*4.2. Variation Characteristics of Landscape Pattern at Landscape Level in Flatland Areas* 4.2.1. Landscape Diversity

From 1995 to 2018, the SHDI in the flatland area was mainly grade II, III and IV, accounting for 78%–80% of the total, while grade I and V were relatively small, with a total proportion between 20% and 22% over the years (Figure 5b). With the construction of roads and the expansion of residential areas, the concentrated contiguity of cultivated land was divided, and the SHDI clearly increased, with the low value area decreasing and the middle and high value areas increasing. The proportion of landscape diversity level I decreased from 11.46% to 7.25%, the proportion of grade II decreased from 24.14% to 19.72%, and the proportion of grade III decreased from 29.36% to 27.46%. It can be seen that although the landscape diversity is on a downward trend, the decline rate becomes ever smaller from grade I to III, indicating that the decrease is mainly caused by the decrease in the low value area. The proportion of class IV increased from 25.52% to 29.50%, and the proportion of class V increased from 9.52% to 15.17%, with the highest increase rate of class

V. In the study period, the periodic changes were obvious, and the changes in the periods 1995–2000 and 2000–2005 were significantly higher than those in the other three periods (Figure 9). In the first four periods, the area with increased SHDI was smaller than that with decreased SHDI. However, a reversal occurred from 2015 to 2018; that is, the area with decreased SHDI was larger than that with increased SHDI, indicating that the landscape complexity began to decline at this stage and the complexity degree decreased.

**Figure 9.** Spatial distribution of SHDI in flatland areas of Yuxi City from 1995 to 2018. (**a**) 1995; (**b**) 2000; (**c**) 2005; (**d**) 2010; (**e**) 2015; (**f**) 2018.

#### 4.2.2. Landscape Shape

The LSI of the flatland area is mainly graded as II, III and IV. The proportion of grade I and II showed a downward trend, and the phased changes were first decreasing and then increasing (Figure 5d). The proportion of class III, IV and V showed an overall upward trend, and the phased changes were first increasing and then decreasing. Taking 2000 or 2005 as the boundary, the proportion of class I decreased rapidly from 1995 to 2000 and then increased and the proportion of class II decreased continuously in the period 1995–2005, then increased, while the proportion of class IV and V both peaked in 2005 and subsequently decreased. During the study period, the phased changes were clear, and the first two periods were completely the opposite to the last three periods. During the periods 1995–2000 and 2000–2005, the area of the LSI that increased was larger than that of the decreased. However, during the periods 2005–2010, 2010–2015 and 2015–2018, the area of the LSI that decreased was greater than that of the increased (Figure 10).

#### 4.2.3. Largest Patch

From 1995 to 2018, the proportion of class I increased from 15.36% to 21.21%, and class II increased from 25.46% to 27.33%, indicating that the low value area of the LPI in the flatland area continued to expand (Figure 5f). However, the proportion of class III decreased from 27.45% to 25.01%, class IV decreased from 19.64% to 17.46%, and class V decreased from 12.09% to 9.06%. This indicated that the medium and high value area continued to shrink and that the landscape fragmentation degree clearly increased. During the study period, the overall change in the LPI decreased, mainly during the periods 1995–2000 and 2000–2005 (Figure 11). During the period 1995–2000, the area of the LPI that decreased was 45,946.49 hm2, which was 4.01 times that of the index increased. From 2000 to 2005, the area of the LPI that decreased reached 41,649.45 hm2, which was 1.51 times the increased area. In the first four time periods, the index decreased areas were greater than the increasing areas, but the index increased area was greater than the decreasing area in the period 2015–2018.

**Figure 10.** Spatial distribution of LSI in flatland areas of Yuxi City from 1995 to 2018. (**a**) 1995; (**b**) 2000; (**c**) 2005; (**d**) 2010; (**e**) 2015; (**f**) 2018.

