Effect of riparian land use on environmental conditions and riparian vegetation in the east African highland streams

Accepted Manuscript Title: Effect of riparian land use on environmental conditions and riparian vegetation in the east African highland streams Authors: Tibebu Alemu, Simon Bahrndorff, Kitessa Hundera, Esayas Alemayehu, Argaw Ambelu PII: DOI: Reference:

S0075-9511(17)30082-8 http://dx.doi.org/doi:10.1016/j.limno.2017.07.001 LIMNO 25591

To appear in: Received date: Revised date: Accepted date:

25-3-2017 13-6-2017 4-7-2017

Please cite this article as: Alemu, Tibebu, Bahrndorff, Simon, Hundera, Kitessa, Alemayehu, Esayas, Ambelu, Argaw, Effect of riparian land use on environmental conditions and riparian vegetation in the east African highland streams.Limnologica http://dx.doi.org/10.1016/j.limno.2017.07.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effect of riparian land use on environmental conditions and riparian vegetation in the east African highland streams Tibebu Alemu1,2*, Simon Bahrndorff1, Kitessa Hundera3, Esayas Alemayehu4, Argaw Ambelu2 1

Section of Biology and Environmental Science, Department of Chemistry and Bioscience, Aalborg University,

Fredrik BajersVej 7H, 9220 Aalborg East, Denmark 2

Department of Environmental Health Sciences and Technology, Jimma University, P.O. Box 378, Jimma,

Ethiopia. 3

Department of Biology, Jimma University, P. O. Box 378, Jimma, Ethiopia.

4

School of Civil and Environmental Engineering, Jimma Institute of Technology, P.O. Box 378, Jimma, Ethiopia

* Correspondence present address Department of Chemistry and Bioscience - Section for Environmental Technology, 9220 Aalborg, DK.E-mail: [email protected]/[email protected].

Highlights 

We investigated the effect of riparian buffer zone deforestation on highland tropical streams.



Stream status was assessed based on riparian index, water quality and plant metrics.



Riparian and floristic quality index score were higher in forest site than agricultural site.



Riparian vegetation index has relation with water quality in tropical African streams.

Abstract

1

Agricultural land use is expanding and at an accelerated rate. In Ethiopia, most of this expansion has occurred in highland areas and involve deforestation of natural riparian vegetation. However, the impacts on the water quality of streams are poorly understood, especially with regard to the influence of land use patterns on highland streams. In this study, we investigated the effects of land use modifications on the water quality and riparian condition of highland streams and examined whether the preservation of riparian vegetation would help mitigate the negative impacts of intensive agriculture practices. Our results show significant differences in the water quality of streams with different land use. Several parameters commonly used to indicate water quality, such as the concentrations of orthophosphate, turbidity, and suspended solids were significantly higher in the agricultural streams than in the forest stream. The preservation of riparian vegetation in the surrounding highland streams was associated with overall better riparian condition, floristic quality, and water quality such as lower turbidity, total suspended solids, orthophosphate, and higher dissolved oxygen. We conclude, that increases in vegetation cover improved riparian condition and water quality relative to other non-vegetated areas. Therefore, we strongly recommend the preservation of riparian vegetation in tropical highland streams surrounded by intensive agriculture. More studies on the effects of best management practices in areas dominated by agriculture can greatly improve our capacity to prevent the degradation of water quality in tropical highland streams of Africa. Keywords: Coefficient of Conservatism, Floristic quality index, Land use, Riparian index, Water quality.

Introduction Riparian areas are transitional zone between aquatic and terrestrial ecosystems and provide many important ecosystem functions (Scott et al., 2009; Hubble et al., 2010; Piechnik et al., 2012). These functions include reduction of nutrient and sediment loading (Yuan et al., 2009), reducing streambank erosion (Berges, 2009), providing shade (Meek et al., 2010), and supplying habitat for both aquatic and terrestrial organisms (Scott et al., 2009; Fernandes et al., 2011). Riparian areas are thus also essential for diminishing the negative impacts of land use practices on streams (Li et al., 2009; Taniwaki et al., 2017). However, many riparian areas of the east African highland are degraded and have been converted into farm land (Demissie et al., 2013), which has significantly altered the aquatic habitats ( e.g., Jun et al., 2011; Uriarte et al., 2011; Lu et al., 2014)

Riparian vegetation are often removed along streams and rivers for agricultural purposes, such as grazing of livestock, providing water for livestock, and crops production (Taniwaki et al., 2017). Previous studies have investigated the negative impacts of removal of riparian vegetation in various parts of the world, e.g., South Africa

2

(Meek et al., 2010), Argentina (Rosso and Cirelli, 2013) China (Bu et al., 2014), Japan (Mori et al., 2009), and Portugal (Fernandes et al., 2011) and include impact on the aquatic life, e.g., by covering spawning areas with sediments, increasing stream temperature, and productivity, and decreasing dissolved oxygen (Sweeney and Newbold, 2014). Especially agricultural practices can increase nutrient loading in aquatic environments, including suspended sediments, organic matter, nitrogen, and phosphorus (Valle Junior et al., 2015), resulting in seasonal algal blooms that decrease levels of dissolved oxygen (Prabu et al., 2010).

The study of riparian plant communities of running streams offers information on the feature of the environment (Thiébaut and Muller, 1998; Riis et al., 2000) and can potentially be used for biomonitoring purposes (Daniel et al., 2006). Riparian plant communities can provide strong causal linkages between biodiversity and ecosystem functioning (Cavaillé et al., 2013), which can be assessed using metrics based on for example riparian condition index (RI) (Rosso and Cirelli, 2013) and floristic quality index (FQI) (Bowers and Boutin, 2008; McIndoe et al., 2008; Malik et al., 2012). be used as biological indicators of stream condition. Biological indicators integrate the spatial and temporal effects of the environmental conditions on inhabiting organisms (Haury et al., 2006; Thiébaut et al., 2006) and are appropriate for evaluating the potential effects of many aspects of changes in aquatic ecosystems (Grasmück et al., 1995; Robach et al., 1996). However, there are few such survey of environmental condition of streams in tropical Africa is rarer (Kennedy et al., 2015).

In this study, we assessed the effects of land use on plant communities in the riparian zone and the water quality of tropical highland streams draining different land uses (forest, mixed, eucalyptus and agriculture). The water quality of the highland streams was assessed using common chemical indicators of stream health; water temperature, dissolved oxygen, conductivity, suspended solids, nitrate, and orthophosphate. Floristic quality of the riparian zone was evaluated using metrics such as riparian condition index, floristic quality index, and species richness. The main hypothesis was that land use predict the water quality of highland streams and riparian conditions, including their floristic composition.

Methods and materials Study area

3

The area of the present study are located in the Gilgle Gibe catchment area in the Jimma Zone, Southwestern Ethiopia (latitude 7°25′–7°55′N and longitude 36°30′–37°22′E) (Fig 1) at altitudes ranging from 3259 meters in the headwater to 1096 meters above sea level in the lowland (Ambelu et al., 2010).

