An Evaluation on the Effect of Water-Saving Renovation on a Large-Scale Irrigation District: A Case Study in the North China Plain

Abstract

The construction of water-saving renovation projects can bring substantial benefits to the development of agriculture, but it may also be accompanied by negative impacts, especially in a large-scale irrigation district. Hence, there is always controversy, and it is vital and necessary to investigate the effectiveness of water-saving renovation. In this study, the Yahekou irrigation district, one of the largest districts in China, is selected as the case to explore the beneficial influence and adverse impact. Kriging interpolation, Pearson correlation analysis, and linear regression analysis are employed to study the temporal evolution, spatial distribution, and domain influencing factors. The results show that the water-saving renovation project in the Yahekou irrigation district had significant water-saving benefits during the period of 1998–2021, and the average annual water use of the irrigation district decreased by 61%. Canal lining is an important influencing factor for the decrease in irrigation water use, with a correlation coefficient of −0.538, B of −18.669, and R² of 0.290. Furthermore, the water-saving renovation of irrigation districts is not the dominant reason for changes in groundwater depth. Meanwhile, the groundwater depth below ground level (the level DBGL) in the irrigation district increased by 82%. The combined effects of atmospheric precipitation, canal lining, river sand mining, and groundwater exploitation have led to a sustained increase in groundwater depth in the study area. The results obtained from this study can provide valuable and scientific reference for agricultural water resource management over the large-scale irrigation district. This article focuses on the impact of water-saving renovation on ecological and technical indicators such as water use and the groundwater DBGL. However, the impact of water-saving renovation in irrigation districts is multifaceted; subsequent research can explore the impact of water-saving renovation projects on society and the economy.

1. Introduction

Water-saving in agriculture is a crucial approach to solving water scarcity issues. Many countries have successfully allocated water resources saved through agricultural production to industrial sectors, residential use, and ecological preservation, thereby ensuring sustainable social development. Since 1998, the Chinese government has been promoting water-saving renovation projects in irrigation districts across China [1]. Upon the completion of water-saving renovation projects in large irrigation districts in China, the irrigation water utilization coefficient will increase by 0.12, and the average gross quota of irrigation per hectare will decrease by approximately 1950 m³ within these areas [2]. Additionally, there will be an expansion of irrigated farmland area by 13.74 million hectares throughout China. Due to their long construction period, significant investment scale, and wide-ranging impact on large irrigation districts, it is essential to evaluate the benefits associated with these water-saving renovation projects. In recent years, numerous scholars worldwide have conducted studies focusing on various aspects such as scale effects, water-saving benefits, economic advantages, and negative consequences related to agricultural water-saving renovation.

Lina [3] explored the impact of water-saving renovation on the ecological and hydrological environment at the regional scale. It was pointed out that agricultural water-saving is the main reason for the changes in the ecological and hydrological environment of the Yangtze River Basin. The water-saving renovations in the Yangtze River Basin fundamentally change the regional ecological and hydrological environment, becoming a controlling factor of the regional ecological and hydrological environment. Xu [4] explored relevant issues concerning agricultural water-saving benefits and spatial scales within research areas. It was highlighted that return flow generated upstream can be reused downstream within an irrigation area, thus indicating that larger spatial scales result in an increased reuse of irrigation waters and higher utilization coefficients. Consequently, it is unrealistic to solely hope to achieve true water conservation in the entire irrigation area or watershed scale by reducing the amount of irrigation water in the field. Contor [5] pointed out that as the efficiency of irrigation water utilization increases, the total water resource consumption will also increase. This is because when the marginal cost of production input equals its marginal benefit, profit is maximized. Therefore, adopting water-saving renovation measures and improving the efficiency of irrigation water use will incentivize irrigators to use more irrigation water to achieve the maximum economic benefits. Pfeiffer [6] noted that water-saving renovation is typically considered a technically effective and politically feasible method to reduce water consumption for agricultural production. However, the assessment of the actual water-saving effect in western Kansas, United States, found that due to changes in crop planting patterns and an increase in irrigation area, local groundwater mining output did not decrease as expected by the project.

Grafton [7] highlighted that many countries have alleviated the increasingly serious water crisis by implementing the water-saving renovation of farmland. However, with the improvement in irrigation water efficiency, the actual water consumption in the region has also increased. The reasons are as follows: (1) Due to the lack of total quantity control and quota management, although water-saving renovation has improved the efficiency of irrigation water use and the guarantee rate of farmland irrigation, it has also increased the irrigated area of farmland, reduced the return water volume, and thus increased the total consumption of irrigation water. (2) The proliferation of water-saving measures, the increase in national water-saving subsidies, and farmers’ urgency to recover investment costs in the early stages of using water-saving irrigation facilities have led them to plant more water-intensive crops to obtain greater economic benefits. Zhang [8] questioned the practice of relying solely on improving irrigation water efficiency to save water. It was pointed out that although the utilization efficiency of irrigation water has been improved through the adoption of water-saving renovation measures, the so-called “efficiency paradox” of total water consumption increasing instead of decreasing due to the unrestricted increase in land irrigation area and the unrestricted use of water resources has emerged after the water-saving renovation and modernization upgrade of the irrigation system. At the same time, the study indicated that in areas with high recycling rates of return water, the impact of improving irrigation water use efficiency on the total regional water use is not significant.

Kang [9] pointed out that there are misconceptions in various sectors of society regarding the advantages and disadvantages of channel lining and the utilization of return water in the water-saving renovation of irrigation districts. He emphasized that the significant energy costs consumed by agricultural irrigation cannot be ignored; at the same time, water-saving renovation measures have improved the irrigation efficiency of irrigation districts and enhanced their ability to resist drought disasters. The related adverse effects can be overcome through a scientific and reasonable summarization of experience and scientifically scheduling during operation.

