Analysis of Temporal and Spatial Changes of Cultivated Land in Karst Areas of Southwest Guangxi Based on Landscape Pattern Indices

Qiuyue YIN, Yu WANG, Jinlei YIN

Abstract Affected by topographical conditions， the distribution of cultivated land in karst areas is relatively fragmented. More effective cultivated land protection measures can be taken in a targeted manner by grasping the spatial and temporal distribution characteristics of regional cultivated land. In this study， with the cultivated land of four county-level administrative regions （Debao County， Jingxi City， Daxin County， and Tiancheng County） in typical karst landform areas in southwestern Guangxi as the research object， combining with the spatial analysis function of the GIS software platform， the spatial distribution characteristics and temporal and spatial changes of cultivated land in the study area were investigated from 2009 to 2020 using the quantitative analysis method of landscape pattern indices， revealing the main problems of cultivated land utilization， and the multi-dimensional management ideas of cultivated land protection in karst areas under the background of rural revitalization strategy were proposed.

Key words Cultivated land protection; Temporal and spatial change; Cultivated land; Landscape pattern index; Karst area; Southwest Guangxi

Received： March 13， 2022  Accepted： May 15， 2022

Qiuyue YIN （1981-）， female， P. R. China， senior engineer， devoted to research about cultivated land protection， natural resource survey and evaluation， and geographic information techniques.

*Corresponding author.

Overview of the Study Area and Data Sources

Overview of the study area

The karst areas distributed in Guangxi Zhuang Autonomous Region （"Gui" for short） are relatively large， accounting for about 35.1% of the total land area in Guangxi， and rocky desertification land accounts for 8.14% of the total land area in Guangxi[1]. The karst areas are mainly distributed in southwestern， central and northeastern Guangxi， among which the distribution in southwestern Guangxi is more concentrated. The four counties （cities） of Debao County， Jingxi City， Daxin County and Tiandeng County in the southwest of Guangxi are one of the most typical and concentrated Fenglin karst distribution areas in China and even in the world. The unique topography directly affects the spatial layout of cultivated land.

Debao County and Jingxi City are subordinate to Baise City， Guangxi， and Daxin County and Tiandeng County are subordinate to Chongzuo City， Guangxi. Among them， Jingxi City and Daxin County are adjacent to Vietnam. The four administrative regions are located between 105°56′-107°29′ east longitude and 22°29′-23°40′ north latitude. The annual average temperature is 20.2 ℃; the annual average rainfall is 1 411.6 mm; the annual frost-free period is 342 d; and the annual sunshine hours are 1 522 h. The four administrative regions have a subtropical monsoon climate， which is very suitable for the growth of tropical and subtropical crops. In terms of topography and geomorphology， the topography of Debao County and Jingxi City is high in the northwest and low in the southeast， dominated by dissolved plateau landforms， most of which are karst peak clusters， solitary peaks and non-karst soil mountains[2]， and they are rich in mineral resources. The terrain of Daxin County is high in the northeast and northwest， gradually sloping to the south， and the middle part is slightly lower， forming a grooved rocky mountain basin， with peaks and forests and valleys formed by limestone mountains as the main terrain. Tiandeng County is dominated by low mountains and hills. The mountainous area accounts for about 78% of the total land area of Tiandeng County， and the karst landform accounts for about 77% of the total land area of Tiandeng. The distributed peaks， forests and valleys are mostly fault zones with high water permeability.

According to the statistics released by the Guangxi Statistical Yearbook in 2009 and 2021， the social and economic indices of the four counties （cities） in southwestern Guangxi， such as the GDP， the added value of the primary industry， and the per capita disposable income of rural residents， are increasing year by year， while the indices such as resident population and total sown area of crops are on a downward trend.

By obtaining the land use data of 4 counties （cities） in Debao County， Jingxi City， Daxin County and Tiandeng County in 2020， we could see that the proportion of cultivated land in the 4 counties （cities） in southwestern Guangxi to the total land area in 2020 was 19%. Specifically， paddy fields accounted for 36% of the total cultivated area， and dry land accounted for 64% of the total cultivated area. The terrain slope of cultivated land is mostly less than 15°， and the thickness of effective soil layer of cultivated land is generally between 60-100 cm. The soil is mainly loam and clay. Compared with the scale of arable land in 2009， it could be seen that the reduction of arable land exceeded 24 000 hm2. After the implementation of the "drought to water" upgrading project， the proportion of paddy fields in the four counties （cities） increased from 34% in 2009 to 36%， and the protection of high-quality cultivated land was effective.

