Land Use Challenges in Emerging Economic Corridors of the Global South: A Case Study of the Laos Economic Corridor

Abstract

Economic corridors play a crucial role in promoting economic growth and facilitating coordinated regional development. However, land use changes associated with the development of emerging economic corridors have become a prominent source of conflict in regional integration in the Global South. This study takes the Laos Economic Corridor as a case study to explore the characteristics and driving mechanisms of land use changes in emerging economic corridor regions. Using global land cover data from 2000 to 2020 (GlobeLand30) and employing spatial statistical analysis, the Random Forest (RFC) algorithm, and the CA-Markov model, the study follows a Pattern–Process–Mechanism–Trend analytical framework to reveal the spatial distribution characteristics and transformation paths of land use within the corridor. The study results indicate that (1) The land use pattern in the Laos Economic Corridor has gradually shifted from a “single-core radial” structure to a “dumbbell-shaped” structure, promoting coordinated regional economic development. (2) A significant unidirectional flow of land use has been established, with forestland being converted into cultivated land and cultivated land being further converted into artificial surfaces. (3) In addition to the natural geographical constraints, the transport infrastructure and the spatial layout of industries are the main drivers for the expansion of ecological land, agricultural land, and built-up land. (4) Spatial planning interventions are essential and urgent: the establishment of land management rules based on the principles of forest conservation and intensive development can effectively control the uncontrolled expansion of artificial areas, significantly reduce the loss of forestland, and ensure the rational allocation of land resources for long-term development. The findings of this study offer valuable insights and reference points for the Global South, enhancing understanding of the spatial development dynamics of economic corridors, informing the optimization of land-use policies, and supporting efforts to promote regional integration and sustainable development.

1. Introduction

The development of economic corridors is regarded as an essential means of promoting regional economic integration and industrial clustering. The principal objective is to facilitate accelerated regional economic growth and development through the enhancement of infrastructure, the streamlining of cross-border trade and investment, and the optimization of the regional flow of resources, finance, and labor [1]. The functions and objectives of developing economic corridors vary depending on different regional development goals. Brunner’s study classified global economic corridors into three categories: (i) fulfilling transit and value-added functions by bridging trade gaps between cross-border regions through infrastructure (B + T); (ii) establishing dense economic networks at technology hubs (High-C); (iii) connecting remote areas to economic centers via corridors to access markets and technologies (C + C) [2]. The Global South is currently experiencing a period of unprecedented economic growth and rapid urbanization, driven by the emergence of new economic corridors focused on infrastructure projects such as transportation routes and energy pipelines (B + T). These corridors are gradually becoming key engines of economic development in these regions. For instance, the Pan-Asian Railway through Southeast Asia, the Mombasa-Nairobi Railway in Kenya, and the “Lobito Corridor” in Southern Africa not only serve as key drivers of economic development and regional integration but also carry the dual objectives of poverty alleviation and promoting sustainable development [3].

However, infrastructure-based economic corridors are highly influenced by land factors due to their immense demand for land, the scarcity of land resources, and the multifaceted economic, social, and ecological functions of land [4]. The Global South faces several land-related conflicts during economic corridor development, such as land grabs [5,6], deforestation [7], agricultural boundary damage [8], and unresolved land claims threatening social equity and local economic development [9]. Therefore, although economic corridors offer the Global South countries great opportunities for economic growth and regional integration, they also present severe land-use challenges.

The concept of “corridor” in spatial planning has a long history. In 1949, Taylor mentioned the idea of economic corridors [10], and Whebell compared different types of economic corridors and their economic patterns, identifying five historical development stages: the initial phase, the commercialization of agriculture, the railway transportation stage, and the highway transportation and urbanization stage. In each stage, transportation played a crucial role [11]. As economic integration has advanced, the modern “corridor” has evolved into a multi-scale concept encompassing local, regional, and transnational scales [12,13], addressing multiple dimensions, such as transportation, space, institutions, and economies [14,15].

The European Union was an early adopter of regional economic corridors with the launch of the Trans-European Networks in Transport (TEN-T) program in 1992. This initiative links highly urbanized and densely populated areas to form a relatively homogeneous economic zone [16,17]. In contrast, economic corridors in Asia and Africa have emerged more recently but have experienced rapid growth [18,19,20]. The concept of economic corridors is increasingly regarded as a significant phenomenon within the Global South, with these regions identified as possessing considerable potential for future growth and development [21,22,23]. The development of these corridors has resulted in accelerated urbanization and enhanced regional economic integration, although this has also led to an increase in environmental pressure [24,25,26]. Linthicum formulated a risk analysis model for infrastructure corridors affected by neighboring land development. This model emphasizes that planners must anticipate and actively manage land development along the corridor to avoid unintended consequences, regrets, and delays [27]. This cautionary note is especially pertinent for countries in the Global South, which are experiencing a rapid acceleration in the construction of economic corridors and a concomitant transition towards urbanization. It is of paramount importance that this issue be treated with the seriousness it deserves.

As a major regional development initiative, the development and construction of economic corridors are closely intertwined with land use. Numerous studies based on transport geography have discussed the relationship between corridors and urbanization, as well as the impact of transport corridor planning on urban land use [28]. One aspect of the research concerns the examination of the expansion of construction land within buffer zones, with a particular focus on infrastructure projects associated with economic corridors and the resulting increase in land economic benefits. Research studies have evidenced that the development of economic corridors typically gives rise to an intensification of land use along the corridor, particularly in areas of strategic importance, such as transportation hubs, industrial parks, and logistics centers [29,30]. Economic corridors enhance accessibility along their routes, prompting a shift in land-use patterns from decentralized to concentrated, thus improving land-use efficiency [31,32,33]. Furthermore, economic corridor development facilitates infrastructure and public service upgrades in surrounding areas, creating an environment more suitable for residential and commercial development. This results in a diversified distribution of land use, where commercial, residential, industrial, and public facility lands are intermingled [34,35]. Such diversification contributes to accelerated regional economic growth [36,37]. Another category of study concentrates on the ecological landscape pattern of the economic corridor region through the utilization of regional land cover analyses, which permit the assessment of the loss of land ecological benefits. The study observes that engineering activities in economic corridors will result in substantial alterations to landscape patterns. Transport engineering activities will contribute to infrastructure development, population pressure, and increased livestock rearing, which will, in turn, precipitate shifts in land use and lead to land fragmentation and degradation [38,39]. The establishment of ecological corridors and the securing of natural resources have been identified as key drivers of development and conservation efforts in many regions around the world. However, these activities have also contributed to peatland loss and exponential deforestation [9], raising concerns about the environmental consequences of this approach and the potential for fair economic development to be undermined.

