Spatiotemporal scale and integrative methods matter for quantifying the driving forces of land cover change
Graphical abstract
Introduction
Land cover patterns have been altered at an unprecedented pace, magnitude and spatial scale during recent decades (Gao et al., 2016). A large amount of natural land (such as forests, grassland and wetland) has been converted into developed land (such as urban and cropland), thus pressures on the environment and ecosystems are increasing (Liang et al., 2017), including atmospheric composition, energy and water balance, degradation of soil and water and declines in biodiversity (Foley et al., 2005). Land cover change (LCC) thus constitutes one of the nine “planetary boundaries” (Rockstrom et al., 2009), and land degradation neutrality was proposed as a UN 2030 global sustainable development goal. To achieve this objective, understanding the processes, trends, causes and consequences of LCCs is a key issue in global change research (Pielke, 2005; Verburg et al., 2009), and its focus has gradually transitioned from ‘globe’ to ‘region’ and from ‘nature’ to ‘humanity’ (Liu et al., 2017).
For land systems science, the causes of LCCs have attracted attention at different spatial and temporal scales, and the richness of explanations has greatly increased, often at the expense of generality (Eunice Quintero-Gallego et al., 2018). Current research on the driving forces of LCCs has focused on different scales and methods, but there are still many questions to be solved. First, most previous studies about drivers have focused on the global, country or state scale (Alexander et al., 2015; Amuti and Luo, 2014; Arowolo and Deng, 2018), and a number of studies also have assessed LCC at small scale (local), such as at the county or town level (Abadie et al., 2018; da Silva et al., 2016). The results from these scales have found similar drivers, such as topography, accessibility, urbanization and other biophysical or socioeconomic factors (Abadie et al., 2018; Bansal et al., 2016; Gao et al., 2016; Hoyos et al., 2018). However, the driving forces on differences and similarities across multiple scales is worthy of more attention. Second, LCC exhibits strong path dependence (Celio et al., 2014). To study human impacts on the natural environment, it is necessary to assess the history of LCCs and interpret the spatiotemporal patterns of such changes relevant to other environmental and human factors, so it is not sufficient to assess only the changes in areas (Wang et al., 2013; Zhou et al., 2008). The various components of LCCs, such as total change, net change and swap change, have been detected and estimated in previous studies (Gao et al., 2016; Lu et al., 2004; Pontius et al., 2004; Wang and Wang, 2013). However, most of these studies have focused only on decreases or increases in one certain land type, such as croplands (Alexander et al., 2015; Arowolo and Deng, 2018; Najmuddin et al., 2018; van Vliet et al., 2015; Wang et al., 2015; Wood et al., 2004; Xie et al., 2014), forests (Abadie et al., 2018; Acacio et al., 2017), grasslands (Monteiro et al., 2011), wetlands (Zheng et al., 2017), rural lands and urban lands (Zhou et al., 2008). Recently, the topic of the change trajectory (change from a certain land cover type to another, such as the transition from forest to grassland or cropland) (Calaboni et al., 2018; Lambin and Meyfroidt, 2010; Yackulic et al., 2011) has received increasing attention in studies on landscape change analyses (Hietel et al., 2004; Kayhko and Skanes, 2006; Wang et al., 2013), and it is particularly useful for addressing this challenge. Some studies have focused on multiple trajectories (Wang et al., 2013; Yeh and Liaw, 2016), but typical or dominant paths were not highlighted, especially those characterized by historical changes, such as revegetation. The identification of driving forces that cause dominant LCCs will help managers establish policies that are designed to prevent or minimize the negative effects of LCCs. Third, previous studies have frequently quantified the driving forces linked to LCCs through using descriptive analysis (Patarasuk and Binford, 2012; Teixeira et al., 2014; Zhu et al., 2010), principal component analysis (PCA) (Abadie et al., 2018; Chen et al., 2017; Du et al., 2014; Hoyos et al., 2018; Li et al., 2018; Meneses et al., 2017; Zhao and Liu, 2014), analytic hierarchy process (AHP) (Thapa and Murayama, 2010), redundancy analysis (RDA) (Gao et al., 2016; Parcerisas et al., 2012), canonical correlation analysis (CCA) (Salvati et al., 2016), and regression analysis (Abadie et al., 2018; Acacio et al., 2017; Kindu et al., 2015; Monteiro et al., 2011; Najmuddin et al., 2018; Yu et al., 2017). Machine learning methods have gradually been used in identifying driving forces, such as neural networks (Jalala et al., 2011; Tan et al., 2014), random forests, and gradient boosting trees (Moore and Lin, 2019). These methods have been used to interpret the links between drivers and LCCs at one certain scale, but they could not be used to identify the relationships between measured variables and quantify the differences across scales.
