Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.
We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.
To purchase hard copies of this book, please contact the representative in India:
CBS Publishers & Distributors Pvt. Ltd.
www.cbspd.com
|
customercare@cbspd.com
Invasive plants have had significant impacts on vegetation communities of Zimbabwe. A study was undertaken to determine current and potential distribution of C. edulis in Zimbabwe using DIVA GIS and MAXENT, and to determine climatic conditions under which the species thrives, together with. The species population structure and its impact on native species. Results indicate that the species has its highest occurrence frequency in Manicaland Province, followed by Matebeleland South Province. Some 13% occurrence points were recorded at an altitude less than 600 m, 21% at an altitude ranging from 600–999 m, 43% at an altitude between 1000 and 1399 m and 23% at an altitude above 1400 m. C. edulis was recorded in areas of maximum temperature range of 34°C and a minimum of 20°C. The species also occurred in regions with a mean precipitation range as low as 60–300 mm and as high as 1000–1261 mm. Further, C. edulis distribution is predicted to expand in the Eastern Highlands (Manicaland), parts of Mazowe and Bindura (Mashonaland Central Province) and parts of Matobo (Matebelaland South Province). Diameter class distributions showed an inverse J-distribution in control sites and in all three sampled sections. An irregular bell-shaped distribution was recorded for co-occurring species on C. edulis occupied sites. It was concluded that C. edulis’ regeneration potential is high and that of competing native species is unstable and has the potential to expand beyond the currently occupied sites.
Department of Biological Sciences and Ecology, University of Zimbabwe, Harare, Zimbabwe
Clemence Zimudzi
Department of Biological Sciences and Ecology, University of Zimbabwe, Harare, Zimbabwe
Shakkie Kativu
Department of Biological Sciences and Ecology, University of Zimbabwe, Harare, Zimbabwe
Brita Stedje
Natural History Museum, University of Oslo, Oslo, Norway
*Address all correspondence to: evengarakana@gmail.com
1. Introduction
Currently, global biodiversity is being threatened greatly by climate change and invasive species [1]. Several dozens of species and variants (including invasives) have been introduced into Zimbabwe [2], a majority of them within the Eastern Highlands where the climate supports the highest plant diversity. These include some suspected variants of Catha edulis (Vahl) Forssk. exEndl. C. edulis naturally occurs in the horn of Africa down to southern Africa and Madagascar. Its centre of origin is believed to be Ethiopia and Kenya [3].
Observations on spatial distribution of C. edulis in Zimbabwe over the past few years indicate that it is spreading, and where this occurs, few other plant species thrive [4]. Immigrants from East Africa are suspected to have introduced different variants of C. edulis to Southern Africa [5]. Some such variants have become more aggressive and currently occupying forest margins in Manicaland, Zimbabwe (Eastern Highlands), an area which forms part of the eastern Africa biodiversity hotspot [6]. The Eastern Highlands ecosystem provides freshwater and other ecosystem services to a significant number of people in the region [6].
The climatic conditions under which C. edulis in Zimbabwe thrives and its current and potential sites of distribution in the country are not known. That, together with the population structure of the species, requires an urgent investigation as the results may be useful in predicting the species distribution trends and in wildlife management. An assessment of the impact of C. edulis on biodiversity of the Eastern Highlands biodiversity hotspot is also needed. Population structure, which partly reflects age and size structure, is indicative of the health and survival capacity of a species [7]. Important life stages of a species can be revealed and aid in wildlife management [8]. Species diversity and evenness make up species composition and the higher the species richness and productivity are the more resilient the ecosystem is [9].
The present study sought to map current and potential distribution of C. edulis in Zimbabwe using MAXENT and DIVA GIS, establish climatic conditions under which the species thrives, determine the population structure of C. edulis and assess its impact on indigenous species within Vumba Forest area of Zimbabwe. Maximum entropy (MAXENT), which relies only on presence data and background environmental information, was the preferred assessment method. The method also accommodates small sample size and allows for gaps in records [10].
This study was conducted in Zimbabwe. About 4% of the country receives an annual rainfall of >1000 mm, and low mean annual temperatures which range from 15 to 18°C. Approximately 32% of the country has annual rainfall below 500 mm, with a high mean annual temperature of 21–25°C [11]. Some 16% of the country is under protection, and includes endemic and critically endangered plant and animal species [12].