It is evident that the construction of economic corridors has significant impacts across various countries and regions, with land-use patterns in corridor areas evolving alongside corridor projects’ implementation [40,41,42]. These changes eventually manifest as either enablers or obstacles to the strategic implementation of economic corridors. Therefore, exploring land-use changes and their driving mechanisms serves as a critical starting point for optimizing economic corridor strategies from a spatial planning perspective. With advancements in remote sensing (RS) and geographic information systems (GIS), research on land use and its driving mechanisms has grown rapidly. Studies have utilized multi-temporal and multi-scale land cover data to reveal spatial and temporal changes in land-use types [43,44], focusing on the influence of natural factors (e.g., topography and climate) and socio-economic factors (e.g., population, economy, and policies) [45,46,47]. Research methods now include statistical models like regression and principal component analysis [48,49] and simulation models like CA-Markov and CLUE-S [50], providing robust tools for analyzing trends and making predictions about land-use changes.

A review of existing studies highlights three key limitations. Firstly, research on economic corridors is abundant in early-planning, industrially advanced northern nations. In contrast, studies on rapidly developing economic corridors in the Global South remain scarce. Furthermore, findings from northern industrial economies often lack direct applicability to emerging economic corridors in Global South countries, emphasizing the need for context-specific research. Secondly, current studies often adopt a single-dimensional focus—either ecological or economic—while economic corridor strategies involve multidimensional impacts (social, economic, and ecological). This has led to a lack of comprehensive evaluations of land-use changes and their broader implications. Thirdly, research predominantly examines quantitative land-use changes, with insufficient focus on the transitional processes between land types. Consequently, systematic pathways leading to land conflicts are not well-identified. Additionally, there is a lack of forward-looking analyses of spatial planning and land-supply coordination policies, limiting effective decision-making for conflict mitigation and optimization.

Comprehensive evaluations of land-use changes and driving factors in emerging economic corridors in Global South countries can fill these research gaps. By deeply analyzing the processes and drivers of land-type transitions, systematic analyses can overcome the limitations of prior single-dimensional discussions. Such studies will better inform the development of sustainable spatial planning strategies and policies for optimizing land use in rapidly evolving regions. This study aims to address the following questions: (1) Under rapid development, what changes have occurred in the spatial patterns of land use within economic corridors in the Global South, and where are the primary land-use conflicts? (2) How can we understand the process of land use type conversions and their driving forces within economic corridors in the Global South based on spatial characteristics? (3) How can spatial interventions help balance economic development with land resource conservation?

This paper constructs a “Pattern-Process-Mechanism-Trend” analytical framework. Firstly, we use the administrative regions traversed by the economic corridor as the basis for analyzing land use changes, which provides a policy-relevant analytical framework compared to traditional buffer zone analysis. In terms of research methods, we incorporate transition intensity analysis into conventional spatial statistical analysis, aiming to combine static “regional pattern” analysis with dynamic “conversion process” analysis to systematically reveal the spatial distribution characteristics of land use and its transformation pathways. Furthermore, we employ the Random Forest (RFC) algorithm for evolutionary mechanism analysis and the CA-Markov model to simulate multi-scenario spatial intervention strategies, which more intuitively illustrates the balance between economic development and land resource conservation. This approach aids in understanding the current characteristics and causes of land use change and provides a scientific basis for developing land use planning and land management policies that align with the efficient development of economic corridors in the Global South.

2. Materials and Methods

2.1. Study Area

As a rapidly growing representative among Global South countries, Laos has seen land become a critical asset in regional development following iterations of land policies such as resettlement, land use planning, and land commercialization. With the Greater Mekong Sub-region (GMS) expanding the North–South Economic Corridor (NSEC) to include the route ”Boten–Oudomxay–Luang Prabang–Vang Vieng–Vientiane–Nong Khai–Udon Thani–Nakhon Ratchasima–Laem Chabang” for the first time, Vientiane was integrated into the GMS corridor network, recognizing its crucial role at the sub-regional center [51]. The gradual construction of the corridor has driven rapid socio-economic and environmental transformations in northern Laos [52,53]. As an emerging development corridor, its land use challenges not only reflect common issues found in the development of economic corridors but also highlight the unique environmental and socio-economic difficulties faced by Global South in their rapid growth.

This study employs the Laos part of the Greater Mekong Sub-region (GMS) North–South Economic Corridor (NSEC) as a case study, including the administrative areas of the Luang Namtha Province, Oudomxay Province, Luang Prabang Province, Vientiane Province, and Vientiane Capital, geographically located between 17°46′ N and 21°36′ N and 100°35′ E and 103°23′ E (Figure 1). Among these, the Vientiane Province experienced administrative changes with the dissolution of the Xaisomboun Saysettha Special Zone and its conversion into a province. This study adopts the 2015 adjusted administrative boundaries as the study area’s border.

2.2. Dataset

The data used in this study consisted of land cover maps, natural resource datasets, and socio-economic datasets (

Table 1. Data types and sources.

Data Type	Data Name	Source
Land cover maps	LUCC2000/2010/2020	GlobeLand30 (http://globeland30.org/, accessed on 5 June 2024)
Land use plan	Land Use Master Plan2030	Lao PDR Geographic Information Platform (https://apps.k4d.la/explorer/, accessed on 10 June 2024) Open Street Map (http://www.openstreetmap.org/, accessed on 10 June 2024)
Spatial restraint data	Lakes
	Rivers
	National Protection Forest Area
Socio-economic dataset	Administrative boundary
	Population density
	Highway
	Railway
	Railway station
	National road
	Rural road
	Provincial road
	Residential road
	SEZs
	Electricity
	Permanent market
	Community office
	Religious site
Natural resource dataset	Soil type
	Average temperature
	Average precipitation
	DEM	USGS (https://www.usgs.gov/, accessed on 10 June 2024)
	Slope

).

2.2.1. Land Cover Maps

The GlobeLand30 dataset was created under the “Global Land Cover Mapping at Finer Resolution” project led by the National Geomatics Center of China (NGCC). The Globland30 products have been shown to have good overall average accuracy and kappa coefficients of 82% and 0.80, respectively [54,55], providing effective support for the study. Considering the data availability and comparability, land cover data from 2000, 2010, and 2020 were used in this study (with a map resolution of 30 m). The land cover types included in the study area were coded as cultivated land (10), forest (20), grassland (30), shrubland (40), wetland (50), water bodies (60), artificial surfaces (80), and bare land (90).

2.2.2. Natural Resource Dataset

A digital elevation model (resolution 90 m) generated from spatial distribution data from SRTM (Shuttle Radar Topography Mission), published by the United States Geological Survey (USGS), was used. The average temperature and precipitation data were obtained from WorldClim.v2 [56]. Spatial data on soil types, lakes, and rivers were obtained from the Geographic Information Platform provided by the Lao National Geographic Department.

2.2.3. Socio-Economic Dataset

The Lao PDR administrative boundary data, population density data, major roads, POI data, and Land Use Master Plan 2030 (Laos) data were obtained from geographic information platforms. In this study, raw socio-economic data, consisting of point and line data, were processed using kernel density estimation, shortest path analysis, and inverse distance-weighted (IDW) interpolation. These methods transformed the discrete raw data into continuous spatial data, enabling spatial matching and correlation analysis with land use changes.

