in Southern Ethiopia:
i) evergreen and semi-evergreen bush land and thickets,
ii) Acacia and Commiphora trees,
iii) shrubby Acacia, Commiphora, and allied genera, and
iv) dwarf shrub grassland or shrub grassland (Angassa, 2005; Dalle et al., 2006b).
The climate of southern Ethiopian rangelands is arid and semiarid, with isolated pocket areas experienced sub-humid climate (Coppock, 1994; Oba & Kotile, 2001). The average annual temperature in the West Guji rangelands is between 24 and 29°C, while the average annual temperature in the Borana rangelands is between 27 and 29°C (Homann, 2004). The maximum temperature in the Duda rangeland ranges between 26.5 and 29.6°C, while the maximum temperature at the Gomole rangeland varies between 26.9 and 29.4°C. The minimum temperature in the Duda rangeland, on the other hand, varies between 13.9 and 16.2°C, while it ranges between 13.9 and 16.3°C in the Gomole rangeland. Furthermore, the annual rainfall in the Borana rangelands ranges from 400 to 700 mm, while the mean annual rainfall in the West Guji rangelands ranges from 550 to 700 mm (Berhanu, 2019). The annual rainfall in the Duda rangeland ranges between 600 and 700 mm, while annual rainfall in the Gomole rangeland ranges between 400 and 650 mm. In general, the rainfall pattern in Southern Ethiopian rangelands is bimodal, with 1) long rainy season or spring rainfall (Bokkeya Ganna) incorporating rainfall from the months of March, April, and May, and 2) short rainy season or autumn rainfall (Bokkeya Hageyya) incorporating rainfall from the months of September, October, and November (Dejene, 2020).
Sampling Method
West Guji and Borana zones were purposely chosen for this study, because comprehensive research on the implications of climate change on rangeland condition has not been conducted. So that, the Duda rangeland was chosen out of 4 west Guji zonal rangelands (Duda, Dawa, Cari and Galana rangelands) and the Gomole rangeland was selected out of 5 Borana zonal rangelands (Gomole, Wayama, Malbe, Dire, and Golbo rangelands). The selections of grazing lands were through Kebeles1 and Reeras2, were categorized into different grazing land management systems by the researchers. The grazing lands in the Duda rangeland's sampled Kebeles were classified into three management systems: communal, private, and open grazing lands. Similarly, total grazing lands in the Gomole rangeland's sampled Reeras were classified as communal, open, or Government Ranch. According to the study, the communal grazing land is thorny bush land that has been set aside for livestock grazing during the dry season. Private grazing land is owned by every family or household and is not subject to any resource utilization rules or regulations. Open grazing land is public land that is continuously grazed throughout the year, with no time for the grass species to recover from livestock grazing. The Government Ranches are government-owned grazing lands that are primarily used for livestock fattening, and dairy farming and are subject to stringent formal rules and regulations governing natural resource utilization. Therefore, three communal grazing lands (Dididimtu Kura, Liban Ilme, and Hema), one private (Urag Karole), and two open (Ilala Sara and Jigesa) were purposely selected from Duda rangeland. On the other hand, three communal grazing lands (Yametu Golba, Harboro, and Didahara), two open grazing lands (Kella Kufa and Yametu), and one government ranch (Dida Tuyura Ranch) were purposely selected from the total grazing lands (communal, open, and Government Ranch) in Gomole rangeland.
Prior to range condition data collection from the field, transects were randomly placed in each sample grazing land. Two 400 m transect lines (2 x400m =800m) were placed in each grazing land. A total of 6-7km transects were laid randomly in six grazing lands with different management systems in three sample Kebeles and Reeras in the Duda and Gomole rangelands, From each grazing land, ten 400 m2 (20×20m) sample plots were chosen randomly from the Duda and Gomole rangelands. The range condition factors were evaluated in a nested plot of 1 m2 (1×1m).