The comprehensive analysis showed that the 4 counties （cities） in the karst areas of southwestern Guangxi have uneven distribution of rainwater overall， and face obvious shortage of engineering water， and the potential hazards of rocky desertification such as soil erosion and natural disasters have a great impact on agricultural production. The development of industrialization and urbanization has improved the local economic level， and the phenomenon of arable land being turned to non-agricultural use is prominent. The resident population is seriously lost to the outside world; and the total sown area of crops is generally decreasing except for Daxin County， and the phenomenon of "non-grain" of cultivated land is common， and the per capita disposable income of rural residents has increased significantly. The reduction of arable land is relatively large， and the protection situation of the quantity of arable land is severe， but the area of arable land per capita does not decrease but increases.

Data sources

This study took 2009， 2013， 2015， 2018 and 2020 as the research time nodes， and 2009-2020 as the research period. The basic data used included vector data of land use status in the study area， satellite remote sensing image data， socio-economic statistical data， rocky desertification land data， etc. Among them， the vector data of the study area and the satellite remote sensing image data came from the annual survey data of land use in the study area and various thematic digital orthophotos; the socio-economic statistical data indices came from the "Guangxi Statistical Yearbook" （2009-2021）; and the rocky desertification land data came from the third rocky desertification monitoring data， and the auxiliary data included the boundary vector data of the administrative districts at all levels in the study area.

Research Methods

Data preprocessing

The acquisition of current land use data

The thematic digital remote sensing images of Debao County， Jingxi City， Daxin County and Tiandeng County in 2009， 2013， 2015， 2018 and 2020 in the karst areas of Southwest Guangxi were integrated and connected by year， and the image data of various years were interpreted， analyzed and processed through the statistical analysis module and spatial analysis module of the ArcGIS software platform， and connected and compared with the current land use vector data. The land use remote sensing monitoring technology was used to obtain information on land use status and its dynamic changes. In order to better connect with the land use data of various years， the land patches from 2009 to 2018 were marked to various land types such as cultivated land， garden land， forest land， construction land and unused land according to Land Use Status Classification （GB/T 21010-2007） used by the second land survey， and the land patches of 2020 were marked according to Technical regulation of the third nationwide land survey （TD/T 1055-2019） used by the third national land survey work， thereby forming a map of the current situation of land use.

The rate of change in cultivated land scale

The overall change trend of the regional cultivated land scale from 2009 to 2020 could be further analyzed by studying the change rate of the cultivated land scale in each year. The following was the mathematical expression of the rate of change in cultivated land scale[3]：

W=（S2-S1）/S1×100%（1）

Where W is the change rate of cultivated land scale in the study area; S2 is the cultivated land area of the study area at the beginning of the study; and S1 is the cultivated land area of the study area at the end of the study.

The dynamic degree of cultivated land change

Focusing on the analysis of cultivated land， from 2009 to 2020， the amount of cultivated land resources in the four counties （cities） showed a downward trend year by year. In order to obtain the changing situation of cultivated land resources more intuitively， the dynamic degree of cultivated land change was proposed for the expression of the change speed of the quantity and spatial distribution of cultivated land resources in each region within a certain period of time， so as to quantitatively analyze the change speed. The dynamic degree of cultivated land change is an important index to reflect the intensity of regional cultivated land change[4]. It is used to indicate the stability of the quantity and distribution of cultivated land resources in a period， which is equivalent to the annual change rate of cultivated land in a region. The following was mathematical expression of the dynamic degree of cultivated land change （K）[3]：

K=Ub-UaUa×1T×100%（2）

In the formula， K is the dynamic degree of cultivated land change in a certain time period， and the unit is %/a; Ua refers to the initial cultivated land area， and Ub refers to the final cultivated land area; and T is the length of the study period， in years. When K is 0， it means that the cultivated land area is generally stable and has not changed; K<0 indicates that the cultivated land area is decreasing， and the smaller the value， the greater the reduction; when K>0， it means that the cultivated land area has increased， and the larger the value range， the larger the increase scale; and the larger the absolute value of K， the more obvious the change of cultivated land area.