2.3. Methods

The research framework mainly includes four steps (Figure 2): (1) Data organization: Based on high-quality time-series LUCC maps from Globelland30, a multi-source spatial dataset was created by performing kernel density estimations, shortest paths, and inverse distance weighting interpolation calculations based on the raw data. (2) Spatial statistical analysis: Using LULC map samples, land use dynamics, land use transition matrices, land use change intensity, land use degree, and spatial centroid were calculated to analyze the spatio-temporal evolution characteristics of land use. (3) Driving effect analysis: The RFC algorithm was used to detect LUCC driving factors, assessing the contribution of natural resource data and socio-economic data on land use expansion (cultivated land, forestland, grassland, and artificial surfaces). (4) Based on the results of the analysis of the spatial and temporal evolution characteristics of land use and the driving mechanism, spatial intervention policy conditions are set up, and a multi-scenario simulation prediction is carried out using the CA-Markov model.

2.3.1. Characteristics of the Spatial and Temporal Evolution of LULC

(1)

Land use dynamic degree

Land use dynamics can be expressed as the rate of land use change, which is an indicator used to calculate the amount of change per unit of time for a given land category [57].

$$\mathrm{K}={\displaystyle \frac{u_{t+1}-u_t}{u_t}}\times {\displaystyle \frac{1}{T}}\times 100\%$$

(1)

K refers to the dynamic degree, u_t is the area of the land use type at the beginning of the monitoring period, u_t+1 is the area where the land use type was converted to the land use type during the monitoring period, and T is the detection period.

(2)

Land use transition matrix

Quantitative and structural change characteristics are important aspects of land use change research, and the more commonly used method of analysis is to calculate a land change transfer matrix based on land use/cover (LULC) maps at two points in time. The land use transition matrix is a result of the quantitative description of system states and state transitions in the system analysis [58]. It can be used to indicate the transformation between different land use types and to reveal the transfer rate between different land use types. The formula is

$${\mathrm{A}}_{\mathrm{i}\mathrm{j}}=\left\{\begin{array}{c}{\mathrm{A}}_{11}\\ {\mathrm{A}}_{21}\\ ⋮\\ A_{n1}\end{array}\begin{array}{c}{\mathrm{A}}_{12}\\ {\mathrm{A}}_{22}\\ ⋮\\ A_{n2}\end{array}\begin{array}{c}\dots \\ \dots \\ ⋮\\ \dots \end{array}\left.\begin{array}{c}{\mathrm{A}}_{1n}\\ {\mathrm{A}}_{2n}\\ ⋮\\ A_{nn}\end{array}\right\}\right.$$

In this formula, A_ij denotes the area of land type i at the beginning of the study period that was converted to land type j at the end of the period; n denotes the land use type; where i = (1, 2, …, n), j = (1, 2, …, n).

(3)

LUCC intensity

Land use change intensity analyses the conversion trends between land classes and their impact on the LULC pattern from both absolute and relative perspectives based on the land use transition matrix [59]. This analysis is divided into two dimensions: the “absolute intensity” of the absolute quantities of conversions between land classes and the “relative intensity”, which reflects the impact of conversion intensity on the land use/land cover structure of the study area. It includes both the “inflow” of a land class being converted to other land classes and the “outflow” from other land classes into a specific land class [60].

Absolute intensity: The absolute inflow intensity of land class i converting to land class n over time interval t is denoted as ${AI}_{in}(i\ne n)$. The average absolute inflow intensity of all other land classes, excluding n, converting into land class n during the same time period is denoted as ${MAI}_n$. Similarly, the absolute outflow intensity of land class mmm converting to land class j at the end of the period is denoted as ${AO}_{mj}$ (m ≠ j), and the average absolute outflow intensity from land class m to all other land classes, excluding m, is denoted as ${MAO}_m$.

$${AI}_{in}={\displaystyle \frac{C_{in}/(u_{t+1}-u_t)}{\sum _{i=1}^I{C}_{in}}}$$

(2)

$${MAI}_n={\displaystyle \frac{\left\{\left[\left(\sum _{i=1}^I{C}_{in}\right)-C_{nn}\right]/\left(I-1\right)\right\}/(u_{t+1}-u_t)}{\sum _{i=1}^I{C}_{in}}}$$

(3)

In this formula, i represents the initial land class code; n represents the inflow land class code; $u_t,u_{t+1}$ refer to the quantity of a particular land use type at the beginning and end of the study period, $C_{in}$ represents the area of land class i converted into land class n during time interval t; $C_{nn}$ represents the area of land class n that remained unchanged during time interval t; I represents the number of land classes at the beginning of the period.

Relative intensity: This mainly reflects the impact of conversion intensity between land classes on the LULC structure of the study area. The relative inflow intensity of land class i converting to land class n over time interval t is denoted as ${RI}_{in}$ (i ≠ n). The average relative inflow intensity of all other land classes, excluding n, converting into land class n during the same time period is denoted as ${MRI}_n$. Similarly, the relative outflow intensity of land class m converting to land class j is denoted as ${RO}_{mj}$ (m ≠ j), and the average relative outflow intensity from land class mmm to all other land classes, excluding m, is denoted as ${MRO}_m$.

$${RI}_{in}={\displaystyle \frac{C_{in}/(u_{t+1}-u_t)}{\sum _{j=1}^J{C}_{ij}}}$$

(4)

$${MRI}_n={\displaystyle \frac{\left[\left(\sum _{i=1}^I{C}_{in}\right)-C_{nn}\right]/(u_{t+1}-u_t)}{\sum _{j=1}^J\left[\left(\sum _{i=1}^I{C}_{ij}\right)-C_{nj}\right]}}$$

(5)

In this formula, $C_{ij}$ represents the area of land class i converted into land class j at the end of the period, $C_{nj}$ represents the area of land class n converted into land class j, and J represents the number of land classes at the end of the period.

By comparing the trends between a land class’s inflow (or outflow) intensity and the average inflow (or outflow) intensity, the tendency or suppression of that land class’s conversion process can be identified. For example, if ${AI}_{in}$ = ${MAI}_n$, it implies that the transfer of land class n from each land class is uniform and less influenced by the land use and cover structure of the study area. If ${AI}_{in}$ > ${MAI}_n$, it suggests that land class n gains more inflow from land class iii than the average level, indicating a tendency for inflow from land class i (red legend). Conversely, if ${AI}_{in}<{MAI}_n$, it means that the inflow from land class i into land class n is below the average level, indicating suppression of inflow from land class i (blue legend). By dividing the map into units through a Confusion Table of land use transitions, an intensity map (Figure 3) is generated, showcasing the stability and systematic characteristics of the land use change process.

(4)

Comprehensive Index of Land Use Degree

The land cover is divided into three categories based on the extent of transformation: ecological, agricultural, and built-up land use levels are assigned hierarchical values (

Table 2. Classification indices of land use degree.

