Impact of invasion : A case study on the ecological and socioeconomic impact of Lantana camara ( L . ) in Abay Millennium Park ( AMP ) , Bahir Dar , Ethiopia

This study was carried out to quantify ecological distribution, abundance and effects of Lantana camara on different land uses and to assess the perception of local community concerning its impacts on environment and biodiversity of the study area, Abay Millennium Park, Bahir Dar. To do so, 20 m x 20 m (trees) inside which is 5 m x 5 m nest plots (shrubs and saplings), and 2 m x 2 m (seedlings) were laid. Vegetation data were collected using purposive sampling technique with quantitative measurements (DBH, density, seedling and sapling count). Furthermore, semi-structured questionnaire was used to collect data about its socio-economic and ecological impacts. Results reveal that L. camara has shown highest invasion (82.1%, grassland) and least invasion (34.6%, riverine forest); cultivated land being intermediate (57.5%). Plant species density was high at the lower class (<6cm DBH, 83.94% of total plants sampled). At species level, L. camara exhibited an inverted J-shaped frequency distribution (62.21 %< 2 cm DBH), screening its greater regeneration potential. There was high diversity and species richness with H’=2.6980, S=45 for natural riverside forest, H’= 1.8173, S=28 cultivated land, and H’=0.3446, S=14, grassland. Economic and ecological results revealed that Lantana was perceived as more disadvantageous; therefore the community highly recommends, if possible, its complete removal from the park.


INTRODUCTION
The deliberate or accidental introduction of non-indigenous species to new habitats has become an increasingly important aspect of global environmental change (Pimentel et al., 2000;Mack et al., 2000;Andersen et al., 2004;Malik and Husain, 2007) and can cause important economic, environmental and social losses (Goulson and Derwent, 2004).Many research works have shown that invasive plant species have broad distribution throughout the world and can directly or indirectly affect the food security of local residents by destroying natural pasture, displace native trees, crops, and reduce grazing potential of rangelands and set limitations for economic development (Manchester and Bullock, 2000;Stohlgren et al., 2001;Pauchard and Alaback, 2004;Anderson et al., 2004;Dogra et al., 2009).Invasion is considered as the second most widespread threat to global biodiversity next to habitat destruction (Sharma et al., 2005) of natural ecosystems worldwide (Haysom and Murphy, 2003;IUCN/SSC/ISSG, 2004;CBD, 2005;Sala et al., 2005;Sharma and Raghubanshi, 2006).Once an invasive species becomes firmly established, its control often becomes difficult and eradication is usually impossible (Primentalet al., 2000).Therefore, exotic species will forever be common components of every ecosystem on Earth.
The impacts of alien species are enormous.They cause alteration in ecosystem processes and community structure, decline in abundance and richness of native flora (Sander, 1998;Stohlgren et al., 2001;Grice, 2006).Globally, the extent of damage caused by invasive species has been estimated to be £1.5 trillion per year, close to 5% of global GDP.In developing countries, where agriculture accounts for a higher proportion of GDP, the negative impact of invasive species on food security and economic performance can be even greater which exacerbate poverty (Pimentel et al., 2000;Manchester and Bullock, 2000;IBC, 2009).
Due to its strong allelopathic properties, aggressiveness and its dense impenetrable thorny thickets, Lantana camara has the potential to interrupt the health and regeneration process of other species by decreasing germination, growth of seedlings and biomass production which in turn increases mortality and decline of plant species, pasture and crops (Gentle and Duggin, 1998;Catherine and Russell, 2005;Sharma et al., 2005;Sharma and Raghubanshi, 2006).A native of Central and South America (Spies and Duplessis, 1987;Day et al., 2003), Lantana camara, as coined by Carlous Linnaeus in 1753 (Swarbrick et al., 1995), contains approximately 270 species and subspecific taxa of woody shrubs (Mendez-Santos, 2002;Bhakta and Ganjewala, 2009).The genus Lantana ranked among the 100 worlds worst invasive alien species (Day et al., 2003;GISP, 2006).The diverse and broad geographic distributions of the species beyond its native range are the reflection of its wide ecological tolerance, ability to conquer diverse habitats and its success on a variety of soil types (Day et al., 2003;ECZ, 2004).It is now a cosmopolitan exotic invader and has been declared as a noxious weed in many parts of the world (Morton, 1994;Benggeli et al., 1998;Goulson and Derwent, 2004).It is particularly a weed of the tropics and sub-tropics becoming naturalized in approximately 60 countries (Day et al., 2003).The research finding of Hailu et al. (2004) and Kassahun et al. (2004), showed the high spread of invasive plant species in Ethiopia becoming a great concern in national parks, lakes, rivers, power dams, and urban green spaces causing huge economic and ecological losses.They had become major threats to biodiversity loss and socioeconomic welfare of the Ethiopians.The prominent ones in which the government declared the need for their control and eradication (Mckee, 2007;EIAR, 2009) include Parthenium hysterophorus, Prosopis juliflora, Eichhornia crassipes, Lantana camara and Acacia species (A.drepanolobium and A. melifera).
L. camara was introduced to Ethiopia as an ornamental plant due to its beautiful aromatic flowers (Binggeli and Desalegn, 2002).However, because of prolific seed production and easy dispersal, it escaped cultivation and become a pest in the social, ecological and economic concerns.Presently, it has spread almost all over the country, but still it is not much perceived as a chronic environmental problem, except in few parts of Ethiopia, such as Oromia and Somali regions (Binggeli and Desalegn, 2002).Currently, there is little information available on spatial distribution of Lantana camara invasion and its potential geographic spread.
Lantana was brought from Addis Ababa by Mr. Alubel Kassa in 1985, the local administrator of Bezawit Palace at that time, and planted as one of the ornamental species in and around the present study area.Since then, it has covered about 70 hectare of the park area, highly spreading and destroying the native biodiversity, converting the beautiful attractive riverine park into homogenous and less attractive sight.While having vast horizons of both ecological and environmental significance in the park so far, due attention is not yet given by the concerned body.Moreover, there is no important ecological investigation, which is done to either indicate the status or the extent of the weeds' distribution and its economic as well as ecological disadvantages in the study area.
Therefore, the goal of this study was to carry out an ecological investigation on the distribution, invasion impact on different land uses and associated plant species of L. camara and assessing the perception of the local community regarding its impacts on the biodiversity of AMP.