Selection of landscape pattern indices

Landscape pattern indices refer to simple quantitative indices that can highly condense landscape pattern information and reflect some aspects of its structural composition and spatial configuration[5]. Fragstats 4 software was used to calculate the landscape pattern indices in this study， and the following main indices were selected for calculation and analysis： patch density （PD）， largest patch index （LPI）， Shannons diversity index （SHDI）， aggregation index （AI）， number of patches （NP）， and landscape shape index （LSI）[6].

Extraction and calculation of land use information

Overall change of cultivated land

According to formulas ① and ②， the changes of cultivated land in four counties （cities） of Debao County， Jingxi City， Daxin County and Tiandeng County in the karst areas of Southwest Guangxi from 2009 to 2020 were calculated （Table 1， Fig. 1）.

Calculation of land use information landscape indices

Fragstats 4 software was used to first calculate the overall pattern index of land use landscape in each study period from 2009 to 2020， and analyze its change trend （Fig. 2）， and then calculate the landscape pattern indices of various land use types in each study period （Table 2， taking cultivated land as an example）.

Results and Analysis

Analysis of changes in the quantity of cultivated land from 2009 to 2020

The overall scale of arable land showed a decreasing trend year by year， with large decreases

In the four time periods from 2009 to 2020， the scale of cultivated land in the four counties （cities） showed an overall decreasing trend， with an average annual net decrease of 300-800 hm2 and an average annual decrease rate of 1.03%. The decrease in the scale of cultivated land increased significantly from 2018 to 2020 （Table 1）， and the decreases in Jingxi City and Debao County were more significant. However， due to serious population loss， the per capita arable land area of these four counties （cities） increased from 0.12 hm2/person in 2009 to 0.15 hm2/person in 2020， showing a rare upward trend. Overlay analysis was performed using ArcGIS software on the land use status maps in 2009 and 2020， to obtain information on the increase and decrease of cultivated land during the eleven-year period. It could be seen that most of decrease in cultivated land from 2009 to 2020 went to forest land， garden land and construction land; and the main sources of newly increased cultivated land were the reclamation of wasteland in unused land， the reclamation of damaged forest land， and the adjustment of agricultural structure due to different technical methods of land type interpretation and identification， and in general， the rate of decrease of cultivated land was higher than that of the increase of cultivated land. After checking the arable land reserve tasks of the overall land use plan （2006-2020） in the 4 counties （cities）， it could be seen that the scale of arable land in these 4 counties （cities） had exceeded the red line of arable land， and the situation of arable land protection is relatively severe.

The outward expansion trend of construction land was obvious， and the rapid development of urbanization accelerated the "non-agriculture" of cultivated land

From 2009 to 2020， the total construction land increased by 11 400 hm2， with an average annual growth of about 1 000 hm2 and an average annual growth rate of 2.74%. In the construction land， transportation land had the largest growth rate， and the increase rate was more prominent during 2018-2020， which was mainly because that the technical requirements of the land use status survey changed. From 2009 to 2018， the technical specifications of the second land survey were used， and most of the land for transportation was marked with linear ground features.  From 2019， the land use status maps were drawn and annotated in accordance with the requirements of the technical regulations of the third national land survey， combining with high-precision images， on-the-spot proof interpretation， and identification of patch land type， and linear features such as roads and water systems were drawn in the form of patches， which resulted in the increase of the scale of transportation land. The expansion of construction land such as township-oriented villages has always been one of the main reasons for the reduction of cultivated land. Karst areas are limited by topography and landforms， and the gentle areas around township-oriented villages are basically high-quality cultivated land. The expansion of township-oriented villages and infrastructure construction will inevitably occupy a large amount of high-quality cultivated land， and the potential of reserve resources of cultivated land in these areas is limited， making it difficult for major construction projects in these areas to achieve a balance of cultivated land occupation and compensation. There are also common phenomena such as the indiscriminate occupation of cultivated land due to house building in rural areas， which has also become a "big problem" that plagues the development of southwestern Guangxi. The demand for construction land is increasing year by year， and the phenomenon of illegal occupation of cultivated land has become more and more prominent， further accelerating the trend of "non-agriculture" of cultivated land.