Land Use Degree	Land Cover Type	Index
Ecological land	Forest, grassland, shrubland, wetland, water bodies	1
Agricultural land	Cultivated land	2
Built-up land	Artificial surfaces	3

), and the level of the assigned value intuitively reflects the degree of land use in the region. The comprehensive index of the region’s land use degree (LUD) is expressed as follows:

$${\mathrm{L}\mathrm{U}\mathrm{D}}_i=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{D}}_{\mathrm{i}}\times {\displaystyle \frac{{\mathrm{C}}_{\mathrm{i}}}{\mathrm{C}}}\times 100\%$$

(6)

In this formula, ${\mathrm{D}}_{\mathrm{i}}$ represents the utilization intensity index of category i patches, ${\mathrm{C}}_{\mathrm{i}}$ represents the area of category i patches in the region, and C represents the total area of the region.

The geometric center migration model of land use is capable of revealing spatiotemporal changes in land types and clearly indicating the direction of those changes.

$${\mathrm{X}}_{\mathrm{t}}={\displaystyle \frac{\sum _{i=1}^n{X}_i{A}_{ti}}{\sum _{i=1}^n{A}_{ti}}},{\mathrm{Y}}_{\mathrm{t}}={\displaystyle \frac{\sum _{i=1}^n{Y}_i{A}_{ti}}{\sum _{i=1}^n{A}_{ti}}}$$

(7)

In this formula, X_t and Y_t represent the horizontal and vertical coordinates, respectively, of a land class’s geometric center in year t, and $A_{ti}$ represents the area of category i in year t.

2.3.2. Analysis of Factors Contributing to Land Expansion

The land expansion analysis strategy (LEAS) can simulate land use change at the patch scale [61]. The algorithm obtains the change and inertia probability for each land class by determining the transformation rules between land use types. The Random Forest classification (RFC) algorithm was applied to examine the contribution of each driving factor to the different land class changes. The algorithm can address the issue of multicollinearity among various variables by extracting random samples from the primary dataset and subsequently calculating the likelihood of k land use types occurring in cell i, denoted as $P_{i,k}^d$.

$$P_{i,k}^d(x)={\displaystyle \frac{\sum _{n=1}^{M}I=\left[h_n\left(x\right)=d\right]}{M}}$$

(8)

In this formula, the value of d ranges from 0 to 1. If d is equal to 1, it signifies that other land use types have been transformed into k. On the other hand, when d is equal to 0, it means that the land use types have been transformed into other land use types except k. x is a vector that comprises numerous driving force factors, while the function I is an indicator function of the decision tree set. Furthermore, $h_n\left(x\right)$ represents the prediction type of the n-th decision tree in vector x. Last, M is a decision tree for the overall number of trees.

2.3.3. Land Use Forecast Model

A preliminary experiment was conducted to compare the 2020 LULC (Land Use and Land Cover) simulation results using the CA-Markov, FLUS, and PLUS models in terms of simulation accuracy, area consistency, and spatial location alignment. The CA-Markov model demonstrated superior simulation performance. Consequently, this study employed the CA-Markov model, supported by the IDRISI 17.0 software platform, to predict and simulate future land use changes. The construction process involves the following steps. (1) Data conversion and reclassification; (2) calculation of the transfer area and transfer probability matrix of land use types in the study area from 2010 to 2020; (3) calculation of the probability map for each land use type based on the analysis of the driving factors using the IDRISI software Ensemble Editor, which allowed the creation of a suitability atlas for each land use type by adjusting the constraints for multi-scenario simulation. (4) Construction of the CA filter using a 5 × 5 neighborhood filter to define the neighborhood. (5) Determination of the starting point and number of iterations.

3. Results

3.1. Spatio-Temporal Characteristics of Land Use Evolution

3.1.1. Overall Land Cover Changes

In terms of overall quantity, forests and grassland account for over 93% of the study area, with cultivated land making up 5.4%, while artificial surfaces are concentrated in the capital city of Vientiane and the historic city of Luang Prabang, comprising only 0.069% (Figure 4). Significant land cover changes have taken place in the corridor region during the study period (

Table 3. Land use change in the study areas for 2000–2020 (unit: km²).

Time (Year)	2000	2010	2020	K (The Single Land Use Dynamic Degree)
				2000–2010	2010–2020	2000–2020
Cultivated land	3132.587	3773.854	3800.890	2.047	0.072	1.067
Forest	49,318.946	48,441.016	47,907.157	−0.178	−0.110	−0.143
Grassland	4399.753	4637.147	4824.366	0.540	0.404	0.483
Shrubland	105.645	150.045	150.192	4.203	0.010	2.108
Wetland	8.083	5.733	4.811	−2.908	−1.608	−2.024
Water bodies	673.974	625.226	773.962	−0.723	2.379	0.742
Artificial surfaces	39.844	45.812	214.57	1.498	36.838	21.927
Bare land	0.035	0.035	2.913	0.000	0.000	0.000

). Consistent with regional development goals, the area of artificial surfaces expanded by approximately 5.97 km² from 2000 to 2010, with a marked increase of about 168.7 km² between 2010 and 2020, with the single dynamic degree rising from 1.49 to 36.83, marking it as the most prominent land type change within the corridor. During the same period, forest and wetland areas saw varying degrees of reduction. From 2000 to 2020, the forest area declined by 1411.78 km² and the wetland area by 3.27 km², with single dynamic degrees of −2.02 and −0.14, respectively. Meanwhile, changes in farmland and shrubs slowed down, with their single dynamic degrees shifting from 2.04 to 0.07 and from 4.2 to 0.01.

3.1.2. Spatial Pattern of Land Use

Using the land use intensity assignment rules (Table 2), the land use intensity of the study area was calculated across different time periods. As shown in

Table 4. LUD of areas along the corridor for 2000–2020.

Administrative Region	Classification Index of Land Use Degree
	2000	2010	2020
Vientiane Capital	37.37	38.63	41.17
Vientiane	8.70	9.75	11.47
Oudomxay	1.65	2.01	2.36
Luangprabang	1.22	1.61	1.65
Luangnamtha	3.37	7.12	7.17
Cumulative index	52.32	59.12	63.81
∆LUD	-	11.5%	7.3%

, land use intensity in the economic corridor area rose from 52.32 to 63.81, mirroring the trend of rapid urbanization, with land development intensity substantially increasing; the dynamic degree reached 11.5 in the first decade and 7.3 in the subsequent decade.

By categorizing land use intensity based on functional classification by degree of land transformation, the spatial pattern shifts in the corridor region can be further observed through the movement of spatial centroids for different functions. As shown in Figure 5, the centroid changes in land use intensity from 2000 to 2020 are most evident in agricultural and built-up land, showing a staged characteristic. In the first stage (2000–2010), the centroid of agricultural land moved southeast with a latitude shift of 1.04°, while the centroid of urban artificial surfaces moved northwest with minimal latitude and longitude changes. In the second stage (2010–2020), agricultural land moved back slightly to the northwest, while urban artificial surfaces continued to move northwest with a latitude shift of 0.91°. Furthermore, the centroid of ecological land in the study area remained centrally located over the two decades, attributed to the large and even distribution of ecological land, resulting in negligible shifts in regional ecological spatial patterns. Additionally, as indicated by the standard deviation ellipse of the spatial distribution for these two types of land, the directional axis weakened as the centroid shifted toward the center, indicating a trend of spatial coordination in corridor region development.

