Machine learning in modelling land-use and land cover-change (LULCC): Current status, challenges and prospects

A protocol to model future land use scenarios using Dinamica-EGO

Land use changes are important drivers of ecosystem change. They depend on ecological, social, economic and political aspects. This work aims to develop a detailed protocol to forecast land use changes using Dinamica-Ego software. It includes the 1) time frame definition, 2) future scenarios definition, 3) identify the major driving forces of land use change, 4) collection and organize data for the modelling process, 5) calculation of landscape metrics for the base year, and 6) Dinamica-Ego modelling. Here, several sub-steps are described that involve calculating the transition matrix, preparing the raster cube, calculating the Weights of Evidence (WoE), assessing multicollinearity, revising the raster cube, validating the land use change model, adjusting the transition matrix and WoE and running the future land use simulation. The protocol explains how to simulate land use changes to 2050, showing scenarios 1) business as usual and 2) urbanization in Kaunas (Lithuania).•The protocol details a step-by-step approach to model land use change using Dinamica-Ego;•This protocol can be replicated in forecasting land use in any urban area;•The results obtained using this protocol were well-validated. Therefore, the reliability is high.

Evaluation of the synergistic change in cultivated land and wetland in northeast China from 1990 to 2035

Wetlands are the most biodiverse ecological landscape in nature and one of the most important natural resources for human beings. In recent years, wetlands in northeast China have been increasingly converted into cultivated land, resulting in significant reduction in wetland area. Currently, the extensive and prolonged use of natural resources, combined with mismanagement and climate change, presents considerable challenges for both governments and future sustainability. This study utilized the PLUS model to analyze the spatial-temporal transformation of cultivated land and wetland in northeast China over the past 30 years and to project land use changes from 2020 to 2035. The analysis quantitatively evaluated the impacts of human activities and climate change. The results showed that: (1) Wetlands in northeast China have been progressively converted into paddy fields or degraded into unused land. (2) Topography, GDP, and temperature are the primary drivers of land use change over the past three decades. (3) There is an urgent need for national macro-policy regulation to mitigate the degradation of cultivated land and wetlands through the rational allocation of land resources.

Enhanced large-scale flood mapping using data-efficient unsupervised framework based on morphological active contour model and single synthetic aperture radar image

Floods are critical hydrological extremes that cause significant environmental damage. Remote sensing data, specifically Synthetic Aperture Radar (SAR), derived flood maps are crucial for detecting and quantifying this damage, enabling effective flood management and damage assessment. However, majority of SAR-based flood mapping frameworks are supervised and often face the problem of data generalisation and data labelling, which presents a challenge for model transferability and rapid mapping. To address this challenge, an unsupervised computer vision-based framework built upon the Morphological Chan-Vese Active contour model (Morph CV ACM) and unitemporal SAR image is proposed for flood extent mapping. In this work, sensitivity analysis of the model parameters is performed to check its applicability for two commonly found flooding patterns (clustered and scattered) in SAR images. Furthermore, a localised version of Morph CV ACM is proposed, which adaptively adjusts the model parameters according to the specific characteristics of flooding patterns, based on an empirically developed formula. The proposed framework is tested to map floods that occurred in North India in 2023 across the flood plains of the Yamuna River. The results were validated against flood reference masks generated by PlanetScope optical images for six different Areas of Interest (AOIs), representing varied land covers and flooding patterns. The novel framework accurately identified flood extents with a high F1 score of 0.935. Flood extents from the proposed framework were also compared with the Otsu segmentation, a widely established unsupervised method, and results indicated a major improvement in detecting flood extents with our framework. The improvements in the performance were attributed to the inherent property of Morph CV ACM to use region-based information to govern the energy equation of the model, leading to accurate flood boundary detection, while its use of morphological metrics enhances resilience to the speckle effect in SAR images. Additionally, the generated flood extents were overlaid on the 10 m resolution land cover map for performance assessment across different land covers. The extents generated from our framework provide enhanced flood mapping accuracy in built-up and agricultural areas, where precise mapping using SAR data is challenging yet crucial for damage assessment. The framework's automation and minimal data requirements make it a valuable tool for near-real-time, large-scale flood mapping, with significant potential to enhance damage assessment and guide effective flood management strategies.

