Abstract

In this study, 96 Environmental Impact Assessment (EIA) reports of hydropower projects from 2001 to 2015 have been collected, and four parts of the contents including the project status, hydrological characteristics, downstream fish status and ecological flow have been analyzed by statistical methods. Thus, we built the ecological flow database of hydropower projects in China, and analyzed the differences of ecological flow release before and after the implementation of the 2006 guideline, based on spatial and temporal characteristics analysis in ArcGIS platform. We also analyzed differences between the calculation methods, release types, safeguard measures and monitoring measures of ecological flow. Meanwhile, we analyzed the ecological benefits of insurance for endangered fish protection. We focused on the effectiveness of ecological flow guarantee and shortages after the 2006 guideline. Therefore, ecological flow practices on hydropower projects in China have been summarized, and we suggested 17% of annual average flow as the ecological flow constraint red line, combined with the flow process demand. Then we put forward the prospect of ecological flow research and practice in China, by strengthening the research of calculation principles, cascade reservoir operation and ecosystem recovery and rehabitation.

Introduction

The river ecosystem is a complex mosaic of interaction and interrelationship of biotic (flora and fauna) and abiotic (hydrological, geomorphic, etc.), and it is now globally agreed that adherence to the minimum environmental flow or ecological flow in a river alters the dynamic equilibrium of the river basin to a new but sustainable dynamic equilibrium (Shekhar, 2016; Chen et al., 2017). In river ecosystem management, identifying clear ecological responses to flow alteration can be a significant challenge because of the complexity of river systems and other factors which may confound the response (Kirby et al., 2014; Summers et al., 2015). Thus, research developed an Adaptive River Management (ARM) framework of river flows in different regions of the world such as Europe, the USA and China. Adaptive management aims to reduce uncertainty by increasing knowledge and understanding, thus enabling improved management decisions over time, which has been used internationally in a wide range of ecological flow management and practice. In 2000, the World Commission on Dams (WCD) indicated that it was not possible to mitigate many of the impacts of dams, including little success with fish passes. Ecological flow is supposed to reduce the adverse regulation impacts on channel, flood plain and estuary ecosystems (Warner, 2014). The Water Framework Directive (WFD) is the first legislation in Europe to explicitly use ecological conditions as the benchmark for the management of ecological flow. Green Hydropower in Switzerland and the Low Impact Hydropower policy in the USA are the most widely used sustainable assessment guidelines in the world now, under the framework of adaptive management. Ecological flow management regulations were promulgated in many countries, river basin scale to hydropower project scale, and adaptive management of ecological flow has been one effective way of river ecosystem protection by maintaining the minimum flow in the river course and flow duration curve for ecological sensitive periods.

Determination of ecological flow duration curves needs to be optimized through adaptive management. The curve was mainly for special and sensitive targets according to research results and practices. Because of the significant differences in river types, reservoir types, aquatic organism types, etc., even though the Tennant method provided a solution in minimum flow management, it led to dramatically low discharges as fixed minimum environmental flows, while the habitat simulation method gave an acceptable ecological flow duration curve. The duration curves should be determined based on regional characteristics diversity, especially for endangered fishes, and improved and optimized through reservoir operation and post-Environmental Impact Assessment (EIA). In many cases, restoring flows back to completely natural conditions would not be feasible because of the important societal benefits which are associated with the flow alterations arising from abstractions, discharges and impoundments. Therefore, it is necessary to identify the limits to which the components of flow regimes can be modified, without having unacceptable ecological impacts, and it is necessary to identify clear ecological responses to flow alteration. Even though the adaptive management of ecological flow also has both active and passive impacts, it is currently the most effective way to optimize the ecological flow duration curve. More practice was needed to improve the passive adaptive management process. Water is crucially important in energy production, and the Green Hydropower policy has been used to manage the ecological flow in Switzerland and prompted stakeholders to release more water for downstream aquatic ecosystems (Karakoyun et al., 2016). Meanwhile, there would be more ecological benefits by integrating ecological operation. In the process of adaptive management implementation, many existing ecological flow monitoring programs have poorly defined objectives, nonjustified indicator choices, weak experimental designs, poor statistical strength, and often focus on outcomes from a single event (King et al., 2015). To improve the efficiency of ecological flow adaptive management, decision support systems have been introduced in several basins such as the Delaware River in the USA and the River Mahaweli in Sri Lanka (Eriyagma & Jinapala, 2014; Maloney et al., 2015). The system that synthesizes the voluminous scientific data and models on these factors will facilitate the management of ecological flow and integration with a river monitoring system. It contains all individual model outputs, computationally integrates these data, and produces the amount of potentially available habitat for a suite of species of interest under each flow release scenario. With the development of ecological flow, the connotation varies from the flow requirements of specific endangered fish species to flow duration curve requirements of sustainable social–ecological aquatic systems (Matthews et al., 2014). The need to deepen the adaptive management ecological flow duration curve and strengthen the range of EIA process is growing.