Population structure studies were carried out in Vumba, Eastern Highlands of the country (Figure 1). Vumba is about 246 km2 with its highest altitude being 1911 m [14]. The mean yearly precipitation is 1800 mm and majority occurs between November and August [6]. Soils are deep and well weathered [6]. Its vegetation comprises miombo woodland which favours high rainfall, evergreen Afromontane forests and montane grassland [14]. Most of the vegetation types have been exposed to severe disturbances which has paved way for the encroachment of such invasive species as Cestrum aurantiacum Lindl., Lantana camara L., Vernonanthura polyanthus (Spreng.) Vega & Dematteis and Solanum mauritiunamScop. [6]. Wattle and eucalypts are also planted for commercial purposes [6]. The following GPS coordinates are of the first, second and third study plots respectively: 19.064640°S, 32.720713°E; 19.078418°S, 32.750628°E; 19.071590°S, 32.744625°E.
Figure 1.
Study area map in Vumba (right), and its location within Zimbabwe (left) Ballings & Wursten [13].
2.2 Mapping of C. edulis populations
Sources for the species occurrence data were the Zimbabwean National Herbarium and Botanic Gardens, Global Biodiversity Information Facility (GBIF) (www.gbif.org) and Zimbabwe Flora (www.zimbabweflora.co.zw). In the course of field studies, new distribution records of C. edulis were recorded. The combined data were checked for duplication. The source for climate variables was WorldClim (www.worldclim.org). WorldClim has a total of 19 bioclimatic variables, 11 temperature and 8 precipitation matrices which represent different annual trends, seasonality and extreme environmental conditions [15]. Pearson correlation coefficients were examined in order to check all the variables for multicollinearity [16]. The first principal component analysis (PCA) analysed only one variable from each set of highly correlated variables (r > 0.95). The main bioclimatic variables were also determined by performing the Correlation analysis (CA) and PCA using the statistical program R version 3.1.3 [16]. Spatial resolution t used was 30arc seconds (about 1 km2 per pixel). This allowed for maximum details [17]. DIVA-GIS [18] was used to map the distribution of C. edulis in the whole of Zimbabwe. It is a geographic information system that has been used in various mapping studies such as the mapping of spatial distribution of Jatropha curcas L. in Malaysia by Shabanimofrad et al. [19] and that of Senecio vulgaris L. in China by Cheng and Xu [15]. The software was used according to the manual guide by Scheldeman et al. [20].
2.3 Predicting Catha edulis potential sites of distribution
MAXENT model was used to predict the potential sites of distribution of C. edulis. One-time Split method was used to partition the occurrence records for use in validating the accuracy of the model’s predictions [21]. 75% was randomly selected to make up the training data (calibration data) and the remaining 25% was the test data (evaluation data) [17]. Jackknife test was used to measure the importance of the climatic factors. MAXENT was run in default settings. Area Under Curve (AUC) was used to evaluate the predictive ability of the model generated by MAXENT [15]. Results were imported and visualised in DIVA-DIS. The software was used according to the manual guide by Scheldeman et al. [20] and explanations by Phillips et al. [22].
2.4 Population structure
The study was carried out in October and November 2020. Three C. edulis occupied sites representing at least 20% of the total area with the species were randomly selected. Three adjacent sites without C. edulis were also selected to be controls and were situated within 100 m from the occupied sites. C. edulis occupied sites where land had been cleared for the construction of electricity power line in the 1960s (pers. comm.) while the control sites had no obvious signs of disturbance.
Three lines of transect measuring 210 to 220 m were placed 60 m each from the disturbed point to the furthest point away from disturbance where C. edulis occurred. Following Walker [23] and Gandiwa & Kativu [24]’s set up, 20 x 10 m sampling plots were systematically placed 65 m apart each. Total plots in the sites occupied by C. edulis and the control sites were 27 each. Three sections were also demarcated in the sites occupied by C. edulis as sections (i) closest, (ii) mid-way, and (iii) furthest from disturbance.
Stem circumferences of each woody species in the study plots were measured at 1.3 m with a tape measure [25] and used to calculate diameter at breast height (circumference/pi). Each stem on multi-stemmed plants was measured and the values summed up to calculate the circumference of the plant [25]. All the woody species in the study plots were also identified in situ or at the National Herbarium in Harare.
Differences in species evenness and composition in the control sites and sites occupied by C. edulis were verified using the equitability test and Shannon-Weiner diversity index [26]. Separate analyses were done for each site and then averaged. Shapiro–Wilk test was used to test for normality in populations from both the sites occupied by C. edulis and the adjacent control sites in SPSS 2007. Significant differences in species richness and evenness in the sites invaded by C. edulis and the control sites were assessed using Independent T-test in Microsoft Excel 2007. Significant differences in means of the three sections studied also assessed using One Way ANOVA in Microsoft Excel 2007.