In summary, the water-saving renovation of irrigation districts has outstanding benefits in improving irrigation water efficiency, energy conservation, and emission reduction, increasing grain yield, and ensuring food security. However, controversies exist regarding whether it will have a negative impact on the environment and whether large-scale water-saving is effective. In response to these controversial points, this article takes the Yahekou irrigation district in China as an example, considering a large-scale irrigation district as a spatial scale and its 24-year water-saving renovation process as a time scale. It aims to study the spatiotemporal evolution and main influencing factors of irrigation water use, grain yield, and groundwater depth in large-scale irrigation districts and conduct an in-depth analysis of the specific impact of water-saving renovation projects on them.

2. Materials and Methods

2.1. Research Methods

(1) Pearson correlation coefficient method: This method is employed to measure the correlation between two variables, X and Y, with values ranging from −1 to 1. The Pearson correlation coefficient between two variables (population) is defined as the quotient of covariance and standard deviation between the two variables [10]. The calculation formula is as follows:

$${\rho }_{X,Y}=\frac{E[(X-{\mu }_X)(Y-{\mu }_Y)]}{{\sigma }_X{\sigma }_Y}$$

(1)

By estimating the covariance and standard deviation of each sample point, the Pearson correlation coefficient of each sample can be obtained, and its calculation formula is as follows [11]:

$$r=\frac{{\displaystyle {\sum }_{i=1}^n(x_i-\overline{x})(y_i-\overline{y})}}{\sqrt{{\displaystyle {\sum }_{i=1}^n(x_i-\overline{x}{)}^2}}\cdot \sqrt{{\displaystyle {\sum }_{i=1}^n{(y_i-\overline{y})}^2}}}$$

(2)

In the formula, r represents the Pearson correlation coefficient, x_i and y_i represent two sequence sample points, and n represents the length of the sequence.

In this study, the primary application of this method is to investigate the influencing factors of comprehensive benefits, water-saving benefits, and changes in groundwater depth.

(2) Regression analysis: This method is a predictive modeling technique used to determine the interrelationships between dependent and independent variables [12]. Linear regression and logistic regression are primarily utilized in this study to examine the interrelationships between independent and dependent variables [13].

Univariate linear regression analysis includes two variables; one is the independent variable, usually represented by x, and the other is the predictor variable, also known as the dependent variable, usually represented by Y. The calculation formula is shown in Equation (3):

$$Y=ax+b$$

(3)

In Equation (3), x is the independent variable, Y is the dependent variable, and a and b are the parameters of the univariate linear regression equation. The calculation formulas are shown in Equations (4) and (5):

$$b=\frac{{\displaystyle \sum Y_i}}{n}-a\frac{{\displaystyle \sum x_i}}{n}$$

(4)

$$a=\frac{n{\displaystyle \sum x_i{Y}_i-{\displaystyle \sum x_i{\displaystyle \sum Y_i}}}}{n{\displaystyle \sum }{x_i}^2-{({\displaystyle \sum x_i})}^2}$$

(5)

Subsequently, this study screens and analyzes the influencing factors of indicators such as channel lining, water-saving benefits, and changes in groundwater depth.

(3) Kriging spatial interpolation method: This method is a regression algorithm that spatially models and predicts (interpolates) random processes and fields based on the covariance function [14]. It primarily calculates the value of unknown sample points by assigning weights to known sample points [15]. The Kriging interpolation method is as follows:

$$Z\left(x\right)={\displaystyle \sum _{i=1}^m{\lambda }_i}Z\left(x_i\right)$$

(6)

where x represents the predicted position, m represents the number of measurements, Z(x) represents the estimated value at the interpolation point, Z(x_i) represents the measured value at the i-th position, and λ_i represents the weight of the measured value at the i-th position.

From Equation (6), the estimated value at the interpolation point is obtained through a weighted sum of known nodes. The weights λ_i are calculated using the variational function. The formula is as follows:

$${\displaystyle \sum _{i=1}^m{\lambda }_i}=1$$

(7)

$$\left[\begin{array}{c}{\lambda }_1\\ {\lambda }_2\\ ⋮\\ {\lambda }_m\\ \mu \end{array}\right]={\left[\begin{array}{ccccc}{\gamma }_{11}& {\gamma }_{12}& \cdots & {\gamma }_{1m}& 1\\ {\gamma }_{21}& {\gamma }_{22}& \cdots & {\gamma }_{2m}& 1\\ ⋮& ⋮& \ddots & ⋮& ⋮\\ {\gamma }_{m1}& {\gamma }_{m2}& \cdots & {\gamma }_{mm}& 1\\ 1& 1& \cdots & 1& 0\end{array}\right]}^{-1}\left[\begin{array}{c}{\gamma }_{10}\\ {\gamma }_{20}\\ ⋮\\ {\gamma }_{m0}\\ 1\end{array}\right]$$

(8)

where μ represents the Lagrange multiplier, and γ_ij represents the value of the variational function.

The value of the variational function between known nodes x_i and x_j is denoted as follows:

$${\gamma }_{ij}=\frac{1}{2}E{\left[Z\left(x_i-x_j\right)\right]}^2=\frac{1}{2}{\left[Z\left(x_i\right)-Z\left(x_j\right)\right]}^2$$

(9)

After calculating the variational function values corresponding to different intervals, a variational function curve γ = γ(d) is fitted. γ₁₀, γ₂₀, …, γ_m0 represent the variational function values between the interpolation points and known nodes, which can be calculated based on the fitted function curve.

In this study, the fitting curve adopts a spherical model.

$$\gamma \left(d\right)=\left\{\begin{array}{c}{C}_0+C_1\left[\frac{1.5d}{a}-0.5{\left(\frac{d}{a}\right)}^3\right]\\ C_0+C_1,d>a\end{array}\right.,0\le d\le a$$

(10)

where C₀ and C₁ are parameters to be determined, and a represents the maximum distance in the first law of geography.