**Figure 11.** Spatial distribution of LPI in flatland areas of Yuxi City from 1995 to 2018. (**a**) 1995; (**b**) 2000; (**c**) 2005; (**d**) 2010; (**e**) 2015; (**f**) 2018.

#### **5. Discussion**

This study showed that the spatial and temporal trends in landscape pattern evolution in mountainous areas and flatland areas were the same: both showed an increased fragmentation degree and decreased connectivity degree, but there were also clear differences between them. At the class level, the fragmentation degree of construction land in the flatland area was significantly higher than that in the mountainous area, but its patch area was larger and the layout was more concentrated than that in the mountainous area, while the construction land in the mountainous area was more dispersed, mainly because the human activity intensity in the flatland area was significantly higher than that in the mountainous area due to the superiority of natural and socio-economic conditions [40]. Similar to the construction land, the landscape index of cultivated land in the flatland area was higher, but the Area\_MN continued to decline, showing clear fragmentation but also a concentrated layout, mainly because the flatland area had flat terrain and sufficient hydrothermal conditions, which were conducive to cultivation [36]. Over time, however, human interference in the cultivated land has intensified [41]. The PD and ED of forest and grass in the flatland area were significantly higher but the Area\_MN and LPI were significantly lower than those in the mountainous area, in fact only about a third of those in the mountainous area. Combined with the implementation time of regional land management policies, mainly affected by the policy of returning cultivated land to forest and grass during the study period, the degree of fragmentation of forest and grass in the flatland area was significantly higher than that in the mountainous area, while the forest and grass in the mountainous area showed an obvious trend of concentration and contiguity. At the landscape level, landscape diversity in the flatland area was significantly higher than that in the mountainous area. In the relatively low-lying area for urban construction, the urban construction and development interspersed the urban patches with the patches of cultivated land, forest and grass, with the consequence that the SHDI continued to rise, and the landscape fragmentation degree also continued to increase. The landscape shape in the mountainous area was simpler than that in the flatland area. Due to the implementation effect of the policy of returning farmland to forest and the rapid improvement of the level of social and economic development, the LPI in the flatland area continued to increase, and the landscape units in the mountainous area dominated by forest and grass tended to be more concentrated and contiguous [42].

Topography plays a key role in the formation of landscape patterns, determining the basic landscape pattern [43], and the differentiated development of the social economy will further affect the change in the landscape pattern. With the increase in topographic relief, the man-made landscape gradually gives way to the natural landscape. In terms of influencing landscape unit distribution, human factors usually dominate in flatland areas, while natural factors usually dominate in mountainous areas. The mountainous area is high in elevation and slope, and the topography is undulating, with the consequence that the accessibility is worse than in the flatland area. At the same time, it can be difficult to meet the demand for high-quality land brought by population growth and the pursuit of a prosperous life, which makes most of the population migrate to the flatland area [44], bringing about the transformation of cultivated land landscapes to forest and grass landscapes. It has been found that the landscape pattern of mountainous areas in southwest China is affected by the landforms of high mountains and river valleys, mainly forest and grass [45], while high-quality arable land mainly continues to be distributed in basins, trough valleys and low mountain valleys [46]. Before 2000, with the increase in population, in order to meet the needs of survival, the mountainous area was blindly reclaimed and the forest was destroyed [47], while the landscape diversity was higher than that of the flatland area [48]. After 2000, due to road construction and settlement expansion, concentrated and contiguous cultivated land was divided, and the landscape diversity of the flatland area increased significantly. However, due to ecological restoration or vegetation degradation, the landscape types in mountainous areas gradually became single, and the landscape diversity declined [37].