In China, there were two kinds of ecological flow concepts: the environmental flow concept was introduced from the Brisbane Declaration which represented a definition for environmental flow; and the ecological flow concept was defined as the discharged flow of reservoir projects for maintaining the structure and functions of segment. The difference in the two concepts was that environmental flow was considered from watershed scale, while ecological flow was from the viewpoint of flow management on a segment scale (Chen et al., 2016; Wu et al., 2017). Thus, we used ecological flow in this paper. Adaptive management methods have been applied by the Chinese government for ecological flow management and practice. The water-short adjacent basins of the Huang (Yellow), Huai and Hai Rivers in north China promoted adaptive management of water resources in these basins (Xia et al., 2012). As guidance for ecological flow adaptive management practice, the government has promulgated many guidelines, regulations and legislations to ensure the effectiveness and optimization of ecological flow release and safeguards over the last 15 years. The ecological flow related legislations used to be focused by the Ministry of Environmental Protection (MEP), Ministry of Water Resources (MWR), Ministry of Transport (MOT), etc. From the 21st century, hydropower development has shown increasing growth, especially in the southwest of China where the ecosystem was relatively more sensitive. To mitigate the impacts of hydropower development in these regions, government and researchers made great efforts to study adaptive management on ecological flow policy and the practice of hydropower projects. There have been some practices in the Yangtze River, the Pearl River and the Yellow River, such as eco-functional classification for environmental flow assessment and adaptive management in the Pearl River Basin, ecological flow adaptive management of the Three Gorges Reservoir based on the relationships between flow and ecological response for protecting the Chinese sturgeon and four major Chinese carps, and 15 years’ water and sediment regulation of Xiaolangdi Reservoir for maintaining environmental flows of the downstream Yellow River (Liu et al., 2008; Wang et al., 2013, 2016). Over 15 years’ management and practice, 10% of the annual average flow in the dam section has been made as an empirical rule for the minimum flow standard of hydropower projects, and the flow duration curve has been made as flow standard in the case of sensitive targets protection. However, there were still some shortages in the management system and the standard of the ecological flow needs to be improved by a comprehensive analysis of the development.

Therefore, we collected the ecological flow data of hydropower projects from 2001 to 2015, and analyzed the adaptive management development over the 15 years by summarizing the ecological flow determination program. We also analyzed the optimization process of ecological flow in different periods, and the shortages of flow standard. The purpose is to improve the management system of ecological flow and mitigate the downstream impacts of hydropower projects.

Materials and method

Database of hydropower projects

We built a hydropower projects database for analysis based on an ArcGIS Engine platform and SQL Server database, which includes the reservoirs and dams in China (Figure 1). The projects analyzed were those from 2001 to 2015 which had passed the EIA and where the data could be downloaded from the website of the Ministry of Environment Protection (www.zhb.gov.cn/gkml/). In the process of EIA, professionals’ opinions were the most important factor for determining and optimizing the ecological flow curve as the final period of the quantitative method. The database had been divided into four parts to include basic information, hydrological characteristic, ecological flow curve and downstream sensitive fishes. All the information was edited in Microsoft Excel and then connected to the SQL Server database, then the Spatial Reference System of GC-WGS-1984 was added in the GIS system to show and edit the data.

Fig. 1.

Hydropower projects distribution from 2001 to 2015.

Fig. 1.

Hydropower projects distribution from 2001 to 2015.

Statistical methods and spatial analysis methods were the two main methods used in this paper. Based on the basic hydropower projects database, implementation, safeguards and monitoring measures of ecological flow were analyzed. By using a geographic information system platform, spatial topography and correlation analysis were used to analyze the spatial and temporal distribution of ecological flow of hydropower projects.

Ecological flow assessment framework

A commonly accepted paradigm in ecological flow management is that a regulated river flow regime should mimic the natural hydrological regime to sustain the key attributes of freshwater ecosystems (Latu et al., 2014). The management system of rivers was not effective in managing ecological flow. Most difficulties were found in the spatial and functional diversities and there are still no guidelines for management and decision making in compromising situations. Until now, there were mainly three regulations in managing ecological flow of hydropower projects in the EIA period published by MEP. EIA practices improved and optimized the regulations and adaptive management system of hydropower projects. In the beginning, there was no unified definition of ecological flow and there were many related concepts such as Environmental Flow, In-stream Flow, Minimum Flow, Minimum Acceptable Flow, etc. Meanwhile, there are no official recommended methods because there were more than 200 types of method after several years’ research and practice. In 2006, ‘Technical Guide for Environmental Impact Assessment of River Ecological Flow, Cold Water, and Fish Passage Facilities for Water Conservation Construction Projects (Trial) EIA Letter [2006] No.4’ was issued by the State Environmental Protection Administration (former MEP), and it defined the concepts and methods of ecological flow in the EIA which stipulated 10% of the annual average flow as the minimum flow. Then, ‘MEP General Office Letter [2012] No.4’ and ‘MEP Issue [2014] No.65’ deeply strengthened the demand for an ecological flow curve and standard of the minimum flow. However, there is still not an accurate value of the minimum flow. In the adaptive management of ecological flow, flow duration curves based on ecological flow assessment framework support the decision making.

Ecological flow assessment framework developed from simple to complex through adaptive management progress. In the beginning, ecological flow was not one of the key points of EIA because of the complexity of flow quantification, and the assessment results were the simple red line constraints; the red line is the minimum flow demand and the value was usually 10% of the annual average flow. Then, following the development of methods such as habitat and holistic methods, the assessment framework became more complex, and the ecological flow duration curves were made to satisfy downstream ecological needs. Some practices of adaptive management have been taken to optimize and improve the duration curve, which made adaptive management an important method in the framework. Until now, the ecological flow assessment framework has been established and, for most of the projects mentioned in this paper, the ecological flow determination methods were in the framework within the hydrological, hydraulics, habitat simulation and holistic methods (Figure 2). The normal progress of ecological flow was usually the last three steps in the five-step framework.

