Population and conservation status of the endangered Dracaena ombet tree in dry Afromontane forests

Abstract

Dracaena ombet, a flagship tree species in arid ecosystems, holds a significant ecological, economic, and socio-cultural value. However, its persistence is currently under threat from both anthropogenic and natural factors. Consequently, the species has been listed as an endangered tree species on the IUCN Red List, requiring urgent conservation actions for its continued existence. To develop effective conservation actions, it is necessary to have information on the population dynamics of the species. A study was conducted in the lowland and midland agroecological zones (sites) within the Desa'a dry Afromontane forest, northern Ethiopia to analyze the population status of D. ombet and identify its site-specific threats. At each site, abundance, health status, diameter, height and threats of the species were collected using 60 sample plots (20 m × 20 m) distributed over six transects (500 m × 20 m) spaced one km apart. The study showed that the D. ombet population was characterized by low abundance and unstable structure. It was further characterized by a substantial number of unhealthy damaged and dead trees. The low abundance of the species with unstable age structure in the dry Afromontane forests can be attributed to various factors such as stem cutting and debarking, leaf defoliation, overgrazing, soil erosion, and competition from expansive shrubs. Alternative livelihood options for the local inhabitants should be introduced to minimize the overexploitation of D. ombet for subsistence use in the dry Afromontane forests. The impacts of overgrazing and soil erosion on D. ombet and its Desa'a habitats should also be addressed through the introduction of community-based exclosures and in-situ soil and water conservation practices, respectively.

Full text

Global Ecology and Conservation 50 (2024) e02809
Available online 15 January 2024
2351-9894/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Population and conservation status of the endangered Dracaena
ombet tree in dry Afromontane forests
Tesfay Gidey
a
,
*
, Emiru Birhane
b
,
c
,
m
, Negasi Solomon
b
,
d
, Tesfay Atsbha
e
,
Ashena Manaye
d
, Hadgu Hishe
b
,
f
, Yirga Gu
g
, Musse Tesfaye
h
, Aklilu Negussie
i
,
Tˆ
ania Soa Oliveira
j
, Joao HN Palma
k
, Petr Madˇ
era
l
, Jose G. Borges
k
a
Department of Plant Science, College of Agriculture and Environmental Sciences, Adigrat University, P.O. Box 50, Adigrat, Ethiopia
b
Department of Land Resource Management and Environmental Protection, College of Dryland Agriculture and Natural Resources, Mekelle
University, P.O. Box 231, Mekelle, Ethiopia
c
Institute of climate and society, Mekelle University, P.O. Box 231, Mekelle, Ethiopia
d
Tigray Institute of Policy Studies, P.O. Box 902, Mekelle, Tigray, Ethiopia
e
Tigray Agricultural Research Institute, P.O. Box 492, Mekelle, Ethiopia
f
KU Leuven, Department of Earth and Environmental Sciences, Division Forest, Nature and Landscape, Celestijnenlaan 200E, P.O. Box 2411, Leuven
3001, Belgium
g
Ethiopian Forest Development, Mekelle Centre, Mekelle, Ethiopia
h
Institute of International Forestry and Forest Products, Technical University of Dresden, Tharandt 01737, Germany
i
WeForest, P.O. Box 25450/1000, Addis Ababa, Ethiopia
j
RAIZ - Forest and Paper Research Institute, Quinta de S˜
ao Francisco, Eixo, Aveiro 3801-501, Portugal
k
Forest Research Centre and Associate Laboratory TERRA, School of Agriculture, University of Lisbon, Tapada da Ajuda s/n, Lisbon 1349-017,
Portugal
l
Department of Forest Botany, Dendrology and Geobiocoenology, Faculty of Forestry and Wood Technology, Mendel University in Brno, Zemˇ
edˇ
elsk´
a,
Brno 1613 00, Czech Republic
m
Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences (NMBU), Ås, Norway
ARTICLE INFO
Keywords:
Agroecological zones
Desa’a forest
Ethiopia
Nubian dragon tree
Overexploitation
Persistence
ABSTRACT
Dracaena ombet, a agship tree species in arid ecosystems, holds a signicant ecological, eco-
nomic, and socio-cultural value. However, its persistence is currently under threat from both
anthropogenic and natural factors. Consequently, the species has been listed as an endangered
tree species on the IUCN Red List, requiring urgent conservation actions for its continued exis-
tence. To develop effective conservation actions, it is necessary to have information on the
population dynamics of the species. A study was conducted in the lowland and midland agro-
ecological zones (sites) within the Desa’a dry Afromontane forest, northern Ethiopia to analyze
the population status of D. ombet and identify its site-specic threats. At each site, abundance,
health status, diameter, height and threats of the species were collected using 60 sample plots (20
m ×20 m) distributed over six transects (500 m ×20 m) spaced one km apart. The study showed
that the D. ombet population was characterized by low abundance and unstable structure. It was
further characterized by a substantial number of unhealthy damaged and dead trees. The low
abundance of the species with unstable age structure in the dry Afromontane forests can be
attributed to various factors such as stem cutting and debarking, leaf defoliation, overgrazing, soil
erosion, and competition from expansive shrubs. Alternative livelihood options for the local in-
habitants should be introduced to minimize the overexploitation of D. ombet for subsistence use in
* Corresponding author.
E-mail address: tglove.gidey@gmail.com (T. Gidey).
Contents lists available at ScienceDirect
Global Ecology and Conservation
journal homepage: www.elsevier.com/locate/gecco
https://doi.org/10.1016/j.gecco.2024.e02809
Received 13 May 2023; Received in revised form 9 January 2024; Accepted 14 January 2024

Global Ecology and Conservation 50 (2024) e02809
2
the dry Afromontane forests. The impacts of overgrazing and soil erosion on D. ombet and its
Desa’a habitats should also be addressed through the introduction of community-based exclosures
and in-situ soil and water conservation practices, respectively.
