Re-Thinking Ecological Flow in Romania: A Sustainable Approach to Water Management for a Healthier Environment

Abstract

Water resources and aquatic ecosystems are facing significant threats due to unsustainable water management practices. To address this challenge in Romania, a sustainable approach to water management is necessary, re-thinking ecological flow. This article proposes a re-thinking of the current approach to ecological flow in Romania by advocating for a more holistic and integrated approach considering environmental factors. The objective of the article was to present a methodology for the establishment of ecological flow that took into account the natural variability of flows. Four ecological flow values related to flood, high-water, medium-water, and low-water hydrological regimes were defined. To establish them, the duration curve of average daily flows was used in conjunction with hydromorphological and biological indicators. The proposed methodology was applied and compared to the existing methodology for the Uz river developed with hydropower use. The methodology represents a transition from the anthropocentric perspective to the sustainability perspective. The proposed methodology is easy to apply, with rigorously defined hydrological, hydraulic, and biological criteria. This research was conducted within the Hydrotechnical Faculty to refine the Romanian legislation regarding the improvement of the ecological status of all rivers.

1. Introduction

The quality of water bodies is influenced by modifications to the hydrological regimes of rivers in the context of economic development, overlapping with climate change.

The anthropization of rivers leads to the modification of the natural conditions of habitats, affecting the aquatic ecosystem. What is worse is that species that require strict protection are threatened with extinction due to reduced flows downstream for all water uses (in particular, abstraction, impoundment, and flow regulation) [1]. Thus, the economic prosperity of human society negatively affects biodiversity.

A measure to reduce the impact on the aquatic environment is to ensure flows that provide optimal conditions for the aquatic ecosystem, imitating the variability of flows in a natural regime and improving the efficiency of the use of water resources. The effective implementation of ecological flow can be achieved by a socio-economically acceptable reduction in the use of water resources by increasing the efficiency of technological processes and turning to other alternative sources of energy, such as wind and photovoltaic energy [2].

In the European Union, concerns regarding the mitigation of the effects of altering hydrological regimes have focused on hydromorphological analysis and measures to reduce hydromorphological pressures, which have been transposed through the Water Framework Directive [3], because the main hydromorphological indicators, such as the river channel, shore, and riparian zone and the continuity of the floodplain, are directly influenced by flow variability.

In Europe, several hydrological methods are used to determine ecological flow, with the predominant method being the employment of average monthly flows [1,4,5,6,7,8,9,10], and the duration curve of average daily flows being rarely implemented [6,11,12,13].

In Romania, ecological flows are implemented through HG 148/2020 [14], and there are no official rules for hydromorphological analysis, only proposals [15].

Guidance Document No. 31 [1] recommends an ecological flow that is variable, imitating the natural variability of river flows over time, because this has an important role in ensuring the good ecological condition of water bodies, including biodiversity, a concept known as the “natural flow paradigm” [16,17,18,19].

The ecological flow variability must be seasonal and “alike”, because native aquatic and riparian biota adapt to this variability [1,16].

According to the research of Wei [20,21], ecological flow calculation includes four categories, with the main methods presented in Figure 1.

The hydrological method is preferred because it is simple. This method is enhanced by using correlations with local physical-geographical conditions and critical periods of the ecosystem, usually in general terms. The hydrological method is a generalizing method, so it is an inappropriate methodology for ecosystems that have particularities.

The hydraulic method is a method that generally defines the hydraulic parameters, such as the wetted perimeter, depth, and velocity, required by the ecosystem. It is a rarely used method and is often applied in tandem with other methods.

Habitat simulation methods are a very good alternative for particular cases, because generalization leads to the problems mentioned above.

Holistic methods comprise a multitude of approaches that take into account ecological, hydraulic, and hydrological conditions and the requirements of human society as limiting factors.

It should be noted that the methods for calculating ecological flow are diverse, and their application can lead to deviations from the proposed objectives due to attempts at simplification and generalization, often made by proposing inappropriate coefficients and referring to the apparent conditions of the ecosystem [14].

The calculation of ecological flow requires an exhaustive approach, considering hydrological data, which is generally sufficient; topographic data (reference sections, with the data necessary for the hydraulic calculation); and data related to the ecosystem (species and their needs). This makes any methodology based on a simplified approach generally easy to apply compared to complex approaches, which are difficult to apply.

The methodology applied in Romania since 2022 [14] is a simple one based on unconvincing hydrological calculations. Ecosystem reporting is achieved by eliminating the current biodiversity studies out of the desire to simplify and generalize the method. Until 2020, environmental flow in Romania was assigned a constant value throughout the year, which led to ecological responses such as local extinction, threats to native species, and reduced nutrients for floodplain plant species [17,22,23,24].