The region was once covered with tropical sub-humid vegetation and coffee forest (Keddi and Moges, 2016), but is now dominated by cropland (wheat, barley, faba bean, sorghum, and maize) and to a lesser extent grazing land, and where riparian areas are to different degrees covered by vegetation (Demissie et al., 2013). The climate of the study area is classified as tropical humid and belongs to the high altitude cool tropic area of the country; this is also the wettest part with an average rainfall of 1550 mm per year and an average temperature of 19°C (Demissie et al., 2013).

Sampling Design Satellite images were used to delineate stream and quantify land use of the study area. Further, remote sensing data and arial photographs (scale of 1:50,000) were used to digitalise the study stream and sub-watersheds. Before actual data collection, a preliminary field survey had been conducted to obtain data on the physical conditions of the study area, including vegetation cover and different land use categories. Afterward, eighteen permanent streams ranging from 1st to 3rd order were selected following the classification by Rosgen (1985). Data derived from the GIS covers were compared with that of field observation and potential sites were randomly positioned on the land use map. The selected streams were representatives of the catchment in terms of botanical composition, management scenario, and geomorphological diversity. Thirty-five sampling sites were located at various locations to encompass the range of a priori land use categories (Forest, Mixed, eucalyptus, and agriculture). Hereafter, riparian areas embedded in primary forest land uses are called forest sites; riparian embedded in primary mixed of pasture and bushland are called mixed sites; riparian embedded in primary eucalyptus plantation are called eucalyptus sites, and riparian embedded in primary agricultural land use are called agricultural sites. This information was used to construct a numerical classification ranking of sites based on land use category (1= agriculture, 2= eucalyptus, 3= mixed and 4= forest) after Kasangaki et al., (2008). Sites were sampled in agriculture (15), mixed (9), eucalyptus plantation (4) and forest (7) land use categories. The altitude of the streams included ranged from 2116 meters in the headwater to 1638 meters above sea level in the downstream area. All map manipulations and spatial analyses were performed using ESRI ArcGIS 10.3 and Erdas Imagine 10.0 software.

4

Following Burton et al., (2005), at each selected riparian sampling site (Fig. 1) 2-4 transects spaced 100 m apart were established extending perpendicularly to the stream edge and upland on both sides of the stream. Then 100 m2 (5 m  20 m) plots were established (with a long face parallel to the stream). Within each plot, all woody vegetation stands were recorded. Herbaceous plant species were collected and recorded using 1m by 1m (1m2) plots, which was placed at the center of the main plot. Adjacent land-use and any evidence of disturbances (stream canalization, grazing, cutting, and firewood collection) were also recorded. Samples were brought back to Jimma University and identified using published flora of Ethiopia. Specimens which were difficult to identify using these methods were brought to the National Herbarium of Addis Ababa University and identified by comparing with authentically identified specimens.

Water samples were collected at thirty-five sampling sites (Fig 1) twice during the season: the wet season (September 2014) and the dry season (January 2015). At each site, water temperature, pH, dissolved oxygen (DO), and electrical conductivity (EC) were measured using a multi-probe meter (HQd4 Single-Input Multi-Parameter Digital Meter, HACH) and turbidity was measured using a Wagtech turbidity meter (Wag-WT3020). At each site one liter of water was collected and stored on ice until return to the Laboratory of Environmental Health at Jimma University, where the samples were analyzed for total suspended sediment (TSS), nitrate and orthophosphate according to standard methods as prescribed by APHA et al., (1995).

Data analysis Presence/absence data of species for each transect and at the sites were pooled to create a single list of identified species per site. Riparian index (RI) was applied using the following metrics; riparian buffer zone, bank stability, and vegetation cover following (Rosso and Cirelli, 2013). The following indices were tested on riparian plant data: species richness (S), the Shannon diversity index (H’), (Shannon, 1948), Simpson index (D) (Simpson, 1946), and Floristic Quality Index (FQI). FQI is one of the assessment means that may show habitat condition of an area (Matthews et al., 2015). The FQI uses measure of ecological conservatism and species richness as measure of habitat quality of an area and ecological conservatism is showed numerically as a coefficient of conservatism (Jog et al., 2006). The values of coefficient conservatism (CC) is ranging from 0 to 10 and is based on individual plant species’ fidelity to definite habitat types and its tolerance to both natural and human disturbance (Bowers and Boutin, 2008). According to Freyman et al. (2016), a value of zero was assigned to exotic taxa and native taxa

5

that act as opportunistic invaders, values of one to three were assigned to taxa that are widely distributed and occur in disturbed ecosystems and a value of four to six were assigned to taxa with a faithfulness to a particular ecosystem, whereas taxa typical of well-established ecosystems that sustain only minor disturbances were assigned a value of seven or eight, and lastly taxa with high degrees of fidelity to a narrow set of stable ecological conditions were assigned a rank of nine or ten. The FQI score for an individual site was calculated as :I'= ∑ (CCi)/ √(N

all species richness),

where I' = the modified FQI score, CCi = the coefficient of conservatism of plant species i,

and N all species = the total number of species both native and non- native (Andreas et al., 2004; Taft et al., 2006).

The Spearman rank correlation was used to determine the relationships between water quality variables, RI, and riparian plant metrics. Significant effects of land use on RI, FQI and species richness were determined using a one-way ANOVA, and significant differences among the groups were identified using Tukey’s multiple range test (α=0.05). All of the ANOVA test adequately met assumption of normality and equality of variance. The Spearman rank correlations and ANOVA were performed using Sigma Plot version 12.0.

A multivariate principal component analysis (PCA) was used to ordinate sampling site in relation to water quality and riparian condition, using PC-ORD 5 software (McCune and Mefford, 2013). This analysis is considered ideal for reducing a large number of equivalent response down to a smaller number of summary variable and is effective in identifying patterns that can be modeled linearly (McCune and Befford, 1999).

Results Environmental conditions The streams were characterized by a wide range of physical, chemical and riparian conditions. Agricultural sites had significantly (Tukey’s post hoc test; p<0.05) higher turbidity, TSS, and orthophosphate, than that of forest, mixed and eucalyptus sites. Forest sites had significantly (Tukey’s post hoc test; p<0.05) higher dissolved oxygen compared to that of the three other land use types. However, water pH, conductivity, and water temperature were not significantly (p>0.05) different among land use types. Water nitrate was significantly (Tukey’s post hoc test; p<0.05) different only between plantation and agricultural sites (Appendix A). The riparian conditions ranged from highly disturbed situations lacking any riparian cover and bank stability (RI = 0.20) to well developed riparian buffer zones and protected banks (RI = 2.85). There was a significant difference in RI between forest and agricultural sites (Tukey’s post hoc test; P<0.001) (Fig. 2).

6

PCA performed with 10 environmental parameters explained 57.9% of data variation on the first two axes (Table 1, Appendix C). The linear (Pearson’s r) relationships between the PCA scores and the individual variables indicated that axis 1 was positively correlated with higher values of water temperature (r = 0.769) turbidity (r = 0.0.877), TSS (r= 0.867) and orthophosphate (r=821) to which most agricultural sites were associated. On the negative side of this axis, forest sites were associated with higher values of DO (r = –0.737), RI (r = –0.792) and altitude (r = –0.356). Axis 2 was associated with pH (r = -0.782) and EC (r = -0.824) having a negative relationship.