The scale of forests and gardens has increased substantially， and the "non-grain" phenomenon of cultivated land is more prominent

Under the background of the rural revitalization strategy， driven by market demand and economic interests， various localities have vigorously promoted the development of agricultural product-related industries such as characteristic fruits and vegetables. In addition， the high cost and low income of planting food crops have made the cultivation of cash crops gradually become the leading industry in many rural areas. Based on the ArcGIS statistical analysis module and spatial analysis module， this study focused on the comparison and analysis of the cultivated land， forest land， garden land and other patches in 2009 and 2020. The overall scale of cultivated land decreased by 24 200 hm2; the overall scale of forest land increased by 326 100 hm2; and the overall scale of garden land increased by 7 400 hm2. To promote rural revitalization， while maintaining the bottom line of food security and the red line of ecological protection， it is also necessary to make comprehensive use of local background resources to properly develop characteristic agricultural industries. The four counties in southwest Guangxi all have a subtropical monsoon climate， which is very suitable for the development of the forest and fruit industry. However， due to the limitation of karst topography， cultivated land is generally distributed from the front of houses to the foot of mountains， and the development of the forest and fruit industry cannot go up the mountains. According to our field research， the "non-grain" of cultivated land in these areas mainly includes the cultivation of non-grain crops or the rotation of grain and non-grain crops， the cultivation of economic forests and fruits， the cultivation of fast-growing forests， and the digging of ponds for aquaculture. The planting structure of agricultural land is adjusted frequently， and the trend of "non-grain" of cultivated land is obvious. It is necessary to seek appropriate measures to coordinate and handle the relationship between "non-grain" of cultivated land and farmers income increase.

The distribution of arable land resources is uneven， and there is less high-quality arable land

Based on the 2020 data of cultivated land in the 4 counties （cities）， the proportion of paddy fields in the secondary land structure of cultivated land was relatively small， only 36%. The cultivated land in the four counties （cities） was mainly dry land， and the quality of paddy fields was generally higher than that of dry land. According to the survey and evaluation results of the quality of cultivated land in Guangxi， these four counties （cities） generally have less high-quality cultivated land resources. The expansion of construction land in cities and towns is gradually encroaching on the surrounding high-quality arable land， further exacerbating the sharp decline in high-quality arable land resources.

The ecological security and environmental problems of cultivated land cannot be ignored

The rocky desertification trend in the karst landform area in southwest Guangxi is obvious. The rocky desertification land area of the 4 counties （cities） was 205 300 hm2， accounting for 19% of the total land area. Due to the scarcity of forest vegetation and low disaster resistance in these rocky desertification areas and the thickness of soil layers in karst areas mostly less than 100 cm， they are easily eroded by rainwater and surface water， resulting in serious soil erosion and frequent geological disasters， while soil erosion further accelerates the rocky desertification of the land， forming a vicious circle， which has a negative impact on the ecological security of regional farmland.

Analysis of fragmentation degree of cultivated land

Overall pattern of land use landscape

Based on the current land use maps and land type attributes of the four counties （cities） in the karst area of Southwest Guangxi in 2009， 2013， 2015， 2018， and 2020， the four landscape pattern indices in each year were calculated by software （Fig. 2）. The total land area of the four counties （cities） in southwestern Guangxi was about 1.08 million hm2. From 2009 to 2020， the index of patch density increased from 16.92 to 31.58， and the largest patch index decreased from 8.36 to 7.81 from 2009 to 2018， but greatly increased to 45.56 in 2020. The Shannons diversity index decreased from 1.26 to 0.94， and the aggregation index decreased from 88.03 to 87.57. It could be seen that in the four counties （cities） in southwestern Guangxi， the fragmentation of each land use type was relatively large and increased year by year， and the land use landscape type was diversified; and the shape of land was more complex， and the degree of concentration and connectivity was decreasing. It shows that the landscape pattern of land use in the four counties （cities） in southwestern Guangxi is generally affected by karst landforms. Limestone mountains in the karst area have a large proportion， and due to the steep terrain and weak soil layer， the land use space is greatly limited. In the context of the current rural revitalization strategy， to effectively improve the local economic level， limited land resources and frequent human activities will face greater conflicts.

