Shortcomings in the water release procedures during peak flood periods.

(Dan Tri Newspaper) - The record-breaking flood release at the Song Ba Ha hydropower plant once again exposed the paradox of "following the correct procedures, yet the downstream areas are still deeply flooded."

On the afternoon of November 19th, the Song Ba Ha hydroelectric reservoir released floodwaters at a rate of approximately 16,100 m³/second, a level that many experts consider the highest ever recorded in the history of hydroelectric power plant operation in Vietnam.

The massive amount of water that poured downstream in a short period of time caused severe flooding in many areas of Dak Lak and the provinces further downstream, disrupting transportation and causing heavy damage to people's homes and property.

Many argue that the operating unit claims the flood release was carried out according to the approved procedures. However, the fact that "the procedure was followed correctly, yet downstream areas were still deeply flooded" once again raises questions about the appropriateness and effectiveness of the current operating procedures.

Speaking with a reporter from Dan Tri newspaper, Associate Professor Dr. Ngo Anh Quan - Deputy Director of the Institute of Hydraulic Engineering, Vietnam Academy of Water Resources Science, emphasized that the problem lies not only in the "operating process," but also depends on many other objective and subjective factors.

These include the permissible operating margins, the reliability of input data, and the coordination mechanisms between reservoirs within the same basin.

According to him, the most important thing is to quickly establish a coordinating body at the river basin level, with sufficient authority and capacity, to ensure that the operation of dams and reservoirs is more synchronized, safe, and efficient.

The "proper procedure" margin

Looking at the recent flood release at the Ba Ha River, as well as many other floods, what, in your opinion, is the core problem in the current reservoir operation, where all units claim to be following "the correct procedures," yet the situation of flood peaks coinciding with the time when the dam has to release water at full capacity keeps recurring?

- In principle, current reservoir operation procedures specify quite clearly the water level thresholds and discharge rates corresponding to each period. Reservoir owners and local management agencies base their operations and decisions regarding flood discharge, water storage, or flow regulation on these parameters.

Assoc. Prof. Dr. Ngo Anh Quan - Deputy Director of the Institute of Hydraulic Engineering, Vietnam Academy of Water Resources Science (Photo: Minh Nhat).

However, in practice, these processes do not fixate on a single value but allow for operation within a flexible range.

The reservoir manager can choose to release water early or late within the permitted time frame; they can also maintain the water level at low or high levels within the specified range.

The problem arises when this buffer zone is designed too wide: under conditions where rainfall and flooding forecasts are assessed as not being severe, operating units tend to retain water longer to optimize power generation, leading to the risk of imbalance when hydrological conditions change suddenly.

When rainfall or inflow forecasts for the reservoir are inaccurate, attempting to "maximize water retention" within permissible limits can lead to adverse situations: when forced to release water to ensure the safety of the structure, a large volume of water will rush downstream in a very short period of time.

On paper, the operating unit can still prove that they followed the correct procedures; but in reality, people in the downstream areas had to endure a sudden surge in floodwaters, causing unpredictable damage.

It is worth noting that this risk does not stem from intentional violations but from the design of an excessively wide margin, while the forecasting system and tools to support the selection of the optimal solution within that margin are not yet accurate or timely enough.

Therefore, it can be understood that the current process has wide margins and lacks supporting analytical tools to enable both the government and hydropower operators to make more accurate decisions, thereby minimizing the risk of flood releases coinciding with peak flood levels?

- That's right. I would like to highlight some shortcomings in the way we currently manage the operation of hydroelectric dams, and also areas that need improvement to ensure the safety of the structures and minimize risks to downstream areas.

Firstly, the current allowable discharge time and flow rate ranges are designed to be relatively "safe" for the reservoir, meaning that as long as they do not compromise the structural integrity of the reservoir, they meet the requirements.

However, in practice, these boundaries need to be more closely linked to downstream flood scenarios. When decision-makers can clearly see: if option A is chosen, how high the water level will rise at each downstream point, for how long, and what the expected damage will be, then the decision made will be more realistic and reduce the risk of creating unfavorable flood peaks.