Optimizing land use spatial patterns to balance urban development and resource-environmental constraints: A case study of China's Central Plains Urban Agglomeration

The unprecedented urbanization of the Central Plains Urban Agglomeration (CPUA) has significantly improved human well-being but has also led to severe land degradation and environmental challenges. Spatial zoning is a crucial tool for advancing sustainable land management in urban agglomerations. However, spatial zoning has become increasingly challenging due to the diversity associated with future land-use changes and the vulnerability of the ecological environment. This study applies a Self-Organizing Map (SOM) neural network model to emphasize the dual pivotal role of future land-use scenarios and resource-environment carrying capacity in optimizing sustainable land management zoning. Focusing on the CPUA in China, the study reveals several key findings: (1) Under the natural evolution scenario, the proportion of cultivated land decreases from 58.76 % to 55.58 %, resulting in a reduction of 9146 km2, while construction land expands from 14.72 % to 21.96 %, with an increase of 23,100 km2 and an average annual growth rate of 1.17 %. (2) The resource-environment carrying capacity across the CPUA is generally low to medium, with an average index value of 35.28. Spatially, the surrounding areas concentrate higher carrying capacity, while the central regions exhibit lower values. The western, northern, and southern edge regions show relatively higher capacities. (3) Based on comprehensive assessments of land-use patterns, ecological quality, and resource-environment carrying capacity, the CPUA is divided into nine distinct sustainable land management zones. Each zone requires tailored strategies that consider its specific resource endowments, ecological conditions, agricultural productivity, and urban development potential. Coordinated infrastructure development and resource-sharing initiatives are essential for promoting sustainable land management throughout the urban agglomeration. The proposed zoning optimization strategy strikes a balance between urban development demands and resource-environment constraints, offering a practical framework for refining land management policies and advancing sustainable development goals in large urban agglomerations.

Land Cover Classification Model Using Multispectral Satellite Images Based on a Deep Learning Synergistic Semantic Segmentation Network

Land cover classification (LCC) using satellite images is one of the rapidly expanding fields in mapping, highlighting the need for updating existing computational classification methods. Advances in technology and the increasing variety of applications have introduced challenges, such as more complex classes and a demand for greater detail. In recent years, deep learning and Convolutional Neural Networks (CNNs) have significantly enhanced the segmentation of satellite images. Since the training of CNNs requires sophisticated and expensive hardware and significant time, using pre-trained networks has become widespread in the segmentation of satellite image. This study proposes a hybrid synergistic semantic segmentation method based on the Deeplab v3+ network and a clustering-based post-processing scheme. The proposed method accurately classifies various land cover (LC) types in multispectral satellite images, including Pastures, Other Built-Up Areas, Water Bodies, Urban Areas, Grasslands, Forest, Farmland, and Others. The post-processing scheme includes a spectral bag-of-words model and K-medoids clustering to refine the Deeplab v3+ outputs and correct possible errors. The simulation results indicate that combining the post-processing scheme with deep learning improves the Matthews correlation coefficient (MCC) by approximately 5.7% compared to the baseline method. Additionally, the proposed approach is robust to data imbalance cases and can dynamically update its codewords over different seasons. Finally, the proposed synergistic semantic segmentation method was compared with several state-of-the-art segmentation methods in satellite images of Italy's Lake Garda (Lago di Garda) region. The results showed that the proposed method outperformed the best existing techniques by at least 6% in terms of MCC.

Temporal adjustment approach for high-resolution continental scale modeling of soil organic carbon

Open-source legacy data available for training soil organic carbon (SOC) models are limited and not uniformly distributed in space or time. While some process-based models predict SOC changes, most of the large-scale data-driven SOC modeling efforts overlook temporal shifts. Accounting for the expected temporal drift allows us to increase the accuracy of dataset available for machine learning models. Here we present an approach for creating proximity-based distance matrices using the legacy data available in contiguous US (CONUS) and generating spatially resolved temporal shift projections that adjust observations to the target date. The approach was evaluated by comparing SOC observations projected to two reference years, SOC1980 and SOC2020 and without temporal adjustment (SOCno-adj). Stocks of SOC projections showed significant differences between SOCno-adj and SOC2020. Baseline estimate of SOC stocks in CONUS croplands (top 1 m) were higher based on SOCno-adj (14.49 Pg C) compared to SOC2020 (13.29 Pg C), for pasture lands 15.49 Pg (SOCno-adj) and 14.22 Pg C (SOC2020), for forest lands at 39.52 Pg C (SOCno-adj) and 40.83 Pg C (SOC2020). The study results confirmed the validity of our methodology, and its capability to enhance SOC stock projections effectively with temporal adjustments. Potential users of this study's outcomes include many stakeholders involved in carbon incentive programs, including farmers, scientists, policy makers, and industry partners.