Landscape pattern analysis of land use types

Fragstats 4 software was used to calculate the four landscape pattern indices of cultivated land， garden land， forest land， construction land， and unused land in each year （taking cultivated land as an example， as shown in Table 2）. The results obtained from comparative analysis were given below.

① Number of patches： In 2009， cultivated land showed the largest patch number， accounting for 39.30% of the total patch number that year， and forest land and unused land ranked second， both accounting for about 20%. From 2009 to 2018， the patch number indices of cultivated land and other land use types generally increased year by year， and by 2020， the patch number indices of cultivated land， garden land and construction land increased significantly， while those of forest land and unused land decreased more. It shows that the fragmentation degree of cultivated land increased during 2009-2020， and the fragmentation degree of forest land decreased most obviously.

② Patch density index： From 2009 to 2018， the patch density index range of each land use type generally showed a gradual growth trend. In 2020， the index value ranges of cultivated land， garden land and construction land increased significantly， while those of forest land and unused land decreased， indicating that the fragmentation degree of cultivated land was increasing year by year.

③ Largest patch index： During the period from 2009 to 2020， the largest patch indices of cultivated land， garden land and unused land gradually decreased， while that of construction land increased slightly， and that of forest land increased significantly， indicating that the fragmentation degree of cultivated land was increasing， while forest land showed a significant trend of increasing concentricity.

④ Landscape shape index： Cultivated land had the largest landscape shape index among all land use types and was increasing year by year. The indices of other land types were basically below 200. The indices of garden land， forest land， construction land and unused land were relatively balanced， with up and down changes. It showed that the shape of cultivated land fields changed frequently and was in a state of irregular change.

Combining the changes of the four landscape indices， it could be seen that ① from 2009 to 2020， the fragmentation degree of cultivated land in the four counties （cities） in southwestern Guangxi generally increased year by year， and the fragmentation degrees of garden land， construction land and unused land also increased， while the fragmentation degree of forest land decreased significantly. ② The analysis of the flow direction of various types showed that the reduced cultivated land mainly flowed to garden land， forest land and construction land. The newly added cultivated land was mainly due to the implementation of a series of land reclamation for unused land. Through the spatial analysis function of ArcGIS， it was found that the scale of the reduced cultivated land was relatively large， but the spatial distribution of its patches was relatively scattered. Meanwhile， affected by the karst topography， the patches of the newly added cultivated land were generally small in size and scattered in locations. These changes made the fragmentation degree of cultivated land increase year by year. ③ The flow of cultivated land to other land types also indirectly affected the fragmentation degrees of garden land and construction land， which was mainly because the newly added patches were relatively scattered. ④ The main reason for the obvious reduction of forest fragmentation was that the phenomenon of "non-grain" of cultivated land planted with fast-growing trees such as Eucalyptus robusta Smith was increasing， and when drawing the annual land use status maps， multiple forest patches with the same attributes and connected locations would be merged into large patches through the comprehensive mapping technology， which would significantly reduce the fragmentation degree of forest land.

Exploration of Multi-dimensional Management Ideas for Cultivated Land Protection in Karst Areas

Strictly implement Chinas control policy on the phenomenon of "non-agriculture" and "non-grain" of cultivated land

Through the establishment of the responsibility system of cultivated land for provincial， municipal and county administrative leaders on curbing "non-agriculture" and "non-grain"， the basic quantity of "non-agriculture" and "non-grain" of arable land and permanent basic farmland should be made clear combining with the third national land survey and effective measures should be taken to gradually solve the stock problem. Meanwhile， we should take the preparation of national land space planning as an opportunity to deepen the control of land use， highlight the life community of mountains， rivers， forests， fields， lakes and grasses， and draw three control lines. Guided by planning， we will optimize the layout of cultivated land， explore the construction of cultivated land protection and agglomeration areas， strictly control the occupation of cultivated land for various constructions， strictly maintain the existing spatial layout of long-term stable utilization of cultivated land， and clarify the importance of cultivated land protection， comprehensive national land improvement and ecological restoration， so as to ensure the accurately implementation of the responsibility for arable land protection. The enthusiasm of farmers and growers to plant grains can be further increased by increasing the subsidies for grain cultivation through the establishment of a financial incentive mechanism and a compensation mechanism for the interests of main grain-producing areas[7]， and increasing grain yield combining with measures such as seed improvement. It can also indirectly slow down the trend of rural labor transfer and employment. We should strictly implement farmland occupation and compensation balance， farmland "in and out balance" and other farmland use control systems， and establish and improve the multi-level linkage farmland "field chief system" protection grid supervision mechanism and the farmland protection reward and punishment mechanism， so as to keep the bottom line of cultivated land through multiple measures and realize the comprehensive "three-in-one" protection of quantity， quality and ecology of cultivated land.

