Scenario Simulation and Efficiency Study of Hydropower Development to Promote Regional Sustainable Development: An Empirical Analysis of a Province in Southwestern China

Abstract

Hydropower is a significant component of China’s contemporary energy framework, with its construction and operation critically contributing to the advancement of sustainable development in the region. However, the influence of hydropower development on regional sustainable development is not evident, and the course of its effect is not clear. In this work, on the basis of assessing the effect of hydropower development on regional sustainable development, a system dynamics (SD) simulation model was created, and 15 distinct development modes were set and tested. The relative driving efficiency of hydropower growth on sustainable development in a province in Southwest China under several scenarios was assessed using the super-efficiency-slacks-based measurement (Super-SBM), and an efficiency analysis was undertaken. The study’s findings demonstrate that: (1) Hydropower development had a complex driving influence on sustainable development in the province in 2015–2022. (2) The relative driving efficiency of the province over the study period exhibited a changing pattern and was at a high level greater than 0.7. (3) New hydropower investment and hydropower generation are the major factors impacting the relative driving efficiency of hydropower development on regional sustainable development in the province. The study’s findings provide a platform and reference for supporting regional sustainable development through hydropower development.

1. Introduction

Hydropower, a clean and low-carbon renewable energy source with mature technology, flexible application, and low cost, is an important part of the modern energy system [1,2] and plays an important role in reducing the dependence on fossil energy, improving production modes, optimizing resource management, and driving the sustainable development of regions [3]. With the continued promotion of regional sustainable development and the implementation of national green development policies, energy development and utilization have become increasingly focused on efficiency enhancement, cleanliness, and low carbon. Hydropower development is regarded as an important production method for clean energy and an important means of securing regional energy supply. It is considered a key solution to meet the growing demand for electricity and to increase renewable energy sources [4], thus it is being given more important responsibilities and higher requirements.

According to research by the International Hydropower Association (IHA), the total global installed hydropower capacity continued its growing speed from 1272 GW to 1360 GW during 2017–2021 [5]. Hydropower may supply electricity and non-energy services (flood control, irrigation, etc.) to regions [6] and has helped the growth and prosperity of many nations in a non-negligible way. In New Zealand [7], where hydropower provided over 60% of the total electricity in 2023, hydropower projects have been developed to accommodate the rising demand for electricity. In Sub-Saharan Africa (SSA) [8], hydropower is regarded as a key component of local development, with major hydropower taking the lion’s share of energy supply in Zambia. In Malaysia [9], hydropower accounted for 67.2% of renewable energy in 2015 and is the country’s backbone. The influence of hydropower on regional development has been widely explored by academics, and various researchers have focused on analyzing and confirming the sustainability of hydropower. Bingsheng Liu et al. [10] presented a time-series-based multi-attribute decision-making framework for identifying sustainable hydropower development scenarios. Yan Zhang [11] employed an equilibrium approach to equate hydropower to a ‘power bridge’ and claimed that hydroelectricity is a contentious source of energy for sustainable development. Mohd Alsaleh et al. [12] indicated that expanding sustainable hydropower would have an influence on land-use change in the EU. Sustainable hydropower arises from a concern for sustainable development. Sustainable development is development that satisfies the requirements of the present without compromising the ability of future generations to meet their own needs [13]. Jiwei ZHU et al. [14] pointed out that minimizing the use of natural resources and renewable energy can help achieve ecological environmental protection and sustainable development. Hailong Du et al. [15] noted that logical hydropower project planning may make hydropower growth sustainable. Therefore, rational hydropower development is a vital strategy to promote sustainable development in the regions.

Exploring the role of hydropower development in fostering regional sustainable development is of considerable relevance in encouraging the building of China’s modern energy system and low-carbon sustainable development. Many approaches have been utilized to analyze the interactions between complex systems, such as system dynamics [16], structural equations [17], and quantitative research methods [18]. Compared to other methodologies, system dynamics can abstract and simulate complicated systems using stock and flow diagrams and mathematical equations, which is an advantage in quantitative analysis. Data Envelopment Analysis (DEA) is able to quantify the efficiency of multi-input, multi-output decision-making units (DMUs) and has been widely employed in the domains of power generation efficiency [19], carbon emission reduction [20], urban greening [21], and regional transport sustainability [22]. SBM adds slack variables to the traditional DEA. The super-efficiency SBM (Super-SBM) is an upgrade of SBM, which overcomes the incapacity of the original SBM to compare the entire efficiency (efficiency value equal to 1) of different DMUs [23]. Super-SBM can estimate an efficiency value greater than 1, allowing for a comparison of DMUs with full efficiency [24].