For each interpolation point, the weights λ_i are calculated individually. Subsequently, the estimated values of the interpolation points are calculated using Equation (6). After all interpolation points have been estimated, the spatial interpolation map can be obtained.

In this study, the ordinary Kriging method within the Kriging method was primarily used to conduct spatial interpolation on the groundwater depth in the irrigation area. This was conducted to obtain the spatial distribution of groundwater depth in the research area and then carry out research on the spatial evolution of groundwater depth.

2.2. Data Sources

The data of this study include groundwater depth, irrigation water use, engineering construction data, capital investment data, meteorological data, vector data of water system distribution, and administrative divisions in the study area.

The groundwater depth data sourced from the Nanyang Hydrological Bureau were measured every 10 days from 1998 to 2021 through 16 groundwater level monitoring wells. A total of 36 measurements were conducted each year, and the average of the 36 measurements was taken as the groundwater depth data for that year. The irrigation water data sourced from the Yahekou irrigation district Management Bureau include the total water use of the irrigation area and the water use per unit area. The total water use was obtained by measuring the head of the two main canals, the Baitong Canal and the Yadong Canal. The water use per unit area was calculated by multiplying the total water use with the irrigation area. The water use of each county was obtained through the measurement of branch canal heads in each county.

Rainfall and other meteorological data were measured by the Yahekou Irrigation District Experimental Station (112.61138 E; 32.9499 N) located in the center of the irrigation district. The engineering construction data and capital investment data are sourced from the Yahekou irrigation district Management Bureau. The water system data and administrative division vector data are both sourced from Nanyang Water Resources Survey and Design Institute.

2.3. An Overview of the Research Area

2.3.1. Natural Conditions

The Yahekou irrigation district was founded in 1958 and carried out a 23-year water-saving renovation project mainly using channel lining from 1998 to 2021. The district is situated in the Nanyang Basin of China, lying between the Tang River and the Bai River. It falls within the Tang Bai River system, which is a tributary of the Han River in the Yangtze River Basin. This district is recognized as one of the top ten self-flowing irrigation districts in China, positioned at approximately longitude 112°39′ to 112°54′ E and latitude 32°27′ to 33°09′ N. It spans a width of 15 to 30 km in the east–west direction and extends approximately 100 km in the north–south direction, covering a total area of approximately 2428 square kilometers [16]. The specific geographical location is shown in Figure 1.

The Yahekou irrigation district experiences a monsoon continental humid and semi-humid climate, characterized by distinct weather patterns across all four seasons. The average annual evaporation on the land surface in the irrigation area is approximately 558 mm, accompanied by an average annual precipitation of about 802.7 mm. Precipitation varies from a maximum of 1290.1 mm to a minimum of only 492.2 mm annually. The rainy season occurs from June to September each year, contributing over 60% of the total annual precipitation. The irrigation area sees an absolute maximum temperature of 43.2 °C and an absolute minimum temperature ranging from 0 to 21.2 °C. The average temperature over many years is recorded at 14.9 °C, with annual sunshine hours exceeding 2000 h. The annual accumulated temperature meets the requirements for the double crop** planting model adopted in the irrigation area [17].

2.3.2. Socioeconomic Benefits

The Yahekou irrigation district benefits counties such as Fangcheng (FC), **nye (XY), Tanghe (TH), Sheqi (SQ), and Wancheng (WC) in Nanyang City (NYC). There are 35 townships and 605 administrative villages within the irrigation district, benefiting a population of approximately 1.83 million people. The total output value of the irrigation district in 2022 is estimated to be approximately CNY 30.866 billion. This irrigation area serves as an important commodity grain base in Henan Province, with the “winter wheat–summer maize” rotation being the primary planting mode. Additionally, other economic crops such as cotton, sweet potato, soybeans, sesame, peanuts, and vegetables are intercropped [18].

The main categories of water usage in the Yahekou irrigation district include agricultural irrigation water, domestic water, industrial water, and ecological water. Under a 50% guarantee rate, the annual domestic water quota in the irrigation area is approximately 74.89 million m³, the agricultural water quota is about 49.585 million m³, the industrial water quota is approximately 69.38 million m³, and the ecological water quota is roughly 5.7 million m³. Under a 90% guarantee rate, the annual domestic water quota remains at about 74.89 million m³, while the water quota for agricultural irrigation increases to about 557.44 million m³. The industrial water quota remains approximately 69.38 million m³, and the ecological water quota remains at about 5.7 million m³.

2.3.3. Irrigation Engineering

The designed irrigation area of the irrigation district is about 158,700 hm², and the effective irrigation area is about 88,400 hm². Through nearly 20 years of water-saving renovation projects in the irrigation district, the actual maximum irrigation area has expanded from approximately 66,700 hm² to approximately 116,700 hm². The distribution and specific engineering quantities of the canal system are shown in Figure 2 and

Table 1. The length of the canal in the Yahekou irrigation district.

Canal Name	Quantity	Length (km)
Main canal	2	86.279
Divided main canal	8	197.968
Branch canal	118	680.657
Divided canal	28	134.125

:

3. Results

3.1. The Outcomes of Water-Saving Renovation in the Irrigation District

During the period of 1998–2021, the Yahekou irrigation district invested USD 156.54 million—June 2024—(CNY 1.13365 billion) in conducting water-saving renovation projects to address various issues, including imperfect original engineering design and construction standards, low channel lining rates, low water resource utilization coefficients, and prominent supply–demand contradictions within the irrigation district. Following the implementation of multiple consecutive water-saving renovation projects, the irrigation district completed a total of 865 km of canal lining at all levels, covering approximately 8.577446 million square meters of lining area, and renovated 5360 buildings. This effort resulted in the establishment of a relatively comprehensive irrigation district engineering system. The overview of water-saving renovation and supporting projects over the years is shown in

Table 2. The statistical table of the water-saving renovation of the Yahekou irrigation district over the years (1998–2021).

