Enhancing Transboundary Water Governance Using African Earth Observation Data Cubes in the Nile River Basin: Insights from the Grand Ethiopian Renaissance Dam and Roseries Dam

Abstract

The construction of the Grand Ethiopian Renaissance Dam (GERD) on the Blue Nile has heightened transboundary water tensions in the Nile River Basin, particularly affecting downstream Sudan and Egypt. This study leverages African Earth Observation Data Cubes, specifically Digital Earth Africa’s Water Observations from Space (WOfS) platform, to quantify the hydrological impacts of GERD’s three filling phases (2019–2022) on Sudan’s Roseires Dam. Using Sentinel-2 satellite data processed through the Open Data Cube framework, we analyzed water extent changes from 2018 to 2023, capturing pre- and post-filling dynamics. Results show that GERD’s water spread area increased from 80 km² in 2019 to 528 km² in 2022, while Roseires Dam’s water extent decreased by 9 km² over the same period, with a notable 5 km² loss prior to GERD’s operation (2018–2019). These changes, validated against PERSIANN-CDR rainfall data, correlate with GERD’s filling operations, alongside climatic factors like evapotranspiration and reduced rainfall. The study highlights the potential of Earth Observation (EO) technologies to support transparent, data-driven transboundary water governance. Despite the Cooperative Framework Agreement (CFA) ratified by six upstream states in 2024, mistrust persists due to Egypt and Sudan’s non-ratification. We propose enhancing the Nile Basin Initiative’s Decision Support System with EO data and AI-driven models to optimize water allocation and foster cooperative filling strategies. Benefit-sharing mechanisms, such as energy trade from GERD, could mitigate downstream losses, aligning with the CFA’s equitable utilization principles and the UN Watercourses Convention. This research underscores the critical role of EO-driven frameworks in resolving Nile Basin conflicts and achieving Sustainable Development Goal 6 for sustainable water management.

1. Introduction

The Nile River Basin, spanning eleven countries and sustaining over 300 million people, is a critical lifeline for water, agriculture, and energy [1]. However, competing national interests, historical inequities from colonial-era treaties, and climate pressures have fueled transboundary tensions, particularly among Ethiopia, Sudan, and Egypt [2]. The Grand Ethiopian Renaissance Dam (GERD), a major hydropower project on the Blue Nile, has intensified debates due to its potential to alter downstream flows, impacting Sudan’s Roseires Dam and Egypt’s water security [3]. The Cooperative Framework Agreement (CFA), ratified by six upstream states in 2024, seeks equitable water-sharing but faces resistance from Egypt and Sudan, underscoring the need for data-driven governance [4,5,6].

Indeed, one of the major concerns surrounding the construction of the GERD is the duration of the reservoir filling process and its potential impact on countries downstream of the GERD that share NR. The filling of the GERD reservoir has the potential to affect the downstream flow of water [7]. Sudan and Egypt relies heavily on the Nile River as a primary source of freshwater, and any significant changes in the river’s flow can have profound implications for the country’s water supply, agriculture, and overall water security [7,8]. The concern arises from the fact that, during the filling phase, Ethiopia has the ability to control the release of water from the dam, potentially reducing the flow downstream of the dam [9,10].

Several previous studies have been conducted on the impacts of GERD on the downstream countries and water resource management and structures. For instance, Wheeler et al. [9] discussed the potential impacts of the GERD on downstream water resources. The construction of GERD offers an opportunity for cooperation among the countries of Ethiopia, Sudan, and Egypt. The article analyzes different filling strategies for the dam and their implications for water supply and power generation in downstream countries. The analysis finds that risks to water diversions in Sudan can be managed through adaptations of Sudanese reservoir operations. Risks to Egyptian users and energy generation can be minimized through agreed annual releases from GERD, a drought management policy for the High Aswan Dam, and a basin-wide cooperative agreement to protect the elevation of Lake Nasser highlighted GERD as part of Ethiopia’s strategy to prioritize renewable energy production and meet the growing energy demand in the country, with opposition from downstream countries like Egypt due to concerns about the impact on water availability and downstream flows. The article emphasizes the need for sustainable energy development and the importance of considering the geopolitical implications of large-scale infrastructure projects like GERD. The study by [2] explores the historical rivalry between Egypt and Ethiopia over control of the Nile and the colonial-era treaties that have shaped Egypt’s hydro-hegemony. However, the author argues that armed confrontation between the two countries is unlikely and that the GERD may ultimately lead to the cessation of Africa’s oldest geopolitical rivalry.

