**1. Introduction**

Land use transformation was first proposed by the British geographer Alan Grainger, inspired by the concept of forest transformation proposed by the Finnish scholar A.S. Mather [1]. The transforming land types mainly include cultivated land [2–4], forest land, urban land [5,6], rural land, and other land use types, and they also include the overall

**Citation:** Sun, Y.; Zhou, Z.; Huang, D.; Chen, Q.; Fang, M. The Spatial and Temporal Evolution Pattern and Transformation of Urban–Rural Construction Land in Karst Mountainous Areas: Qixingguan District of Guizhou, Southwest China. *Land* **2022**, *11*, 1734. https:// doi.org/10.3390/land11101734

Academic Editor: Xiaoyong Bai

Received: 28 August 2022 Accepted: 1 October 2022 Published: 7 October 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

regional land use type [7,8]. With the acceleration of urbanization and increasing human activities, the transformation of urban–rural construction land in land use transformation has become an important socioeconomic phenomenon worldwide [9,10]. Urban–rural construction land is the core component of the land use system. It is not only an important indicator to control urban expansion and urban–rural construction, but also the main component of urban spatial planning and the spatial carrier of human non-agricultural economic production activities [11–13]. At present, under circumstances in which global warming is causing glaciers to melt and inundate some coastal areas, forests are sharply reduced, the environmental carrying capacity is fragile, and human activities affect vegetation restoration [14–16], how to realize scientific and sustainable utilization of limited land resources has become a key issue for the future development of the world. Whether urban–rural construction land can be reasonably controlled and allocated will directly affect the protection of cultivated land and the development of urbanization.

In recent years, many scholars have studied the evolution and driving factors of urban–rural construction land by applying current land use data and using different methods [17–20]. Weber and Puissant extracted land cover from 1986 to 1996 with time series satellite images (SPOT XS), used a prediction model to carry out an empirical analysis of the expansion characteristics of local construction land in Tunisia, and explained the development trend of the city in the future [21]. Based on Landsat satellite observation data, Masek et al. used the band reflection of NDVI to distinguish urban and agricultural land, and dynamically monitored the expansion and evolution process of urban construction land in Washington, DC, from 1973 to 1996. In conjunction with census data, they found that urban expansion was strongly correlated with the regional economic development level [22]. Saizen et al. quantitatively analyzed land use changes in the Osaka metropolitan area of Japan from 1979 to 1996 through GIS raster data, and the results showed that the main reason for the continuous increase in idle land in the suburbs was urban sprawl [23]. Ustaoglu and Aydmoglu used an integrated geographic information system (GIS) and multicriteria decision analysis (MCA) approach to assess the suitability of land use in the Pendikc area of eastern Istanbul, Turkey, for residential, industrial, commercial, and recreational development in the city [24].

Mann used a regression analysis model to study the main driving factors of changes in per capita construction land in rural areas and suggested that the implementation of incentive measures for construction land management by local governments could limit the expansion of construction land to a certain extent [25]. Colsaet et al. analyzed the scientific literature on the occupation of agricultural land by urban construction land and the determinants of urban expansion from 1990 to 2017, and suggested that population and income growth, transportation infrastructure, and car use were the main driving factors [26]. Bittner et al. studied the spatial evolution of land use in peri-urban areas of Israel based on time-series land use data supplemented by summary statistical analysis, and indicated that rural reorganization would have an important impact on the economic society and ecological environment [27]. Diogo and Koomen studied the process of land use change in Portugal between 1990 and 2006 and analyzed the impact of different driving forces on the formation of land use patterns during this period. They concluded that land expansion was positively correlated with economic development, and the driving influence of economic factors on land use change remained stable in a certain period.

The deployment of new infrastructure and the gradual implementation of territorial space planning policies will also affect land expansion [28]. However, the results of previous works illustrate that the main driving factors for the evolution of urban–rural construction land are the economic and social development level, social living conditions, and the policy and institutional environment, etc. [29,30]. Overall, most studies on the influencing factors of spatiotemporal changes of urban–rural construction land are based on linear analysis, trend analysis, and correlation analysis; however, it is still difficult to quantitatively decompose the influencing factors of such land changes.

The above studies are of great significance for understanding the driving factors of urban–rural construction land changes, but the disadvantage of the methods used is that they assume there is a constant and significant linear relationship between the driving forces and land changes across the entire time series. An idealized linear model or qualitative description can reveal the complex driving forces; in fact, there is no strict linear relationship between the transformation of urban–rural construction land and socioeconomic development, population density [31], or urban residential development [32]. Although studies have applied algorithms such as K-means [33] for classification and partitioning, statistical methods for spatial differentiation need to be further developed.

Geographical detector comprises a group of statistical methods that detect spatial differentiation and reveal driving forces [34]. Therefore, this study applied the geographic detector model to analyze the characteristics and driving mechanisms of the transformation of urban–rural construction land in typical karst mountain areas to explain the interactions between social and economic activities and changes in such land. The results of the present study are intended to help local governments explore whether any unreasonable, unsustainable land use resulted from unsustainable human activities and development practices, and thus adapt to current village planning and urbanization strategies and optimize the efficiency of land resource allocation. Moreover, this study can also provide reference for the urban–rural development of other karst areas and promote the integration of urban and rural transformation and land use transformation.

#### **2. Materials and Methods**

#### *2.1. Study Area*

The present study area is located within Qixingguan District (27◦03 –27◦46 N, 104◦51 – 105◦55 E), Guizhou Province, southwest China, on the slopes of the Yunnan–Guizhou Plateau, sloping toward the eastern low mountains (Figure 1) and covering an area of 3411.14 km2. The study area is adjacent to Yunnan Province in the west and Sichuan Province in the north and is the transportation and logistics hub of southwest China [35]. The elevation ranges from 456 to 2210 m. The terrain is high in the west and low in the east. The landform type is mainly mid-size mountains and hills. The western area has a concentration of high mountains, mid-size mountains, and valleys, and the terrain is steep, with ravines and mountains intertwined in the northeast. Most of the central area consists of river valley flats and middle mountains [36]. The type of soil in this area is mainly yellow soil (43.27%), which develops from limestone, followed by lime soil, yellow brown soil, paddy soil, and coarse bone soil [37]. This area is located in the hinterlands of the Wumeng Mountains. It is a typical karst mountain area with completely developed karst landforms and severe rocky desertification, resulting in a low ecological environment capacity and a fragile ecologic system. In terms of economic development, in 2020, the gross regional product of Qixingguan District was RMB 50.006 billion, a year-over-year increase of 4.4%; the fixed asset investment of the region increased by 3.7% over the previous year; the total fiscal revenue of the region was RMB 11.749 billion, an increase of −2.4% over the previous year; and the general public budget revenue was RMB 2.572 billion, an increase of −9.1% over the previous year.

**Figure 1.** Topographic and location map of study area.