Study area
The study was carried out at Bezawit Palace in AMP, Bahir Dar zuriaworeda, Ethiopia.The topography of the park is characterized by gentle undulating plateau with meandering river course and slow flow of water, close to the source of Abay River (Blue Nile), Lake Tana.The vegetation structure of the park is composed of evergreen, moist riverine forest with scattered islands of dense forests, dominated by high diversity of woody plant species (Abraham, 2009).This riverian park, located at 110 28' to 110 38 ' N and 370 23' to 370 36 E (Berhanu et al., 2007), covers 4680 ha and extends up to 39 km along the course of Abay River.The present study site (Bezawit Palace) is the hilltop of the AMP where the weed had begun to spread (Figure 1).

Data source and material
In this study, to investigate the relative ecological distribution, socioeconomic and ecological impacts of L. camara, both primary and secondary data sources were used.Primary data were generated from preliminary survey, field work and the responses of the local people, agricultural experts and park managers who are involved directly or indirectly with the problems of the weed.Questionnaire, interview and group discussion were used as the primary data collecting instruments.On the other hand, secondary data were obtained from Bahir Dar city master plan, published materials including books, journals, research articles and census reports.

Field data collection
A reconnaissance survey and group discussion was carried out with the residents of Bezawit kebele.This was conducted in Oct 16-20, 2010 to collect base line information, observe vegetation distribution and to determine the number of transect lines and plots that would be laid.Subsequently, field survey was carried out from Oct. 21 to Nov. 20, 2010,with the help of massive technical staff, to determine whether there was a gradient in indigenous plant diversity at different land uses caused by of L.camara invasion and to identifythreatened plant species.The length and the width of the plots were measured using tape measure and ropes.Total count of each plant species along transect was recorded.The variation of habitats with respect to altitude was measured with GPS.