3.1.3. Land Use Transition Pathways

Based on LUCC spatial data, the land use transition matrix (

Table 5. LUCC transition matrix of the study areas for 2000–2020 (unit: km²).

2000–2010	Cultivated Land	Forest	Grassland	Shrubland	Wetland	Water Bodies	Artificial Surfaces	Bare Land	Total2000
Cultivated land	2788.064	263.417	66.011	0.481	0.077	7.174	7.366	-	3132.588
Forest	821.761	46,893.545	1546.565	27.840	0.149	28.726	0.361	-	49,318.946
Grassland	136.937	1215.149	2976.462	46.704	0.817	23.265	0.420	-	4399.754
Shrubland	0.368	17.269	13.811	74.081	-	0.115	0.002	-	105.646
Wetland	0.756	1.201	1.097	0.005	1.521	3.505	-	-	8.084
Waterbodies	24.087	50.216	33.148	0.936	3.170	562.246	0.172	-	673.975
Artificial surfaces	1.882	0.221	0.055	-	-	0.195	37.492	-	39.845
Bare land	-	-	-	-	-	-	-	0.035	0.035
Total 2010	3773.855	48,441.0	4637.147	150.045	5.733	625.226	45.813	0.035	57,678.872
2010–2020	Cultivated land	Forest	Grassland	Shrubland	Wetland	Waterbodies	Artificial surfaces	Bare land	Total2010
Cultivated land	3339.460	198.458	91.042	0.383	0.045	27.807	116.659	-	3773.855
Forest	368.350	46,481.315	1394.863	26.321	0.141	131.140	36.220	2.668	48,441.017
Grassland	83.543	1178.429	3304.231	17.086	0.098	36.077	17.458	0.227	4637.147
Shrubland	0.592	24.976	17.658	106.187	-	0.493	0.140	-	150.045
Wetland	0.050	0.074	0.019	-	3.975	1.616	-	-	5.733
Waterbodies	7.153	23.796	15.769	0.216	0.552	576.761	0.961	0.019	625.226
Artificial surfaces	1.743	0.077	0.783	-	-	0.069	43.141	-	45.813
Bare land	-	0.034	0.001	-	-	-	-	-	0.035
Total 2020	3800.891	47,907.1	4824.366	150.192	4.811	773.962	214.578	2.913	57,678.872

) was generated through data processing with ArcGIS, and plots showing the transformation of land use in the corridor area were extracted (Figure 6). In terms of quantity and structure of land use transition, the first phase (2000–2010) was marked by transitions between cultivated land, forest, and grassland, with increases primarily in cultivated land. Forest-to-cultivated-land conversions covered 821.7 km², grassland-to-cultivated-land accounted for 136.9 km², and cultivated-land-to-forest transformations amounted to 263.41 km². In the second phase (2010–2020), although conversions among forest, grassland, and cultivated land continued, the transformation between forest and cultivated land slowed down, with notable increases in artificial surfaces and water bodies toward the end. Forest-to-cultivated-land conversions decreased to 368.3 km², while cultivated-land-to-forest transformations dropped to 198.45 km². Newly added water bodies covered an area of 148.7 km², approximately 131.1 km² of which originated from forest, and newly added artificial surfaces covered 168.76 km², around 116.6 km² of which came from cultivated land.

In terms of land use transition direction and intensity, intensity maps were used to connect transition directions with time phases, reflecting variations in transition intensity among different land types. Figure 7a shows the LUCC intensity map for the study area from 2000 to 2010. Cultivated land exhibited an absolute tendency for conversion to forest, with a relative tendency for conversion to water bodies, while showing systematic inhibition for conversion to shrubs and bare land. Forest-to-cultivated-land transitions were absolutely inclined, with a systematic tendency toward grassland. Grassland and cultivated land exhibited relative inclination, while wetland-to-cultivated-land transitions showed relative inclination. Water bodies showed a relative inclination towards conversion to shrubs and artificial surfaces, while artificial surfaces exhibited a relative inclination towards water bodies. Shrubs and bare land predominantly exhibited systematic inhibition. Figure 7b illustrates the LUCC intensity map from 2010 to 2020, with cultivated land and artificial surfaces showing systematic tendencies. Forest-to-cultivated-land transitions remained absolutely inclined, with forest-to-grassland conversions also systematically inclined. Grassland-to-artificial surfaces showed a relative inclination, while shrubs, wetlands, and bare land continued to demonstrate inhibited transformation characteristics.

3.2. Contribution Analysis of Driving Factors

As evidenced by the findings of the land use spatial pattern evolution analysis presented in Section 3.1, the most significant changes in land cover within the study area have occurred in cultivated land, forestland, grassland, and artificial surfaces. In order to gain a deeper understanding of the spatial patterns of land use and their evolution within the corridor region, this analysis links the spatial expansion characteristics of the four land types (cultivated land, forestland, grassland, and artificial surfaces) from 2000 to 2020 with the potential driving factors. By calculating the probability of land-use expansion based on a range of factors, the study aims to illuminate the influence of these factors on the land-use patterns observed in the corridor region.

Land use change is a complex process that involves the spatial interaction of natural, geographic, and socio-economic factors [62]. Natural factors such as soil type, average annual temperature, precipitation, and topography have a significant impact on cultivated land use in a region [63]. The availability and proximity of transportation and public services also significantly influence the direction and speed of regional land use change [64,65]. In addition, changes in population size help to assess the overall development status of the region [66,67]. Moreover, Laos is a country where the population widely practices religion, and religious facilities play a crucial role in social life [68]. Considering the characteristics of the study area and insights from related research, 16 independent variables were selected from three categories: natural geography, socio-economics, and spatial facilities (Table 1). The PLUS model’s LEAS (Land Expansion Analysis Strategy) module was used to analyze the driving forces behind the expansion of cultivated land, forestland, grassland, and artificial surfaces in the study area from 2000 to 2020. This analysis partially explains the causes of land use changes in the region(Figure 8). Ultimately, the contribution of each factor to land expansion was calculated (with RMSE values all below 0.2), and the driving factors explained over 80% of the land expansion changes.

The results of the driving factors analysis indicate that natural geographical conditions are key drivers of land use patterns in the study area. However, as infrastructure and socio-economic development in the corridor area progress, the influence of socio-economic factors becomes increasingly apparent. From Figure 8a, it is clear that natural climate conditions are the dominant driving factor for the expansion of cultivated land. Newly cultivated land is primarily distributed in areas with higher average annual temperatures and relatively lower rainfall. Additionally, socio-economic factors such as special economic zones, population density, and railway construction have also influenced the distribution of cultivated land to some extent. Climate conditions play a decisive role in the expansion of cultivated land, as favorable conditions improve land productivity. Accessibility and policy guidance also influence this, with newly cultivated land being located closer to economic activity centers and transportation facilities. Figure 8b,c show that the distribution of new grasslands and forests is highly similar, with elevation, population density, average rainfall, and highways being the primary driving factors. Forests and grasslands are widely distributed in the study area, particularly in regions with lower average rainfall. Besides natural factors, population density also contributes to the growth of forests in some areas. Driven by economic benefits, the scale of rubber plantations continues to expand [69,70,71]. Moreover, improved road systems allow residents to work in urban areas or shift from subsistence farming to more intensive commercial crop production, reducing the need for deforestation.