Landscape development usually involves urbanization and industrialization, which can lead to significant LCCs. Therefore, it is especially critical to identify the main LCC processes in economically underdeveloped and ecologically fragile areas because of the relatively rapid pace and extent of LCC and its effects on fragile ecosystems (Gao et al., 2016; Li et al., 2012; Liu et al., 2005; Serra et al., 2008; Sleeter et al., 2013). Over the past several decades, China has been one of the most rapidly developing countries, and it has experienced continuous urbanization, population increases and ecological restoration (Chen et al., 2019). The largest ecological restoration program (GFGP: Grain for Green Program) in China has been implemented since 1999, and it has caused increasing landscape changes that have received widespread attention (Du et al., 2014; Liu et al., 2005; Wang et al., 2012). The Loess Plateau is a representative dryland in China. It was the 1st pilot and demonstration area for GFGP, and it has experienced extensive ecological restoration and significant economic development in the last 20 years. Therefore, the Loess Plateau is a suitable research region with typical LCC processes, including agricultural intensification, urbanization, and ecological restoration.
The studies on these LCC processes and their driving forces need to pay more attention on differences across spatiotemporal scales to provide reasonable suggestions on landscape planning and management. In this paper, we examined the historical dynamics (between 1990 and 2015) of land cover area at both regional and town-cluster scales on the Loess Plateau to deepen our understanding of possible spatiotemporal differences in the driving forces in arid and semiarid regions. The objectives of this paper are to: (1) examine typical land cover transitions at regional and town-cluster scales in three time intervals (1990–2000, 2000–2010, and 2010–2015); (2) quantify the differences in the driving forces (biophysical and socioeconomic factors) of LCCs using integrated methods at two spatial scales, i.e., using structural equation modeling (SEM) and a mixed-effects model (MEM); (3) identify the temporal differences in the impacts of the driving forces; and (4) discuss the differentiated landscape management and planning processes at a small spatial scale for sustainable development in the region.
Section snippets
Study area
The Loess Plateau (33°43′-41°16′ N, 100°54′-114°33′ E) is located in northern China (Fig. 1) and covers an area of 6.24 × 105 km2. It is characterized by inland arid and semiarid climate with concentrated rainfall, intensive evaporation and water shortages. The mean annual precipitation ranges from 700 mm in the southeast to 200 mm in the northwest, and the mean annual temperature ranges from 15.3 °C in the southeast to −3.1 °C in the northwest. Most of the research region is covered by thick
Historical trend of land cover pattern
In the last 25 years, there was a net gain in forests and grasslands throughout the Loess Plateau, especially between 2000 and 2010; however, there was a significant decrease in croplands (Fig. 3a). The area of built-up lands has increased steadily since 1990, reflecting a steady land urbanization process.
