case study 7 building a dam

Strengthening Urban Engagement of Universities in Africa and Asia

Case study seven: environmental impacts of dam construction in Kurdistan- the Zheveh dam in Sanandaj

This is the seventh case study from SUEUAA (Strengthening Urban Engagement in Universities in Asia and Africa), written by Dr Nematollah Azizi of the University of Kurdistan. The project is interested in how Universities in the Global South can contribute to solving geographical, economic and social issues in their cities. The project is looking at six cities in six different countries: Harare (Zimbabwe), Dar-es-Salaam (Tanzania), Johannesburg (South Africa), Manilla (Philippines), Duhok (Iraq), and Sanandaj (Iran). The SUEUAA project involves academics from each of these cities, who will be carrying out fieldwork, and elite interviews with decision makers in the city and senior academics from local Universities, to better understand the ability of Universities to respond to city issues.

In this seventh case study, we look at how the University of Kurdistan in Iran has responded to the negative environmental impacts on both urban and surroundings areas of dam construction in Kurdistan.

The increasing demand of humans for fresh water sources as well as energy production has led to constant attention to damming as one of the main pillars of developmental approaches. Moreover sustainable management of scarce agricultural water resources is an unavoidable necessity for countries facing the incremental water demands and suffering from water deficiencies, such as Iran [1]. In order to reserve the limited existent water resources, dam building can potentially help the more efficient use and allocation of them. But the ecological and environmental damages caused by the construction of dams must be assessed critically.

Kurdistan province, with a total area of 29,000 square kilometres, has 17 permanent rivers, and 24 rivers of seasonality. The region produces 6.6% of the country’s waterflow, and has 3.1% of the country’s precipitation. According to the governor of Kurdistan Regional Water Authority, three first-class watersheds in the country are located in Kurdistan [2]. Despite this, and in the line with the country’s policies for sustainable development,  in the time of the Shah only two Gheshlag and Kazemi dams were built in the province, which were used exclusively for the drinking water supply source in Sanandaj and West Azarbaijan.

Therefore due to the rainfall and the good potential of surface waters in Kurdistan province and the use of water and soil resources in recent twenty years, the dam construction industry has entered a new stage in Kurdistan. Recently, the construction of 12 dams was completed; with another eight other dams under construction and the study plan for the construction of another 14 dams has begun in Kurdistan [3, 4].

The dam construction industry in Kurdistan has an important job, as it has to balance a number of competing issues: controlling and developing water resources, supply increased drinking water, generate hydro-electricity, supply water to agriculture and industry, and also to develop sustainable flood control and food management policies. These have to be understood within the wider context of improving the environment of the rivers (including reinforcing of underground acquifiers in the plains), and encourage local employment and recreational opportunities. However, dam construction risks negative environmental and large scale social impact if these are not handled correctly.

One such consequence is the negative impact on rural populations. Construction of dams has led to the compulsory purchase of land and displacement of large part of the rural population, including those from historic sites. This leads to an increase in displaced and marginalised populations, poverty, and unemployment in Sanandaj. In addition to this, the allocated water from the dams cover less than 10% of Kurdistan’s agricultural lands, meaning the resources produced are not being given to local people. Finally, it also creates negative environmental impacts affecting the regions climate and ecosystem. These include: water pollution; air pollution; soil contamination; sound pollution; sediment (worsened through policies of sedimentation in the Province of Kurdistan); regional climate change; change in the shape of the earth; and destroying ecosystem (affecting plants and indigenous animals’ habitats).

Many departments at the University of Kurdistan have been engaged toward lowering the negative environmental impacts on both urban and surroundings areas. There have been studies exploring the impact of the dam construction on the natural and social environments. The findings of this, which will be discussed in more depth in a later publication, have found that implementing the policy of further dam construction without true comprehensive assessment of its consequences, especially for target communities, can practically lead to destruction of surface water resources and deprivation of natural and human ecosystems from these essential resources [5].

Findings of this study were used to successfully petition both dam construction project managers, and local officials to reconsider their construction plans. Academics in the Department of Urban and Natural Resources at the University of Kurdistan consulted with these officials to decrease the height of the Javeh and Daryan dams by 30 metres.

Academics at the University of Kurdistan believe that the evidence base they have collated can be used to stop damages in a number of sensitive areas, and may be used to employ more effective strategies and new methods for maximising outputs of natural resources. These new methods should have an evidence base behind them, with all future construction projects requiring a critical appraisal.

References:

[1] Azizi, V. (2015). A reflection on Environmental impacts of Reflection on the Environmental Impacts of Dam Construction: A Case Study of Jahve Dam. Project Report, University of Kurdistan.

[2] Kurdistan Regional Water Authority (2018). Dam Construction In Kurdistan: Emerging Area for Development.  http://www.kdrw.ir/

[3] Amini, A., and  Goftari, Z. (2017). Assessment of the Influence of Zhaveh Dam on the Neighboring Rural Areas: Sanandaj County.  Journal of Geography and Regional Development, 15 (1): 47-52.

[4] MOSHANIR (2018). ZHAVEH RESERVOIR DAM & WATER TRANSFER SYSTEM http://www.moshanir.co.ir/ En/zhaveh.aspx

• SUEUAA Case Studies
• Dar es Salaam
• Johannesburg
• Philippine Normal University
• University of Dar es Salaam
• University of Duhok
• University of Johannesburg
• University of Kurdistan
• University of Zimbabwe

Project in-depth: The Three Gorges Dam, China

Introduction to the Dam

Three Gorges Dam, China is the world’s largest hydroelectric facility. In 1994, work on the project started with the goal of not only creating power to fuel China’s rapid economic expansion but also controlling the country’s longest river, protecting millions of people from deadly floods, and elevating the project to a point of great technological achievement and national pride. Initially, the entire project took about two decades to complete, Construction on the Three Gorges Dam was completed in 2008 it cost 200 billion yuan ($28.6 billion). The dam is significantly larger than Brazil’s 12,600MW Itaipu facility, standing 185 meters tall and 2,309 meters wide as one of the largest hydroelectric plants in the world.

NSPD: Water Quality, Ecosystems

Stakeholder Types: Federated state/territorial/provincial government, Sovereign state/national/federal government, Local Government, Environmental interest, Community or organized citizens.

Sun Yat-sen first stated his plan to construct the dam in 1919. In addition to endorsing the project to mitigate the ongoing threat of flooding, Mao Zedong also revealed his plan to construct the dam in one of his most well-known poems, “Swimming” (1956). At the end of the first phase, two Chinese equipment vendors were crucial. 

Working with the two foreign groups, Harbin Power Equipment and Dongfang Electrical Machinery benefited from stringent technology transfer rules. Dongfang collaborated with the Voith General Electric and Siemens consortium, while Harbin worked with the Chinese company Oriental Motor and the Alstom, ABB, and Kvaerner grouping. Nearly the construction of the final two units of the first phase took place in China. Chinese groups were given construction chores. Contracts totaling $800 million were awarded to Gezhouba Share Holding, Yichang Qingyun Hydropower Joint Management, and Yichang Three Gorges Project Construction 378 Joint Management just before the equipment announcements. The powerplant and dikes were constructed as part of the work.

Time of Construction 

Because the Three Gorges Project needed to manage flood discharge on a massive scale, if a typical architecture of the flood-discharge orifices had been used, a large leading edge would have been needed for the discharge sections. The length of the flood-discharge dam sections had to be as short as possible while still meeting the requirements of energy dissipation and anti-scouring due to the large installed capacity of the power station and the large number of installed units. In addition, construction diversion and navigation needed to be considered.

Excavation and earth-filling 

To build the plant, 102.83 million cubic meters of rock and soil had to be excavated, and 31.98 million cubic meters had to be filled in.

Concrete and metal placement 

Additionally, 27.94 million cubic meters of concrete had to be placed, and a 256,500-ton metal frame had to be installed. 

Hydro turbine generator 

Production of hydroelectric power started small in 2003 and grew steadily as more turbine generators were added over the years until 2012 when all 32 of the dam’s turbine generator units were in use. With those units and two more generators, the dam could produce 22,500 megawatts of energy, making it the world’s most productive hydroelectric dam. With an annual power generating volume of 111.88 terawatt hours in 2020, the hydropower project set a new world record.

The Three Gorges Dam is a gravity structure made of concrete that is straight-crested and spans 2,335 meters (7,660 feet) with a maximum height of 185 meters (607 feet). Its design calls for 463,000 metric tons of steel and 28 million cubic meters (37 million cubic yards) of concrete. Large portions of the Qutang, Wu, and Xiling gorges are submerged by the dam for around 600 km (375 miles) upstream. 

This creates a massive deepwater reservoir that enables oceangoing freighters to travel 2,250 km (1,400 miles) inland on the East China Sea from Shanghai to the inland city of Chongqing. The complex’s five-tier ship locks, which let vessels weighing up to 10,000 tons pass the dam, and ship lift, which enables vessels weighing up to 3,000 tons to bypass the ship locks and pass the dam more rapidly, aid in the navigation of the dam and reservoir. The lift was the largest ship lift in the world when it was finished in late 2015. It measured 120 meters (394 feet) long, 18 meters (59 feet) broad, and 3.5 meters (11 feet) deep.

Context and debate surrounding the Three Gorges Dam

The concept for the Three Gorges Dam was first floated by Chinese Nationalist Party leaders in the 1920s. However, it gained fresh momentum in 1953 when Mao Zedong, the Chinese leader, ordered feasibility assessments of several locations. In-depth project planning started in 1955. The dam was not without its critics, despite the claims of its supporters that it would prevent catastrophic floods along the Yangtze, ease inland trade, and supply much-needed electricity for central China. The Three Gorges project was criticized from the moment the designs were put forth until they were completed. 

The threat of a dam collapse, the 1.3 million people (some critics claimed the number was closer to 1.9 million) who were uprooted from over 1,500 cities, towns, and villages along the river, and the devastation of numerous unique architectural and archaeological sites along with breathtaking landscapes were among the main issues. 

In addition, there were worries—some of which came true—that the reservoir would become contaminated by industrial and human waste from towns and that the massive volume of water it contained may cause landslides and earthquakes. Several smaller, far less expensive, and less problematic dams on the Yangtze tributaries, according to some Chinese and foreign engineers, could produce as much electricity as the Three Gorges Dam and regulate flooding almost as effectively. Building those dams, would, they claimed, allow the government to fulfill its top priorities without taking unnecessary chances.

An Environmental Catastrophe?

Chinese officials assert that the Three Gorges Dam has been successful in preventing floodwaters from spreading, despite the destruction. China Three Gorges Corporation, the dam’s operator, informed China’s official news agency Xinhua that 18.2 billion cubic meters of potential floodwater had been caught by the dam. 

The dam “effectively reduced the speed and extent of water level rises” on the middle and lower reaches of the Yangtze, an official from the water resources ministry told the state-run publication China Youth Daily. However, some geologists claim that the poor effectiveness of the Three Gorges Dam in preventing flooding has been exposed, given that numerous gauging stations monitoring river flows in the Yangtze basin are witnessing record-high water levels this summer it involved uprooting over a million people along the Yangtze River.

Furthermore, the government’s claim that the dam could shield the nearby villages from a “once-in-a-century flood” has been contested on multiple occasions. The Yangtze basin experienced its highest average rainfall in over 60 years since June, which led to the river and its numerous tributaries overflowing, reinforcing those fears. Over 158 persons have lost their lives or disappeared, 3.67 million residents have had to relocate, and 54.8 million individuals have been impacted, resulting in horrifying financial losses of 144 billion yuan ($20.5 billion).

Concerns about sustainable development and proper water management have surfaced internationally due to the project’s far-reaching effects. All of the measures as mentioned earlier place a strong emphasis on cooperative fact-finding, mutual benefits discussions, technical expertise, inclusion, transparency, and collaborative adaptive management , all of which are progressively enhancing Chinese governance in the areas of water management and dam construction.

China: The Three Gorges Dam Hydroelectric Project (no date) China: The Three Gorges Dam Hydroelectric Project – AquaPedia Case Study Database. Available at: https://aquapedia.waterdiplomacy.org/wiki/index.php?title=China%3A_The_Three_Gorges_Dam_Hydroelectric_Project (Accessed: 01 December 2023). 

Hvistendahl, M. (2013) China’s Three Gorges Dam: An environmental catastrophe? Scientific American. Available at: https://www.scientificamerican.com/article/chinas-three-gorges-dam-disaster/ (Accessed: 28 December 2023). 

Three Gorges Dam hydroelectric power plant, China (2021) Power Technology. Available at: https://www.power-technology.com/projects/gorges/?cf-view (Accessed: 01 December 2023). 

Three Gorges Dam (2023) Encyclopædia Britannica. Available at: https://www.britannica.com/topic/Three-Gorges-Dam (Accessed: 01 December 2023). 

Gan, N. (2020) China’s Three Gorges Dam is one of the largest ever created. was it worth it? CNN. Available at: https://edition.cnn.com/style/article/china-three-gorges-dam-intl-hnk-dst/index.html (Accessed: 01 December 2023). 

 (2022)The Three Gorges Project. Available at: https://www.engineering.org.cn/en/article/35384/detail (Accessed: 01 December 2023). 

A recent graduate, passionate about learning tangible and intangible concepts and ideas relating to space, time and people, is mostly interested in looking at how built spaces is a medium of cultural and social identity. Architecture for her is constant search. she is interested in representing built designs better with graphics,drawings and writing.

Assessment and Evaluation in Decolonized Design Education

Why travel to Kochi this December?

Related posts.