Secondly, the economic benefits of water storage for electricity generation have not yet been systematically compared with the costs of flood damage. If we had flood maps and models estimating damage corresponding to each release option, we could quantify the difference between increased electricity revenue and increased risk to the population.

This is an important basis for making more balanced decisions between economic benefits and social safety.

The Song Ba Ha hydroelectric reservoir releases floodwaters (Photo: Trung Thi).

Currently, these comparisons are still quite vague because operational scenarios have not been fully developed and flood release options have not been systematically analyzed by specialized agencies for each specific situation.

Therefore, within the permissible limits, the decision to release floodwaters in an emergency has not been truly optimized to both ensure the safety of the structure and minimize damage to people in downstream areas.

If we continue to improve analytical tools, update flood maps, and develop multi-objective operational scenarios, decision-making will become increasingly accurate, proactive, and substantive, thereby best serving the lives of the people.

The person signing the order needs to know how far this level of flooding will occur.

So, in your opinion, how should these gaps be narrowed to reduce the risk of "following the correct procedures but still getting deeply involved"?

- I believe we need to shift from a "safety boundary for the reservoir" mindset to a "safety boundary for the entire river basin" mindset. This means that each operational boundary must be directly linked to a downstream flood scenario, instead of just requiring "ensuring water levels do not exceed thresholds A and B."

Currently, many regulations only specify water levels and discharge rates, but fail to clearly answer practical questions such as: if this option is chosen, which downstream areas will be flooded, how deep will the flooding be, and for how long will it last?

The area where the Song Ba Ha Hydropower Plant is located (red dot), and the downstream area that may be affected (blue dot) (Image: Reproduced from a map of the Vietnam Surveying, Mapping and Geographic Information Agency).

Each discharge level in the process should be associated with a set of flood maps and a brief description of the downstream impact.

With this data, the person issuing the order not only looks at the numbers in the reservoir but also visually sees the expected consequences below, making decisions within the boundary range that are more cautious, more realistic, and reduce the risk of exacerbating the flood peak.

The operational margin needs to be gradually narrowed as data systems and forecasting capabilities improve. In the initial stages, when data is incomplete, we can accept a wide margin to maintain the safety of the structure.

However, with a denser monitoring network and more accurate forecasting models, we have a solid basis to refine the process, narrow the "discretionary" zone, and create a more transparent and efficient operating framework.

In process design and boundary construction, socio-economic factors must be considered from the outset. It is impossible to optimize for power generation benefits while ignoring the costs of flood damage.

If we have models of damage corresponding to each discharge level, we can make a relative comparison between the additional revenue generated from water storage for power generation and the estimated risks and costs of damage if downstream water levels exceed a certain threshold.

Once this picture is clearly quantified, accepting earlier discharge, sacrificing some electricity production but significantly reducing risks to residential areas, becomes much more reasonable, transparent, and persuasive.

All of the above solutions, if implemented synchronously, will help to make reservoir operation more proactive, scientific, and geared towards the highest goal: protecting the safety of the people and ensuring sustainable development of the entire river basin.

Data gap

To optimize operations, as you said, the input data must be very good. What is the current state of the monitoring and forecasting system, sir?

- In terms of regulations, we already have fairly clear standards for the placement of rain gauges, water level gauges, and flow rate gauges in each river basin; there are regulations on the types of equipment and minimum requirements that must be met. Many construction projects have also installed monitoring systems in accordance with these standards.

According to Associate Professor Quan, another difficulty is that monitoring data within the same river basin is sometimes scattered due to different management units (Photo: Minh Nhat).

However, from the perspective of serving precise and real-time operations, the current system still has quite a few limitations. In some river basins, the density of measuring stations is not high enough to accurately describe the spatial distribution of rainfall; many devices are old and no longer reliable, while climate change is causing floods to occur more quickly and extremely.

Another difficulty is that monitoring data within the same watershed is sometimes scattered due to different management units. The mechanisms for connecting and sharing information are still not clearly defined, making it difficult to aggregate data and develop unified operational scenarios.

If we standardize and synchronize data systems early on, invest in equipment upgrades, and build a seamless information sharing mechanism between parties, the quality of input data will be significantly improved. This is a crucial foundation for narrowing operational margins, increasing proactiveness, and reducing risks for downstream areas.