Hydropower is China’s primary source of renewable energy [25], and it plays an important role in ensuring regional energy supply and driving sustainable regional development. However, the link between hydropower development and regional sustainable development has yet to be understood, and the direction of its involvement is still uncertain. Therefore, this study aims to explore the impact relationship of hydropower development on regional sustainable development, identify the path of impact, and measure the relative driving efficiency of hydropower development on regional sustainable development by using different scenarios set as decision-making units. The contributions of this study are as follows:

(1)

The SD model was built, which can dynamically explain the interplay between hydropower growth and regional sustainable development.

(2)

The relative driving efficiency of hydropower development on regional sustainable development was examined using the Super-SBM model.

(3)

The technical efficiency (TE) curves, inputs, and outputs of several models were evaluated, and two essential factors for boosting regional sustainable development were filtered out.

2. Materials and Methods

2.1. Data Sources

The data for this study were acquired mostly from numerous statistics yearbooks and data portals, including the China Macroeconomic Database, the China Finance and Taxation Database, the China Electricity Statistical Yearbook, the China Environmental Statistical Yearbook, the province’s Statistical Yearbook, etc. The price of water and the price of energy usage were picked from the price monitoring table issued by the Development and Reform Commission of the province.

2.2. Methodology

2.2.1. Entropy Weight Method and System Dynamics

The entropy weight technique is an objective method of indicator assignment with the assistance of the information entropy of indicator data to estimate the relative weights of indicators [26]. In this work, the entropy weight approach is utilized to determine the weights of the indicators of the water electronic system. System dynamics is a thorough multidisciplinary study of nonlinear feedback systems [27], with the assistance of which the SD model of the interaction between hydropower development and regional sustainable development is created.

2.2.2. Super-SBM

As a non-parametric efficiency assessment approach, the Data Envelopment Analysis (DEA) method is frequently used to quantify the relative efficiency of decision units with many inputs and outputs [28]. Many derivative models have emerged from modern DEA, one of which is slacks-based measurement (SBM), proposed by Tone [29], which can directly reflect the improvement of slacks in efficiency measures with excess inputs and shortfalls in outputs [30], and, at the same time, is both input- and output-oriented in addition to non-oriented. Compared with classic DEA models, SBM can prevent radial and angle-induced bias to some extent [31]. Super-SBM is a further development of SBM, which may compare DMUs with the same relative efficiency of 1. The non-oriented, global Super-SBM model [32], incorporating non-desired outcomes, is illustrated in Equation (1).

$$\begin{gathered} min{\gamma }^{*}={\displaystyle \frac{1+\sum _{i=1}^m{(s}_i^{-}/x_{ik}^t)/\left(m\right)}{1-\left(\sum _{r=1}^{q_1}{s}_r^{+}/y_{rk}^t+\sum _{C=1}^{q_2}{s}_c^{-}/y_{ck}^t\right)/(q_1+q_2)}} \\ \mathit{\text{s.t.}}\left\{\begin{array}{l}{x}_{ik}^t\ge {\sum }_{t=1}^T{\sum }_{j=1,j\ne k}^n{x}_{ij}^t{\lambda }_j-s_i^{-},i=1,\cdots ,m\\ y_{rk}^t\le {\sum }_{t=1}^T{\sum }_{j=1,j\ne k}^n{y}_{rj}^t{\lambda }_j+s_r^{+},r=1,\cdots ,q_1\\ y_{ck}^t\ge {\sum }_{t=1}^T{\sum }_{j=1,j\ne k}^n{y}_{cj}^t{\lambda }_j-s_c^{-},c=1,\cdots ,q_2\\ 1-({\sum }_{r=1}^{q_1}{s}_r^{+}/y_{rk}^t+{\sum }_{C=1}^{q_2}{s}_c^{-}/y_{ck}^t)/(q_1+q_2)>0\\ \hfill {\lambda }_{ij},{\mathrm{s}}_i^{-},{\mathrm{s}}_c^{-},{\mathrm{s}}_r^{-}\ge 0\hfill \end{array}\right. \end{gathered}$$

