Analysis of Impact of Well-Facilitated Farmland Construction—Engineering Measures on Farmland Quality

Abstract

We studied the influence and correlation of soil improvement, farmland ecological protection, soil fertilization, and field infrastructure construction on the quality grade of well-fertilized farmland in the engineering measures of well-fertilized farmland construction. Taking Xiao County of the Anhui Province as the study area, based on the software platforms of SPSS, ArcGIS10.6, and the county farmland resource management information system, we investigated the farmland quality changes of well-facilitated farmland before and after construction using the fuzzy evaluation method and analytic hierarchy process. We used principal component analysis and the gray relational method to analyze the impact and correlation of various engineering measures on farmland quality. The farmland quality grade in the study area was improved by a 0.59 grade after the construction of the well-facilitated farmland. Well-facilitated farmland construction engineering measures mainly affected the farmland quality through 12 factors, such as the soil bulk density, tillage layer texture, irrigation and drainage guarantee rates, forest network density, and field road accessibility. There is a strong correlation between these factors and the characteristics of farmland quality; the degrees of correlation were 0.865–0.610, respectively. The highest correlation degree was 0.939 between the deep plowing and deep loosening soil improvement project and the improvement of the well-facilitated basic farmland quality; this was followed by soil fertilization with an increased application of organic fertilizer, farmland ecological protection, and the field infrastructure project with correlations of 0.936, 0.857, and 0.563, respectively. Represented by the improvement of farmland fertility, the soil improvement project had the strongest impact on well-facilitated farmland quality. The soil fertility project, farmland ecological protection project, and the field infrastructure project were the second most important, with very close degrees of correlation.

1. Introduction

With the continuous growth of China’s population, the tightening of resources and environmental carrying capacity, and the continuous upgrading of consumption structures [1], coupled with the superimposed influence of unexpected events such as the new coronavirus pneumonia spreading worldwide, rising energy prices have spurred increasing demand for biofuels and extreme disasters, the volatile international grain trade market is a serious threat to the stable supply of the global food system [2]. China has issued multiple principal documents emphasizing the need to speed up agricultural modernization and enhance the ability to guarantee the supply of food and key agricultural products [3]. The construction of well-facilitated farmland is an essential strategic measure to ensure farmland resource security and national food security and to promote the high-quality development of agriculture [4]. The Chinese government has proposed developing 1.2 billion hectares of well-facilitated farmland, upgrading 280 million hectares, and stabilizing the grain production capacity to more than 1.2 trillion pounds by 2030 [5,6,7]. Many Chinese scholars have studied the construction of well-facilitated basic farmland in detail and generally believe that the development of well-facilitated basic farmland can not only improve the farmland quality grade [8,9,10,11] but can also bring about social [12], economic [13], and comprehensive benefits [14,15].

Based on criteria such as farmland area and drainage facilities, farmland in the United States is classified as basic agricultural land, special agricultural land, locally important agricultural land, and nationally important agricultural land, which are used and protected at different levels [16]. Rural land consolidation projects are comprehensive in Germany. Comprehensive studies have been conducted not only regarding plot adjustment [17], soil improvement [18], and road planning [19] but also regarding eco-environmental protection [19] and landscape studies [20]. The United Kingdom applies the results of farmland potential assessment to basic farmland protection to permanently protect high-quality basic farmland [21]. In agricultural land consolidation in Japan, scattered and fragmented farmland is transformed into concentrated and continuous land through the exchange of land-use rights, facilitating large-scale mechanized production [22]. The Netherlands eliminates finely divided farmland through land consolidation and reorganization, emphasizing agricultural infrastructure construction, and improving land quality [23]. The construction of well-facilitated farmland in China is based on the elimination of the main farmland limiting factors and the overall improvement of farmland productivity by carrying out infrastructure construction such as plot consolidation, irrigation, and drainage, field roads, farmland protection, and eco-environmental protection and farmland fertility enhancement activities such as soil improvement and fertilization [6]. Based on the perspective of farmland quality grade, Zhang [24] found that farmland quality in their project area was improved to varying degrees after the construction of well-facilitated basic farmland. Qian et al. [25] conducted an assessment study of farmland quality and site conditions based on the construction of well-facilitated basic farmland in Donggang city. The authors concluded that the construction area of well-facilitated basic farmland in the study region could be divided into priority construction areas, farmland natural quality conditions, and farmland site conditions. Xu [26] analyzed the logical relationship between the engineering content of well-facilitated basic farmland construction and farmland production capacity and measured the potential of well-facilitated basic farmland construction to improve farmland production capacity. However, there is a lack of systematic research concerning the correlation between engineering measures of well-facilitated farmland construction and farmland quality grade. The focus of well-facilitated farmland construction engineering measures in different regions varies due to natural geographic and socio-economic differences [27,28] and restrictive factors of agricultural production [29,30]. In addition, there are significant differences in the improvement and correlation degree of farmland quality after construction [31,32]. Based on the summary and generalization of the well-facilitated farmland construction engineering measures and according to the relevant requirements of well-facilitated farmland construction, in this study, we investigated the impact and correlation of well-facilitated farmland construction engineering measures, including soil improvement, farmland ecological protection, soil fertilization, and field infrastructure projects on the well-facilitated farmland quality grade. This study enriches and improves the research into well-facilitated farmland engineering construction measures, providing a scientific reference for Xiao County to achieve the goal of farmland quality enhancement as well as a reference for similar regions and countries.