Methods for establishing ecological flow must take into account the alterations in the flow components of magnitude, frequency, and duration [25,26].

The ecological flow must take into account the seasonal dynamics of the flow, comprising four values corresponding to the hydrological regime (low flow, medium waters, high waters, and floods), and generally be greater than 25% of the multiannual average flow [27]. In general, for the area of Eastern Europe, hydrological regimes can be classified based on the duration curve of the average daily flows. Thus, wet conditions correspond to a duration interval of 0 to 10%, moderately wet conditions correspond to a duration interval of 10–40%, moderately dry conditions are associated with a duration interval of 40–60%, dry conditions correspond to a duration interval of 60–90%, and very dry conditions correspond to a duration interval of 90–100% [12].

For the methodology proposed in this article, a similar classification approach was used, considering the influence of hydrological regimes on hydromorphological indicators, where moderate conditions, dry conditions, and very dry conditions become low-water regimes, and wet conditions are divided into floods and high waters with changes in duration intervals.

Thus, we considered four hydrological regimes: floods for a duration interval of 0–0.274%, high-water periods for a duration interval of 0.274–5%, medium-water periods for a duration interval of 5–30%, and low-water periods for a duration interval of 30–100%. This classification allowed us to highlight the optimal ecological conditions for biodiversity.

Below, the novel elements of our method are presented, such as the definition of the four hydrological regimes, namely floods, high waters, medium waters, and low waters; the use of the duration curve of the average daily flow to define the hydrological regimes and corresponding ecological flow values; the holistic approach integrating hydrology and river hydraulics with biodiversity; and ensuring the daily variability of the ecological flow using hydrometric systems.

This article presents the exhaustive research carried out in the Faculty of Hydrotechnics regarding the development of a methodology for determining ecological flow to replace the existing method.

2. Methodology

In this article, a methodology for determining ecological flow is proposed as a measure to reduce hydromorphological pressures, especially for highly modified water bodies, and to ensure biodiversity.

The proposed methodology represents an improvement upon the existing approach, being practical and transparent and imitating the seasonal dynamics of flows. The methodology relies on a holistic approach, with the determination of some values reported on the duration curve of the average daily flow, taking into account ecology and economy in order to reflect the overall situation of the river ecosystem and the water demand for the most important water users. Thus, the methodology represents an easy-to-apply hydrological synthesis with prospects for improving the environment in which we live.

2.1. The Existing Methodology

The existing methodology [14] is a hydrological method that uses as basic data the multiannual average monthly flows of the last 30–50 years under natural conditions. Thus, at least one reference period that reflects the changing climate is included in the analysis.

The ecological flow has three distinct values, trying to imitate periods of low-, medium-, and high-water hydrological regimes; flood periods are not taken into account.

The calculation of the ecological flow is carried out for each month of the year according to the multiannual average monthly flow by applying several coefficients that take into account the altitude (mountain, hill, or plain). The coefficients have four distinct values, namely 0.20, 0.25, 0.30, and 0.35, which are chosen according to the local physical-geographical and climatic conditions of the hydrographic basin in the calculation section; whether the location lies in a protected area; and the characteristics of the ecosystem, including the reproduction period of the ichthyofauna. The maximum values of the coefficients are chosen for areas that require a higher level of protection.

Based on the values of the monthly ecological flows, three distinct values are established for the hydrological regime. The low-water ecological flow is calculated using the maximum value between the average monthly minimum annual flow with the probability of exceeding 95% ($Q_{mam95\%}$—minimum annual average monthly flow with 95% exceedance probability) and the lowest value among the monthly ecological flows. The average ecological flow value represents the median of the series of monthly ecological flows. The high-water ecological flow value is obtained by the arithmetic mean of the first four values of the monthly ecological flows, in descending order.

These three ecological flow values are correlated with the average monthly flow hydrological forecast developed by the National Institute of Water Management (INHGA). The hydrological forecast is divided into five forecast classes as a percentage of the multiyear average monthly flow. These forecast classes are >100%; 80–100%; 50–80%; 30–50%; and <30%. The forecasts have a high degree of uncertainty, so for the months outside the ichthyofauna reproduction period, the ecological flows are reduced by a class of hydrological regime.