(1)

where ${\gamma }^{\mathrm{*}}$ is the efficiency value and $n$ is the total number of DMUs, the model contains $m$ input indications, $q_1$ desired outputs, and $q_2$ non-desired outputs. $x_{ij},y_{rj},\text{ and }{y}_{cj}$ are the i-th class of inputs, the r-th class of desired outputs, and the c-th class of non-desired outputs, respectively, of the decision unit j. $s_i^{-},s_r^{+},\text{ and }{s}_c^{-}$ are the slack variables.

3. Results

3.1. Building and Testing the SD Model

3.1.1. Analysis of the Structure of the SD Model

In this study, the research object is separated and divided into the hydropower development subsystem and the regional sustainable development subsystem. Referring to recent research [33,34], the regional sustainable development subsystem is generally characterized by aspects of economic growth, social stability, and environmental friendliness. The indices of economic growth may be roughly classified as economic size, industrial economy, consumer dynamics, and government finances. Indicators of social stability can be generically classified as scientific research, social security, population and employment, infrastructure, and transport communications. Environmentally friendly indicators are essentially classified as pollution control, investment in control, and efficacy of control. The hydropower development subsystem, on the other hand, consists of four dimensions: investment, construction, output, and consumption. Figure 1 presents a structural analysis of the system relevant to this research to elucidate the modeling concepts. The indicators within the hydroelectric development sub-system aim to impact regional development by affecting the pertinent indicators in the subsystems of society, economics, and environment. Figure 1 incorporates the Sustainable Development Goals (SDGs) symbols to examine the SDGs relevant to hydropower development, drawing on recent research [35,36]. It is important to acknowledge that hydropower development contributes to just some Sustainable Development Goals (SDGs).

Hydropower development has an impact on achieving the SDGs of clean energy, clean water, decent employment, industrial innovation, climate action, and a terrestrial ecosystem. The hydropower development subsystem directly affects or constitutes a sustainable development system, influencing regional sustainability by providing power security, reducing carbon emissions, promoting economic and financial growth, optimizing industrial structure, and influencing regional employment [37]. Enhancing hydropower production’s sustainability will have a direct impact on regional sustainable development. Hydropower, in particular, will help reduce the use of fossil energy compared to the generation of an equivalent quantity of thermal power, thereby reducing greenhouse gas emissions and helping to mitigate climate change [38]. At the same time, the operation of the power plant will provide more clean energy for the region [39], which will stimulate employment and industrial innovation as well as optimize the industrial structure [40], thereby supporting sustainable development in the region.

3.1.2. SD Model Boundaries and Hypotheses

The primary difficulty that has to be solved to construct a system dynamics model is to mimic the process of sustainable development in an area that includes hydropower development, and it is not feasible to identify all the influencing components since the real system belongs to a complex system. Therefore, system boundaries are specified, and fundamental hypotheses are made to facilitate the modeling process. The SD model’s temporal boundary is 2015–2022, whereas the geographic boundary is the administrative boundary of a province in southwest China. The essential hypotheses are as follows: (1) Hydropower development comprises both the construction and operating eras, i.e., the province is always in the state of “hydropower development” during the sample period. (2) The system dynamics technique only has to examine the core problem; thus, it is sufficient to pick the core influencing components to construct a closed loop in the modeling process. (3) The core indicators of the subsystems can indicate the dynamic behavior of the subsystems, i.e., the development of the subsystems can be deduced by utilizing the core variables pertaining to the subsystems.

3.1.3. Illustrating and Analyzing Stock and Flow Diagrams

This study’s stock and flow diagram is a visual representation of the SD model. Compared with the causal loop diagram, the stock and flow diagram has a more rigorous mathematical logic since it can be directly replicated. The stock and flow diagram is given in Figure 2. Since the statistical time period of the statistics obtained in this study is 1 year, the simulation step was set to 1 year as well. Referring to the existing studies [41,42], the equations of the variables in the stock and flow diagram were mainly set as table functions (e.g., GDP growth rate of the primary industry, GDP growth rate of the secondary industry), numerical fitting equations (e.g., turnover of the technological market, greening coverage rate of built-up areas), and ordinary equations (e.g., impact value of the tertiary industry, year-end resident population), and so on.