2. Materials and Methods

2.1. Study Area Overview

Xiao County is part of the North China Plain. It is one of China’s major grain-producing areas, along with being one of the demonstration counties for China’s commercial grain base. The study area was the well-facilitated farmland construction project area of Xiao County in 2020, distributed in nine project areas of Xinzhuang Township, Ma Jing Township, Zhao Zhuang Township, Wang Zhai Township, Ding Li Township, Liu Tao Township, Yang Lou Township, Zhao Zhuang Township, and Wang Zhai Township, with a total construction scale of 100,000 hectares. Xiao County is located at 116°31′–117°12′ east longitude and 33°56′–34°29′ north latitude, which is the southernmost tip of the Yellow-Huaihai Plain. The topography of the study area is relatively gentle. The landscape is dominated by plains and intermountain basins. The annual average temperature is 14.6 °C. The soil types are mostly yellow cinnamon soil and fluvo-aquic soil. The location of the study area is shown in Figure 1.

2.2. Data Sources

2.2.1. Cartographic Data

The Xiao County soil map, Xiao County administrative division map, Xiao County land-use status map, and Xiao County’s new well-facilitated farmland implementation distribution map of 2020 are the basic maps for the farmland quality survey and quality assessment. The cartographic data are mainly provided by the Bureau of Agriculture and Rural Affairs of Xiao County.

2.2.2. Textual Data

The data mainly include the results of the second soil census, relevant data on soil testing and formula fertilization, the ownership code table, Xiao County’s soil log, the feasibility report and engineering design scheme of the new well-facilitated farmland construction projects in 2020, and the new well-facilitated farmland construction self-inspection report (including the quality construction measures summary table of the new well-facilitated farmland).

2.2.3. Field Survey

According to the relevant assessment requirements, such as the well-facilitated farmland’s Construction Assessment Standards (GBT33130-2016) [33] and the Farmland Quality Grade (GB/T 33469-2016) [33], the site layout, survey, sampling, and recording of the new well-facilitated project area were carried out with the combination of specific conditions of the new well-facilitated project area. According to the area and distribution characteristics of the new well-facilitated farmland, the survey points distribution map is shown in Figure 2.

2.3. Research Methods

Based on the sources of the Xiao County farmland quality survey and assessment in 2019, we extracted the data from 1685 plots before the well-facilitated farmland construction. First, the analytic hierarchy process was used to assess the farmland quality grade of the project area before and after the construction of the new well-facilitated farmland in 2019 and 2020. Next, based on previous research on farmland quality characteristics, the impact of four engineering measures on well-facilitated farmland quality was selected for gray relational analysis. Based on the specific measures of the four projects, 12 representative indicators were analyzed as secondary influencing factors in the principal component analysis. The component scores were then saved as the corresponding data of the four projects. Gray relational analysis was carried out for the four projects. Figure 3 shows the specific process.