The study by [8] used a one-dimensional analysis model called the Hydrologic Engineering Center River Analysis System (HEC-RAS) to investigate the impact of GERD failure. They highlight the potential catastrophic effects on Sudan, specifically on the Roseires, Sennar, and Merowe dams, as well as on the city of Khartoum. The study also suggests that the High Aswan Dam in Egypt would be at risk if GERD were to fail, along with the consequences of the failure of the dams in Sudan. The article emphasizes the importance of analyzing different scenarios related to dam failure and the need for early warning systems and emergency evacuation plans to mitigate the potential loss of life and property. The research by [7] used Hydrologic Engineering Center River Analysis System (HEC-RAS) one-dimensional modeling, using hydrological data to simulate GERD failure scenarios and their impacts on downstream dams (e.g., Roseires, Sennar, Merowe, and High Aswan Dam). The study highlights the catastrophic risks of GERD failure, particularly to Sudan’s Roseires, Sennar, and Merowe dams, and Egypt’s High Aswan Dam. It underscores the need for early warning systems and emergency evacuation plans to mitigate potential loss of life and property in downstream regions. The authors of Mordos et al. [11] studied the impacts of GERD on river inflows, water levels, runoff, and hydrograph shapes are investigated and compared to average baselines. The study concludes that during the first filling of GERD, the runoff of the Blue Nile will decrease by 30% and, in the long run, changes in hydrograph and water levels are expected. The researchers in [12] discussed the impact of the GERD on the performance of the High Aswan dam. The study used a simulation model to assess the potential impact of the dam on the filling and operation phases of the High Aswan dam. The results indicate that the planned 6-year filling period is sufficient to fill the reservoir, with little impact on the current irrigation water demands from High Aswan dam in Egypt. However, there will be a reduction in annual energy output from the High Aswan dam during the filling and after filling stages of GERD.

Recent studies shed more light on these tensions and propose several solutions. The authors of [13] model GERD’s filling scenarios, projecting a 17% reduction in Egypt’s Nile flow under non-cooperative strategies, heightening tensions given Egypt’s near-total reliance on the river. Salhi and Benabdelouahab (2023) [13] utilize Earth Observation (EO) data to quantify downstream flow reductions of 5–7% during GERD’s first two filling phases (2020–2021), noting that cooperative releases could mitigate impacts on Sudan and Egypt, yet unilateral actions persist. Similarly, [13,14] highlight Ethiopia’s independent filling approach, with Sudan facing increased flood risks and Egypt anticipating agricultural losses, despite the Cooperative Framework Agreement’s (CFA) 2024 ratification by six upstream states. These findings underscore ongoing mistrust, incomplete CFA adoption, and the need for EO-driven, cooperative frameworks to resolve Nile Basin conflicts [15]. The study by [4] used climate and socio-economic uncertainty models, incorporating hydrological data, climate projections, and socio-economic scenarios affecting Nile Basin water management. The research demonstrates that adaptive management strategies can mitigate downstream risks caused by GERD’s operation under varying climate and socio-economic conditions. It emphasizes cooperative approaches to balance water needs across Nile Basin countries, reducing potential conflicts.

Effective transboundary water management requires transparent data to monitor impacts and foster cooperation. Earth Observation (EO) technologies, such as DE Africa’s Water Observations from Space (WOfS), offer high-resolution tools to track water extent changes, supporting evidence-based decision-making [16]. EO plays a vital role in understanding our planet, managing its resources, and addressing global challenges, such as climate change, natural disasters, and sustainable development. It provides valuable insights and data-driven information for informed decision-making at various scales, from local to global [17,18,19,20]. The volume of EO data is increasing rapidly, leading to the need for efficient methods of managing and analyzing this vast amount of information [13,21,22]. One approach to managing EO data is through the use of Earth Observation Data Cubes [23,24]. These data cubes are multi-dimensional structures that organize EO data based on spatial and temporal dimensions [25]. They allow users to access and analyze EO data based on coordinates, rather than relying on file names or directory structures. EO data cubes can be implemented on different scales, from local to global, and can be hosted on various infrastructures, including cloud-based systems [23,26,27]. DE Africa is one the EO initiatives that operates a digital platform in Africa with the goal of providing free and unrestricted access to EO data and services for everyone. Its purpose is to enhance Africa’s capacity to utilize EO-based information and address the challenges of sustainable development through Open Data Cubes (ODC) [28,29].