1. Introduction
The genus Dracaena includes about 160–190 plant species (Govaerts et al., 2021). Some, commonly known as dragon trees, exhibit
a tree growth habit (Marrero et al., 1998). Unlike other monocots, tree-like Dracaena achieve stem and root thickening due to a
secondary thickening meristem (Hubalkova et al., 2017). These trees are mainly found in the tropics. They are native to Africa, but they
are also present in Asia, the Mediterranean, central America, and northern Australia (Govaerts et al., 2021). Dragon trees comprise 10
arborescent species (Wilkin et al., 2012), all growing in seasonally arid climates with an annual rainfall of 200–500 mm and mean
temperature between 18 and 20 ◦C (Marrero et al., 1998; Adolt and Pavlis, 2004). Dragon trees are well adapted to capture horizontal
precipitation (Nadezhdina et al., 2018) and their distribution is sometimes associated with seasonal cloud forests (De Sanctis et al.,
2013; Kalivodov´
a et al., 2020). Dragon trees are threatened by overexploitation, overgrazing and habitat degradation despite their
economic and ecological values in dryland areas (Ghazali et al., 2008; Al-Okaishi, 2020; Madˇ
era et al., 2020). The effects of these
anthropogenic pressures have been intensied by climate change in the last decade, as highlighted by Elnoby et al. (2017) and Vahalík
et al. (2023) in the case of Dracaena ombet Kotschy and Peyr. and D. cinnabari populations. These populations have also become more
scattered with small and isolated patches with unbalanced age structures, where often the tree seedlings are missing (Leng´
alov´
a et al.,
2020; Madˇ
era et al., 2020). As a result, D. ombet, D. cinnabari, D. draco, D. tamaranae and D. serrulata have been listed on the Inter-
national Union for Conservation of Nature (IUCN) Red List as threatened species (IUCN, 2017) among the total ten dragon trees (Wilkin
et al., 2012).
Dracaena ombet is a agship species of dryland areas of the north-east African region with substantial economic and ecologic
benets. For example, its fruits are edible and eaten by local communities as a supplement to their meagre diets (Ghazali et al., 2008).
The fruits and leaves are used as livestock fodder, and resin extracted from its stem is used for traditional medicine (Mohammed,
2015). Furthermore, the species is vital for the livelihood of the local people through the harvest of its plant parts for making household
utensils and farm implements (Ghazali et al., 2008; Gidey et al., 2023). It is also valuable for soil and water conservation, carbon
sequestration, shade and adaptation to the impacts of climate change (Kamel et al., 2014; Madˇ
era et al., 2020). D. ombet is native to
Egypt, Sudan, Ethiopia, Eritrea, Somalia, Djibouti and Saudi Arabia (Marrero et al., 1998; Ghazali et al., 2008), typically situated in
between 1000–1800 m altitude with an annual rainfall of 200–500 mm (Thulin, 1995; Kamel et al., 2014). However, the current
suitable habitats of the species were predicted to contract due to climate change in Ethiopia (Birhane et al., 2023) and Sudan (Andersen
et al., 2022).
In Ethiopia, two subspecies of D. ombet are found, subsp. ombet, which is the focus of this study, and throughout the paper referred
to as D. ombet occurs in the northern dry Afromontane forests (Aynekulu, 2011), while subsp. schizantha is situated on the escar-
pemnets in eastern Ethiopia to the east of the Rift Valley and northern Somalia (Bos and Teketay, 1997), mainly in open habitats, often
in association with the evergreen shrub, Buxus hildebrandtii (Kitaba, 2006). Dry Afromontane forest is one of the 14 ecosystem types of
Ethiopia and is characterized by a long dry season (October–mid-June). Its average annual temperature varies between 14 and 20 ◦C
and the annual rainfall from 700 to 1100 mm, with most of the rain recorded in mid-June to mid-September (Ethiopian Biodiversity
Institute, 2022). The ecosystem is the second most rich vegetation type in the country, and a habitat for a large number of endemic and
unique animal and plant species. Despite the substantial values of the ecosystem to humans, it has been deteriorated by various
disturbances including overexploitation and land-use change (Aynekulu, 2011; Ethiopian Biodiversity Institute, 2022; Birhane et al.,
2023).
The health and population structure of D. ombet in dry Afromontane forests is currently impacted by anthropogenic and natural
factors (Ghazali et al., 2008; Gidey et al., 2023). For example, in northern and eastern Africa, the local communities harvest intensively
the species for making various household materials (Ghazali et al., 2008; WeForest, 2018). They also overgraze its habitats (Kamel
et al., 2014; WeForest, 2018). In addition, natural phenomena such as soil erosion and encroachment by expansive shrubs occasionally
occur in the species’ habitats (Ghazali et al., 2008; Mohammed, 2015; Haile et al., 2021; Gidey et al., 2023). These factors have then
led to deforestation and degradation of the remaining populations of the species (Ghazali et al., 2008; WeForest, 2018). Nevertheless,
the impacts of anthropogenic and natural pressures on D. ombet and other dry Afromontane trees depend on the characteristics of their
natural habitats including altitudinal features and edaphic properties (Aynekulu, 2011; Birhanu et al., 2021; Hishe et al., 2021a). For
instance, Vahalik et al. (2020) found a larger number of D. serrulata trees in the higher altitudes of the Dhofar Mountains of Oman,
characterized by low pressures (e.g., livestock grazing) than at the lower altitudes. Similar situations have also been observed for
several dry Afromontane trees elsewhere (Aynekulu et al., 2016; Birhanu et al., 2021). Understanding the site-specic (altitude-based)
information related to population dynamics and threats may contribute to sustainable conservation of D. ombet in dry Afromontane
forests (Leng´
alov´
a et al., 2020; Andersen et al., 2022). This study aims 1) to assess the abundance, health status and population
structure of D. ombet; 2) to evaluate the relationships between the abundance and health status of the species, and altitude and soil
variables; and 3) to determine site-specic threats to the species and its habitats in two agroecological zones, representing different
altitudinal gradients. We hypothesized that 1) the abundance, health status and structural characteristics of the species will be higher
in the midland areas than the lowlands because of lower level of disturbances in the midland areas, 2) there will be positive corre-
lations between abundance and health status of the species, and altitude and soil variables, and 3) site-specic threats will be identied
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
3
for each study area due to higher level disturbances in the lowland areas than the midlands.
2. Methods
2.1. Study species
Dracaena ombet Heuglin ex Kotschy & Peyr. (Asparagaceae) is an evergreen, long-lived tree that grows to about 5–8 m height, with a
densely packed umbrella-shaped crown (Kotschy, 1867; Brown and Mies, 2012). Its trunk is forked, grey in colour, and divided into
several branches, with sword-shaped leaves, about 1 m long and 5 cm wide. The leaves are thick, smooth on the lower and upper
surfaces, and erect in position to reduce transpiration (Kotschy, 1867). The inner part of the branches and the roots consist of ber, this
is an indicator that the species can resist drought, and it probably depends on mist and rain fall rather than ground water (Ghazali et al.,
2008). As a result, D. ombet establishes and grows on shallow, rocky soils where no permanent ground water occurs (Ghazali et al.,
2008; Aynekulu, 2011). D. ombet has white owers, and the fruits are spherical esh berries each containing one to three small seeds
(Bauerov´
a et al., 2020). The tree produces seed once per year between April and July. Although seed production may be high, many
seeds do not germinate due to lack of viability (Ghazali et al., 2008; Birhane et al., 2023; Gidey et al., 2023). D. ombet seeds seem to
have a long dormancy period (could be more than a decade) in the soil, which enable them to germinate and establish under favourable
conditions (Ghazali et al., 2008). The seeds are manly dispersed by birds (Gonz´
alez-Castro et al., 2019), water and wind (Ghazali et al.,
2008). Vegetative propagation for D. ombet was attempted using axillary buds, but was not successful (Ghazali et al., 2008).