Sampling technique and field protocol
Data on vegetation were collected using systematic sampling techniques to include as much vegetation as possible that can represent the vegetation of the study area.To analyze the spatial distribution and effect of L. camera on neighboring plants (woody plants and shrubs), starting from 50 m away from the primary plantation of Lantana around the Palace, three transect lines were laid down to the hill of Bezawit along three directions or land uses of natural forest (South-East or towards Abay River), cultivated (South and South-East) and grassland (North-East).Within 50 m interval, 12 plots per land use (a total of 36 plots) with 20 m x 20 m were taken for sampling of trees (Gyansharma and Raghubanshi, 2007) and within each sample plot, nested plots, 5 m x 5 m (saplings and shrubs), and 2 m x2 m (seedlings) were taken (ECZ, 2004).
The collected specimens were identified with the assistance of botanists and rarely by comparing with already identified ones from the herbarium and authenticated using Flora of Ethiopia and Eritrea (Hedberg andEdwards, 1989, 1995;Azene et al., 1993).Moreover, diameter at breast height (DBH) was measured at 1.3 m above the ground for trees (DBH > 5.00 cm and height > 3.00 m) and 0.3 m for seedlings (DBH < 2 cm and 0.5 m height) saplings and shrubs (6 cm < DBH >2 cm and 0.50 m to 3.00 m height) (Tefera et al., 2005) to estimate the regeneration status of the forest in comparison with the regeneration potential of L. camara.

Socioeconomic survey
The socio economic data was collected from Nov. 21 to 30, 2010.This was done to assess the perception and awareness of the local people towards the species, and the data was collected through semi-structured questionnaires and participatory rural appraisal (PRA) interview (Martin, 1995).A total of 60 households, 55 people from local residents who have faced problems against Lantana invasion and 5 people from agro-forestry and park development experts of the area were selected to be representative using purposive sampling technique.Households (Both male and female) who were absent during data collection were replaced with the other households.

Data analysis
Both descriptive and inferential statistics were used for data presentation and analysis.MS-Excel was used for data organization, density, relative abundance and list of species.The quantitative structural analysis was made using data from density, abundance, and frequency distribution of each species in the study sites.

Shannon Wiener's diversity index (H')
Species diversity indices were analyzed using software PAST (Hammer et al., 2001).Shannon diversity index accounts for both diversity and evenness of the species present in a community.This means due consideration was given to the species richness (composition) and evenness (equitability) within the given land or community.
Where, H' = Shannon-Wiener index; Pi = n/N is the proportion of individuals or the abundance of i th species expressed as a proportional of total cover in the sample (ranges 0 to 1); n= no of individuals of a given species; N= total no of individuals found; ln = natural log ( that is base 2.718); Σ = Summation symbol; and ln = log bases (natural logarithms).
Equitability or evenness index was calculated from the ratio of observed diversity to maximum diversity using the equation;

DBH and regeneration status
For population structural analysis, diameter class distribution for each species was done to investigate the regeneration status of the forest (Peters, 1996).Population structure bar graphs (density versus DBH class) were used to discuss the different patterns of population structure in relation to the distribution of L. camara so that total number of individuals was grouped into different arbitrary diameter classes.

Density
Density is referred to as the total number of plants per hectare, and it is useful for evaluating areal coverage of each species.

Frequency
It is calculated by a number of quadrates in which a particular species occurred divided by the total number of quadrates.

Bezawit forest patch
Bezawit forest patch has been highly threatened by the invasion of L. camara in the past; there are still many recorded native species of shrubs and trees.The result of the study showed that the area was locally dominated by L. camara (1379 individuals/ha), followed by the second highly spreading and threatening plant species; Agave sisalana (159 ind/ha).Other relatively abundant plants with more than 50 ind/ha includes: Securinea virosa(114 ind/ha), Calpurnia aurea (93 ind/ha), and Argemone mexicana (75 ind/ha) (Table 3).
Although this forest patch is highly affected by the invasion of L. camara in the study area, the compositions of vegetation are relatively intact and are much more diverse (with H=3.9215) and dense with 1898 ind/ha (Abraham, 2009) when compared to the other forest patches of the park.In the northeast direction of the study area; towards the grassland, there has been modification through re-vegetation with indigenous tree species mainly Syzygium guineese, Milleia ferruginea, Cordia africana, Acacia abyssinica, Croton macrostachyus and Olea europea sub spp.capsidata by the government and private sectors of the city during the Ethiopian millennium celebration.As the respondents replied, this natural vegetation was intensively deforested as a result of illegal cutting of trees for construction, firewood and expansion of farmlands.This indicates there was high past disturbance of the area which created more access or open spaces for the fast invasion of Lantana, corresponding with the colonizing nature of the species where disturbances are very common (Gentle and Duggin, 1997;Goodland et al., 1998;Day et al., 2003;Stock, 2005).Moreover, the absence of shade effects is believed to accelerate the intensity of Lantana (Plate 1) invasion and severe damage of the land uses of the study area.