This study focuses on quantifying the hydrological impacts of GERD’s filling on downstream water bodies, specifically Sudan’s Roseires Dam, using satellite-based observations. By leveraging Digital Earth Africa’s Earth Observation Data Cubes and the Water Observations from Space (WOfS) platform, we analyze changes in water extent before and during the filling phases of GERD. The objective is to provide an evidence-based assessment of how large-scale upstream interventions affect downstream water availability, supporting data-driven decision-making for transboundary water governance. By juxtaposing pre- and post-GERD filling scenarios, the study aims to discern the nuanced impacts and implications of large-scale hydraulic interventions on regional water resources. By harnessing the advanced capabilities of DE Africa, this investigation not only offers real-time monitoring and management of these critical dams but also sets a new standard for sustainable water resource governance in the region. The study contributes to a growing body of research that supports transparent, data-driven approaches to transboundary water governance in the Nile Basin.

2. Methodology

2.1. Study Area

The Blue Nile River, originating in the Ethiopian highlands and flowing into Sudan, is the principal tributary of the Nile River, contributing 59–62% of its total annual flow. This study focuses on the transboundary segment of the Blue Nile, where GERD and Roseires Dam serve as critical hydrological infrastructure. The GERD, located 60 km upstream of the Sudan-Ethiopia border (Figure 1), regulates upstream flows, while the Roseires Dam, situated midstream in Sudan, supports irrigation, hydropower, and flood control. The region’s rugged terrain, plateaus, and river valleys shape its hydrological dynamics, with the Ethiopian highlands acting as a key water catchment area for the Nile Basin.

The Nile Basin’s colonial-era treaties and the 2024 Cooperative Framework Agreement (CFA) shape water-sharing disputes. While six upstream nations ratified the CFA, Egypt and Sudan’s non-participation underscores unresolved tensions. Earth observation data (e.g., Sentinel-2, DE Africa’s WOfS) are critical for transparent monitoring of dam impacts.

2.1.1. GERD

The GERD is Africa’s largest hydropower project, with a storage capacity of 74 billion m³ (Figure 1 and Table 1). The dam has a 145-m height and 1.8-km length, enabling significant flow regulation and is located near the Sudan border. The reservoir spans 1874 km², altering local hydrology and sediment transport. Ethiopia emphasizes GERD’s role in hydropower generation (15,759 GWh/year) and flood mitigation, while downstream nations highlight risks of reduced flow during filling phases (Figure 1 and Table 1). The study area encompasses a diverse range of geographic features, including rugged terrain, plateaus, and river valleys. It spans the Ethiopian highlands, which serve as a crucial water catchment area for the Nile River Basin [30].

GERD holds immense hydrological significance, as it directly influences the flow and distribution of water resources within the Nile River Basin [30,31]. The Blue Nile, originating from Lake Tana in Ethiopia, contributes a substantial portion of the Nile’s water volume. With the construction of the Ethiopian Renaissance Dam, the study area has become a focal point for water resource management, presenting both opportunities and challenges for downstream countries [32,33,34].

Ethiopia has made repeated efforts to assure that the construction of the GERD will not lead to a reduction in downstream water flow. They have argued that the choice of the dam’s location was made with consideration for the downstream countries, as the dam’s primary purpose is to generate hydroelectric power with minimal water consumption for irrigation purposes. However, the focus of the dialogue has shifted over the past five years from Ethiopia’s considerations regarding site selection to concerns about the dam’s size and the potential impact during the initial filling phase. The Ethiopian government maintains that the filling process will have minimal negative effects and, in fact, will benefit downstream riparian countries in terms of flood control, reduction of siltation, improved irrigation, and water conservation [35].

Table 1. GERD Properties [36,37].

Parameter	Description
Total storage volume	7.4 × 1010 m3
Height	145 m
Length	1.8 km
Area	1874 km2
Annual energy production	15,759 GWh/yr
Discharge rate	1547 m3/s
Purpose	Hydropower

Table 1. GERD Properties [36,37].

Parameter	Description
Total storage volume	7.4 × 1010 m3
Height	145 m
Length	1.8 km
Area	1874 km2
Annual energy production	15,759 GWh/yr
Discharge rate	1547 m3/s
Purpose	Hydropower

2.1.2. Roseires Dam

Sudan has implemented various water control structures and dams along the Nile River and its tributaries. Specifically, two dams, the Sennar Dam and the Roseires Dam, are situated on the Blue Nile River in Sudan. The Roseires Dam was initially constructed in 1966, and its height and volume were increased in 2013. The primary functions of the dam include power generation, flood control, and irrigation purposes. Currently, the dam consists of a one-kilometer-long concrete buttress dam with a height of 78 m, accompanied by a 24-km earthen dam on both sides, reaching a height of 48 m. The reservoir formed by the dam has a surface area of 627 square kilometers and a volume of 7.4 billion cubic meters. The dam’s vicinity to the Ethiopian border and sediment accumulation issues have necessitated a height increase in order to address these concerns. Additionally, there are plans to utilize over a million and a half hectares of land for irrigation purposes in the area surrounding the dam (

Table 2. Roseires dam properties.