Participant selection for focus group discussion (FGD)
FGD participants were sought to identify trends of range condition, herbaceous and bush species in the grazing lands of various management systems in the study rangelands. Thus, five participants were chosen purposely from one grazing land users in the Kebele of the Duda rangeland and five participants were also picked from one grazing land users in the Reera of the Gomole rangeland. A total of 45 peoples were chosen from three grazing lands in three Kebeles of the Duda rangeland (e.g., 5 peoples × 3 grazing lands × 3 Kebeles = 45 peoples). Similarly, 45 peoples were chosen from 3 grazing lands in 3 Reeras in the Gomole rangeland (e.g., 5 peoples × 3 grazing lands × 3 Reeras = 45 peoples).
Data collection
a) Climate data
Ethiopia's National Meteorological Agency (NMA) has provided merged station-satellite climate data (monthly minimum and maximum temperatures, as well as monthly rainfall) for the Kebeles of Duda and Reeras of Gomole rangelands for the past 38 years (1981-2018). Experts from the national meteorological agency evaluated the data's quality. Data for the Duda and Gomole rangelands were downloaded at 4.16×4.16 km intervals with a grid cell size of 0.0375 degrees.
b) Range condition factors data
Rangeland condition data were collected at the end of the long rainy season, which lasted from early June to the end of August 2021. This time period was chosen for data collection because most grass species were expected to be in blooming stage. The assessment was carried out across rangeland grazing lands, following the paths of Angassa and Dalle (Angassa et al., 2006; Dalle et al., 2006b). Grazing lands were sampled for assessment of range condition factors such as grass species botanical composition, basal cover, litter cover, number of seedlings, age distribution, soil erosion, and soil compaction. In the Duda rangeland, for example, data were collected from grazing lands of communal, private, and open. Similarly to the Gomole rangeland, the required information was gathered from communal, open, and Government Ranch grazing lands. Thus, the grass composition, basal cover, and litter cover were rated from 0 to 10, whereas grass seedlings, age distribution, soil erosion, and compaction were rated from 0 to 5 (Dalle et al., 2006b). The following rangeland condition ratings were implemented: very poor (10 points), poor (11-20 points), fair (21-30 points), good (31-40 points), and excellent (41-50 points) (Angassa et al., 2006). A total rangeland condition score was obtained from sum values of seven condition factors: Plant composition, basal cover, and litter cover, number of seedlings, age distribution, soil erosion and soil compaction.
c) Focus group discussion (FGD) data
FGD was used to collect community perception about trends in range condition factors as well as trends in herbaceous and bush species. Three focus groups were conducted separately with users of similar grazing land. Thus, 15 people took part in one FGD in the Duda or Gomole rangelands. In general, three separate FGDs were conducted in the Duda rangeland with users of communal, private, and open grazing lands, as well as three separate FGDs in the Gomole rangeland with users of communal, open, and Government Ranch grazing lands.
The second hand herbaceous plant data for Duda rangeland between 2010 and 2017 was obtained from the Agriculture and Natural Resource office of Dugda Dawa district in West Guji. Similarly, herbaceous plant data of Gomole rangeland from 2009 to 2018 was obtained from the Agriculture and Natural Resource office of Gomole district in Borana Zone.
Data analysis
a) Mann Kendall trend and Sen’s slope
Mann-Kendall trend test and Sen's slope non parametric test tools were used to estimate the trend of temperature and rainfall for the study areas over the last 38 years (1981-2018), whereas Sen's slope estimator was used to estimate the magnitude of temperature and rainfall trend or change (Ayal et al., 2018).
Eqs 1, 2, represents the Mann-Kendall trend test:
Eq1
Eq2
Where:
S is the Mann-Kendal trend test statistic;
xi and xj are the season/annual time series sequential data values, where i and j (j > I and
N is the time series length (Eshetu et al., 2018).
If xj - xi > 0, sign (xj - xi) = 1;
if x = 0, sign (xj - xi) = 0; and
if xj - xi = 0, sign (xj - xi) = -1.
As a result, if the p-values are greater than the significant level, there is a statistically significant trend vice versa.
Sen's slope estimator was used to test the slope of the trend. Eq3 represents Sen's slope estimator mathematically (Eq. 3).
Eq3
Where:
xi and xj are the values of the variable at time steps i and j, respectively.
A value close to zero indicates that there is no change in slope.