Current spatial distribution and location altitude of C. edulis in Zimbabwe is as illustrated in Figure 2. Manicaland Province recorded the highest number of C. edulis presence points, followed by Matebeleland South Province. Of the 161 C. edulis presence points recorded, 13% occur at an altitude less than 600 m represented by yellow triangles. 21% occur at an altitude ranging from 600 to 999 m and represented by orange squares. The highest presence points recorded (43%) occur at an altitude between 1000 and 1399 m, and this is followed by 23% which occur at an altitude above 1400 m.
Figure 2.
Distribution and altitudinal location of C. edulis in Zimbabwe.
Spatial distribution of C. edulis in Zimbabwe and maximum temperature of warmest month at each site are illustrated in Figure 3. Sampled C. edulis occupies sites with four temperature ranges were recorded out of five. The occupied sites include those with a maximum temperature range of 31–34°C represented by white, 28–31°C represented by yellow, 24–28°C represented by neon/light green and those with the maximum temperature range of 20–24°C represented by forest/dark green.
Figure 3.
Distribution of C. edulis with location maximum temperatures of warmest month in Zimbabwe.
Spatial distribution and sites’ average precipitation of warmest month of C. edulis are as illustrated in Figure 4. The sampled C. edulis is shown to be occupying sites with all five different precipitation ranges. The occupied sites include those with a mean precipitation range of 60 to 300 mm represented by white, 300 to 500 mm represented by yellow, 500 to 800 mm represented by neon-green, 800 to 1000 mm represented by pine-green and those with a mean precipitation range of 1000 to 1261 mm represented by tea-green.
Figure 4.
Distribution of C. edulis with mean precipitation of warmest month in Zimbabwe.
Potential sites of distribution of C. edulis in Zimbabwe are illustrated in Figure 5. Further C. edulis distribution is predicted in the Eastern Highlands (Manicaland Province), parts of Mazowe and Bindura (Mashonaland Central Province) and parts of Matobo (Matebelaland South Province) with a probability of 0.5 to 1 represented by red.
Figure 5.
Potential sites of distribution of C. edulis in Zimbabwe.
Table 1 shows the percentage contributions of the 19 bioclimatic variables that were used in modelling the potential sites of spread of C. edulis in Zimbabwe. Results show that precipitation of driest month (BIO14), mean temperature of warmest quarter (BIO10) and maximum temperature of warmest month (BIO5) had the highest contribution in the modelling of C. edulis distribution in Zimbabwe shown in Figure 4. Their total contribution was 74.1%.
Variable
Name of variable
Percent contribution
Permutation importance
BIO14
Precipitation of driest month
53.7
27.3
BIO10
Mean temperature of warmest quarter
12.6
2.9
BIO5
Maximum temperature of warmest month
7.8
8.4
BIO19
Precipitation of coldest month
6.7
0
BIO16
Precipitation of wettest month
3.1
10.6
BIO7
Temperature annual range
2.6
0
BIO9
Mean temperature of driest quarter
2.3
12.8
BIO4
Temperature seasonality
2.1
3.7
BIO15
Precipitation seasonality
1.4
11.3
BIO18
Precipitation of warmest month
1.3
7.6
BIO3
Isothermality
1.3
4.9
BIO17
Precipitation of driest month
1.2
2.1
BIO13
Precipitation of wettest month
1.2
2.9
BIO8
Mean temperature of wettest month
1.1
0
BIO12
Annual precipitation
1
3.3
BIO6
Mean temperature of coldest month
0.6
0.1
BIO2
Mean diurnal range
0.1
2.2
BIO11
Mean temperature of coldest quarter
0
0
BIO1
Annual mean temperature
0
0
Table 1.
Relevance of each bioclimatic variable for the model.
Figure 6 shows relative importance of the variables with respect to training gain after jackknife test had been done. BIO14 generated the highest gain when used as the only bioclimatic variable of the model and its omission also resulted in the strongest decrease of gain among all the variables. This is consistent with the relevance of bioclim variables test (Table 1).
Figure 6.
Results of jackknife evaluation of the relative importance of the variables with respect to training gain.
Figure 7 shows the predictive ability of the model in Figure 5 generated by Maxent using Area Under Curve (AUC) of the Receiver Operating Characteristic (ROC) curve. Random prediction (black line) is a reference line. Both training data (red line) and test data (blue line) shows that the model had a high predictive ability with a high AUC value of 0.95.
Figure 7.
Evaluation of the predictive ability of the model generated by Maxent using AUC of the ROC curve.