Floristic composition
A total of 55 species of woody plants (trees, shrubs and climbers) representing 38 families were recorded including L. camara, from Bezawit forest patch of AMP.All the specimens were found sparsely intermingled with the dense stand of Lantana thickets, and of the 55 sampled plants, 49 were used in floristic and structural analysis (Table 2).Since they were collected from the spaces outside the selected sampling quadrates (still within the study area), the six ones were not considered in the analysis, but they were, indeed, taken into account in the total specimens in order to make the complete floristic list of all land uses.
From all species collected and identified, 24, 22 and 3 were trees, shrubs and climbers, respectively (Figure 2).Thus, Fabaceae was the most diverse family in species number that comprised seven species; Maytenus Plate 1.The intensity of L. camara invasion in the grassland of AMP.

Species diversity, richness and equitability
The output of computation (Shannon-Wiener diversity index) of vegetation data collected in comparison with L. camara is shown in Table 1.The mean diversity index of woody plants across three land uses was H'=1.6199 with the maximum H'=2.6980 was at the natural forest and the minimum H'='0.3446 was at the grass land in which the intermediate H'=1.8173 being cultivated.These values were obtained by subtracting the index of L. camara (these values within parentheses) from the total diversity index, including L. camara (those values without parentheses).
Based on the standardized values of Wiener diversity index, in Table 1 there was medium diversity of plant species (excluding L. camara) with H'= 2.6980 (2<H'<3) in the NF followed by low diversity, H'= 1.8173 (1<H<2) in the CL and very low species diversity (H<1) in the GL with H = 0.3446, indicating adverse effect of L. camara in disruption of succession, which aggressively reduced the indigenous plant diversities in which H' in NF > CL > GL.A low value of evenness indicates the dominance of an environment by one or a few species, while others are present with few individuals (Zerihun et al., 1999).
Thus, in the current study, species evenness of woody plants of the GL (E=0.6802) was lower than that of the CL (E=0.8749) and the maximum was recorded from the NF forest (E=0.8802),indicating a high dominance and effect of L. camara in the GL followed by CL.Species richness (S), equitability (E) and diversity (H') are also positively correlated (Krebs, 1999), with the ratio of mean diversity and evenness indexes of Lantana to other woody plants (0.2825/1.6199 and 0.6953/0.8118),respectively (Table 1).This revealed that there was better diversity of woody plants compared to the distribution and low diversity of L. camara, and the relatively high equitability in the study area.Considering species richness as one measure of diversity (Tivy, 1993), the NF has high species richness (S=45) followed by the disturbed site (S=28) and the GL which is left with the poorest species abundance (S=14).This reveals that threat and subsequent extinction of species in the grassland are very common due to the negative effect of the weed (Csurhes and Edwards, 1998;Daehler, 1998;Humphries et al., 1991).

Density, frequency and abundance of woody plants
The total density of tree and shrub species excluding L. camara in NF, CL and GL is 1,457, 513 and 543 individuals/ha, respectively.However, the density of A. sisalana (the second devastating species of plants in the area), were 42 ind/ha in CL but 435 ind/ha in GL and totally absent in the NF (Table 2).Therefore, the total density of other woody plants excluding these two species (A.sisalana and L. camara) becomes 471 ind/ha in the CL and and 109 ind/ha in the GL.Consequently, density ratio of L. camara to other affected woody plant species in the NF, CL and GL was found to be 773/1457, 698/471 and 2665/109 ind/ha of the sampled area.This revealed that the effect of Lantana still remained more severe towards the GL than the other habitats.
Analysis of the mean density of other woody plants was done for comparison of the land uses under the influence of L. camara, and turned out to be 30.35,10.69 and 11.31ind/ha in NF, CL and GL, respectively (Table 2), corresponding to the total density recorded in the same area by taking into account the influence of A. sisalana in GL.Moreover, the mean density of Lantana in all habitats (1379 individuals/ha) was found to be about 46 times greater than the mean density of woody plants in the NF and 125 times greater than the remaining land use types.
The frequency of L. camara in all land uses or its mean frequency was found to be 100% which means that it has distributed in all plots of the sampled area.Nevertheless, the mean frequency of other woody plant species was 37.51% in NF, 17.90% in the CL and 8.16% in GL.