The main disadvantage of this methodology, in addition to the lack of a holistic approach, is the fact that it does not improve the ecological status of water bodies (especially those located in protected areas). This is highlighted by an analysis of the ecological flow values characteristic of the low-water hydrological regime (representing a period of eight months of the year), which are generally much lower than the values of the environmental flow (characterizing the situation before the implementation of the methodology). Additionally, the methodology does not take into account all four hydrological regimes that characterize water courses. Another disadvantage of this methodology is the reporting of the biological ichthyofauna component based on outdated materials and studies, completely ignoring the existing environmental studies that are carried out regularly and reflect the real situation of the ichthyofauna characteristic of watercourses in Romania. This aspect, contributing to the method’s generalized character, is also evident in the case study presented in this article. Additionally, the application of this methodology is disadvantageous in that it is based on monthly hydrological forecasts, which are characterized by a high degree of uncertainty.

For the ecological flow defined by HG 148, the increase in flow values over a period of 2–4 months of the year (depending on the hydrological forecast) is much reduced or non-existent in the case of run-of-the-river facilities, because they have an installation coefficient of at most 1.5–2.5, and flow is ensured through overtopping. Furthermore, dams remain under the previous conditions (environmental flow) due to their exemption from the application of the method based on the analysis of the disproportionate costs this would entail, so no improvement is achieved.

2.2. The Proposed Methodology

The recommendations in [1] and other specialized materials [28,29,30] supported the application of a methodology with four ecological flow values that imitate periods of different hydrological regimes.

The methodology is based on the tributary flow input and its partial return to the river according to the hydrological regime, thus ensuring better variability compared to the approach of using a constant value for long periods of the year (eight months). The advantage of the proposed methodology is that the variability of the ecological flow does not depend on seasonal period but takes into account the hydrological regimes of the tributary flows. Thus, during the period of low waters (usually associated with the July–February period in the existing methodology), periods of medium and high waters or even floods can be recorded: instead of ensuring a constant value during this period, a variable flow depending on the tributary flow is implemented according to the hydrological regime.

Figure 2 shows a flow chart of the proposed methodology for determining ecological flows.

In this figure, HIS is the habitat suitability index, WUA is the weighted usable area index [31,32,33,34,35], SWDI is the Shannon–Wiener diversity index [36,37,38], and PI is the Pielou index [36,37,38,39].

The proposed methodology considered the duration curve of the average daily flow from the last 30 years under natural conditions. The principle behind the choice of average daily flows was that they describe more accurately the hydrological characteristics of the river. The variation with the daily flow frequency is also important for biological processes, because native aquatic and riparian biota are adapted to this variability.

The methodology incorporated a holistic approach (hydrological, hydraulic, and biological) but established thresholds, durations, and flows based on the annual hydrological regime. The analysis was based on rivers with varied catchment areas and physical-geographic characteristics, as well as biological characteristics from up-to-data environmental studies.

Hydrological data were based on chronological series of average daily flows transformed into duration curves of the average daily flow expressed in modulo flows ($k=Q_{day}/Q_m$).

Figure 3 shows part of the analyzed duration curves characteristic of several watercourses with varied morphometric elements.

Table 1. The morphometric characteristics of the analyzed rivers.

River	Watershed Area (FBH)	Average Altitude (Hm)	$\mathbf{Multiannual} \mathbf{Average} \mathbf{Flow} \mathbf{\left(}{\mathit{Q}}_{\mathit{m}}\mathbf{\right)}$
	(km2)	(m)	(m3/s)
Viseu	1555	1020	34.6
Crisul Negru	3750	351	30.3
Mures	27,056	618	186
Capra	79.4	1625	2.72
Prigor	141	729	1.71
Badeni	27.6	1438	0.455
Ialomita	6233	494	33.8

shows the morphometric elements of the rivers included in Figure 3.

Based on the analyzed duration curves, the characteristic durations of the hydrological regimes specific to Romania were established, namely the regimes of low waters, medium waters, high waters, and floods.

The duration of the low-water regime was established as being in the range of durations with a probability of 30–100%. The 30% limit for the duration curves of the daily average flows was established considering that the values of the multiannual average flow were in the range of 25–35%, and the duration of 30% represented the middle of this range.

The duration of the medium-water regime was established as being in the range of durations with a probability of 5–30%. The threshold at a duration of 5% was selected on the grounds that it generally corresponded to a value of three times the average multiannual flow, considered characteristic of periods of high waters [40,41,42].

The duration of the high-water and flood regimes was determined to be characteristic of durations with a probability <5%. The high-water regimes are characterized by high flows over relatively long periods, and flood regimes are characterized by very high waters over relatively short periods. Thus, it was necessary to impose a duration threshold delimiting these two sub-regimes. The threshold was set as the duration with the probability of 0.274%. Following research [15] carried out within the Faculty of Hydrotechnics regarding the improvement of the norm for the delimitation of the bankfull channel, this corresponded to the duration of the bankfull discharge, which is the main hydrological component for the lateral connectivity of water bodies. Thus, the duration of the high-water regime was between the thresholds of 0.275 and 5%, and the flood regime had a duration <0.274%.