Table 2 lists the woody plant species observed in the study plots. Sites invaded by C. edulis had 6 species while sites unoccupied by C. edulis had 15 species. Trema orientalis and Harungana madagascariensis were only found in the sites occupied by C. edulis while Bersama abyssinica, Heteropyxis dehniae and Acacia abyssinica were recorded in both invaded and control sites.
Woody species recorded in sites C. edulis invaded and non-invaded sites. X shows that the species is present in the site.
Sites occupied by C. edulis and the adjacent control sites had greater Shapiro–Wilk test values (0.22 and 0.21, respectively) than the alpha value (0.05). The populations are therefore normally distributed. Parametric tests were used to check for differences in species richness and evenness among the populations in the study sites. Means of the sections close to, mid-way and furthest from disturbance were also subjected to parametric tests to check for differences among study sections.
Sites occupied by C. edulis’ had mean equitability value of 0.11 while the adjacent control sites had a value of 0.45. The latter is closer to 1 while the former is closer to 0. Control sites also had a higher Shannon diversity index of 2.65 compared to that of the sites occupied by the C. edulis (0.73). T Stat values in both Equitability test and Shannon diversity index (63.82 and 60.56, respectively) are higher than 2.78 which is the t Critical value. Equitability test and Shannon diversity index’s P values of 3.6x10−7and 4.45x10−7 are less than 0.05 which is the Alpha value. Therefore, the differences in species evenness and diversity in sites occupied by C. edulis and the unoccupied sites is significant.
Results of diameter class distribution of C. edulis stems in the section close to, middle section and furthest from disturbance respectively are shown in Figure 8 and all three sections display an inverse J-distribution.
Figure 8.
Diameter class distribution of the mean number of C. edulis stems observed in sections close to, mid-way and furthest from C. edulis invaded sites.
Percentages of mean number of C. edulis stems recorded in the section close to, mid-way and furthest from disturbance in the three diameter classes are summarised in Table 3. 0-8 cm diameter class had an average of 502 stems. Section closest to the disturbance had the lowest percentage (20.1%) while the one furthest from the disturbance had the highest percentage (46.4%). The second diameter class (8<16 cm) had an average of 125 stems and 18.4% of these stems were recorded in the section furthest from disturbance while 57.6% were in the section closest to the disturbance. 16<24 cm diameter class had zero recordings in the section furthest from disturbance while 83.3% of the 18 stems recorded in this diameter class were from the section closest to the disturbance.
Diameter class (cm)
near the disturbance
Middle section from the disturbance
Furthest section from the disturbance
0 ≥ 8
(233) 46.40%
(168) 33.50%
(101) 20.10%
8 ≤ 16
(23) 18.40%
(30) 24%
(72) 57.60%
16 ≤ 24
(0) 0%
(3) 16.70%
(15) 83.30%
Table 3.
Percentages of C. edulis stems recorded in the three diameter classes in study sections.
Results of the One Way Anova show that at least means of C. edulis stems in the three diameter classes and the three sections had a significant difference as the 0.05 alpha value was greater than 6.8x10−7, 1.2x10−6, and 6.6x10−5 p values. T-test results showed that the means of C. edulis stems in all the three sections were significantly different as 0.05 alpha value was greater than 3.7x10−5, 1.7x10−4, 1.4x10−3, 0.031, 2.31x10−4, 8.36x10−5, 0.035, 0.01, 0.005 P values.
Results of diameter class distribution are shown in Figure 9. An inverse J distribution is indicated for both native plant stems in uninvaded sites and also for C. edulis stems in sites it exclusively occupies. An irregular bell shaped distribution is indicated for native stems co-occurring with C. edulis with zero stems being recorded in classes 0–8, 21–24 and 29–35 cm.
Figure 9.
Diameter class distribution of the (1) mean number of native stems co-occurring with C. edulis, (2) mean number of native stems invaded sites (3) mean number of C. edulis stems in sites invaded sites.
Altitude, temperature and precipitation were the primary parameters used in constructing the maps because of their crucial role in determining plant distribution, growth and persistence [27]. Altitude influences such microclimatic conditions as temperature and precipitation [28]. C. edulis was found to be predominantly occurring in the Eastern Highlands of Zimbabwe, a region characterised by relatively high altitude, high rainfall and low temperatures. This observation is consistent with what was reported in Yemen and Ethiopia by Al-hebshi & Skaug [29], Zahran et al. [30] and Kandari et al. [31]. C. edulis, being an evergreen tree [31], is well adapted to these conditions. Broad-leaved subtropical evergreen trees have low water use efficiency. Hence, they thrive in high rainfall conditions. They use the stored, readily available underground water for evaporative cooling during the dry season to avoid excessive light and heat stress which enables them to retain their leaves throughout the dry season [32].