A negative slope value represents the strength of a negative trend, while
A positive slope value represents the strength of a positive trend.
b) Inferential statistics
Analysis of variance (ANOVA) was used to test whether there was any significant difference in range conditions and factors among different grazing lands.
Pearson bivariate correlation was used to test the level of association between climate and herbaceous plants.
Multiple linear regression method was used to estimate the effect of climate change on the rangelands conditions in Duda and Gomole. The dependent variable (rangeland condition) and independent variables (rainfall and temperature), Eq4 denotes the model specification:
Eq4
Where:
Y is rangeland condition in 2021,
B0 is the constant
B1, B2, B3, B4are regression coefficients
X1 is the rate of annual rainfall change from 1981 to 2018,
X2 is the rate of seasonal rainfall from 1981 to 2018,
X3 is the rate annual mean temperature,
X4 is the trend seasonal mean temperature,
Data were statistically analyzed using STATA V.14.1 and XLSTAT V.2020 software packages.
Results
Climate trend and magnitude of change
Spring total rainfall in the Duda rangeland decreases significantly by 0.24 mm per season (P< 0.05) over 38 years (1981-2018), while autumn and annual rainfall increases over 38 years, respectively (Table 1). However, in the Gomole rangeland, spring and autumn rainfall increase by 0.08 and 0.99 mm, respectively, while annual rainfall increases by 4.10 mm between 1981 and 2018 (Table 1).
From 1981 to 2018, mean annual maximum and mean annual temperatures decreased in both the Duda and Gomole rangelands. However, the mean annual minimum temperature rises by 0.008 and 0.007°C, respectively, in Duda and Gomole rangelands, respectively (Table 2).
Table 1. Trend of seasonal and annual rainfall and magnitude of change (1981-2018) in the Duda and Gomole rangelands
Study areas | Mann-Kendall trend (Z test) | Sen’s Slope (Q) | ||||
Spring | Autumn | Annual | Spring | Autumn | Annual | |
Duda rangeland | -0.25** | 1.13 | 0.68 | -0.24** | 0.62 | 1.70 |
Gomole rangeland | 0.20 | 2.61 | 2.19 | 0.08 | 0.99 | 4.10 |
**= Significant at 0.01
Table 2. Trend of mean maximum, minimum and annual temperature and magnitude of change (1981-2018) in the Duda and Gomole rangelands
Study areas | Mann-Kendall trend test (Z) | Sen’s Slope (Q) | ||||
Tmax | Tmin | MAT | Tmax | Tmin | MAT | |
Duda rangeland | -2.0 | 1.0 | -1.2 | -0.021 | 0.008 | -0.008 |
Gomole rangeland | -0.8 | 1.8 | -0.5 | -0.009 | 0.007 | -0.003 |
Tmax= Annual maximum temperature,
Tmin= Annual minimum temperature,
MAT=Mean annual temperature,
Herbaceous plants presentation alongside rainfall and temperature conditions
Long-term herbaceous vegetation survey reports compiled by the Agriculture and Natural Resource Office Rangeland department of Dugda Dawa district in West Guji Zone and Gomole district in Borana Zone revealed a trend of herbaceous species at the Duda and Gomole rangelands.
In the Duda rangeland (Fig.2a), of all herbaceous species identified by survey reports (Cenchrus ciliaris, Chrysopogon aucheri, Cynodon dactylon, Digitaria milanjiana, Digitaria naghellensis and Panicum maximum) except Pennisetum mezianum increased between 2010 and 2017 (Fig. 2a). In relation to these, the annual rainfall and maximum temperature decreases, while the minimum temperature increases significantly ((R2=0.72) between 2010 and 2017 in the Duda rangeland (Fig.2b).
Furthermore, in the Gomole rangeland, out of the total herbaceous species (Bothriochloa insculpta, Cenchrus ciliaris, Chrysopogon aucheri, Cyperus bulbosus, Digitaria naghellensis, Panicum maximum, Pennisetum mezianum, and Sporobolus pellucidus), Chrysopogon aucheri increases, despite Bothriochloa insculpta and Cenchrus ciliari decrease between 2009 and 2018 years (Fig. 2c). Consistent with the finding, the maximum temperature significantly decreases (R2=0.47) between 2009 and 2018 years in the Gomole rangeland (Fig. 2d).