Distribution of L. camara on different land uses
Although the density of Lantana towards the FL was to some extent greater than its density towards the cultivated or disturbed site, its relative abundance and rigorous effect still remained in the CL (Figure 2, Plate 3).However, compared to other land use types, there was high dense stand of Lantana in the GL (2665 ind/ha) due may be to the presence of intense tree canopy in the NF, imposing heavy shade effect on Lantana and hence preventing sun light, an essential factor for the hasty growth of Lantana (Reader and Bricker, 1994;Stock, 2005;Gentle and Duggin, 1997;Goodland et al., 1998;ARMCANZ and NZECCFM, 2000).
On the other hand, according to the research work of Gentle and Duggin (1998) and Prieur-Richard and Lavorel (2000), the presence of enough nutrients and the absence of shade effect on the open and/or disturbed sites, greatly favor L. camara to flourish and be widely distributed.Accordingly, as the competition of resources with Latana is very severe for them, the native species in the area declined progressively from time to time (Swarbrick et al., 1995;ARMCANZ and NZECCFM, 2000).Equivalently, our current finding indicated that the relative abundance and invasion pattern of L. camara showed an inverted "J" shape distribution, where 82.1% of the GL was invaded followed by 57.5% of the CL and lastly 34.6% of the FL habitat (Figure 3).Since the suppression and deleterious effect of the species is much more severe on the GL, there would be a serious fodder problem to the society practicing mixed farming due to the modification of the habitats and the overall ecosystem of the area (Csurhes and Edwards, 1998;Daehler, 1998;Mack et al., 2000).

Population structure and regeneration status
The entire analysis of population structure of all tree species resulted in three different patterns (Figure 4a-c), except Melia azedarch which occurred singly.The first pattern is represented by L. camara (Figure 4a), and all species in this group have high density in the lower DBH class and gradually decreased with increasing DBH (positively skewed).They show inverted "J" curve pattern which further indicates good reproduction and recruitment.Other species included in this group are Securinega virosa, Calpurnia aurea, Carissa sapinarum, Acanthus eminus, Jaminiuum grandifeorum, Matenus arbutifolia, Pterollobium stellatum, Argemone mexicana, Osris guadriparitita, Celtis africana and Pittostigma thonningii.
The second pattern is represented by Croton macrostachus (Figure 4b).The density of all species in this group increases with increasing DBH up to some points and then decreases with increasing DBH afterwards (shows Gaussean curve).Tree species in this group are, Ackanthera schmiperi, Sennaea singunea, Ecluea racemosa, Ocium lamifolium, Pottosporum vividifolia, Clutia abyssinica and combretum species.
The third pattern is represented by Sapium elliptum (Figure 4c).Species included in this pattern are Cussonia holisti, Dombeya torrid, Rhus vulgaris, Capparis tomentosa, Lannea schimperi, Gordonia termifolia, Cordia africana, and Milletia ferruginea.The number of individual trees in each species increases with increasing DBH and most of the remaining species were found to be included in this pattern, except the recorded data of a few species which showed decreasing pattern at the beginning     and then increase up to some limits and then decrease.
Population structure or distribution of individuals of each species in arbitrarily diameter size classes enables to provide the overall regeneration profile of the study species (Simon and Girma, 2004) and it is used to predict the trend of the population of that particular species (Peters, 1996).For that reason, composition and density of seedlings and saplings of each threatened tree species recorded from Bezawit forest were included in this study so as to compare the regeneration status of the forest against the seedling recruitment capacity of L. camara.To do so, all tree species in the study area were categorized into three arbitrary diameter classes, seedlings (< 2 cm DBH), saplings plus shrubs (2 cm < DBH < 6 cm) and trees (> 6 cm).
In this investigation, 1216.6 of seedlings, 746.52 of saplings plus shrubs and 268.61 trees of individuals per hectare were recorded from the total sampled plants.As a whole, when the diameter classes of all woody species were analyzed together, the forest patch showed somewhat an inverted J-shaped frequency distribution with abundant individuals at the lower diameter classes more by seedlings (49.69%) and then by shrubs and saplings with (33.25%) showing decrement in number of individuals as the diameter classes increased, thereby suggesting good regeneration status of the forest (Figure 5).Nevertheless, when considering the proportion of seedlings, more than 70.44% (856.94seedlings/ha) of the total seedling count were contributed by L. camara, screening its greatest regeneration ability (Figure 4a) and its effect on the decline of species diversity and regeneration (Murali and Siddapa, 2001;Sharma et al., 2005;Sharma and Raghubanshi, 2006).Thus, the results obviously indicate the threatened status of the forest community in the park.
Besides, L. camara with its high density and age categories (above the diameter of 4.9 m and 5.3 m), are determinant so as to influence the floristic diversity (Catherine and Russel, 2005).Accordingly, at this juncture, although the proportion of L. camara at seedling level was high, its effect on plant diversity was observed to be more severe as its size (age) increases so that from the graph (Figure 5), one can see that the strength of depletion both in the number of plants and species was more pronounced in Lantana groves with DBH class of > 2cm and < 6 cm.