The results, however, show that C. edulis also occurs in other parts of Zimbabwe which are at low altitude and characterised by low rainfall and high temperatures. Similar results were summarised by Zahran et al. [30] where different variants of C. edulis were distinguished by the Yemenis farmers based on different altitude and climatic conditions under which they thrive. Most plants in dry areas have shallow roots which allow quick uptake of moisture and nutrients near the soil surface [33].
Dessie & Kinlund [34] also reported that C. edulis was expanding into relatively higher temperature, lower rainfall and lower altitudinal zones, resulting in the decline of forests in Wondo Genet, Ethiopia. The present study observed similar range expansion trends for C. dulis occupied sites in the Eastern Highlands of Zimbabwe. Dessie & Kinlund [34] attributed range expansion of C. edulis to various factors, including the species ecological adaptability and its genetic variation [35].
Plants with broader climatic and altitudinal needs are reported to be highly adaptive, with potential to become invasive [36]. Most non-invasive species biological activities and growth occur within specific narrow temperature ranges. Such species require specific amounts of precipitation for physiological processes [37]. However, species with dominance and invasive tendencies tend to tolerate a wide range of climatic and edaphic conditions [36]. In a study to assess invasion risk of plants by Higgins & Richardson [38], a physiologically based species distribution model was used, and it concluded that species that tolerate a wider range of environmental conditions tend to be invasive. Findings of the present study showed that C. edulis in Zimbabwe thrives under broad climatic conditions, and has the potential to expand its range of distribution in the Eastern Highlands of the country.
Sites occupied by C. edulis had significantly different species evenness and diversity from those not occupied by the C. edulis. Other studies involving plant species with invasive properties also showed similar trends [39, 40]. C. edulis has a competitive advantage as it produces seeds in large quantities, allowing it to occupy new extended habitats [41, 42].
Plots further away from the point of disturbance had notably younger and higher numbers of C. edulis stems in comparison to those closest to the disturbance. C. edulis individuals nearest the disturbance were the first ones to occupy the site hence are the oldest. In conducive conditions, the combination of C. edulis being tall plus producing relatively smaller seeds [43], allows it to disperse its seeds furthest away from the parent plant [44]. This reduces intra specific competition and enhances increased range distribution for the species [45]. Seeds from tall trees are dispersed from a higher point which is significantly open thereby reduces chances of other plants disturbing them. Shrubs and seedlings of C. edulis are therefore more densely populated than the adults [34] in a similar sized area, thus, comprising a higher population density in comparison with a site occupied predominantly by adults.
Fifty percent of the species found in sites disturbed areas are pioneer species [46]. The height, densely packed establishment of C. edulis and its closed canopy formation tend to suppress the growth of co-occurring pioneer species. Their seedlings are deprived of light due to the shade formed by C. edulis’ closed canopy growth form [34]. This strategy has been observed in other species for example, Impatiens glandulifera [40]. This species reduces evenness and diversity in habitats it invades by forming a closed canopy over seeds of co-occurring species, thereby depriving them of light which disturbs germination and also cause stunted growth [40].
The observed inverse J-distribution for C. edulis implies that the species is in a healthy regenerating state [47]. This distribution pattern is usually displayed when the lower diameter classes have a higher number of individuals which steadily reduce towards the higher diameter classes. Such a distribution ensures sustainability of the population as there will be numerous seeds that can be recruited into the following growth stages [48]. An irregular-bell shaped distribution like the one shown by the co-occuring native species in sites occupied by C. edulis suggests that the rate of regeneration is lower that the mortality rate at certain diameter classes [48] therefore the population will be in poor regeneration potential [47]. Hence, their populations are under threat. Native species in sites not occupied by C. edulis showed a healthy regeneration potential.
The main driving assumption in population diameter class distribution structures is that a population with more stems in the lower diameter classes compared to the higher ones is constantly regenerating, while that with fewer stems in the lower classes compared to the higher classes is in decline [49]. However, this assumption is not true for all species populations as some have displayed the reverse J-shape distribution but well known to be declining [7]. Some species which do not display the reverse J-shape distribution are actually known to be increasing or their populations are in stable states [50]. These observations have been credited to the different growth rates sometimes found among different size classes. The population diameter distribution method and interpretation, however, has been used successfully in many studies by Souza [47, 48, 51]. While it is not recommended as a sole assessment method, it is a useful basis for management decisions in the absence of complimentary demographic approaches [49].