2.3.1. Field Survey and Analysis Methods

Site Layout

Before the field survey, Arc GIS and OWI maps were used as reliable tools. Based on the characteristics of the area and the distribution of the new well-facilitated farmland in the study area, the survey sampling was designed on the basis of 1000 mu/point, with a total of 100 points.

Survey

Information on field irrigation, drainage, and farmland forest networks was obtained through field observation. In the field, a soil auger was used to obtain information on the effective soil thickness, the type of barrier layer, and the texture and configuration of the field. In addition, information on soil texture was also obtained in the field by hand measurement. A small sample of soil was taken in the palm of the hand and crushed with the fingers, and new or invasive bodies, such as fine gravel and coarse organic matter, were picked out. Once finely crushed, the sample was ready for wet testing. Information was collected from field observations of well-facilitated farmland in the study area. The survey was combined with sampling, while the surveyors and samplers were trained in the subject matter and relevant skills prior to the survey to ensure that the survey was scientific and standardized.

Sampling

Soil samples were collected in September 2021. In this study, sampling was conducted using the quadrat method to maintain the representativeness of the soil samples and the five-point sampling method to maintain the homogeneity of the soil samples. Mixed soil samples, soil profile samples, soil ring knife samples, and soil texture samples were collected.

Testing of samples

The soil’s pH was determined using the potentiometric method, HJ 962-2018. Soil organic matter was determined using the farmland quality grade GB/T 33469-2016 Appendix C [33]. The determination of the quick-acting potassium content of soils was carried out using the NY/T 889-2004 standard [34]; The soil’s available phosphorus was tested according to the soil test, part 7. Effective phosphorus in soil was determined using NY/T 1121.7-2014 [34]. The bulk density of the soil was determined according to the method specified in NY/T 1121.4. Soil texture was determined in accordance with NY/T 1121.3 [34] in the soil, part 15 [7]. The remaining indicators were evaluated at the test facility.

2.3.2. Principal Component Analysis

The aim of the principal component analysis is to reassemble the original variables into a new set of independent comprehensive variables, from which several comprehensive variables can be extracted according to practical needs to reflect as much information as possible regarding the original variables. This statistical method is also a mathematical way to deal with dimensionality reduction.

2.3.3. Gray Relational Analysis

The gray relational analysis is a type of multi-factor statistical analysis. In a gray system, if the researcher wants to know the relative strength of a certain indicator item impacted by other factors, these influencing factors are not ranked, but rather a gray relational coefficient is calculated for each factor, which represents the strength of the relationship between the factor and the variable of interest. Factors with superior gray relationship coefficients are considered to be more strongly related to the variable of interest; this will indicate which factors have a better correlation with the indicator of interest. Further, the system operation can be effectively adjusted.

• Determination of characteristic sequence and parent sequence

Taking the farmland quality, X₀, of the well-facilitated farmland in Xiao County in 2020 as the reference sequence and the score X₁…X₄ of the four engineering measures as the comparison sequence, the gray relational analysis was carried out [35].

Characteristic sequence:

$$[\begin{array}{cccc}{X}_1^{\prime }& X_2^{\prime }& \cdots & X_n^{\prime }\end{array}\left]=\right[\begin{array}{cccc}{x}_1^{\prime }\left(1\right)& x_2^{\prime }\left(1\right)& \cdots & x_n^{\prime }\left(1\right)\\ x_1^{\prime }\left(2\right)& x_2^{\prime }\left(2\right)& \cdots & x_n^{\prime }\left(2\right)\\ ⋮& ⋮& & ⋮\\ x_1^{\prime }\left(m\right)& x_2^{\prime }\left(m\right)& \cdots & x_n^{\prime }\left(m\right)\end{array}]$$

(1)

Reference sequence:

$$X_0^{\prime }={\left(x_0^{\prime }\left(1\right),x_0^{\prime }\left(2\right),\dots ,x_0^{\prime }\left(m\right)\right)}^T$$

(2)

2.

Dimensionless data processing

It is necessary to convert data into dimensionless forms to eliminate the influence of the differences in the unit and magnitude between the project score data and the well-facilitated farmland quality grade data. Since the project score data and the well-facilitated farmland quality grade data may differ in dimensions, it is not convenient to compare these directly, or it is difficult to draw exact conclusions from the comparison. Therefore, the data were averaged prior to the gray relational degree analysis [36].