The comparison and analysis of landscape spatial patterns between mountainous areas and flatland areas based on a grid method and further microscopization and refinement from the scale [49] are of great importance for clarifying the difference in landscape patterns between mountainous areas and flatland areas on the micro scale. The difference in the spatial and temporal evolution of landscape patterns in the mountainous area and flatland area leads to the consideration of coordinated and sustainable development of mountains and flatlands. Appropriate human intervention appears to help enhance the diversity of the landscape, while inappropriate human intervention will exacerbate the problem of landscape fragmentation [50–52]. It is the differentiated characteristics of the social and economic development that form the differentiation in landscape pattern evolution between mountainous areas and flatland areas. The level of "landscape diversity" is inversely correlated with that of "biological diversity". Broken "landscape diversity" is not conducive to "biological diversity" [53], because the contact surface between the landscape system and the environment is large, and the "hinterland" is not deep, which is not conducive to the recovery of some species in the "biological chain". The decline in the landscape diversity in mountainous areas is the result of the connectivity of forest and grass. The enhancement in the connectivity of forest and grass in the mountainous area further enriches biodiversity and is conducive to the restoration of the ecosystem (animal habitat and reproduction) [54]. The landscape diversity of the flatland area is a response to the development of the society and economy. The expansion of urban construction land is conducive to population agglomeration and job creation [55]. The development of the transportation industry, although it has brought about an increase in landscape fragmentation, has facilitated the circulation of people and materials [41]. The in-depth intersection of arable land patches and urban construction land patches further expands the rural and urban interface, making it more conducive to the connection between the sales and consumption chains of agricultural products. All in all, the mountainous area provides an ecological barrier to the social and economic development of the flatland area. While the social and economic development level of the flatland area is constantly improved, it feeds the mountainous area and provides the economic foundation for the further optimization of the mountainous area and the construction of an ecological environment.

## **6. Conclusions**

Despite the uncertainties in the interpretation accuracy of the landscape units, as well as the small extent of the study area used, some meaningful conclusions can be drawn from this research. Preliminary results show that the analysis of mountain–flatland landscape pattern evolution based on the grid scale can effectively reveal the variation difference and coupling law.

In the past 24 years, the landscape pattern of the mountainous area and the flatland area in Yuxi City has shown periodic changes, and the trend of its evolution is consistent with the laws of human social and economic development. With the further development of the social economy, the landscape fragmentation and landscape diversity in the flatland area are clearly higher than those in the mountainous area, and the degree of landscape fragmentation is further intensified, while the landscape shape in the mountainous area is simpler than that in the flatland area, and the trend in landscape concentration and contiguity is obvious. The natural landscape of forest and grass in mountainous areas continues to expand and tends to be contiguous, while the man-made landscape in flatland areas constantly increases and shows fragmentation, which reflects the pattern characteristics formed by the coupling evolution of land use between the mountain and flatland. There is a coupling linkage relationship between the landscape pattern evolution of the mountainous area and the flatland area. The urban expansion and the increase in construction land in the flatland area are mutually causal with the decrease in cultivated land and the increase in forest and grass in the mountainous area.

In future studies, we can further improve the interpretation accuracy of the landscape units and carry out further studies by taking the whole southwest mountainous area as the research area, in order to find the general coupling laws of landscape pattern evolution between mountainous areas and flatland areas. The rationality and universality of the law will be verified by statistical inspection and analysis, based on which the coordinated and sustainable development countermeasures of mountainous areas and flatland areas in the southwest karst region will be formulated.

**Author Contributions:** Conceptualization, L.W. and B.X.; methodology, L.W.; software, L.W.; validation, L.W. and J.Z.; formal analysis, L.W.; investigation, J.Z.; resources, J.Z.; data curation, L.W.; writing—original draft preparation, L.W.; writing—review and editing, B.X.; visualization, J.Z.; supervision, B.X.; project administration, L.W.; funding acquisition, L.W. and J.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China, grant number 42161041, Science and Technology planning Project of Yunnan Province, grant number 202305AC160089, and the Youth Project of Science and Technology Agency in Yunnan Province, grant number 202101BA070001-275, as well as the General Project of Science and Technology Agency in Yunnan Province, grant number 202101BA070001-078.

**Data Availability Statement:** Not applicable.

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

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