According to him, to improve the quality of data used for dam operation, what kind of investment and reorganization of the monitoring system and data sharing mechanism are needed to ensure safer and more efficient operation?

- Technically, we need to move towards a more modern generation of monitoring stations, equipped with sensors that measure rainfall, water levels, and flow rates in real time, and continuously transmit data to a central hub.

Based on that data, it is entirely possible to integrate big data analytics software and mathematical models to simulate rainfall and floods, predict flooding, and support decision-making in specific situations.

Alongside investing in equipment, we need a unified policy framework for hydrological data.

Within this policy framework, the State must clearly define: which data is mandatory shared data; and which data can be provided as a service at an appropriate cost.

Monitoring stations funded by the state budget can provide data to businesses; conversely, businesses that install stations within their projects also have an obligation to share data with management agencies, especially in situations of heavy rain and floods where timely information is crucial.

Currently, the Ministry of Agriculture and Environment has issued many important documents related to the operation of interconnected reservoirs during the flood season, and has assigned the task of building an information system and mathematical models to support the regulation and distribution of water in major river basins, aiming for real-time operation. These are correct directions that need to be further promoted.

The key for the next phase is to concretize these directives by establishing a synchronized network of monitoring stations, an interconnected database, and unified modeling tools for each river basin; thereby expanding implementation from a pilot scale to a large-scale operation.

Once the data infrastructure and models are perfected, reservoir operation will become increasingly scientific, transparent, and efficient, contributing to minimizing risks for the people and maximizing the benefits of water resources.

There needs to be an agency with sufficient authority to coordinate the operation of hydropower plants.

Besides the technical issues mentioned above, in your opinion, are there any other shortcomings in the current management and operation of hydroelectric dams?

- Rivers do not flow along administrative boundaries. A reservoir located in one province may release water that causes flooding downstream in another province. Within the same river basin, numerous hydroelectric, irrigation, and domestic water reservoir projects may be involved in water storage and release.

Ban Ve Hydropower Plant operates flood discharge procedures (Photo: Linh Chi).

Currently, when floods occur, hydropower plants report to relevant ministries, departments, and the province where the project is located; the parties exchange information and consult, and then the Chairman of the Provincial People's Committee makes a decision. This approach may still be heavily bureaucratic, while flood flows follow the hydrological patterns of the entire river basin, independent of administrative boundaries.

Globally, basin-based management models have been widely adopted. The Mekong River Commission is a prime example, where multiple countries consult with each other before implementing projects that could impact the shared river.

In the field of irrigation, we also have the Department of Irrigation Construction Management coordinating irrigation water supply on a basin-wide scale, rather than on a provincial basis.

What we need now is the next step: forming a truly effective coordinating body, with clearly defined functions, responsibilities, and resources.

At the large basin level, one can envision a Committee or Center for Dam and Reservoir Safety Coordination, which would not replace the role of the investor or the Provincial People's Committee, but would perform several key tasks: Building and operating a shared data system for the entire basin.

This includes observational data, meteorological and hydrological forecasts, flood maps, and technical information on structures.

Maintaining and regularly updating mathematical models is crucial for providing operational recommendations for individual reservoirs or groups of reservoirs under various scenarios. In situations like the Ba Ha River crisis, this agency must act as the central point for data aggregation, quickly calculating scenarios, and submitting recommendations along with risk assessments to the decision-making body.

This agency also serves as the focal point for long-term strategic consultation for the entire basin: proposing adjustments to the inter-reservoir operating procedures, determining investment priorities for upgrading monitoring systems, or issuing warnings when downstream land-use planning risks encroaching on flood drainage areas.

Without such a coordinating body, each incident would be handled in a " каждый себя" (every man for himself) manner; procedures might be adjusted locally, but the overall risk to the entire basin would not be reduced much.

Thank you for the conversation!

Source: https://dantri.com.vn/khoa-hoc/bat-cap-khoang-bien-trong-nhung-lan-xa-nuoc-dung-quy-trinh-giua-dinh-lu-20251211121539371.htm