3.

Correlation coefficients

The correlation coefficients of various projects corresponding to the well-facilitated farmland quality grade were calculated using the following equation:

$$X_0^{\prime }={\left(x_0^{\prime }\left(1\right),x_0^{\prime }\left(2\right),\dots ,x_0^{\prime }\left(m\right)\right)}^T\begin{array}{c}\gamma \left(x_0\left(k\right),x_i\left(k\right)\right)=\frac{\Delta min+\rho \Delta max}{{\Delta }_{ik}+\rho \Delta max}\\ \Delta min=\underset{i}{min}{\mathrm{m}}_k\left|x_0\left(k\right)-x_i\left(k\right)\right|\\ \Delta max=\underset{i}{max}{\mathrm{m}}_k\left|x_0\left(k\right)-x_i\left(k\right)\right|\\ {\Delta }_{ik}=\left|x_0\left(k\right)-x_i\left(k\right)\right|\end{array}$$

(3)

In the equation, ρ is the distinguishing coefficient and takes values in (0, 1). The smaller the distinguishing coefficient, the greater the difference between the correlation coefficients and the stronger the differentiation ability, which is usually taken as 0.5. X₀ is the farmland quality sequence, X_i is the engineering measures sequence, and γ [X₀(k), X_i(k)] is the gray correlation coefficient between the farmland quality sequence and engineering measures sequence.

3. Results and Analysis

3.1. Changes in Farmland Quality Grade after Well-Facilitated Construction

The assessment results are shown in Figure 3. The distribution of farmland grades in the project area was grades 3 to 8 before the well-facilitated construction in 2019. The area of medium-grade farmland (grade 4 to 6) was the largest, with an area of 6109.46 hectares, accounting for 91.64% of the total farmland in the whole project area. The area of high-grade farmland (grade 3) was the smallest, with an area of 18.49 hectares, accounting for 0.24%. Low-grade farmland (grade 7 to 8) covered 538.71 hectares, accounting for 8.08%. After the completion of the new well-facilitated construction in 2020, the quality of farmland in the total project area was distributed among 3 to 8 grades. Most of the farmland was medium-grade and high-grade land between 3 to 6 grades. The area was 6307.13 hectares, accounting for 94.61% of the total farmland in the whole project area. Low-grade land was between 7 to 8 grades with an area of 359.54 hectares, accounting for 5.39%.

Comparing the change in the quality area before and after the completion of the well-facilitated farmland showed that after the construction of the well-facilitated farmland in 2020, the total area of low-grade land projects decreased from 538.71 hectares to 359.53 hectares. The area of medium-grade land dropped significantly by 106.70 hectares. Moreover, the area of high-grade land went up significantly. The area of grade 3 land increased by 285.87 hectares, up 4.31% of the total percentage. Therefore, the farmland quality grade in Xiao County significantly improved after the establishment of well-facilitated farmland in 2020. The average farmland quality grades were calculated before and after the establishment of the well-facilitated farmland. The average farmland quality grade was 5.65 for the total project area in 2019 and 5.06 after the establishment of the new well-facilitated farmland in 2020. The average farmland quality grade was improved by 0.59 grade, as shown in Figure 4.

3.2. Main Factor Indicators of Farmland Quality Analysis in Well-Facilitated Basic Farmland Impacted by Engineering Construction

Engineering construction measures that affect the farmland quality of well-facilitated fields are typical of water conservancy, agriculture, forest networking, and field road factors. The four engineering measures of the well-facilitated construction in Xiao County are soil fertilization, soil improvement, farmland infrastructure, and farmland ecological protection projects. The soil fertilization project mainly contains measures to increase the application of organic fertilizer, returning straw to the field, and deep plowing and deep loosening to improve the farmland’s soil fertility. The soil improvement project mainly includes measures such as foreign soil and the application of soil conditioner and organic fertilizer to improve the soil texture. The farmland infrastructure project includes measures of irrigation and drainage and field road construction. The farmland ecological protection project includes the construction of a farmland protection forest network and ditch management. The above four engineering measures have led to changes in the farmland quality of well-facilitated fields. Twelve representative factors, such as organic matter, soil bulk density, field–forest density, and irrigation and drainage assurance rates, were selected to analyze the degree of impact. According to the technical process requirements of the farmland fertility assessment determined in the National Technical Regulations for Farmland Fertility Survey and Quality Assessment, the indicators selected are shown in

Table 1. Indicator selection of engineering measures influencing the factors.