Parameter	Description
Height	78 m
Area	290 km2
Annual energy production	1605 GWh/yr
Discharge rate (spillways)	694 m3/s
Purpose	Irrigation and hydropower

).

The Sennar Dam is located approximately 350 km south of Khartoum and has a height of 33 m and a length of 3019 m. It forms a reservoir that stretches for 80.5 km. The dam became operational on 15 July 1925, initially providing irrigation for 122,000 hectares of land on the Gezira plain, as documented in the British Medical Journal on 16 January 1926. Over time, the irrigated area expanded to 860,000 hectares by the 1990s, with cotton being the primary crop.

2.2. Study Period

The analysis of water extent for the two dams was conducted over a 5-year study period, specifically from 2018 to 2022. This timeframe was selected to account for the pre-dam effect by including data from 2018–2019, which is prior to the commencement of the GERD operation. Additionally, the years 2020 to 2022 were chosen to capture the three distinct phases of the GERD filling process. This filling process began in 2020 and continued annually until 2023. By considering this 5-year period, the study aims to comprehensively analyze the changes in water extent associated with the GERD operation and filling phases and its possible effect on downstream dam.

2.3. DE Africa’s ODC and WOfS

DE Africa offers a regular, dependable, and functional service that empowers African nations to monitor changes across their countries and the continent with an unprecedented level of detail (this can be found at DE Africa). This service provides valuable insights into a wide range of issues, such as flooding, droughts, soil and coastal erosion, agriculture, forest cover, land use and land cover changes, water availability and quality, and transformations in human settlements [38].

DE Africa utilizes technology and services developed in Australia through Digital Earth Australia, adapting them to create a continental-scale platform and program. The primary goal is to democratize the capacity to process and analyze satellite data, making it accessible to a broader audience. Similar to the functioning of a weather service, DE Africa aims to provide routine decision-ready products and services that assist in informed decision-making processes.

Water Observations from Space (WOfS) is a continental-scale service provided by DE Africa that enables users to understand water availability and trends across Africa. It allows mapping, assessment, and visualization of surface water, providing valuable insights into water-related issues. By translating satellite imagery into easily understandable information, WOfS supports the UN’s Sustainable Development Goal 6 of ensuring the availability and sustainable management of water resources. It aids in improving access to safe drinking water, addressing water-related disasters, and facilitating better governance and management of water resources. The data-driven mapping and analysis tools offered by WOfS contribute to informed decision-making, sustainable land and water use, and effective water resource planning and management.

2.4. Sentinel Satellites

Sentinel satellites are being increasingly utilized to enhance the monitoring of water masses, such as lakes and water bodies [39,40,41]. These satellites are part of the European Union’s Copernicus program and provide high-quality Earth observation data, including both optical and radar data. The services are developed based on the utilization of LANDSAT and SPOT wavelengths. Specifically, the Sentinel-2 (S-2) mission consists of a constellation of two satellites, S-2A and S-2B. These satellites were launched on 23 June 2015, and 7 March 2017, respectively, and are positioned in the same sun-synchronous orbit. Importantly, they are in phase with each other at a 180° separation. The types of data used and their respective sources are presented in

Table 3. Data types used, along with their sources and links for this study.

Data Type	Source	Links
Sentinel-2	European Union’s Copernicus program	Sentinel-2
Rainfall data	NOAA climate data records	PERSIAN-CDR
Evapotranspiration data	Food and Agriculture Organization (FAO)	WaPOR evapotranspiration

.

2.5. Data Processing Environment

The processing of the image is performed by using DE Africa’s Sandbox platform. This platform utilizes a web-based interactive programming environment called JupyterLab, which incorporates algorithms known as “Jupyter notebooks”. These notebooks employ python packages and functions to carry out analysis tasks. The images are stored in the Cloud Optimized Geotiff format, enabling efficient retrieval from the cloud. Additionally, they are accompanied by metadata that adheres to the Spatio-Temporal Asset Catalog specification.