Monitoring of agricultural progress in rice-wheat rotation area based on UAV RGB images

Real-time monitoring of rice-wheat rotation areas is crucial for improving agricultural productivity and ensuring the overall yield of rice and wheat. However, the current monitoring methods mainly rely on manual recording and observation, leading to low monitoring efficiency. This study addresses the challenges of monitoring agricultural progress and the time-consuming and labor-intensive nature of the monitoring process. By integrating Unmanned aerial vehicle (UAV) image analysis technology and deep learning techniques, we proposed a method for precise monitoring of agricultural progress in rice-wheat rotation areas. The proposed method was initially used to extract color, texture, and convolutional features from RGB images for model construction. Then, redundant features were removed through feature correlation analysis. Additionally, activation layer features suitable for agricultural progress classification were proposed using the deep learning framework, enhancing classification accuracy. The results showed that the classification accuracies obtained by combining Color+Texture, Color+L08CON, Color+ResNet50, and Color+Texture+L08CON with the random forest model were 0.91, 0.99, 0.98, and 0.99, respectively. In contrast, the model using only color features had an accuracy of 85.3%, which is significantly lower than that of the multi-feature combination models. Color feature extraction took the shortest processing time (0.19 s) for a single image. The proposed Color+L08CON method achieved high accuracy with a processing time of 1.25 s, much faster than directly using deep learning models. This method effectively meets the need for real-time monitoring of agricultural progress.

Effects of recent urbanization on carbon and nitrogen burial rates of sedimentary records in a tropical coastal lagoon (Brazil)

Land use and land cover changes (LULCC) are a global environmental issue that has impacted biogeochemical cycles worldwide. Sedimentary records can demonstrate the effects of LULCC on aquatic ecosystems, where the recent urbanization has been linked to changes in carbon and nitrogen burial. In this study, we reconstructed long-term LULCC and sedimentary records of carbon (C), nitrogen (N), phosphorus (P), and sediment burial rates in a eutrophic tropical coastal lagoon affected by recent urban expansion. Based on analyses of 30 years of satellite imagery and sedimentary records from 1932 to 2013, we revealed that urban expansion over low-productivity agricultural-pasture areas increased siltation and C, N, P concentrations and fluxes in the coastal lagoon. Large temporal variability of such parameters revealed not only the effects of LULCC on the lagoon's burial rates, but also the influence of artificial sand barrier openings, which connect the studied lagoon to the sea, reducing C, N, P, and particle deposition in the sediment. Our results support multi-proxy methods to assess the relationships between recent urbanization, rising C, N, and P burial rates, and the eutrophication process. We highlight that artificial sandbar openings, the current eutrophication management strategy for coastal lagoons, are ineffective in reducing the eutrophication state, even in the recent scenario of decreasing C, N, and P burial rates.

Generating high-resolution land use and land cover maps for the greater Mariño watershed in 2019 with machine learning

Land Use and Land Cover (LULC) maps are important tools for environmental planning and social-ecological modeling, as they provide critical information for evaluating risks, managing natural resources, and facilitating effective decision-making. This study aimed to generate a very high spatial resolution (0.5 m) and detailed (21 classes) LULC map for the greater Mariño watershed (Peru) in 2019, using the MORINGA processing chain. This new method for LULC mapping consisted in a supervised object-based LULC classification, using the random forest algorithm along with multi-sensor satellite imagery from which spectral and textural predictors were derived (a very high spatial resolution Pléiades image and a time serie of high spatial resolution Sentinel-2 images). The random forest classifier showed a very good performance and the LULC map was further improved through additional post-treatment steps that included cross-checking with external GIS data sources and manual correction using photointerpretation, resulting in a more accurate and reliable map. The final LULC provides new information for environmental management and monitoring in the greater Mariño watershed. With this study we contribute to the efforts to develop standardized and replicable methodologies for high-resolution and high-accuracy LULC mapping, which is crucial for informed decision-making and conservation strategies.