For the biological and hydromorphological process reference sections, hydraulic data were obtained, consisting of cross-sections with hydraulic slopes and associated roughness. Based on these data, limnimetric keys were determined and later calibrated following hydrometric measurement campaigns. The 2D modeling results, i.e., the water flow depths and velocities for each flow threshold value, were correlated with habitat suitability criteria [43].

For the synthesis of the hydromorphological results, correlations with linearizable power functions (Leopold and Maddock) were used in the technical reports for the synthetic characteristics of the bankfull channel (width B, average depth H, velocity V) [44,45,46].

$$B\left(Q\right)=a\times Q^b$$

(1)

$$H\left(Q\right)=c\times Q^f$$

(2)

$$V\left(Q\right)=k\times Q^m$$

(3)

The relations were logarithmized for linearization; thus, the coefficients $a,c,k$ and the exponents $b,f,m$ were obtained by the least squares method with the flow and depth vectors as calibration data, under the condition that the product of the coefficients was one and the sum of the exponents was one. The depth function of flow was thus more simply defined for the water depth range from thalweg to bankfull. The coefficients $a,c$, and $k$ depended on the roughness of the channel and the hydraulic slope of the river, and the exponents $b, f$, and $m$ depended on the shape of the channel.

Another solution was to use numerical computation based on the inverse of the function defining Chezy’s relation, which requires iterative numerical computation.

The proposed methodology focused on ensuring optimal conditions for biodiversity, also reflected by the hydromorphological indicators that provided a data centralization solution for decision making.

Biodiversity is the most important characteristic of water bodies. Reference ecological conditions (depth, width, speed, and temperature) and optimal ecological conditions were defined according to recent ichthyofauna and environmental studies. The parameters of the ecological conditions were chosen based on up-to-date mandatory environmental studies (ichthyofauna, riparian, phytobenthos, etc.) for the reference sections of the considered water bodies established following the analysis.

After the multicriteria analysis (hydrological, hydraulic, and biodiversity-based), the ecological flow thresholds associated with the four hydrological regimes were established. Natural hydrological regimes are modified by flow sampling, affecting the river length. Thus, the proposed methodology optimized the water requirements for various applications and the water requirements for ensuring biodiversity. The modified hydrological regime would not actually approach the natural hydrological regime in terms of flow durations and magnitudes, but it would preserve for shorter durations the important characteristics of the natural hydrological regime. In the proposed methodology, four duration intervals were defined for the duration curve of the average daily flow for the hydrological regimes of low waters, medium waters, high waters, and floods.

Thus, for the hydrological regime of low waters, the ecological flow threshold corresponding to the exceedance probability $Q_{90\%}$ was chosen from the curve of the duration of average daily flows. This had a constant value during the hydrological regime of low waters (30–100%). The choice of the flow threshold took into account the values of the hydromorphological indicators (depth, width, and speed) that were optimal for biodiversity.

During the hydrological regime of medium waters, the value of the ecological flow corresponded to the flow with an exceedance probability of $Q_{80\%}$ based on the duration curve of the average daily flows. This flow had a constant value throughout the hydrological regime of medium waters (5–30%). In establishing this flow, the particularities of the biodiversity component were taken into account, especially the reproduction period of the ichthyofauna. This period was established based on up-to-date environmental studies. It should be noted that during this period, a compromise was made between water management and biodiversity, so instead of a flow close in magnitude to the average flow, a flow with a value of approximately 30% was accepted ($Q_m$).

The hydrological regime of high waters was characterized by an ecological flow corresponding to an exceedance probability of 30% from the duration curve of the average daily flows ($Q_{30\%}$). This value was constantly ensured throughout the hydrological regime of high waters (0.274–5%). The value of the ecological flow was approximately $Q_m$, below the characteristic values of the hydrological regime of high waters, which vary from $Q_m$ to approximately $3\times Q_m$.

Downstream of the dams for water reservoirs, which are high-volume and therefore subject to multiannual regularization, there is no flow even during high-water periods. The multiannual average flow was selected as a compromise between the environment and water users. Thus, the river does not have flows specific to periods of high water. It is true that in smaller catchments, which have an installed flow of no more than three times the average multiyear flow of the river, this does not represent a problem, as large water flows are transferred downstream.