To sum up, in Duda rangeland, the trends of herbaceous species (Cenchrus ciliaris, Cynodon dactylon, Digitaria naghellensi, Digitaria milanjiana and Panicum maximum) significantly decrease, except Pennisetum mezianum increases between 2010 and 2017 years. The increase in Pennisetum mezianum was due to mainly less desirability of the species for livestock production (Table 4).
In the Gomole rangeland, the trends of highly desirable species (Cenchrus ciliaris and Bothriochloa insculpta) significantly decreased, despite Chrysopogon aucheri significantly increases in the rangeland. The increase of Chrysopogon aucheri was due to intermediate desirability (increaser) for livestock production (Table 4).
Fig. 2. Herbaceous plants presentation alongside climate elements in the study rangelands
Correlation between climate factors and herbaceous plants
The study found a positive correlation between annual rainfall and herbaceous plants like Cynodon dactylon, Digitaria milanjiana, Panicum maximum and Digitaria naghellensis (P<0.05) in the Duda rangeland, indicating by increasing annual rainfall the frequency of plant species increased. For Pennisetum mezianum here was weak negative correlation between annual rainfall in the same area. In the Gomole rangeland, there was positive correlation between Chrysopogon aucheri, with annual rainfall (P<0.05) and its frequency significantly increased over time (Table 3).
In the Duda rangeland, there was no correlation between herbaceous plants with maximum temperature, whereas in the Gomole rangeland, Digitaria naghellensis and Bothriochloa insculpta had positive correlation with maximum temperature (P<0.05), indicating their frequency were increased over times (Table 3).
In the Duda rangeland, there were strong negative correlations between the herbaceous plants such as Cenchrus ciliaris, Digitaria naghellensis, Cynodon dactylon, Digitaria milanjiana and Panicum maximum with minimum temperature (P<0.01), indicating that in by risen minimum temperature, the frequency of species were deceased (Table 3). In contrast there was positive correlation between Pennisetum mezianum with minimum temperature (P<0.01) indicating by risen minimum temperature, the frequency of this species was decreased (Table 3). In the Gomole rangeland, there were negative correlations between Cenchrus ciliaris with minimum temperature (P<0.05), similar trend was observed for Bothriochloa insculpta indicating that in by risen minimum temperature, the frequency of the two species were deceased (Table 3).
Table 3. Correlation between climate and major herbaceous plants at Duda rangeland (2010-2017) and Gomole rangeland (2009-2018)
Herbaceous plants | Duda rangeland (2010-2017) |
| Herbaceous plants | Gomole rangeland (2009-2018) | ||||
| Rainfall | T max | T min |
|
| Rainfall | T max | T min |
Cenchrus ciliaris | 0.56 | 0.18 | -0.86** |
| Cenchrus ciliaris | 0.37 | 0.57 | -0.65* |
Chrysopogon aucheri | 0.08 | -0.39 | -0.39 |
| Chrysopogon aucheri | 0.70* | 0.02 | -0.47 |
Digitaria naghellensis | 0.70* | 0.36 | -0.82* |
| Digitaria naghellensis | 0.25 | 0.81* | -0.38 |
Pennisetum mezianum | -0.54 | -0.36 | 0.86** |
| Pennisetum mezianum | 0.46 | -0.35 | -0.20 |
Cynodon dactylon | 0.69* | 0.25 | -0.83** |
| Bothriochloa insculpta | 0.33 | 0.66* | -0.54 |
Digitaria milanjiana | 0.70* | 0.36 | -0.77* |
| Cyperus bulbosus | 0.14 | -0.55 | 0.06 |
Panicum maximum | 0.67* | 0.30 | -0.80* |
| Sporobolus pellucidus | 0.28 | -0.58 | -0.30 |
T max= Annual maximum temperature,
T min= Annual minimum temperature,
MAT=Mean annual temperature;
*, **= The correlation coefficients are significant at the 0.05 and 0.01 probability levels, respectively
Range condition factors
For assessment of range condition, the collected data for grass botanical composition, basal cover, litter cover, number of seedlings, age distribution, soil erosion, and soil compaction in both reigns are as follows:
a) Grass composition
The study results revealed that 31 herbaceous species from the Duda rangeland and 33 herbaceous species from the Gomole rangeland were used as livestock fodder (Table 4). There was a significant variation regarding the grass composition among the grazing lands managed under different management systems in the Duda and Gomole rangelands (P<0.05). For example, at the Duda rangeland, grass composition remained fair in communal grazing land (score 5.0), very poor at private (3.6), and open (4.6) grazing lands. Furthermore, one-way ANOVA result showed a significant difference in grass composition among communal, open, and private grazing lands in the Duda rangeland (P<0.05). Similarly, the one way ANOVA result confirmed that there was a significant difference in the composition of grasses in the Gomole rangeland's communal, open, and government ranch grazing lands (P<0.05). The grass composition in the Duda rangeland was generally poor, whereas the composition in the Gomole rangeland was fair (Table 5).