  • (1)

    Management unit classification

Unlike biodiversity or conservation eco-regions and freshwater regions, the management unit of ecological flow regions was more complex. Much research has been carried out in developing simulation tools for managing ecological flow, and classification of ecological flow regions in the Mediterranean watershed (Alcázar & Palau, 2010; Steinfeld et al., 2015). Because of the integration of society and ecology, a social–ecological framework has been proposed to integrate multiple objects for ecological flow management (Martin et al., 2014). In the EIA process of hydropower projects in China, only objects in the river course and bank were assessed. Even though there is still no regionalization for a management unit, the water resources region was often used to manage ecological flow.
Fig. 2.

Ecological flow assessment framework.

Fig. 2.

Ecological flow assessment framework.

  • (2)

    Stream reach classification

After a management unit was made, the next step was to classify the stream reach. An eco-functional classification was proposed for river management in the Pearl River Basin; the method views ecological functions as fundamental characteristics of riverine systems and provides a framework for dividing a basin into eco-specific categories according to the heterogeneity of the primary ecological functions (Wang et al., 2016). Downstream reaches were of the most concern in the EIA process, for the reach length was normally 10–100 km at most.

  • (3)

    Impacted downstream survey

When the dam was constructed, there were impacts on the downstream ecosystem. There should be a comprehensive survey of downstream including sensitive targets. The survey was mainly for sensitive aquatic organisms, especially for endangered fishes, and other factors such as wetland, avifauna, pollution dilution and reserves were also in the survey lists. With the classification of stream reaches, there should be a hydrological analysis to classify natural flow regimes to support ecological flow calculation and management (Tavassoli et al., 2014). Hydrological analysis formed a critical part in the assessment framework as a connection with spatial regionalization and river courses objects.

  • (4)

    Ecological flow calculation

Calculation methods of ecological flow worldwide are now composed of mainly four types including hydrological, hydraulic, habitat and holistic methods (Arthington et al., 2004; Olsen et al., 2013). The first three kinds of method were mostly used in the EIA progress. Ecological flow adaptive management aims to improve the ecological condition of the rivers; it also represents an opportunity to collect quantitative data which will help to gain a broader understanding of environmental flow requirements (EFR). As a river monitoring system had not yet been established, a task of top priority is to collect more data for ecological flow calculation, and finally the ecological flow component could be made as a result of this assessment (Fitzhugh, 2015).

  • (5)

    Ecological flow optimization

There needs to be a comprehensive analysis of ecological flow duration curves by professionals in multiple fields such as hydrology, hydraulics, biology, etc. After ecological flow curves have been determined, the next step was adaptive management and practice. The curve was optimized through adaptive management of reservoir regulations, while ecological flow was more suitable for downstream ecosystems.

Results

Statistics of hydropower projects

Temporal and spatial distribution

The total number of hydropower projects passed by the EIA is 96, of which 37 were approved from 2001 to 2005 while 59 projects were approved after 2006. The temporal distribution of the projects is depicted in Figure 3, showing the number of projects in each year.

Fig. 3.

Temporal and spatial distribution of hydropower projects.

Fig. 3.

Temporal and spatial distribution of hydropower projects.

Spatial distribution was discussed by geographic, provincial and watershed regions. The projects were mainly in the south region, the number being 78. Separately, there were seven, six and five projects in the Qinghai-Tibet, northern and northwest regions, respectively. Sichuan and Yunnan provinces accounted for 37 and 21 projects, respectively, due to the large hydropower potential in the southwest rivers.

For watershed scale, the projects were mainly in southwest rivers (Table 1). Hydropower projects in the Yangtze River basin were mainly in the Dadu River and Jinsha River, respectively 15 and 19 (including the Yalong River and Hengjiang River which are tributaries of the Jinsha river), and projects in the southwest rivers were mainly in the Lancang River.

Table 1.

Watershed distribution of hydropower projects.

River Basins Rivers Quantity Total 
River Basins in Southeast Region Oujiang River 
Yellow River Region Yellow River 
Longqing River 
Songhua River Region Songhua River 
River Basins in Northwest Region Kashi River 
Kai River 
River Basins in Southwest Region The River 16 
Dayingjiang River 
Red River 
Lancang River 
Niyang River 
Pearl River Region Beipanjiang River 
Hongshui River 
Yangtze River Region Yuanshui River 62 
Zishui River 
Dadu River 15 
East River 
Duhe River 
Fujiang River 
Ganjiang River 
Hanjiang River 
Heishuihe River 
Jinshajiang River 12 
Jiulong River 
Loushui River 
Muli River 
Nanya River 
Qingshui River 
Wujiang River 
Yalong River 
Hengjiang River 
River Basins Rivers Quantity Total 
River Basins in Southeast Region Oujiang River 
Yellow River Region Yellow River 
Longqing River 
Songhua River Region Songhua River 
River Basins in Northwest Region Kashi River 
Kai River 
River Basins in Southwest Region The River 16 
Dayingjiang River 
Red River 
Lancang River 
Niyang River 
Pearl River Region Beipanjiang River 
Hongshui River 
Yangtze River Region Yuanshui River 62 
Zishui River 
Dadu River 15 
East River 
Duhe River 
Fujiang River 
Ganjiang River 
Hanjiang River 
Heishuihe River 
Jinshajiang River 12 
Jiulong River 
Loushui River 
Muli River 
Nanya River 
Qingshui River 
Wujiang River 
Yalong River 
Hengjiang River 

Functional differences statistics

Functional differences from the three items include types of hydropower station, behaviors of reservoir regulation and development task of hydropower projects (Table 2). Seventy-five of the 96 projects were mainly dam type hydropower stations, and 11 projects were the diversion type. The diversion type hydropower stations result in the degradation of downstream ecosystem health more easily than the other two types because of the dewatering and water reduced segment. Fifty-nine projects were daily regulation reservoirs, and numbers of degree of regulation (DOR) of the projects were small. There were several development tasks of each project, and in most of the projects the first task was power generation because they were hydropower projects. The following four tasks were shipping, flooding control, irrigation and water supply. Only five projects set ecological protection as one of the development tasks.

