The spillage of water from Akosombo dam and the path forward

Water management, especially in dammed rivers, is always a delicate balance between conservation and safety. A sudden unanticipated spill from a dam, do not only underscores the need for robust forecasting systems but also underlines the potential consequences of being unprepared.

The primary consequence of an unanticipated spill is flooding. Without warnings, communities downstream are left vulnerable, homes can be submerged, agricultural lands ruined, and in the worst cases, lives can be lost.

Beyond these immediate physical consequences, there are lasting socio-economic ramifications as displaced communities, damaged infrastructure, and increased health risks could compound the crisis. The unprepared uncontrolled spillage can’t just be a failure of a system, but perhaps a sequence of overlooked events and underestimated risks.

An article written by Michael Agbenyegah of the Volta River Authority, who until his demise, was also the head of their Survey Unit on “Monitoring dam spills with geospatial technology.”

In that article, the author held that VRA has in 2010 developed an Emergency Preparedness Action Plan (EPP) and an Environmental Management Plan (EMP) for both Akosombo and Kpong Dam to mitigate the impacts caused by the existence and operation of the dams during emergency situations such as flooding during dam spilling, to keep the loss of life and property to the barest minimum.

If this was truly so, then VRA, like the proverbial Ostrich, would have gone back to sleep since the spillage in 2010 or else this was just research on some activities which ought to have been done such as deploying Geospatial technology to accurately analyze and delineate areas likely to be affected by flooding during dam spilling or dam break!

According to records in the Daily Graphic News report on Wednesday, 20th October 2010, the Volta Dam reached a level of 274.80 feet close to the maximum capacity of 278 feet which led to the execution of a controlled spillage.

Whiles controlled spillage of excess water was being carried out, the water level kept rising until it reached an all-time highest level of 277.54 feet on 8th November 2010. This necessitated the option of uncontrolled spillage and hence its six flood gates were opened causing some flooding in communities downstream. It was also reported that the year 2010 was the time highest level has been recorded by the dam within the last 20 years. Records have further revealed that, there have been spillages in 1966, 1967, 1968, 1969, 1970, 1971, 1972, 1974, 1991.

However, the Kpong Dam was constructed between 1977 and 1982 to be fed by water from Akosombo, hence the lack of controlled spillage between 1974 and 1991. Although it cannot be overlooked that VRA always starts with controlled spillage, the timing is always wrong and very late.

However, in order to prevent the water from overtopping the dam, uncontrolled spillage is done.

Unlike the 2010 spillage, which took place after the rainy season, this year's spillage is happening in the rainy season.

This will make it more disastrous because a lot more water is expected in the Lake from its main tributaries (Black &White Volta, Oti and Afram rivers etc). There is also the increased inflow that may come in the case of the Bagre Dam spillage in Burkina Faso. We should therefore fasten our seat belts because more houses will not just be flooded but submerged if we still fail to prepare and take proper remedial actions.

VRA has being taking water level data since 1965, additionally, rainfall data over the same period is also available. The over 70 years of water level data combined with rainfall data in the Volta basin can be useful in modelling the trend of water rise for timely controlled spillage.

The data can also be simulated to show all areas likely to be inundated and the extent as well as properties and people that are likely to be affected in any future events.

While the past cannot be changed, the future can be prepared for to avoid similar occurrences.

First the vast Historical tide gauge data must be analysed to reveal the nuances of dam behaviour especially in the face of changing environmental conditions and climate.

Steps for remedy
The following roadmap is being proposed to curtail similar scenarios in the future:

1. Enhanced Data Analytics: With advancements in AI and machine learning, predicting dam behavior can be more accurate. By feeding systems more diverse data, from rainfall patterns to upstream river activities and land use land cover data, predictive models can be made more comprehensive.

2. Regularly scheduled checks and maintenance can ensure that dams function as expected and investing in new technology to provide better real-time data.
3. Communities living downstream need to be involved in disaster preparedness plans. Regular drills, education on evacuation procedures, and training on flood safety can equip them to handle emergencies.

4. Development of early warning systems, thus, efficient Communication such as automated SMS alerts to localized weather apps to ensuring that warnings reach people in the surrounding communities on time.

5. Engagement of experts such as Geospatialists, hydrologists, environmental scientists, and climate experts to provide insights into emerging risks.

6. Creating a predictive model and a hazard map using tide gauge data from the dam, topographic maps, and other relevant information that are essential tool for disaster risk reduction and early warning. This includes time-Series Analysis to help identify patterns in water height levels using techniques like ARIMA, Prophet, or LSTM to analyzing factors like rainfall, upstream water inputs, that might affect dam levels and performing threshold Analysis to identify when the spill gates need to be opened.

7. Preparation of Hazard Maps using Geospatial Techniques. This activity includes:

a. Flooding Simulation: Using the topographic maps/DEMs to simulate water flow downstream when the spill gates are opened.

b. Vulnerability Assessment: Identify areas with high population density, critical infrastructure, or other socio-economic factors that would make them particularly vulnerable to flooding.

c. Combine Information: Overlay the flood-prone areas with the vulnerability assessment to create a comprehensive hazard map showing regions at high, medium, and low risk.