Engineering Measures	Principal Factors
Soil fertilization project	pH, organic matter, available phosphorus, rapidly available potassium
Soil improvement project	Texture configuration, tillage layer texture, soil bulk density
Farmland infrastructure project	Field road accessibility, irrigation guarantee rate, drainage guarantee rate
Farmland ecological protect project	Field–forest density, biodiversity

.

3.3. Principal Component Analysis of Factors Affecting Farmland Quality of Well-Facilitated Basic Farmland by Engineering Construction

3.3.1. Principal Component Analysis Test

The results of the principal component analysis test of the secondary influencing factors (

Table 2. KMO test and Bartlett test.

KMO Value		0.712
Bartlett sphericity test	Approximate chi-square	7230.051
	df	66.000
	p	0.000 ***

Note: *** represents significance levels of 1%.

) showed that the value of KMO was 0.712, which meets the premise requirements of principal component analysis. The Bartlett sphericity test (p < 0.05) showed that the data could be used for principal component analysis. The research data are suitable for principal component analysis.

3.3.2. Corresponding Relationship between Principal Components and Analytical Items

The principal component analysis is used to concentrate data information, condensing multiple analysis items into several key general indicators [35,36,37]. The principal component analysis is suitable for the extraction analysis of multiple indicators. Secondary factor indicators have different dimensions and orders of magnitude. In order to avoid its influence on the results, the raw data must be standardized prior to the principal component analysis. The standardized data were analyzed by principal component analysis. Four principal components were extracted from the 12 indicators. The eigenvalues, variance contribution rates, and cumulative variance contribution rates are shown in

Table 3. Eigenvalue and contribution rate of principal components.

Principal Component	Characteristic Root	Variance Explanation Rate (%)	Cumulative Contribution Rate (%)
1	3.210	26.751	26.751
2	2.322	19.353	46.104
3	1.444	12.030	58.133
4	0.930	7.753	65.886

. A total of four principal components were extracted from this analysis. The variance of each principal component is the eigenvalue, indicating how much the original information can be described by the corresponding component. Their eigenvalues were all greater than 0.9. The variance interpretation rate was 26.75%, 19.35%, 12.03%, and 7.75%, respectively. The cumulative variance explanation rate was 65.88%. The results showed that these four principal components reflect the most information of the four engineering measures.

The extracted principal component load matrix is shown in

Table 4. The principal component load matrix of engineering construction influencing indicators.

Name	Load Coefficient				Commonality (Common Factor Variance)
	Principal Component 1	Principal Component 2	Principal Component 3	Principal Component 4
Field road accessibility	0.806	0.238	−0.086	−0.106	0.724
Irrigation guarantee rate	0.827	0.239	−0.162	−0.186	0.802
Drainage guarantee rate	0.894	0.274	−0.126	−0.079	0.896
Forest network coverage	0.645	0.175	0.191	0.415	0.654
Soil bulk density	0.286	−0.457	0.220	0.240	0.397
Biodiversity	−0.207	0.728	−0.046	0.511	0.836
Rapidly available potassium	−0.309	0.435	0.411	−0.293	0.540
Organic matter	−0.184	0.842	0.045	0.238	0.802
Available phosphorus	−0.405	0.422	0.282	−0.186	0.457
Texture configuration	0.397	−0.283	0.518	0.179	0.537
Tillage layer texture	0.015	−0.262	0.720	0.159	0.613
pH	−0.296	−0.373	−0.519	0.391	0.649