Table 4. Herbaceous species desirability (forage value) for livestock production
Botanical name | Duda Rangeland | Gomole Rangeland |
Andropogon chinensis | Highly Desirable | Least Desirable |
Aristida kenyensis | Intermediate | Intermediate |
Barleria spinisepala | Least Desirable | Highly Desirable |
Borhrichloa radicans | Not identified | Least Desirable |
Bothriochloa insculpta | Highly Desirable | Highly Desirable |
Cenchrus ciliaris | Highly Desirable | Highly Desirable |
Chloris roxburghiana | Not identified | Intermediate |
Chlorophytum gallabatense | Highly Desirable | Highly Desirable |
Chrysopogon aucheri | Intermediate | Highly Desirable |
Commelina africana | Intermediate | Highly Desirable |
Cynodon dactylon | Highly Desirable | Highly Desirable |
Cyperus bulbosus | Intermediate | Least Desirable |
Cymbopogon commutatus | Intermediate | Intermediate |
Cyperus sp | Intermediate | Intermediate |
Dactyloctenium aegyptium | Least Desirable | Highly Desirable |
Digitaria milanjiana | Highly Desirable | Highly Desirable |
Digitaria naghellensis | Highly Desirable | Highly Desirable |
Eleusine intermedia | Intermediate | Least Desirable |
Eragrostis capitulifera | Intermediate | Intermediate |
Eragrostis papposa | Intermediate | Intermediate |
Harpachne schimperi | Intermediate | Intermediate |
Heteropogon contortus | Highly Desirable | Least Desirable |
Indigofera volkensii | Not identified | Intermediate |
Leptothrium senegalense | Intermediate | Intermediate |
Loudetia flavida | Highly Desirable | Intermediate |
Melinis repens | Least Desirable | Least Desirable |
Ozoroa insignis | Intermediate | Intermediate |
Panicum maximum | Intermediate | Highly Desirable |
Pennisetum mezianum | Least Desirable | Least Desirable |
Setaria verticillata | Highly Desirable | Highly Desirable |
Sporobolus discosporus | Intermediate | Intermediate |
Sporobolus pellucidus | Intermediate | Intermediate |
Themeda triandra | Highly Desirable | Intermediate |
b) Basal and litter cover
According to the result, communal grazing land in the Duda rangeland had poor basal and litter covers (scores of 4.3, 4.1), while private grazing land had poor basal and litter covers (3.5, 3.3), and open grazing land had also poor basal cover (3.7) and poor litter cover (4.1). However, the basal and litter covers in communal grazing land in Gomole rangeland were found fair (5.5, 5.5). Similarly, the basal and litter cover in the rangeland's government ranch were found to be fair (5.9, 5.5), whereas the basal and litter covers in the open grazing lands were poor (4.1) and very poor (3.9), respectively. Furthermore, one way ANOVA results show that there was no significant variation amongst the basal and litter covers in the Duda rangeland's communal, open, and private grazing land management systems, whereas there was a significant variation among basal and litter covers in the Gomole rangeland's amongst grazing land management systems (P<0.05). Overall, grass litter and basal covers were found to be very poor and poor in the Duda rangeland, respectively, whereas grass litter and basal covers were found to be fair in the Gomole rangeland (Table 5).