In addition, the setting of the original hydropower index equation is based on the weights of the indicators computed by the entropy weighting approach. For indicators that had an effect relationship conceptually but were not numerically clear, this research chose to link the two, but it is sufficient to lessen the influence of the input indicators in the specification of the equations so that they form a closed loop. It is worth mentioning that the influence of hydropower development on regional sustainable development is correlative rather than causative.

The eight indicator variables using data fitting were selected to test the simulated and actual value errors, including the value added of wholesale and retail trade; industrial water; technology market turnover; value added of transport, storage, and postal services; additional area for soil erosion control; road mileage; greening coverage rate of built-up areas; and investment in industrial waste gas treatment. As demonstrated in Figure 3, the absolute magnitude of the error between the simulated and real values of these eight indicators is less than 10 percent. The historical data test findings even suggest that this study’s SD model can efficiently and reasonably accurately predict the link between hydropower development and sustainable development in the province.

3.2. Calculating and Analyzing the Efficiency of Hydropower for Sustainable Regional Development

3.2.1. Measurement Metrics for the Super-SBM Model

The effect of the efficiency of hydropower development on regional sustainable development was measured with the use of the indicator system of the aforementioned SD model. Based on the concept of the DEA technique [43,44], the representative and strongly correlated indicators in hydropower development and regional sustainable development subsystems were selected as the measurement indicators of the Super-SBM model. The indicators of the hydropower development subsystem were set as input variables of the model, including new hydropower investment, installed hydropower capacity, and hydropower generation, coded as $I_1$, $I_2$, and $I_3$, respectively. The indicators of the total regional GDP impact value, installed capacity, power generation, electricity consumption, and area of soil erosion control in the sustainable development system were set as output variables, coded as $O_1,O_2,O_3,O_4$, and $O_5$, respectively. The filtered efficiency measurement indicator system is provided in

Table 1. Indicator system for efficiency measurement of the Super-SBM model.

Category	Variables	Code	Unit
Input	New hydropower investment	$I_1$	100 million CNY
	Installed hydropower capacity	$I_2$	10,000 kW
	Hydropower generation	$I_3$	100 million kWh
Output	Total regional GDP impact value	$O_1$	100 million CNY
	Installed capacity	$O_2$	10,000 kW
	Power generation	$O_3$	100 million kWh
	Electricity consumption	$O_4$	100 million kWh
	Area of soil erosion control	$O_5$	thousand hectares

.

3.2.2. Setting up Multiple Hydropower Development Modes

In view of the requirement for the number of DMUs when the Super-SBM model is employed for efficiency measurement [45], distinct development modes were specified so that they generated 120 DMUs. Distinct development modalities were treated as panel data in the efficiency measurement procedure. The mode setting was carried out on the basis of developing the SD model, and the objective of establishing different modes was achieved by modifying the values of input indicators in the Super-SBM indicator system. The results of the modification of the input indicators were merged in an unordered method, and the outcomes are displayed in Figure 4, setting a total of 15 development modes. White circles reflect unaltered values of indicators, whereas colored circles need to be updated. There are a total of 14 development modes ($S_1~S_{14}$) for altering the values of the input indicators, with the indicator values of $S_1~S_7$ being 120% of the original values and the indicator values of $S_8~S_{14}$ being 80% of the original values. The base model’s values ($S_{15}$) are based on historical data for a province in southwest China from 2015 to 2022. In the SD model, $I_1$, $I_2$, and $I_3$ represent the new hydropower investment, installed hydropower capacity, and hydropower generation, respectively.

