Figure 4. The performance of the re-regulation reservoirs to re-regulate
the discharge for P (40%, 2) during 500 hours. (a) illustrates how the
reservoir limits the discharge to the thresholds defined in the
algorithm and achieves optimal conditions whenever possible. (b)
distinctly shows the ramping rates before and after implementing the
re-regulation reservoir.
Discussion
Re-regulation reservoirs (RRR) have emerged as a promising method for
mitigating the ecological impacts of hydropeaking. While the efficacy of
RRRs in mitigating hydropeaking impacts is still being assessed, early
studies suggest that this novel approach may be a valuable tool for
enhancing the ecological sustainability of hydropower operations.
Studies have shown that RRRs can reduce the adverse impacts of
hydropeaking downstream of hydropower plants and reduce the deleterious
effects of hydropeaking on aquatic ecosystems (Tonolla et al., 2017).
Compared to other mitigation measures, such as the installation of
downstream flow control devices or modifying the operation of hydropower
facilities, RRRs offer several advantages. These include greater
flexibility and adaptability to changing environmental conditions,
power, and water demand without increasing the operational cost of power
systems. While RRRs also possess the potential to provide other
services, such as water storage and flood control (Anindito et al.,
2019; Premstaller et al., 2017; Tonolla et al., 2017). In practice, RRRs
could be an artificial structure, natural water reservoir, flood plain
or any other storage possibility near by the HPP. Our study did not
prioritize the examination of structural or operational designs.
Instead, we focused on exploring theoretical approaches to mitigate
hydropeaking and developed a model to operate and estimate the required
RRR volumes.
This study tested and determined the RRR volumes required to meet the
objectives and priorities outlined in section 2.3 of the Kemijoki River
downstream from Taivalkoski HPP using a re-regulation algorithm
developed specifically for this study. The flow data of the Taivalkoski
HPP was scaled down to create a base case that can be used for other
similar HPPs. While we recognize that re-regulation is generally more
manageable in smaller river systems due to their smaller reservoir
capacities and volumes, our case data from a large river system can
still be leveraged to explore theoretical possibilities for
re-regulation practices. The results indicate that for most of the
tested permutations, the required volume of the RRRs increased as the
thresholds for peak and minimum hourly flows and the ramping rates
became more stringent (Figure 2). The observed outcome can be attributed
to two factors. Firstly, as the thresholds became more stringent, they
activated the conditions that enable water storage by the algorithm at
more time steps than those that facilitate water release. The flow
pattern controls the algorithm’s decision to store or release water.
Secondly, a more stringent threshold necessitates storing or releasing
larger quantities of water. Considering both factors, i.e., increased
water storage events and an increased water volume to be stored in a
singular event, led to the requirement of greater RRR volumes as the
thresholds became more stringent. Nonetheless, for some permutations,
this trend was not observed. One example is the needed reservoir volumes
for P (10%, 3.5) and P (10%, 4) were larger than the reservoir volume
needed for P (10%, 2.5) that had more stringent ramp rate threshold.
For this particular case, the water released in proportion to the water
stored for P (10%, 3.5) and P (10%, 4) was less efficient than that of
P (10%, 2.5) due to algorithms increased propensity to store water
rather than release water. As such, inadequate (i.e., too slow) water
release back into the waterway might increase the required RRR volume or
lead to small volume availability during high flows. Furthermore, if a
flow adjustment of 10% is deemed sufficient, it is more beneficial then
to consider permutation P (10%, 2.5) since it has more stringent ramp
rate thresholds while requiring a smaller RRR volume. Whereas, if the
daily peak and minimum flows are of greater concern than the ramp rates
due to the ecological needs of the river, it would be more efficient to
consider permutation P (20%, 4) than P (10%, 4) which requires a
smaller RRR volume. The increased flow adjustment threshold in P (20%,
4) activated the conditions that enable water release by the algorithm
at more time steps than those that allow water storage. Thus, resulting
in a reduction of the required RRR volumes when compared to P (10%, 4).
This highlights the significance of maintaining a proportional balance
between storing and releasing water. As such, choosing the optimal
re-regulation of reservoir volume is related to the river regime and
sub-daily flow patterns. These findings highlight the importance of
careful consideration of the unique characteristics of a given river and
its ecosystem when designing RRRs for hydropeaking mitigation.
To achieve greater efficiency, RRRs can be optimized by tailoring their
design to the ecological needs of the river and the desired mitigation
objectives. Hayes et al. (2019), proposed a mitigation approach that is
specific for the life cycle stages of fish. For instance, if the
objective is to improve the larval life stage of trout and grayling
rather than eliminating the stranding of fish, it might be more
efficient then to have less stringent ramp rate thresholds resulting in
the reduction of the required volume of RRRs. Another important factor
to consider when optimizing the reservoir is the RRR operation period.
It might be more efficient to adjust the threshold values and objectives
according to the seasons, days, or time steps that are deemed
significant to the ecological status of river rather than using constant
threshold values for the whole year. Furthermore, excluding periods with
less ecological significance from the operation period would reduce the
required volume of RRRs. For the case studied in this work, excluding
July and August from the operation period would be the most impactful on
the required RRR volume as the highest flow occurs in the late spring or
early summer. However, the re-regulation operation period should be
tailored according to the ecological needs of the river in order to
reduce the required RRR volume. Overall, these findings highlight the
potential for optimizing the design and operation of RRRs for
hydropeaking mitigation, based on the unique needs of the local
ecosystem and the desired mitigation objectives. By doing so, a greater
efficiency and a reduction in the required RRR volume, while minimizing
the impact of hydropeaking on downstream ecosystem can be achieved.
The model design in this work can be used to determine the required RRR
volumes for other HPP in any other river system of various sizes and
flows. However, the range of thresholds for hydropeaking should be
defined to best suit the flow pattern and river size. Furthermore, the
model could be enhanced by incorporating additional conditions into the
algorithm such as RRR operational period (season), timing of
hydropeaking event (day and night), water temperature fluctuations, and
sediment loads, while also accounting for water supply, irrigation, and
recreational needs. Additionally, it may be useful to investigate a
model with a RRR that is distant from the HPP. This would present the
opportunity to study corridors connecting HPPs with RRRs as additional
storage volumes. Also, a model that incorporates flow velocity and water
losses from the RRR would further enhance the understanding, use, and
optimization of RRRs.
Conclusion
We investigated in this work the capabilities of RRRs to limit flow
components to a range of thresholds, such as peak flow, minimum flow,
and ramping rates. The results indicate that as the thresholds become
more stringent, the required RRR volume increase. In certain cases,
however, this trend was not observed, indicating that the design of RRRs
should be tailored according to mitigation objectives and ecological
needs of river systems. The model developed in this study can aid HPP
operators, authorities, and researchers in designing, optimizing, and
rating the feasibility of introducing RRRs to river systems. However,
several avenues for future research can be pursued to further advance
the understanding, use, and optimization of RRRs. One promising area of
investigation is the development of more sophisticated models that
incorporate, RRR operational period (season), timing of hydropeaking
event (day or night), as well as water temperature fluctuations and
sediment loads. Finally, additional research is needed to better
understand the ecological impacts of hydropeaking, including its effects
on fish populations, water quality, and overall ecosystem health. By
advancing our understanding of hydropeaking and its impacts, we can
develop more effective strategies for managing our water resources and
minimizing the negative impacts of hydroelectric power generation.