This ensures maximum discoverability and compatibility with other spatial datasets. The datasets are indexed within DE Africa’s Open Data Cube (ODC) database and can be accessed through the ODC interface [42,43,44].

2.6. Data Processing and Water Mapping

The processing of the sentinel data follows the following steps:

• Important python functions and packages for image processing are loaded in the notebook
• Initiate the datacube database to enable the loading and visualization of stored Earth observation data.
• The “load_ard” function applies a mask to remove low-quality pixels, retaining images with a majority of good pixels.
• Reliable images are reconstructed using a geomedian algorithm, considering a specified percentage.
• A Boolean function (“True” and “False”) is utilized to process water observations in the images.
• This function extracts pixels recognized as “Water” or “Non-water,” enabling water surface classification and tracking over time.
• The calculation of the Modified Normalized Difference Water Index (MNDWI) is then performed to evaluate water presence and characteristics.

2.7. Validation

The findings of this study were rigorously validated against climatological data and GERD’s filling dynamics, ensuring the accuracy and reliability of the assessments. Rainfall data, crucial for understanding the regional hydrological context, were retrieved from the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks—Climate Data Record (PERSIAN-CDR). There are numerous works in the literature that validate and use the PERSIANN-CDR for hydrological and climatological research [45,46,47,48,49,50]. This dataset is widely recognized for its long-term (1983–present), global, daily precipitation records at 0.25-degree spatial resolution, and has been validated in multiple regions and contexts.

To further strengthen our findings, we incorporated evapotranspiration (ET) data for both dams into our validation framework. This additional parameter helps account for water loss through atmospheric processes, providing a more comprehensive understanding of the water balance dynamics in the study area [51]. By analyzing ET patterns alongside precipitation data, we could better distinguish between climate-driven water fluctuations and those attributable to dam operations. This multi-parameter validation approach enhances the robustness of our conclusions regarding the hydrological impacts of GERD’s filling phases on downstream water resources. The ET analysis also helped verify whether observed water extent changes aligned with expected evaporation patterns given the regional climate conditions [52].

3. Result and Discussion

3.1. Result

In 2018 and 2019, the water extent of the Blue Nile remained constant. This was because the river followed its natural course before the filling of the Grand Ethiopian Renaissance Dam (GERD) commenced. However, starting in 2020, the river began to accumulate water due to the initiation of the first filling phase of the GERD, as announced by the Ethiopian government. The filling of the dam took place over four rounds, utilizing the seasonal availability of rainfall. From 2020 to 2022, during the first three rounds of GERD filling, the water extent of the dam steadily increased (Figure 2).

The initial stage of filling the reservoir started in July 2020 and, by August 2020, the water level rose to 540 m, which is 40 m higher than the riverbed situated at an elevation of 500 m above sea level. The second phase of filling was finished by 19 July 2021, reaching a water level of approximately 575 m. Subsequently, the third filling concluded on 12 August 2022, raising the water level to 600 m. Finally, the fourth filling was completed by 10 September 2023, resulting in water levels reaching around 625 m [53,54,55,56].

The water spread area of Roseires dam in Sudan experienced a decrease during this period, primarily attributed to the filling processes (Figure 2). The annual observed area of both dams was also assessed. As depicted in Figure 3, the annual water spread area for GERD remained consistent from 2017 to 2019 due to the absence of water infrastructure, such as a dam capturing the flowing river. However, during the first filling of the dam in 2020, the water spread area increased from approximately 80 km² to around 235 km². Subsequently, it further expanded to about 310 km² in 2021, indicating a total water gain of approximately 75 km² during the second filling period. In 2022, during the third filling period of GERD, the water occupied a total area of 528 km².

In contrast, Roseires dam in Sudan exhibited a general decreasing trend before and after the operation of GERD. The water spread area was around 474 km² in 2017, which then decreased to 469 km² in 2018. From 2018 to 2019, the area remained relatively constant at about 469 km². However, during the first filling period of GERD in 2020, the area decreased to approximately 466 km². The area remained similar in 2021 and, during the third filling, the water area dramatically decreased from 466 to about 460 km² (Figure 3).

Figure 4 illustrates the comparative water extent area of the two dams during the initial two filling periods. Prior to the construction of GERD, the water area for GERD was approximately 32 km², while Roseires dam had a much larger pre-dam water area of 474 km². After the operation of GERD and the first two fillings, the water surface area of GERD expanded to around 284 km², gaining approximately 252 km² of additional area.