Year	Canal Lining (m)	Canal Lining (m2)	Anti-Seepage Reinforcement (m)	Landslide Control (m)	Canal Side Road (m)	Bridge	Canal System Buildings	Drainage Canal Lining (m)	Investment (USD Million)
1998	4150	178,601	/	1747	/	/	1	/	2.07
1999	5204	71,834	6800	/	/	/	315	/	2.49
2000	9200	93,830	3750	/	/	/	24	/	1.66
2001	18,500	108,673	/	/	/	/	186	/	1.10
2002	/	203,328	5283	/	/	/	286	/	2.22
2003	41,542	1071,149	3970	/	/	/	46	/	2.21
2004	24,443	109,184	100	/	/	/	46	/	2.23
2005	11,798	184,685	/	500	/	/	27	/	2.76
2006	15,965	170,309	/	150	5000	/	6	/	2.49
2007	14,000	117,163	80	330	3500	7	9	/	1.10
2008	59,745	903,696	/	/	8000	31	17	/	7.73
2009	16,295	214,976	31,300	/	/	12	2	/	2.97
2010	66,931	736,242	/	2600	3000	44	249	/	10.02
2011	57,028	569,251	/	1408	/	65	119	464	6.68
2012	94,225	784,854	900	40	5000	59	350	1400	12.02
2013	82,212	513,547	/	/	9767	221	408	/	11.58
2014	35,902	197,017	6118	/	10,400	114	291	2917	11.14
2015	83,635	799,890	/	1777	6000	177	287	3400	22.68
2016	163,545	895,414	/	17,198	2200	481	854	2247	28.90
2017	9677	162,648	560	/	11,550	395	571	/	5.25
2018	15,132	225,318	/	/	/	71	127	/	7.27
2019	30,175	253,226	/	/	/	146	176	/	8.58
2020	1500	12,600	/	/	/	/	14	/	1.13
2021	/	0	/	/	/	/	4	/	0.26

:

The content of water-saving renovation projects in the Yahekou irrigation district is not fixed, and the construction pattern is not significant. In order to facilitate subsequent research, this study analyzes the project outcomes from the perspective of spatiotemporal changes.

3.1.1. Spatial Pattern of Water-Saving Renovation

The traditional calculation of the channel lining rate primarily relies on the length of the lined channels in the irrigation district. However, due to variations in heights, widths, and construction standards of the lining, the general proportion of the lining length to the total length is often used as the channel lining rate, which may not be scientifically or reasonably accurate [19]. This study adopts the area of channel lining as the primary calculation indicator and analyzes construction patterns and the main influencing factors by examining the area of different levels of channel lining across various years.

This study commences from 1998, the initiation year of the project, with each 3-year interval as the time scale. Each county within the irrigation district is considered as the spatial scale to organize and summarize changes in the lining area of water-saving renovation channels, using the proportion of the lining area to the total lining area every three years as a specific indicator. Subsequently, ArcGIS 10.2 is utilized for visualization purposes to depict the spatial changes in channel lining in the Yahekou irrigation district. The specific analysis results are illustrated in Figure 3:

Figure 3 takes each county within the irrigation district as the spatial scale, and the specific indicator is the proportion of the area of lining every three years to the total lining area. The transition from cold to warm tones in this figure represents a continuous increase in the proportion of lining, expressed in%: 0.00–15.00% is deep blue, 15.00–40.00% is green, 40.01–60.00% is yellow, 60.00–90.00% is orange, and 90.01–100% is red.

Figure 3 visually illustrates that, spatially, the water-saving renovation project generally progresses from north to south within the Yahekou irrigation district. In 1998, the project commenced in the upstream counties located in the northern part of the irrigation district. It subsequently focused on the central region of the district and was ultimately completed in the southern part of the district between 2019 and 2021.

3.1.2. Time Pattern of Water-Saving Renovation

The year 1998 marks the initiation of the water-saving renovation project in the Yahekou irrigation district. Utilizing a 3-year interval as the time scale, this study analyzes changes in the proportion of the project volume to the total project volume in the irrigation district, as well as changes in the lining area in each county every 3 years. Subsequently, Origin 2019 is employed to analyze the temporal variation pattern of the lining of the water-saving renovation channels in the irrigation district, thus revealing the construction time characteristics of the channel lining in the Yahekou irrigation district. The specific changes in channel lining over time are depicted in Figure 4:

Figure 4 illustrates that, temporally, the water-saving renovation project in the Yahekou irrigation district primarily focused on the period from 2007 to 2018, during which the engineering volume accounted for approximately 71.35% of the total engineering volume. Progress in water-saving renovation projects was relatively slow from 1998 to 2006, constituting approximately 25.55% of the total project volume during this timeframe. The majority of water-saving renovation projects were essentially completed from 2019 to 2021, with the engineering volume during this period accounting for about 3.1% of the total engineering volume.

3.2. Temporal and Spatial Evolution of Irrigation Water Use

The overall water use in various counties and districts of the Yahekou irrigation district showed a fluctuating and decreasing trend from 1998 to 2021. During this period, the average annual total water use in the irrigation district decreased by about 61%. SQ and FC have the most significant water-saving benefits, reducing usage by about 94% and 80%, respectively; TH and XY have poor water-saving benefits, with a decrease of about 37% and 39%, respectively; the water-saving benefits in WC are average, reducing usage by about 66%.