Project in-depth: Frank Sinatra Residence by Paul Williams

Project in-depth: Hale Lana, Hawaii, by Olson Kundig

Project in-depth: Hale Mau’u, Hawaii, by Walker Warner Architects

Project in-septh: Musubi, Hawaii, by Craig Steely Architecture

Project in-depth: Kaufmann House by Richard Neutra

Statesman Building, Delhi

• Architectural Community
• Architectural Facts
• RTF Architectural Reviews
• Architectural styles
• City and Architecture
• Fun & Architecture
• History of Architecture
• Design Studio Portfolios
• Designing for typologies
• RTF Design Inspiration
• Architecture News
• Career Advice
• Case Studies
• Construction & Materials
• Covid and Architecture
• Interior Design
• Know Your Architects
• Landscape Architecture
• Materials & Construction
• Product Design
• RTF Fresh Perspectives
• Sustainable Architecture
• Top Architects
• Travel and Architecture
• Rethinking The Future Awards 2022
• RTF Awards 2021 | Results
• GADA 2021 | Results
• RTF Awards 2020 | Results
• ACD Awards 2020 | Results
• GADA 2019 | Results
• ACD Awards 2018 | Results
• GADA 2018 | Results
• RTF Awards 2017 | Results
• RTF Sustainability Awards 2017 | Results
• RTF Sustainability Awards 2016 | Results
• RTF Sustainability Awards 2015 | Results
• RTF Awards 2014 | Results
• RTF Architectural Visualization Competition 2020 – Results
• Architectural Photography Competition 2020 – Results
• Designer’s Days of Quarantine Contest – Results
• Urban Sketching Competition May 2020 – Results
• RTF Essay Writing Competition April 2020 – Results
• Architectural Photography Competition 2019 – Finalists
• The Ultimate Thesis Guide
• Introduction to Landscape Architecture
• Perfect Guide to Architecting Your Career
• How to Design Architecture Portfolio
• How to Design Streets
• Introduction to Urban Design
• Introduction to Product Design
• Complete Guide to Dissertation Writing
• Introduction to Skyscraper Design
• Educational
• Hospitality
• Institutional
• Office Buildings
• Public Building
• Residential
• Sports & Recreation
• Temporary Structure
• Commercial Interior Design
• Corporate Interior Design
• Healthcare Interior Design
• Hospitality Interior Design
• Residential Interior Design
• Sustainability
• Transportation
• Urban Design
• Host your Course with RTF
• Architectural Writing Training Programme | WFH
• Editorial Internship | In-office
• Graphic Design Internship
• Research Internship | WFH
• Research Internship | New Delhi
• RTF | About RTF
• Submit Your Story

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 18 January 2021

Role of dams in reducing global flood exposure under climate change

• Julien Boulange   ORCID: orcid.org/0000-0003-2167-8761 1 ,
• Naota Hanasaki   ORCID: orcid.org/0000-0002-5092-7563 1 ,
• Dai Yamazaki   ORCID: orcid.org/0000-0002-6478-1841 2 &
• Yadu Pokhrel   ORCID: orcid.org/0000-0002-1367-216X 3  

Nature Communications volume  12 , Article number:  417 ( 2021 ) Cite this article

26k Accesses

145 Citations

188 Altmetric

Metrics details

• Climate-change impacts

Globally, flood risk is projected to increase in the future due to climate change and population growth. Here, we quantify the role of dams in flood mitigation, previously unaccounted for in global flood studies, by simulating the floodplain dynamics and flow regulation by dams. We show that, ignoring flow regulation by dams, the average number of people exposed to flooding below dams amount to 9.1 and 15.3 million per year, by the end of the 21 st century (holding population constant), for the representative concentration pathway (RCP) 2.6 and 6.0, respectively. Accounting for dams reduces the number of people exposed to floods by 20.6 and 12.9% (for RCP2.6 and RCP6.0, respectively). While environmental problems caused by dams warrant further investigations, our results indicate that consideration of dams significantly affect the estimation of future population exposure to flood, emphasizing the need to integrate them in model-based impact analysis of climate change.

Similar content being viewed by others

Floods and rivers: a circular causality perspective

The impacts of tidal wetland loss and coastal development on storm surge damages to people and property: a Hurricane Ike case-study

Extreme Coastal Water Levels Exacerbate Fluvial Flood Hazards in Northwestern Europe

Introduction.

Global warming is expected to increase flood risk by altering the distribution, variability, and intensity of precipitation events 1 , 2 . While global estimates of populations exposed to river flooding vary widely across studies, a 4–20 fold increase by the end of the 21 st century is commonly predicted 3 , 4 , 5 . To mitigate the destructive potential of floods and maximize water availability for human consumption, an estimated 2.8 million dams 6 have been constructed globally with a total water impoundment capacity ranging from 7,000 to 10,000 km 3 , which represents over one-sixth of the annual continental discharge to global oceans 7 , 8 , 9 . Currently, about half of the planet’s major river systems are regulated by dams 10 , 11 and only 23% of rivers worldwide flow uninterrupted to the ocean 6 . By regulating water flow, dams generally alter the frequency, duration, and timing of annual flooding events 12 . With more than 3,700 major dams planned or under construction worldwide 13 , understanding the role of dams in climate impact studies has become increasingly important. Previous studies on flood prediction, however, have neglected the role of dams 3 , 14 due to data scarcity 15 , difficulties in parameterizing reservoir outflows, and challenges in implementing features of dams that function at a scale smaller than those accounted for by global-scale models.

Previous global-scale analyses of floods have reconstructed historical flood patterns 16 , 17 to forecast future floods considering climate change 3 , 14 and/or socio-economic development factors 18 , 19 . A key conclusion of the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC) was that the number of people exposed annually to the equivalent of a historical 100-year river flood was projected to triple when compared to high and low emission scenarios. However, despite the regulation of most large rivers by dams, the extent to which their alterations of river and floodplain dynamics interacts with flooding, and the exposure of populations to floods in response to climate change remains largely unknown since dams have not been physically integrated into global flood-impact studies 3 , 14 , 15 , 20 . The few studies that have accounted for dams and/or flood protection have underscored the importance of considering dam-induced changes in streamflow characteristics in flood-hazard modelling 21 , 22 , 23 . In the contiguous United States (CONUS), dams are reported to reduce total flood exposure by 9% (protecting approximately 590 million people) owing to the medium to high dam attenuation effects on the 100-year return period discharge of 62% of CONUS hydrological units 22 .

Here, we provide the first global assessment of the role of dams in reducing future flood risk under climate change by using a modelling framework that integrates state-of-the-art global hydrological model with a new generation of global hydrodynamics model. Specifically, the modelling framework quantifies changes in the frequency of historical 100-year return period floods when dams are considered and estimates the global population at a reduced risk of flood exposure. Throughout this study, a flooding event is defined as extreme discharge associated with a 100-year return period (probability). We specifically investigated flood frequency (number of floods per year), the associated maximum flooded area, and populations exposed to these floods.

Streamflow regulation

Robust and reliable estimates of future river floods rely on two critical components: accurate reproduction of river discharge and appropriate prediction of floodplain inundation dynamics. In this study, we used two different models to simulate these critical processes globally. River discharge considering dams was simulated by H08, while flood inundation dynamics were simulated by the CaMa-Flood model. H08 is a global hydrological model that considers human interactions with the hydrological cycle. CaMa-Flood is an advanced river hydrodynamics model with an emphasis on efficient flow computation at the global scale (see Methods). Two global flood simulations were performed: one considering dams and one not considering dams. In total, four bias-corrected global circulation models (GCMs) combined with three radiative forcing scenarios (historical, RCP2.6, and RCP6.0) were used to force the models (see Methods).

The H08 model has been widely used and validated in global studies and accurately reproduces monthly river discharge in basins heavily affected by anthropogenic activities 24 . At the global scale, the H08 model has been benchmarked against other global hydrological models (GHMs) and has performed relatively well for reproducing the magnitude of high flows associated with different return periods 25 . H08 has also been calibrated and validated at finer spatial and temporal resolutions in multiple regional analyses, including the Chao-Phraya basin, the Ganges–Brahmaputra–Meghna basin, and Kyushu Island, among others 26 , 27 , 28 . Critical to these faithful discharge reproductions is the scheme used for dam operations. While improvements to the dam operation scheme implemented in H08 have been recently proposed 29 , 30 , it is still regarded as the benchmark to beat, given its ability to capture observed reservoir storage variation with high accuracy 31 . CaMa-Flood has also been extensively used and validated. It is capable of faithfully reproducing historical flood patterns 32 , 33 , 34 and daily measurements at river gauging stations across the globe 33 , partly owing to the integration of satellite-based topography data 35 . While both models have been widely used for climate impact assessments, they have never been coupled to analyze global-scale floods, leaving a gap in our understanding of the potential role of dams in reducing future flood risks. While the GCMs employed in this analysis were not assimilated, and consequently do not reproduce the exact timing of historical weather events, we nevertheless confirmed that our coupling framework can satisfactorily reproduce observed monthly discharges before and after dam construction (see Supplementary Figs.  13 – 23 ) and that its predicted maximum discharges in 33 large basins were reasonably similar to available observations (see Supplementary Fig.  24 ). We further compared global patterns of future floods with a previous publication 3 (Supplementary Figs.  1 and   5a, b ). We also compared the historical and predicted populations exposed to 100-year floods with information from published literature and a public database (see Supplementary Table  2 and Supplementary Note  1 ).

Population exposure to floods

Results indicate that, driven by climate change, the risk of floods will increase in the future. However, owing to the implementation of dams in our simulations, on average (range from the first and third quartiles in bracket represent uncertainty from the GCM ensemble), populations exposed to flooding below dams decreased by 16.3% (5.7–30.7%) in the RCP2.6 scenario and 12.8% (4.2–27.5%) in the RCP6.0 scenario, respectively, compared to the RCP simulations not considering dams (over 2006–2099, see Fig.  1 ). The decrease in the number of people exposed to floods due to the implementation of dams was highest during the last decade of the 21 st century for both RCPs. On average, 9.1 (4.6–18.1) million people were exposed to river floods in RCP2.6 (no dams) compared to 7.2 (3.5–15.1) million people in the simulation with dams. In the RCP6.0 scenario, the population exposed to river floods increased considerably to 15.3 (8.3–27.2) million and 13.4 (7.3–24.3) million for the simulations without and with dams, respectively. Large differences, consistent across experiments, in the number of people exposed to floods between the GCMs were apparent (Fig.  1b ). When population growth was taken into consideration using Shared Socioeconomic Pathways (SSPs) (see Methods), accounting for dams reduced populations exposed to flooding below dams by 20.6–32.0% for RCP2.6 and 7.0–16.8% for RCP6.0 (lowest and highest values across the five SSPs).

a 5-year moving averages of the population living below dams exposed to the historical 100-year river flood for historical (grey line) and future simulations for 2 RCPs and experiments (colour lines). The uncertainty range represent the spread among GCMs. b The 95 th and 5 th range (whiskers), median (horizontal lines in each bar), and 1 st and 3 rd quartiles (height of box) and individual mean values among GCMs (markers) of the population exposed to the historical 100-year flood for grid-cells located below dams over the 2070–2099 period.

Return period of future floods

Downstream of dams, historical 100-year floods occurred less frequently in the experiment considering dams than in the experiment with no dams for: (on average and ± standard deviation across GCMs), 66.6 ± 4.2% and 60.8 ± 12.7% of the grid-cells in RCP2.6 and RCP6.0, respectively (Fig.  2 , Supplementary Fig.  5c ). These results are similar to other regional- and country-scale analyses. For example, in the US, medium or large dam-attenuation effects were reported for 62% of hydrologic units 22 . Likewise, a study in Canada revealed that dams totally prevented flows with a return period greater than the historical 10-year recurrence 36 (see additional comparison with existing studies in Supplementary Note  3 ). Particularly prominent reductions in future flood frequency were observed along major sections of rivers containing multiple high-capacity dams (e.g. the Mississippi, Danube, and Paraná; see Supplementary Fig.  2 ). Reductions in 100-year flood frequencies in the experiments involving dams decreased moving downstream, becoming relatively small (or negligible) at the river mouth (e.g. in the Amazon, Congo, and Lena; see grey cells in Fig.  2 ). In a few locations (blue cells in Fig.  2 ), the presence of dams increased the frequency of historical 100-year floods compared to experiment without dams (6.7 ± 2.4% and 4.6 ± 1.1% for RCP2.6 and RCP6.0, respectively). This behaviour was connected to sporadic overflow events referred to as the pulsing effect by Masaki et al. 37 and has been documented for some rivers in the US 23 . Although water released from dams was regulated through the majority of the simulation period, pulsing events can result in a dam failure to prevent flooding, distorting the distributions of extreme discharge, and compromising the fitting of the extreme discharge to a Gumbel distribution (see Methods). In such cases, the definition of the 100-year flood is rather ambiguous, and while great efforts are made to prevent overflow 29 , not all are reflected in the generic scheme for dam (see Methods). Note that since the lead time before major storms is generally too short for preventive dams emptying, pulsing may not be totally averted in global dam simulations.

Grid-cells belonging to Köppen–Geiger regions BWk , BWh (hot and cold desert climates, respectively), and EF (ice cap climate) and for which the 30-year return period discharge was lower than 5 m s −1 were systematically screened out (see Methods). The case for representative concentration pathway (RCP) 6.0 is shown (RCP2.6 available in Supplementary Fig.  5c ).

Evolution of future floods for individual catchments

Median changes in the occurrence of historical 100-year river floods and the maximum flooded areas in the experiment considering dams relative to the experiment not considering dams were computed over the 2070–2099 period for 14 catchments (see Methods for the selection of catchments). Figure  3 indicates that the historical 100-year floods occurred less frequently in the experiment with dams, decreasing, on average, across catchments by 36.5% (26.6–49.1%) for RCP2.6 and 35.5% (28.8–46.6%) for RCP6.0. Similarly, the maximum flooded area in the catchments shrank on average by 22.5% (19.8–40.5%) and 25.9% (12.1–34.5%), for RCP2.6 and RCP6.0, respectively. These reductions in the occurrence of 100-year floods and maximum flooded areas were robust to the choice of extreme discharge indices used for identifying flood events (see Methods), with the exception of two catchments that experienced pulsing from dams (Supplementary Fig.  7 ). We note that by employing alternative extreme discharge indices (see Methods) to identify flood events, the eventual influence of pulsing events on the occurrence of 100-year floods and maximum flooded areas was largely filtered.

a Occurrence of the historical 100-year river flood and, b annual maximum flooded area over the period 2070–2099, given two experiments (with and without dams), and tow representative concentration pathways (RCP). The box-and-whisker plots include the 95 th and 5 th range (whiskers), median (horizontal lines in each bar), and 1 st and 3 rd quartiles (height of box) of the annual values obtained for all four global circulation models.