Figure 8d shows that the expansion of artificial surfaces is primarily driven by socio-economic factors such as population density, the number of permanent markets, government offices, and special economic zones. Population density is the main driver of artificial surface expansion, and areas closer to public service facilities are more likely to be converted into artificial surfaces. Additionally, regions near special economic zones and major transportation routes are more prone to the emergence of new artificial surfaces. The establishment of special economic zones attracts substantial investments and businesses, spurring economic activity and population movement in surrounding areas, thus increasing land demand. The construction of rapid transportation corridors mainly contributes to economic growth and increased land use intensity [72], ultimately changing land use patterns.

3.3. Multi-Scenario Simulation of LULC

The study, using Laos’ land use data from 2010 to 2020, sets up multiple scenario models based on the results of analyses of land use evolution trends and driving mechanisms.

(1)

Baseline Development (BD)

This is mainly a reference scenario based on historical trends from the recent period (2010–2020). It models the natural spatial distribution of land use types without considering the impact of policies or constraints on land use conversions.

(2)

Investment-prioritized development (IPD)

This scenario focuses on the socio-economic impacts of railway construction and the development of special economic zones (SEZs) within the economic corridor. It increases the absolute tendency of other land types to convert into artificial surfaces and the tendency of cultivated land to convert into other land types while reducing the relative tendency of artificial surfaces to convert back to cultivated land. Therefore, in comparison to the Baseline Development scenario, the probability of cultivated land, forest, and grassland converting into artificial surfaces is increased by 20%, while the probability of all land types except artificial surfaces converting into cultivated land is also increased by 20%. Furthermore, the probability of artificial surfaces converting into other land types (excluding cultivated land) is reduced by 50%. In this scenario, the Lao government’s proposed SEZs are treated as key development and construction areas.

(3)

Harmonious development (HD)

This scenario promotes balanced development between ecological safety and economic construction. It considers the development potential brought by the opening of the China–Laos Railway and aligns with the Lao government’s vision for future ecological protection and restoration. It increases the absolute tendency of other land types to convert into artificial surfaces, but to achieve sustainable development goals, it also enhances the relative tendency for other land types to convert into forests while suppressing the expansion of artificial surfaces. As in the Baseline Development scenario, the probability of cultivated land, forest, and grassland converting into artificial surfaces is increased by 20%, but the probability of cultivated land and grassland converting into forest is raised by 30%. The probability of artificial surfaces converting into other land types (excluding cultivated land) is reduced by 20%. Additionally, the National Biodiversity Conservation Area (NBCA) of Laos is included as an ecological protection restriction area, while SEZs remain key development areas.

In addition, the study carefully considers changes in water bodies during the simulation. Based on historical land use trends, the growth of water bodies is primarily due to reservoirs created by hydropower projects, which are based on scientifically informed site selections and major regional decisions. Therefore, the simulation includes suitability maps for hydropower and water regions from the Laos Land Use Master Plan (2030). The results of the multi-scenario simulation are shown in Figure 9.

The simulation results indicate that in the BD scenario, the artificial surface in the study area expands to 375.23 km², with the expansion relying on the radiating influence of existing built-up areas. The development primarily occurs in an outward expansion from concentrated built-up areas, such as those observed in the cities of Vientiane and Luang Prabang. This expansion continues to demonstrate a pattern of artificial surfaces encroaching upon cultivated land and cultivated land encroaching upon forested areas (Figure 9b). In terms of land conversion, the area of cultivated land is forecast to be 3830.94 km² by 2030, with the majority of expansion occurring along the boundaries of settlements and transportation infrastructure. Conversely, the area of forestland is estimated to decline to 47,724.11 km², becoming an important source of land encroachment, with losses observed across the entire region. In the context of an IPO, the objective of optimizing economic benefits has resulted in a notable expansion of both the artificial surface and cultivated land, accompanied by an intensification of forestland loss. In terms of spatial distribution, new artificial surfaces are observed in economic zones such as Boten and Vientiane, as well as in proximity to railway stations and highway intersections along the corridor. The prioritized development and construction in designated investment zones have effectively controlled the sprawl of artificial surfaces in Vientiane and Luang Prabang (Figure 9c). However, the delineation of forest boundaries in proximity to urban areas persists, and the phenomenon of forestland being fragmented into discrete patches of grassland remains a notable trend. In the HB scenario, the expansion of both artificial surfaces and cultivated land slows down but still shows growth. By 2030, the area of artificial surfaces is projected to increase to 404.82 km², while cultivated land will grow to 3846.00 km². From a spatial perspective, artificial surfaces are becoming increasingly concentrated, with key cities along the corridor demonstrating substantial growth. Concurrently, the constraints on land conversion in NBCA (National Biodiversity Conservation Areas) are substantially slowing the loss of forestland, resulting in a closer alignment of the simulation results with the targets outlined in Laos’ Master Plan for 2030 (Figure 9d).

4. Discussion

4.1. Evolution of Spatial Patterns of Land Use in Economic Corridors

The initial land cover in the northern Lao section of the economic corridor predominantly consisted of traditional agricultural and resource collection areas. Due to geographical constraints and limited economic and technological conditions, the corridor region has retained a large amount of primary forest and grassland, while infrastructure and concentrated development are mainly found in the capital area. A similar situation can be observed in projects such as the Lamu Port-South Sudan–Ethiopia Transport Corridor (LAPSSET) in East Africa [73] and India’s Delhi–Mumbai Industrial Corridor (DMIC) [74]. In these new economic corridors, much of the land remains in its natural state, exhibiting a “single-core radiation” spatial pattern where development is concentrated around a single economic center. This land cover characteristic reflects the foundational phase of economic corridor development in most Global South countries, aligning with the findings of this study.

As the economic corridor developed, land use intensity increased alongside Laos’ rapid urbanization trend, particularly since the early 2000s, driven by government-supported infrastructure and increased foreign direct investment [31,75]. Notable differences emerged among administrative regions within the corridor: Vientiane, as the capital, remained the most intensively developed area with linear growth in land use intensity, while Vientiane Province, adjacent to the capital, also experienced moderate growth. Meanwhile, Luang Namtha Province, on the other end of the corridor, saw its land use intensity double between 2000 and 2010, matching Vientiane’s growth rate and becoming the corridor’s most significant area of development. By contrast, Luang Prabang and Oudomxay Provinces in the corridor’s central section showed minimal change.