3.2.1. Spatial Evolution of Irrigation Water Use

In ArcGIS 10.2, the spatial variation pattern of the annual average water use in the Yahekou irrigation district from 1998 to 2021 was studied and analyzed, resulting in spatial variation maps of water use in each county and district of the Yahekou irrigation district, as depicted in Figure 5:

From Figure 5, the following observations can be made: (1) The overall water use of the Yahekou irrigation district is exhibiting a downward trend. (2) WC emerges as the county with the highest annual total water use in the Yahekou irrigation district, followed by XY as the second largest county in terms of the annual total water use. TH ranks third with the lowest annual total water use, while FC and SQ exhibit the lowest annual total water use. Notably, the annual total water use of each county within the irrigation area does not exhibit obvious upstream and downstream distribution characteristics. Additionally, the total water use ranking of each county is relatively consistent with the ranking of each county’s size within the irrigation area.

Using ArcGIS 10.2 to analyze the spatial variation pattern of water use per unit area in the Yahekou irrigation district from 1998 to 2021, a spatial variation map of water use per unit area in the Yahekou irrigation district was obtained, as displayed in Figure 6:

From Figure 6, the following observations can be made: (1) The overall water use per unit area of each county in the irrigation district demonstrates a downward trend. (2) Within the irrigation district, FC and SQ exhibit the least water use per unit area, while XY downstream of the irrigation district experiences higher water use per unit area. Conversely, WC and TH, major grain-producing counties, display the highest water use per unit area.

3.2.2. Time Evolution of Irrigation Water Use

Using Origin 2019 to analyze the temporal changes in total water use in each county and district of the Yahekou irrigation district from 1998 to 2021, a graph illustrating the temporal changes in water use in each county and district was obtained, as displayed in Figure 7:

From Figure 7, it can be observed that the total water use of each county and district exhibits a fluctuating downward trend over time, with peaks mainly occurring between 2013 and 2015 and 2007 and 2009.

Furthermore, employing Origin 2019 to analyze the temporal variation in water use per unit area in various counties and districts of the Yahekou irrigation district from 1998 to 2021, a graph illustrating the temporal variation in water use per unit area in each county and district was obtained, as shown in Figure 8.

From Figure 8, the following observations can be made: (1) The water use per unit area in each county and district of the Yahekou irrigation district demonstrates a fluctuating downward trend over time, with peak fluctuations mainly occurring around 2013–2015 and 2007–2009. This trend aligns with the temporal variation trend of the annual total water use mentioned earlier. (2) TH and WC, the major grain-producing counties in the Yahekou irrigation district, exhibit the highest water use per unit area, while FC and SQ have the lowest water use per unit area.

3.3. Temporal and Spatial Evolution of Groundwater Depth in Irrigation District

In preliminary research, some water-consuming farmers in the irrigation district also reported that after the water-saving renovation, the groundwater level in the irrigation district significantly decreased, leading to issues such as house cracking (Figure 9a), foundation settlement (Figure 9b), perennial interruption of natural river flow (Figure 9c), and road settlement (Figure 9d).

Related studies have also pointed out a series of rural environmental problems caused by water-saving irrigation [20,21].

To address this concern, this study aims to analyze the spatiotemporal evolution of groundwater depth in the irrigation district from 1998 to 2021.

To objectively analyze the spatiotemporal evolution of groundwater depth in the irrigation district, this study selected sixteen representative observation wells, with a continuous record, in the Yahekou irrigation district and its surrounding areas, as depicted in Figure 10:

3.3.1. Spatial Evolution of Groundwater Depth

A summary of the changes in the groundwater depth below ground level of 16 observation wells in the Yahekou irrigation district and its surrounding areas during the period of 1998–2021 was conducted. The change in the average depth to groundwater was analyzed over the project duration from initiation in 1998/2001 to completion in 2019/2021. It was found that, except for a decrease of about 35% in groundwater depth in Yanzhuang Village, Gaomiao Township, Wancheng District, the groundwater depth of other observation wells has increased to varying degrees. The location with the greatest changes in groundwater depth is Zhoumiao Village, Cha’an Township, Wancheng District, with an increase of about 3.78 times. The average groundwater depth of 16 observation wells has increased by about 1.25 times.

In practice, the groundwater depth in the Yahekou irrigation district showed an overall fluctuating and increasing trend during the period of 1998–2021. This study imported data from 16 monitoring wells in the Yahekou irrigation district and its surrounding areas into ArcGIS 10.2. The Kriging spatial interpolation method was used to conduct spatial modeling and prediction interpolation based on covariance functions. The spatial variation pattern of groundwater depth in the irrigation district was analyzed, and the spatial evolution pattern of groundwater depth in the Yahekou irrigation district during the period of 1998–2021 was obtained, as shown in Figure 11:

From the figure, the following observations can be made more intuitively: (1) The groundwater depth in the irrigation district is distributed in a belt shape from northeast to southwest, with the shallowest groundwater depth in the areas of Qiaotou Town in SQ, Cha’an Town in WC, and Tonghe Town in TH in the northeast of the irrigation district. The area with the highest groundwater depth is located on the first terrace of the Tang River in TH, the southeastern part of the irrigation district. Most of WC in the central part of the irrigation district is located in the same depth zone, with a relatively shallow depth. Although FC in the northern part of the irrigation district and XY in the southern part are not in the same depth zone, the burial depth is roughly the same. (2) During the period of 1998–2021, the overall groundwater depth in the irrigation district showed a continuous increasing trend. In the southern part of the irrigation district, there is little change in the groundwater depth around the first terrace of the Tanghe River in TH, while in the central northern part of the irrigation district, there is a significant change in the groundwater depth in Qiaotou Town of SQ.

3.3.2. Time Evolution of Groundwater Burial Depth

Using Origin 2019, a visualization of the temporal variation in groundwater depth in the Yahekou irrigation district was obtained, as displayed in Figure 12:

(1) During the period of 1998–2021, the groundwater depth in the irrigation district exhibited an overall increasing trend, i.e., the groundwater level was falling further below the surface. From 1998 to 2006, the groundwater depth in the irrigation district showed little change and a slight decreasing trend. However, from 2007 to 2021, the groundwater depth in the irrigation district exhibited a wave-like upward trend. (2) Analyzing the measured data of 16 monitoring wells, it was found that during the period of 1998–2006, the groundwater depth fluctuated slightly and remained relatively stable. The average depth in 2006 decreased by 5% compared to 1998. However, during the period of 2007–2021, the groundwater depth in the irrigation district significantly increased, with an average depth reaching 0.93 times deeper than the initial measurement in 1998.