Enhancing spatio-temporal environmental analyses: A machine learning superpixel-based approach

The progressive evolution of the spatial and temporal resolutions of Earth observation satellites has brought multiple benefits to scientific research. The increasing volume of data with higher frequencies and spatial resolutions offers precise and timely information, making it an invaluable tool for environmental analysis and enhanced decision-making. However, this presents a formidable challenge for large-scale environmental analyses and socioeconomic applications based on spatial time series, often compelling researchers to resort to lower-resolution imagery, which can introduce uncertainty and impact results. In response to this, our key contribution is a novel machine learning approach for dense geospatial time series rooted in superpixel segmentation, which serves as a preliminary step in mitigating the high dimensionality of data in large-scale applications. This approach, while effectively reducing dimensionality, preserves valuable information to the maximum extent, thereby substantially enhancing data accuracy and subsequent environmental analyses. This method was empirically applied within the context of a comprehensive case study encompassing the 2002-2022 period with 8-d-frequency-normalized difference vegetation index data at 250-m resolution in an area spanning 43,470 km2. The efficacy of this methodology was assessed through a comparative analysis, comparing our results with those derived from 1000-m-resolution satellite data and an existing superpixel algorithm for time series data. An evaluation of the time-series deviations revealed that using coarser-resolution pixels introduced an error that exceeded that of the proposed algorithm by 25 % and that the proposed methodology outperformed other algorithms by more than 9 %. Notably, this methodological innovation concurrently facilitates the aggregation of pixels sharing similar land-cover classifications, thus mitigating subpixel heterogeneity within the dataset. Further, the proposed methodology, which is used as a preprocessing step, improves the clustering of pixels according to their time series and can enhance large-scale environmental analyses across a wide range of applications.

Prediction of land use for the next 30 years using the PLUS model's multi-scenario simulation in Guizhou Province, China

Land use changes significantly impact the structure and functioning of ecosystems. The current research focus lies in how to utilize economic and policy instruments to regulate conflicts among stakeholders effectively. The objective is to facilitate rational planning and sustainable development of land utilization resources. The PLUS model integrates a rule-based mining method for land expansion analysis and a CA model based on multi-type stochastic seeding mechanism, which can be used to mine the driving factors of land expansion and predict the patch-level evolution of land use landscapes. Using the PLUS model, a simulation was conducted to study the future land use distribution in the research area over the next 30 years. Based on land use data from Guizhou Province in 2000, 2010, and 2020, a total of 16 driving factors were selected from three aspects: geographical environment, transportation network, and socio-economic conditions. Four scenarios, namely natural development, urban development, ecological conservation, and farmland rotection, were established. Comparative analysis of the simulated differences among the various scenarios was performed. (1) The overall accuracy of the land use simulation using the PLUS model in the study area was 0.983, with a Kappa coefficient of 0.972 and a FoM coefficient of 0.509. The research accuracy meets the simulation requirements. (2) Through the simulation of four different scenarios, the study investigated the land use changes in Guizhou Province over the next 30 years. Each scenario exhibited distinct impacts on land utilization. Comprehensive comparison of the different simulation results revealed that the farmland protection scenario aligns with the sustainable development goals of the research area. Currently, there is a relative scarcity of research on land use simulation, particularly in model application, for Guizhou Province. This study aims to provide a reference for the rational planning of land resources and high-quality urban construction in Guizhou, promoting the high-quality economic development in tandem with advanced ecological and environmental protection.

SLMFNet: Enhancing land cover classification of remote sensing images through selective attentions and multi-level feature fusion

Land cover classification (LCC) is of paramount importance for assessing environmental changes in remote sensing images (RSIs) as it involves assigning categorical labels to ground objects. The growing availability of multi-source RSIs presents an opportunity for intelligent LCC through semantic segmentation, offering a comprehensive understanding of ground objects. Nonetheless, the heterogeneous appearances of terrains and objects contribute to significant intra-class variance and inter-class similarity at various scales, adding complexity to this task. In response, we introduce SLMFNet, an innovative encoder-decoder segmentation network that adeptly addresses this challenge. To mitigate the sparse and imbalanced distribution of RSIs, we incorporate selective attention modules (SAMs) aimed at enhancing the distinguishability of learned representations by integrating contextual affinities within spatial and channel domains through a compact number of matrix operations. Precisely, the selective position attention module (SPAM) employs spatial pyramid pooling (SPP) to resample feature anchors and compute contextual affinities. In tandem, the selective channel attention module (SCAM) concentrates on capturing channel-wise affinity. Initially, feature maps are aggregated into fewer channels, followed by the generation of pairwise channel attention maps between the aggregated channels and all channels. To harness fine-grained details across multiple scales, we introduce a multi-level feature fusion decoder with data-dependent upsampling (MLFD) to meticulously recover and merge feature maps at diverse scales using a trainable projection matrix. Empirical results on the ISPRS Potsdam and DeepGlobe datasets underscore the superior performance of SLMFNet compared to various state-of-the-art methods. Ablation studies affirm the efficacy and precision of SAMs in the proposed model.