The proportion of each land cover in the six categories of township areas varied (Fig. 3b). Croplands and built-up lands accounted for the largest percentage in the class I and II township
The spatiotemporal scale dependence of driving forces
A large number of previous studies have evaluated the impact of socio-economic decisions and policies on land cover at the regional and local scale (Dupin et al., 2018; Ferrara et al., 2016). Differences can be seen in the amount and spatial allocation schemes between them (Wang et al., 2016). For example, large-scale arrangements only apply to the amount or proportion of land cover in an entire region, whereas local-scale planning requires meticulous attention to the use of each street and
Conclusions
This paper focused on the land change processes in a typical dryland region, i.e., the Loess Plateau in China, from both spatial and temporal perspectives, and we suggested how causal associations with multiple biophysical and socioeconomic forces and their variations could be quantified. We made new attempts to at least partly fill the knowledge gaps in land change science by comprehensively investigating LCC trajectories and the spatiotemporal variations in the underlying natural and
CRediT authorship contribution statement
Ying Luo:Conceptualization, Methodology, Writing - original draft, Writing - review & editing.Yihe Lü:Conceptualization, Supervision, Writing - review & editing, Funding acquisition.Lue Liu:Investigation.Haibin Liang:Investigation.Ting Li:Data curation.Yanjiao Ren:Data curation.
Acknowledgments
The study was funded by the Chinese Academy of Sciences [XDA23070201], the National Key Research and Development Program of China [No. 2016YFC0501601], and the International Partnership Program of Chinese Academy of Sciences [No. 121311KYSB20170004]. We also thank Zengnao Ren for assisting in the collection of socioeconomic data.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (95)
- et al.
Drivers for global agricultural land use change: the nexus of diet, population, yield and bioenergy
Global Environmental Change-Human and Policy Dimensions
(2015) - et al.
Modeling land use decisions with Bayesian networks: spatially explicit analysis of driving forces on land use change
Environ. Model Softw.
(2014) - et al.
Urban driving forces and megacity expansion threats. Study case in the Mexico City periphery
Habitat International
(2017) - et al.
Shaping the role of ‘fast’ and ‘slow’ drivers of change in forest-shrubland socio-ecological systems
J. Environ. Manag.
(2016) - et al.
Optimization of land use structure and spatial pattern for the semi-arid loess hilly-gully region in China
Catena
(2010) - et al.
Historical dynamics of ecosystem services and land management policies in Switzerland
Ecol. Indic.
(2019) - et al.
Change trajectories and key biotopes - assessing landscape dynamics and sustainability
Landsc. Urban Plan.
(2006) - et al.
Land use transitions: Socio-ecological feedback versus socio-economic change
Land Use Policy
(2010) - et al.
Risk management for optimal land use planning integrating ecosystem services values: a case study in Changsha, middle China
Sci. Total Environ.
(2017) - et al.
Spatial and temporal patterns of China’s cropland during 1990-2000: An analysis based on Landsat TM data
Remote Sens. Environ.
(2005)
Drivers of land cover and land use changes in St. Louis metropolitan area over the past 40 years characterized by remote sensing and census population data
Int. J. Appl. Earth Obs. Geoinf.
Monitoring land use/land cover transformations from 1945 to 2007 in two peri-urban mountainous areas of Athens metropolitan area, Greece
Sci. Total Environ.
Spatial modeling of agricultural land use change at global scale
Ecol. Model.
Assessment of land cover changes and spatial drivers behind loss of permanent meadows in the lowlands of Italian Alps
Landsc. Urban Plan.
Bundles of ecosystem (dis)services and multifunctionality across European landscapes
Ecol. Indic.
Long-term socio-economic and spatial pattern drivers of land cover change in a Caribbean tropical moist forest, the Cockpit Country, Jamaica
Agriculture Ecosystems & Environment
Land use changes, landscape ecology and their socioeconomic driving forces in the Spanish Mediterranean coast (El Maresme County, 1850-2005)
Environ. Sci. Pol.
Longitudinal analysis of the road network development and land-cover change in Lop Buri province, Thailand, 1989-2006
Appl. Geogr.
Detecting important categorical land changes while accounting for persistence
Agric. Ecosyst. Environ.
‘Land-use mixite’: evaluating urban hierarchy and the urban-to-rural gradient with an evenness-based approach
Ecol. Indic.
Integrating supply, flow and demand to enhance the understanding of interactions among multiple ecosystem services
Sci. Total Environ.