4. Discussion

4.1. Analysis of Water-Saving Renovation

In the previous text, the construction rules of water-saving renovation projects in the irrigation district have been preliminarily elaborated. Based on relevant research [22] and the actual situation of the research area, it is preliminarily determined that the spatial and temporal changes in water-saving renovation projects in the irrigation district are affected by the investment of project funds. During the period of 1998–2006, funds were limited, and the focus was on solving prominent issues in key positions such as headworks. From 2007 to 2019, with sufficient funds available, key efforts were made to promote water-saving renovation projects in counties and districts located in the central hub of the irrigation district, such as WC and TH. In the final stage, key efforts were made to promote water-saving renovation projects in counties such as XY, located at the downstream of the irrigation district.

Next, the Pearson correlation analysis method is used to analyze the correlation between the annual changes in the water-saving renovation channel lining area and the annual investment amount in the irrigation district, further verifying this inference. The specific results are shown in

Table 3. The correlation analysis index between the annual change in the canal lining area and annual capital investment.

Variable Correlation Indicators (r)		Capital Investment (10,000 CNY/a)
Changes in channel lining area (m2/a)	Correlation coefficient	0.617 **
	Significance (bilateral)	0.004

Note: ** Significant correlation at 0.01 level (bilateral).

.

From Table 3, it can be observed that the change in the canal lining area in the irrigation district is closely related to annual capital investment, with a bilateral significance index of 0.004, indicating a significant positive correlation between these two variables at the 0.01 level, with a bilateral correlation coefficient of 0.617.

Next, this study takes the change in channel lining area as the dependent variable and annual capital investment as the independent variable. Linear regression analysis is conducted using SPSS 25.0 software to verify the above inference. Based on the regression analysis results,

Table 4. The regression analysis index between the annual change in the canal lining area and annual capital investment.

Variable	Regression Coefficient (B)	Significance Level (Sig)	Coefficient of Determination (R2)
Changes in channel lining area (m2/a)	33.712	0.004	0.381

is obtained.

From Table 4, it can be observed that the overall significance level of the model is 0.004 (p < 0.01), with the change in the canal lining area as the dependent variable and the annual investment as the independent variable. In this regression analysis, the regression coefficient is 33.712, indicating that for every CNY 10,000 increase in investment, the lining area of the canal system will increase by 33.712 m². The coefficient of determination is 0.381, indicating that the influence of capital investment factors accounts for 38.1% of the total influence factors that cause changes in channel lining area.

From the correlation and regression analysis results in this section, it can be seen that capital investment is an important influencing factor for the construction of the water-saving renovation project in the Yahekou irrigation district. This is similar to Hur and Zhong’s conclusion on the financial analysis of similar water conservancy projects, which states that capital investment affects project progress [23,24].

4.2. Analysis of Changes in Irrigation Water Use

Water conservation is the primary objective of water-saving renovation projects in irrigation districts. This study posits that the lining project of water-saving renovation channels in the irrigation district has led to a reduction in irrigation water use. To assess this claim, Pearson correlation analysis was performed using SPSS 25.0 software to examine the correlation between the change in the lining area of water-saving renovation channels and the related water usage indicators within the irrigation district. The results are presented in

Table 5. The correlation analysis coefficient of the irrigation water index.

Benefit-Related Indicators		Changes in Channel Lining Area
Changes in total irrigation water use	Correlation coefficient (r)	−0.538 **
	Significance (bilateral)	0.010
Changes in total irrigation water saved	Correlation coefficient (r)	0.547 **
	Significance (bilateral)	0.003

Note: ** Significant correlation at 0.01 level (bilateral). The “irrigation water index” refers to “total irrigation water use (m³)” and “total irrigation water saved (m³)”.

.

From Table 5, it is evident that the change in the canal lining area in the irrigation district is closely linked to the change in the total irrigation water use, with a bilateral significance index of 0.010. This indicates a significant negative correlation between these two variables at the 0.01 level, with a correlation coefficient of −0.538 on both sides. The change in the canal lining area in the irrigation district is closely linked to the change in the total irrigation water saved, with a bilateral significance index of 0.003. This indicates a significant positive correlation between these two variables at the 0.01 level, with a correlation coefficient of 0.547 on both sides.

Subsequently, this study employed linear regression analysis using SPSS 25.0 software to further validate the aforementioned inference. The regression analysis results are summarized in

Table 6. The regression analysis coefficient table of the irrigation water index.

Variable	Regression Coefficient (B)	Significance Level (Sig)	R2
Changes in total irrigation water use	−18.669	0.010	0.290
Changes in total irrigation water saved	11.939	0.003	0.300

Note: The “irrigation water index” refers to “total irrigation water use (m³)” and “total irrigation water saved (m³)”.

.

Table 6 illustrates that, with the change in the canal lining area as the independent variable and the change in the total irrigation water use and total irrigation water saved as the dependent variable, the overall significance level of the model is 0.010, 0.003. This suggests that the established regression equation model effectively fits the sample data and holds statistical significance. Additionally, the obtained histogram exhibits good symmetry, the distribution of each point in the P-P diagram is linear, and the scatter plot distribution is relatively concentrated, indicating a successful test. In this regression analysis, the regression coefficient is −18.669 and 11.939, indicating that for every additional square meter in the lining area of the irrigation district canal system per annum, the total irrigation water use will decrease by 18.669 m³, and the total irrigation water saved will increase by 11.939 m³. The determination coefficient R² is 0.290 and 0.300, signifying that the influence of water-saving renovation channel lining factors accounts for 29.0% and 30.0% of the total factors affecting changes in irrigation water use and saved in the Yahekou irrigation district.