Table 2.

Functional statistics of hydropower projects.

Statistical items Index Quantity 
Hydropower station type Dam type 75 
Mixed type 10 
Diversion type 11 
Behavior of reservoir regulation Incomplete annual regulation 
Incomplete regulation 
Over year regulation 
Seasonal regulation 
Annual regulation 
Weekly regulation 
Daily regulation 59 
No regulation performance 
Development task Power generation 96 
Irrigation 13 
Tourism 
Flood control 18 
Shipping 27 
Sediment interception 
Energy storage 
Water supply 11 
Aquaculture 
Ecological protection 
Re-regulation 
Statistical items Index Quantity 
Hydropower station type Dam type 75 
Mixed type 10 
Diversion type 11 
Behavior of reservoir regulation Incomplete annual regulation 
Incomplete regulation 
Over year regulation 
Seasonal regulation 
Annual regulation 
Weekly regulation 
Daily regulation 59 
No regulation performance 
Development task Power generation 96 
Irrigation 13 
Tourism 
Flood control 18 
Shipping 27 
Sediment interception 
Energy storage 
Water supply 11 
Aquaculture 
Ecological protection 
Re-regulation 

Ecological flow optimization process

Temporal variation of ecological flow objects

Due to functional diversities of ecosystem services for maintaining downstream aquatic health, there are multiple targets to meet, such as the baseflow, baseload electricity generation, shipping baseflow, etc. Table 3 shows the development of ecological flow optimization process in a temporal scale. Before 2000, there was no requirement for ecological flow and stakeholders of most hydropower plants did not take downstream into consideration. There was still no protected target under consideration until 2001, but this situation changed in 2002. Maintaining the baseflow of the downstream reach was the first target to meet, but there was still no guide for quantifying the amount of ecological flow. Until 2006, the need of baseflow was one of the most important targets to meet, integrated with domestic water, shipping water and power generation water requirements. Pollutants dilution water requirements and aquatic organisms were being taken into consideration in the following years in the EIA process, with enrichment of basic data in the river ecosystem monitoring. However, the flow duration curve in the spawning period was hard to quantify. From the view of whole river ecosystem, habitat protection and restoration became more and more important in ecological flow determination. In practice, from 2010 to 2015, there was more consideration for fish habitat restoration or river protection and restoration. With the increasing river ecosystem and fish resources degradation (especially endangered fish species), the demands for ecological flow discharging were increasingly strict, for example, natural flow duration discharge was demanded for maintaining fish spawning on the basis of fish habitat simulation.

Table 3.

Temporal variation of ecological flow targets.

Year Targets to meet 
2001 None 
2002 Baseflow 
2003 Minimum monthly average flow in dry seasons 
2004 Baseflow, baseload electricity generation, water for industry and residents’ life 
2005 Integrated baseflow and domestic water, shipping baseflow 
2006 Baseload power generation, shipping water 
2007 Landscape needs 
2008 Domestic water requirement, industrial and agricultural water requirement, pollutants dilution water requirement 
2009 Aquatic organisms water requirement 
2010 Aquatic organisms water requirement, flow duration curve in spawning period 
2011 Baseflow and shipping needs 
2012 Eco-hydraulics needs 
2013 Habitat protection, integrated natural flow and R2-Cross method 
2014 Habitat protection and restoration 
2015 River restoration and river protection 
Year Targets to meet 
2001 None 
2002 Baseflow 
2003 Minimum monthly average flow in dry seasons 
2004 Baseflow, baseload electricity generation, water for industry and residents’ life 
2005 Integrated baseflow and domestic water, shipping baseflow 
2006 Baseload power generation, shipping water 
2007 Landscape needs 
2008 Domestic water requirement, industrial and agricultural water requirement, pollutants dilution water requirement 
2009 Aquatic organisms water requirement 
2010 Aquatic organisms water requirement, flow duration curve in spawning period 
2011 Baseflow and shipping needs 
2012 Eco-hydraulics needs 
2013 Habitat protection, integrated natural flow and R2-Cross method 
2014 Habitat protection and restoration 
2015 River restoration and river protection 

Spatial diversity of regional ecological flow

The ecological flow was finally determined based on the regional segment characteristics and aquatic organisms, which reflected the difference in ecological flow methods' selection. As statistical results show (Table 4), there are mainly four kinds of ecological flow method in the northern region, which focused on a guaranteed 10% of annual average flow as minimum flow, coupled with a shipping baseflow or baseload power generation flow. In the south region of China with larger runoff, large differences are shown in the methods. More methods were chosen to calculate ecological flow, and an envelope curve formula was used to determine the flow curve as results, even though the Tennant method was still the most used method. There were also more hydraulic and habitat simulation methods being used in the south region of China except for hydrology methods, because there were more endangered fishes in larger rivers like the Yangtze River. Especially in the Sichuan and Yunan province, ecological flow was combined, based on ecology and hydrology. More methods need to be taken into consideration to determine the enveloped curve of south rivers than those of other regions. However, the minimum flow threshold was also more or less 10% of the annual average flow. There is generally ephemeral stream in the northwest and Qinghai-Tibet region, and the river ecosystem in these regions is more sensitive and fragile. So, there were more eco-hydraulics and habitat simulation methods being used to protect the aquatic system.

Table 4.