c) Number of grass seedling
In the Duda rangeland, the average score of grass seedling in the communal (score 1.7), open (1.4), and private (1.8) grazing lands remained very poor, whereas grass seedlings in the Gomole rangeland's communal (2.0) and government ranch (2.2) grazing lands found poor as compared to the very poor grass seedling in open grazing lands (1.2). To summarize, there was no significant variation among number of grass seedlings in the grazing lands of different management systems in the Duda rangeland, but there was significant variation of number of grass seedlings in the grazing lands of management systems in the Gomole rangeland (P<0.05). In general, the number of grass seedlings was very low in both the Duda and Gomole rangelands (Table 5).
d) Soil erosion and compaction
The study results show that in the Duda rangeland, communal grazing land has the lowest soil erosion (score 2.2), while private (1.7) and open (1.6) grazing lands have the highest soil erosion. Soil erosion was found to be lower in the communal (2.7) and Government Ranch (2.6) areas of the Gomole rangeland, while it seems medium in the open grazing land (2.1). Furthermore, in the Duda rangeland, soil compaction was highest in all grazing land management systems, including communal (1.9), private (1.7), and open (1.9) grazing lands, whereas in the Gomole rangeland, soil compaction was found to be medium in communal (2.5), open (2.2), and Government Ranch (2.7) grazing lands. In general, there was a significant difference in soil erosion manifestation in the Duda rangeland grazing land management systems (at P<0.05). Similarly, soil erosion in the Gomole rangeland varied significantly across grazing land management systems (at P<0.05). However, there was no significant variation in soil compaction among different grazing land management systems in both Duda and Gomole rangelands. Soil erosion was generally greater in the Duda rangeland compared to intermediate soil erosion in the Gomole rangeland (Table 5).
Overall rangeland condition
The range condition in the Duda rangeland's communal grazing land (21.2) was fair, but the condition in the rangeland’s private and open grazing lands was poor (17.0, 19.7), whereas the range conditions of all types of grazing lands remained fair in the Gomole rangeland. Government Ranch (27.6) and communal (26.6) grazing lands were found to be slightly better than open grazing land. In the Duda rangeland, there was a significant difference in range conditions between communal, private, and open grazing lands (P<0.05). Similarly, in the Gomole, there was a significant difference in range condition between communal, open, and Government Ranch grazing lands (P<0.05) (Table 5). To summarize, the Duda rangeland was in poor condition, whereas the Gomole rangeland was in fair condition (Table 5).
Table 5. Mean range condition and factors of different grazing lands management systems
Factors name (Scores) | Duda rangeland (Mean ± SD) |
| Gomole rangeland (Mean ± SD) | ||||||
| Communal | Private | Open | Mean |
| Communal | Open | Government | Mean |
Grass composition | 5.0±1.1 | 3.4±0.9 | 4.6±1.6 | 4.6 |
| 5.8±0.9 | 4.6±1.2 | 6.2±0.8 | 5.5 |
Grass basal cover | 4.3±0.9 | 3.5±1.3 | 3.7±1.5 | 3.9 |
| 5.5±0.7 | 4.1±1.2 | 5.9±0.6 | 5.1 |
Litter cover | 4.1±0.9 | 3.3±1.1 | 4.1±1.5 | 4.0 |
| 5.5±1.0 | 3.9±1.5 | 5.5±0.5 | 5.0 |
Grass seedlings No. | 1.7±0.7 | 1.4±0.8 | 1.8±0.9 | 1.7 |
| 2.0±0.7 | 1.2±1.1 | 2.2±0.9 | 1.8 |
Grass age distribution | 2.0±0.7 | 1.7±0.6 | 2.2±0.9 | 2.0 |
| 2.6±0.8 | 2.0±0.7 | 2.6±0.5 | 2.4 |
Soil erosion | 2.2±0.6 | 1.7±0.9 | 1.6±0.8 | 1.9 |
| 2.7±0.7 | 2.1±0.8 | 2.6±0.4 | 2.5 |
Soil compaction | 1.9±0.7 | 1.9±0.5 | 1.7±0.7 | 1.8 |
| 2.5±0.6 | 2.2±0.9 | 2.7±0.6 | 2.5 |
Range condition score | 21.2±2.8 | 17.0±4.1 | 19.7±5.5 | 19.9 |
| 26.6±3.9 | 20.2±6.0 | 27.6±1.7 | 24.6 |
Rangeland condition class | Fair | Poor | Poor | Poor |
| Fair | Fair | Fair | Fair |
Perceived rangeland condition trend
Pastoralists perceived that in the Duda rangeland climate change, in the form of recurrent drought, increased bush encroachment due to bun of fire was the major drivers of decreased desirable herbaceous species in communal grazing lands. Human and livestock population pressures, as well as associated grazing land overgrazing and crop cultivation expansion in rangelands, were identified as drivers of rangeland condition deterioration in the Duda rangeland's communal, private and open grazing lands.