Socio-economic and ecological impact
The socio economic survey identified the following uses of L. camara in the specified area.
The above table implies that the economic value of the weed was generally low or its negative effect is more perceived by the local people (Tables 4 and 5) rather than its use in any case.The utilization of the plant as a live fence/hedge around the villages and farm lands in the study area was also mentioned most frequently (35%), corresponding to reports from other part of the world (ARMCAN and ANECCFM, 2000;Binggeli, 2003;Day et al., 2003).Leaf of L. camara for animal fodder and the varieties of its flowers was the second advantage of the plant.While the livestock are forced to eat the leaves where there is no an alternative feed around, particularly during drought seasons.The local people used to cut the stand of Lantana from the field to remove and manage its invasion, but not purposely for fuel.It was also common to observe bees collecting nectar from flowers.
Although extensively used in the overseas (Bhakta and Ganjewala, 2009), none of the respondents acknowledged the medicinal value of the plant in their localities.This may be due to lack of knowledge and experience in the traditional use of plants in the study area.Lantana can also make the area, in which it is planted, very attractive and interesting because of its year around varietal flowers (Holm et al., 1991;Swarbrick et al., 1995), but here the economic value of the plant as ornamental was placed in the third rank.In this particular study, the ecological benefits of the weed were found to be better and therefore, the main advantage of the plant in the study area was highly recommended for its significant role in reducing soil erosion (80% of the respondents), followed by its importance in reducing wind speed and compacting desertification.Likewise, the local people benefited from the plant as shade tree when it appears especially in single stumps.
In highly invaded areas, in the GL, 91.67% of the respondents (Table 4) noted that the most severe problems appeared to be in the reduction of grasses and herbs under its canopy.Many respondents (75%) also acknowledged secondly that the negative effect of Lantana was great in decreasing the biodiversity of neighboring plants by computing resources, mechanical suppression and aggressiveness (Gentle and Duggin, 1998).
Furthermore, they (58.33% of the respondents) added that the plant narrowed the size of their farm lands and create difficulties in cultivation of crops.The prickly impenetrable thickets of Lantana, not only completely hinder movement of their animals, prevent them in searching of food, but also puncture their skins and cause injury.

Perception and attitude of the local people
Raising public awareness is absolutely crucial for successful AIS prevention and management (Goodland et al., 1998;Binggeli, 2003).The socio-economic survey revealed that different respondents have similar view about how (it was purposely introduced), why (it was believed to be used as ornamental/hedge), and where (it was planted around the palace) the species was established in the study area.About 90% of the local people (farmers and agropastoralists) have negative attitude towards the species (Table 5) and they are highly supportive for its total removal, while some governmental employees (park development and agroforestry experts) did not support its complete eradication, and have mixed attitude towards the species.They elaborated their reasons in that the plant has its own advantage for example, such species are ecologically successful and important in conserving genetic resources in the environment.