Riparian vegetation and land use We identified 107 species, represented by 90 genera and 49 families. Of these, (95%) species were native while (5%) species were non-native. The most species-rich families were: Fabaceae (14 species) Asteraceae (10 species), Lamiaceae (7 species) Euphorbiaceae and Poaceae (6 species each). The most frequent species was Croton macrostachyus, which was found at 80% of the study sites. The second most frequent species was Senna didymobotrya found on 60 % of the study sites. Species richness (S), Shannon diversity (H), and Simpson diversity index (D) were calculated based on the riparian plant assemblage for each sample streams. Species richness ranged from 25 species at SE-01 site (forest sites) to two species at ONK-01 and KK-03 (agricultural sites). Shannon diversity ranged from 0.69 to 3.22 at, KK-03 and SE-01 respectively, similar to species richness. Simpson index ranged from a low of 0.50 at ONK-01 and KK-03 (agricultural sites) to a high of 0.96 at SE-01 (forest sites). The modified FQI ranged from 2.12 in the agricultural sites to 27.11 in a forest sites. Comparisons across land use show that species richness and FQI values varied per land use, with the highest scores in the forest sites and the lower score in the agricultural sites. A Tukey, pairwise comparison test between the mean FQI values suggested that forest land use differed significantly (p<0.001) from mixed, plantation and agricultural land use type (Fig 3). The average FQI (20.58) for forested sites were more than double the average for agricultural sites 6.95. Species richness also significantly deferred (p< 0.001) between forest sites and agricultural sites, but there is no significant (p>0.001) differences between forest and mixed sites.

Riparian plant metrics and environmental variables Riparian index was positively correlated with land use (r=0.759, p<0.0 and dissolved oxygen (r=0.656, p<0.01), and negatively correlated with water temperature (temperature=-0.-0.603, p<0.01), turbidity (r=-0.584, p< 0.01), TSS (r=-0.610, p=<0.01) and orthophosphate (-0.619, p<0.01). Floristic quality index and species richness showed

7

similar correlations with environmental variables. FQI was positively correlated with land use (r=0.805, p < 0.01) DO (r=673, p<0.01) and RI (r= 0.710, p< 0.01). Similarly, a positive correlation was between species richness and four environmental variables: land use (r=0.668, p<0.01), DO (r=0.558, p<0.01), and RI (r=0.539, p<0.01). A significant negative relationship was found between FQI and four water parameters: water temperature (r=-498, p<0.01), turbidity (r=- 528, p<0.01), TSS (r=-672, p<0.01), and orthophosphate (r=-529, p<0.01). Likewise, species richness negatively correlated with water temperature (r=-0.467, p<0.01), turbidity (r=-0.492, p<0.01), TSS (r=-0.589, p<0.01) and orthophosphate (r=-0.500, p<0.01. Likewise, both Shannon and Simpson diversity index showed the same trend with FQI and species richness.

Discussion Environmental condition Riparian land use has substantial effects on aquatic habitats and biological communities (Jun et al., 2011) and loss of natural riparian vegetation can significantly affect the physico-chemical properties of streams (Allan, 2004). Our findings revealed that water quality and riparian condition index were closely associated with the predominant land use in the surrounding landscape. The result of both the PCA and analysis of variance (ANOVA) show definite patterns among land use and water quality parameters. The results thus suggest that impaired areas are having more impact on water quality than the atural habitat. In a similar study in the Bwindi Impenetrable National Park Uganda, Kasangaki et al. (2008) found that deforested sites had poor water quality conditions compared to forested riparian sites. As for major pollutants, such as TSS, turbidity and orthophosphate, the present data highlight higher values in agricultural site than forest site, indicating pollutant were more associated with human land use modification (Busulwa and Bailey, 2004; Melaku et al., 2007; Monteiro et al., 2016). Here, the multiple statistical analysis indicates a significant relationship between water quality variable, land use, and riparian condition index. For example, the agricultural sites generally had the highest turbidity, TSS, and orthophosphate values, while the forest sites generally exhibited low turbidity, TSS and orthophosphate values. The mean turbidity of agricultural sites was four times higher compared to the mean value of forested sites despite comparable elevations. The high turbidity, TSS, and orthophosphate in the agricultural sites are most likely due to the high load of suspended materials, increased runoff from agricultural fields, and riverbank erosion (Buck et al., 2004). Similarly, riparian condition index was generally higher at forested sites, which is consistent with (Rosso and Cirelli, 2013) who reported that protected streams have higher riparian condition index scores than degraded streams. The lower riparian condition index score at agricultural sites could be due to modification of the riparian

8

buffer zone. This situation commonly affect the surface water quality, channel morphology, and the biological properties of streams (Hubble et al., 2010).

Riparian plant metrics and environmental variables Although it is well established that riparian vegetation provide many important functions (Souza et al., 2013; Rosso and Cirelli, 2013; Taniwaki et al., 2017), removal of vegetation and degradation in general result in poor functioning riparian areas (Méndez-Toribio et al., 2014). Particularly in agricultural watersheds, the effects of riparian vegetation loss negatively affect stream water quality and aquatic food webs (Moore et al., 2014), which affect species composition and structure of the biota (Iñiguez–Armijos et al., 2014). In our survey, it was evident that local anthropogenic activities at agricultural sites impaired riparian plant composition and diversity. Riparian vegetation was extremely vulnerable to anthropogenic impacts resulting from the expansion of agricultural land (Aguiar et al., 2001; Larrañaga et al., 2009). As the disturbance in an area increases, the plant community exhibits a decrease in floristic quality (Bowers and Boutin, 2008).

Riparian plant metrics were biological elements that are included as potential indicators of ecological quality, for use in the bioassessment and monitoring of the ecological status of water bodies (Ferreira and Aguiar, 2006). Physio- chemical parameter such as TSS, turbidity, water temperature, and orthophosphate showed a negative correlation with plant metrics (FQI, species richness, and diversity indices). Riparian zone enclosed by forest has higher score of metrics and characterized by better water quality than channelized agricultural sites (Sweeney and Newbold, 2014). The tested metrics are negatively correlated with land modification and supported the results found by Miller and Wardrop (2006). This may indicate the usefulness of riparian plant metrics for detecting impact of anthropogenic activities along streams and rivers of tropical Africa.

Due to their importance, integral riparian areas are essential for diminishing negative impacts of land use practices on streams. Research data collected so far suggests that increases in vegetation cover improved riparian condition and water quality relative to other non-vegetated areas (Uriarte et al., 2011). The observations of water quality variation in the studied streams suggest that riparian vegetation along rivers and streams significantly reduces not only riverbank erosion but also filter sediment coming from upslope agricultural field (Li et al., 2009) and provide important ecosystem functions for the maintenance of stream habitat quality (Hickey and Doran, 2004; Helmers

9

et al., 2008). Thus, restoration of riparian ecosystems should be a high priority for water resource conservation in particular along the east African highland streams, where there are intensive agricultural activities.