The previous analysis highlighted that water use in the Yahekou irrigation district exhibited a fluctuating downward trend, with peak fluctuations primarily occurring around 2013–2015 and 2007–2009. During the period from 2012 to 2016, the district experienced five consecutive “dry years”, characterized by insufficient precipitation. This resulted in an increase in irrigation water use between 2013 and 2015. Conversely, from 2007 to 2009, the total precipitation in the district was deemed sufficient. However, a further analysis of precipitation distribution within these years revealed significant variations. For instance, during the summer planting and emergence period of crops in 2007, high temperatures coupled with minimal rainfall led to low soil moisture levels, hindering the normal growth of crops like wheat. Conversely, in the autumn of 2007, frequent rainfall occurred, causing water saturation and runoff. Similarly, in the summer of 2008, despite being dry and rainy, the district experienced minimal rainfall, resulting in decreased soil moisture content. In 2009, uneven rainfall distribution during the wheat growth period negatively impacted crop development, with dry conditions prevailing during critical growth stages. Additionally, excessive rainfall during the autumn of 2009 led to reduced crop yields due to flooding. Despite sufficient total rainfall during 2007–2009, the uneven distribution of rainfall throughout the year contributed to increased water use in the irrigation area during this period, aligning with the findings of Zai Songmei et al. [25].

Considering the spatiotemporal evolution and correlation and regression analysis results of irrigation water use, along with the actual conditions in the research area, it can be concluded that the water-saving renovation of channel lining is a significant factor influencing changes in irrigation water use in the Yahekou irrigation district. Additionally, the uneven distribution of precipitation within and between years may also impact the spatiotemporal evolution of irrigation water use in the district.

4.3. Groundwater Impact Analysis

The related literature [26,27] suggests that changes in the groundwater environment in the irrigation district are influenced by factors such as water-saving renovation projects and weather conditions.

It is speculated that changes in groundwater depth are closely related to the implementation of water-saving renovation projects and variations in precipitation. In SPSS software, Pearson correlation analysis is used to verify the above inference. Based on the results of the correlation analysis,

Table 7. The correlation of the groundwater depth and estimated driving force factor.

Benefit-Related Indicators		The Cumulative Lining Area of the County (m2)	Annual Average Precipitation (mm)
The groundwater depth in FC (m)	Correlation coefficient	0.906 **	−0.768 *
	Significance (bilateral)	0.002	0.026
The groundwater depth in SQ (m)	Correlation coefficient	0.791 *	−0.831 *
	Significance (bilateral)	0.019	0.011
The groundwater depth in WC (m)	Correlation coefficient	0.824 *	−0.837 **
	Significance (bilateral)	0.012	0.010
The groundwater depth in XY (m)	Correlation coefficient	0.942 **	−0.790 *
	Significance (bilateral)	0.000	0.020
The groundwater depth in TH (m)	Correlation coefficient	0.533	−0.832 *
	Significance (bilateral)	0.174	0.010

Note: ** Significant correlation at 0.01 level (bilateral), * significant correlation at 0.05 level (bilateral).

is generated.

From Table 7, it can be observed that changes in the groundwater depth in the irrigation district are closely associated with the implementation of water-saving renovation projects, such as canal lining, as well as annual precipitation.

The relationship between changes in groundwater depth and the cumulative lining area of FC and XY, which collectively cover about 35% of the total area of the Yahekou irrigation district, is more closely correlated (significant positive correlation at the 0.01 level, bilateral) than the change in annual precipitation (significant negative correlation at the 0.05 level, bilateral). Similarly, the relationship between changes in the groundwater depth in WC, SQ, and TH, covering approximately 65% of the total area of the Yahekou irrigation district, and the change in the annual average precipitation in these counties is more closely linked (significant negative correlation at the 0.01 and 0.05 levels, bilateral) than the change in the cumulative lining area (significant positive or no correlation at the 0.05 level, bilateral). Overall, it can be inferred that changes in groundwater depth in the Yahekou irrigation district are most closely associated with changes in annual precipitation and are further influenced by water-saving renovation channel lining projects.

Next, this study considers changes in the canal lining area and annual precipitation in each county as independent variables and changes in groundwater depth in each county as dependent variables. Linear regression analysis is conducted using SPSS 25.0 software to verify the above inference. Based on the regression analysis results,

Table 8. The regression analysis coefficient between the groundwater depth and estimated influencing factors.

Regression Analysis Coefficient	The Groundwater Depth in FC (m)	The Groundwater Depth in SQ (m)	The Groundwater Depth in WC (m)	The Groundwater Depth in XY (m)	The Groundwater Depth in TH (m)
Regression coefficient of channel lining area (B)	4.08 × 10−2	5.67 × 10−2	5.23 × 10−3	2.81 × 10−2	1.06 × 10−2
Significance (bilateral) of channel lining area (Sig)	0.002	0.019	0.012	0	0.174
Canal lining area R2	0.82	0.625	0.678	0.888	0.284
Annual average precipitation regression coefficient (B)	−1.5 × 10−2	−1.5 × 10−2	−0.8 × 10−2	−1.4 × 10−2	−0.8 × 10−2
Annual average precipitation significance level (Sig)	0.026	0.011	0.01	0.02	0.01
Annual average precipitation (R2)	0.589	0.690	0.700	0.624	0.692

is derived.