3.2.3. Examining Hydropower’s Effectiveness in Contributing to Regional Sustainable Development

(1)

Macro facilitation

Based on the indicator system and development mode of the Super-SBM model constructed in the previous section, different development modes were simulated by the SD model, and the simulated values of development modes, such as $S_1~S_{15},$ were obtained. Differences between the indicator values of $S_7$ and $S_{15}$ (the results of which differ very little from the differences between $S_{15}$ and $S_{14}$) are displayed in Figure 5f for both input and output indicators. As indicated, Figure 5a–e illustrate the values of installed capacity, electricity generation, electricity consumption, additional area for soil erosion control, and the total impact value of regional GDP under the three modes ($S_7,S_{14},S_{15}$), respectively. The output indicator value corresponding to $S_{15}$ is between $S_7$ and $S_{14}$, implying that the change in the input indicator is positively associated with the change in the output indicator. The difference between the indicators $S_7$ and $S_{15}$ corresponding to Figure 5f is greater than 0, which means that in the SD model set up in this study, an overall increase in the values of the indicators in the dimension to which the hydropower development sub-system belongs will lead to an increase in the values of the affected indicators in the regional sustainable development system. The results of the model run indicate that hydropower development has a positive macro-level contribution to regional sustainable development. This result aligns with the findings of references [46,47] and accurately reflects the factual reality of the province.

(2)

Efficiency analysis

We used the SD model to simulate the development modes outlined above, and we obtained simulation data for each mode. When using DEA to assess efficiency, standardizing or normalizing the data will skew it. Therefore, the simulation results were immediately employed to carry out the relative efficiency evaluation by iDEA Ultra Data Envelopment Analysis Software v5 (Software Copyright Registration No. 2023SR1344621). A total of 120 DMUs for the 15 development models were considered as 15 sets of panel data, and the Super-SBM model, which is non-oriented, globally referenced, and does not take into consideration the non-desired outcomes, was chosen to carry out the computations. Setting the returns to scale variable provides relative efficiencies, as indicated in Figure 6, and the efficiency selected was technical efficiency (TE). The set development model was separated into two groups for analysis. The first group is the development model with a larger input scale than $S_{15}$, which includes $S_1,S_2,S_3,S_4,S_5,S_6$ and $S_7$, as shown in Figure 6e. The second set of development models, with input sizes less than $S_{15}$, includes $S_9,S_{10},S_{11},S_{12},S_{13}$ and $S_{14}$, as seen in Figure 6f. On the other hand, $S_{15}$ serves as a benchmark curve, and its technical efficiency value exceeds 0.7.

The first group (Figure 6e) has efficiency effective values (≥1) spread mostly in the years 2016, 2018, 2019, and 2022, and the development patterns with technical efficiency roughly above $S_{15}$ are $S_3,S_5,S_6,$ and $S_7$. $S_3$ and $S_5$ are more clearly above $S_{15}$ (Figure 6a), and the technical efficiency values in 2019 and 2020 have a slight rise in efficiency values compared to $S_{15}$. $S_6$ and $S_7$ (Figure 6c) exhibit minor improvements in efficiency values compared to $S_{15}$; however, $S_1$ and $S_4$ (Figure 6c) have lower efficiency values than $S_{15}$ at specific time intervals. The efficiency frontiers in the second group (Figure 6f) are mainly concentrated in the years 2016, 2019, and 2022, and the technical efficiencies of almost all the development models in the second group are at the upper end of the $S_{15}$ curve, especially $S_9$ and $S_{11}$ (Figure 6b), followed by $S_{13}$ and $S_{14}$ (Figure 6d), and then by $S_8$, $S_{10}$, and $S_{12}$ (Figure 6d).

Comparing the efficiency value curves in the two groups, we observed several intriguing features. (1) The efficiency curves of $S_3$ and $S_5$ are greater than $S_{15}$ for the same increase in the value of input indicators, whereas the efficiency curves of $S_1$ and $S_4$ are lower than $S_{15}$. (2) Decreasing the value of input indicators leads to greater measured technical efficiency than that of $S_{15}$, and this phenomenon happens in the efficiency curves of the second group, with the curves of $S_9$ and $S_{11}$ being the more noticeable ones. (3) The time points at which altering the input indicators creates the technical efficiency change are predominantly 2018, 2020, and 2021. Figure 4 and Figure 5 may assist us in explaining these events. Combining phenomenon (1), phenomenon (2), and Figure 4, raising the value of $I_3$ and not reducing the value of $I_3$ would make the efficiency curves of $S_3$, $S_5$, $S_9$, and $S_{11}$ much greater than those of $S_{15}$. Increasing the value of $I_1$ makes the efficiency curves of $S_1$ and $S_4$ lower than that of $S_{15}$, while lowering the value of $I_1$ makes the efficiency curves of $S_8$ and $S_{11}$ higher than that of $S_{15}$,illustrating the role of the alteration of $I_3$ and $I_1$ in boosting the efficiency of hydropower development and fostering sustainable regional development. Combining phenomenon (3), phenomenon (4), and Figure 5, it is concluded that the relationship between output variables and input variables is complex and non-linear, and the proportion of increase in output variables is not only affected by the input variables; for example, the input variables are increased by 20 percent in all of $S_7$ in Figure 4, but the proportion of change in the output indicators is either higher or lower than 20 percent in different years.

