*3.2. Analysis of Driving Factors of Urban–Rural Construction Land Transformation* 3.2.1. Analysis of Detection Factor Influence

By testing the driving factors of urban–rural construction land, the influence of testing factor on driving factors was revealed. The PD value of each driving factor was calculated by the geographic detector (Table 3), and the influence of each factor on urban–rural construction land was determined. Table 3 shows the order of impact of various driving factors on urban–rural construction land (from high to low): total fixed asset investment in society as a whole > total fiscal revenue > road network density > total population at the end of the year > urbanization rate > per capita GDP > total annual wages of employees per unit > average GDP > total industrial output > population density > average slope > average elevation. According to the detection results for the driving factors, the total fixed assets investment for the whole society had the greatest impact on the growth of urban–rural land with an explanatory power of 0.819, which is similar to the results of studies conducted in China and internationally [46,47]. It is believed that total fixed assets investment improves people's living standards, stimulates the development of urban real estate, and promotes the expansion of urban construction land.

Figure 5 shows that the PD values of population density and total industrial output from 2009 to 2020 generally show increasing trends, and the PD values of other driving factors have increases and decreases during the monitoring period. From 2009 to 2013, except for increased PD values, average slope, average elevation, road network density, population density, and total industrial output value, the PD values of other driving factors showed decreasing trends to varying degrees. From 2013 to 2017, the PD values of total fixed asset investment of the whole society, average slope, average elevation, road network density, etc., showed a decreasing trend. From 2017 to 2020, the PD values of other driving factors showed an increasing trend, except for values for per capita GDP, total population at the end of the year, total fiscal revenue, and road network density. To summarize, the explanatory power of the driving factors during each monitoring period for Qixingguan District regarding the expansion of urban–rural construction land varies. Therefore, time series monitoring of driving factors would have important guiding significance for predicting the expansion of urban–rural construction land.

**Figure 5.** Change in PD value of detection factor in Seven Star Pass area from 2009 to 2020. X1, per capita GDP; X2, average GDP; X3, urbanization rate; X4, population density; X5, total population at end of year; X6, total industrial output; X7, total fixed assets investment of whole society; X8, total annual wages of employees per unit; X9, total fiscal revenue; X10, average slope; X11, average elevation; X12, road network density.


**Table 3.** PD values of detection factors.

#### 3.2.2. Interaction Analysis of Detection Factors

The interaction detection in the geographic detector mainly identifies the degree of influence of interactions between driving factors on changes in urban–rural construction land in Qixingguan District. On this basis, interaction detection evaluates whether the influence of factors is independent or if it increases or decreases the explanatory power of the evolution of urban–rural construction land due to the interaction. Table 4 shows the relationships between driving factors affecting the evolution of urban–rural construction land by using the interaction detector. The results indicate that there are no factors that independently affect urban–rural construction land, and the interaction effect between driving factors is mainly manifested in the relationship of nonlinear enhancement and mutual enhancement, and the explanatory power of the interaction between all factors increases significantly: X9∩X3 (0.915) > X5∩X3 (0.891) > X7∩X3 (0.882) > X12∩X3 (0.794) > X10∩X3 (0.707) > X6∩X3 (0.642) > X4∩X3 (0.591) > X11∩X3 (0.544). In particular, a significant interaction exists between other factors and the completion of fixed assets and total fiscal revenue in the whole society, and the interaction between total fiscal revenue and total industrial output has the strongest explanatory power. Overall, the influences of the driving factors on the evolution of urban–rural construction land are not independent, nor do they represent a simple superposition process. Rather, a mutual or nonlinear enhancement effect is shown, and the influence of the interactions between detecting factors on the spatial differentiation of urban–rural construction land is greater than that of a single factor.

**Table 4.** Interaction of detection factors.


#### *3.3. Discussion*

3.3.1. Unreasonable Structure of Urban and Rural Construction Land

Although the average annual growth rate of urban land in the study area (8.87%) is higher than that of rural residential land (2.18%), the total rural residential land area exceeds that of urban construction land. With regard to the proportion of urban–rural construction land in the study area, the proportion of rural residential land is the largest, and the map plot is broken. Although the proportion of the area has decreased since 2009, the proportion of rural residential land was still as high as 41.75% by 2020. Moreover, the number of map spots was almost twice that of urban construction land. For urban land, the area continued to increase in the four monitoring periods of the study; the area ratio increased from 21.85% in 2009 to 28.86% in 2017, it but decreased by 4.44% during the period 2017–2020. During the 13th Five-Year Plan period in China, the urbanization rate of

the resident population in the study area increased to 54%, indicating that the proportion of land and population in rural areas is seriously imbalanced as a result of less population occupying more land. China's special long-term dual urban–rural structure is mainly manifested in a continuous increase in the demand for construction land in cities, and the continuously supply still occasioned an increasing shortage of construction land. However, in the countryside, it is manifested in the continuous increase in the population moving to cities, and the construction land does not decrease but increases. A large amount of rural construction land is thus idle [51].