Management and control options
Even if no action is taken yet by the regional government to control its invasion, the community is aware of the problem, and has substantial interest to control and manage the plant.Lantana is difficult to control (ECZ, 2004), as it will coppice and form denser thickets if it is simply slashed, not uprooted.The edible pods, mostly by birds, also contribute for its long range dispersal mechanisms (Gentle and Duggin, 1997;Binggeli and Desalegn, 2002).At broad scale, there are basically three methods of prevention and management options (mechanical, chemical and biological control), all of which can be used together within an integrated management  , 2000).In this study however, the local people (91.67%) (Table 6) have tried to use only physical methods (hand grubbing, uprooting, and cutting).These physical and mechanical (stickraking, bulldozing, ploughing) techniques are in fact effective and mainly suitable for medium sized infestations thinning and pruning methods of control were also frequently used, especially following the street of the palace.Nevertheless, in the assessment of the study area, no biological and chemical control methods were used.However, a study conducted in South Africa proved that biocontrol agents including Teleonemiascrupulosa (Hemiptera), Octotomascabripennis (Coleoptera), Uroplatagirardi (Coleoptera) and Ophiomyialantanae (Diptera) were successful in controlling Lantana invasion (Cilliers and Neser, 1991;Broughton, 2000).Furthermore, a number of herbicides are registered for control of Lantana.If carefully managed, fire has also been proved to bring effective control, particularly wellsuited to dense infestations (ARMCAN and ANECCFM, 2000).

Conclusion
L. camara (L.), a highly aggressive exotic environmental weed in many countries, has significant adverse effects on biodiversity.It forms dense thickets, suppressing native vegetation and seedlings through shading, nutrient competition, smothering and allelopathy.The present study has demonstrated that L. camara impact negatively on native vegetation structure and composition.The effects on native vegetation are direct, through smothering and allelopathic means, and indirect through changes in soil properties.However, the result of this study indicates that L. camara has no equal distribution on each land use types and so does it impact equally.The plant was highly abundant and distributed in the grass, agricultural and forest lands of the study area respectively, due to absence of shade effect on the grassland and frequent disturbance in the former land uses (Plate 3).L. camara occupied 62.20% (1379 individuals/ha) of area coverage in the sampled study area (1.44 ha), with high proportion of its seedlings (60.20%).This implies that the greater regeneration Alemu and Terefe 143 capacity and potential threat of the weed on the environment.On the other hand, Lantana has at most the following advantages, usually as ornamental or hedge plant, avoidance of soil erosion and in some cases, it may provide animal fodder.However, at the current situation, the disadvantages massively outweighed the advantages.Consequently, nearly every one of the respondents in the study area agreed for its comprehensive removal, except for a few government employees.

Recommendation
The study provides strong evidence that L. camara invasion has reduced biodiversity and negatively affecting other ecosystem processes in AMP and possibly in other areas of its occurrence in Ethiopia.Consequently, without timely action for its control or removal plus at its present rate of distribution, it will not take a long time for Latana to totally invade and destroy the park's native vegetation.Based on our results, tools proposed by the respondents and the researchers for the long term and sustainable control of Lantana could include the following: first of all, the dilemma on the benefits and adverse impact of L. camara could be solved by enhancing the profits through and improving/implementing all proper management options beyond the only physical and mechanical options which are currently applied in control mechanism of the weed.Secondly, the principles underpinning for strategic control need collective action including all levels of governments, NGOs and community groups.On the other hand, creation of community awareness about the regarding the negative impacts of lantana's invasion would optimize control success and/or utilization options.Thirdly, thinning and pruning of the plant should be adopted before flowering or seed producing seasons so as to reduce its seed dispersal capacities.Moreover, integration of such techniques is likely to result in improved control and mitigation strategies.While this study has highlighted some impacts on native forest vegetation, grasses, and agricultural lands together with the perception of local communities towards the weed, impacts on other taxa such as birds, invertebrates and large mammals have not been addressed yet.Therefore, our results suggest a long-term study would provide a much more detailed understanding about the impact of L. camara invasion in the park, for example, to conduct the more promising areas for future research which include modeling and mapping invasion risk potential that will result much more comprehensive information about the whole biodiversity status of the park.

Figure 1 .
Figure 1.Map of the study area.

Figure 2 .
Figure 2. Proportion and density of plant life forms in the AMP

Figure 3 .
Figure 3.The relative abundance of L. camara on difffernt habitats.

Table 1 .
Comparisons of diversity indices of woody plant species on different land uses versus L. camara distribution.
Habitats Species richness (S) Diversity index (H') H'maxEvenness (H'/H'max)(x)Values of L. camara, indicating its contribution (influence) for each diversity index (the values without parentheses minus v alues within parentheses results indexes of other plant species).

Table 2 .
List of woody species including L. camara recorded from three land uses with their densities and frequencies.

Table 3 .
Economical and ecological use values of L. camara.

Table 4 .
Adverse impacts of L. camara.

Table 5 .
People perception and attitude against the impact of L. camara.

Table 6 .
Methods used to control invasion of L. camara in the study area.