4. Discussion

In this work, the combination of SD and Super-SBM takes a major step forward in investigating the driving effect of hydropower growth on regional sustainable development. Based on the objective statistical data of a province in Southwest China, the SD model of the influence of hydropower development on regional sustainable development was created by integrating the entropy weight method and the system dynamics method. Then, the Super-SBM efficiency measurement index system was designed to emphasize the influence of hydropower development on social, economic, and environmental indicators in the sustainable development subsystem.

The set of 15 development models was simulated using the SD model, resulting in 15 sets of data. The comparison study demonstrates that raising the value of hydropower input indicators leads to a rise in the value of the impacted indicators in the regional sustainable development subsystem, while reducing the value of hydropower input indicators leads to a decline. This indicates that, under the SD model built into this work, the influence of hydropower development on regional sustainable development belongs to a positive driving effect. This result is comparable to previous research, which verifies the sustainability of hydropower installations. On this premise, the Super-SBM model in DEA was used to quantify the effectiveness of hydropower development in driving regional sustainable development. It is worth mentioning that the efficiency assessed by the DEA algorithm is relative efficiency [48]. In addition, by viewing the development mode as a panel of data for different areas, the assessed technical efficiency is believed to have originated in locations where there were only slight variances in the values of the indicators and where there were regular variations. If the development mode is taken as an enhancement of the base mode $S_{15}$, then it emerged from the preceding studies that the input variables hydropower generation ($I_3$) and new hydropower investment ($I_1$) are essential variables. It is also plausible that a reasonable rise in $I_3$ combined with a moderate fall in $I_1$ is a realistic strategy to boost hydropower growth efficiency while driving regional sustainable development. Meanwhile, the technical efficiency of $S_{15}$ tested in this study was effective in 2016, 2019, and 2022, suggesting that the actual output in these years exceeded the “ideal output” in the non-parametric Super-SBM algorithm.

This study provides a fresh way of thinking about whether hydropower development has a driving influence on regional sustainable development from a macro perspective. Combining the SD with the DEA successfully answers the presence of a beneficial influence of hydropower development on the sustainable development of the study region and displays the course of its impact through a stock and flow diagram. This technique may measure the relative efficiency of a study area. However, it is difficult to apply this approach to different regions due to the volume and variability of data required by DEA. In the future, there is a need to explore an approach that can measure the relative efficiency of multiple regions simultaneously.

5. Conclusions

Based on the objective statistical data of a province in Southwest China, this study analyzed the influence of hydropower development on regional sustainable development and confirmed the non-linear and complicated driving effect of hydropower growth on regional sustainable development. Starting from the input and output viewpoints, the simulation data of diverse development modes were utilized as the input data of Super-SBM, which realized the effective relationship between the SD model and the Super-SBM model. Finally, the essential role indicators were found through the study. The primary conclusions are as follows:

(1)

The SD model emphasizes hydropower development’s influence, link, and role in regional sustainable development.

(2)

The influence of hydropower development on regional sustainable development is depicted in the SD model as the correlating impact of hydropower development on economic growth, social stability, environmental friendliness, etc. The findings of the simulation data comparison of different development modes demonstrate that hydropower development has a driving function in regional sustainable development, and it has a sort of complicated non-linear role.

(3)

The driving efficiency of hydropower development on regional sustainable development in the province from 2015 to 2022 was variable, and the technical efficiency value was maintained at a high level over 0.7.

(4)

Changing the input variables will influence the output variables and, consequently, the technical efficiency. Meanwhile, the province’s hydropower output and new hydropower investment are crucial variables that determine the relative driving efficiency of hydropower development for regional sustainable development.