The present study area is located in an underdeveloped area of karst mountains in southwest China, and the problem of unbalanced and insufficient urban and rural development is particularly prominent. Compared with China's economically developed regions (e.g., Yangtze River Delta) [52], the main features are that the structure of urban– rural construction land is unreasonable, the area of rural residential land is larger than the area of urban land, the population and residential areas are obviously scattered, and the driving forces of urban and rural development and urbanization are mainly external (e.g., state investment, project-driven development, administrative promotion), thus the development of urban–rural integration is limited by economic development, difficulty integrating urban and rural resources, etc. In other countries, such as the United States [53], Japan [54], the United Kingdom [55], and France [56], urban–rural integration has basically been achieved because it was developed earlier and their economies are more developed, resulting in fewer significant differences between urban and rural areas. However, China is still in the early stage of urban–rural integration; especially in the karst mountain area with fragile ecological environment, the promotion of urban–rural development is costly, difficult, and slow.

To address the problems in urban and rural land planning, Qixingguan District should formulate policies to control the expansion of urban construction land and the consolidation of rural land based on the experience of developed countries to deal with the relationship between urban and rural land to a certain extent. It is necessary to pay closer attention to the management of rural construction land, and to optimize the allocation of rural land and agricultural development through rural land consolidation, sorting out more land that can free up more land targets for the expansion of the central urban area. Moreover, urban land should be controlled within a certain range, in order to realize rational urban development and adjust the structure and layout of rural residential and urban land. This would satisfy the common development of urbanization and industrialization and improve rural development and urban–rural relations, which would improve the efficiency and effectiveness of land use, and at the same time avoid the continuous spread of the "hollowing out" phenomenon in rural areas.

#### 3.3.2. Transformation of Cultivated Land into Construction Land

From 2009 to 2020, the main source of new urban–rural construction land in the study area was agricultural land (more than 89%, of which the minimum occupied cultivated land was 57.72%). Cultivated land converted to new urban–rural construction land amounted to 13,328.91 hm2, but construction occupying cultivated land slowed down to a certain extent. From 2009 to 2020, the urban–rural construction land mainly transformed into reclaimed construction land. The area of reclaimed land transformed into cultivated land was 3786.97 hm2; 61.20% of the transferred urban land was reclaimed as cultivated land, while 57.86% of the transferred rural residential land was reclaimed as cultivated land.

In terms of stages, the proportion of urban–rural construction land reclaimed as cultivated land shows a fluctuating trend, but the area of cultivated land transformed to urban–rural construction land continues to increase. In terms of area, the reduction of urban–rural construction land to grassland, garden, water, and other types of land is much smaller than the area of cultivated land and forest land, indicating that, in recent years, a large number of supplementary cultivated land projects were carried out in the study area, which will contribute to the protection of cultivated land to a certain extent. It is worth noting that under the condition that the land area remains stable, if the current growth trend is maintained and the cultivated land is recklessly constructed, it will lead to expanded urban development and the unreasonable and disorderly growth of rural residential land. The "red line" of cultivated land will be broken. At the same time, it will inevitably lead to wasted land resources and increased carrying capacity of land. Therefore, in the process of land development in the study area, more attention should be given to excavating the existing rural residential land, reforming and improving the countryside homestead management, and promoting the efficient and intensive use of urban–rural construction land by the use of clearing property rights, compensation systems for the use of resources, and standardized circulation. At the same time, it is also necessary to strengthen the supervision of the red line of cultivated land protection, resolutely implement the policy of linkage between urban land taking and rural land giving (LUTRG), and prohibit construction and development in non-construction areas, especially in karst areas with a fragile ecological environment.

#### 3.3.3. Regular Evaluation of Planning and Dynamic Revision

From the detection results of driving factors (Section 3.3), the completion of fixed assets in the whole society has the greatest impact on the growth of urban and rural land use, and the PD value is 0.819, which agrees with previous works in other regions [57,58]. The completion of fixed asset investment in the whole society improved people's living standards, stimulated the development of urban real estate, and promoted the expansion of urban construction land. At the same time, our results imply that the PD value of each driving factor and the interaction between factors of urban and rural construction land are not constant in different monitoring periods and show dynamic changes with fluctuation. For example, the PD value of road network density as a single factor of urban–rural construction land was lower than 23% in 2009 and 2020, but it was greater than 70% in 2013 and 2017. However, China's current laws and regulations, such as those regarding land use planning or urban and rural planning, focus on vision planning, in an effort to develop an ideal planning scheme to achieve the "ultimate goal" throughout the planning period, while ignoring the suitability between the processes of "static" planning and "dynamic" implementation. In addition, there is limited ability to quickly and synchronously adjust the planning scheme when identifying changing factors, resulting in a greatly reduced implementation effect.