Regarding the hydrological regime characteristic of floods, the discharge threshold was established as the bankfull discharge value. Following research within the Faculty of Hydrotechnics aiming to improve the existing regulations for the delimitation of the bankfull channel, bankfull discharge corresponded to the discharge with an exceedance probability of $Q_{0.274\%}$ based on the duration curve of the average daily discharges.

Thus, during periods of hydrological drought, it is prohibited to reduce the flow. This corresponds to a flow threshold value comprising 90% of the flow duration curve (FDC), and the period of high water for reservoirs and large dams is improved by controlled discharges under bankfull channel characteristic flows.

The application of the ecological flow values must take into account the provisions and recommendations of the European legislation regarding the disproportionality of costs and technical infeasibility [1,14].

The Romanian legislation [14] states that, if, following the analysis of disproportionate costs, the ecological flows are not implemented, the initial regulatory conditions are kept, namely the maintenance of the environmental $Q_{95\%}$ flow.

It was thus necessary to introduce an additional condition: if, following the calculation of disproportionate costs, the implementation of the ecological flow is not possible, a step-by-step increase in the environmental flow to $Q_{90\%}$ for run-off rivers and $Q_{95\%}$ for the reservoir regularization of annual (multiannual) flow should be implemented. During periods of high-water flow and floods, run-off river power plants with or without pondage discharge, thus implicitly ensuring the high-water flow without restrictions. This phased application involves the involvement of society as a whole in increasing the efficiency of water use and thus reducing its consumption in favor of increasing river flows. This can be achieved through successive stages over the six-year periods of the water management plans, increasing flows by $1\%\times Q_m$ every six years, thus reaching the proposed values in a 30-year period.

Additional measures could be imposed to reduce hydropeaking by building rectifier basins or abandoning the operation of the weightage.

Water intakes are used for collecting water from rivers. The monitoring necessary for the implementation of ecological flows was carried out with the help of limnimetric keys in the upstream and downstream sections of the water intakes or even using the limnimetric keys of the intakes. The identification of the hydrological regimes was achieved via the readings on the calibrated limnimetric keys in the upstream sections, and the ecological discharge corresponding to the hydrological regime was ensured by checking the calibrated limnimetric keys in the downstream section. Currently, daily readings are taken per meter and flow balance to ensure environmental flow for all water uses, so the procedure for ensuring and monitoring ecological flow is easy to implement.

3. Case Study

The Uz River is the right tributary of the Trotus River from the hydrographic basin of the Siret River (cadastral code XII).

It is located in the eastern part of Romania in the territory of the Harghita and Bacau counties, between the geographical coordinates 46°19′39.0″ N 26°02′51.2″ E and 46°21′55.5″ N 26°30′54.4″ E (Figure 4).

The Uz River has a length of 50 km, a watershed area of 469 km², an average altitude of 972 m, an average slope of 17‰, and a sinuosity coefficient of 1.26.

Its main tributaries are the Başca (cadastral code XII −1.69.22.3) and Bârzăuţa (cadastral code XII −1.69.22.5) rivers, both of which are right tributaries [47].

ROSCI0327 Nemira-Lapos is a protected site on this river.

Along the course of the Uz river and its two main tributaries, a hydropower facility consisting of five Tyrolean-type water intakes and six small hydropower plants can be found.

Immediately upstream of the confluence with the Trotus river, one finds the Uz Valley gravity dam, with a height of 84 m and a reservoir of 98 million cubic meters. These are used for the supply of drinking and industrial water, including for a small hydropower plant.

The water measurement locations coincided with the water intake locations. The water intake has an installed flow of approximately 1.5 times the multiannual average flow of the river, so in operation during the periods when the flows are below the installed flow, a flow of approximately 20–25% of the multiannual average flow is left downstream. It can thus be said that this arrangement influences the hydrological regime during periods of average flows, having no influence during periods of hydrological drought and during periods of high waters and floods.

The main morphometric characteristics of the rivers are presented in

Table 2. The morphometric characteristics of the Uz River and its tributaries.

River	Length (km)	Average Slope (‰)	Sinuosity Coefficient (−)	Average Altitude (m)	Watershed Area (km2)
Uz	50	17	1.26	972	469
Oreg	6	29	1.3	1124	12
Başca	10	27	1.32	1123	25
Copuria	6	113	1.12	1001	10
Bârzăuţa	26	22	1.31	1091	156
Apa Lina	15	17	1.22	1030	48

[48].

In the considered location, the zonal relationship for calculating the average specific flow as a function of the average altitude was obtained using the equation: $q_m=5.382\times {10}^{-5}\times H_m^{1.753}$.