Conclusions There is a strong consensus that riparian vegetation constitutes an important component of freshwater ecosystems (Hough-Snee et al., 2015). Thus, effective water management in freshwater ecosystems calls for understanding of how the surrounding terrestrial landscape, including the riparian zone, determines the structure and functioning of streams and rivers also in tropical highlands. This study showed that lotic ecosystems exposed to agricultural land use displayed significantly different conditions regarding in the physico-chemical (e.g., turbidity, TSS, and orthophosphate) than forest sites. Additionally, we provide evidence suggesting that both RI and FQI significantly correlated with riparian land use, thus, restoration of riparian ecosystems should be of high priority for water resource conservation in the particular along East African highland streams

Acknowledgement This research was supported by the International Foundation for Science (IFS) (grant no. W/5387). The authors express their sincerest gratitude to Mr. Melaku Wondafrash for his assistances in plant specimen identification and to all survey members involved in the project for their help in sampling and analysis. We also thank comments and suggestions from two anonymous reviewers.

References Aguiar, F.C., Ferreira, M.T., Moreira, I., 2001. Exotic and native vegetation establishment following channelization of a western Iberian river. Regul. Rivers Res. Manag. 17, 509–526. doi:10.1002/rrr.642 Allan, J.D., 2004. Landscape and riverscape: The Influence of Land Use on Stream Ecosystems. Annu. Rev. Ecol. Evol. Syst. 35, 257–284. doi:10.1146/annurev.ecolsys.35.120202.110122 Ambelu, A., Lock, K., Goethals, P., 2010. Comparison of modelling techniques to predict macroinvertebrate community

composition

in

rivers

of

Ethiopia.

Ecol.

Inform.

5,

147–152.

doi:10.1016/j.ecoinf.2009.12.004

10

Andreas, B.K., Mack, J.J., McCormac, J.S., 2004. Floristic Quality Assessment Index (FQAI) for vascular plants and mosses for the State of Ohio. Ohio EPA. APHA, AWWA, WPCF, 1995. Standard Methods for the Examination of Water and Wastewater, 19th ed. American Public Health Association, Washington DC. Berges, S.A., 2009. Ecosystem services of riparian areas: Stream bank stability and avian habitat (Graduate Theses and Dissertations. Paper 11065). Bowers, K., Boutin, C., 2008. Evaluating the relationship between floristic quality and measures of plant biodiversity along stream bank habitats. Ecol. Indic. 8, 466–475. doi:10.1016/j.ecolind.2007.05.001 Bu, H., Meng, W., Zhang, Y., 2014. Spatial and seasonal characteristics of river water chemistry in the Taizi River in Northeast China. Environ. Monit. Assess. 186, 3619–3632. doi:10.1007/s10661-014-3644-6 Buck, O., Niyogi, D.K., Townsend, C.R., 2004. Scale-dependence of land use effects on water quality of streams in agricultural catchments. Environ. Pollut. 130, 287–299. Burton, M.L., Samuelson, L.J., Pan, S., 2005. Riparian woody plant diversity and forest structure along an urbanrural gradient. Urban Ecosyst. 8, 93–106. Busulwa, H.S., Bailey, R.G., 2004. Aspects of the physico-chemical environment of the Rwenzori rivers, Uganda. Afr. J. Ecol. 42, 87–92. Cavaillé, P., Dommanget, F., Daumergue, N., Loucougaray, G., Spiegelberger, T., Tabacchi, E., Evette, A., 2013. Biodiversity assessment following a naturality gradient of riverbank protection structures in French prealps rivers. Ecol. Eng. 53, 23–30. doi:10.1016/j.ecoleng.2012.12.105 Daniel, H., Bernez, I., Haury, J., 2006. Relationships between macrophytic vegetation and physical features of river habitats: the need for a morphological approach, in: Macrophytes in Aquatic Ecosystems: From Biology to Management. Springer, pp. 11–17. Demissie, T., Saathoff, F., Seleshi, Y., Gebissa, A., 2013. Evaluating the Effectiveness of Best Management Practices in Gilgel Gibe Basin Watershed—Ethiopia. J. Civ. Eng. Archit. 7, 1240–1252. Fernandes, M.R., Aguiar, F.C., Ferreira, M.T., 2011. Assessing riparian vegetation structure and the influence of land use using landscape metrics and geostatistical tools. Landsc. Urban Plan. 99, 166–177. doi:10.1016/j.landurbplan.2010.11.001 Ferreira, M.T., Aguiar, F.C., 2006. Riparian and aquatic vegetation in Mediterranean-type streams (western Iberia). Limnetica 25, 411–424.

11

Freyman, W.A., Masters, L.A., Packard, S., 2016. The Universal Floristic Quality Assessment (FQA) Calculator: an online tool for ecological assessment and monitoring. Methods Ecol. Evol. 7, 380–383. doi:10.1111/2041-210X.12491 Grasmück, N., Haury, J., Léglize, L., Muller, S., 1995. Assessment of the bio-indicator capacity of aquatic macrophytes using multivariate analysis, in: Space Partition within Aquatic Ecosystems. Springer, pp. 115–122. Haury, J., Peltre, M.-C., Trémolières, M., Barbe, J., Thiébaut, G., Bernez, I., Daniel, H., Chatenet, P., HaanArchipof, G., Muller, S., Dutartre, A., Laplace-Treyture, C., Cazaubon, A., Lambert-Servien, E., 2006. A new method to assess water trophy and organic pollution – the Macrophyte Biological Index for Rivers (IBMR): its application to different types of river and pollution. Hydrobiologia 570, 153–158. doi:10.1007/s10750-006-0175-3 Helmers, M.J., Isenhart, T.M., Dosskey, M.G., Dabney, S.M., Strock, J.S., 2008. Buffers and vegetative filter strips (Water Quality Concerns Workshop). the American Society of Agricultural and Biological Engineers, Michigan. Hickey, M.B.C., Doran, B., 2004. A review of the efficiency of buffer strips for the maintenance and enhancement of riparian ecosystems. Water Qual. Res. J. Can. 39, 311–317. Hough-Snee, N., Roper, B.B., Wheaton, J.M., Lokteff, R.L., 2015. Riparian Vegetation Communities of the American Pacific Northwest are Tied to Multi-Scale Environmental Filters. River Res. Appl. 31, 1151– 1165. doi:10.1002/rra.2815 Hubble, T.C.T., Docker, B.B., Rutherfurd, I.D., 2010. The role of riparian trees in maintaining riverbank stability: A

review

of

Australian

experience

and

practice.

Ecol.

Eng.

36,

292–304.

doi:10.1016/j.ecoleng.2009.04.006 Iñiguez–Armijos, C., Leiva, A., Frede, H., Hampel, H., Breuer, L., 2014. Deforestation and Benthic Indicators: How Much Vegetation Cover Is Needed to Sustain Healthy Andean Streams? PLoS ONE 9, 1–10. doi:10.1371/journal.pone.0105869 Jog, S., Kindscher, K., Questad, E., Foster, B., Loring, H., 2006. Floristic Quality as an Indicator of Native Species Diversity

in

Managed

Grasslands.