In the same timeframe, Roseires dam experienced a decrease in water extent area of approximately 5 km², while gaining around 3 km². However, during the third filling period, both GERD and Roseires dam underwent significant changes in water extent area. The water area of GERD expanded from approximately 284–528 km², resulting in a total gain of 244 km². In contrast, Roseires dam experienced an increased water loss, going from 5 to 10 km² (Figure 4 and Figure 5)

The variation of evapotranspiration for the study areas is shown in Figure 6 below. Between 2018 and 2023 the average actual evapotranspiration at both GERD and Roseires Dam followed a broadly similar arc: an initial rise, a mid-period peak, and a dip in 2022. GERD consistently showed higher values—roughly 350–400 mm above Roseires each year—indicating greater atmospheric water loss in its catchment. GERD dipped slightly from 2018 to 2019, surged to a maximum of about 1250 mm in 2021, then eased back to roughly 1130 mm in 2022 and reached its local maximum of 1257 mm in 2023. Roseires climbed from about 750 mm in 2018 to 830 mm in 2019, slipped to 800 mm in 2020, reached near 840 mm in 2021, and fell to roughly 790 mm in 2022 and which then rose to its peak, 991.7 mm in 2023. The synchronized peaks and subsequent declines suggest common regional climatic influences.

3.2. Discussion

The water spread area of the GERD continually increased, while that of Roseires dam decreased during the study period, with a strong negative correlation factor of −0.91. This can be attributed to many factors, like climatology, the water filling and releasing policy implemented, the starting conditions of the reservoirs, etc., even though it remains that the main factor is thought to be the mega project of Ethiopia, i.e., GERD operation [9,11,57,58]. The potential impacts of the GERD on downstream water resources were discussed by [9], providing various filling strategies for the dam and their consequences on water supply and power generation in the downstream countries. The analysis indicates that the risks associated with water diversions in Sudan can be effectively managed by adapting Sudanese reservoir operations. To mitigate risks to Egyptian users and energy generation, the article suggests implementing agreed annual releases from GERD, adopting a drought management policy for the High Aswan Dam, and establishing a basin-wide cooperative agreement to safeguard the elevation of Lake Nasser.

Figure 7 below shows the average yearly and monthly rainfall for both dams from 2017–2023. As is depicted in the figure, the average yearly rainfall for both dams saw a decline from 2017–2018 and then sharp increase from 2018–2020 for two consecutive years. Both areas then experienced a sharp decrease in average yearly rainfall from 2020–2021 and this almost remained stable from 2021–2022.

The correlation between average yearly rainfall and the water extents of the two reservoirs can be summarized as follows:

• The reduction in the water extent of the Roseires dam from 2017 to 2018, a period without the presence of GERD, can be largely attributed to decreased rainfall in both areas and other climatic elements, such as evaporation (Figure 3 and Figure 6).
• The decline in the water extent of the Roseires dam from 2019 to 2020, coinciding with a significant rise in average rainfall, is more likely linked to the initiation of the first filling of GERD and other climatological factors.
• The decrease in the water extent of the Roseires dam from 2020 to 2021, during a period of notable rainfall decrease, can be associated with the commencement of the second filling of GERD and the prevailing rainfall conditions.
• Despite a substantial expansion in the water spread area at GERD during its third filling phase from 2021 to 2022, with consistent rainfall in both regions, the water spread area at the Roseires dam decreased. This suggests the influence of GERD operations and the presence of additional factors impacting the water extent at the Roseires dam.

According to [58], the GERD filling stages have negative impacts on downstream water budgets and this study recommends reutilizing water through several appropriate approaches, reorganizing the distribution of water, improving water transportations and eliminating losses by lining irrigation canals and drainage canals, making evaporative condensers to utilize it, and increasing abstraction of groundwater, while taking into account the necessary precautions. The study by [59] analyzed the changes in water surface coverage or water extent in the Blue Nile River associated with the construction and operation of the Grand Ethiopian Renaissance Dam (GERD). The findings reveal significant changes in the water surface coverage area, particularly following the filling stages of the GERD.

According to Figure 3, the water surface area of the Roseires Dam decreased prior to the operation of GERD. This decrease in water extent can potentially be attributed to climate change, as documented in [58] and the increase in evapotranspiration from 751–828 mm in 2018 and 2019, respectively, and other factors, like sedimentation [60].