By categorizing land use levels based on the functional degree of land transformation, we can further refine land use analysis and observe the corridor region’s spatial pattern transformation through the trends in spatial centers of various functions. Between 2000 and 2020, the shifts in the spatial centers of agricultural and built-up land reveal a fundamental shift in the region’s spatial structure. Although both land types generally show a net increase, the spatial distribution of land changes indicates that cultivated land in the southern corridor grew significantly more than in the north. However, as a traditional agricultural area in Laos, the corridor has seen a substantial increase in built-up land over these two decades. This change is driven by adjustments in regional policy and the implementation of comprehensive development strategies. The Lao government has promoted village settlement and agricultural technology in shifting cultivation areas, effectively fostering agricultural development in the southern corridor plains along the Mekong River [76]. Despite the limited effectiveness of these measures in the northern highlands due to geographical and ecological constraints, improved transportation infrastructure and the establishment of border economic zones have introduced new growth points in the north. This has led to a transformation of the corridor’s spatial pattern from a “single-core radial” structure toward a “dumbbell-shaped” structure.

4.2. Land Use Structure and Transition Pathways in Economic Corridors

This study uses a land-use transition matrix to understand the structure and transition pathways of land use in the economic corridor area, which encompasses data on changes in land quantity, structure, and direction. In terms of quantity and structure, the reciprocal transformation between forestland and grassland dominates, with widespread spatial distribution. The primary feature of the 2000–2010 phase was the conversion between forestland and cultivated land, where the ecological costs were relatively high due to lower development intensity. In contrast, during 2010–2020, forestland loss remained the dominant trend, with commercial logging leading to increased forest degradation and fragmentation. More than one-third of the study area was converted to grassland and degraded forest, while the area of artificial surfaces showed a sustained growth trend, increasing by 168.76 km² from 2010 to 2020. This phase was characterized by the conversion of cultivated land into artificial surfaces, with the construction of hydropower facilities significantly expanding water areas.

As observed by some scholars, economic corridor strategies typically involve large-scale infrastructure projects such as roads, railways, and industrial parks, which inevitably lead to an increase in artificial surfaces [77,78,79]. However, in comparison, Global South countries—especially those with lower population densities—often experience large-scale development in previously undeveloped or minimally developed areas during economic corridor construction, resulting in more substantial increases in artificial surfaces [40]. However, in comparison, this characteristic is less pronounced in Laos’ economic corridors; artificial surface expansion in Laos remains largely concentrated around the peripheries of pre-existing urban core areas. The transformation of natural land is currently the most controversial issue in the construction of economic corridors in Global South countries. Critical studies have highlighted the negative impacts of natural land reduction on ecosystems, particularly concerning biodiversity and habitat protection. Wettasin (2023), in his research on Thailand’s Eastern Economic Corridor, pointed out that land-use changes in the corridor and adjacent provinces have significantly reduced suitable habitats, contributing to the migration of wild Asian elephants into rural agricultural areas and increasing crop destruction incidents [80]. Similarly, Li’s study on the China-Mongolia-Russia Economic Corridor explored this phenomenon more systematically, noting that human activities affect vegetation cover in the corridor region. However, the study showed mixed results, with both vegetation improvement and degradation. Vegetation improvement is primarily attributed to human interventions, such as policy adjustments and afforestation efforts, whereas natural vegetation cover is more adversely affected by factors like climate change and overgrazing. This suggests that economic corridors do not inevitably lead to a loss of natural land cover [81]. While the contributing factors for natural land loss may differ, the findings also support the significant trend of ecological land conversion observed in this study.

In summary, the land use challenges faced by Global South countries in economic corridor construction primarily revolve around understanding landscape changes. On the one hand, the baseline of natural land should serve as a reference for the impact of land development. In comparison, most developable land in developed countries has already been utilized, so economic corridor construction typically involves the expansion of existing urban and infrastructure networks, resulting in relatively smaller increments. On the other hand, in Global South countries like Laos, Cambodia, and Kenya, economic development is prioritized, and large-scale infrastructure construction is often seen as a crucial driver of economic growth. While these projects lead to natural land loss, they also bring essential infrastructure, employment opportunities, and economic activities, thus fostering overall economic development. Therefore, landscape changes resulting from economic corridor construction are a common feature in the Global South. The balance between development and environmental protection is a recurring theme in discussions on sustainable development [82]. This requires that countries in the Global South develop comprehensive land use planning and management systems that are informed by an understanding of their unique landscape characteristics and that incorporate regular monitoring of land use and change.

With regard to land-use transfer pathways, there has been a clear and absolute trend towards the conversion of forestland into cultivated land. However, transitions such as cultivated land to shrubland, artificial surfaces to forestland, and bare land to cultivated land exhibit inhibition tendencies. These patterns align with general trends seen in land system changes across the Global South [58,83]. E Economic growth drives cultivated land expansion, particularly for crops like rubber and corn, making the conversion between cultivated land and forestland a recurring theme, especially in Southeast Asia, as observed in LUCC analyses [84,85]. Moreover, unique transition patterns in ecologically fragile and rapidly urbanizing regions also emerged, with a systematic tendency for forestland to convert to grassland and a strong conversion trend from cultivated land to water areas, as well as wetland and grassland to water bodies. These trends not only reflect Laos’s specific approach to forest and hydropower resource development but also exemplify common resource management challenges faced by many Global South countries.

When comparing land use transfer trends between the two stages, some noteworthy changes emerge: (1) Increased Tendency for Forestland Conversion: While there is an overall trend for forestland conversion to other land types, the high transition volume between forestland and grassland makes its impact less apparent. Though both are ecological land types, their ecosystem services differ, underscoring the need to address the ongoing loss of forestland, which highlights the urgency of strengthening forestland protection. (2) Restriction on Cultivated land Reversion to Forestland: Cultivated land shows a growing tendency to convert into artificial surfaces, indicating a systematic bias towards cultivated land for expansion rather than other land types. Urbanization, fueled by economic corridor development, has thus placed cultivated land protection under greater pressure as cultivated land becomes the primary target for conversion to artificial surfaces. (3) Intensified Conversion of Grassland to Artificial surfaces: Though the absolute number of grassland conversions is low, the intensity of conversion to artificial surfaces is above average, reflecting a trend toward artificial surface expansion. In conclusion, analyzing the transition intensities reveals that the unidirectional flows from forestland to cultivated land and cultivated land to artificial surfaces create potential land-use conflicts. With continuous growth in urban land demand, the systematic land-use transition pathways, once established, are likely to self-reinforce, exacerbating functional imbalances in land allocation.

4.3. Trends and Challenges in Land Use Development in Economic Corridors

The evolution of land use patterns in these corridor regions is a complex process. While geographic and climatic conditions remain the primary determinants of land use patterns, the series of projects within the economic corridor has significantly influenced the region’s socio-economic and physical space, resulting in notable interactive impacts on land use patterns. On the one hand, factors such as accessibility and government policy have resulted in the expansion of cultivated land in the proximity of economic activity centers and transportation facilities. Conversely, the expansion of grasslands and forests is closely related to population density and the construction of highways. On the other hand, the expansion of artificial surfaces is significantly influenced by population density, the location of special economic zones, and major transportation routes. This illustrates that the development of economic corridors, through infrastructure improvements and strategic industrial spatial planning, can effectively accelerate urban development in the region. Consequently, it acts as an intrinsic force shaping the evolution of land use patterns and enhances land use efficiency.