Spatial diversity of regional ecological flow methods.

Region Method 
Northern Region Jilin ① 5% of annual average flow 
Shaanxi ① Max (10% of annual average flow & shipping baseflow) 
  ② Tennant method 
  ③ Shipping baseflow 
  ④ Baseload power generation flow 
Southern Region Guangxi ① Shipping baseflow 
Guizhou ① 10% of annual average flow 
 ② 23% of annual average flow 
 ③ 5% of annual average flow 
 ④ Baseload power generation flow 
 ⑤ 98% probability of shipping water 
 ⑥ Landscape water demand 
 Hubei ① Minimum monthly average flow in 10 years 
  ② 10% of annual average flow 
  ③ Max (10% of annual average flow & shipping baseflow) 
  ④ Max (10% of annual average flow & industrial and agricultural water & pollutants diluted with water) 
 Jiangxi ① Shipping baseflow 
 Sichuan ① Tennant method 
  ② 10% of annual average flow 
  ③ 5% of annual average flow 
  ④ Shipping baseflow 
  ⑤ Baseload power generation flow 
  ⑥ 10% of average flow in dry season 
  ⑦ Minimum flow in history 
  ⑧ R2-Cross method 
  ⑨ Habitat simulation method 
  ⑩ Wetted perimeter method 
  ⑪ Flow duration curve method 
  ⑫ Eco-hydraulics method 
 Yunnan ① Max (flow duration curve method, wetted perimeter method & eco-hydraulics method) 
  ② 10% of annual average flow 
  ③ Max (10% of annual average flow & shipping baseflow) 
  ④ 15% of annual average flow 
  ⑤ Minimum flow in history 
  ⑥ Eco-hydraulics method 
  ⑦ Max (considering power generation & river ecological water demand & urban water supply) 
 Zhejiang ① Minimum monthly average flow in dry season 
Northwest Region Gansu ① Baseload power generation flow 
Qinghai-Tibet Region Tibet ① 5% of annual average flow 
 ② Eco-hydraulics method 
 Qinghai ① Baseload power generation flow 
  ② Tennant method 
  ③ Eco-hydraulics method 
  ④ Habitat simulation method 
Region Method 
Northern Region Jilin ① 5% of annual average flow 
Shaanxi ① Max (10% of annual average flow & shipping baseflow) 
  ② Tennant method 
  ③ Shipping baseflow 
  ④ Baseload power generation flow 
Southern Region Guangxi ① Shipping baseflow 
Guizhou ① 10% of annual average flow 
 ② 23% of annual average flow 
 ③ 5% of annual average flow 
 ④ Baseload power generation flow 
 ⑤ 98% probability of shipping water 
 ⑥ Landscape water demand 
 Hubei ① Minimum monthly average flow in 10 years 
  ② 10% of annual average flow 
  ③ Max (10% of annual average flow & shipping baseflow) 
  ④ Max (10% of annual average flow & industrial and agricultural water & pollutants diluted with water) 
 Jiangxi ① Shipping baseflow 
 Sichuan ① Tennant method 
  ② 10% of annual average flow 
  ③ 5% of annual average flow 
  ④ Shipping baseflow 
  ⑤ Baseload power generation flow 
  ⑥ 10% of average flow in dry season 
  ⑦ Minimum flow in history 
  ⑧ R2-Cross method 
  ⑨ Habitat simulation method 
  ⑩ Wetted perimeter method 
  ⑪ Flow duration curve method 
  ⑫ Eco-hydraulics method 
 Yunnan ① Max (flow duration curve method, wetted perimeter method & eco-hydraulics method) 
  ② 10% of annual average flow 
  ③ Max (10% of annual average flow & shipping baseflow) 
  ④ 15% of annual average flow 
  ⑤ Minimum flow in history 
  ⑥ Eco-hydraulics method 
  ⑦ Max (considering power generation & river ecological water demand & urban water supply) 
 Zhejiang ① Minimum monthly average flow in dry season 
Northwest Region Gansu ① Baseload power generation flow 
Qinghai-Tibet Region Tibet ① 5% of annual average flow 
 ② Eco-hydraulics method 
 Qinghai ① Baseload power generation flow 
  ② Tennant method 
  ③ Eco-hydraulics method 
  ④ Habitat simulation method 

Ecological flow safeguards optimization

The ecological flow should be released continuously in a year, and many measures have been taken to safeguard it. From the statistic of ecological flow practice, there were mainly four kinds of safeguard measures that made full use of water resources and improved the effectiveness of ecological flow (Table 5). Ecological flow automatic monitoring, forecasting and remote transmission systems have been established in a few hydropower projects, to ensure the authority and accuracy of the data. Meanwhile, it is also one part of ecological flow adaptive management to feed back the release plan and management system. The optimized ecological flow based on monitoring data would be more efficient for downstream ecosystem protection and restoration:

  • 1.

    Based on the layout of hydropower plants, a separated mini-generator set of ecological flow turbines were set into the main part for ecological flow release; at the same time, there were to a certain extent hydroelectric benefits when separate mini-generators were operated.

  • 2.

    Flow was released through power plants by undertaking baseload power generation. However, an ecological flow release schedule needs to be set as a rule for reservoir operation plans, so that ecological flow components can be recognized from the hydroelectric parts.

  • 3.

    There is a need for separated ecological flow release facilities, such as ecological flow diversion facilities, ecological flow drainage hole, spillway or flushing gallery, ecological flow drainage pipe, etc. The operation plan of these facilities should also be cleared in the design period.

  • 4.