Nat.

Areas

J.

26,

149–167.

doi:10.3375/0885-

8608(2006)26[149:FQAAIO]2.0.CO;2 Jun, Y.-C., Kim, N.-Y., Kwon, S.-J., Han, S.-C., Hwang, I.-C., Park, J.-H., Won, D.-H., Byun, M.-S., Kong, H.Y., Lee, J.-E., Hwang, S.-J., 2011. Effects of land use on benthic macroinvertebrate communities:

12

Comparison of two mountain streams in Korea. Ann. Limnol. - Int. J. Limnol. 47, S35–S49. doi:10.1051/limn/2011018 Kasangaki, A., Chapman, L.J., Balirwa, J., 2008. Land use and the ecology of benthic macroinvertebrate assemblages of high-altitude rainforest streams in Uganda. Freshw. Biol. 53, 681–697. doi:10.1111/j.1365-2427.2007.01925.x Keddi, B., Moges, A., 2016. Identification of Soil Erosion Hotspots in Jimma Zone (Ethiopia) Using GIS Based Approach. Ethiop. J. Environ. Stud. Manag. 8, 926–938. doi:10.4314/ejesm.v8i2.7S Kennedy, M.P., Lang, P., Grimaldo, J.T., Martins, S.V., Bruce, A., Hastie, A., Lowe, S., Ali, M.M., Sichingabula, H., Dallas, H., Briggs, J., Murphy, K.J., 2015. Environmental drivers of aquatic macrophyte communities in southern tropical African rivers: Zambia as a case study. Aquat. Bot. 124, 19–28. doi:10.1016/j.aquabot.2015.03.002 Larrañaga, A., Basaguren, A., Pozo, J., 2009. Impacts of Eucalyptus globulus Plantations on Physiology and Population Densities of Invertebrates Inhabiting Iberian Atlantic Streams. Int. Rev. Hydrobiol. 94, 497– 511. doi:10.1002/iroh.200811156 Li, S., Gu, S., Tan, X., Zhang, Q., 2009. Water quality in the upper Han River basin, China: The impacts of land use/land

cover

in

riparian

buffer

zone.

J.

Hazard.

Mater.

165,

317–324.

doi:10.1016/j.jhazmat.2008.09.123 Lu, Y.H., Canuel, E.A., Bauer, J.E., Chambers, R.M., 2014. Effects of watershed land use on sources and nutritional value of particulate organic matter in temperate headwater streams. Aquat. Sci. 76, 419–436. doi:10.1007/s00027-014-0344-9 Malik, R.N., Shinwari, Z.K., Waheed, H., 2012. Linkages between spatial variations in riparian vegetation and floristic quality to the environmental heterogeneity a case study of river soan and its associated streams, Pakistan. Pak J Bot 44, 187–197. Matthews, J.W., Spyreas, G., Long, C.M., 2015. A null model test of Floristic Quality Assessment: Are plant species’ Coefficients of Conservatism valid? Ecol. Indic. 52, 1–7. doi:10.1016/j.ecolind.2014.11.017 McCune, B., Befford, M., 1999. PC-ORD multivariate analysis of Ecological data. Gleneden Beach, Oregon. McCune, B., Mefford, M., 2013. PC-ORD multivariate analysis of Ecological data. Version 5.0. MjM software. Gleneden Beach, Oregon. circulation 127, e6–e245.

13

McIndoe, J.M., Rothrock, P.E., Reber, R.T., Ruch, D.G., 2008. Monitoring tallgrass prairie restoration performance using floristic quality assessment, in: Proceedings of the Indiana Academy of Science. pp. 16–28. Meek, C.S., Richardson, D.M., Mucina, L., 2010. A river runs through it: Land-use and the composition of vegetation along a riparian corridor in the Cape Floristic Region, South Africa. Biol. Conserv. 143, 156– 164. doi:10.1016/j.biocon.2009.09.021 Melaku, S., Wondimu, T., Dams, R., Moens, L., 2007. Pollution status of Tinishu Akaki River and its tributaries (Ethiopia) evaluated using physico-chemical parameters, major ions, and nutrients. Bull. Chem. Soc. Ethiop. 21, 13–22. Méndez-Toribio, M., Zermeño-Hernández, I., Ibarra-Manríquez, G., 2014. Effect of land use on the structure and diversity of riparian vegetation in the Duero river watershed in Michoacán, Mexico. Plant Ecol. 215, 285–296. doi:10.1007/s11258-014-0297-z Miller, S.J., Wardrop, D.H., 2006. Adapting the floristic quality assessment index to indicate anthropogenic disturbance in central Pennsylvania wetlands. Ecol. Indic. 6, 313–326. Monteiro, J.A.F., Kamali, B., Srinivasan, R., Abbaspour, K., Gücker, B., 2016. Modelling the effect of riparian vegetation restoration on sediment transport in a human-impacted Brazilian catchment: Modelling Riparian Restoration. Ecohydrology 9, 1289–1303. doi:10.1002/eco.1726 Moore, J.W., Lambert, T.D., Heady, W.N., Honig, S.E., Osterback, A.-M.K., Phillis, C.C., Quiros, A.L., Retford, N.A., Herbst, D.B., 2014. Anthropogenic land-use signals propagate through stream food webs in a California, USA, watershed. Limnologica 46, 124–130. doi:10.1016/j.limno.2014.01.005 Mori, T., Murakami, M., Saitoh, T., 2009. Latitudinal gradients in stream invertebrate assemblages at a regional scale on Hokkaido Island, Japan: Latitudinal gradients in stream invertebrate assemblages. Freshw. Biol. 55, 1520–1532. doi:10.1111/j.1365-2427.2009.02363.x Piechnik, D.A., Goslee, S.C., Veith, T.L., Bishop, J.A., Brooks, R.P., 2012. Topographic placement of management practices in riparian zones to reduce water quality impacts from pastures. Landsc. Ecol. 27, 1307–1319. doi:10.1007/s10980-012-9783-7 Prabu, P.C., Wondimu, L., Tesso, M., 2010. Assessment of water quality of Huluka and Alaltu rivers of Ambo, Ethiopia. J. Agric. Sci. Technol. 13, 131–138. Riis, T., Sand-Jensen, K., Vestergaard, O., 2000. Plant communities in lowland Danish streams: species composition and environmental factors. Aquat. Bot. 66, 255–272.