Furthermore, during the period 2020–2022, it is noted that the water extent of the Roseires Dam continued to decrease. This decrease is largely attributed to the filling stages of the GERD, as mentioned in [58,61] and other factors like sedimentation, as there is no significant loss because of evaporation. This implies that the construction and operation of the GERD have had an impact on the water extent of the Roseires Dam, leading to a further decrease in its water surface area.

The analysis by [13,53,62] also supports the findings of this study for Roseries dam’s water extent fluctuation during this study period. The proposed filling strategies of the GERD were retrieved for a period of three years using the best-performing simulation of CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data) for the study mentioned. The analysis of the filling strategies revealed that the GERD stored a certain percentage of the monthly inflow during specific periods. Specifically, in July 2020, the GERD stored 14% of the monthly inflow, while in July 2021, it stored 41%. In July and August 2022, the GERD stored 37% and 32% of the respective monthly inflows. On an annual basis, the GERD retained 5.2% and 7.4% of the annual inflow during the first two filling phases and between 12.9% and 13.7% during the third phase [53]. Indeed, the findings mentioned in the research support the observations made in the present study. The present study indicates that, during the first and second phases of filling, there was little change in water extent. However, during the subsequent phases, a significant fluctuation in water extent was observed in both the GERD and the Roseires Dam. This aligns with the documented increase in storage capacity and the subsequent impacts on water extent due to the filling stages of the GERD.

The relationship between surface water extent, rainfall, and actual evapotranspiration (ET) reflects the combined effects of climatic conditions and dam operations. Rainfall acts as the primary water input, while ET represents atmospheric loss. Surface water extent results from the balance between these two, modified by human interventions such as the GERD filling.

For instance, the decline in Roseires Dam’s surface water from 2018 to 2019 occurred before GERD operation and coincided with low rainfall and rising ET—from 751 mm to 828 mm—indicating climate-driven impact. However, from 2020 onward, despite increased rainfall, Roseires continued to lose water while GERD’s extent expanded, highlighting the influence of upstream water retention. Even in 2022, with relatively stable rainfall and ET, Roseires’ area declined as GERD entered its third filling phase.

This integrated analysis confirms that surface water variability is influenced not only by natural factors but also by upstream regulation, reinforcing the need for coordinated transboundary water management.

3.3. Implications for Transboundary Water Governance in the Nile River Basin

The hydrological impacts of GERD’s 500 km² water extent gain and Roseires Dam’s 9 km² loss over 2018–2022, derived from DE Africa-WOfS data, highlight the need for robust transboundary water governance in the Nile River Basin. These findings reflect the challenges in managing shared resources amid historical inequities, geopolitical tensions, and climate pressures. Effective governance, supported by the Cooperative Framework Agreement (CFA) ratified by six upstream states in 2024, requires enhanced data sharing, multi-national decision support, and institutional cooperation [63].

3.3.1. Data Sharing and Transparency

DE Africa’s open-access EO data, used to quantify GERD’s expansion from 80 km² to 528 km² and Roseires Dam’s reduction from 474 km² to 460 km² (Figure 3), enables transparent monitoring of transboundary water dynamics. Such transparency is vital to build trust, given mistrust from colonial-era treaties [64]. Standardized protocols for sharing EO-derived data, aligned with the CFA’s call for information exchange [16,65], could ensure equitable access via a centralized repository, reducing disputes during GERD’s filling phases when Roseires Dam lost about 6 km² (2021–2022).

3.3.2. Multi-National Decision Support

The 6 km² water loss at Roseires Dam during GERD’s third filling phase underscores the need for multi-national decision support systems. The NBI’s Decision Support System (NB-DSS) can integrate EO data and AI-driven models to forecast impacts and optimize water allocation [25,43]. Cooperative filling strategies, as proposed by Wheeler et al. (2020) [3], could mitigate downstream effects, supporting the CFA’s equitable utilization principle. Such systems would address climate-induced flow reductions of 5–15% by 2050 [66].

3.3.3. Institutional Cooperation

The NBI facilitates cooperation, but Egypt and Sudan’s non-ratification of the CFA limits its authority [19]. A permanent Nile River Basin Commission, as envisioned by the CFA, could institutionalize joint monitoring and adaptive management, leveraging EO data to track changes (Figure 3, Figure 4 and Figure 5) [21]. Benefit-sharing, like GERD’s energy trade, aligns with CFA goals and could offset Roseires Dam’s losses [18].

3.3.4. Broader Implications

The GERD–Roseires dynamics reflect Nile Basin governance challenges. The CFA’s framework, combined with DE Africa’s data, NB-DSS, and a strengthened NBI, can foster equitable, sustainable management, aligning with UN Watercourses Convention principles and SDG 6.

