Water abstraction of invasive Prosopis juliflora and native Senegalia senegal trees: A comparative study in the Great Rift Valley Area, Ethiopia.

Besides direct water abstraction, natural water scarcity in semi-arid and arid regions may be further exacerbated by human-assisted changes in vegetation composition, including the invasion by non-native plant species. Water abstraction by the invasive tree Prosopis juliflora and by the native Senegalia senegal was compared in the dry Great Rift Valley, Ethiopia. Transpiration rates were quantified using the heat ratio method on six trees each of P. julifora and S. senegal, growing adjacent to each other in the same environment. Water use for P. juliflora trees ranges from 1 to 26 L/day (an average of 4.74 ± 1.97), and that of S. senegal trees from 1 to 38 L/day (an average of 5.48 ± 5.29 during two study years). For both species, soil heat, latent heat, and soil moisture status influenced the rates of sap flow of trees; in addition, water use by P. juliflora trees was related to vapor pressure deficit; the higher the vapor pressure deficit, the higher the water abstraction by P. juliflora. Stand densities of pure P. juliflora and S. senegal were 1200-1600 trees and 400-600 trees per ha, respectively. At the stand scale, P. juliflora consumed approximately 6636 L/day/ha (transpiration: 242 mm per year) and S. senegal stands consumed 2723 L/day/ha (transpiration: 87 mm per year). That is, P. juliflora stands consumed three times more water than S. senegal stands, because of two reasons: (1) P. juliflora stands are denser than S. senegal stands, and denser stands consume more water than less dense stands, and (2) P. juliflora is evergreen and uses water all year-round, while S. senegal sheds its leaves during the peak dry seasons. Our findings suggest that, compared to S. senegal, P. juliflora invasion results in severe impacts on groundwater resources of the drylands of Ethiopia, with direct and indirect consequences to ecosystem services and rural livelihoods.

Monitoring and assessment of environmental resources in the changing landscape of Ethiopia: a focus on forests and water.

Monitoring and Assessment (M&A) of environmental resources aims to support the formulation of policies and follow up on outcomes of their implementation. In this study, the state of M&A is explored for Ethiopia with a focus on forests and water resources. The study is intended to serve as recommendations for future M&A applications in Ethiopia, as well as fulfillment of SDGs and other national and international commitments. Expert meetings, key informant interviews, and selected document analysis served as sources of information. The findings were summarized using qualitative grading and institutional mapping. Basic results of the study are that monitoring data on climate and streamflow are standardized in forms that can be communicated to policymakers. Scantier and less standardized are environmental data on soils, sediment transport, forests, biodiversity, and air quality. Water quality, soil moisture, groundwater level, forest biomass, and soil carbon are rarely monitored and can only be found in reports or studies for the fulfillment of academic degree requirements. Resources like nutrient fluxes have rarely been documented, not at all in some cases. There is considerable scope for tapping both technological advances and experiences of citizen science and local participation in environmental governance to rapidly expand and improve monitoring from local to regional and national scales. The study showed that there is a need for establishing a coordinated national system for monitoring and assessing the status of the environment, including the use of natural resources. Communicating such data to the scientific and wider public will support evidence-based planning and policy-making towards national development.

Evaluating the Effectiveness of Best Management Practices On Soil Erosion Reduction Using the SWAT Model: for the Case of Gumara Watershed, Abbay (Upper Blue Nile) Basin.