14

Robach, F., Thiébaut, G., Trémolières, M., Muller, S., 1996. A reference system for continental running waters: plant communities as bioindicators of increasing eutrophication in alkaline and acidic waters in northeast France, in: Management and Ecology of Freshwater Plants. Springer, pp. 67–76. Rosgen, D.L., 1985. A stream classification system, in: Riparian Ecosystems and Their Management: Reconciling Conflicting Uses. First North American Riparian Conference, Arizona. pp. 91–95. Rosso, J.J., Cirelli, A.F., 2013. Effects of land use on environmental conditions and macrophytes in prairie lotic ecosystems. Limnologica 43, 18–26. doi:10.1016/j.limno.2012.06.001 Scott, M.L., Nagler, P.L., Glenn, E.P., Valdes-Casillas, C., Erker, J.A., Reynolds, E.W., Shafroth, P.B., GomezLimon, E., Jones, C.L., 2009. Assessing the extent and diversity of riparian ecosystems in Sonora, Mexico. Biodivers. Conserv. 18, 247–269. doi:10.1007/s10531-008-9473-6 Shannon, C.E., 1948. A mathematical theory of communication. Bell Syst. Tech. J. 27, 379–423. Simpson, E.H., 1946. Mesure of Diversity. Nature 163, 688. Souza, A.L.T. de, Fonseca, D.G., Libório, R.A., Tanaka, M.O., 2013. Influence of riparian vegetation and forest structure on the water quality of rural low-order streams in SE Brazil. For. Ecol. Manag. 298, 12–18. doi:10.1016/j.foreco.2013.02.022 Sweeney, B.W., Newbold, J.D., 2014. Streamside Forest Buffer Width Needed to Protect Stream Water Quality, Habitat, and Organisms: A Literature Review. JAWRA J. Am. Water Resour. Assoc. 50, 560–584. doi:10.1111/jawr.12203 Taft, J.B., Hauser, C., Robertson, K.R., 2006. Estimating floristic integrity in tallgrass prairie. Biol. Conserv. 131, 42–51. doi:10.1016/j.biocon.2006.02.006 Taniwaki, R.H., Cassiano, C.C., Filoso, S., Ferraz, S.F. de B., Camargo, P.B. de, Martinelli, L.A., 2017. Impacts of converting low-intensity pastureland to high-intensity bioenergy cropland on the water quality of tropical streams in Brazil. Sci. Total Environ. 584–585, 339–347. doi:10.1016/j.scitotenv.2016.12.150 Thiébaut, G., Muller, S., 1998. The impact of eutrophication on aquaticmacrophyte diversity in weakly mineralized streams in the Northern Vosges mountains (NE France). Biodivers. Conserv. 7, 1051–1068. Thiébaut, G., Tixier, G., Guérold, F., Muller, S., 2006. Comparison of different biological indices for the assessment of river quality: application to the upper river Moselle (France). Hydrobiologia 570, 159– 164. doi:10.1007/s10750-006-0176-2 Uriarte, M., Yackulic, C.B., Lim, Y., Arce-Nazario, J.A., 2011. Influence of land use on water quality in a tropical landscape: a multi-scale analysis. Landsc. Ecol. 26, 1151–1164. doi:10.1007/s10980-011-9642-y

15

Valle Junior, R.F., Varandas, S.G.P., Pacheco, F.A.L., Pereira, V.R., Santos, C.F., Cortes, R.M.V., Sanches Fernandes, L.F., 2015. Impacts of land use conflicts on riverine ecosystems. Land Use Policy 43, 48–62. doi:10.1016/j.landusepol.2014.10.015 Yuan, Y., Bingner, R.L., Locke, M.A., 2009. A review of effectiveness of vegetative buffers on sediment trapping in agricultural areas. Ecohydrology 2, 321–336.

16

Figure caption

Fig. 1. Map showing (A) location of the study area, (B) Gilgele Gibe watershed and the sampling sites. Streams are indicated by a solid blue line and sampling sites are indicated by solid red dots. The black dashed line delineates sub-watershed boundaries of the Gilgle Gibe river.

Fig. 2. Riparian index (RI) along four land use; FO (Forest), MX (mixed), PL (eucalyptus plantation) and AG (agriculture). Lower RI score indicating higher disturbances. Each solid line at the center describes the median; the boxes define the 25th and 75th percentile; whisker delimiting the minimum and maximum values, and dots indicates the outliers. Different letters were significantly different (Tukey’s post hoc test, P<0.05).

Fig. 3. Average FQI (open circles) and species richness (solid circles) across land use. Similar letters were not significantly different (Tukey, P<0.05). Land use classes are abbreviated as follows: FO= Forest, MX= Mixed, PL= Eucalyptus plantation and AG= Agriculture.

17

Fig 1.

18

Fig 2.

19

Fig 3.

20

Appendix C

Principal components analysis (PCA) plot of the abiotic variables measured along the highland streams in southwestern Ethiopia, considering four different riparian land use. Environmental variables are indicated by vectors. EC (electrical conductivity), Orto p (orthophosphate), DO (dissolved oxygen), TSS (total suspended solids), Tem (water temperature), and RI (riparian index).

21

Table 1. Loadings of the biotic variables on the first two principal components (PC) and proportion of variance explained by each component. Bold values highlight the variables composing factor structure.

parameter

PC1

PC2

Eigenvalue % variance explained Cumulative % variance explained pH Dissolved oxygen Electrical conductivity Water temperature Turbidity Total suspended solid Nitrate Orthophosphate Riparian index Altitude

4.21 42.10 42.10 -0.006 -0.735 0.015 0.769 0.877 0.867 0.360 0.821 -0.793 -0.356

1.58 15.79 57.89 -0.782 -0.098 -0.824 0.278 0.232 -0.088 -0.111 -0.210 0.279 -0.076

22

Appendix A The mean and standard deviation (SD) the water quality parameters among 4 a priori land use categories. Values across the rows with the same letter code are not significantly different (The Tukey’s post hoc test; p <0.05).

Forest

Mixed

Eucalyptus

Agriculture

pH

7.25 ± 0.20a

7.30 ± 0.16a

7.35±0.22a

7.32±0.22a

DO

7.23±0.17a

6.65±0.56ab

6.31±0.58b

6.18±0.51b

EC

90.31±14.69a

115.92±27.59a

112.44±39.15a

110.86±28.65a

Temperature

19.5±1.13a

19.81±1.67a

20.31±1.58a

21.73±2.32a

Turbidity

27.30±24.63a

48.87±30.14a

21.10±18.49a

120.04±70.52b

TSS

8.57±5.05a

30.46±16.25a

25.91±22.73a

55.26±21.76b

Nitrate

2.24±0.58ab

2.89±0.99ab

1.36±1.5a

3.02±0.89b

Orto-phosphate

0.07±0.07a

0.13±0.08a

0.09±0.08a

0.28±0.14b

23

Appendix B Spearman rank correlations among environmental variables and average water physicochemical for both rainy and dry season in the Gilgal Gibe river southwestern Ethiopia. DO (dissolved oxygen), EC (electrical conductivity), TSS (total suspended solid), Orto P (orthophosphate), RI (riparian index), FQI (floristic quality index), S (species richness), D (Simpson diversity index) and H (Shannon diversity index). Land use Land use pH DO EC