The interplay between GERD’s water gain and Roseires Dam’s loss reflects the broader challenges of managing shared water resources in the Nile Basin, where historical inequities and climate pressures exacerbate tensions. Benefit-sharing models, such as energy trade from GERD to downstream states, could offset water losses and foster mutual gains, as proposed by [61,67]. The study’s reliance on DE Africa’s EO data underscores the potential of technology to support transparent, data-driven governance. By establishing transparent data-sharing mechanisms, multi-national decision support systems, and robust institutional frameworks, riparian countries can mitigate the adverse impacts of large-scale projects like GERD, while maximizing shared benefits, such as flood control, hydropower generation, and irrigation efficiency. These strategies align with the principles of equitable and reasonable utilization and no significant harm, as outlined in the UN Watercourses Convention [23,42,68], and are critical for achieving Sustainable Development Goal 6 in the Nile Basin.

3.4. Significance and Limitations of the Study

This study provides critical insights into the transboundary hydrological dynamics of the Nile River Basin, particularly focusing on the impacts of the GERD on downstream water bodies such as the Roseires Dam. By leveraging EO data from DE Africa and its WOfS platform, this research contributes to the growing body of data-driven transboundary water governance approaches. The innovative use of EO data cubes allows for consistent, spatially comprehensive, and temporally rich analysis of water extent changes—offering transparency and actionable information to policymakers and stakeholders involved in Nile Basin management.

However, the study has several limitations. First, the analysis is based solely on surface water extent and does not account for subsurface hydrological changes or water quality parameters, which may also influence downstream water availability. Second, the attribution of water extent changes to specific factors such as GERD operations or sedimentation remains partially inferential, as disentangling their individual contributions requires coupled hydrological and socio-political models. Additionally, while the DE Africa platform offers high temporal frequency, cloud coverage and satellite data gaps may influence classification accuracy during certain months.

Future research should integrate hydrological modeling, in situ measurements, and socioeconomic datasets to provide a more holistic assessment of transboundary impacts. Moreover, extending the analysis to include other downstream infrastructures, such as Sennar and Merowe Dams, could offer a broader understanding of basin-wide impacts, enhancing regional cooperation and adaptive management strategies.

4. Conclusions and Recommendations

4.1. Conclusions

This study demonstrates the significant hydrological impacts of the GERD downstream water resources, particularly Sudan’s Roseires Dam, using Digital Earth Africa’s Earth Observation Data Cubes and WOfS platform. From 2018 to 2022, GERD’s water extent expanded from 80 km² to 528 km², while Roseires Dam experienced a reduction of about 9 km² loss during the same period. The decrease in Roseires Dam’s water extent from 2018 to 2019, prior to GERD’s filling, can be attributed to increased evaporation (from 751 mm to 828 mm) and other factors such as sedimentation. These changes, validated against PERSIANN-CDR rainfall data, are primarily driven by GERD’s filling operations post-2020, compounded by climatic factors like evapotranspiration and fluctuating rainfall, sedimentation, etc. The findings highlight the critical role of Earth Observation (EO) technologies in providing transparent, evidence-based data to monitor transboundary water dynamics. Despite the Cooperative Framework Agreement (CFA) ratified by six upstream states in 2024, the lack of full adoption by Egypt and Sudan underscores persistent mistrust and governance challenges. The study emphasizes that EO-driven data, integrated with the Nile Basin Initiative’s Decision Support System (NBI-DSS), can foster cooperative water management, aligning with the CFA’s equitable utilization principles and the UN Watercourses Convention. The GERD–Roseires dynamics reflect broader Nile Basin challenges, where benefit-sharing models, such as energy trade, could mitigate downstream impacts and promote sustainable development.

4.2. Recommendations

• Enhance Data Sharing and Transparency: Establish standardized protocols for sharing EO-derived data through a centralized repository, as supported by the CFA, to build trust and reduce disputes during GERD’s filling phases.
• Strengthen Multi-National Decision Support: Integrate EO data and AI-driven models into the NBI-DSS to forecast impacts and optimize water allocation, enabling cooperative filling strategies to mitigate downstream effects.
• Promote Institutional Cooperation: Establish a permanent Nile River Basin Commission to institutionalize joint monitoring and adaptive management, leveraging EO data to track water extent changes.
• Implement Benefit-Sharing Mechanisms: Promote energy trade from GERD to offset downstream water losses, fostering mutual gains and aligning with CFA goals.