Catha edulis has stable populations with high potential for regeneration in the Eastern Highlands of Zimbabwe, while co-occurring indigenous species are in a decline or unstable state. The persistence of C. edulis in sites furthest away from points of disturbance suggests that disturbance is not the only determinant factor for the invasion taking place, which is a worrisome observation for the continued existence of the indigenous species in the area. Future studies must focus on elucidating mechanisms that support and encourage the dominance of C. edulis and potentially suppressing co-occurring native species in areas occupied by C. edulis.
The Norwegian Partnership Programme for Global Academic Cooperation (NORPART) financially supported the study.
References
1.WWF. 2020. Living Planet Report 2020 – Bending the curve of biodiversity loss. Almond REA, Grooten M, Peterson T. https://www.zsl.org/sites/default/files/LRP%202020%20Full%20report.pdf
2.Maroyi A. Local plant use and traditional conservation practices in Nhema communal area, Zimbabwe. International Journal of African Renaissance Studies. 2012;7(1):109-128
3.Tembrock LR, Simmons MP, Richards CM, Reeves PA, Reilley A, Curto MA, et al. Phylogeography of the wild and cultivated stimulant plant qat (Catha edulis, celastraceae) in areas of historical cultivation. American Journal of Botany. 2017;104(4):538-549. DOI: 10.3732/ajb.1600437
4.Wursten B, Ballings P, Palgrave, MC, Hyde M. Catha edulis (Vahl) Forssk.exEndl. 2016. Retrieved July 6, 2017. http://www.zimbabweflora.co.zw/speciesdata/species.php?species_id=137020
5.Surve I. Chewing the khat is illegal, says Sanab. Independent Online (IOL). 2000. [Retrieved from]: https://www.iol.co.za/news/south-africa/chewing-the-khat-is-illegal-says-san b-35917
6.Timberlake J, Ballings P, Vidal J Jr, Wursten B, Hyde M, Mapaura A, et al. Mountains of the mist: A first plant checklist for the Bvumba Mountains, Manica Highlands (Zimbabwe-Mozambique). Phyto Keys. 2020;145:93-129
7.Paul A, Khan MF, Das AK. Population structure and regeneration status of rhododendrons in temperate mixed broad-leaved forests of western Arunachal Pradesh, India. Geology Ecology and Landscapes. 2019;3(3):168e186. DOI: 10.1080/24749508.2018.1525671
8.Meekins JF, Mccarthy BC. Effect of population density on the demography of an invasive plant (Alliariapetiolata, Brassicaceae) population in a southeastern Ohio Forest. The American Midland Naturalist. 2006;147(2):256-278
9.Dyderski MK, Jagodziński AM. Impact of invasive tree species on natural regeneration species composition, diversity and density. Forests. 2020;11:456-476
10.Ma B, Sun J. Predicting the distribution of Stipapurpurea across the Tibetan Plateau via the MaxEntmodel. BMC Ecology. 2018;18(1):1-12. DOI: 10.1186/s12898-018-0165-0
11.Mugandani R, Wuta M, Makarau A, Chipindu B. Re-classification of agro-ecological regions of Zimbabwe in conformity with climate variability and change. African Crop Science Journal. 2012;20(s2):361-369. DOI: 10.4314/acsj.v20i2
12.Kumirai A, Msimanga A, Munyikwa D, Chidavaenzi R, Muringaniza SJ. Nomination dossier for the proposed Matobo Hills World Heritage. 2003. www.whc.unesco.org/uploads/nominations/306rev.pdf
13.Ballings P, Wursten B. Vumba – Facts. 2001. https://vumba-nature.com/wp/vumba-nature/vumba-facts/. [Retrieved from]: 22 January 2021
14.Clark VR, Timberlake JR, Hyde MA, Mapaura A, Coates Palgrave M, et al. A first comprehensive account of floristic diversity and endemism on the Nyanga Massif, Manica Highlands (Zimbabwe-Mozambique). Kirkia. 2017;19:1-53
15.Cheng D, XU L. Predicting the potential distributions of Senecio vulgaris L. in China. PeerJ Preprints. 2015. DOI: 10.7287/peerj.preprints.1612v1
16.Perez LV. Principal component analysis to address multicollinearity. 2017. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwi75Ijrt-b3AhWoRkEAHflsB9MQFnoECB0QAQ&url=https%3A%2F%2Fwww.whitman.edu%2FDocuments%2FAcademics%2FMathematics%2F2017%2FPerez.pdf&usg=AOvVaw2OOzdrgT--uRZ2WyeooKOZ