According to the changes in the influence of driving factors and the actual situation in the study area, the present study suggests that formulating land-related laws and regulations does not mean planning an ultimate blueprint, but a strategic and structural type of flexible and dynamic planning. Based on this, this study suggests that detecting the driving factors of urban–rural construction land in the study area should be included in the city-level urban physical examination category of "territorial space safety". The formation of the annual urban physical examination index system should be optimized, the key driving factors of urban–rural construction land in each monitoring period in the study area should be identified, and the implementation of the plan should be regularly scheduled, evaluated, and adjusted during planning. This not only could improve the utilization of urban–rural construction land resources and the efficiency of land resource utilization, but also more accurately predict the reasonable demand and spatial expansion direction of urban–rural construction land in the future.

#### 3.3.4. Research Prospects

The following aspects should receive more attention in future work: (1) Research on the transformation of urban–rural construction land only considers the transfer between construction land and non-construction land, and mainly focuses on its dominant characteristics (quantity structure, source destination, spatial differentiation, etc.). Thus, it is necessary to consider urban–rural construction land in the next step in terms of the transformation between types and the analysis of the interaction mechanism of strengthening the hidden characteristics (input–output, utilization efficiency, economic density, function, etc.), to gain an in-depth understanding of the transformation characteristics and driving mechanism of urban–rural construction land. (2) The acquisition of basic data and indicators was limited to only four periods of land use data (2009, 2013, 2017 and 2020), which was not able to fully reflect the long-term and continuous process of urban–rural construction land transformation, resulting in insufficient research depth. It is extremely important for research to obtain multitemporal land use data and make comparisons with the natural environmental conditions, economic and social levels, policy and institutional environment, and other data in the same period. (3) The spatial distribution law, evolution process, and driving mechanism of urban–rural construction land reflect the interactive human–land relationship. Based on the limited availability and accessibility of data in this study, the selection of driving indicators of urban–rural construction land transformation needs to be further improved, and in the selection of indicators, the categories can be further increased, such as forest cover [59], technological progress, food security [60], karst basin water re-sources [61], etc., to clarify the mechanism of the driving factors of urban–rural construction land transformation more clearly. (4) Research results on carbon peaking and carbon neutralization surge since the double carbon target was proposed [62,63], the carbon emissions of urban–rural construction have attracted more attention [64]. In order to optimize carbon neutrality in new urbanization construction, more studies should be conducted that combine the processes of carbon emission reduction and carbon trading, relying on this huge carbon sink [65] to achieve the goal of carbon neutrality [66].

#### **4. Conclusions**

Using spatial analysis and geographical detectors, this study conducted a detailed analysis of the spatial and temporal evolution characteristics of urban–rural construction land in the study area and explored the effects of their driving factors. The main research conclusions are as follows:

(1) With regard to time series, the urban–rural construction land increased in Qixingguan District from 2009 to 2020, and most of it came from agricultural land. The proportion of cultivated land among the various land use types is not less than 57.72%. The fastest growing land types are rural residential, urban, and transportation land. The expansion of rural residential land is greater than that of urban land, and it continues to grow. With regard to the spatial distribution pattern, the spatial distribution expands from the surrounding area of the central city; until 2020, the distribution was medium high and high density in the northern, eastern, and southwestern townships of the study area, and medium low density in the central urban area.

(2) From 2009 to 2020, the ranking of various driving factors on urban and rural construction land in terms of impact was as follows (from high to low): total fixed asset investment in society as a whole > total fiscal revenue > road network density > total population at the end of the year > urbanization rate > per capita GDP > total annual wages of employees per unit > average GDP > total industrial output > population density > average slope > average elevation. The PD value of total fixed asset investment in the whole society and total fiscal revenue is above 60%; thus, they have become the main driving factors affecting the change of urban–rural construction land.

(3) The driving factors have interactive effects in terms of their impact on urban– rural construction land, and these effects show mutual and nonlinear enhancement. The present study reveals that the PD value of the driving factors in Qixingguan District in each monitoring period has changed. This study reveals the impact of these driving factors on urban–rural construction land and provides a foundation for studying the dynamics of the transformation of such land.

**Author Contributions:** Y.S.: Conceptualization, Methodology, Software, Writing—original draft. Z.Z.: Resources, Data curation, Formal analysis, Project administration, Funding acquisition. D.H.: Visualization. Q.C.: Supervision. M.F.: Investigation. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the State's Key Project of Research and Development Plan of China (2018YFB0505400), the National Natural Science Foundation of China (41661088), and the Guizhou Province High-level Innovative Talent Training Plan "Hundred" Level Talents (Qiankehe Platform Talents [2016]5674).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** This study was supported financially by the State's Key Project of Research.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**