    A combination of ecological flow turbine and measures in item 3 above, the combined measures were more effective in protecting ecological flow because there are multiple measures, and it has been widely used in projects in recent years.

Table 5.

Statistic of ecological flow safeguards' development.

Year Ecological flow safeguard measures 
2001 None 
2002 None 
2003 ① Separated mini-generator set 
 ② Ecological flow diversion facilities 
2004 ① By baseload power generation 
 ② Ecological flow diversion facilities 
 ③ Ecological flow drainage hole 
 ④ Spillway or flushing gallery 
2005 ① By baseload power generation 
 ② Ecological flow turbines 
 ③ Ecological flow drainage hole 
2006 ① By baseload power generation 
 ② Separated mini-generator set 
 ③ By baseload power generation 
 ④ Ecological flow drainage hole 
 ⑤ Ecological flow drainage pipe 
 ⑥ Combining sediment flushing outlets and spillway 
2007 ① By baseload power generation 
 ② Ecological flow drainage pipe 
2008 ① Separated mini-generator 
 ② Ecological flow turbines 
 ③ Ecological flow drainage hole 
2009 ① By baseload power generation 
 ② Ecological flow turbines 
 ③ Ecological flow spillway 
2010 ① By baseload power generation 
2011 ① By baseload power generation 
 ② Reservoir joint operation 
 ③ Ecological flow turbine combined with optimal operation 
2012 ① By baseload power generation 
 ② Ecological flow discharge sluice 
2013 ① Ecological flow drainage hole 
 ② Ecological flow turbines 
 ③ Ecological flow discharge sluice 
 ④ Ecological flow spillway 
2014 ① Ecological flow drainage pipe 
 ② Ecological flow turbine combined with discharge sluice 
2015 ① Ecological flow turbines 
 ② Ecological flow discharge sluice 
 ③ By baseload power generation 
 ④ Ecological flow drainage pipe 
 ⑤ Multiple combination of above measures 
Year Ecological flow safeguard measures 
2001 None 
2002 None 
2003 ① Separated mini-generator set 
 ② Ecological flow diversion facilities 
2004 ① By baseload power generation 
 ② Ecological flow diversion facilities 
 ③ Ecological flow drainage hole 
 ④ Spillway or flushing gallery 
2005 ① By baseload power generation 
 ② Ecological flow turbines 
 ③ Ecological flow drainage hole 
2006 ① By baseload power generation 
 ② Separated mini-generator set 
 ③ By baseload power generation 
 ④ Ecological flow drainage hole 
 ⑤ Ecological flow drainage pipe 
 ⑥ Combining sediment flushing outlets and spillway 
2007 ① By baseload power generation 
 ② Ecological flow drainage pipe 
2008 ① Separated mini-generator 
 ② Ecological flow turbines 
 ③ Ecological flow drainage hole 
2009 ① By baseload power generation 
 ② Ecological flow turbines 
 ③ Ecological flow spillway 
2010 ① By baseload power generation 
2011 ① By baseload power generation 
 ② Reservoir joint operation 
 ③ Ecological flow turbine combined with optimal operation 
2012 ① By baseload power generation 
 ② Ecological flow discharge sluice 
2013 ① Ecological flow drainage hole 
 ② Ecological flow turbines 
 ③ Ecological flow discharge sluice 
 ④ Ecological flow spillway 
2014 ① Ecological flow drainage pipe 
 ② Ecological flow turbine combined with discharge sluice 
2015 ① Ecological flow turbines 
 ② Ecological flow discharge sluice 
 ③ By baseload power generation 
 ④ Ecological flow drainage pipe 
 ⑤ Multiple combination of above measures 

In the statistics of 96 hydroelectric projects (Figure 4), 79 of them clearly implemented ecological flow safeguard measures and the percentage showed an increasing trend. Among them, 44 projects implemented safeguard measures in which there were dewatering reaches in the downstream, and there were 67 projects with endangered fishes in the downstream reaches.

Fig. 4.

Variation amount of projects with ecological flow safeguards.

Fig. 4.

Variation amount of projects with ecological flow safeguards.

Improvement of minimum ecological flow

Minimum flow threshold development

Ecological flow management is a compromise proposal of research, practice and management. From 2001 to 2006, ecological flow calculation methods were not clear; therefore, several projects did not set ecological flow and safeguards. After 2006, the main method was based on ‘Letter [2006] No.4’ as the threshold demand of government management. The method is 10% of the annual average flow as the minimum ecological flow threshold, taking other targets into account to determine the ecological flow, such as downstream aquatic organisms, shipping, pollution dilution, sediment flushing, etc. Stakeholders of hydroplants lack the initiative to release ecological flow so they will choose the minimum flow threshold as the implemented guide. Meanwhile, it is difficult for them to optimize the ecological flow because the management system was not perfect. It was not until 2012 that the ecological flow duration curve became an important part of implementation due to ‘Letter [2012] No.4’.

In the development process of hydropower practice and management, the minimum flow threshold and ecological flow duration curve were the two main indices for EIA, and the minimum flow threshold determination and its optimization method were the focus content of this paper (Figure 5). The projects were concentrated in Sichuan and Yunnan provinces in the southwest of China, where 79 of the total amount set flow thresholds and safeguards. For most of them, 10% of the annual average flow was the minimum flow threshold of the ecological flow, and the ecological flow value was approximately 15–20% of the annual average flow because of the combination with other water targeting needs. It is difficult to put this into practice even if there is a complex ecological flow release curve for downstream fishes, while flow threshold was one of the most effective indexes for management. We could summarize from the figure that the minimum ecological flow threshold has raised more than 10% of annual average flow, and the fluctuation ranges of maximum and minimum ecological flow reduced after 2008. The average of ecological flow was also affected by the amount of projects, so the value did not change too much in different years.