The machine-learning-based simulation prediction for land-use patterns in the corridor region by 2030 reveals that if current development speed and an inertia-driven land supply model (BD) are maintained, urban expansion will depend solely on the growth of existing built-up areas. This trend suggests that unregulated sprawl in Vientiane and Luang Prabang will further strain resource capacity and likely intensify polarization in Laos’s capital city. The region will face a common ecological deterioration challenge observed in Global South countries, where forestland is the primary type of land at risk for conversion, impeding sustainable development goals [86,87]. Based on the analysis of drivers behind land-use evolution within the economic corridor indicates that strengthening regional socio-economic impacts through corridor-led projects like railway lines and special economic zones (IPO) can somewhat mitigate unregulated artificial surface expansion and improve regional development equity. This managed approach would yield superior investment returns compared to the Baseline Development model [31]. However, the control of forest loss trends remains a significant challenge. It is important to highlight that the implementation of the development model of ecological protection and centralized investment guidance (HB), through the planning of priority development and construction-led centralized investment zones, the strengthening of the management of ecological-type land use, and the establishment of restrictions on forestland conversion in the NBCA protection zones, can help to mitigate the loss of forestland to a certain extent while maintaining the growth of man-made land surface and arable land in line with the goal of sustainable development. Despite the limitations of the multi-scenario projection, which was based on historical trends and a paucity of socio-economic data, the simulation results offer insights into the efficacy of different spatial planning interventions in controlling land use patterns.

Therefore, many Global South countries are currently in a phase where economic corridor development is accelerating urbanization and improving socio-economic levels. Land expansion for economic development is inevitable, but adopting a development principle that balances intensive growth with ecological protection and guides land use through spatial planning interventions will better support sustainable development. This is of great importance for the rational allocation of resources and land reserves in Global South countries.

4.4. Policy Implications

Land is the fundamental resource for achieving the 2030 Sustainable Development Goals. The issues of disorderly expansion of construction space and ecological loss arising from the evolution of land use patterns during economic corridor development can be seen as challenges and trade-offs faced by Global South countries in their development processes, but they should not be considered an inevitable path. Global South countries can learn from the experiences of similar nations and combine them with their own realities to institutionalize changes and explore more sustainable development paths.

Firstly, the issues of disorderly expansion of construction space and ecological loss faced by Global South countries like Laos are often due to land supply and management systems lagging behind the dynamic demand for land expansion in the rapid development process. It is essential to improve land supply methods and management systems. In terms of land supply, a zoning-based planning intervention in the economic corridor’s land supply can be adopted. Through resource and environmental carrying capacity evaluations and development suitability assessments, concentrated development zones and ecological protection control zones should be scientifically delineated to harness agglomeration effects while avoiding excessive interference with key natural ecosystems. Additionally, the overall land use structure and spatial layout of the economic corridor should be controlled to ensure the sustainable development of land resources and socio-economic systems.

Secondly, given the “dumbbell-shaped” spatial structure trend observed in the Laos Economic Corridor, we should follow the mechanism of influencing the spatial layout of transport infrastructure and industries to promote the land development mode of supporting infrastructure and industrial projects on a regional scale and strengthen the development of nodes in Luang Prabang and Vang Vieng in the middle section of the corridor. Priority should be given to the concentrated development of land around transport hubs, and transportation hubs, logistics centers, and industrial parks should be strategically laid out to promote the formation of industrial clusters or economic centers. Through these nodes, surrounding areas can be driven to develop their land and industries. Implementing a “node-radiation” development model would increase land use efficiency through concentrated development while preventing resource waste and land use imbalance caused by inefficient economic activities, thereby promoting the coordinated development of the corridor region.

5. Conclusions

The development strategy of economic corridors is widely regarded as an essential path for regional economic growth. However, emerging corridors in Global South countries face more significant land conflict challenges compared to those in industrialized Northern countries due to factors like limited infrastructure, incomplete policies, and vulnerable ecosystems. Addressing these challenges requires a reassessment of the interwoven economic and land use dynamics of economic corridors, balancing economic development with land resource preservation. This study utilized GlobeLand30 data from 2000, 2010, and 2020, combining spatial statistical analysis and machine learning to create a framework of static pattern analysis, dynamic process analysis, mechanism identification, and future trend prediction. Key findings include

• Shift to a Dumbbell Structure in Land Use Patterns: With the construction of economic corridors, the land use pattern in Laos’ corridor regions has gradually shifted from a “single-core radial” structure to a “dumbbell-shaped” structure. The development and restructuring of land-use functions have created new growth points, potentially alleviating issues of unbalanced spatial development typical of the Global South, where metropolitan areas often develop disproportionately. However, due to a weak economic foundation and an export-oriented economy, the corridor tends to favor this two-end aggregation, reinforcing a dumbbell pattern.
• Land Use Conflicts Through Unidirectional Conversion: In the evolution of spatial land use, the unidirectional flow from forestland to cultivated land and cultivated land to artificial surfaces serves as a potential source of land-use conflict. From 2000 to 2020, there was a substantial transformation between forestland, grassland, and cultivated land in the Laos corridor. A systematic trend was observed in the conversion of forestland to grassland, with forestland loss intensifying over time. This process of unidirectional transition poses a risk for land use conflicts in Global South economic corridors, necessitating appropriate guidance and timely control.
• Mechanisms of Land Use Pattern Evolution: Although geographic and climatic conditions still play a decisive role in shaping cultivated land, forestland, and grassland, the series of projects within economic corridors have reshaped the socio-economic and spatial environment of these areas. Key factors such as transportation infrastructure and industrial spatial layout drive changes in the patterns of ecological land, agricultural land, and built-up land. The 2030 forecast underscores the need for urgent spatial planning intervention, with land management rules that prioritize both forest protection and intensive development to balance economic growth and environmental preservation.

From a theoretical perspective, this study provides a framework for understanding the spatial and temporal evolution characteristics of land use and cover in emerging economic corridors in the Global South and the mechanisms that influence them. This framework, in contrast to traditional one-dimensional approaches, reveals the complexity and diversity of land-use changes and identifies systematic paths of land conflict. From a practical perspective, this study addresses the conflict between regional economic development and land resources in Global South countries, modeling the effectiveness of spatial intervention policies. This helps urban policymakers evaluate spatial planning and governance policies, offering useful recommendations for government agencies coordinating land supply and management systems. The methods and results presented here also serve as a significant reference for the rapidly urbanizing Global South facing similar pressures on economic growth and environmental resources.

However, this study faces certain limitations, including constraints on time and data availability, with economic data and land-use efficiency not fully integrated into the analytical framework. Moreover, the influence of the corridor on broader connected areas has not been thoroughly analyzed, with numerous potential hidden factors yet to be fully considered. Future research should examine broader inter-regional interactions and cross-regional land-use change, promoting more comprehensive and standardized research to contribute to sustainable development.