Exploring optical descriptors for rapid estimation of coastal sediment organic carbon and nearby land-use classifications via machine learning models

This study utilizes ultraviolet and fluorescence spectroscopic indices of dissolved organic matter (DOM) from sediments, combined with machine learning (ML) models, to develop an optimized predictive model for estimating sediment total organic carbon (TOC) and identifying adjacent land-use types in coastal sediments from the Yellow and Bohai Seas. Our results indicate that ML models surpass traditional regression techniques in estimating TOC and classifying land-use types. Penalized Least Squares Regression (PLR) and Cubist models show exceptional TOC estimation capabilities, with PLR exhibiting the lowest training error and Cubist achieving a correlation coefficient 0.79. In land-use classification, Support Vector Machines achieved 85.6 % accuracy in training and 92.2 % in testing. Maximum fluorescence intensity and ultraviolet absorbance at 254 nm were crucial factors influencing TOC variations in coastal sediments. This study underscores the efficacy of ML models utilizing DOM optical indices for near real-time estimation of marine sediment TOC and land-use classification.

Land-use intensification dominates China's land provisioning services: From the perspective of land system science

A pressing challenge to global sustainability is meeting the escalating needs of a growing population while safeguarding land resources from degradation. In recent decades, China's rapid growth, expanding population, urban sprawl, and diminishing high-quality farmland have presented a compelling case suitable for exploring solutions and challenges related to this critical issue. Therefore, there is an urgent need for comprehensive and detailed information regarding land systems. Here, we developed the first fine-scale dataset of the China Land System at a spatial resolution of 1 km, covering the period from 2000 to 2015. By leveraging this comprehensive land information, we identified five primary types of land systems and their respective subsystems, thereby delineating distinct patterns of human-environmental interaction. Land system dynamics followed diverse developmental trajectories characterized by incremental shifts toward more functionally centralized systems. Land use intensification played a significant role in increasing the population capacity and food production in China, contributing nearly 93.94% and 84.99%, respectively. In contrast, land cover changes accounted for only 4.69% and 11.43%, respectively. These findings underscore the tendency of previous studies to overestimate the impact of land cover change and underestimate the influence of land use intensification in meeting the growing demands of land-based production. This study emphasizes the importance of transcending traditional land cover-based approaches and integrating land systems into land representation and global land change scenario simulations to promote sustainability.

Spatiotemporal assessment of the nexus between urban sprawl and land surface temperature as microclimatic effect: implications for urban planning

Rapid urbanisation has led to significant environmental and climatic changes worldwide, especially in urban heat islands where increased land surface temperature (LST) poses a major challenge to sustainable urban living. In the city of Abha in southwestern Saudi Arabia, a region experiencing rapid urban growth, the impact of such expansion on LST and the resulting microclimatic changes are still poorly understood. This study aims to explore the dynamics of urban sprawl and its direct impact on LST to provide important insights for urban planning and climate change mitigation strategies. Using the random forest (RF) algorithm optimised for land use and land cover (LULC) mapping, LULC models were derived that had an overall accuracy of 87.70%, 86.27% and 93.53% for 1990, 2000 and 2020, respectively. The mono-window algorithm facilitated the derivation of LST, while Markovian transition matrices and spatial linear regression models assessed LULC dynamics and LST trends. Notably, built-up areas grew from 69.40 km2 in 1990 to 338.74 km2 in 2020, while LST in urban areas showed a pronounced warming trend, with temperatures increasing from an average of 43.71 °C in 1990 to 50.46 °C in 2020. Six landscape fragmentation indices were then calculated for urban areas over three decades. The results show that the Largest Patch Index (LPI) increases from 22.78 in 1990 to 65.24 in 2020, and the number of patches (NP) escalates from 2,531 in 1990 to an impressive 10,710 in 2020. Further regression analyses highlighted the morphological changes in the cities and attributed almost 97% of the LST variability to these urban patch dynamics. In addition, water bodies showed a cooling trend with a temperature decrease from 33.76 °C in 2000 to 29.69 °C in 2020, suggesting an anthropogenic influence. The conclusion emphasises the urgent need for sustainable urban planning to counteract the warming trends associated with urban sprawl and promote climate resilience.