This study was conducted to evaluate the effectiveness of best management practices (BMPs) to reduce soil erosion in Gumara watershed of the Abbay (Upper Blue Nile) Basin using the Soil and Water Assessment Tool (SWAT) model. The model was calibrated (1995-2002) and validated (2003-2007) using the SWAT-CUP based on observed streamflow and sediment yield data at the watershed outlet. The study evaluated four individual BMP Scenarios; namely, filter strips (FS), stone/soil bunds (SSB), grassed waterways (GW) and reforestation of croplands (RC), and three blended BMP Scenarios, which combines individual BMPS of FS and RC (FS & RC), GW and RC (GW & RC), and SSB and GW (SSB & GW). Mean annual sediment yield at the baseline conditions was estimated at 19.7 t ha-1yr-1, which was reduced by 13.7, 30.5, 16.2 and 25.9% in the FS, SSB, GW, and RC Scenarios, respectively at the watershed scale. The highest reduction efficiency of 34% was achieved through the implementations of the SSB & GW Scenario. The GW & RC, and FS & RC Scenarios reduced the baseline sediment yield by 32% and 29.9%, respectively. The study therefore concluded that the combined Scenarios mainly SSB & GW, and GW & RC can be applied to reduce the high soil erosion in the Gumera watershed, and similar agro-ecological watersheds in Ethiopia. In cases where applying the combined scenarios is not possible, the SSB Scenario can yield significant soil erosion reduction.

Water use of Prosopis juliflora and its impacts on catchment water budget and rural livelihoods in Afar Region, Ethiopia.

Dense impenetrable thickets of invasive trees and shrubs compete with other water users and thus disrupt ecosystem functioning and services. This study assessed water use by the evergreen Prosopis juliflora, one of the dominant invasive tree species in semi-arid and arid ecosystems in the tropical regions of Eastern Africa. The objectives of the study were to (1) analyze the seasonal water use patterns of P. juliflora in various locations in Afar Region, Ethiopia, (2) up-scale the water use from individual tree transpiration and stand evapotranspiration (ET) to the entire invaded area, and 3) estimate the monetary value of water lost due to the invasion. The sap flow rates of individual P. juliflora trees were measured using the heat ratio method while stand ET was quantified using the eddy covariance method. Transpiration by individual trees ranged from 1-36 L/day, with an average of 7 L of water per tree per day. The daily average transpiration of a Prosopis tree was about 3.4 (± 0.5) mm and the daily average ET of a dense Prosopis stand was about 3.7 (± 1.6) mm. Using a fractional cover map of P. juliflora (over an area of 1.18 million ha), water use of P. juliflora in Afar Region was estimated to be approximately 3.1-3.3 billion m3/yr. This volume of water would be sufficient to irrigate about 460,000 ha of cotton or 330,000 ha of sugar cane, the main crops in the area, which would generate an estimated net benefit of approximately US$ 320 million and US$ 470 million per growing season from cotton and sugarcane, respectively. Hence, P. juliflora invasion in the Afar Region has serious impacts on water availability and on the provision of other ecosystem services and ultimately on rural livelihoods.

Implications of land use/land cover dynamics and Prosopis invasion on ecosystem service values in Afar Region, Ethiopia.

Land use/land cover (LULC) dynamics and the resulting changes in ecosystems, as well as the services they provide, are a consequence of human activities and environmental drivers, such as invasive alien plant species. This study assessed the changes in LULC and ecosystem service values (ESVs) in the Afar National Regional State, Ethiopia, which experiences a rapid invasion by the alien tree Prosopis juliflora (Swartz DC). Landsat satellite data of 1986, 2000 and 2017 were used in Random Forest algorithm to assess LULC changes in the last 31 years, to calculate net changes for different LULC types and the associated changes in ESVs. Kappa accuracies of 88% and higher were obtained for the three LULC classifications. Post-classification change analyses for the period between 1986 and 2017 revealed a positive net change for Prosopis invaded areas, cropland, salt flats, settlements and waterbodies. The rate of Prosopis invasion was estimated at 31,127 ha per year. Negative net changes were found for grassland, bareland, bush-shrub-woodland, and natural forests. According to the local community representatives, the four most important drivers of LULC dynamics were climate change, frequent droughts, invasive species and weak traditional law. Based on two different ESVs estimations, the ecosystem changes caused by LULC changes resulted in an average loss of ESVs in the study area of about US$ 602 million (range US$ 112 to 1091 million) over the last 31 years. With an increase in area by 965,000 ha, Prosopis-invaded land was the highest net change during the study period, followed by grassland (-599,000 ha), bareland (-329,000 ha) and bush-shrub-woodland (-327,000 ha). Our study provides evidence that LULC changes in the Afar Region have led to a significant loss in ESVs, with serious consequences for the livelihoods of the rural people.