2.2. Study sites
The Desa’a dry Afromontane forest is located in northern Ethiopia, on the border between the Tigray and Afar regions (Fig. 1). The
forest lies between 13◦20’–14◦10’N latitude and 39◦32’–39◦55’E longitude, with an altitudinal range between 900 and 3100 m. It
covers about 154,000 ha (WeForest, 2018). The forest has high economic, ecological and biodiversity conservation, as well as climate
change mitigation and adaptation values at local and global scales (Aynekulu, 2011; WeForest, 2018). For example, it has been a
source of livelihood and other ecological services for nearly half a million people (WeForest, 2018). Desa’a was also selected by the
Ethiopian government as one of the biodiversity hubs for the implementation of the international climate change mitigation pro-
grammes such as the REDD
+
initiatives (Tetemke et al., 2019).
Desa’a is characterized by varied climate, geology and topography (Nyssen et al., 2005). Altitude is the major inuencing factor for
the amount of rainfall in the forest area (Nyssen et al., 2005). The average annual temperature ranges between 13 to 25 ◦C and rainfall
is from 400 to 700 mm (Hishe et al., 2021a). The dominant soil types are Leptosols, Cambisols, Vertisols, Regosols and Arenosols
(BoANR, 1997). The variability in climate, altitude and edaphic conditions makes the forest an important habitat for about 90 woody
Fig. 1. Geographic locations of the study sites within the Desa’a dry Afromontane forest, northern Ethiopia.
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
4
Table 1
Descriptions of indicators used to categorize different threats to D. ombet and its habitats in the Desa’a dry Afromontane forest, northern Ethiopia.
Threats Threat level
High Medium Low Absent
Stem cutting Most trees are badly cut and chopped; cut stumps are
visible
Trees are partially damaged with the stem, e.g.
branches are cut
Signs are the same with the medium but on the
sporadic scale
No signs of cutting
Stem debarking Most trees are badly debarked Some trees are sparsely debarked Trees are rarely debarked No signs of debark
Leaf defoliation Trees are badly defoliated; branches are highly detached Trees are defoliated; branches are detached Trees are slightly defoliated; branches are detached No signs of
defoliation
Grazing pressure Cattle and goats are seen; fresh dungs are seen; trampled
plants
Dungs at one or two places are seen; hoof-mark is seen Dungs and hoof may not be visible No signs of grazing
Fire Stems are blackened; the crown is burnt, and some trees
are dead
Stems are blackened; the crown is burnt, but trees are
not dead
Understory vegetation burns are rarely seen No signs of re
Soil erosion Small gullies and exposed roots are seen Rill erosion is seen Only surface erosion is seen No signs of soil
erosion
Weed infestation Weeds occupy more than 50 % of the plot Weeds occupy about 10–50 % of the plot Weeds occupy less than 10 % of the plot No weeds
Shrub
encroachment
Shrubs occupy more than 50 % of the plot Shrubs occupy about 25–50 % of the plot Shrubs occupy less than 25 % of the plot No shrubs
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
5
plant species including the threatened D. ombet and Dobera glabra trees (WeForest, 2018). Juniperus procera and Olea europaea subsp.
cuspidata are the dominant trees in the highland parts of the forest while D. ombet and D. glabra dominate the lowlands (Aynekulu,
2011; WeForest, 2018). The forest is also a home for various threatened wildlife species including the hamadryas baboon (Papio
hamadryas), the spotted hyaena (Crocuta crocuta), the African striped ground squirrel (Xerus erythropus) and the cinereous bunting
(Emberiza cineracea) (WeForest, 2018). Cynodon dactylon is the major grass species found in the forest and is highly preferred by
livestock and wild grazers. Desa’a is one of the few signicant remnants protected dry Afromontane forests of Ethiopia, and is among
the 58 national forest priority areas designated to conserve biodiversity (Woldemichael et al., 2011). Despite the forest is a protected
area, it is highly degraded by overexploitation and land-use change resulting in the reduction of forest cover and biodiversity
(Aynekulu, 2011; Giday et al., 2018; Hishe et al., 2021a).
According to Bekele-Tesemma (2007) Ethiopian agroecological zone classications based on altitude, the Desa’a forest encom-
passes three zones: lowland (500–1500 m), midland (1500–2300 m) and highland (2300–3200 m) with an area cover of 15 %, 51 %
and 34 %, respectively (WeForest, 2018). D. ombet is present within the Vachellia etbaica communities in Desa’a forest at 1000–2000 m
altitude (Aynekulu, 2011; Birhane et al., 2023). At this altitude, two agroecological zones, representing different altitude gradients
were selected for this study (Fig. 1). The lowland agroecological zone, hereafter referred to as ’Lowland site’ is located at lower al-
titudes (1000–1500 m) while the midland agroecological zone, hereafter referred to as ’Midland site’ is situated at altitude between
1500 and 2000 m. The lowlands are characterized by low amount of rainfall and high temperatures whereas the midlands have
medium amount of rainfall and moderate temperatures (Bekele-Tesemma, 2007).
2.3. Sampling design and data collection
The comprehensive land-use map of the Desa’a forest produced by WeForest (2018) and a reconnaissance survey were rst used to
locate the study agroecological zones. Then, in each zone (site), six parallel transects of 500 m long and 20 m wide (1 ha) set one km
apart were established along an altitudinal gradient from the edge of each site. Each transect was divided into 10 plots of 20 m ×20 m
set 50 m apart. A total of 60 plots were established at each site. The eld inventory was conducted from April to August 2020. In each
plot, the abundance of the species was counted, and the diameter at breast height (1.3 m above the ground level – DBH), and the
diameter at stump height of living trees (0.3 m above the ground level – DSH, for plants that had a height <1.3 m) were measured using
a diameter tape and a calliper, respectively. Height was measured using bamboo stick 5 m long, graduated with 10 cm markings. The
health status of the trees was determined through visual observation by comparing living and non-living plant parts. The altitude of
each plot was recorded using a hand-held Garmin 74 GPS.