1.000

pH -0.091 1.000

DO 0.667

EC **

-0.173

Tem -0.513

Turbidity **

-0.621

**

TSS -0.736

Nitrate **

Orto P

-0.283

-0.656

**

RI 0.759

FQI **

0.805

S **

0.668

D **

0.668

H **

0.659**

0.151

0.363*

-0.209

-0.117

0.164

0.075

0.094

-0.109

0.083

0.180

0.201

0.205

1.000

0.021

-0.541**

-0.547**

-0.608**

-0.040

-0.509**

0.656**

0.673**

0.558**

0.556**

0.590**

1.000

-0.418*

-0.094

0.172

-0.010

0.114

-0.188

0.020

0.170

0.170

0.189

**

*

0.045

0.462

0.725**

0.238

0.815**

-0.584**

-0.528**

1.000

0.451**

0.631**

-0.610**

1.000

0.306 1.000

Tem

1.000

Turbidity

0.638

1.000

TSS Nitrate Orto P RI FQI S D H

0.338

**

**

-0.486**

-0.492**

-0.494**

-0.457**

-0.672**

-0.589**

-0.586**

-0.559**

-0.098

-0.242

-0.270

-0.269

-0.229

-0.619**

-0.529**

-0.500**

-0.500**

-0.468**

1.000

0.710**

0.539**

0.531**

0.545**

1.000

0.895**

0.896**

0.886**

1.000

0.999**

0.991**

1.000

0.990**

-0.603

**

-0.498

**

-0.467

**

-0.464

1.000 **. Correlation is significant at the 0.01 level. *. Correlation is significant at the 0.05 level.

24

Appendix D

Environmental conditions of sampled streams by means of riparian plant metrics, riparian and physico-chemistry variables. Riparian plant metrics: FQI (floristic quality index), S (species richness), D (Simpson diversity index) and H (Shannon diversity index). Riparian condition and Index. RBW (riparian buffer width), Veg. cover (vegetation cover, Bank sta. (bank stability) and RI (riparian index). Physico chemistry: pH, DO (dissolved oxygen), EC (electrical conductivity), Tem (water temperature), turbidity, TSS (total suspended solids), nitrate and Orto P (orthophosphate).

Physico- chemistry

Riparian condition and index

Riparian plant index

pH

DO

EC

Tem

Turbidity

TSS

Nitrate

Orto P

RBW

RBW%

Veg. cover

Bank sta.

RI

FQI

S

D

H

SK-02

7.37

6.98

104.10

19.5

10.09

4.45

2.36

0.03

12

60

65

100

2.25

22.8

16

0.94

2.77

SK-01

7.41

7.33

101.65

18.5

9.23

3.50

2.04

0.04

20

100

85

100

2.85

27.1

23

0.96

3.14

ONL-02

6.88

7.09

62.35

21.3

67.70

5.72

1.73

0.22

20

100

35

95

2.30

20.5

11

0.91

2.40

ONL-01

7.33

7.35

79.60

20.7

57.90

5.59

1.62

0.11

10

50

40

100

1.90

17.7

10

0.91

2.30

SBF-01

7.35

7.13

91.80

19.7

19.35

16.00

2.40

0.06

10

50

85

89

2.24

15.9

12

0.92

2.49

SE-01

7.40

7.41

94.90

18.3

13.45

14.50

3.38

0.02

7

35

45

100

1.80

26.8

25

0.96

3.22

DO-01

7.05

7.39

97.80

18.7

13.41

10.25

2.19

0.07

6

30

40

90

1.60

13.3

9

0.89

2.20

DQ-01

7.03

7.71

141.50

18.1

14.65

12.03

4.59

0.07

15

75

75

100

2.50

13.2

7

0.86

1.95

Code

25

SCH-01

7.48

6.95

103.05

17.6

35.05

49.50

3.44

0.07

10

50

90

85

2.25

15.0

13

0.92

2.57

KK-01

7.13

6.36

87.10

22.0

87.24

11.63

2.81

0.29

7

35

55

74

1.64

16.3

20

0.95

2.99

SW-03

7.53

6.56

96.75

20.0

36.45

40.88

2.64

0.08

5

25

65

70

1.60

12.4

8

0.88

2.08

SB-01

7.32

6.59

156.95

18.1

38.91

24.62

1.77

0.16

3.5

17.5

25

70

1.13

20.1

20

0.95

2.99

SB-02

7.21

5.60

155.55

19.1

45.47

26.38

1.36

0.06

3.5

17.5

20

75

1.13

12.7

15

0.93

2.71

SW-01

7.40

6.89

96.70

20.4

24.30

30.13

3.18

0.10

2

10

75

60

1.45

16.1

13

0.92

2.57

SC-03

7.33

6.55

110.15

21.3

108.57

58.63

2.51

0.20

2

10

20

50

0.80

15.8

13

0.92

2.57

SE-03

7.32

6.65

95.50

21.7

49.22

20.38

3.75

0.15

1

5

5

55

0.65

8.1

8

0.88

2.08

SCH-03

7.63

6.75

103.40

19.8

46.50

48.13

3.10

0.21

5

25

40

80

1.45

14.2

10

0.91

2.30

SBE-02

7.33

6.59

83.53

21.4

7.53

6.25

0.07

0.03

4

20

30

80

1.30

4.5

6

0.83

1.79

SBE-01

7.08

5.46

92.95

21.7

7.24

6.37

0.03

0.04

2

10

20

70

1.00

9.9

8

0.88

1.32

SBU-01

7.35

6.46

169.90

18.3

23.11

42.88

2.22

0.10

1.5

7.5

35

65

1.08

8.7

9

0.89

2.20

KB-01

7.11

5.99

88.00

22.9

237.65

82.37

2.39

0.12

8

40

25

70

1.35

9.5

8

0.88

2.08

DL-01

7.02

6.71

113.90

19.1

47.87

26.75

3.64

0.14

6

30

35

75

1.40

9.9

11

0.91

2.40

SCH-02

7.53

6.42

103.35

18.3

51.60

62.00

3.76

0.30

5

25

55

70

1.50

4.9

5

0.80

1.61

KC-01

7.19

5.09

67.35

22.5

130.85

56.25

2.54

0.29

3

15

12

65

0.92

2.1

2

0.50

0.69

SM-01

7.79

7.22

158.65

21.5

51.42

27.63

2.18

0.15

3

15

3

70

0.88

13.8

16

0.94

2.77

SG-01

7.38

5.76

151.10

21.1

129.15

50.38

2.25

0.39

3

15

2

20

0.37

6.4

3

0.67

1.10

DO-02

7.38

6.64

136.45

22.1

66.97

36.51

2.46

0.26

3

15

20

35

0.70

9.2

8

0.88

2.08

KK-02

7.11

6.08

87.90

23.4

163.97

59.38

2.72

0.13

3

15

10

35

0.60

5.8

3

0.67

1.10

KK-03

6.95

5.96

84.45

21.8

238.90

92.50

4.06

0.25

4.5

22.5

12

55

0.90

2.1

2

0.50

0.69

SW-02

7.27

6.08

100.45

20.3

38.65

42.00

3.28

0.19

2

10

50

45

1.05

10.7

5

0.80

1.61

SE-02

7.52

6.22

95.60

21.8

50.80

27.75

3.58

0.25

0.5

2.5

5

10

0.18

12.3

10

0.91

2.30

SG-02

7.51

6.04

160.95

20.6

137.30

58.13

2.28

0.48

1.5

7.5

6

15

0.29

7.8

8

0.88

2.08

SC-01

7.33

6.46

100.05

21.4

117.02

55.50

2.34

0.27

1

5

15

10

0.30

4.0

5

0.80

1.61

SC-02

7.44

6.44

96.55

20.7

116.51

57.13

2.58

0.33

1

5

12

10

0.27

3.5

3

0.67

1.10

ONK-01

7.34

5.62

118.10

28.6

221.95

94.61

5.29

0.65

1

5

5

10

0.20

2.1

2

0.50

0.69

26

27

Appendix E Lists of plant identified in the study sites.