Performances of machine learning algorithms for mapping fractional cover of an invasive plant species in a dryland ecosystem.

In recent years, an increasing number of distribution maps of invasive alien plant species (IAPS) have been published using different machine learning algorithms (MLAs). However, for designing spatially explicit management strategies, distribution maps should include information on the local cover/abundance of the IAPS. This study compares the performances of five MLAs: gradient boosting machine in two different implementations, random forest, support vector machine and deep learning neural network, one ensemble model and a generalized linear model; thereby identifying the best-performing ones in mapping the fractional cover/abundance and distribution of IPAS, in this case called Prosopis juliflora (SW. DC.). Field level Prosopis cover and spatial datasets of seventeen biophysical and anthropogenic variables were collected, processed, and used to train and validate the algorithms so as to generate fractional cover maps of Prosopis in the dryland ecosystem of the Afar Region, Ethiopia. Out of the seven tested algorithms, random forest performed the best with an accuracy of 92% and sensitivity and specificity >0.89. The next best-performing algorithms were the ensemble model and gradient boosting machine with an accuracy of 89% and 88%, respectively. The other tested algorithms achieved comparably low performances. The strong explanatory variables for Prosopis distributions in all models were NDVI, elevation, distance to villages and distance to rivers; rainfall, temperature, near-infrared and red reflectance, whereas topographic variables, except for elevation, did not contribute much to the current distribution of Prosopis. According to the random forest model, a total of 1.173 million ha (12.33% of the study region) was found to be invaded by Prosopis to varying degrees of cover. Our findings demonstrate that MLAs can be successfully used to develop fractional cover maps of plant species, particularly IAPS so as to design targeted and spatially explicit management strategies.

Modelling the current fractional cover of an invasive alien plant and drivers of its invasion in a dryland ecosystem.

The development of spatially differentiated management strategies against invasive alien plant species requires a detailed understanding of their current distribution and of the level of invasion across the invaded range. The objectives of this study were to estimate the current fractional cover gradient of invasive trees of the genus Prosopis in the Afar Region, Ethiopia, and to identify drivers of its invasion. We used seventeen explanatory variables describing Landsat 8 image reflectance, topography, climate and landscape structures to model the current cover of Prosopis across the invaded range using the random forest (RF) algorithm. Validation of the RF algorithm confirmed high model performance with an accuracy of 92% and a Kappa-coefficient of 0.8. We found that, within 35 years after its introduction, Prosopis has invaded approximately 1.17 million ha at different cover levels in the Afar Region (12.3% of the surface). Normalized difference vegetation index (NDVI) and elevation showed the highest explanatory power among the 17 variables, in terms of both the invader's overall distribution as well as areas with high cover. Villages and linear landscape structures (rivers and roads) were found to be more important drivers of future Prosopis invasion than environmental variables, such as climate and topography, suggesting that Prosopis is likely to continue spreading and increasing in abundance in the case study area if left uncontrolled. We discuss how information on the fractional cover and the drivers of invasion can help in developing spatially-explicit management recommendations against a target invasive plant species.

Variability, trends, and teleconnections of stream flows with large-scale climate signals in the Omo-Ghibe River Basin, Ethiopia.