Soil samples were taken from the four corners and the centre of each plot using an ’X’ pattern to a 30 cm soil depth (Negash and
Starr, 2015). Accordingly, a total of 60 composite soil samples were collected of each site for some soil physical and chemical analysis.
Prior to the analysis, the soil samples were cleaned from plant roots and other debris, then air-dried and sieved.
To assess the conservation threats to the species and its habitats, various threat indicators were assessed by observation within the
plot (Table 1). These indicators were adopted from various sources (e.g., Aynekulu, 2011; Negussie and Delaet, 2017; Giday et al.,
2018; WeForest, 2018; Haile et al., 2021; Gidey et al., 2023).
2.4. Soil analysis
The soil samples were analysed in the soil laboratory of the Department of Land Resource Management and Environmental Pro-
tection at Mekelle University in Ethiopia. Soil texture was determined using the hydrometer method (Gee and Bauder, 1982). The pH of
the soil was measured with a pH meter in a 1:2.5 soil to water ratio suspension. Electrical conductivity was measured by a conductivity
meter in a 1:5 soil to water ratio suspension. Soil organic carbon was analysed using the Walkley and Black methods (Walkley and
Black, 1934). Total nitrogen and available soil phosphorus were determined using the Micro-Kjeldhahl and Olsen methods, respec-
tively (Ryan et al., 2001). Exchangeable potassium and cation exchange capacity were determined using the ammonium acetate
Table 2
Soil physical and chemical properties (0–30 cm depth) of the study sites.
Soil parameters Lowland site
Mean (±SEM)
Midland site
Mean (±SEM)
Sand (%) 63 (±6.6) 67 (±8.2)
Clay (%) 19 (±5.5) 18 (±6.2)
Silt (%) 17 (±4.4) 17 (±4.8)
Soil pH (1:2.5 H
2
O) 8.4 (±2.8) 8.0 (±3.2)
EC (1:5 H
2
O) 0.16 (±0.5) 0.16 (±0.6)
OC (%) 2.35 (±0.7) 2.55 (±0.8)
TN (%) 0.20 (±0.2) 0.28 (±0.3)
Av. P (ppm) 5.46 (±2.2) 5.66 (±2.4)
Ex. K (ppm) 306 (±10.8) 300 (±11.0)
CEC (meq/100 g) 26.38 (±7.8) 24.4 (±7.2)
EC =electrical conductivity, OC =organic carbon, OM =organic matter, TN =total nitrogen, Av.
P=available phosphorus, Ex. K =exchangeable potassium, CEC =cation exchange capacity
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
6
method (Van Reeuwijk, 1993). Soil physical and chemical properties of the study sites are presented in Table 2.
2.5. Data analysis
To analyse the abundance at each site, the D. ombet trees were divided into three ontogenetic stages based on ndings from previous
studies and opinions of the local communities (Ghazali et al., 2008; Andersen et al., 2022; Birhane et al., 2023). Accordingly, D. ombet
trees were divided into seedlings (DSH <2 cm and height <0.5 m), saplings (DSH =2–8 cm and height =0.5–1.3 m or DBH =2–8 cm
and height =1.3–1.5 m) and mature trees (DBH >8 cm and height >1.5 m). The health status of each tree or sapling was categorized
as healthy (70 % of the tree parts were living), moderately healthy (35–70 %), unhealthy (1–35 %) or dead (0 %) (Ghazali et al., 2008).
The species’ population structure was further characterized graphically using size-class distributions (SCDs). The density (e.g., the
total density, the densities of seedlings, saplings, and mature trees) and the health status of the species between the sites were
compared using the Kruskal-Wallis nonparametric test as the data did not meet the assumption of normality. The health status was
further analysed using descriptive statistics such as percentages. To assess the correlation between the species’ abundance and health
status, as well as altitude and soil variables (Table 2), Spearman’s correlation analysis was performed. Signicant differences were
considered at p <0.05. The Statistical Analysis Software (SAS) version 9.2 was used for data analysis. The conservation threats for the
species and its habitats were categorized into high, medium, low, and absent levels according to the corresponding indicators pre-
sented in Table 1.
3. Results
3.1. Abundance, health status and structure
The total density of D. ombet ranged from 173 to 235 trees ha
−1
, with a signicantly higher value recorded in the midland site
(Table 3). The mean density in the sites was 204 ±4.2 trees ha
−1
, with a relatively higher number of mature trees than saplings and
seedlings. As to the health status, about 47 % of trees were healthy and the remaining 27 % and 26 % were unhealthy and dead,
respectively (Table 3). Tree health was not signicantly different between the sites (Table 3). The SCDs highlighted that the population
lacked regeneration, seedlings (DSH <2 cm) and saplings (e.g., DBH between 2 to 5 cm) (Fig. 2). The highest individuals were found in
stem diameter class of 20–25 and 15–20 cm, respectively. Besides, the upper SCDs were characterized by a dearth of mature trees (DBH
>25 cm) (Fig. 2).
Within the same rows, the means with the same letters are not signicantly different at p <0.05. SEM =standard error of the mean.
3.2. Effects of altitude and soil variables
The density of D. ombet was signicantly positively correlated with altitude, silt soil particles, soil organic carbon and soil nitrogen.
However, tree density was negatively correlated with the sand fraction of the soil, soil pH, available phosphorus and exchangeable
potassium. The correlations between the tree health status and soil nitrogen were positive while the correlations of the tree health with
clay fraction and soil pH were signicantly negative (Table 4).
3.3. Conservation threats
At the sites, D. ombet and its habitats were severely threatened by stem cutting and debarking, leaf defoliation, overgrazing, soil
erosion and competition from expansive shrubs (Table 5; Fig. 3). However, the effects of weed infestation and re on the tree and its
habitats appeared to be minor (Table 5). Intensive cutting, debarking and defoliation of the trees by local communities were observed
during the period of sampling. We found 30 D. ombet trees ha
−1
felled, 20 trees ha
−1
defoliated and 18 trees ha
−1
partially debarked.
Cattle, goats and camels have also been seen browsing and trampling on the species. In addition, multiple small gullies and rills were
often seen around the root zones of the species. Expansive shrubs of Cadia purpurea and Tarchonanthus camphoratus were the major
encroaching of the species’ habitats.
Table 3
Density and health status (trees ha
−1
) of D. ombet in the study sites.