In Table 8, it can be observed that the overall significance level (Sig) of the regression model established with the change in channel lining area (m²/a) in TH as the independent variable and the change in groundwater depth (m) as the dependent variable is 0.174 (p > 0.05), indicating that the regression equation model established has a poor fit to the sample and lacks statistical significance. Conversely, for changes in the canal lining area (m²/a) in FC, SQ, WC, and XY as independent variables and changes in groundwater depth (m) as dependent variables, the overall significance levels (Sig) of the model are 0.002, 0.019, 0.012, and 0.000, respectively. The established regression equation models exhibit a good fit for the sample and have statistical significance. Additionally, the obtained histograms display good symmetry, the distribution of each point in the P-P diagram is straight, and the scatter plot distribution is relatively concentrated, indicating a passing test. In this regression analysis, the regression coefficients are 4.084 × 10⁻², 5.670 × 10⁻², 5.227 × 10⁻³, and 2.814 × 10⁻², respectively. This suggests that for every additional square meter in the lining area of the irrigation system in FC, SQ, WC, and XY, the groundwater depth increases by 4.084 × 10⁻² m, 5.670 × 10⁻² m, 5.227 × 10⁻³ m, and 2.814 × 10⁻² m, respectively. Moreover, the determination coefficients are 0.820, 0.625, 0.678, and 0.888, respectively, indicating that the fitted equations can explain 82.0%, 62.5%, 67.8%, and 88.8% of the changes in groundwater depth in FC, SQ, WC, and XY within the Yahekou irrigation district.

Taking the average annual precipitation (mm) of FC, SQ, WC, XY, and TH as independent variables and the change in groundwater depth (m) as the dependent variable, the overall significance levels (Sig) of the model are 0.026, 0.011, 0.01, 0.02, and 0.01, respectively. This indicates that the established regression model has a good fit for the sample, and the regression equation is effective, with statistical significance. Additionally, the histograms obtained in the regression analysis exhibit good symmetry, the distribution of each point in the P-P diagram is straight, and the scatter plot distribution is relatively concentrated, passing the test. In this regression analysis, the regression coefficients are −1.5 × 10⁻², −1.5 × 10⁻², −0.8 × 10⁻², −1.4 × 10⁻², and −0.8 × 10⁻², respectively. This suggests that for every 1 mm increase in precipitation in the irrigation area, the groundwater depth in FC, SQ, WC, XY, and TH decreases by 1.5 × 10⁻² m, 1.5 × 10⁻² m, 0.8 × 10⁻² m, 1.4 × 10⁻² m, and 0.8 × 10⁻² m, respectively. The determination coefficients are 0.589, 0.690, 0.700, 0.624, and 0.692, respectively, indicating that the fitted regression equation can explain 58.9%, 69.0%, 70.0%, 62.4%, and 69.2% of the changes in groundwater depth in FC, SQ, WC, XY, and TH within the Yahekou irrigation district.

From the spatiotemporal evolution of groundwater depth in the irrigation district and its correlation and regression analysis with channel lining area and annual precipitation, it can be observed that atmospheric precipitation is the primary influencing factor of groundwater depth in the irrigation district, while channel lining only significantly impacts groundwater depth in certain counties and districts.

Additionally, the Yahekou irrigation district is situated between the Bai River and the Tang River. Apart from atmospheric precipitation and irrigation water supply, lateral river replenishment is also a key source of groundwater supply in the district. In recent years, there has been extensive sand mining in rivers such as Baihe and Tanghe within the irrigation district. With continued sand extraction from the riverbeds, the riverbeds are being continuously lowered, causing a reduction in the aquifer’s ability to store underground groundwater. This leads to a change in the hydraulic connection between underground groundwater and shallow groundwater in the irrigation area. Consequently, the Yahekou irrigation district not only fails to receive river recharge but also has to discharge groundwater into the river.

As a semi-natural and semi-artificial ecosystem, the artificial exploitation of groundwater in irrigation areas significantly contributes to changes in groundwater depth; this is consistent with the research findings of the relevant literature [12,28]. With the implementation of high-standard farmland and land improvement projects in the irrigation district, the coverage area of well irrigation continues to expand. Approximately 65.03% of the farmland in the irrigation district can be irrigated with groundwater, with about 30.6% relying solely on groundwater for irrigation. The water output of a single well typically ranges from 8 to 40 m³/h. Despite an overall tap water penetration rate of over 80% in the irrigation district, most farmers still utilize underground water intake and storage equipment such as pressure wells, small pumps, and pressure tanks at home. Groundwater remains the primary source of household daily water use. In the Yahekou irrigation district, approximately 85.32% of water users predominantly mix groundwater and surface water in their daily lives, with an average daily water consumption ranging from 45 to 80 L per person.

The relevant literature [29,30] indicates that the variation in groundwater depth is primarily influenced by both groundwater recharge and discharge. Therefore, it is scientifically unsound to solely attribute the continuous increase in groundwater depth in the Yahekou irrigation district to the water-saving transformation of the irrigation channel lining project.

5. Conclusions

(1) The water-saving renovation project in the Yahekou irrigation district has yielded significant water-saving benefits. From 1998 to 2021, despite fluctuations in total water use and water use per unit area in the irrigation district, the overall trend showed a notable decline. The average annual water use in the irrigation area has decreased by approximately 0.61 times. Canal lining emerges as a crucial factor contributing to the reduction in irrigation water use (r = −0.538 **; B = −18.669; R² = 0.290).

(2) The water-saving renovation of the irrigation district is not the primary cause of changes in groundwater depth. During the period of 1998–2021, the average underground burial depth in the irrigation district increased by 0.82 times. Data analysis and field research reveal that the combined effects of atmospheric precipitation, channel lining, river sand mining, and groundwater exploitation have led to a continuous increase in groundwater depth in the Yahekou irrigation district. It is scientifically unsound to attribute the continuous increase in groundwater depth solely to the water-saving transformation of the irrigation channel lining project.

(3) The impact of water-saving renovation in irrigation districts is multifaceted, and the comprehensive impact of water-saving renovation projects on the economy, society, and other aspects will be studied in a later stage.