The duration curve of the average daily flows was obtained from the time series of average daily flows at the Uz Valley reference hydrometric station for a period of 40 years. From the duration curve of the flow rate, the ecological flow values corresponding to the four hydrological regimes were obtained.

Table 3. The threshold values of ecological flows.

Flows from the Duration Curve of Average Daily Flows
$k_{0.274\%}$	$k_{30\%}$	$k_{80\%}$	$k_{90\%}$
9.37	1.01	0.297	0.207

presents the threshold values of the ecological flows corresponding to the hydrological regimes.

From the point of view of the Romanian legislation regarding river typology, the Uz River belongs to the RO01 typology, being part of the mountain typological class. According to this typology, the fish species that should be found in this location are brown trout (Salmo trutta), grayling (Thymallus thymallus), and chub (Squalius cephalus). However, in reality, the ichthyofauna of the site featured bullhead (Cottus gobio), Petenyi’s barbel (Barbus petenyi), brown trout (Salmo trutta), minnow (Phoxinus phoxinus), and stone loach (Barbatula barbatula). Among these species, two are protected at the European level (Cottus gobio and Barbus petenyi). Cottus gobio has protection status according to [48,49]. Barbus petenyi has protection status according to [48,49,50,51].

Figure 5 shows the five species of fish found at the site, according to information obtained over 10 years of specialized studies [52].

The critical periods characteristic of these species are highlighted in

Table 4. Critical periods of fish species found at the site.

Fish Species Name	Breeding Period
	Months of the Year
	1	2	3	4	5	6	7	8	9	10	11	12
Cottus gobio			X	X	X
Barbus petenyi					X	X	X
Salmo trutta										X	X	X
Phoxinus phoxinus				X	X	X
Barbatula barbatula				X	X	X

.

Along the course of the Uz River and its two main tributaries, a hydropower facility consisting of five Tyrolean-water intakes and six small hydropower plants can be found.

The maximum height of the water intakes is 1.45 m, and all sections are provided with fish ladders for ichthyofauna with increased mobility (Barbus petenyi, Salmo trutta, Phoxinus phoxinus, and Barbatula barbatula) and lateral passages specially designed to ensure optimal conditions for sedentary ichthyofauna with low mobility, such as Cottus gobio.

4. Results

To observe the differences in the ecological flow calculated by the new methodology and the existing methodology [14], the river Uz was chosen, which features a run-of-the-river hydropower arrangement with five water intakes and six small hydropower plants (SHPP), representing a model of good practices in terms of the impact on the environment.

At the initiative of ichthyofauna specialists [53,54], with a beneficiary’s help, alternative passages to the fish ladders were created to ensure the mobility of sedentary fish species.

Until 2023, the existing environmental flow (salubrious flow) represented 21–26% of $Q_m$ and was constantly ensured, regardless of the hydrological regime.

According to the legislation in force, for the July–February period, the ecological flow is constant and represents 13% of $Q_m$, practically half of the environmental flow. This is a consequence of inadvertently choosing the ecological flow value as the maximum of two values from two different datasets, namely the minimum annual average monthly flow with a 95% excess probability and the lowest value of the monthly ecological flows, which represents a fraction of the average monthly flows.

For the forecast of average monthly flows below 80%, in the March–June period the ecological flow is 28.5% of $Q_m$. This period was established on the basis of outdated ichthyofauna studies [55] as the breeding period of the potential fish fauna.

In the INHGA studies, it was wrongly stated that the fish fauna was represented only by brown trout, grayling, and chub species. In reality, the species include Cottus gobio, Barbus petenyi, Salmo trutta, Phoxinus phoxinus, and Barbatula barbatula.

Regarding the forecast of the average monthly flows above 80%, the ecological flow increases only for the month of March to 62% of $Q_m$.

It was observed that regardless of the fact that in the autumn months the hydrological regime is often medium and/or high waters, the flow remains at 13% of $Q_m$.

A considerable disadvantage of the methodology presented in [14] is that it does not ensure the variability of the flows according to the hydrograph and imposes seasonal variability for a period of eight months without variability and with lower flows compared to the existing conditions, i.e., lower values than those of a normal hydrological year.

Figure 6 shows the duration curve.

Table 5. Ecological flows according to HG148.

Class of forecast

Months of year

I

II

III

IV

V

VI

VII

VIII

IX

X

XI

XII

>100%

62%

29%

80–100%

50–80%

13%

30–80%

<30%

The yellow, green and blue colors represent the low, medium and high–water regimes.

presents the ecological flow conditions according to HG148 [14].