The 100-year return extreme discharge expected in the future (2070–2099) was calculated for all combinations of RCPs and experiments (Supplementary Fig.  8 ) along the main river of the 14 catchments. Downstream of dams, the experiment considering dams always produced a lower 100-year discharge than that produced by the experiment not considering dams. For catchments located in regions where annual precipitation and/or snowmelt is forecast to decrease in the future (the Mississippi, Volga, and Euphrates; see Supplementary Figs.  1 and  8a, c, d ), the RCP2.6 simulations produced higher 100-year discharges than those in the RCP6.0. However, simulations employing the RCP6.0 scenario and the experiment not considering dams generally produced the highest 100-year discharges. For catchments containing few dams on the mainstem river, future 100-year return extreme discharges in both experiments (with and without dams) were similar at the river mouth (Supplementary Fig.  8i, k, l, m, n ). However, in other catchments, the 100-year extreme discharges were clearly reduced in the experiments considering dams (Fig.  3 and Supplementary Fig.  7 ), resulting in reduced flood exposure to populations residing downstream of dams. In addition, the reductions in 100-year extreme discharge in the Amazon, Congo, and Mekong rivers were relatively small due to the small cumulative storage capacity of the mainstem dams compared to the discharge volume generated in these basins.

Explicitly considering dams in climate-impact studies of floods significantly offsets the population size exposed to river floods. Downstream of dams at the end of the 21 st century, a 100-year flood was, on average, indicated to occur once every 107 (79–168) years for RCP2.6 and once every 79 years (55–103) in the experiments not considering dams (see Supplementary Fig.  8 ). In RCP6.0, the historical 100-year flood occurred more frequently: once every 59 years (39–110) and 46 years (33–75) for the experiments considering and not considering dams, respectively (see Supplementary Fig.  8 ). In most catchments, dams reduced both the frequency of floods and the extent of flooded areas. Our findings were robust to the selection of indices used to identify floods although the pulsing effect of dams was identified as compromising estimates in some catchments. This problem could be partially mitigated by revising the reservoir operation method used in the present study by accounting for future precipitation variabilities and cascade-dams. Since our large-scale modelling considers daily precipitation, potential dam failure due to increased extreme precipitation events 38 (resulting in downstream flooding) is not fully considered here, nor are the construction and filling phases of a dam’s life cycle. Nevertheless, neglecting the morphological, environmental, and societal impact of dams 39 , our results imply that dams significantly decrease the risk of future global floods in terms of both frequency and intensity, protecting 1.4 (0.7–3.1) and 2.3 (0.8–3.7) million people at the end of the 21 st century, for RCP2.6 and RCP6.0, respectively.

The aging dam landscape faces new temperature, snow, discharge, and floods patterns that increase the risk of hydrological failure 40 , 41 . To maintain historical levels of flood protection in the face of climate change, new dam release operations will be required. In addition, precise and reliable hydro-meteorological forecasts will be invaluable for maximizing flood protection and avoiding untimely and excessive outflows. By focusing solely on the role of dams in reducing global flood exposure under climate change, the results of this study are perceived as over emphasizing the benefits of dams (see Supplementary Note  2 ). However, given the many negative environmental and social impacts of dams 39 , comprehensive assessments that consider both potential benefits and adverse effects are necessary for the sustainable development of water resources. Furthermore, future analyses of global flood risks would benefit from: addressing the disparities and uncertainties associated with global dam and river datasets (e.g. location, characteristics, networks); developing realistic future population projections that account for population behaviour; enhancing historical GCM scenarios by assimilating past observations; and archiving and referencing historical reservoir operations, streamflow, and inundation for robust model validation.

Two hydrological models were used in this study. H08 is an open-source global hydrological model (GHM) that explicitly considers human water abstraction from six major water sources including dams 24 . The reservoir operation scheme in H08 is a generic one; that is, it is not tailored to a specific site. A detailed description can be found in Hanasaki et al. 31 . Outflow from dams is computed in two steps: considering the water currently available in the reservoir, a provisional annual total release is computed, and is then adjusted every month according to changes in storage, inflow, and water demand below the dams. The algorithm distinguishes two classes of dams: irrigation and non-irrigation dams, which influences the computation of monthly water release. It should be noted that, while the storage capacity used in the simulations corresponded to that reported in the Global reservoirs and Dams database (GRanD), the actual storage capacity of dams is expected to be lower due to the allocation of dead and surcharge storages. As a result, the allocated dam storage in the present simulations is likely to have been overestimated. The most recent version of the H08 model, which participated in ISIMIP2b, was employed 24 . Simulations were carried out at a spatial resolution of 0.5° by 0.5°, and a 1-day interval.

CaMa-Flood is a new generation of global river routing model that relies on HydroSHEDS 42 topography to simulate floodplain dynamics and backwater effects by explicitly solving the local inertia equation 33 . The model was reported to outperform other GHMs for reproducing historical discharge 43 . The CaMa-Flood model requires only daily runoff as an input, and by computing the inflow from upstream cells and outflow to downstream, the evolution of water storage can be predicted. In this study, three output variables were used: the total discharge exiting a grid-cell (sum of river discharge and floodplain flow), the flooded area, and the flooded fraction of a grid-cell. To output the latter two variables, CaMa-Flood assesses whether water currently stored in a grid-cell exceeds the total storage of the river section. When this is the case, excess water is then stored in the floodplain, for which topography (dictated by HydroSHEDS) controls the flood stage (water level and flooded area).

To simulate the effects of water regulation due to anthropogenic activities on floodplain dynamics, the H08 and CaMa-Flood models were coupled because, in its current global version (v3.62), the global version of CaMa-Flood cannot simulate dam operations despite being essential for assessing flood risk. Hence, the H08 model is required for accurate forecasts of dam outflow. To ensure compatibility between the models, the river network originally used in CaMa-Flood was employed in both models. The coupling procedure is as follows: simulations with the H08 model are conducted; the daily runoff predicted by H08 is used as a forcing input in CaMa-Flood; in grid-cells containing major dam(s), 44 the river discharge produced by H08 (following the reservoir operating rule) is imposed onto the CaMa-Flood model (Supplementary Fig.  3a ); the difference in daily discharge between the two models due to water regulation is added to the hypothetical storage associated with every dam but without interacting with the river or floodplain to close the water balance.

For grid cells that are neither downstream nor upstream of dams (light blue locations in Supplementary Fig.  3 ), experiments considering and not considering dams produced the same discharge outputs. In contrast, for grid cells located below and above dams, the daily discharge simulated by the experiments considering dams can change compared to the experiments not considering dams due to water regulation (below dams) and the impossibility of the backwater effect and its propagation (above dams).

The four general circulation models (GCMs; GFDL-ESM2M, HadGEM2-ES, IPSL-CM5A-LR, and MIROC5) implemented in the ISIMIP2b protocol participated in the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC). The forcing data consisted of precipitation, temperature, solar radiation (short and long wave downward), wind speed, specific humidity, and surface pressure which were bias corrected 45 and downscaled to a 0.5° by 0.5°-grid resolution. Here we used three radiative forcing scenarios: historical climate (1861–2005), and two future scenarios consisting of a low greenhouse gas concentration emission scenario (RCP2.6; 2006–2099) and a medium–high greenhouse gas concentration (RCP6.0; 2006–2099). Note that the historical climate scenario does not attempt to reproduce the exact day-to-day historical climate but rather gives a consistent evolution of the climate under a given climatic forcing.

Dam specifications (location, storage capacity, and construction year) are provided in GRanD 44 , 46 . The dams were georeferenced to the river network employed in CaMa-Flood, iteratively adjusting dam locations when necessary until the catchment areas of each dam reported in GRanD corresponded to ± 10% of the catchment area in CaMa-Flood 47 .

Experiments

For the future scenarios (RCP2.6 and RCP6.0), two experiments were considered. In the first experiment, dams were not implemented, therefore this simulation is analogous to the simulations conducted in previous studies 3 , 14 . In contrast, in the second experiment, the effect of major global dams on water regulation, hence floodplain dynamics, were considered. Due to water regulation, the future return period (in years) associated with the historical 100-year extreme discharge might change compared to that obtained for the experiment not considering dams (Supplementary Fig.  8 ). These potential differences were used to quantify the effect of dams on the potential reduction in the future return period of the historical 100-year flood.

The H08 model has been extensively validated in catchments located in India, the US, China, Europe, and South America for predicting river discharge, total water storage anomalies, groundwater, and water transfer 24 . Across these major catchments, the average Nash–Sutcliffe efficiency ( NSE ) obtained when comparing daily observed and simulated discharge was positive. Benchmarked against GHMs, H08 was reported to perform relatively well for reproducing historical daily discharges 25 . More relevant to the context of this study, the same study 25 highlighted that the H08 model was among the top four GHMs best able to reproduce the magnitude of extreme discharge and the maximum flows associated with different return periods.

The ability of the CaMa-Flood model to reproduce floodplain inundation was reported in the Amazon basin, where it performed well 33 . In addition, the discharges produced by CaMa-Flood have been evaluated against gauge observations in 30 major river basins 33 . CaMa-Flood has also been benchmarked against nine GHMs, including the H08 model, at 1701 gauge locations 43 . Generally, discharge simulations using CaMa-Flood produce lower and later peak discharges compared to those predicted by other GHMs, resulting in more accurate reproduction of observations 43 .

The quality of discharge data produced by nine GHMs, including the H08 model used in this study, was evaluated and compared against calibrated regional hydrological models in 11 large river basins 48 . While regional models generally outperformed GHMs in most regions, GHMs reproduced the intra-annual variability of water discharge reasonably well. Extreme discharges are strongly related to floods, 5 and the inclusion of human activity in hydrological simulations, such as in H08 has been reported to greatly improve the reproduction of hydrological extremes 49 . The predicted return period for the historical 100-year discharge obtained in the experiment not considering dams was compared to the literature. Global estimates of populations exposed to river floods were also compared to those reported in the literature (Supplementary Table  2 ). We evaluated how the coupled model reproduced river discharges before and after the implementation of dams at key locations. We separated our observation dataset into two parts: pre- and post-dam construction. We then compared our dam and no-dam simulations to the relevant observations. Supplementary Table  3 lists the dam locations of the dams and their key characteristics.

Definition of flood event and extreme discharge

We compared the frequency of historical (1975–2004) and predicted future (RCP2.6 and RCP6.0; 2070–2099) flood events using given two experiments: an experiment in which no dams were considered (analogous to previous studies 3 , 4 , 5 ), and an experiment considering global dams (Supplementary Fig.  2 ) 50 . Flood events were defined as the historical 100-year return extreme discharge, that is, the extreme discharge with a probability of exceeding 1/100 in any given year.

Two annual-extreme discharge indices were used in this analysis to assess the robustness of our findings expressed by the spread (or consistency) of results from multiple GCMs and extreme indices. We primarily focused on the maximum annual daily discharge ( P max ) since it is the preferred index used in the literature 3 , 4 , 5 , 14 . The alternative indicator is the annual 5 th percentile ( P 05 ) of daily discharge.

Before fitting the Gumbel distribution to estimate the 100-year river discharge, we initially compared the two series of extreme discharges in the dam and no-dam experiments. Run-of-the-river dams tend to alter the natural flow regimes only negligibly. For such locations, the fitted Gumbel distribution should be identical in both experiments. In contrast, in rivers heavily regulated by dams, it is possible that the extreme discharge series obtained for the experiment considering dams included many identical or tied values. We initially computed the absolute difference between the annual discharge extremes obtained by the simulation not considering dams minus the simulation considering dams and compared that difference to a given threshold (150 m 3  s −1 , or an annual difference of 5 m 3  s −1 between the extreme discharge generated for the experiments with and without dams). When the threshold was exceeded, the extreme discharge series were considered dissimilar and therefore treated separately. In contrast, when the threshold was not exceeded, the two extreme discharge series were considered similar and all data were pooled before moving to the fitting phase. We assessed the sensitivity of our results to alternative thresholds, with those results reported in Supplementary Table  1 .

Fitting of Gumbel distribution

The extreme discharges were first ranked in ascending order and fitted to a Gumbel distribution using the L-moment method 51 . As a result of the comparison protocol, the number of data to fit was either 60 (experiments with and without dams produced similar extreme discharges and were pooled) or 30 (experiments with and without dams produced different extreme discharges). The fitting process is identical to that described in detail in the Supplementary Note  2 of Hirabayashi et al. 3 .

Assessment of goodness of fit

The goodness of fit of the annual extreme discharge to the Gumbel distribution was assessed using the probability plot correlation coefficient test (PPCC) 52 . While other methods can be used to assess the goodness of fit of the Gumbel distribution, the PPCC has been reported to outperform most of them in terms of rejection performance 53 . The PPCCs were computed for all historical simulations and are reported in Supplementary Fig.  9 . A PPCC score close to 1 indicates that the distribution of the extreme series is well fitted by the Gumbel distribution. For a sample size of 30, the critical PPCC score at the 95 th level of significance was reported 52 to be approximately 0.96.

A bootstrap methodology was used to assess the influence of the 30-year samples on the fitted Gumbel distribution 54 . We generated 1000 bootstrap estimates for every GCM and all experiments. We did not explore all combinations of bootstrap estimates and GCMs due to the high computational cost (1012 estimates for a given year and a single experiment). Instead, we ranked the estimates in descending order before taking the average across GCMs (1000 estimates for a given year and a single experiment). While simple, this method has the advantage of reporting the broadest confidence intervals since the lowest and highest estimates among GCMs are averaged.

In the reported global maps, we masked grid-cells belonging to the Köppen–Geiger regions BWk (hot desert climates), BWh (cold desert climates), and EF (ice cap climates) which discharge corresponding to the historical 30-year return period was less than 5 m 3  s −1 (Supplementary Fig.  4 ). In such grid cells, flooding is not a problem due to the low volume of water discharge. As a result, the goodness of fit of the Gumbel distributions was generally low (as indicated by a low PPCC score in Supplementary Fig.  9 ).

Population exposure

The population dataset, created by the Socioeconomic Data and Applications Center (SEDAC), consists of the Gridded Population of the World (GPW, v4.11) for the year 2010 55 . The population was fixed at 2010 to assess only the effect of climate change on population exposure to floods. To increase the accuracy of our exposure assessment, the original 0.5° resolution flooding depths were downscaled to a resolution of 0.005°. The file containing flooding depth resulting from historical 100-year floods was constructed annually following a two-step procedure. First, we determined the 0.5° grid cells experiencing a 100-year flood as indicated by the annual discharge extreme exceeding the 100-year historical discharge extreme. Second, for such grid cells, we extracted the maximum annual flooding depth, while the flooding depth of other grid cells was set to zero. The files were then downscaled to a 0.005° resolution using routines implemented in CaMa-Flood 33 (see model description). Population exposure to river floods was assessed by overlaying the population and flooding-depth datasets. When flooding water was present in a 0.005° cell, the population within that cell was considered exposed to flooding.