Land-cover and land-use change in a Mediterranean landscape: a spatial analysis of driving forces integrating biophysical and human factors
Appl. Geogr.
Land-cover change in the conterminous United States from 1973 to 2000
Global Environmental Change-Human and Policy Dimensions
Unpacking ecosystem service bundles: towards predictive mapping of synergies and trade-offs between ecosystem services
Global Environmental Change-Human and Policy Dimensions
Predicting the drivers of behavioral intention to use mobile learning: a hybrid SEM-neural networks approach
Comput. Hum. Behav.
Systematic processes of land use/land cover change to identify relevant driving forces: implications on water quality
Sci. Total Environ.
Drivers of urban growth in the Kathmandu valley, Nepal: examining the efficacy of the analytic hierarchy process
Appl. Geogr.
Manifestations and underlying drivers of agricultural land use change in Europe
Landsc. Urban Plan.
Land use-cover change processes in highly biodiverse areas: the case of Oaxaca, Mexico
Global Environmental Change-Human and Policy Dimensions
From land cover change to land function dynamics: a major challenge to improve land characterization
J. Environ. Manag.
Land-use changes and policy dimension driving forces in China: present, trend and future
Land Use Policy
Comparative analysis of land use/cover change trajectories and their driving forces in two small watersheds in the western Loess Plateau of China
Int. J. Appl. Earth Obs. Geoinf.
Driving forces of changes in the water and sediment relationship in the Yellow River
Sci. Total Environ.
Understanding the drivers of agricultural land use change in south-central Senegal
J. Arid Environ.
Socio-ecological changes on the Loess Plateau of China after grain to green program
Sci. Total Environ.
Forest recovery since 1860 in a Mediterranean region: drivers and implications for land use and land cover spatial distribution
Landsc. Ecol.
Landscape dynamics in Mediterranean oak forests under global change: understanding the role of anthropogenic and environmental drivers across forest types
Glob. Chang. Biol.
Analysis of land cover change and its driving forces in a desert oasis landscape of Xinjiang, Northwest China
Solid Earth
Land use/land cover change and statistical modelling of cultivated land change drivers in Nigeria
Reg. Environ. Chang.
An analysis of land use and land cover dynamics and causative drivers in a thickly populated Yamuna river basin of India
Appl. Ecol. Environ. Res.
The forest transition in Sao Paulo, Brazil: historical patterns and potential drivers
Ecol. Soc.
Land-use drivers of forest fragmentation vary with spatial scale
Glob. Ecol. Biogeogr.
China and India lead in greening of the world through land-use management
Nature sustainability
Land cover change in different altitudes of Guizhou-Guangxi karst mountain area, China: patterns and drivers
J. Mt. Sci.
Factors affecting driver speed choice along two-lane rural highway transition zones
J. Transp. Eng.
Drivers of land change: human-environment interactions and the Atlantic forest transition in the Paraiba Valley, Brazil
Land Use Policy
Spatial pattern of land use change and its driving force in Jiangsu Province
Int. J. Environ. Res. Public Health
Cited by (26)
Analysis and simulation of the driving mechanism and ecological effects of land cover change in the Weihe River basin, China
2023, Journal of Environmental ManagementLand cover change in global drylands: A review
2023, Science of the Total EnvironmentQuantifying the spatial differentiation mechanism of land use degree
2022, HeliyonCitation Excerpt :Meanwhile, machine learning methods such as neural networks and random forests have increasingly been used to identify drivers (Jalala et al., 2011; Tan et al., 2014; Moore and Lin, 2019). However, an important shortcoming of these methods is that they assume a significant linear relationship between drivers and changes in land use degree over the entire time series and cannot be used to identify spatially differential relationships between the variables and quantifiers being measured (Luo et al., 2020; Peng et al., 2019b). Spatial differentiation is the spatial manifestation of natural and socio-economic processes.
Pathways from the payment for ecosystem services program to ecological and socio-economic outcomes
2022, Ecological Indicators