17.Yi YJ, Cheng X, Yang ZF, Zhang SH. Maxentmodeling for predicting the potential distribution of endangered medicinal plant (H. riparia Lour) in Yunnan, China. Ecological Engineering. 2016;92:260-269. DOI: 10.1016/j.ecoleng.2016.04.010
18.Hijmans RJ, Guarino L, Jarvis A, O’Brien R, Mathur P, Cruz CBM, et al. DIVA-GIS Version 5.2 Manual. 2005. Retrieved July 7, 2017, from http://www.diva-gis.org/docs/DIVA-GIS5_manual.pdf
19.Shabanimofrad M, Yusop MR, Saad MS, Megat Wahab PE, Biabanikhanehkahdani A, Latif MA. Diversity of physic nut (Jatrophacurcasi) in Malaysia: Application of DIVA-geographic information system and cluster analysis. Australian Journal of Crop Science. 2011;5(4):361-368
20.Scheldeman X, Van Zonneveld M. Training Manual on Spatial Analysis of Plant Diversity and Distribution. Analysis. Vol. 3282010. Rome: Biodiversity International; pp. 11-177. DOI: 10.1021/ja0267831
21.Pearson RG. Species’ distribution modeling for conservation educators and practitioners. Lessons in Conservation. 2010;3:54-89. DOI: 10.1016/S0140-6736(10)61462-6
22.Phillips SJ, Dudík M, Schapire R. A maximum entropy approach to species distribution modeling. In Proceedings of the 21st International Conference on Machine Learning. New York: ACM Press. 2004. (pp. 655-662)
23.Walker BH. An approach to the monitoring of changes in the composition and utilisation of woodland and savanna vegetation. South African Journal of Wildlife Resources. 1976;6(1):1-32
24.Gandiwa E, Kativu S. Influence of fire frequency on Colophospermummopane and Combretumapiculatum woodland structure and composition in Northen Gonarezhou National Park, Zimbabwe. Koedoe. 2009;51(1) Art.#685:13. DOI: 10.4102/koedoe. v51i1.685
25.Campbell BM, Cunliffe RN, Gambiza J. Vegetation structure and small-scale pattern in miombowoodland, Marondera, Zimbabwe. Bothalia. 1995;25(1):121-126
26.Shannon CE, Weaver W. The Mathematical Theory of Communication. Urbana: University of Illinois Press; 1949
27.Jin XM, Zhang Y, Schaepman ME, Clevers JGPW, SU Z, Clevers J, et al. Impact of elevation and aspect on the spatial distribution of vegetation in the Qilian and mountain area with remote sensing data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2006;37:1385-1390
28.He X, Hou E, Liu Y, Wen D. Altitudinal patterns and controls of plant and soil nutrient concentrations and stoichiometry in subtropical China. Scientific Reports. 2016;6(24261):1-9. DOI: 10.1038/srep24261
29.Al-Hebshi NN, Skaug N. Khat (Catha edulis)— an updated review. Addiction Biology. 2005;10(December):299-307. DOI: 10.1080/13556210500353020
30.Zahran MA, Khedr A, Dahmash A, El-Ameir YA. Qat farms in Yemen: Ecology, dangerous impacts and future promise. Egyptian Journal of Basic and Applied Sciences. 2013;1(1):1-8. DOI: 10.1016/j.ejbas.2013.09.002
31.Kandari LS, Yadav HR, Thakur AK, Kandari T. Chat (Catha edulis): A socio economic crop in Harar Region, Eastern Ethiopia. SpringerPlus. 2014;3(579):1-9. DOI: 10.1186/2193-1801-3-579
32.Tomlinson KW, Poorter L, Sterck FJ, Borghetti F, Ward D, De Bie S, et al. Leaf adaptations of evergreen and deciduous trees of semi-arid and humid savannas on three continents. Journal of Ecology. 2013;101(2):430-440. DOI: 10.1111/1365-2745.12056
33.Baudena M, Dekker SC, Van Bodegom PM, Cuesta B, Higgins SI, Lehsten V, et al. Forests, savannas, and grasslands: Bridging the knowledge gap between ecology and dynamic global vegetation models. Biogeosciences. 2015;12(6):1833-1848. DOI: 10.5194/bg-12-1833-2015
34.Dessie G, Kinlund P. Khat expansion and forest decline in Wondo Genet, Ethiopia. Geografiska Annaler: Series B, Human Geography. 2008;90(2):187-203. DOI: 10.1111/j.1468-0467.2008.00286.x