The proposed methodology imitated the natural hydrograph in certain periods, resulting in variable flows depending on the hydrological regime, thus eliminating the shortcomings of the existing methodology, which features ecological flows with fixed monthly thresholds.

The value of the ecological flow for the hydrological regime of low waters was 21% of $Q_m$, compared to 13% in the existing methodology.

The value of the ecological flow for the hydrological regime of medium waters was 30% of $Q_m$; this value was also determined for the July–February period.

For the hydrological regime of high waters, the value of the ecological flow was close to the multiannual average flow, and for the period in which floods are recorded, the ecological flow was approximately 9.4% of $Q_m$, thus ensuring lateral connectivity.

Figure 7 shows the duration curve and threshold values for the proposed methodology. When the threshold values are reached, they lead to the modification of the flows downstream of the water catchment sections, the dynamic aspect being the fact that the upstream flow is checked every day, and the flow corresponding to the thresholds is maintained downstream, according to Figure 8. This dynamism is not identical to that of the river but imitates the important periods.

A shift in the duration curve can be observed from the 90% duration with lower values to also ensure water uses but much higher values in important periods compared to the existing methodology.

The proposed methodology analyzed daily flows (inflow) through monitoring and thus established daily flow variations, imitating nature, if necessary, especially for periods of high waters and floods, when the most frequent changes occur.

Figure 8 shows a simulation of the ecological flows for the year 2018 using the existing method and the proposed method.

The difference between the inflow volumes and the volumes consumed by the water intakes is called the returned volume. The returned volumes downstream of the water intakes were 18% of the 2018 stock for the old methodology and 37% for the new methodology.

According to the multiyear average stock, for the old methodology the returned volume was 14%, and for the new methodology it was 29%.

In the case of the environmental flow, the returned volume from the average multiyear stock would have been 21%.

5. Discussions

The implementation of ecological flow in Romania could have been a response to infringement clause no. 2015/4036—water—micro-hydropower plants in Romania [56] regarding the incorrect application of [3] and [38], in the approval process of micro-hydropower plants.

The methodology described in HG 148/2020 [14] represents one step forward and two steps back, as an opportunistic approach to an important topic in water management with “zero” environmental contributions. It has not improved the ecological status of water bodies. Before 2020, the environmental flow ($Q_{sal}$) was used in Romania, with a constant value, which was defined as the flow with a 95% probability of exceeding the duration curve of average daily flows ($Q_{95\%}$). The ecological flow ($Q_{ec}$), according to HG 148 [14], is a variable flow with three values depending on the hydrological regime and the hydrological forecast. The characteristic value of the hydrological regime of low waters, for eight months of the year, is generally much lower than the environmental flow value. This represents a real threat to the ecological condition of water bodies, including hydromorphological indicators. The most important criticism is the reduction in flows for very long periods of at least eight months a year, especially for rivers in natural protection areas. The advantage of increasing the flow values for 2–4 months per year depending on the hydrological forecast is non-existent for run-of-the-river facilities.

The methodology provided in HG 148/2020 used a hydrological method simplistically correlated with the critical periods of fish species, without updating the ichthyofauna studies or taking hydromorphology into account and disregarding hydraulic criteria, such as the minimum water depth for protected ichthyofauna.

This method was applied to several sites (ROSCI0122 Făgăraș Mountains, Crișul Negru, etc.) designated for the conservation of Romanogobio uranoscopus, which has a reproduction period from May to June [55], and Barbus petenyi and Sabanejewia balcanica, which have a reproduction period in the months of May–August [55]. In the eight-month period of the existing regulations, the reduction in flows for a long duration within the reproduction periods, as well as the pre-development periods, would lead to a high probability of extinction. Existing legislation considers only part of the reproductive period as critical, completely ignoring the larval period.

Table 6. Ecological flows vs. environmental flow.

River/Section	Location (Stereo 70 Coordinates)		Qsal	Qec	Relative Decrease
	X	Y	(m3/s)	(m3/s)	(%)
Uz, S1	588,177	535,287	0.187	0.100	47
Uz, S2	591,789	535,387	0.250	0.131	48
Baska	594,435	533,770	0.055	0.034	38
Uz, S4	594,186	535,746	0.351	0.183	48
Uz, S5	596,901	538,721	0.457	0.220	52
Barzauta	598,123	533,676	0.375	0.211	44
Vistea	467,134	480,336	0.202	0.107	47
Sâmbăta	483,838	460,648	0.138	0.058	58
Viștișoara	481,882	461,806	0.078	0.036	56
Porumbacu	462,540	457,880	0.152	0.122	20
Călinești	441,401	433,684	0.152	0.113	26
Taia	375,799	447,525	0.114	0.098	14
Ausel	377,447	448,071	0.251	0.189	25

presents the ecological discharge situation in ten rivers compared to the environmental flow.