. The principal component load matrix reflects the corresponding relationship and correlation degree between the secondary influencing factors and the four principal components. In this study, indicators with load values greater than 0.400 were selected as explanatory variables. The irrigation guarantee rate, drainage guarantee rate, and field road accessibility jointly affect Principal Component 1. Plus, all of these were positive loads. Hence, Principal Component 1 mainly reflects irrigation, waterlogging control, and road construction in field infrastructure construction. Soil bulk density, biodiversity, organic matter, available phosphorus, and rapidly available potassium jointly affect Principal Component 2. The positive load of organic matter was the largest, followed by the positive load of biodiversity, rapidly available potassium, and available phosphorus, playing a positive role in Principal Component 2. Principal Component 2 mainly reflects the measures of the soil fertilization project, including increasing the application of organic fertilizer, returning straw to the field, and deep plowing and deep loosening. The texture configuration, tillage layer texture, and pH were highly correlated with Principal Component 3. Therefore, Principal Component 3 reflects measures such as the application of soil conditioners in the soil improvement project. There was a high correlation between the forest network coverage and Principal Component 4. Thus, Principal Component 4 reflects the planting of trees and the construction of the farmland protective forest network in the farmland ecological protection project. In addition, we can also see that certain influencing factors affect two or more principal components at the same time, which can be discriminatively attributed according to professional knowledge. The commonality (the common factor variance) was less than 0.4, indicating a very weak relationship between the principal components and research items, while the opposite indicates a strong correlation. The correlations between the 12 secondary indicators and principal components were all greater than 0.4, which can be used in the study of principal component extraction.

According to the correspondence of each indicator, the study named Principal Component 1 was a field infrastructure project, Principal Component 2 was a soil fertilization project, Principal Component 3 was a soil improvement project, and Principal Component 4 was a farmland ecological protection project.

Table 5. Results of the principal component weight.

Name	Variance Explanation Rate	Cumulative Variance Explanation Rate	Weight
Field infrastructure project	0.268	0.268	40.602%
Soil fertilization project	0.194	0.461	29.373%
Soil improvement project	0.12	0.581	18.258%
Farmland ecological protect project	0.078	0.659	11.767%

shows the results of the principal component weight analysis based on information such as the load coefficient. The equation is the variance interpretation rate/the cumulative variance interpretation rate after the rotation. The weight calculation results of the principal component analysis showed that the weights of the field infrastructure project, soil fertilization project, soil improvement project, and farmland ecological protection project were 40.602%, 29.373%, 18.258%, and 11.767%, respectively. The maximum value of the indicator weight was 40.602% for the field infrastructure project, whereas the minimum value was 11.767% for the farmland ecological protection project, as shown in Figure 5.

3.3.3. Principal Component Nomenclature

According to the correspondence of each indicator, the study named Principal Component 1 was a field infrastructure project, Principal Component 2 was a soil fertilization project, Principal Component 3 was a soil improvement project, and Principal Component 4 was a farmland ecological protection project. Table 5 shows the results of the principal component weight analysis based on information such as the load coefficient. The equation is the variance interpretation rate/the cumulative variance interpretation rate after the rotation. The weight calculation results of the principal component analysis showed that the weights of the field infrastructure project, soil fertilization project, soil improvement project, and farmland ecological protection project were 40.602%, 29.373%, 18.258%, and 11.767%, respectively. The maximum value of the indicator weight was 40.602% for the field infrastructure project, whereas the minimum value was 11.767% for the farmland ecological protection project.

3.3.4. Correlation Order of Well-Facilitated Basic Farmland Engineering Construction Measures to Farmland Quality

The weighted mean values of the correlation coefficients between various engineering constructions and the corresponding elements of the well-facilitated farmland quality grade were calculated, respectively, with the aim of reflecting the correlation relationship between each engineering construction and the well-facilitated farmland quality grade. This is called the correlation degree and is denoted as [35]:

$$r_{0i}=\frac{1}{m}{\displaystyle \sum }_{k=1}^m{W}_k{\zeta }_i\left(k\right)$$

(4)

In the equation, r_0i is the correlation degree between the farmland quality sequence and engineering measures sequence, and m is the number of engineering measures.

The data obtained from the principal component analysis were sorted and analyzed in a gray relational analysis. The results are shown in

Table 6. Results of principal component correlation degree.