We accounted for population growth in a separate analysis using population projections from 2006 to 2099 based on shared socioeconomic pathways (SSPs) 1 to 5 provided in the ISIMIP2b framework. The time-varying population datasets were first downscaled to a 0.005° resolution. Population exposure to flooding was then determined using the procedure described above.

Catchment selection

Catchments were selected by ensuring that downstream areas were wide, densely populated, and contained major dams. More specifically, the following criteria were used: at least 10 grid cells below dams, a population of at least 5 million residing on the entire main river channel, and the capacity of dams divided by their annual inflow averaged over the number of dams present on the main river channel had to be higher than 0.1. While 15 catchments initially fulfilled these criteria, the Nile catchment was removed from our analysis since a significant portion of its upper section falls within the Köppen–Geiger region BWh (Supplementary Fig.  4 ), which was (partially) screened out of the analysis. The locations of the remaining 14 catchments are given in Supplementary Fig.  6 .

Catchment flood analysis

The analysis consisted of two parts: identifying in which grid cells a flood occurred and extracting the corresponding flooded area for those cells. First, daily discharge, collected annually for the 2070–2099 period, in all grid-cells composing the catchments was converted to annual extreme discharges (considering two indices) and compared to the 100-year return extreme discharge. When the annual extreme discharge was higher than that of the historical 100-year return discharge, a flood was considered to occur in that year. Second, for grid cells where a flood occurred, the maximum flooded area of the grid cell was collected. Finally, we presented the aggregated sum of flood occurrence and flooded area of grid-cells located downstream of dams.

Data availability

The H08 model is open source and its source code is available online ( http://h08.nies.go.jp/h08/index.html ). The source code of the CaMa-Flood model can be requested from D.Y. All input data are available through the ISIMIP2b protocol which is freely accessible ( https://www.isimip.org/ ). Detail explanations regarding the coupling procedure, including the new variables introduced in the model and the source file to edit, are available online ( https://zenodo.org/record/3701166 ).

Code availability

Computer code used for analysis and graphic preparation is available online with explanation ( https://zenodo.org/record/3701166 ).

Prein, A. F. et al. The future intensification of hourly precipitation extremes. Nat. Clim. Change 7 , 48 (2016).

Article   ADS   Google Scholar  

Milly, P. C. D., Wetherald, R. T., Dunne, K. A. & Delworth, T. L. Increasing risk of great floods in a changing climate. Nature 415 , 514–517 (2002).

Article   CAS   PubMed   Google Scholar  

Hirabayashi, Y. et al. Global flood risk under climate change. Nat. Clim. Change 3 , 816 (2013).

Jongman, B., Ward, P. J. & Aerts, J. C. J. H. Global exposure to river and coastal flooding: long term trends and changes. Glob. Environ. Change 22 , 823–835 (2012).

Article   Google Scholar  

Ward, P. J. et al. Assessing flood risk at the global scale: model setup, results, and sensitivity. Environ. Res. Lett. 8 , 044019 (2013).

Grill, G. et al. Mapping the world’s free-flowing rivers. Nature 569 , 215–221 (2019).

Article   ADS   CAS   PubMed   Google Scholar  

Chao, B. F., Wu, Y. H. & Li, Y. S. Impact of artificial reservoir water impoundment on global sea level. Science 320 , 212 (2008).

Oki, T. & Kanae, S. Global hydrological cycles and world water resources. Science 313 , 1068 (2006).

Abbott, B. W. et al. Human domination of the global water cycle absent from depictions and perceptions. Nat. Geosci. 12 , 533–540 (2019).

Article   ADS   CAS   Google Scholar  

Dynesius, M. & Nilsson, C. Fragmentation and flow regulation of river systems in the northern third of the world. Science 266 , 753 (1994).

Poff, N. L. & Schmidt, J. C. How dams can go with the flow. Science 353 , 1099 (2016).

Voeroesmarty, C. J. et al. The storage and aging of continental runoff in large reservoir systems of the world. Ambio 26 , 210–219 (1997).

Google Scholar  

Zarfl, C., Lumsdon, A. E., Berlekamp, J., Tydecks, L. & Tockner, K. A global boom in hydropower dam construction. Aquat. Sci. 77 , 161–170 (2015).

Dottori, F. et al. Increased human and economic losses from river flooding with anthropogenic warming. Nat. Clim. Change 8 , 781–786 (2018).

Sampson, C. C. et al. A high-resolution global flood hazard model. Water Resour. Res. 51 , 7358–7381 (2015).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Ward, P. J. et al. Usefulness and limitations of global flood risk models. Nat. Clim. Change 5 , 712 (2015).

Ward, P. J. et al. Strong influence of El Niño Southern Oscillation on flood risk around the world. Proc. Natl Acad. Sci. 111 , 15659–15664 (2014).

Winsemius, H. C. et al. Global drivers of future river flood risk. Nat. Clim. Change 6 , 381 (2015).

Jongman, B. et al. Declining vulnerability to river floods and the global benefits of adaptation. Proc. Natl Acad. Sci. 112 , E2271–E2280 (2015).

Pappenberger, F., Dutra, E., Wetterhall, F. & Cloke, H. L. Deriving global flood hazard maps of fluvial floods through a physical model cascade. Hydrol. Earth Syst. Sci. 16 , 4143–4156 (2012).

Lim, W. H. et al. Long-term changes in global socioeconomic benefits of flood defenses and residual risk based on CMIP5 climate models. Earth’s Future 6 , 938–954 (2018).

Zhao, G., Bates, P. & Neal, J. The impact of dams on design floods in the conterminous US. Water Resour. Res. 56 , e2019WR025380 (2020).

Mei, X., Van Gelder, P. H. A. J. M., Dai, Z. & Tang, Z. Impact of dams on flood occurrence of selected rivers in the United States. Front. Earth Sci. 11 , 268–282 (2017).

Hanasaki, N., Yoshikawa, S., Pokhrel, Y. & Kanae, S. A global hydrological simulation to specify the sources of water used by humans. Hydrol. Earth Syst. Sci. 22 , 789–817 (2018).

Zaherpour, J. et al. Worldwide evaluation of mean and extreme runoff from six global-scale hydrological models that account for human impacts. Environ. Res. Lett. 13 , 065015 (2018).

Mateo, C. M. et al. Assessing the impacts of reservoir operation to floodplain inundation by combining hydrological, reservoir management, and hydrodynamic models. Water Resour. Res. 50 , 7245–7266 (2014).

Hanasaki, N., Fujiwara, M., Maji, A. & Seto, S. On the applicability of the H08 global water resources model to the Kyusyu Island. J. Jpn. Soc. Civ. Eng., Ser. B1 74 , I_109–I_114 (2018).

Masood, M., Yeh, P. J.-F., Hanasaki, N. & Takeuchi, K. Model study of the impacts of future climate change on the hydrology of Ganges–Brahmaputra–Meghna basin. Hydrol. Earth Syst. Sci. 19 , 747–770 (2015).

Rougé, C. et al. Coordination and control: limits in standard representations of multi-reservoir operations in hydrological modeling. Hydrol. Earth Syst. Sci. Discuss. 2019 , 1–37 (2019).

Shin, S., Pokhrel, Y. & Miguez-Macho, G. High-resolution modeling of reservoir release and storage dynamics at the continental scale. Water Resour. Res. 55 , 787–810 (2019).

Hanasaki, N., Kanae, S. & Oki, T. A reservoir operation scheme for global river routing models. J. Hydrol. 327 , 22–41 (2006).

Yamazaki, D. et al. Analysis of the water level dynamics simulated by a global river model: a case study in the Amazon river. Water Resour. Res. 48 , W09508 (2012).

Yamazaki, D., Kanae, S., Kim, H. & Oki, T. A physically based description of floodplain inundation dynamics in a global river routing model. Water Resour. Res. 47 , W04501 (2011).

Yamazaki, D., Sato, T., Kanae, S., Hirabayashi, Y. & Bates, P. D. Regional flood dynamics in a bifurcating mega delta simulated in a global river model. Geophys. Res. Lett. 41 , 3127–3135 (2014).

Yamazaki, D. et al. Development of the global width database for large rivers. Water Resour. Res. 50 , 3467–3480 (2014).

Assani, A. A., Stichelbout, É., Roy, A. G. & Petit, F. Comparison of impacts of dams on the annual maximum flow characteristics in three regulated hydrologic regimes in Québec (Canada). Hydrological Process. 20 , 3485–3501 (2006).

Masaki, Y., Hanasaki, N., Takahashi, K. & Hijioka, Y. Consequences of implementing a reservoir operation algorithm in a global hydrological model under multiple meteorological forcing. Hydrological Sci. J. 63 , 1047–1061 (2018).

Hollins, X. L., Eisenberg, A. D. & Seager, P. T. Risk and resilience at the Oroville dam. Infrastructures 3 , 49–65 (2018).

Best, J. Anthropogenic stresses on the world’s big rivers. Nat. Geosci. 12 , 7–21 (2019).

Mallakpour, I., AghaKouchak, A. & Sadegh, M. Climate-induced changes in the risk of hydrological failure of major dams in California. Geophys. Res. Lett. 46 , 2130–2139 (2019).

Ehsani, N., Vörösmarty, C. J., Fekete, B. M. & Stakhiv, E. Z. Reservoir operations under climate change: storage capacity options to mitigate risk. J. Hydrol. 555 , 435–446 (2017).

Lehner, B., Verdin, K. & Jarvis, A. New global hydrography derived from spaceborne elevation data. Eos, Trans., Am. Geophys. Union 89 , 93–94 (2008).

Zhao, F. et al. The critical role of the routing scheme in simulating peak river discharge in global hydrological models. Environ. Res. Lett. 12 , 075003 (2017).

Lehner, B. et al. High-resolution mapping of the world’s reservoirs and dams for sustainable river-flow management. Front. Ecol. Environ. 9 , 494–502 (2011).

Lange, S. Trend-preserving bias adjustment and statistical downscaling with ISIMIP3BASD (v1.0). Geosci. Model Dev. 12 , 3055–3070 (2019).

Lehner, B. et al. Global Reservoir and Dam Database, Version 1 (GRanDv1): Dams, Revision 01. (2011).

Masaki, Y. et al. Intercomparison of global river discharge simulations focusing on dam operation—multiple models analysis in two case-study river basins, Missouri–Mississippi and Green–Colorado. Environ. Res. Lett. 12 , 055002 (2017).

Hattermann, F. F. et al. Cross‐scale intercomparison of climate change impacts simulated by regional and global hydrological models in eleven large river basins. Clim. Change 141 , 561–576 (2017).

Veldkamp, T. I. E. et al. Human impact parameterizations in global hydrological models improve estimates of monthly discharges and hydrological extremes: a multi-model validation study. Environ. Res. Lett. 13 , 055008 (2018).

Lehner, B., Döll, P., Alcamo, J., Henrichs, T. & Kaspar, F. Estimating the impact of global change on flood and drought risks in Europe: a continental, integrated analysis. Clim. Change 75 , 273–299 (2006).

Hosking, J. R. M. L-Moments: analysis and estimation of distributions using linear combinations of order statistics. J. R. Stat. Soc. Ser. B 52 , 105–124 (1990).

MathSciNet   MATH   Google Scholar  

Vogel, R. M. The probability plot correlation coefficient test for the normal, lognormal, and gumbel distributional hypotheses. Water Resour. Res. 22 , 587–590 (1986).

Heo, J.-H., Kho, Y. W., Shin, H., Kim, S. & Kim, T. Regression equations of probability plot correlation coefficient test statistics from several probability distributions. J. Hydrol. 355 , 1–15 (2008).

James, G., Witten, D., Hastie, T. & Tibshirani, R. An Introduction to Statistical Learning: with Applications in R . (Springer Publishing Company, Incorporated, 2014).

Center for International Earth Science Information Network - CIESIN - Columbia University. Gridded Population of the World, Version 4 (GPWv4): Basic Demographic Characteristics, Revision 11. (2018).

Download references

Acknowledgements

This work was mainly supported by Environment Research and Technology Development Fund (2RF-1802) of the Environmental Restoration and Conservation Agency (grant number JPMEERF20182R02), Japan. It was partially supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI grant number 16H06291. Y.P. acknowledges the support from the National Science Foundation (CAREER Award, grant number 1752729).

Author information

Authors and affiliations.

National Institute for Environmental Studies (NIES), Tsukuba, Japan

Julien Boulange & Naota Hanasaki

Institute of Industrial Science, The University of Tokyo, Tokyo, Japan

Dai Yamazaki

Department of Civil and Environmental Engineering, Michigan State University, East Lansing, MI, USA

Yadu Pokhrel

You can also search for this author in PubMed   Google Scholar

Contributions

J.B. carried out the simulation and analysis. J.B., N.H., D.Y., and Y.P. commented on and edited the manuscript.

Corresponding author

Correspondence to Julien Boulange .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Brenden Jongman and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Boulange, J., Hanasaki, N., Yamazaki, D. et al. Role of dams in reducing global flood exposure under climate change. Nat Commun 12 , 417 (2021). https://doi.org/10.1038/s41467-020-20704-0

Download citation

Received : 23 March 2020

Accepted : 04 December 2020

Published : 18 January 2021

DOI : https://doi.org/10.1038/s41467-020-20704-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Case Study:

Teton dam (idaho, 1976), description & background, lessons learned, other resources, quick facts.

Location : Idaho, USA

Year Constructed : 1975

Drainage Area : 853 sq. mi.

Type : Earthfill

Height : 305 ft.

Primary Purpose : Flood Control, Hydropower

Date of Incident : June 5, 1976

Evacuation : Yes

Fatalities : 11

Property Damage : $400 Million

Teton Dam was located in southeastern Idaho about 15 miles from Rexburg in the valley of the Teton River. The dam and its reservoir were the principal elements of the Teton Basin Project designed by the Bureau of Reclamation to control flooding as well as provide a source of hydropower, irrigation, and drinking water. Construction on the Teton Dam, reservoir, and powerhouse began in 1972 and by November 1975 the zoned earthfill embankment was essentially complete with a structural height of 305 feet and a crest length of 3,100 feet. Less than one year later, the dam experienced catastrophic failure on June 5, 1976 during its first filling. Failure of the Teton Dam and subsequent draining of the reservoir caused the deaths of 11 people and approximately $400 million in damages.

