INTRODUCTION:
Northwestern Ethiopia has long been considered an important regional resource, especially as these Ethiopian Highlands are the source of the Blue Nile, and the origin of approximately 85% of downstream waters1. Most recently, this region has become the renewed center of attention with the Grand Renaissance Dam proposal, which includes expanding an existing dam, along with the construction of four more dams along the upper Blue Nile basin2,3.
Given the well established possible social, environmental and economic impacts of dams4 and the already existing concern and condemnation over dams built in Ethiopia in the past two decades5,6,7 it is critical to consider the results of these potential dam projects and the implications it may have for local populations and the international region. The purpose of this project is to conduct an analysis of the potential downstream area and population affects of the four newly proposed dams. This study will primarily be limited to a.) downstream impacts and b.) area and population because of the almost complete absence of publicly available data on the exact locations, and technical information of the proposed dams that may allow for predicting what the resulting upstream flooded area will be, and any hydrologic analysis of the dams.
Figure 1: Study Area
METHODOLOGY:
Figure 2 provides an outline of the methodology used in this project. To begin, the primary data sets used are the Global Drainage Basin Dataset (GDBD ) and a population data with 2.5’ resolution. First this data was prepared by setting the appropriate projections and clipping the data sets to all Drainage Basins (DB) that intersected Ethiopia. The next step was to vectorize the population raster in order to make it usable for quantitative analysis of populations and eventually join this data set to the DB data. The next step involved geocoding the dam locations, however, due to no official location data being released, the placement was approximated based on a leaked image of the projected dam locations in Figure 3.
Figure 3 (Source8)
Once the dams were placed, it was possible to select various interpretations of the downstream flow including: an automated main river branch selection, a manual selection of the downstream path for two of the four dams, and finally the main stream along with associated tributaries. This selection was limited to the geographic region of Ethiopia for practicality and limitations of population data in Sudan. These river selections were used to identify the influenced DB, which then allowed for examination of area and population. DB area was chosen as the primary selection method because it ensures the downstream path is accurate and also indicates area that is hydrologically dependent on the upstream path, providing a greater probability that the population in those DB would be affected by upstream changes in hydrology. However, a buffer analysis of the main stream at 5,10,25,50 and 100km was additionally created to provide a more geographically continuous selection of population. Also in addition population center data was examined for significance compared to the whole country. Finally this data was aggregated, and organized for presentation.
GENERAL RESULTS:
The results initially provide a very wide spread of data, ranging from potentially as few as 17,000-18,000 affected people downstream of the Mendiya Dam to an extreme of over 20 to 21 million people affected downstream of the Chara Chara Weir. As Figure 4 demonstrates, there is a wide potential range of affected area, and is ultimately dependent on how far and how many upstream dams are built. Figure 4 also provides a more conservative estimation of influence areas as it bases its area selection on only the main stream, providing a range of 133,000 people affected downstream of Mendiya Dam to over 5.5 million affected if Chara Chara Wier is built. Additionally, though tributaries are not directly dependent on upstream flow, their regions may be affected by flooding patterns of the main branch, and also influence tributary fisheries9. Figure 5 demonstrates if the area selection is expanded to include tributaries downstream the potential affected population dramatically changes to over 1.8 million affected people downstream of Mendiya to almost 21 million downstream of Chara Chara Weir.
Figure 4: Main Stream Analysis
Figure 5: Main Stream with Tributaries Analysis
CHARA CHARA WEIR EXAMPLE:
Additional analysis was also conducted, and as an example case Chara Chara Weir will be used because it can serve as a limited aggregate of downstream impacts of the other three proposed dams. Figure 6 shows the results of a buffer analysis conducted on the main stream for Chara Chara Weir. Though the buffers do directly follow the hydrology of the DB they also provide a useful, and possibly more realistic, examination of the range of potential affected populations. The range is still very significant, with over 1 million affected people downstream in only 5km on the river. This number grows to close to 2 million in 10km and approaches 19.5 million within 100km.