Parameters Lowland site
Mean (±SEM)
Midland site
Mean (±SEM)
Both sites
Mean (±SEM)
Total density 182
b
(±4.6) 225
a
(±3.9) 204 (±4.2)
Seedlings 0
a
(±0.0) 0
a
(±0.0) 0 (±0.0)
Saplings 24
b
(±2.8) 45
a
(±3.2) 35 (±3.0)
Mature trees 158
b
(±4.5) 180
a
(±4.8) 169 (±4.6)
Healthy trees 29
a
(±2.4) 36
a
(±3.0) 34 (±2.7)
Moderately healthy trees 56
a
(±2.5) 66
a
(±2.8) 61 (±2.6)
Unhealthy trees 50
a
(±4.2) 61
a
(±4.5) 55 (±4.3)
Dead trees 47
a
(±6.0) 62
a
(±5.6) 54 (±5.8)
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
7
4. Discussion
Results showed that the D. ombet populations in the lowland and midland areas of Desa’a dry Afromontane forest were charac-
terized by low abundance and unstable population structure. The latter is evinced by lack of regeneration, seedlings (DSH <2 cm) and
mature trees (DBH >25 cm). About half the number of trees was unhealthy or dead. This low abundance and unstable structure was
mainly due to the overexploitation of mature trees (DBH >8 cm) – stem cutting and debarking as well as leaf defoliation. For example,
the local communities harvested intensively the mature trees for subsistence income. Namely, they collected their stems, bark and
leaves for preparing household and agricultural implements to be sold in the local markets. These threats in combination with other
pressures (e.g., overgrazing and soil erosion) are believed to have been affecting the species in the Desa’a forest for a long time and may
Fig. 2. DBH size-class distributions of D. ombet populations in the Desa’a dry Afromontane forest, northern Ethiopia.
Table 4
Correlations of the density and health status of D. ombet, with altitude and soil variables (the two sites were pooled
together).
Parameters D. ombet total density
(trees ha
−1
)
D. ombet healthy trees
(trees ha
−1
)
Altitude (m) 0.77 * 0.29
Sand (%) –0.11 0.03
Clay (%) 0.20 –0.15
Silt (%) 0.23 * 0.18
pH –0.15 –0.01
EC 0.15 0.07
OC (%) 0.16 * 0.14
TN (%) 0.17 * 0.22 *
Av. P (ppm) –0.11 0.07
Ex. K (ppm) –0.05 0.21
CEC (meq/100 g) 0.12 0.19
EC =electrical conductivity, OC =organic carbon, TN =total nitrogen, Av. P =available phosphorus, Ex. K
=exchangeable potassium, CEC =cation exchange capacity.
*=indicated signicant effects at p <0.05.
Table 5
Impacts of different conservation threats on D. ombet and its habitats in the Desa’a dry Afromontane forest, northern Ethiopia.
Threats
Lowland site Midland site
High Medium Low Absent High Medium Low Absent
Stem cutting ✓ ✓
Stem debarking ✓ ✓
Defoliation ✓ ✓
Overgrazing ✓ ✓
Fire ✓ ✓
Soil erosion ✓ ✓
Weed infestation ✓ ✓
Shrub encroachment ✓ ✓
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
8
threaten persistence of the population (WeForest, 2018; Aynekulu, 2011; Hishe et al., 2021b) and to be dominated by old trees
(Leng´
alov´
a et al., 2020) which where only conned to small eroded areas (Aynekulu, 2011). Similarly, mature trees of the species have
been heavily exploited for rope making and fodder in Egypt and Sudan (Kamel et al., 2014; Mohammed, 2015). Consequently, the
abundance and health status of the species were severely impacted. For instance, a detailed inventory of the trees in the Red Sea Hills,
Egypt recorded a total of 353 mature trees, 54 % of which were unhealthy. Its population structure was also very unstable as there was
no regeneration (Ghazali et al., 2008; Kamel et al., 2014). Overexploitation of D. ombet impacted its persistence in various ways
including decrease in the number of mature trees, or compromised health of mature trees, both of which may compromise seed
production (Ghazali et al., 2008).
Results also showed that saplings of the species were browsed and trampled by cattle, goats and camels in the Desa’a forest. Local
communities also over-defoliate the mature trees for livestock feed, particularly during drought periods. These pressures increase the
vulnerability of the saplings and mature trees to soil erosion, disease and pests (Ghazali et al., 2008; Elnoby et al., 2017). Overgrazing
habitats of the species was also identied as the major factor for the poor regeneration and high seedling mortality for D. ombet and
other woody species in the Desa’a forest. It reduces the number of viable seeds on the soil surface and of emerged seedlings (Giday
et al., 2018; WeForest, 2018; Hishe et al., 2021a). Similarly, in Egypt, particularly in the Gabel Elba areas, overgrazing habitat of the
species threatened severely the regeneration and seedlings of the species (Ghazali et al., 2008; Kamel et al., 2014). Moreover, goat
overgrazing was found as a key factor for D. cinnabari seedlings mortality in the island of Socotra, Yemen (Madˇ
era et al., 2018).
Overgrazing by livestock was also mentioned elsewhere as the major pressure for the low regeneration and seedling growth for other
dragon trees, e.g., D. tamaranae (Almeida and Censo, 2003) and D. serrulata (Vahalik et al., 2020).
The study recorded a substantial number of small gullies and rills around the root zones of D. ombet and its habitats. These could be
due to the frequent runoff and soil erosion that occurred in the Desa’a forest (Negussie and Delaet, 2017; WeForest, 2018; Hishe et al.,
2021a). Soil erosion signicantly impacted the regeneration and overall growth of the woody species in the Desa’a forest including
D. ombet. It washes away their viable seeds from natural habitats, uprooting the established small seedlings and leaching out important
soil nutrients (Berihu et al., 2017; WeForest, 2018). Similar ndings conrmed that soil erosion reduced substantially the D. cinnabari
seedlings’ growth in the island of Socotra, Yemen (Pietsch et al., 2013). The same observations were also recorded for other tree species
in Afromontane forests in northern Ethiopia (Aynekulu et al., 2016; Birhanu et al., 2021).
The expansive light-demanding C. purpurea and T. camphoratus pioneer shrubs are the major invaders of D. ombet habitats. This
result agrees with previous studies conducted in the Desa’a forest. For example, Haile et al. (2021) and Hishe et al. (2021b) found that
the aforementioned shrubs were the key encroachers of the high- and midland portions of the forest that negatively threaten the overall
growth of the climax tree species such as J. procera and O. europaea subsp. cuspidata through competing for soil resources (e.g., water
Fig. 3. Different threats to D. ombet and its habitats in the Desa’a forest: A) Stem cutting, B) Stem debarking, C) Defoliation and D) Encroachment by
expansive shrubs (Gidey et al., 2023).