Assessment Item	Correlation Degree	Rank
Soil improvement project	0.939	1
Soil fertilization project	0.936	2
Farmland ecological protect project	0.857	3
Field infrastructure project	0.563	4

. The gray correlation coefficients were all greater than 0.5, indicating that the above engineering measures have a high correlation with the well-facilitated farmland quality in Xiao County and a strong influence on the well-facilitated farmland quality. In 2020, the correlation degree of well-facilitated farmland quality affected by the construction of the soil improvement project was the highest at 0.939, followed by the soil fertilization project at 0.936, and the farmland ecological protection project at 0.857; the field infrastructure project was the lowest at 0.563.

3.3.5. Correlation Order of Principal Factors

The gray relational analysis of the secondary influencing factors showed that there was a significant difference in the correlation degree between each factor and well-facilitated farmland quality (refer to

Table 7. Correlation degree of engineering construction secondary influencing indicators.

Assessment Item	Correlation Degree	Rank
Soil bulk density	0.865	1
pH	0.858	2
Biodiversity	0.824	3
Organic matter	0.791	4
Irrigation guarantee rate	0.769	5
Rapidly available potassium	0.747	6
Texture configuration	0.725	7
Available phosphorus	0.713	8
Drainage guarantee rate	0.711	9
Tillage layer texture	0.667	10
Field road accessibility	0.612	11
Field–forest density	0.610	12

). The soil bulk density had the strongest correlation with the farmland quality grade, indicating that soil improvement measures, such as deep plowing and deep loosening, play a key role in improving the soil’s properties. There was a high correlation among the pH, biodiversity, and organic matter, indicating that the soil fertilization measures, such as increasing the organic fertilizer, can improve soil fertility and well-facilitated farmland quality. The correlation degrees of the irrigation guarantee rate and drainage guarantee rate were more than 0.7, indicating that water hydraulic engineering is essential to the well-facilitated farmland quality. Furthermore, the correlation degree of each indicator was more than 0.5, indicating that each indicator is closely correlated to the well-facilitated farmland quality.

According to the comparison of the primary and secondary indicators, the factors affecting agricultural production have the greatest contribution to well-facilitated basic farmland quality, followed by factors affecting the field infrastructure project. The factors with the least contribution were those affecting field roads. The soil bulk density, pH, biodiversity, and organic matter showed the greatest correlation with the well-facilitated farmland quality in Xiao County, which are key factors affecting the quality. The irrigation guarantee rate, rapidly available potassium, and texture configuration were the second leading factors correlated to the well-facilitated farmland quality in Xiao County; yet, these are also important factors affecting well-facilitated farmland quality. The correlation between the forest network density and well-facilitated farmland quality was the lowest, and the impact on the quality was relatively weak. The results of the correlation degree analysis of the above secondary indicators also verified the correlation order of the four engineering measures to the well-facilitated farmland quality: the soil improvement project > soil fertilization project > farmland ecological protection project > field infrastructure project.