Scientific Name

Family

Adiantum species Acacia species Acacia spp Acanthus polystachius Delile Ageratum houstonianum Mill. Albizia gummifera (J.F Gmel.) CA.Sm. Allophylus abyssinicus (Hochst.) Radlk. Apodytes dimidiata E. Mey. ex Arn. Asparagus africanus Lam. Bersama abyssinica Fresen subsp. abyssinica Bidens macroptera. (Sch. Bip. ex Chiov.) Mesfin Bidens pilosa L. Brucea antidysenterica J.F. Mill Brugmansia suaveolens (Humb. & Bonpl. ex Willd.) Bercht. & Presl Calpurnia aurea (Ait.) Benth Canarina abyssinica Engl. Canthium oligocarpum Hiern Carissa edulis (Forssk.) Vahl Cassipourea malosana (Baker) Alston Clausena anisata (Willd.) Hook.f. ex Benth. Clematis hirsuta Guill. & Perr. Clerodendrum myricoides (Hochst.) Steane & Mabb. Clutia abyssinica Jaub. & Spach Coffea arabica L Combretum paniculatum Vent. Commelina benghalensis L. Conyza bonariensis (L.) Cronq. Cordia africana Lam Croton macrostachyus Hochst. Ex Delile Cynodon dactylon (L.) Pers. Cyperus papyrus L Cyperus sesquiflorus (Torr.) Mattf & Kiik. Cyperus spp Dalbergia lactea Vatke

Adiantaceae Fabaceae Fabaceae Acanthaceae Asteraceae Fabaceae Sapindaceae Icacinaceae Asparagaceae Melianthaceae Asteraceae Asteraceae Simaroubaceae Solanaceae Fabaceae Campanulaceae Rubiaceae Apocynaceae Rhizophoraceae Rutaceae Ranunculaceae Lamiaceae Euphorbiaceae Rubiaceae Combretaceae Commelinaceae Asteraceae Boraginaceae Euphorbiaceae Poaceae Cyperaceae Cyperaceae Cyperaceae Fabaceae 28

Datura stramonium L. Dioscorea schimperiana Kunth Ehretia cymosa Thonn Ekebergia capensis Sparrm. Entada abyssinica A. Rich Erythrococca trichogyne (Muell. Arg.) Prain Eucalyptus grandis W.Hill Euclea racemosa L. Ficus sur Forssk Ficus thonningii Blume Ficus vasta Forssk. Galiniera saxifraga (Hochst.) Bridson Galinsogo parviflora Cav Gouania longispicata Engl. Grewia ferruginea Hochst. ex A. Rich. Guizotia scabra (Vis.) Chiov. Hibiscus berberidifolius A.Rich. Hibiscus macranthus Hochst. exA. Rich Hygrophila schulli (Hamilt.) MR. & S.M Almeida Hyparrhenia cymbaria (L.) Stapf Hypericum quartinianum A. Rich. Impatiens hochstetteri Warb. Indigofera sp. Lablab purpureus (L.) Sweet Lagenaria abyssinica (Hook.f.) C.Jeffrey Laggera crispata (Vahl) Hepper & J.R.I.Wood Lippia adoensis Hochst. Ludwigia abyssinica A. Rich. Maesa lanceolata Forssk Manilkara butugi Chiov. Maytenus arbutifolia (A. Rich.) Wilczek Millettia ferruginea (Hochst.) Bak Mimusops kummel Bruce ex A.DC. Myrsine africana L. Ocimum gratissimum L. Ocimum lamiifolium Hochst. Ex Benth Olea welwitschii (Knobl.) Gilg & Schellenb. Oncoba spinosa Forssk. Penniseturn macroururn Trin.

Solanaceae Dioscoreaceae Boraginaceae Meliaceae Fabaceae Euphorbiaceae Myrtaceae Ebenaceae Moraceae Moraceae Moraceae Rubiaceae Asteraceae Rhamnaceae Tiliaceae Asteraceae Malvaceae Malvaceae Acanthaceae Poaceae Gutiiferae Balsaminaceae Fabaceae Fabaceae Cucurbitaceae Asteraceae Verbenaceae Onagraceae Myrsinaceae Sapotaceae Celastraceae Fabaceae Sapotaceae Myrsinaceae Lamiaceae Lamiaceae Oleaceae Flacourtiaceae Poaceae 29

Penniseturn polystachion (L.) Schult. Penniseturn thunbergii Kunth Persicaria senegalensis (Meisn.) Soják Phoenix reclinata Jacq. Phyllanthus ovalifolius Forssk. Phytolacca dodecandra L'Her. Piper capense Lf. Pittosporum viridiflorum Sims Plectranthus alpinus (Vatke) Ryding Plectranthus punctatus (L.f.) L'Her Podocarpus falcatus (Thunb.) R.Br. ex Mirb. Pterolobium stellatum (Forssk.) Brenan Pycnostachys abyssinica Fresen Pycnostachys eminii Gürke Rhamnus prinoides L'Hér. Ricinus communis L. Rubus spp Rumex nepalensis Spreng. Rytigynia neglecta (Hiern) Robyns Salix mucronata Thunb. Sapium ellipticum (Krauss) Pax Senna didymobotrya (Fresen.) H.S. Irwin& Bameby Senna petersiana (Bolle) Lock Sesbania sesban (L.) Merr Setaria megaphylla (Steud.) T. Durand & Schinz Shirakiopsis ellipticum (Hochst.) Esser Sida rhombifolia L. Solanum anguivi Lam. Syzygium guineense (Willd.) DC Teclea nobilis Del. Trichilia dregeana Sond. Vernonia amygdalina Dellile Vernonia auriculifera Hiern Vernonia hochstetteri Sch. Bip. ex Walp.

Poaceae Poaceae Polygonaceae Arecaceae Euphorbiaceae Phytolaccaceae Piperaceae Pittosporaceae Lamiaceae Lamiaceae Podocarpaceae Fabaceae Lamiaceae Lamiaceae Rhamnaceae Euphorbiaceae Rouceae Polygonaceae Rubiaceae Salicaceae Euphorbiaceae Fabaceae Fabaceae Fabaceae Poaceae Euphorbiaceae Malvaceae Solanaceae Myrtaceae Rutaceae Meliaceae Asteraceae Asteraceae Asteraceae

30