On June 3, 1976, two small seeps were observed at the downstream toe of the dam  which released clear seepage and measured less than ¼ cfs. When the rest of the dam was inspected, however, no further evidence of seepage was noted. Only two days later, in the early morning hours of June 5, 1976, 20 to 30 cfs of clear seepage was observed exiting rock joints near the right abutment of Teton Dam. Shortly thereafter, the seepage became muddy and a sinkhole developed on the downstream slope of the  embankment dam . Flow developed from the sinkhole and dozers were sent in an unsuccessful attempt to fill the resulting eroded area with  riprap . Flow through the open void that developed by scour (a piping mechanism) continued to increase and the embankment eventually breached at the dam crest around 11:55 a.m. (only several hours after the first sign of muddy seepage). The resulting rapid release of the entire contents of the reservoir flooded five counties, inundated over 300 square miles, and traveled a distance of 155 miles downstream.

Downstream view of Teton Dam under construction circa 1975.

Investigations of the Teton Dam failure attributed the catastrophe to a series of design and construction related deficiencies. These inadequacies related primarily to the foundation treatment at the dam and adherence to the overall construction schedule. It was determined the most probable physical failure mode was cracking of the dam’s impervious core due to internal erosion initiated by hydraulic fracturing of the key trench fill material.

Teton Dam was located in an area with highly permeable foundation materials. During investigation of the failure, it was discovered that proper treatment of such foundation material was not implemented. It appeared that the dam’s designers did not take the site specific geological conditions into account when developing the structure. This oversight was exacerbated by the lack of communication between the design and construction engineers about the proper preparation of the dam foundation. As a result, not only was the foundation treated inadequately, but that treatment was also inconsistent. Although a key trench was constructed in an attempt to prevent seepage through the pervious embankment, the slush  grouting at the key trench was insufficient. In addition, slush grouting was inexplicably stopped once El. 5200 was reached. The post-failure review panel also determined that the rock surface at the right abutment was not adequately sealed under the dam’s impervious core. Working together or alone, either of these foundation design deficiencies would have provided optimal conditions for internal erosion of the core.

Although a three- gated spillway existed at crest elevation near the right abutment at the time of failure, the dam’s powerhouse, auxiliary spillway , and  outlet works remained unfinished. Construction delays caused deviation from the original schedule but the first filling of the reservoir was not postponed. Therefore, when the unexpectedly rapid filling of the reservoir occurred, there was no operable low level outlet works for dewatering the reservoir. Although investigations of the incident proved that failure of the dam due to improperly treated foundation materials was most likely inevitable, the lack of a proper reservoir low level outlet works contributed to the severity of the failure and reservoir release. Had the first filling been managed using a low level outlet works, the volume of water suddenly released from behind the Teton dam could have been reduced. Twenty-four hour monitoring of the dam and reservoir during its first filling may have also contributed to the lessening of downstream consequences.

Other factors related to the failure of the dam included a lack of external review of the project plans and specifications that may have discovered some of the design deficiencies. Furthermore, the dam was not designed with any type of secondary defenses against seepage and relied fully on the grout curtain and key trench.

References:

(1) Barnes, M. (1992).  Famous Failures:  Revisiting Major Dam Catastrophes.  ASDSO Annual Conference .  Baltimore:  Association of State Dam Safety Officials.

(2) Graham, W. (2009).  Avoiding Disaster:  Assuring Warning Compliance.  ASDSO Annual Conference .  Hollywood:  Association of State Dam Safety Officials.

(3) Graham, W. (2008).  The Teton Dam Failure:  An Effective Warning and Evacuation.  ASDSO Annual Conference .  Indian Wells:  Association of State Dam Safety Officials .

(4) Independent Panel to Review Cause of Teton Dam Failure. (1976).  Report to U.S. Department of Interior and State of Idaho on Failure of Teton Dam .  Washington D.C.:  U.S. Government Printing Office.

(5) Snorteland, N. (2009).  Fontenelle Dam, Ririe Dam, and Teton Dam:  An Examination of the Influence of Organizational Culture on Decision-Making.  ASDSO Annual Conference .  Hollywood:  Association of State Dam Safety Officials. 

All dams need an operable means of drawing down the reservoir.

Dam failure sites offer an important opportunity for education and memorialization.

Dam incidents and failures can fundamentally be attributed to human factors.

Early Warning Systems can provide real-time information on the health of a dam, conditions during incidents, and advanced warning to evacuate ahead of dam failure flooding.

• High and significant hazard embankment dams should have internal filter and seepage collection systems.

Stability of the dam foundation and other geologic features must be considered during dam design.

The first filling of a reservoir should be planned, controlled, and monitored.

Timely warning and rapid public response are critical to saving lives during a dam emergency.

Additional lessons learned (not yet developed).

• External independent peer review of designs and decisions is an effective means of providing quality assurance and reducing the risk associated with design oversights and deficiencies.

Report to U.S. Department of Interior and State of Idaho on Failure of Teton Dam

Author: Independent Panel to Review Cause of Teton Failure

Investigation Report

Major Historical Dam Failures with Modes of Failure

Author: K. Mills

Presentation at Oregon Dam Safety Conference

Famous Failures: Revisiting Major Dam Catastrophes

Author: M. Barnes

Technical paper published by Association of State Dam Safety Officials

Fontenelle Dam, Ririe Dam, and Teton Dam: An Examination of the Influence of Organizational Culture on Decision-Making

Author: N. Snorteland

Technical Paper published by Association of State Dam Safety Officials

Avoiding Disaster: Assuring Warning Compliance

Author: W. Graham

The Teton Dam Failure: An Effective Warning and Evacuation

Additional resources not available for download.

• VandenBerge, D., Duncan, J., & Brandon, T. (2011). Lessons Learned from Dam Failures . Virginia Polytechnic Institute and State University.
• WordPress.org
• Documentation
• Learn WordPress

An assessment of ecological impacts of dam construction on vegetation. Case study of Nandoni Dam

• August 2016
• Thesis for: MSc Environmental science
• Advisor: R Baxter

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

Supplementary resource (1)

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

Jump to navigation

• What is the NACLA Report?
• Web Exclusives
• Recent Articles
• News and Analysis
• Border Wars
• Rebel Currents
• NACLA-Global Voices
• The Other Side of Paradise
• Extractives in Latin America
• Mexico, Bewildered and Contested
• Red Hot Burning Peace
• Cuadernos Colombianos
• Manufacturing Contempt
• Photo Essays
• Brazil on Fire
• NACLA Radio
• Upcoming events
• Past Events
• Staff & Board
• Jobs/Internships
• Write for NACLA

Search form

Brazil’s native peoples and the belo monte dam: a case study.

Fueled by ample natural resources, Brazil’s economy has soared in the last seven years under the helm of President Luiz Inácio Lula da Silva. As a part of its GDP growth strategy, the Lula government has prioritized large-scale infrastructure development with its Accelerated Growth Program (PAC)—a public-private partnership to fund the expansion of Brazil’s transport, energy, sanitation, and housing infrastructure, launched in 2007. The Brazilian Finance Ministry attributed Brazil’s recently increasing GDP growth (from 4% in 2006 to 6.1% in 2007, 5.1% in 2008) to the program’s success. 1

Belo Monte, a proposed hydroelectric dam in the Xingu River Basin of the Amazonian rainforest, is the PAC’s flagship project. Belo Monte is the world’s largest dam complex in development, and if completed it will be the world’s third-largest dam, after China’s Three Gorges Dam and the joint Brazilian-Paraguayan dam Itaipu. 2 The Belo Monte project will divert the flow of the Xingu River, a massive, 1,700-mile tributary of the Amazon that stretches from the savannas of the western state of Mato Grosso to the northern jungles of Pará. Operating at full capacity, the dam will generate up to 11,233 megawatts, with most of the electricity going to local mining operations before being routed to the metropolitan areas of Rio de Janeiro and São Paulo, about 1,800 miles away.

The Brazilian government has mounted a powerful pro¬paganda campaign to convince its citizens and the world that this $30 billion megadam, funded almost entirely by Brazil’s National Development Bank, is a sustainable way to fuel continued economic growth and human development. Moreover, the government argues that the dam is a model of “green,” renewable energy, based on the premise that it will flood a relatively small area of surrounding forest and produce large amounts of electricity with fewer emissions than a generator powered by fossil fuels. Excluded from the public discussion, however, has been the incalculable social and environmental destruction that the dam threatens to wreak on the Xingu River Basin, a global center of biodiversity that supports an extensive network of tributaries, primary forests, and some 25,000 indigenous people from 18 ethnic groups.

The dam will divert more than 80% of the Xingu River’s flow through two massive canals, flooding more than 193 square miles of forest and part of the city of Altamira, while a 62-mile stretch of the river known as the Big Bend will be left in permanent drought. This is expected to cause a significant decline in the water table, leading to substantial losses of aquatic and terrestrial fauna. The dam threatens to devastate the surrounding rainforest ecosystem and will displace between 20,000 and 40,000 people, both rural and urban, destroying their livelihoods with little or no compensation.

In addition to the hundreds of riverine communities, about 800 people from the Juruna, Xikrín, Arara, Xipaia, Kuruaya, Kayapó, and other indigenous ethnicities in the surrounding region will no longer be able to depend on the river for survival. Receding waters would make it impossible for local communities to travel by boat to sell their produce or buy staples. Upstream communities, including the Kayapó, would lose migratory fish species essential to their diet. And for the peoples who call the river basin home—from the Kayapó of the upper reaches of the Xingu’s tributaries to the Arara, who live alongside its waterways—the Big Bend is the cradle of civilization. The word Xingu means “house of God” to indigenous groups, and its destruction will represent nothing less than a cosmological catastrophe to them.

Despite these concerns, which have been publicly aired by the people who live near the dam project, by experts in numerous fields, and even by some dissident government administrators, the official line has not budged. At a press conference in February, after the Belo Monte project was granted an environmental license, Minister of Environment Carlos Minc flat out rejected the criticisms. “Not a single Indian will be displaced,” Minc said. He conceded only that the peoples of the Xingu will be “indirectly affected.” 3

Even as the Lula government has pursued Belo Monte and other megaprojects, it has postured itself as concerned with sustainability, conservation, and indigenous rights. At the beginning of his presidency, Lula appointed ex–rubber tapper Marina Silva as head of the Ministry of the Environment, where she organized local River Basin Committees to help manage water resources. And in 2007 Brazil voted for the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP), which guarantees the right to self-determination, including free, prior, and informed consent, for projects that affect their communities. But despite these and other steps, the Lula government has on balance perpetuated the Brazilian state’s history of violating indigenous rights and ignoring environmental conservation in the name of progress.

The Lula government’s efforts to portray itself as sensi¬tive to indigenous rights draw on a recent history of re¬forms and legal commitments beginning with the creation of the National Indian Foundation (FUNAI) in 1967. An entire governmental department dedicated to protecting the interests of indigenous people, FUNAI was founded after a scandalous 5,000-page report commissioned by the Ministry of the Interior revealed systemic persecution of indigenous peoples in Brazil, including torture, slavery, sexual abuse, land theft, and even the complete extermination of 80 tribes. The military government created FUNAI in part as a public relations effort to construct a humanitarian, “racially democratic” image. 4 It was no accident, however, that the newly birthed FUNAI was an agency of the Ministry of the Interior, whose principal mission was to exploit Brazil’s natural resources. FUNAI immediately adopted a policy of protection by assimilation, encouraging communities to give up traditional ways of life in order to join mainstream culture—in short, freeing up native lands by eliminating cultural diversity.

After the return to democracy, Brazil’s 1988 Constitution included several hard-won provisions guaranteeing indigenous rights. The most important of these was Article 231, which recognized both the cultural and territorial rights of indigenous peoples based on their heritage, establishing their right to permanently live in traditional territories, and guaranteeing the exclusive use of natural resources necessary for securing their cultural integrity and physical well-being. In 1989, Brazil ratified Convention 169 of the International Labor Organization, which entitled indigenous peoples to participate in the decision-making process of infrastructural development in their territories and codified their right to be fully consulted before exploration or exploitation activities begin.

Although the 1988 constitutional provisions were an important gain for indigenous cultural independence and property rights, the enforcement of these rights remains tenuous and often contingent on external political leverage. Because of this, Belo Monte’s development process and that of other infrastructure projects have been alarmingly deficient in fulfilling the state’s legal obligations to indigenous peoples. Leaders from the Xingu River Basin have made it clear that their right to consultation on Belo Monte as spelled out in the Constitution, UNRIP, and ILO 169 has not been honored.

José Carlos Arara of the Arara people on the Xingu’s Big Bend, for example, has denounced the government’s claims that he and other leaders took part in an official meeting with the government regarding Belo Monte, as mandated by the licensing process. He even has video footage of government officials stating that their 2009 meeting with local leadership was an unofficial consultation, clearly promising that an official audience would take place. In total, only four public hearings were held on Belo Monte, offering little or no access to the remote populations who stand to be most affected by the project. In addition, security forces impeded the entrance of civil society representatives to the hearings, and the few public queries that were voiced were dismissed, ridiculed, and evasively answered.

Belo Monte has violated almost all the articles of UNDRIP, especially those that detail rights to full consultation and participation (including veto power) in development projects that are on or affect indigenous territories, or that affect indigenous cultural practices or survival. Brazil has furthermore violated Article 231.3, Chapter VIII, of its own Constitution, which guarantees indigenous peoples’ right to challenge the exploitation of water resources on their lands. And despite laws and policies promising environmental protection and community participation, Brazil’s official environmental impact assessment (EIA), which is required for the Belo Monte project to receive a construction license, has received harsh criticism from national and international experts. The EIA barely covers even the minimum amount of information required by Brazilian legislation. 5

A specialist panel of 40 independent international experts analyzed the 2009 EIA in response to a request from social justice and environmental watchdog groups in the Altamira region. The panel determined that processes at public hearings were forced and accelerated, while the little information made available to the public was both misleading and incomplete. The EIA itself provided insufficient studies and analysis of the dam’s impact on sedimentation and the water table, did not include the likely effects upon aquatic mammals or the probability of deforestation in the greater region, and blatantly omitted any analysis of the cultural, social, or economic impacts on communities downstream. The panel also concluded that the EIA also contained numerous quantitative inaccuracies and methodological inconsistencies, including overestimations of energy generation, underestimations of the size of the affected rural population, and severe negligence in the overall evaluation of health and environmental risks, and water security. The panel also determined that at least two indigenous areas should be added in the list of directly affected areas because the dam will cut off river flow in the Paquiçamba and the Arara do Maia reserves.