4. Discussion

• The farmland quality grade improved by 0.59 grade after the construction of well-facilitated fields. The gray correlation degree was the highest between the soil improvement and soil fertilization projects. This is different from the studies by Zhang et al. [38] and Chen et al. [39], who found that after the construction of well-facilitated farmland, the farmland quality grade improved, and the irrigation and drainage indicators showed the highest contribution to the improvement of the farmland quality grade. The results of the present study showed that the influencing factors of the farmland’s soil properties and fertility have the greatest contribution to well-facilitated basic farmland quality with the highest correlation degrees. Potential reasons for this situation are the different field conditions in each area, along with the different research methods used. Zhang [38] used the Telfer method to determine the weights and membership of relevant indicators, and the comprehensive index of the farmland quality was used to derive the formula for the influence of individual evaluation indicators on the comprehensive grade of the farmland quality, which is different from the principal component and gray correlation method used in this paper; the evaluation indicators selected by Chen [39] are different from this paper, and she includes indicators such as the degree of soil salinization, irrigation water source, and topographic slope to analyze the influence of well-facilitated farmland construction on the selected factors, which is different from the evaluation indices selected in this paper. In the construction of well-facilitated farmland, the unique regional characteristics should be considered, as well as the formulation of scientific and reasonable content and key points of engineering construction based on local conditions, and differential construction engineering measures should be carried out in an orderly manner. The results of this study showed that the correlation degree of the farmland ecological protection project was 0.857, second only to the soil improvement and fertilization projects, indicating that in the construction of well-facilitated farmland, more attention is needed for the improvement of basic farmland ecology in the future [40,41]. Tang [42] stated that on the basis of determining the ecosystem service value of farmland in different grades, the influence of well-facilitated basic farmland construction on the farmland ecosystem service value should not be underestimated. Well-facilitated basic farmland construction can not only improve the regional production capacity but can also significantly enhance the regional ecological level. The protection and promotion of the ecosystem [42] should not be overlooked.
• The results of this study showed that the correlation degree of the farmland ecological protection project was 0.857, second only to the soil improvement and fertilization projects, indicating that in the construction of well-facilitated farmland, more attention is needed for the improvement of basic farmland ecology in the future [40,41]. Tang [42] stated that on the basis of determining the ecosystem service value of farmland in different grades, the influence of well-facilitated basic farmland construction on the farmland’s ecosystem service value should not be underestimated. Well-facilitated basic farmland construction can not only improve the regional production capacity but can also significantly enhance the regional ecological level. The protection and promotion of the ecosystem [42] should not be overlooked. The results of the correlation order of the secondary influencing factors showed that the degree of correlation of organic matter was lower than that of the soil bulk density. The possible reason is that, although well-facilitated farmland construction increases the application of organic fertilizer, deep plowing and deep loosening [43,44] affect the absorption of the soil’s organic matter, which perhaps invisibly reduces the contribution of organic matter to the improvement of farmland quality [45]. This also verified that the correlation degree of the soil improvement project among the primary influencing factors of the engineering measures is higher than that of the soil fertilization project.
• The well-facilitated farmland construction project showed strong regional, complex, and systematic characteristics. On the theoretical level, the number of dimensions selected for analysis in this study was relatively small. The selected influencing indicators of engineering in the review measures for the well-facilitated farmland construction in the farmland quality grade, and the data obtained have certain limitations. Some variables (the field size [46], ditch density [47], and soil microorganisms [48]) were not included in the analysis. There is also a lack of theoretical support in the quantification of individual indicators (such as biodiversity)
• At this stage, the research on well-facilitated farmland construction is not in-depth [8], detailed [9], and systematic [10], and the research results are difficult to be quickly applied in reality [11]. In the future, research on the construction of well-facilitated farmland should introduce landscape ecological construction [41], landscape pattern construction [42], aesthetic theory [43], etc., in the macroscopic research of land remediation and soil reconstruction, the soil’s physical, chemical and, biological improvement techniques [45], the efficient and rational use of water resources [40], agronomic measures [44], and changes in land use patterns in the microscopic research. In addition, the research will be combined with information technology to develop new ideas and methods for the construction of well-facilitated farmland. The correlation degree of the engineering measures, such as land leveling and the promotion of science and technology on the farmland quality grade, is worth analyzing, which enhances the impact degree of well-facilitated farmland construction engineering measures on the farmland quality grade. This can provide effective and intuitive examples for subsequent proposed projects and is conducive to the selection of well-facilitated farmland engineering types suitable for the region.

5. Conclusions

This paper investigates the changes in the quality level of farmland before and after the construction of well-facilitated farmland and the effects of the well-facilitated engineering construction measures on the quality level of farmland. The research results will provide a reference for future engineering measures of well-facilitated farmland construction.

• The farmland quality grade of well-facilitated farmland was improved by 0.59 grade after the construction of the engineering measures. The quality grade was 5.65 before the construction and 5.06 after the construction.
• According to a gray relational analysis, the degree of correlation between the soil improvement project and well-facilitated farmland quality was the highest at 0.939, followed by the soil fertilization project at 0.936 and the farmland ecological protection project at 0.857; the field infrastructure project was the lowest at 0.563.
• The correlation degree of the secondary influencing factors of the engineering measures from high to low was the soil bulk density, pH, biodiversity, organic matter, irrigation guarantee rate, rapidly available potassium, texture configuration, available phosphorus, drainage guarantee rate, tillage layer texture, field road accessibility, and field–forest density.
• The principal component analysis and gray correlation can effectively analyze and evaluate the indicators, and the final results obtained are easy to compare and select, which can be used as the analysis method for the impact of well-facilitated farmland construction on the quality of farmland.