35.Al-Thobhani MAH, Sathyanarayana NB, Simon L, Sondur S. First comparative genotypic study on Khat(CathaedulisForsk.) genotypes from Yemen. Middle Eastern and Russian Journal of Plant Science and Biotechnology. 2008;2(1), 1-8. [Retrieved from]: http://www.globalsciencebooks.info/Online/GSBOnline/images/0812/MERJPSB_2 (1&2)/MERJPSB_2(1)1-8o.pdf
36.Ratnayake RMCS. Why plant species become invasive? Characters related to successful biological invasion. In: National Symposium on Invasive Alien Species. 2014. (pp. 22-42). [Retrieved from]: https://www.researchgate.net/publication/273452847_Why_plant_species_become_invasive
37.Llewellyn LM. Range plant growth and development are affected by climatic factors. 2013. [Retrieved from]: https://www.ag.ndsu.edu/DickinsonREC/grazing-handbook-files/topic1-volume1-ch2-report1.pdf
38.Higgins SI, Richardson DM. Invasive plants have broader physiological niches. Proceedings of the National Academy of Sciences of the United States of America. 2014;111(29):10610-10614. DOI: 10.1073/pnas.1406075111
39.Hejda M, Pysek P, Jarosík V. Impact of invasive plants on the species richness, diversity and composition of invaded communities. Journal of Ecology. 2009;97:393-403
40.Hulme PE, Bremner ET. Assessing the impact of Impatiens gland ulifera on riparian habitats: Partitioning diversity components following species removal. Journal of Applied Ecology. 2006;43:43-50
41.Landenberger RE, Kota NL, McGraw JB. Seed dispersal of the non-native invasive tree Ailanthus altissima into contrasting environments. Plant Ecology. 2007;192(1):55-70
42.Wan J, Wang C, Yu F. Wind effects on habitat distributions of wind-dispersed invasive plants across different biomes on a global scale: Assessment using six species. Ecological Informatics. 2017;42:38-45
43.van der Vossen, H.A.M., 2002. Catha edulis (Vahl) Forssk.exEndl. [Internet] Record from PROTA4U. Oyen, L.P.A. & Lemmens, R.H.M.J. (Editors). PROTA (Plant Resources of Tropical Africa/Ressourcesvégétales de l’Afriquetropicale), Wageningen, Netherlands. https://uses.plantnet-project.org/en/Catha_edulis_(PROTA). Accessed 3 September 2018.
44.Thomson F, Moles A, Auld TD, Kingsford RT. Seed dispersal distance is more strongly correlated with plant height than with seed mass. Journal of Ecology. 2011;99(6):1299-1307
45.Yang X, Zhang W, He Q. Effects of intraspecific competition on growth, architecture and biomass allocation of Quercus Liaotungensis. Journal of Plant Interactions. 2019;14(1):284-294. DOI: 10.1080/17429145.2019.1629656
46.Hyde MA, Wursten BT, Ballings P, Coates Palgrave M. Flora of Zimbabwe: Home page.2020. https://www.zimbabweflora.co.zw/index.php, retrieved 24 December 2020
47.Mekonen T, Ayele B, Ashagrie Y. Woody plant species diversity, structure and regeneration status of Woynwuha Natural Forest, North West Ethiopia. Asian Journal of Ethnopharmacology and Medicinal Foods. 2015;01(01):07-19
48.Nabasumba D, Eilu G, Bahati J, Katwesigye G. Utilisation and conservation status of indigenous woody plant species in a sedentary pastoral production system in south-western Uganda. Uganda Journal of Agricultural Sciences. 2016;17(2):125-137
49.Virillo CB, Martins FR, Tamashiro JY, dos Santos FAM. Is size structure a good measure of future trends of plantpopulations? An empirical approach using five woody species from the Cerrado (Brazilian savanna). ActaBotanicaBrasilica. 2011;25(3):593-600
50.Condit R, Sukumar R, Hubbell SP, Foster RB. Predicting population trends from size distributions: A direct test in a tropical tree community. The American Naturalist. 1998;152:496-509
51.Souza AF. Ecological interpretation of multiple population size structures in trees: The case of Araucaria angustifolia in South America. Austral Ecology. 2007;32:524-533. DOI: 10.1111/j.1442-9993.2007.01724.x
Written By
Evelyn Ngarakana, Clemence Zimudzi, Shakkie Kativu and Brita Stedje
Submitted: 22 June 2022Reviewed: 13 July 2022Published: 09 September 2022
Open access peer-reviewed chapter