Several Brazilian administrators have also raised concerns about the Belo Monte project. Two senior officials at the government’s environmental agency IBAMA, Leozildo Tabajara da Silva Benjamin and Sebastião Custódio Pires, resigned in 2009, citing high-level political pressure to approve the project. In April, Judge Antonio Carlos de Almeida Campelo suspended the dam’s preliminary license, writing in his decision that “it remains proven, unequivocally, that Belo Monte’s plant will exploit the hydroelectric potential of areas occupied by indigenous people who would be directly affected by the construction and development of the project.” Campelo filed three injunctions against Belo Monte’s license, but they were quickly dismissed by a higher court. In June, Campelo was removed from ruling on environmental cases altogether in an administrative maneuver. 6

“The Lula government is clearly pressuring the courts to approve Belo Monte against the rights and interests of Indigenous people and the local populations of the Xingu, yet it is our lives at stake,” said Sheyla Yakarepi Juruna, a leader of the Juruna people. “Even so, the people affected by this dam are united and determined to stop the project, we will not give up this fight.” 7 Belo Monte will be funded almost entirely by Brazil’s National Development Bank (BNDES), which is linked to Brazil’s Ministry of Development, Industry and Foreign Trade (MDIC) and is bankrolled by Brazil’s workers’ pension and treasury funds. 8 The primary financier of government development initiatives, BNDES is expected to finance up to 80% of Belo Monte’s costs, which could exceed $30 billion. As a public institution, BNDES has received persistent criticism for its lack of transparency and public participation in its lending process and has few, if any, social or environmental safeguards for its projects. 9

The indigenous groups throughout the Xingu river basin—including the Kayapó, Arara, Juruna, and Xipaia tribes—have consistently and adamantly spoken out against government plans to build the Belo Monte dam. Indigenous leaders have repeatedly promised to lay down their lives to defend the river upon which they depend for survival. On April 23, Kayapó groups began a monthlong blockade of the BR-80 highway where it crosses the Xingu River, disrupting a major transport artery for commercial goods. In a May 1 interview with the French channel TF1, Chief Raoni of the Kayapó said, “I have asked my warriors to prepare for war, and I have spoken of this with other tribes from the upper Xingu. We will not let them [build this dam].”

The state-run electric company Electronorte has attempted to buy off some Xingu-area peoples with a financial compensation package in exchange for their public support for the project. In reality, this “compensation” package is composed of services that are already constitutionally guaranteed to indigenous people, yet are treated by Electronorte and the media as handouts because of a lack of public information. There have also been threats that social services will be withdrawn entirely from communities that continue to oppose the Belo Monte project. This has effectively fractured a previously united indigenous re¬sistance to the dam, and the Xicrim people—citing fears of retaliation, cutoff services, and no compensation for their communities—have refused to participate in further resistance activities or dialogues.

After a June 4 meeting of leaders from 10 indigenous nations, Kayapó leader Megaron Txucarramãe stated: “I am sad and angry, not angry with our indigenous brothers, but I am angry with the government, with Electronorte, who worked to fool them, using money to pit us against each other, demobilizing our struggle against Belo Monte. We have decided not to fight and rather to step back and converse among ourselves. We are not going to fight with our brothers. We will ask the NGOs to help us go to Brasilia to speak with Lula; that is what the leaders agreed here.”

José Carlos Arara explained, “Our ancestors are there inside this land, our blood is inside the land, and we have to pass on this land with the story of our ancestors to our children. We don’t want to fight, but we are ready to fight for our land if we are threatened. We want to live on our land in peace with all that we have there.”

But contrary to promises made in the media and in the Brazilian constitution, neither Lula nor other government officials have been willing to speak with indigenous groups in any formal setting to discuss Belo Monte, im¬pacts of its construction, or residents’ concerns. Instead of enforcing its own laws, Brazil prefers to turn a blind eye, just as miners, ranchers, and developers have ignored all indigenous constitutional protections. Even more chilling, plans have already been developed for over 100 new dams in the region.

Part of the larger problem is that Brazil’s indigenous peoples still have ambiguous legal standing. It remains unclear if indigenous individuals—historically ascribed a unique position as statutory minors, or “wards of the state”—can even claim citizenship. This has encouraged persistent discrimination by the non-indigenous community, and has allowed corporate and economic interests to take priority in the development and management of indigenous lands. Moreover, linguistic and geographical barriers, together with unfamiliarity with the political system, have led indigenous Brazilians to depend heavily upon outside mediators to defend their rights, a situation that continues today.

Throughout the 1980s and 1990s, environmentalists and indigenous groups formed partnerships to protect native lands and to fight against environmentally devastating development projects. Support from local and international civil society gave Brazilian indigenous leaders a new independence from FUNAI, and an opportunity to take griev¬ances directly to international supporters and a global audience. Attention from Avatar director James Cameron and two of its stars, Sigourney Weaver and Joel David Moore, has renewed a global interest in the plight of indigenous people in Brazil today. And this time around, the stakes are higher than ever before. Unfortunately, mainstream Brazilian media and many government officials perpetuate conspiracy theories that indigenous rights are simply a strategy for foreign control of Brazilian natural resources.

This support enabled the successful resistance against the first Xingu dams more than 20 years ago. The Kayapó of the Upper Xingu Basin, who had prior success advocating for the removal of illegal gold miners and loggers from their lands, were among the most vocal indigenous opponents of the government’s original plans to dam the Xingu, known then as the Kararoâ project. In 1988, two Kayapó leaders flew to Washington and spoke directly to officials at the World Bank, the U.S. Treasury, and Congress about the dam project. In 1989, dozens of indigenous nations came together for the Conference of Indigenous Peoples of the Xingu, staging a protest in Altamira to demand that the Xingu River remain free of dams. Their steadfast resistance, together with broad-based international support, contributed to the World Bank’s decision to suspend Brazil’s first power-sector loan in 1989, forcing the government to postpone the Kararoâ dam complex.

The arguments for building the Belo Monte dam do not begin to justify the true costs of its construction, nor do they hide the Brazilian government’s hypocrisy in pursuing expedited development and permitting processes while paying lip service to indigenous rights and environmental conservation. The Belo Monte dam project defies constitutional and international policies, and will have devastating, permanent social and environmental consequences. The fight to stop Belo Monte and conserve the Xingu region is critical to preserving indigenous cultural identity and to ensuring fair treatment in future land management and developmental activities. Slated for completion in 2015, the dam will likely continue to be a source of conflict in the Amazon in the months and years ahead.

1. Guy Burton, “Lula and Economic Development,” GlobalAffairs.es, June 18, 2010.

2. Sue Branford, “Belo Monte—Indians Threaten ‘River of Blood,’ ” Latin America Bureau, lab.org.uk, April 26, 2010

3. Quoted in Tom Philips “Brazil to Build Controversial Belo Monte Hydroelectric Dam in Amazon Rainforest,” The Guardian (London), February 2, 2010.

4. Seth Garfield, Indigenous Struggle at the Heart of Brazil: State Policy, Frontier Expansion, and the Xavante Indians, 1937–1988 (Duke University Press, 2001), 143.

5. Amazon Watch, “Ten Myths the Brazilian Government Wants You to Believe About Belo Monte,” fact sheet, 2010.

6. Telma Monteiro, “Portaria cria vara federal ambiental no Pará e novo juiz assume ações contra Belo Monte,” telmadmonteiro.blogspot.com, June 4, 2010; Amazon Watch, “Update: Growing Legal Crisis Around Belo Monte,” amazonwatch.org/newsroom, June 11, 2010.

7. Movimento Xingu Vivo Para Sempre, International Rivers, Amazon Watch, “Belo Monte Dam Auction Goes Forward in Brazil after Court Overrules Second Injunction,” press release, amazonwatch.org/newsroom, April 20, 2010.

8. International Rivers and Amazon Watch, “Lack of Private Sector in Belo Monte Consortium Signals Investor Concerns Over Financial Risks,” press release, internationalrivers.org, July 16, 2010.

9. Plataforma Bndes, “A Letter From Peoples Affected by Projects Financed by BNDES,” plataformabndes.org.br, November 25, 2009.

Sara Diamond is a San Francisco–based researcher and consultant specializing in tropical ecology, conservation, and policy. Christian Poirier is the Brazil Program Coordinator at Amazon Watch.

Like this article? Support our work. D onate  now. 

The Role of Geological Investigations for Dam Siting: Mosul Dam a Case Study

• Original Paper
• Open access
• Published: 18 December 2019
• Volume 38 , pages 2085–2096, ( 2020 )

Cite this article

You have full access to this open access article

• Varoujan K. Sissakian 1 , 2 ,
• Nasrat Adamo 3 &
• Nadhir Al-Ansari   ORCID: orcid.org/0000-0002-6790-2653 4  

36k Accesses

17 Citations

7 Altmetric

Explore all metrics

Dams are engineering structures constructed for different purposes. They are of different sizes, shapes and types. In all cases, many essential studies should be carried out before deciding the location, type and size of the dam. Among those studies is the geological investigations which should be carried out to deduce the geological conditions in the most relevant site, depth of the foundations and their types, cut-off depth, type of the available construction materials, and type of the expected geological hazards. Without proper geological investigations, the siting of a dam will cause serious hazards during construction and during commissioning of the dam. In this study, Mosul Dam case is considered as the consequences of inadequate geological investigations which were carried out by the contractor and supervised by Swiss Consultant. The location of the dam site and its foundations are built over a highly karstified area, where gypsum and limestone beds are exposed and exist deep under the ground surface, and even deeper than the foundations. Accordingly, grouting treatment was carried out and still on going, but all the attempts to have a safe and relevant dam were in vain. In this study we have provided the essential studies which should be included during the geological investigation to have a safe and sound dam.

Similar content being viewed by others

Construction Aspects of Dams

Risk and Stability Evaluation of Klego Dam, Boyolali, Indonesia

Comparative Study on the Design Process of a Shallow Subway Station Isfahan Metro, Iran, Based on Risk Assessment

Avoid common mistakes on your manuscript.

1 Introduction

Geological investigation for selecting and locating dam sites is one of the most significant studies which should be carried out in different scales and stages before deciding the best location for a dam. Therefore, an adequate assessment of site geologic and geotechnical conditions is one of the most significant aspects of a dam safety evaluation. Evaluation of the safety of a new dam requires, among other things, that its site, abutments, foundation and reservoir have been adequately examined, explored, and investigated so that the geological conditions are fully understood as much as possible.

The geological investigations should include four main topics; these are (Woodward 2005 ):

The geology of the dam site including the foundation for the dam itself and the sites for other structures such as spillway, diversion tunnel and outlet works. To check whether the dam foundation has sufficient strength and durability to support the type of dam proposed, whether the foundation is watertight, especially, when karstified rocks occur in the site and in deeper horizons bellow the foundations.

The geology of the area to be occupied by the reservoir once the dam is completed. Whether the storage area is watertight or are there areas of cavernous limestone and/or gypsum which might lead to the dam not retaining water.

Stability of the slopes in the dam site and reservoir area whether landslides into the reservoir are possible which might cause a wave of water to be pushed over the top of the dam.

Finding sources of the construction materials which will be needed to build the dam in nearby areas of the dam site including all required types like: aggregates of different types and sizes, filling materials in the core and both surfaces (if the dam is of earth-fill type).

The main aim of this article is to shed light on the role of the geological investigations in dam siting and to elucidate the consequences when the investigations are inadequate and/or the acquired data is miss-interpreted, which means the interpretation of the acquired data was not relevantly performed. Accordingly, wrong conclusions may be achieved. Mosul Dam (Fig.  1 ) case is presented as a good and unique example for inadequate geological investigation in dam siting and the consequences which were the reason for calling the dam as “the most dangerous dam in the world” (Al-Ansari et al. 2015a , b , 2017 ).

Satellite image showing the location of Mosul Dam

1.1 Previous Studies

Studies concerning the role of the geological investigation in dam siting are enormous. Moreover, many guidelines, instruction booklets, safety codes are also available at different sources. Among the mentioned sources, but not limited to; are: Swiss Consultants Consortium ( 1984 ), Woodward ( 2005 ), Kocbay and Kilic ( 2006 ), Kelly et al. ( 2007 ), Fraser ( 2001 ), Nezhad et al. ( 2012 ), Bell ( 2013 ), Al-Ansari et al. ( 2015a , b ), Sissakian et al. ( 2015 ), Adamo et al. ( 2017 ), Jayanath et al. ( 2017 ), Kanik and Ersoy ( 2019 ) and Poorbehzadi et al. ( 2019 ). All the mentioned publications either deal with a certain dam with its problematic geological conditions and/or mention the necessary steps which were considered during the geological investigation for dam siting to have a safe dam site during constructions and after commissioning of the dam.

However, we have specified some of the studies mentioning their main subjects. Cetin et al. ( 2000 ) explained how the settlements in Ataturk Dam in Turkey were solved. Dreybrodt et al. ( 2002 ) and Hiller et al. ( 2011 ) discussed the problems exerted by karst forms below dam foundations. Bonacci ( 2008 ) described how the water losses were treated in Boljuncia reservoir in Croatia. Hiller et al. ( 2012 ) discussed the karstification problems in Birs Weir at Basel in Switzerland. Milanovic ( 2011 ) discussed the karstification problems in gypsum karstification in dam sites. Andreo et al. ( 2015 ) explained the process of environmental investigation in karst systems. Mozafari and Raeisi ( 2016 ) discussed the karstification problems in the foundations of Salman Farsi Dam in Iran. Milanovic ( 2018 ) discussed the engineering problems caused by karstification in dam sites.

2 Materials and Methods

To fulfil the aims of the current study, tens relevant published articles were reviewed in order to present the most necessary steps which should be followed during performing geological investigation to select a compatible and safe dam site. Moreover, Mosul Dam case is presented as a typical example of performing inadequate geological investigation and wrong interpretation of the acquired data from these geological investigation (Al-Ansari et al. 2015a , b ; Sissakian et al. 2015 , 2017 ).