Figure 6: Chara Chara Weir Downstream Buffer
These number are all significant percentages of the population, though it is also important to exam in conjunction with population data , the location of population centers, especially considering the population data is only at the resolution of 2.5’. Figure 7 examines the number of documented population centers within the DB affected downstream. Of 1120 documented centers in Ethiopia, 58 appear in this area, amounting to 4.75%. Additionally, of 239 population centers within a 100km radius of the main stream, these 58 account for 24.3%. Though the population center data point did not provide populations associated with each location, Figure 7 still demonstrates that at a minimum of 4.75%, a significant percentage of the population will be affected.
Figure 7: Population Center Locations
DISCUSSION: DATA LIMITATIONS
Though there are some significant implications of even these elementary area and population estimates, there are also some very important data limitations. As previously mentioned there is essentially no publically available data on the exact location or technical details, even Egypt, which will be greatly affected by the dams, has been provided very limited information10. With such limited data it is impossible to include the important affects upstream, such as the number of displaced people or loss of agriculturally suitable land that will occur as reservoirs are created. Additionally, because the Nile River has high seasonable variation and is subject to an flooding and desiccation regime, without reliable data on existing stream behavior and planned flooding after the dams are built it would be difficult to make a more realistic area estimation based on areas dependent on flooding for agriculture and aquaculture as was examined with a recent Nigerian Dam11.
Additional limitations include it is difficult to compare potential benefits, such as electricity generated from the dams. Electricity is essential, especially for the high population growth rate, though with no data on the infrastructure of affected regions, and reports that the Ethiopian government plans to sell a significant portion of their power production12, it is difficult to predict if locals will have access to this electricity, or even reservoir water.
DISCUSSION: IMPLICATIONS
There are some important implications from the data though it may be limited, namely it provides a basic working area and population to further investigate adverse impacts of dam construction. This project sets a basic geographic and population scale that is necessary when planning to mitigate future impacts or deal with projected scenarios. One important adverse has been strongly documented in Nigeria, and consists of possible shift to lower quality crops during the wet season, and less dry season, if any, cultivation due to reduced flooding13. If this occurs then an area of Africa already well known for periodic famines will further reduce its ability to survive these events and undermine what economic development is occurring. This data is useful then for preliminary planning and preparation, and can set a reasonable parameter for what to expect.
CONCLUSION:
In conclusion, the analysis of possible downstream data is severely limited by accessible data; however it is still critical to attempt to analyze various ranges of area and population that will undergo significant changes. With known adverse impacts of dam construction such as environmental degradation, interruption of economic and agricultural wellbeing and potential flooding risks, local analysis is important for adequate future humanitarian response and development planning.
REFERENCES:
1. http://www.terradaily.com/reports/Ethiopian_dams_on_Nile_stir_river_rivalry_999.html
2. http://www.internationalrivers.org/en/node/6701
3. http://danielberhane.wordpress.com/2011/07/01/leaked-map-of-ethiopias-4-new-dams-on-nile/
4. Scudder, T(2006). The future of large dams: dealing with social, environmental, institutional and political costs. Sterling, VA: Earthscan.
6. http://www.stopgibe3.org/pdf/Ethiopias%20Gibe%203%20Dam%20-%20Sowing%20Hunger%20and%20Conflict.pdf
8. http://danielberhane.wordpress.com/2011/07/01/leaked-map-of-ethiopias-4-new-dams-on-nile/
9. http://www.mekonginfo.org/mrc_en/doclib.nsf/0/BDAB7F0428BA45A647256DDD000FF1EF/$FILE/FULLTEXT.pdf
10. http://www.reuters.com/article/2011/04/18/us-ethiopia-nile-idUSTRE73H59220110418
11. http://www.jstor.org/stable/622179
12. http://www.reuters.com/article/2011/04/18/us-ethiopia-nile-idUSTRE73H59220110418
13. http://www.jstor.org/stable/622179