The response from pastoralists FGD in the Gomole rangeland, on the other hand, showed that frequent drought incidences, bush encroachment, and overgrazing were the major drivers of decreased desirable herbaceous species. Likewise, the response showed that in the open and communal grazing lands, livestock population pressure, bush encroachment, and severe drought frequencies were the major causes of desirable herbaceous species change, whereas in the Government Ranch, bush encroachment, drought recurrence, and overgrazing were among the major drivers of desirable herbaceous species declining and disappearance. In general, the conditions of all grazing land systems in the rangelands have deteriorated as a result of climate change manifestations such as frequent drought and contaminant bush encroachment and herbaceous species change, and thereby water runoff and soil erosion in the grazing lands.
Furthermore, the FGD participants in the Duda rangeland revealed the most important remedial measures for their deteriorating rangeland condition. They indicated that biophysical measures or land conservation activities such as soil and water conservation, bush clearing and burning, and planting of desirable forage species in rangeland (i.e., Cenchrus ciliaris, Digitaria milanjiana, and Digitaria naghellensis) can be completed in a short period of time. Participants also indicated that improved rangeland conditions would be critical in managing the effects of climate change-induced water stress on vegetation growth and rangeland degradation problems. Furthermore, the problem of livestock forage scarcity could be addressed. Ultimately, livestock overgrazing was identified as a greater threat to rangeland sustainability by FGD participants in the Gomole rangeland, but it is primarily manageable through a shift in livestock species from grazers (i.e., cattle) to browsers (i.e., camel and goat).
Climate change implications on the conditions of the rangelands
Multiple linear regression model equation was used to estimate the effects of climate change on the conditions of the Duda and Gomole rangelands. The model results show that the coefficient of determination (R2=0.46) confirmed that 46% of the variability of Duda rangeland condition in 2021 was explained by trend of annual rainfall and trend of mean annual temperature of the rangeland (Table 6).
According to the results in Table 7, the result indicates that the trend of annual rainfall and trend of autumn rainfall accounted R2=32% of the variability of the Gomole rangeland condition of the year 2021. Besides, the total annual rainfall increment in the Gomole rangeland accompanied with the rangeland condition improvement by 11.19, and 8.29 for annual and autumn rainfall, respectively (Table 7).
Table 6. Multiple linear regression ANOVA between climate data and rangeland condition in Duda and Gomole, regions, Southern Ethiopia
Source | Duda rangeland |
| Source | Gomole rangeland | ||||||
| df | SS | MS | F |
|
| df | SS | MS | F |
Regression | 3 | 512.19 | 170.73 | 16.9** |
| Regression | 2 | 560.67 | 280.33 | 13.44** |
Residual | 56 | 590.55 | 10.54 |
|
| Residual | 57 | 1188.59 | 20.85 |
|
Total | 59 | 1102.7 | 18.69 |
|
| Total | 59 | 1749.26 | 29.64 |
|
**= The regression models are significant at 0.01 probability levels.
Table 7. Multiple linear regression equations between climate data and rangeland condition in Duda and Gomole, regions, Southern Ethiopia
Locations | Regression equation | R2 values |
Duda rangeland |
| R2=0.46 |
Gomole rangeland |
| R2=0.32 |