It is possible to observe a decrease in the flows characteristic of the hydrological regime of low waters (eight months of the year), which is the most vulnerable period for biodiversity.

Moreover, the government authorities that prepare ecological flow studies have published studies approving new water-consuming and polluting activities (intensive fish farming), thus increasing the degree of deterioration of rivers in areas already affected by other activities.

The methodology proposed in this article used updated environmental studies requested by the governmental environmental agencies in Romania. Reporting the duration curve of the average daily flow was chosen as the implementation method, as it is easy for water users to apply and understand. The methodology involved the phasing of the increase in environmental flows for those who claim disproportionate costs.

The proposed methodology relied on a holistic method (hydrological, hydraulic, and biological) but established thresholds, durations, and flows correlated with the annual hydrological regime.

This research was based on correlating the hydrological regime with the needs of the ichthyofauna in critical periods (reproduction and larval), achieving a compromise with the use of water.

To establish the flows, the curve of the duration of the average daily flows was used together with hydromorphological and biological indicators synthesized from over 80 environmental, ichthyofauna, and hydrological studies. For example, for the hydraulic aspect, the ichthyologists requested a minimum of 10 cm water depth. Thus, the minimum thresholds resulting from the hydraulic calculations led to flows that corresponded to a duration of 90% on the duration curve of average daily flows. The calculation criteria were diverse, but, importantly, the results were simple and easy to apply. Thus, a synthesis was imposed, namely a graphic representation of the duration curve with the thresholds for the anthropogenic hydrological regimes (modified through the use of water). This synthesis was much more suggestive and easier to apply.

The proposed methodology has the advantage that it is variable for wet years, normal years, and dry years, in contrast to the existing methodology, which does not differentiate between wet years and normal years.

Lateral connectivity was ensured by imposing a flow rate greater than the bankfull discharge, in contrast to the existing methodology, which does not foresee downstream flows higher than 0.7% of $Q_m$, resulting in downstream river sections of water accumulations with annual and multiannual regularization condemned to a perpetual regime of low waters, well below the multiannual average flow.

6. Conclusions

The existing methodology for the determination of ecological flows in Romania is based only on hydrological approaches, representing a step backwards compared to the situation before its implementation and leading to no real improvement in the water bodies, especially those located in natural protected areas.

The methodology proposed in this article represents a holistic approach based on a complete and complex analysis and research carried out in the Faculty of Hydrotechnics regarding the development of regulations to replace the existing legislation for determining the ecological flows in Romanian water courses.

In this article, the methodology was applied and compared to the existing methodology on the Uz river, which features (from upstream to downstream) six small hydropower plants in a cascade, as well as a dam and the Uz Valley reservoir.

The main conclusions are summarized as follows:

(1)

The proposed methodology took a holistic approach, and the calculation methods presented were simple, concise, and unambiguous.

(2)

From the point of view of the hydrological component, the methodology respected and ensured the natural variability of the flow, characterized by four hydrological regimes: floods, high waters, medium waters, and low waters. The methodology used the duration curve of average daily flows, which provided a higher degree of confidence than the duration curve of average monthly flows regarding the hydrological regimes of the rivers.

(3)

The biodiversity component was of particular importance in the development and application of the methodology, which took into account both the potential biodiversity and the capacity to sustain it, as well as the existing environmental studies regularly conducted for the majority of rivers in Romania, clearly highlighting the real biodiversity situation in the rivers and the measures necessary to support it.

(4)

The application and monitoring of ecological flows were carried out in real time according to the hydrological regime, which was based on the limnimetric key calibrated via the hydrometric measurement campaigns of the hydrotechnical construction evacuators.

(5)

The opportunity to apply ecological flow is provided following the analysis of costs, thus taking into account the recommendations of the European Guidance Document No. 31. If the analysis results in disproportionate costs (thus supporting the maintenance of the environmental flow), the methodology proposed in this article provides for increasing the environmental flow from $Q_{95\%}$ to $Q_{90\%}$ (but no less than 100 L/s) for the run of river, and for the reservoir regularization of annual and multiannual flow, the environmental flow is still increased to $Q_{95\%}$. These increases are implemented once every 6 years with 1% of the multiyear average flow until the target values are reached.

(6)

Additional measures are required to reduce hydropeaking by creating rectifier basins or abandoning the exploitation of weightage. This is a measure specific to the hydropower sector.