This study assesses variability, trends, and teleconnections of stream flow with large-scale climate signals (global sea surface temperatures (SSTs)) for the Omo-Ghibe River Basin of Ethiopia. Fourteen hydrological indices of variability and extremes were defined from daily stream flow data series and analyzed for two common periods, which are 1972-2006 for 5 stations and 1982-2006 for 15 stations. The Mann-Kendall's test was used to detect trends at 0.05 significance level, and simple correlation analysis was applied to evaluate associations between the selected stream flow indices and SSTs. We found weak and mixed (upward and downward) trend signals for annual and wet (Kiremt) season flows. Indices generated for high-flow (flood) magnitudes showed the same weak trend signals. However, trend tests for flood frequencies and low-flow magnitudes showed little evidences of increasing change. It was also found that El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) are the major anomalies affecting stream flow variability in the Omo-Ghibe Basin. The strongest associations are observed between ENSO/Niño3.4 and the stream flow in August and September, mean Kiremt flow (July-September), and flood frequency (peak over threshold on average three peaks per year (POT3_Fre)). The findings of this study provide a general overview on the long-term stream flow variability and predictability of stream flows for the Omo-Ghibe River Basin.

Effects of land use and land cover on selected soil quality indicators in the headwater area of the Blue Nile basin of Ethiopia.

Understanding changes in soil quality resulting from land use and land management changes is important to design sustainable land management plans or interventions. This study evaluated the influence of land use and land cover (LULC) on key soil quality indicators (SQIs) within a small watershed (Jedeb) in the Blue Nile Basin of Ethiopia. Factor analysis based on principal component analysis (PCA) was used to determine different SQIs. Surface (0-15 cm) soil samples with four replications were collected from five main LULC types in the watershed (i.e., natural woody vegetation, plantation forest, grassland, cultivated land, and barren land) and at two elevation classes (upland and midland), and 13 soil properties were measured for each replicate. A factorial (2 × 5) multivariate analysis of variance (MANOVA) showed that LULC and altitude together significantly affected organic matter (OM) levels. However, LULC alone significantly affected bulk density and altitude alone significantly affected bulk density, soil acidity, and silt content. Afforestation of barren land with eucalypt trees can significantly increase the soil OM in the midland part but not in the upland part. Soils under grassland had a significantly higher bulk density than did soils under natural woody vegetation indicating that de-vegetation and conversion to grassland could lead to soil compaction. Thus, the historical LULC change in the Jedeb watershed has resulted in the loss of soil OM and increased soil compaction. The study shows that a land use and management system can be monitored if it degrades or maintains or improves the soil using key soil quality indicators.

Summer rains and dry seasons in the upper Blue Nile Basin: the predictability of half a century of past and future spatiotemporal patterns.

During the last 100 years the Ethiopian upper Blue Nile Basin (BNB) has undergone major changes in land use, and is now potentially facing changes in climate. Rainfall over BNB supplies over two-thirds of the water to the Nile and supports a large local population living mainly on subsistence agriculture. Regional food security is sensitive to both the amount and timing of rain and is already an important political challenge that will be further complicated if scenarios of climate change are realized. In this study a simple spatial model of the timing and duration of summer rains (Kiremt) and dry season (Bega), and annual rain over the upper BNB was established from observed data between 1952 and 2004. The model was used to explore potential impacts of climate change on these rains, using a down-scaled ECHAM5/MP1-OM scenario between 2050 and 2100. Over the observed period the amount, onset and duration of Kiremt rains and rain-free Bega days have exhibited a consistent spatial pattern. The spatially averaged annual rainfall was 1490 mm of which 93% was Kiremt rain. The average Kiremt rain and number of rainy days was higher in the southwest (322 days) and decreased towards the north (136 days). Under the 2050-2100 scenario, the annual mean rainfall is predicted to increase by 6% and maintain the same spatial pattern as in the past. A larger change in annual rainfall is expected in the southwest (ca. +130 mm) with a gradually smaller change towards the north (ca. +70 mm). Results highlight the need to account for the characteristic spatiotemporal zonation when planning water management and climate adaptation within the upper BNB. The presented simple spatial resolved models of the presence of Kiremt and annual total rainfall could be used as a baseline for such long-term planning.