Fig. 5.

Minimum flow threshold development of hydropower projects.

Fig. 5.

Minimum flow threshold development of hydropower projects.

Regression and correlation analysis

On a large scale, like national scale or basin scale, the current minimum ecological flow makes it difficult to meet ecological water requirements in different regions. In global water assessment, the calculated global annual EFR for fair ecological conditions represent between 25% and 46% of the mean annual flow (MAF) (Pastor et al., 2014), and as a result of adaptive management, the management threshold value should be increased and the flow duration curve should be strengthened. In the statistics of ecological flow, multiple targets including aquatic organisms have been taken into consideration which mainly followed the ecological flow assessment framework. Minimum flow thresholds of ecological flow have been set in most projects and the ecological flow duration curve has been set in 19 projects, mainly considering fish spawning, breeding and other sensitive periods. It is an effective index to calculate the percentage of annual average flow as the ecological flow threshold both in the Tennant method and in the China ecological flow management system. So, we also decided to make the correlation between the minimum threshold value and annual average flow, and then propose a new minimum flow threshold value of ecological flow.

Through establishing the correlation between the minimum flow threshold and annual average flow, a regression equation in difference periods has been established (Figure 6). The 12 months have been divided into two parts, general period and special period (fish spawning, breeding and other sensitive periods). For the general period and special period, it shows a linear fit of annual average flow and the equation was y=0.25x–31, y=0.255x+2.7. As a result, we could conclude that 25% of the annual average flow could be the minimum flow threshold, while it is much larger than 31 m3/s in the general period and 2.7 m3/s in the special period. When the annual average flow is not much larger than 31 m3/s, we should consider the frequency of the flow, to determine whether 25% of the annual average flow was still suitable for it.

Fig. 6.

Correlation of ecological flow threshold and annual average flow.

Fig. 6.

Correlation of ecological flow threshold and annual average flow.

From another point of view such as management demand, the frequency of minimum flow threshold value was analyzed for further adaptive management targets (Figure 7). If 25% of the annual average flow is set as the new flow threshold, there are only 17 projects which could reach the new target. From Figure 6, we could explain that 25% of annual average flow could be the minimum flow threshold in the general period when the annual average flow is much larger than 31 m3/s. This is the theoretical result from the regression equation. Also from Figure 7, from the frequency distribution of minimum flow threshold value, we can find that at the point of 25% of annual average flow, over 90% of the projects that were operated for years could not reach this standard. In the current situation, the threshold values were already 10% of the annual average flow, so we suggested an inflection point at 17% of the annual average flow as the new threshold, while half of the projects could meet the minimum ecological flow threshold, this is from the viewpoint of management. So it reached a compromise between the theoretical results and management practice, 17% of annual average flow was chosen while the annual average flow was not much larger than 31 m3/s. Even though the minimum ecological flow threshold was raised, which would be more eco-friendly, it also needs to be tested by the further adaptive management of hydropower projects. This paper was just analyzing the possibility of raising the threshold value, and how it could be in the management system.

Fig. 7.

Frequency distribution of minimum flow threshold value.

Fig. 7.

Frequency distribution of minimum flow threshold value.

Conclusion

In conclusion, we found an improvement of ecological flow development from the 15 years’ practice of adaptive management. The minimum ecological flow threshold standard improved from none to 10% of annual average flow, and the ecological flow duration curve had been proposed for protecting downstream objects in the special period. Ecological flow had been developed from guaranteeing the minimum flow, baseload hydropower flow and shipping baseflow to protect and restore the fish habitat and river ecosystem. Most of the projects had built ecological flow safeguard measures (including engineering measures and non-engineering measures) for guaranteeing the flow release, especially projects with endangered fish species in downstream reaches and projects with dewatering reaches downstream of the dam. Despite achievements in the last 15 years, there were also many problems existing in the practice of adaptive management. For river ecosystem restoration, we proposed to level up the minimum ecological flow threshold standard to 25% of annual average flow, and to 17% of the annual average flow in the general period as the minimum flow threshold value when the flow is not much larger than 31 m3/s. To quantify the flow, the minimum flow standard should be determined and safeguard measures should be made. In the process of flow determination, spatial and temporal diversity should be analyzed and factors such as aquatic organism differences and dam types should be taken into consideration. Cascade reservoirs operation should be the measures to prevent dewatering reaches in the future, and ecological flow should be one of the indicators in river protection and restoration practice.

Acknowledgements

This study was funded by General Financial Grant from the China Postdoctoral Science Foundation (2016M592404).