Remote sensing and GIS-based study of land use/cover dynamics, driving factors, and implications in southern Ethiopia, with special reference to the Legabora watershed

This paper investigates the trends, drivers, and consequences of LULC changes in Legabora watershed, Ethiopia, by utilizing remote sensing and geographic systems. Landsat Maltispectiral scanner (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper (ETM+), and Operational Land Imager (OLI) images of years 1976, 1991, 2001, and 2022, respectively, were used to study the dynamics of LULC. Essential image pre-processing steps were carefully carried out to correct distortions caused by sensor limitations. Eight main LULC categories were identified based on supervised image categorization methods and the maximum likelihood classification algorithm.The findings of change detection and cross-tabulation matrix demonstrate that there has been a significant increase in the area of cropland 345.1 ha/year, settlement 5.9 ha/year, forest 38.2 ha/year, and degraded lands 2.56 ha/year, respectively, over the period between 1976 and 2022. In contrast, considerable decreases were observed in grasslands (-248 ha/year) and shrublands (-144 ha/year), whereas other LULC categories augmented. The results revealed that the overall accuracy rates stood at 88.3 %, 88.4 %, and 85.6 % for 1976, 1991, and 2022, respectively. The overall kappa coefficient demonstrated values of 0.86 %, 0.86 %, and 0.83 % for the same period. Surveyed respondents perceived population growth, settlement, agricultural expansion, and infrastructure development as the most noticeable drivers of these LULC changes. In contrast, deforestation, land degradation, lack of livestock fodder, and biodiversity loss were identified as the main consequences of LULC changes. The factors and implications addressed in this study may be helpful tool for the formulation and implementation of evidence-based land use policies and strategies within in the study area and elsewhere.

Spatio-temporal dynamics of land use transitions associated with human activities over Eurasian Steppe: Evidence from improved residual analysis

We presented a framework to evaluate the land use transformations over the Eurasian Steppe (EUS) driven by human activities from 2000 to 2020. Framework involves three main components: (1) evaluate the spatial-temporal dynamics of land use transitions by utilizing the land change modeler (LCM) and remote sensing data; (2) quantifying the individual contributions of climate change and human activities using improved residual trend analysis (IRTA) and pixel-based partial correlation coefficient (PCC); and (3) quantifying the contributions of land use transitions to Leaf Area Index Intensity (LAII) by using the linear regression. Research findings indicate an increase in cropland (+1.17 % = 104,217 km2) over EUS, while a - 0.80 % reduction over Uzbekistan and - 0.16 % over Tajikistan. From 2000 to 2020 a slight increase in grassland was observed over the EUS region by 0.05 %. The detailed findings confirm an increase (0.24 % = 21,248.62 km2) of grassland over the 1st half (2000-2010) and a decrease (-0.19 % = -16,490.50 km2) in the 2nd period (2011-2020), with a notable decline over Kazakhstan (-0.54 % = 13,690 km2), Tajikistan (-0.18 % = 1483 km2), and Volgograd (-0.79 % = 4346 km2). Area of surface water bodies has declined with an alarming rate over Kazakhstan (-0.40 % = 10,261 km2) and Uzbekistan (-2.22 % = 8943 km2). Additionally, dominant contributions of human activities to induced LULC transitions were observed over the Chinese region, Mongolia, Uzbekistan, and Volgograd regions, with approximately 87 %, 83 %, 92 %, and 47 %, respectively, causing effective transitions to 12,997 km2 of cropland, 24,645 km2 of grassland, 16,763 km2 of sparse vegetation in China, and 12,731.2 km2 to grassland and 15,356.1 km2 to sparse vegetation in Mongolia. Kazakhstan had mixed climate-human impact with human-driven transitions of 48,568 km2 of bare land to sparse vegetation, 27,741 km2 to grassland, and 49,789 km2 to cropland on the eastern sides. Southern regions near Uzbekistan had climatic dominancy, and 8472 km2 of water bodies turned into bare soil. LAII shows an increasing trend rate of 0.63 year-1, particularly over human-dominant regions. This study can guide knowledge of oscillations and reduce adverse impacts on ecosystems and their supply services.