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
9
and nutrients). These shrubs are also expected to decrease the diversity and composition of the remnant Desa’a forest via their rapid
colonization capacity and ability to produce plenty of high-quality seeds for regeneration (Aynekulu, 2011; Haile et al., 2021). A
similar nding was also reported on the rapid colonization of C. purpurea in the Girat Kahisu Hugumburda forest, in northern Ethiopia
(Aynekulu et al., 2016). On the other hand, some shrub species can serve as nurse plants allowing the D. cinnabari seedlings to survive
under grazing pressure (Rejˇ
zek et al., 2016).
The abundance of the species was inuenced by altitude and soil properties. This is consistent with previous studies on various dry
Afromontane trees in northern Ethiopia (Aynekulu, 2011; Birhanu et al., 2021; Hishe et al., 2021a). At the sites, the most important
factors affecting the density of the tree were altitude and amounts of silt, organic carbon and nitrogen in the soil. Overall density of the
tree was higher in the midland site than in the lowland. This could be due to the former site receiving higher annual rainfall and
horizontal precipitation (Kalivodov´
a et al., 2020). This result agrees with Vahalik et al. (2020) who recorded different densities of
D. serrulata trees in three areas within the Dhofar Mountains of Oman, at different altitudes. The Jabal Samhan area, characterized by
relatively higher altitudes (1075–1579 m) and moderate rainfall included 1835 trees while the Jabal al Qamar area characterized by
lower altitudes (659–1082 m) and rainfall had only 552 trees. However, tree health showed a non-signicant difference between the
two sites. This could be associated with the uniform occurrence of the threats for the species at the sites, regardless of their altitudinal
differences. This result is consistent with Kamel et al. (2014) who recorded a total of 353 D. ombet trees in the Gabel Elba, Egypt , 54 %
of which were unhealthy, and distributed more or less evenly over the tree’s preferred habitats.
This study and other recent studies (e.g., WeForest, 2018; Leng´
alov´
a et al., 2020) evidenced that D. ombet is currently under en-
dangered conditions due to anthropogenic and natural factors, which threaten its populations in dry Afromontane forests. Urgent
conservation actions are then necessary to maintain its existence. For instance, the overexploitation of the tree for subsistence use
should be minimized by introducing alternative livelihood strategies. In the Desa’a forest, various environmentally friendly and
non-destructive alternative livelihoods have been suggested, including the gathering of non-timber forest products (e.g., medicinal
plants, edible fruits and honey production), poultry farming and home gardening (Tamba et al., 2021; Gidey et al., 2023). Overgrazing
of D. ombet habitats can also be reduced by introducing livestock exclosures. The latter are critical to enhance the conservation of the
species as it improves the microclimate of the area, increasing the abundance of viable seeds for regeneration and protecting the
emerged small seedlings (Ghazali et al., 2008; Tamba et al., 2021). Moreover, the degradation of Desa’a forest by soil erosion and
runoff can be minimized through community-based construction of soil and water conservation structures like in-situ micro basines,
deep and shallow trenches and terraces. These strategies may contribute to ecological restoration of the forest and succession of its
threatened tree species like D. ombet and others via improving soil fertility and moisture retention (Negussie and Delaet, 2017). The
impacts of the expansive shrubs on D. ombet and its Desa’a habitats should also be addressed, namely by using them for preparing
biochar and bioenergy feedstocks (Haile et al., 2021) or as a substitute of local fuel wood instead of other native tree species (WeForest,
2018).
5. Conclusions
This research showed that the prevailing D.ombet populations in the Desa’a dry Afromontane forest of northern Ethiopia are
characterized by low abundance and unstable structure. It also revealed that about 53 % of the prevailing populations were unhealthy
and damaged. The low abundance, unstable structure and poor health status were primarily due to stem cutting and debarking, leaf
defoliation, overgrazing, soil erosion and competition from expansive shrubs. Therefore, to achieve sustainable D. ombet conservation
in dry Afromontane forests, it may be crucial to implement conservation measures that address both anthropogenic and natural factors.
This can be achieved through a combination of community engagement and policy interventions. To reduce the overexploitation of the
species, different measures such as the use of available alternative livelihoods (e.g., collecting non-timber forest products, poultry
farming) should be introduced. Livestock overgrazing within the species habitats should also be minimized through the introduction of
community-based exclosures. Furthermore, the impacts of soil erosion and of the expansive shrubs on D.ombet and its Desa’a habitats
should be addressed via the construction of soil and water conservation structures and the introduction of proper utilization tech-
niques, respectively. Future research should focus on studying the species habitat preferences and reproductive biology. This
knowledge can inform the development of targeted conservation actions, such as habitat restoration and assisted regeneration and
seedling growth techniques.
CRediT authorship contribution statement
Gidey Tesfay: Writing – review & editing, Writing – original draft, Resources, Methodology, Funding acquisition, Formal analysis,
Data curation, Conceptualization. Birhane Emiru: Writing – review & editing, Writing – original draft, Supervision, Resources,
Methodology, Formal analysis, Conceptualization. Solomon Negasi: Writing – review & editing, Writing – original draft, Resources,
Methodology, Formal analysis, Data curation, Conceptualization. Atsbha Tesfay: Writing – review & editing, Writing – original draft,
Methodology, Data curation, Conceptualization. Manaye Ashena: Writing – review & editing, Writing – original draft, Methodology,
Formal analysis, Data curation, Conceptualization. Hishe Hadgu: Writing – review & editing, Writing – original draft, Methodology,
Formal analysis, Data curation. Gu Yirga: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data
curation, Conceptualization. Tesfaye Musse: Writing – review & editing, Writing – original draft, Methodology, Data curation,
Conceptualization. Negussie Aklilu: Writing – review & editing, Writing – original draft, Resources, Methodology, Formal analysis,
Data curation, Conceptualization. Oliveira Tˆ
ania Soa: Writing – review & editing, Writing – original draft, Validation, Methodology,
Conceptualization. Palma Joao HN: Writing – review & editing, Writing – original draft, Methodology, Conceptualization. Madˇ
era
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
10
Petr: Writing – review & editing, Writing – original draft, Supervision, Methodology, Formal analysis, Conceptualization. Borges Jose
G: Writing – review & editing, Writing – original draft, Supervision, Methodology, Formal analysis, Conceptualization.
Declaration of Competing Interest
The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to
inuence the work reported in this paper.
Data Availability
Data will be made available on request.