References

References
Arthington
,
A. H.
,
Tharme
,
R. E.
,
Brizga
,
S. O.
,
Pusey
,
B. J.
,
Kennard
,
M. J.
,
Welcomme
,
R. L.
&
Petr
,
T.
, (
2004
).
Environmental flow assessment with emphasis on holistic methodologies
.
Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries
19
(
1
),
74
81
.
Chen
,
A.
,
Sui
,
X.
,
Wang
,
D.
,
Liao
,
W.
,
Ge
,
H.
&
Tao
,
J.
, (
2016
).
Landscape and avifauna changes as an indicator of Yellow River Delta Wetland restoration
.
Ecological Engineering
86
,
162
173
.
Chen
,
A.
,
Wu
,
M.
,
Chen
,
K.
,
Sun
,
Z.
,
Shen
,
C.
&
Wang
,
P.
, (
2017
).
Main issues in environmental protection research and practice of water conservancy and hydropower projects in China
.
Water Science and Engineering
4
(
9
),
312
323
.
Eriyagma
,
N.
&
Jinapala
,
K.
, (
2014
).
Developing tools to link environmental flows science and its practice in Sri Lanka
. In:
Icwrs2014
,
Bologna
,
Italy
,
June
, pp.
204
209
.
Fitzhugh
,
T. W.
, (
2015
).
Efcam: a method for assessing alteration of environmental flow components
.
River Research & Applications
30
(
7
),
825
844
.
King
,
A. J.
,
Gawne
,
B.
,
Beesley
,
L.
,
Koehn
,
J. D.
,
Nielsen
,
D. L.
&
Price
,
A.
, (
2015
).
Improving ecological response monitoring of environmental flows
.
Environmental Management
55
(
5
),
991
1005
.
Kirby
,
J. M.
,
Connor
,
J.
,
Ahmad
,
M. D.
,
Gao
,
L.
&
Mainuddin
,
M.
, (
2014
).
Climate change and environmental water reallocation in the Murray–Darling Basin: impacts on flows, diversions and economic returns to irrigation
.
Journal of Hydrology
518
(
Part A
),
120
129
.
Latu
,
K.
,
Malano
,
H. M.
,
Costelloe
,
J. F.
&
Peterson
,
T. J.
, (
2014
).
Estimation of the environmental risk of regulated river flow
.
Journal of Hydrology
517
,
74
82
.
Liu
,
X.
,
Lian
,
Y.
,
Huang
,
J.
,
Ke
,
S.
,
Zhang
,
J.
&
Wang
,
X.
, (
2008
).
Environmental flows of the Yellow River
.
Science and Technology Review
17
,
24
30
.
Maloney
,
K. O.
,
Talbert
,
C. B.
,
Cole
,
J. C.
,
Galbraith
,
H. S.
,
Blakeslee
,
C. J.
,
Hanson
,
L.
&
Holmquist-Johnson
,
C. L.
, (
2015
).
An integrated riverine environmental flow decision support system (REFDSS) to evaluate the ecological effects of alternative flow scenarios on river ecosystems
.
Fundamental and Applied Limnology
186
(
1–2
),
171
192
.
Martin
,
D. M.
,
Harrison Atlas
,
D.
,
Sutfin
,
N. A.
&
Poff
,
N. L.
, (
2014
).
A social-ecological framework to integrate multiple objectives for environmental flows management
.
Journal of Contemporary Water Research & Education
153
(
1
),
49
58
.
Matthews
,
J. H.
,
Forslund
,
A.
,
McClain
,
M. E.
&
Tharme
,
R. E.
, (
2014
).
More than the fish: environmental flows for good policy and governance, poverty alleviation and climate adaptation
.
Aquatic Procedia
2
,
16
23
.
Olsen
,
M.
,
Troldborg
,
L.
,
Henriksen
,
H. J.
,
Conallin
,
J.
,
Refsgaard
,
J. C.
&
Boegh
,
E.
, (
2013
).
Evaluation of a typical hydrological model in relation to environmental flows
.
Journal of Hydrology
507
,
52
62
.
Pastor
,
A. V.
,
Ludwig
,
F.
,
Biemans
,
H.
,
Hoff
,
H.
&
Kabat
,
P.
, (
2014
).
Accounting for environmental flow requirements in global water assessments
.
Hydrology and Earth System Sciences
18
(
12
),
5041
5059
.
Shekhar
,
S.
, (
2016
).
Environmental flows and river rejuvenation
.
Journal of the Geological Society of India
88
(
6
),
813
814
.
Steinfeld
,
C. M. M.
,
Kingsford
,
R. T.
,
Webster
,
E. C.
&
Sharma
,
A.
, (
2015
).
A simulation tool for managing environmental flows in regulated rivers
.
Environmental Modelling and Software
73
(
C
),
117
132
.
Summers
,
M. F.
,
Holman
,
I. P.
&
Grabowski
,
R. C.
, (
2015
).
Adaptive management of river flows in Europe: a transferable framework for implementation
.
Journal of Hydrology
531
(
Part 3
),
696
705
.
Tavassoli
,
H. R.
,
Tahershamsi
,
A.
&
Acreman
,
M.
, (
2014
).
Classification of natural flow regimes in Iran to support environmental flow management
.
Hydrological Sciences Journal
59
(
3–4
),
517
529
.
Wang
,
J.
,
Dong
,
Z.
,
Liao
,
W.
,
Li
,
C.
,
Feng
,
S.
,
Luo
,
H.
&
Peng
,
Q.
, (
2013
).
An environmental flow assessment method based on the relationships between flow and ecological response: a case study of the three gorges reservoir and its downstream reach
.
Science China Technological Sciences
56
(
6
),
1471
1484
.
Wang
,
B.
,
Shao
,
D.
,
Mu
,
G.
,
Wang
,
Z.
&
Li
,
X.
, (
2016
).
An eco-functional classification for environmental flow assessment in the Pearl River Basin in Guangdong, China
.
Science China Technological Sciences
59
(
2
),
265
275
.
Warner
,
R. F.
, (
2014
).
Environmental flows in two highly regulated rivers: the Hawkesbury Nepean in Australia and the Durance in France
.
Water and Environment Journal
28
(
3
),
365
381
.
Wu
,
M.
,
Shi
,
P.
,
Chen
,
A.
,
Shen
,
C.
&
Wang
,
P. Y.
, (
2017
).
Impacts of DEM resolution and area threshold value uncertainty on the drainage network derived using SWAT
.
Water SA
43
(
3
),
450
462
.