The role of satellite remote sensing in mitigating and adapting to global climate change

Climate change has critical adverse impacts on human society and poses severe challenges to global sustainable development. Information on essential climate variables (ECVs) that reflects the substantial changes that have occurred on Earth is critical for assessing the influence of climate change. Satellite remote sensing (SRS) technology has led to a new era of observations and provides multiscale information on ECVs that is independent of in situ measurements and model simulations. This enhances our understanding of climate change from space and supports policy-making in combating climate change. However, it remains challenging to remotely retrieve ECVs due to the complexity of the climate system. We provide an update on the studies on the role of SRS in climate change research, specifically in monitoring and quantifying ECVs in the atmosphere (greenhouse gases, clouds and aerosols), ocean (sea surface temperature, sea ice melt and sea level rise, ocean currents and mesoscale eddies, phytoplankton and ocean productivity), and terrestrial ecosystems (land use and land cover change and carbon flux, water resource and hydrological hazards, solar-induced chlorophyll fluorescence and terrestrial gross primary production). The benefits and challenges of applying SRS in climate change studies are also examined and discussed. This work will help us apply SRS and recommend future SRS studies to mitigate and adapt to global climate change.

Modeling Spatiotemporal Patterns of Land Use/Land Cover Change in Central Malawi Using a Neural Network Model

We examine Land Use Land Cover Change (LULCC) in the Dedza and Ntcheu districts of Central Malawi and model anthropogenic and environmental drivers. We present an integrative approach to understanding heterogenous landscape interactions and short- to long-term shocks and how they inform future land management and policy in Malawi. Landsat 30-m satellite imagery for 2001, 2009, and 2019 was used to identify and quantify LULCC outcomes based on eight input classes: agriculture, built-up areas, barren, water, wetlands, forest-mixed vegetation, shrub-woodland, and other. A Multilayer Perceptron (MLP) neural network was developed to examine land-cover transitions based on the drivers; elevation, slope, soil texture, population density and distance from roads and rivers. Agriculture is projected to dominate the landscape by 2050. Dedza has a higher probability of future land conversion to agriculture (0.45 to 0.70) than Ntcheu (0.30 to 0.45). These findings suggest that future land management initiatives should focus on spatiotemporal patterns in land cover and develop multidimensional policies that promote land conservation in the local context.

Spatial Simulation of Land-Use Development of Feixi County, China, Based on Optimized Productive–Living–Ecological Functions

Rural revitalization places higher demands on the productive–living–ecological (P-L-E) spaces of towns and cities. It is necessary, therefore, to identify, evaluate, and optimize P-L-E spaces to better guide spatial planning. Existing studies typically evaluate a single space, lacking a comprehensive consideration of whole-area integration. This study, therefore, developed a coupled spatial/developmental suitability evaluation system for Feixi County, Anhui Province, China, combining spatial quality evaluation, a coupled coordination model, and future land-use simulation (FLUS) model. The spatial quality of Feixi County in 2010, 2015, and 2020 was obtained by applying the evaluation system to the spatial development pattern. The results were analyzed and verified using the landscape pattern index and development suitability evaluation. The results showed the following: (1) The coupling coordination degree of the region increased from 0.131 to 0.372, changing from low to moderate coordination. (2) Based on the FLUS model to better capture the uncertainty and stochastic basis of the development in the study area. The kappa coefficient and Figure of Merit (FoM) index of the land-use simulation accuracy verification index were 0.7647 and 0.0508, respectively, and the logistic regression ROC values were above 0.75, thus meeting accuracy requirements. This demonstrated that the simulation model—based on a factor library of the evaluation of resource and environmental carrying capacity and suitability for development and construction—could better reflect future land-use changes. (3) The simulation showed that under the baseline development scenario, the area’s spatial layout is too concentrated in terms of construction land, ignoring P-L-E coordination. Under the ecological optimization scenario, high-quality ecological space is ensured, but other types of spaces are lacking. Under the comprehensive guidance scenario, lagging ecological space is optimized and P-L-E spatial development is enhanced through aggregation, clustering, concentration and integration. This way, the spatial quantity structure and distribution form can meet P-L-E spatial development needs in Feixi County. In this study, on the basis of scientific assessment of the current P-L-E space, the FLUS model was applied to carry out a scenario simulation according to different objectives. Moreover, based on the construction of the coupling system of human–nature system, the driving factors were improved to enhance the prediction accuracy of the FLUS model. This study’s findings can help improve the scientificity, flexibility and management efficiency of Feixi County’s P-L-E spatial layout, thereby supporting its sustainable development.