Acknowledgements
We kindly acknowledge the Mohamed Bin Zayed Species Conservation Fund (grant numbers: 180519008 and 212526902), the
Wild Planet Trust, the Van Tienhoven Foundation (grant number, 20128), the Foundation Franklinia (grant number, 2020-15), the
Rufford Foundation (grant numbers: 21680-1, 26273-2, 31671-B and 40760-D) and People’s Trust for Endangered Species for
nancially supported to Tesfay Gidey to conduct the study. The study was further partially funded by the Forest Research Centre, a
research unit funded by Fundaç˜
ao para a Ciˆ
encia e a Tecnologia I.P. (FCT), Portugal (UIDB/00239/2020) as well as by the project ref.
H2020-MSCA-RISE-2020/101007950, with the title “DecisionES - Decision Support for the Supply of Ecosystem Services under Global
Change”, funded by the Marie Curie International Staff Exchange Scheme. Furthermore, we acknowledge the Institute of International
Education-Scholars Rescue Fund (IIE-SRF), Norwegian University of Life Sciences (NMBU), Faculty of Environmental Sciences and
Natural Resource Management (MINA), and NORGLOBAL 2 project “Towards a climate-smart policy and management framework for
conservation and use of dry forest ecosystem services and resources in Ethiopia (grant number: 303600)” for supporting the research
stay of Emiru Birhane at NMBU. We also acknowledge WeForest Ethiopia for the logistic support and granting access to their database.
References
Adolt, R., Pavlis, J., 2004. Age structure and growth of Dracaena cinnabari populations on Socotra. Trees Struct. Funct. 18, 43–53.
Almeida, P., Censo, R., 2003. Distribucion, habitat y estado de conservation de Dracaena tamaranae Bot. Macaronesica 24, 39–56.
Al-Okaishi, A., 2020. Local management system of dragon’s blood tree resin in rmihin, Socotra Island, Yemen. Forests 11, 1–17. https://doi.org/10.3390/
f11040389.
Andersen, G.L., Krzywinski, K., Talib, M., Musa, M., 2022. Endangered Dracaena ombet population in the Red Sea Hills, Sudan, recovers after abrupt change. Front.
Environ. Sci. 10, 793583 https://doi.org/10.3389/fenvs.2022.793583.
Aynekulu E. (2011) Forest diversity in fragmented landscapes of northern Ethiopia and implications for conservation. PhD thesis, pp. 141, University of Bonn,
Germany.
Aynekulu, E., Aerts, R., Denich, M., Negussie, A., et al., 2016. Plant diversity and regeneration in a disturbed isolated dry Afromontane forest in northern Ethiopia.
Folia Geobot. 51, 115–127.
Bauerov´
a, L., Munie, S.A., Houˇ
skov´
a, K., Habrov´
a, H., 2020. Germination of Dracaena cinnabari Balf.f. seeds under controlled temperature conditions. Forests 11, 521.
Bekele-Tesemma (2007) Useful trees of Ethiopia: identication, propagation and management in 17 agroecological zones. RELMA in ICRAF Project, Nairobi, pp. 552.
Berihu, T., Girmay, G., Sebhatleab, M., Berhane, E., et al., 2017. Soil carbon and nitrogen losses following deforestation in Ethiopia. Agron. Sustain Dev. 37, 1–12.
https://doi.org/10.1007/s13593-016-0408-4.
Birhane, E., Gidey, T., Abrha, H., et al., 2023. Impact of land-use and climate change on the population structure and distribution range of the rare and endangered
Dracaena ombet and Dobera glabra in northern Ethiopia. J. Nar. Conserv 76, 126506.
Birhanu, L., Bekele, T., Tesfaw, B., Demissew, S., 2021. Relationships between topographic factors, soil and plant communities in a dry Afromontane forest patches of
North western Ethiopia. PLoS ONE 16, e0247966.
BoANR (1997) Desa’a state forest management plan, National Regional State of Tigray. Mekelle, Ethiopia.
Brown, G., Mies, B., 2012. Vegetation Ecology of Socotra. The Springers press,, Dordrecht, the Nethrlands.
Bos, J.J., Teketay D., 1997. Dracaenaceae. In: Edwards, S., Sebsebe Demissew, Inga Hedberg) (Eds.), Flora of Ethiopia and Eritrea, 4. Addis Ababa University,
Ethiopia; Uppsala University, Sweden, pp. 76–82.
De Sanctis, M., Adeeb, A., Farcomeni, A., et al., 2013. Classication and distribution patterns of plant communities on Socotra Island, Yemen. Appl. Veg. Sci. 16,
148–165.
Elnoby, S., Moustafa, A., Mansour, S., 2017. Impact of climate change on the endangered Nubian dragon tree (Dracaena ombet) in the South Eastern of Egypt. Catrina
16, 23–28.
Ethiopian Biodiversity Institute (2022) National ecosystems assessments of Ethiopia: Syntheses of the status of biodiversity and ecosystem services, and scenarios of
change. Addis Ababa, Ethiopia, pp. 672.
Gee, G., Bauder, J., 1982. Particle size analysis in methods of soil analysis. Agronomy no. 9, 2nd edn..,. Soil Science Society of America,, Medison, USA.
Ghazali U., El Baily H., Dora A. (2008) The globally endangered Dracaena ombet monitoring and assessment project in Gabel Elba protected area, Egypt. Final Report,
Conservation Leadership Programme, pp. 111.
Giday, K., Humnessa, B., Muys, B., et al., 2018. Effects of livestock grazing on key vegetation attributes of a remnant forest reserve: the case of Desa’a forest in
northern Ethiopia. Glob. Ecol. Conserv 14, 1–12.
Gidey, T., Birhane, E., Manaye, A., et al., 2023. Prioritizing forest conservation strategies using a multi-attribute decision model to address concerns with the survival
of the endangered dragon tree (Dracaena ombet Kotschy and Peyr.). J. Nar. Conserv 73, 1–8.
Gonz´
alez-Castro, A., P´
erez-P´
erez, D., Romero, J., Nogales, M., 2019. Unraveling the seed dispersal system of an insular “Ghost” Dragon Tree (Dracaena draco) in the
wild. Front. Ecol. Evol. 7, 39.
Govaerts R., Zonneveld B., Zona S. (2021) World Checklist of Asparagaceae. Facilitated by the Royal Botanic Gardens, Kew. Available online: http://apps.kew.org/
wcsp/ (accessed on 16, July 2022).
Haile, M., Birhane, E., Mekonen, M., Adaramola, M., 2021. Expansive shrubs: expansion factors and ecological impacts in northern Ethiopia. J. Nar. Conserv 61, 1–11.