3 Necessary Steps in Geological Investigation

The necessary steps which should be followed during performing geological investigation for dam siting are briefly mentioned hereinafter. However, it is strongly recommended to perform the geological mapping within the investigation by the national geological survey office with contribution of university and other concerned specialists. Because the geologists in a national geological office have more regional data which can be used in the interpretation of the acquired data, especially the subsurface data. This is attributed to the need for required necessary regional geological data in the dam site and reservoir area.

Geological Maps Studying available geological maps for the selected dam sit; if the dam site is already selected. Otherwise, many alternative sites should be recommended depending on the geological data acquired from the geological maps and then should be ranked using other necessary data which are concerned with dam siting.

Geological Mapping Geological mapping at a scale of 1: 5000 should be performed by well experienced geologists having excellent engineering geological background. The geological maps should present;

Type of the exposed rocks and their thicknesses in the dam site.

Mechanical and geotechnical properties of the exposed rocks in the dam site and deeper than the foundations (more than cut-off depth).

To elucidate if there are karstified rocks (gypsum and limestone) and/or expansive clays.

Presenting all existing faults and other structural elements which shed light on the existence of active faults.

Presenting all Neotectonic evidences.

Drilling Operations Boreholes should be drilled in the dam site and reservoir area. The number, depth and spacing of the boreholes depend on:

Type of the dam.

Height of the dam.

Geological complexity of the dam site and reservoir area.

However, the following aspects should be considered:

All boreholes should be drilled by full core recovery type.

The core recovery should not be less than 85%.

RQD should be calculated.

Depth of the boreholes should not be less than the deepest karstified bedrocks, expansive clays, fractured and/or sheared zones, active faults (if any); otherwise deeper than the cut-off depth of the foundations. In such case the depth should be at least 1.5 H, where (H) is the maximum hydraulic head acting on the foundation as a rule of thumb.

The site geologists should be well experienced in core description, especially karstified rocks and/or karst filling materials.

Systematic sampling of the extracted core in order to apply required geotechnical tests which will provide the mechanical properties of the penetrated rocks.

Applying chemical analyses for the cored rocks and unconsolidated materials, especially those which will be used in construction of different parts of the dam. +

Applying colored photography of the extracted core before sampling but after being cleaned from the drilling mud and/or fluids. The colored photos may be used during the construction of the dam or even during commissioning when needed for certain use.

Applying geophysical logging for the all drilled boreholes to indicate:

To correct the drilling depths of the penetrated rocks.

To indicated the mechanical properties of the penetrated rocks.

To indicate cavities, voids, fractured and/or sheared zones in the borehole.

Full scale Lugeon field permeability tests may be necessary in important cases when permeability is questionable

Karstification Karstification is one of the main significant processes which should be studied to indicate whether the dam site and reservoir area suffer from karstification or otherwise. Karstification can be of two main types:

Surface Karst When karstification is on surface, then its indications can be seen and recognized by experienced geologists, among those indications are:

Presence of karst forms such as sinkholes (of all types and sizes, including active and inactive forms).

Presence of circular and/or crescent-shaped cracks on surface which may indicate the presence of shallow karst forms (Fig.  2 ).

False dipping and crossing of valleys by the dipping rocks without obeying the V-rule

Presence of blind valleys.

Subsurface Karst This type is more dangerous than the surface karst, because it may not be detected in the dam site, foundation area, reservoir area and other structures of the dam by the site geologist. Accordingly, significant problems will arise during construction and commissioning of the dam (Jassim et al. 1987 , Fig.  3 ). The presence of subsurface karst forms can be detected by:

Subsurface karst forms (red arrows) appeared after excavation of an industrial site within the rocks of the Fatha Formation in Shiekh Ibrahim anticline 45 km south of Mosul Dam. Note the crossing of valleys by the dipping rocks without obeying the V-rule (Blue arrow)

Interpretation of high resolution satellite images, where different indications can be recognized by well experienced geologist, among them are (Sissakian and Abdul-Jabbar 2005 ):

Presence of false dipping towards certain side.

Presence of abnormal valleys.

Crossing of rock beds to the valleys without obeying the V-rule of dipping beds (Figs.  4 and 5 ).

Crossing of valleys by the dipping rocks without obeying the V-rule

Interpreted satellite image near Derbendikhan Dam showing unstable slopes in red lines and dashed orange lines shows cliffs (after Sissakian et al. 2019 )

Falling of the drilling pipes during continuous core drilling operation.

Loss of the drilling water and/or muds during continuous core drilling operation.

Loss of the core.

Presence of Terra Rosa (red soil, karst filling material) is indication for karstification. Such clayey soil was interpreted as “bauxite” in Mosul Dam by Swiss Consultants Consortium ( 1989 ) and Wakeley et al. ( 2007 ) and interpreted as Terra Rosa by Sissakian et al. ( 2017 ).

Interpretation of the logging data to indicate karst cavities or otherwise.

Slope Stability Analysis All slopes (natural and man-made) in the dam site and reservoir area should be studied and analyzed to recognize their activity and all other slope stability problems. All existing types should be carefully mapped (Fig.  2 ) and analyzed, and to indicate if they are active or inactive forms. Moreover, all prone areas for mass movements should be clearly indicated and treated relevantly.

Hydrogeological Studies The groundwater in the dam site and reservoir area should be studied and mapped including:

Depth of the groundwater.

Type of the groundwater.

Type of the aquifer(s).

Corrosiveness of the groundwater and salts contents.

Seepages of the groundwater; their locations and quantity.

Checking for new seepages during constructions and commissioning, and the difference in the yield quantities of existing seepages.

Geophysical Studies Different types of geophysical studies should be performed in the dam site and reservoir area to indicate the following data:

To indicate groundwater level by means of Electrical Method.

To indicate subsurface caverns (Karst forms) by means of Micro-gravity Method, or even by using geo-radar investigations in important cases.

To indicate the mechanical properties of the rocks and the depth of the weathering zone by means of seismic refraction and gravity Methods.

Seismicity Study Seismic zonation map should be consulted during the site investigation. In seismically active area, it is necessary to assess the degree of earthquake tremors and design must include provisions for the added loading and increased stresses. Historical seismicity studies are also required in case of important dams to have a complete file of the seismic events that might hit the site.

Burrow Areas Construction of a dam requires large quantity of construction materials like soil, rock, concrete and aggregates. The most economical type of dam will often be the one for which materials are to be found in sufficient quantity at a reasonable distance from the site. So availability of such materials nearby the proposed site should be assessed during the geological investigation. Different laboratory and in-situe tests should be applied to elucidate the characteristics of the materials found in burrow areas and which will be used in the dam’s construction.

4 Mosul Dam: A Case Study

4.1 history of mosul dam project.

The investigations for building Mosul Dam project started in 1950 and it was referred to as Aski Mosul Dam. The location of the dam was suggested in 1953 to be at a village called “Dhaw Al-Qamar”, which is located 12 km north of Aski Mosul village. The dam was designed so that its capacity reaches 8.7 km 3 at 320 m (a.s.l.) while the maximum elevation of the dam reaches 324 m (a.s.l.). In 1956, the Iraqi Government asked Harza Company to perform a new site survey and design for the dam. In 1960, Harza Company suggested two sites for the dam; different from those suggested earlier by other companies, because the dam will be built on highly soluble gypsum and very thin clay beds. The first suggested site was to build a dam with a storage capacity of 7.8 km 3 and the other site was with a storage capacity of 13.5 km 3 . In 1962, the Iraqi Government asked Techno-prom Export (Soviet company) to perform another investigation for the site of Mosul Dam. The company suggested a new site that is 600 m south of the site suggested by Harza Company. The dam was designed with a storage capacity of 7.7 km 3 (Al-Ansari et al. 2015a , b ).

All the above companies suggested that the dam should be of an earthfill type with compressed clay core, but there were different views about the exact location of the dam. Grouting was suggested to be performed under the dam, spillway and the electricity generation station as foundation treatment. In addition, they suggested that detailed geological investigation should be performed before any construction activities should begin. In view of these reports, the Iraqi Government asked a Finish company “AmitranVoima” in 1965 to carry out new investigations. The company suggested a site, which is located 60 km northwest of Mosul city and they pointed out that the geology of the area is very complex and required further investigations, therefore a Yugoslavian company (Geotechinka) worked on further geological investigations at the suggested site in 1972. AmitranVoima then again carried out more investigation in 1973. All the reports were then studied by the International Board of Dams Experts which was appointed by the Government in 1974 and recommended extra geological investigations. In a further step the Iraqi Directorate General for Dams asked a French company (Soletanch) to perform more geological investigation on the suggested site. This was done during 1974–1978). Later on in 1978, the Swiss Consultants Consortium was asked to be the consultants for Mosul Dam project and a consortium of German and Italian companies (GIMOD) was asked to execute the civil and steel works of the project in 1980. The work started on 25th January, 1981 and finished 24th July, 1986 (Al-Ansari et al. 2015a , b ).

4.2 Details of Mosul Dam

Mosul Dam is one of the most important strategic projects in Iraq for the management of its water resources. The dam was constructed on the Tigris River (Fig.  1 ), it is located 60 km northwest of Mosul city. The dam is 113 m high, 3650 m long including the spillway, has a 10 m top width and the crest level is 341 m (a.s.l.). The dam is faced with rock and has an earth fill with a clay core. The dam was designed to impound 11.11 km 3 of water at normal operation level of 330 m (a.s.l.), including 8.16 km 3 and 2.95 km 3 of live storage and dead storage, respectively. The dam has a concrete spillway located on the left abutment.

4.3 Geology of the Mosul Dam Site

The oldest exposed rocks in Mosul Dam site belong to the Fatha Formation of Middle Miocene age; however, in the reservoir area, the oldest exposed rocks belong to the Pila Spi Formation of Late Eocene age (Fig.  6 ). The exposed formations in the dam site and reservoir area are described briefly hereinafter (Sissakian and Al-Jiburi 2014 ).

Geological map of Mosul Dam and reservoir area (after Sissakian and Fouad 2012 )

Pila Spi Formation (Late Eocene): small outcrops of the Pila Spi Formation are exposed in the extreme northwestern side of the reservoir (Fig.  6 ). The formation consists of well bedded, dolomite, dolomitic limestone, limestone and rare marl. The exposed thickness of the formation is few meters only.

Euphrates Formation (Early Miocene): the Euphrates Formation is exposed in the core of some anticlines in the reservoir area (Fig.  6 ), as well in the foundations of the dam. The formation consists of well bedded, hard limestone, marly limestone and dolomitic limestone. Some of the limestone beds are karstified as indicated by the presence of sinkholes. The thickness of the formation in nearby areas ranges from (15–50) m.

Jeribe Formation (Middle Miocene): the Jeribe Formation is not recorded to be exposed in the dam site and reservoir area. However, the formation is recorded in the foundations of the dam (Sissakian et al. 2014 ). The formation consists of well bedded, hard limestone. The thickness in near surroundings is about 60 m.

Fatha Formation (Middle Miocene): the Fatha Formation is widely exposed in the dam site and reservoir area (Fig.  6 ). The formation is characterized by cyclic sediments, the thickness of the formation is variable; in Butmah is 392 m, in Ain Zala 325. The formation consists of two members; these are:

Lower Member the Lower Member consists of cyclic sediments, each cycle consists of green marl, limestone and gypsum. The abutments of the dam are located within this member. The rocks of this member are highly karstified, not only in the dam site and the reservoir area, but else-where in Iraq.

Upper Member the Upper Member consists of cyclic sediments, each cycle consists of green marl, red claystone limestone and gypsum; in the uppermost part reddish brown sandstone is present. This member covers majority of the reservoir area.

Injana Formation (Late Miocene): the Injana Formation is exposed in the eastern, northeastern and some parts of the northern banks of the reservoir (Fig.  6 ). The formation consists of fining upwards cyclic sediments of reddish brown sandstone, siltstone and claystone. The thickness of the formation is variable ranges from (200–330) m.

Mukdadiya Formation (Late Miocene–Pliocene): the Mukdadiya Formation is exposed in the eastern bank of the reservoir (Fig.  6 ). The formation consists of fining upwards cyclic sediments of grey sandstone, siltstone and claystone; some of the sandstone beds are pebbly. The thickness of the formation is variable ranges from (100–230) m.

Mosul Dam site and reservoir area are located within the Low Folded Zone; within the Outer Platform of the Arabian Plate. The Outer Platform is also part of the Zagros Fold–Thrust Belt (Fouad 2007 ). Although many deep seated faults and surface faults occur in the dam site and reservoir area, but without any significant effect on the stability of the dam.

4.4 Karstification Problems in Mosul Dam

The geology of Mosul Dam site is characterized by the presence of four layers of brecciated gypsum (GB 0, GB 1, GB 2 and GB 3) within the Fatha Formation. These layers have thicknesses which range of (8–18) m (Swiss Consultants Consortium 1984 , 1989 ). The GB 0 is at a depth of 80 m from the ground surface in the river section, while GB 3 was uncovered in the excavation of the foundations of the spillway (Fig.  7 ).

Karst line location under the dam (after Adamo and Al-Ansari 2016 ). Note the extension of the karst line which crosses different beds irrespective to the dipping of the beds and the presence of many gypsum beds (GB 0) below the line

The importance of these gypsum layers stems from their resistance to take grout materials during the construction of the deep grout curtain under the dam. In addition, they could not keep the grout material when subjected to the rising hydrostatic pressure due to the impounding of the reservoir. Therefore, the grouting process was not effective although it had started with construction of the dam and is still on going. The failing to find proper solutions for the continuous seepage seemed to originate from the miss-interpretation of the basic geological facts, which means not recognizing geological facts, for example the encountered Terra Rosa (a good indication for karstification) in the boreholes of Mosul Dam was interpreted as “Bauxite” although such bauxite also indicates karst depressions. Moreover, miss-judgement of gypsum rock behavior in this environment and its dissolution phenomenon, in addition to the peculiar nature of the brecciated gypsum in not accepting grouting materials. This has led to the current maintenance work on the grout curtain which continued from 1985 until today.

The main reason for failing in the grouting process is attributed to miss-interpretation of the acquired data from core description and from applying the Lugeon tests to identify the depth of the karstified gypsum beds below the foundations. Figures  7 and 8 show constructed geological cross sections indicating the karst line running in varying depths along the axis of the dam (Modacom 1984 ). It is clear that many gypsum beds exist below the constructed karst line; therefore, grouting to fill the karst cavities in the foundations will not be effective because karstified gypsum beds still occur below the depth to which grouting is applied.