Diversified exploration of cultivated land resource protection methods

In order to effectively improve the current severe situation of arable land protection， it is necessary for the upper-level decision-making groups to conduct in-depth research in different regions to understand the actual situation in rural areas across the country under the current national conditions， so as to analyze the root causes of various phenomena， coordinate the balance between the urgent needs of rural economic development in karst areas and the task of farmland protection and explore more feasible and operational incentive mechanisms for farmland protection. Local daily farmland protection cannot be managed and controlled by the government alone， and it requires both administrative and market means to improve the economic benefits of farmland operators through different methods and to stimulate farmers and other business entities to spontaneously protect farmland.

Improve the ecological protection system of cultivated land

In the past， arable land protection policies focused on resource security， ignoring the issue of ecological security， especially in karst rocky desertification areas， where human activities have brought more and more obvious negative impacts on the security of arable land resources， and the arable land ecosystem has also been damaged to some extent. It is necessary to improve the ecological protection mechanism of cultivated land and further enhance the ecological security value of cultivated land. From the perspective of the cultivated land ecosystem， the sustainable and intensive utilization mode of cultivated land resources can be studied， and the biodiversity and soil health of cultivated land can be improved by multiple means of promoting agriculture through science and technology， so as to promote the ecological protection and restoration of cultivated land. We can make full use of the third national soil census work platform to master the most detailed soil resource information and provide a reference for the targeted implementation of soil fertilization and the rational distribution of characteristic agricultural production in the future. We should continue in-depth exploration of the operation mechanism of farmland ecological protection compensation and accelerate the implementation of the system.

Strengthen the publicity of policies related to land management

Land management publicity is not limited to the annual "Land Day"， and localities can effectively expand the depth and coverage of publicity through the combination of online and offline， publicity and guidance. The focus of publicity should be deeply integrated with the hotspots of land use that the masses care about and are concerned about， including special planning， real estate registration， land use approval process， rural housing construction， facility agricultural land filing， land transfer， rural collective construction land， etc. We can explain and publicize relevant legal knowledge such as the importance of sticking to the red line of cultivated land， the protection of permanent basic farmland and the penalty clauses for destroying cultivated land.

References

[1] HE F， LIU C， XIAO ZQ. The exploration of farming and conservation in the karst rocky desertification mountain area from the perspective of ecological civilization： A case study on Guangxi[J]. China Land， 2020（3）： 52-54. （in Chinese）.

[2] LIN Y. The comparative analysis of land use change and landscape pattern in China-Vietnam border[D]. Guilin： Guangxi Normal University， 2017. （in Chinese）.

[3] ZHANG H， MA L. Analysis of temporal and spatial evolution and driving forces of cultivated land in Cele Oasis from 1972 to 2018[J]. Journal of Heibei University of Environmental Engineering， 2021， 31（6）： 37-41. （in Chinese）.

[4] LIU CY， ZHU KW， LIU JP. Evolution and prediction of land cover and biodiversity function in Chongqing section of Three Gorges Reservoir Area[J]. Transactions of the Chinese Society of Agricultural Engineering， 2017， 33（19）： 258-267. （in Chinese）.

[5] SHEN M. Research on the change of urban forest landscape pattern in Nanjing[D]. Nanjing： Nanjing Agricultural University， 2008. （in Chinese）.

[6] LU JF， AN YL. Analysis on landscape pattern of land use in Qianxi karst area of Guizhou[J]. Guizhou Agricultural Sciences， 2013， 41（5）： 173-175. （in Chinese）.

[7] XU N， WANG XJ. A research review of benefit compensation in Major food-producing areas [J]. Journal of Anhui University of Technology： Social Sciences Edition， 2021（1）： 12-14. （in Chinese）.