Hishe, H., Giday, K., Fremout, T., Negussie, A., et al., 2021a. Environmental and anthropogenic factors affecting natural regeneration of degraded dry Afromontane
forest. Restor. Ecol. 29, e13471.
T. Gidey et al.

Global Ecology and Conservation 50 (2024) e02809
11
Hishe, H., Oosterlynck, L., Giday, K., et al., 2021b. A combination of climate, tree diversity and local human disturbance determine the stability of dry Afromontane
forests. For. Ecosyst. 8 (1), 16.
Hubalkova, I., Houska, J., Kubicek, J., et al., 2017. Preliminary anatomical study on secondary thickening parts of endemic Dracaena cinnabari Bal.l. from the
Soqotra island. Wood Res. 62, 67–77.
IUCN (2017) IUCN Red List of threatened Species: IUCN Global Species Programme Red List Unit. Cambridge, UK. Available online: www.iucnredlist.org (accessed on
01, January 2022).
Kalivodov´
a, H., Culek, M., ˇ
Cerm´
ak, M., Madˇ
era, P., Habrov´
a, H., 2020. Potential importance of Socotra dragon’s blood tree cloud forests and woodlands for capturing
horizontal precipitation. Rend. Lince-.-. Sci. Fis. E Nat. 31, 607–621.
Kamel, M., Ghazaly, U., Callmander, W., 2014. Conservation status of the endangered Nubian dragon tree Dracaena ombet in Gebel Elba National. Oryx 49, 704–709.
https://doi.org/10.1017/S0030605313001385.
Kitaba N.T. (2006) Ecology and plant use diversity in Sof Umer area of Bale, Southeastern Ethiopia. MSc thesis, Addis Ababa University, pp. 107.
Kotschy, Perritsch (1867) Dracaenaceae Dracaena ombet Pl. Tinn. 47.
Leng´
alov´
a, K., Kalivodov´
a, H., Habrov´
a, H., et al., 2020. First age-estimation model for Dracaena ombet and Dracaena draco subsp. caboverdeana. Forests 11, 1–14.
https://doi.org/10.3390/f11030264.
Madˇ
era, P., Habrova, H., Senfeldr, M., Kholova, I., et al., 2018. Growth dynamics of endemic Dracaena cinnabari Balf.f. of Socotra Island suggest essential elements for
a conservation strategy. Biologia 74, 339–349.
Madˇ
era, P., Forrest, A., Hanacek, P., et al., 2020. What we know and what we do not know about dragon trees? Forests 11, 1–35. https://doi.org/10.3390/f11020236.
Marrero, A., Almeida Perez, S., Martin-Gonzalez, M., 1998. A new species of the wild dragon tree, Dracaena (Dracaenaceae) from gran Canaria and its taxonomic and
biogeographic implications. Bot. J. Linn. Soc. 128, 291–314.
Mohammed U. (2015) Trans-boundary conservation of dragon tree come as part for establish the regional Dragon tree conservation consortium in north Africa and
south West Asia. Report for Rufford Foundation, pp. 21.
Nadezhdina, N., Al-Okaishi, A., Madˇ
era, P., 2018. Sap ow measurements in a Socotra Dragon’s Blood Tree (Dracaena cinnabari) in its Area of Origin. Trop. Plant Biol.
11, 107–118.
Negash, M., Starr, M., 2015. Biomass and soil carbon stocks of indigenous agroforestry systems on the south-eastern Rift Valley escarpment, Ethiopia. Plant Soil 393,
95–107.
Negussie A., Delaet K. (2017) Manual for forest landscape restoration projects in Ethiopia: Pre-and post-tree planting management, monitoring and evaluation.
Brussels, Belgium, pp. 77.
Nyssen, J., Vandenreyken, H., Poesen, J., 2005. Rainfall erosivity and variability in the northern Ethiopian highlands. J. Hydrol. 311, 172–187.
Pietsch, D., Kuhn, P., Morris, M., Mubarak, G., 2013. Land improvement on a dry tropical island in the Arabian Sea. J. Land. Ecol. 6, 109–123.
Rejˇ
zek, M., Sv´
atek, M., ˇ
Sebesta, J., Adolt, R., Madˇ
era, P., Matula, R., 2016. Loss of a single tree species will lead to an overall decline in plant diversity: effect of
Dracaena cinnabari Balf. f. on the vegetation of Socotra Island. Biol. Conserv 196, 165–172.
Ryan, J., Estefan, G., Rashid, A., 2001. Soil and plant analysis laboratory manual. Jointly published by the International Center for Agricultural Research in the Dry
Areas (ICARDA) and the National Agricultural Research Center (NARC), 2nd edn..,. ICARDA,, Aleppo, Syria.
Tamba, Y., Wafula, J., Whitney, C., Luedeling, E., et al., 2021. Stochastic simulation of restoration outcomes for a dry afromontane forest landscape in northern
Ethiopia. Policy Econ. 124, 102403.
Tetemke, B., Birhane, E., Rannestad, M., Eid, T., 2019. Allometric models for predicting aboveground biomass of trees in the dry Afromontane. Forests 10, 1114–1129.
Thulin, M., 1995. In: Thulin, M. (Ed.), Dracaenaceae. In Flora of Somalia, Vol. 4. Royal Botanic Gardens, Kew London, UK, p. 29.
Vahalik, P., Patocka, P., Drapela, K., et al., 2020. The conservation status and population mapping of the endangered Dracaena serrulata in the Dhofar Mountains,
Oman. Forests 11, 322–337.
Vahalík, P., Van Damme, K., N´
etek, R., Habrov´
a, H., et al., 2023. UAV inventory of the last remaining Dragon Tree forest on Earth. Forests 14, 766.
Van Reeuwijk P. (1993) Procedures for soil analysis, 4th edn. International Soil Reference and Information Center, the Netherlands.
Walkley, A., Black, I., 1934. An examination of the Degtjareff method for determining soil organic matter, and a proposed modication of the chromic acid titration
method. Soil Sci. 37, 29–38.
WeForest (2018) Desa’a state forest management plan. Final report, Mekelle, Ethiopia, pp. 369.
Wilkin, P., Suksathan, P., Keeratikiat, K., Welzen, P., 2012. A new threatened endemic species from central and northeastern Thailand, Dracaena jayniana
(Asparagaceae: tribe Nolinoideae). Kew Bull. 67, 697–705.
Woldemichael, L., Bekele, T., Nemomissa, S., 2011. Vegetation composition in Hugumbirda-Gratkhassu national forest priority area, South Tigray. Momona Ethio J.
Sci. 2, 27–48.
T. Gidey et al.