Geological section starts from (East) at upper left corner to (West) at bottom lower (Modacom 1984 in Adamo and Al-Ansari 2016 ). Note the constructed karst line (dashed red line) and note the gypsum beds below the karst line. For legend, refer to Fig.  9 )

Another miss-interpretation is the encountered Terra-rossa (reddish brown clayey soil) in the boreholes and even in the excavations which was described as “Bauxite”. The presence of the Terra-rossa is good indication for karstification. Moreover, the used lithological terms (Fig.  9 ) is good indication that the site geologist was not familiar with karst filling sediments; accordingly, he missed interpreting the existing karst forms below the dam’s foundation (Figs.  7 and 8 ).

Lithological legend for Fig.  7

5 Discussion

In the current study, we have emphasized the role of the geological investigation in dam siting and the consequences when the investigation is inadequate. This means, the aforementioned instructions and necessary works were not performed, either in quantity or quality. Moreover, we have emphasized on the karstification problems, since large parts of Iraq are karstified and suffer from existing problems.

To have adequate geological investigation and a safe dam site, it is necessary to perform the geological investigation in the following three steps.

Preliminary Investigations This investigation should provide a first general impression of the engineering and geological aspects of the proposed site(s). The field work generally would include preliminary field geologic mapping, some preliminary hand auger holes for soil and overburden sampling, a limited number of core holes into rock and possibly some preliminary seismic refraction lines. This information would be used to answer questions raised by an office study. The data would also be used to plan the type, location, and amount of explorations and laboratory testing required for future, more detailed investigations. Air photos and satellite images can also give good guide in interpreting the geological forms at the site(s).

Initial Design Investigations These investigations would be undertaken to provide more detailed information on foundation characteristics on a particular site or several sites, and to provide data for preliminary considerations of the design requirements and construction methods. This phase of field investigation should include surface and subsurface exploration and sampling through borings, test pits, test trenches, material testing, geologic mapping, and additional geophysical surveys to supplement drilling. Data developed from these activities should be used to compare alternative sites, to analyze different types of structures that might serve the same purpose, and to develop economic evaluations of the sites. An end product of this investigation is to rank the studied sites (usually three sites).

Final Design Investigation These investigations would be primarily composed of detailed drilling, sampling, and testing concentrated on specific features at the selected project site; and should be specifically planned to provide the engineer with information that is necessary to design structures, estimate quantities, determine rates of construction progress, develop cost estimates, prepare plans and specifications, and obtain bids.

6 Conclusions

The main conclusion of the current study is the performance of geological investigation in different levels is very necessary in dam siting and having a safe dam without or with minimum maintenance works to keep the dam as safe as possible. Moreover, geological investigation should be carried out by well experienced geologists, geotechnical engineers with a relevant consultant firm. Since we have emphasized on karstification and related problems; therefore, the geologists and geotechnical engineers should be well experienced in karst forms and karst filling sediments. The geological investigation should cover the dam site, the structures of the dam, foundation and reservoir area.

Adamo N, Al-Ansari NA (2016) Mosul Dam full story: engineering problems. J Earth Sci Geotech Eng 6:213–244

Google Scholar  

Adamo N, Al-Ansari NA, Laue J Sven, Knutsson S, Sissakian V (2017) Risk management concepts in dam safety evaluation: Mosul Dam as a case study. J Civ Eng Archit 11:635–652

Al-Ansari NA, Adamo N, Issa IE, Sissakian V, Knutsson S (2015a) Geological and engineering investigations of the most dangerous dam in the world. ScienPress, London

Al-Ansari NA, Adamo N, Issa IE, Sissakian V, Knutsson S (2015b) Mystery of Mosul Dam the most dangerous dam in the world: karstification and Sinkholes. J Earth Sci Geotech Eng 5:33–45

Al-Ansari N, Adamo N, Sissakian V, Knutsso NS, Laue J (2017) Is Mosul Dam the most dangerous dam in the world? Review of previous work and possible solutions. Engineering 9:801–823

Article   Google Scholar  

Andreo B, Carrasco F, Durán JJ, Jiménez P, LaMoreaux JW (2015) Hydrogeological and environmental investigations in karst systems. Springer, Berlin. https://doi.org/10.1007/978-3-642-17435-3

Book   Google Scholar  

Bell FG (2013) Engineering geology and geotechniques. Elsevier, Amsterdam

Bonacci O (2008) Water losses from a reservoir built in karst: the example of the Boljunčica reservoir (Istria, Croatia). Environ Geol 58(2):339–345

Cetin H, Laman M, Ertunç A (2000) Settlement and slaking problems in the world’s fourth largest rock-fill dam, the Ataturk Dam in Turkey. Eng Geol 56(3–4):225–244

Dreybrodt W, Romanov D, Gabrovsek F (2002) Karstification below dam sites: a model of increasing leakage from reservoirs. Environ Geol 42(5):518–524

Fouad SF (2007) Tectonic and structural evolution. In: Geology of the western desert. Iraqi journal of geology and mining, special issue no 1. pp 29–50

Fraser WA (2001) Engineering geology considerations for specifying dam foundation objectives. Division of Safety of Dams, California Department of Water Resources. https://water.ca.gov/LegacyFiles/damsafety/docs/egc.pdf . Accessed 20 Sep 2019

Hiller T, Kaufman G, Romanov D (2011) Karstification beneath dam sites: from conceptual models to realistic scenarios. J Hydrol 398(3–4):202–211

Hiller T, Romanov D, Kaufman G, Huggenberger P (2012) Karstification beneath the Birs weir in Basel/Switzerland: a 3D modeling approach. J Hydrol 448:181–194

Jassim SZ, Sissakian VK, Taufiq JM (1987) Final report on the detailed geological mapping of Atshan Wax Plant Area. Iraq Geological Survey Library Report No. 1522

Jayanath MGS, Gunatilake J, Pitawala HMTGA (2017) Geological and geophysical investigation at the construction site of the Kaluganga Main Dam, Sri Lanka. In: Proceedings of the 33rd technical session of Geological Society of Sri Lanka, 2017. http://www.gsslweb.org . Accessed 12 June 2019

Kanık M, Ersoy H (2019) Evaluation of the engineering geological investigation of the Ayvali dam site (NE Turkey). Arab J Geosci. https://doi.org/10.1007/s12517-019-4243-1

Kelly J, Wakeley LD, Broadfoot SW, Pearson ML, McGill TE, Jorgeson JD, Talbot CA, McGrath CJ (2007) Geologic setting of Mosul Dam and its engineering implications. Final report, U.S. Army Engineer District, Gulf Region, Baghdad

Kocbay A, Kilic R (2006) Engineering geological assessment of the Obruk dam site, Corum, Turkey. Eng Geol 87(3–4):141–148

Milanović P (2011) Dams and reservoirs in Karst. In: van Beynen P (ed) Karst management. Springer, Berlin, pp 47–73. https://doi.org/10.1007/978-94-007-1207-2

Chapter   Google Scholar  

Milanović P (2018) Engineering karstology of dams and reservoirs. CRC Press, Boca Raton

Modacom JV (1984) Permeability of parent rocks. Mosul Dam project report, Drawing No.FR 82303. Mosul Dam Library, Mosul, Iraq

Mozafari M, Raeisi E (2016) Salman Farsi Dam reservoir, a successful project on a karstified foundation, SW Iran. Environ Earth Sci. https://doi.org/10.1007/s12665-016-5844-6

Nezhad HZ, Ajalloeian R, Azimian A (2012) Evaluation of geological and engineering geological properties of Cheshmeh-Asheq dam site. Electron J Geotech Eng 17(2):2633–2644

Poorbehzadi K, Yazdi A, Teshnizi ES, Dabiri R (2019) Investigating of Geotechnical Parameters of Alluvial Foundation in Zaram-Rud Dam Site, North Iran. Int J Min Eng Technol 1(1):33–44

Sissakian VK, Abdul-Jabbar MF (2005) Site selection problems in gypsum-bearing formations. A case study from north of Iraq. Iraqi Bull Geol Min 1(2):45–52

Sissakian VK, Al-Jiburi BM (2014) Stratigraphy. In: Geology of the high folded zone. Iraqi bulletin of geology and mining, special issue no 6. pp 73–161

Sissakian VK, Kadhum TH, Abdul Jab'bar MF (2014) Geomorphology. In: The geology of the high folded zone. Iraqi bulletin of geology and mining, special issue no 6. pp 7–56

Sissakian V, Al-Ansari NA, Issa IE, Adamo N, Knutsson S (2015) Mystery of Mosul Dam the most dangerous dam in the world: general geology. J Earth Sci Geotech Eng 5:1–13

Sissakian V, Adamao N, Al-Ansari N, Knutsson S, Laue J (2017) Defects in foundation desigsn due to miss-interpretation of the geological data. A case study of Mosul Dam. Sci Res Eng 9(7):1–15. https://doi.org/10.4236/eng.2017.90742

Sissakian V, Fouad SF (2012) Geological map of iraq, scale 1:1000000, 4th ed. Iraq Geological Survey Publications, Baghdad, Iraq

Sissakian VK, Sdiq SB, Xailifar GH (2019) Slope stability of Derbendi Khan Dam and near surroundings. A reconnaissance Study. Kurdistan Region, NE Iraq. J Zankoy Sulaimani 20(3–4):57–72

Swiss Consultants Consortium (1984) Mosul Dam flood wave. Summary, vol 1. Mosul Dam Library, Internal report

Swiss Consultants Consortium (1989) Mosul Dam project main scheme, Final report, and as built drawings. Swiss Consultants Consortium, Section 2.2

Wakeley LD, Kelley JR, Talbot CA, Pearson ML, Broadfoot SW (2007) Geologic conceptual model of Mosul Dam. U.S. Army Engineer Division, Gulf Region, Baghdad

Woodward R (2005) Geology of dams. Internet data. http://members.optusnet.com.au/~engineeringgeologist/page19.Html . Last accessed 12 June 2019

Download references

Acknowledgements

Open access funding provided by Lulea University of Technology.

Author information

Authors and affiliations.

University of Kurdistan Hewler, Erbil, Iraq

Varoujan K. Sissakian

Erbil, Iraq

Norrköping, Sweden

Nasrat Adamo

Lulea University of Technology, Lulea, Sweden

Nadhir Al-Ansari

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Nadhir Al-Ansari .

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Sissakian, V.K., Adamo, N. & Al-Ansari, N. The Role of Geological Investigations for Dam Siting: Mosul Dam a Case Study. Geotech Geol Eng 38 , 2085–2096 (2020). https://doi.org/10.1007/s10706-019-01150-2

Download citation

Received : 07 October 2019

Accepted : 10 December 2019

Published : 18 December 2019

Issue Date : April 2020

DOI : https://doi.org/10.1007/s10706-019-01150-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Geological investigations
• Karstification
• Find a journal
• Publish with us
• Track your research

To read this content please select one of the options below:

Please note you do not have access to teaching notes, conflict dynamics in a dam construction project: a case study.

Built Environment Project and Asset Management

ISSN : 2044-124X

Article publication date: 18 November 2011

The purpose of this paper is to present a model for a comprehensive and integrated approach to managing interface conflict from the early stages of a dam construction project.

Design/methodology/approach

A case study methodology is adopted. Following comprehensive literature review, qualitative data were gathered from case studies through interviews conducted on the Middle Marsyangdi Hydroelectric Project (MMHEP) dam project in Nepal. Causal loop diagrams on the typical evolution of key indicators of interface conflict were then developed and a simulate‐able model of interface conflict was derived using system dynamic modeling technique. The model was then simulated to derive viable policies for future management of dam construction projects in developing countries.

The study reveals that interface conflicts at the construction stage of projects are caused mainly by lack of effective Environmental Impact Assessment, public participation and mutual consultation, on timely basis and accurate information from the early stages of projects. The system dynamic model is able to replicate general behavior of evolution of interface conflict in a dam construction project. Furthermore, the study explored three viable policies to avoid and minimize interface conflict in the construction stage of a dam project. The policies were tested and demonstrated to be useful in improving the value of projects to stakeholders. It is demonstrated that a combination of policies is better than adopting a single policy to stakeholder management.

Originality/value

The paper demonstrates the utility of system dynamics as a modeling tool for understanding the dynamics of conflicts on dam construction projects. The model should be helpful to policy makers on large projects, especially those likely to be subject to social and environmental conflict. Policies derived from the model have the potential of being used to assess and take proactive measures to manage conflicts effectively and efficiently from early in a project's life.

• Developing countries
• Dam construction
• Project management
• Interface conflict
• System dynamics
• Sustainable construction

Kishor Mahato, B. and Ogunlana, S.O. (2011), "Conflict dynamics in a dam construction project: a case study", Built Environment Project and Asset Management , Vol. 1 No. 2, pp. 176-194. https://doi.org/10.1108/20441241111180424

Emerald Group Publishing Limited

Copyright © 2011, Emerald Group Publishing Limited

Related articles

All feedback is valuable.

Please share your general feedback

Report an issue or find answers to frequently asked questions

Contact Customer Support

Project Management: Achieving Competitive Advantage, Fifth Edition by Jeffrey K. Pinto

Get full access to Project Management: Achieving Competitive Advantage, Fifth Edition and 60K+ other titles, with a free 10-day trial of O'Reilly.

There are also live events, courses curated by job role, and more.

CASE STUDY 8.1 The Hidden Costs of Infrastructure Projects—The Case of Building Dams

In recent years there has been a surge in the construction of large dams in both developing and developed ­countries including Brazil, Ethiopia, Pakistan, and China. There are several reasons for this increased interest in dam building. Demand for electricity is expected to double worldwide between 2010 and 2035. Thus, hydroelectric power is seen as a cheap, available option for countries with river systems that allow for dams, particularly as it is a preferred option to dirtier coal-burning plants. Other arguments in favor of building these dams include their use for flood control, crop irrigation, inland transportation, urban water supplies, and as a job creators. ...

Get Project Management: Achieving Competitive Advantage, Fifth Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Don’t leave empty-handed

Get Mark Richards’s Software Architecture Patterns ebook to better understand how to design components—and how they should interact.

It’s yours, free.

Check it out now on O’Reilly

Dive in for free with a 10-day trial of the O’Reilly learning platform—then explore all the other resources our members count on to build skills and solve problems every day.