Land Cover Change Research Articles

Quantifying the land use land cover change and its effect on sediment yield in upper watersheds of Bilate River, Ethiopia

ABSTRACT The effects of land use land cover (LULC) changes on sediment yield (SY) are crucial for downstream river ecology. Understanding LULC change rates and impacts on SY in watersheds is essential for water management. This study assessed LULC dynamics and their influence on SY in the Upper Bilate Watershed (UBW), Ethiopia. Using supervised classification for Landsat images from 1992, 2002, 2012, and 2022, we estimated LULC changes. The Soil and Water Assessment Tool (SWAT) and partial least squares regression (PLSR) simulated LULC effects on sediment generation. Results indicated a 56% impact on the study area from 1992 to 2022. Agriculture and settlements increased by 384.8 and 76.6 km2, while wetlands and grasslands decreased by 212.7 and 107.7 km2. Major conversions were from forestland and wetlands to agriculture. SY effects were most pronounced at the watershed outlet and varied significantly within sub-watersheds. Four sub-watersheds were the highest contributors to SY in the 2022 LULC classification scenario. Increased agriculture and settlements, coupled with reduced wetlands, forests, and grasslands, were key SY influencers. The PLSR model highlighted agriculture, wetlands, and forests as dominant LULC classes affecting SY. These findings underscore the need for LULC-based watershed management to prevent wetland degradation and sediment accumulation in Lake Abaya.

MO-SAM: Testing the reliability and limits of mine feature delineation using Segment Anything Model to democratize mine observation and research

The purpose of this paper is to leverage the growth of AI-enabled tools to support the democratization of mine observation (MO) research. Mining is essential to meet projected demand for renewable energy technologies crucial to global climate mitigation objectives, but all mining activities pose local and regional challenges to environmental sustainability. Such challenges can be mitigated by good governance, but unequal access among stakeholders to accurately interpreted satellite imagery can weaken good governance. Using readily available software—QGIS, and Segment Anything Model (SAM)—this paper develops and tests the reliability of MO-SAM, a new method to identify and delineate features within the spatially-explicit mine extent at a high level of detail. It focuses on dry tailings, waste dumps, and stockpiles in above-ground mining areas. While we intend for MO-SAM to be used generally, this study tested it on mining areas for energy-critical materials: lithium (Li), cobalt (Co), rare earth elements (REE), and platinum group elements (PGE), selected for their importance to the global transition to renewable energy. MO-SAM demonstrates generalizability through prompt engineering, but performance limitations were observed in imagery with complex mining landscape scenarios, including spatial variations in image morphology and boundary sharpness. Our analysis provides data-driven insights to support advances in the use of MO-SAM for analyzing and monitoring large-scale mining activities with greater speed than methods that rely on manual delineation, and with greater precision than practices that focus primarily on changes in the spatially-explicit mine extent. It also provides insights into the importance of multidisciplinary human expertise in designing processes for and assessing the accuracy of AI-assisted remote sensing image segmentation as well as in evaluating the significance of the land use and land cover changes identified. This has widespread potential to advance the multidisciplinary application of AI for scientific and public interest, particularly in research on global scale human-environment interactions such as industrial mining activities. This is methodologically significant because the potential and limitations of using large pre-trained image segmentation models such as SAM for analyzing remote sensing data is an emergent and underexplored issue. The results can help advance the utilization of large pre-trained segmentation models for remote sensing imagery analysis to support sustainability research and policy.

Assessing land degradation and drought vulnerability in the Trans-Gangetic Plains using Google Earth Engine and remote sensing for SDG 15.3.1 monitoring.

The Trans-Gangetic Plains (TGP) is a distinct region in northwestern India encompassing the states of Punjab, Haryana, the union territories of Delhi and Chandigarh, and the Ganganagar district of Rajasthan. Although the region plays a critical role in India's food security through substantial wheat and rice production, it is increasingly threatened by land degradation and drought vulnerability. Existing studies often lack integrated, large-scale assessments combining land degradation and drought analysis in this region. Addressing this gap, the present study utilizes Google Earth Engine and remote sensing-based SDG 15.3.1 indicators to evaluate land productivity, soil organic carbon loss, land cover change, and drought susceptibility across the TGP. The assessment reveals a 214% surge in artificial land cover between 2001 and 2022, primarily at the expense of grasslands, wetlands, and croplands, leading to declines in biodiversity, carbon sequestration, and soil water retention. Additionally, 2.93% of the area showed soil organic carbon degradation, further stressing ecosystem health. Despite relative stability in agricultural productivity, recurrent moderate to severe droughts affected over 75% of the region in peak years (e.g., 2002, 2009, 2014), largely driven by El Niño events and groundwater overextraction. The findings highlight that the western and central TGP regions are particularly vulnerable and emphasize the need for integrated site-specific soil and water conservation strategies. This study demonstrates the effectiveness of open-source remote sensing platforms for large-scale degradation monitoring and suggests that integrating climate-smart agriculture and afforestation programs will be vital to enhancing resilience and achieving long-term sustainability.

The biogeophysical impacts of land cover change on climate extremes in the Arctic and Boreal regions

The biogeophysical impacts of land cover change on climate extremes in the Arctic and Boreal regions

Coastal carbon at risk: forecasting the impacts of sea-level rise on future land cover

IntroductionCoastal land cover (LC) is in constant flux and shaped by human activity and natural forces. These shifts have profound implications for climate resilience, as LC change can either enhance or diminish the landscape’s capacity to store and sequester carbon.MethodsThis study investigates the impact of sea-level rise (SLR) on carbon storage and sequestration within the coastal Superfund and industrially contaminated areas of Aberdeen Proving Ground (APG) and its adjacent environment, located in the northern Chesapeake Bay, Maryland. Leveraging the MOLUSCE plugin in QGIS and the InVEST model, this study integrates historical LC data with predictive modeling techniques, including artificial neural networks, multi-layer perceptron, and Cellular Automata.ResultsProjections for 2061 reveal that, under a no-SLR scenario and non-submerged aquatic vegetation (SAV) scenario, APG retains 4,059,312 Mg C in storage, losing -54,087 Mg C sequestration and -$42.06 million net present value (NPV). The NPV is changed to -$40.57 million for the Low SAV scenario and -$38.86 million for the High SAV scenario for 2061 under the no-SLR scenario. However, with SLR, storage declines to 3,894,892 Mg C, and sequestration losses escalate to -218,505.75 Mg C, representing -$169.93 million NPV for the non-SAV scenario. The amount of NPV is changed to -168.44 million and -$166.73 million for the Low and High SAV scenarios.DiscussionThese findings underscore the accelerating carbon debt imposed by SLR and the urgent need for adaptive strategies. Coastal preservation techniques, such as living shorelines and thin-layer placement, have emerged as critical strategies for mitigating carbon losses and enhancing resilience. By quantifying the ecological and economic consequences of SLR-driven LC change, this study advances the understanding of carbon dynamics in vulnerable coastal landscapes and reinforces the necessity of proactive management to sustain their climate-regulating functions.

Modeling land cover changes using an enhanced Markov-future land use simulation model with spatial distribution considerations: a case study in the Yellow River Basin

ABSTRACT The traditional Markov-future land use simulation (FLUS) model for land use prediction primarily emphasizes the quantity changes and spatial distribution of land use types, but it neglects the influence of their inherent spatial characteristics on the prediction precision. This study introduces an innovative approach by designing shape control parameters and developing an enhanced Markov-FLUS model. The model integrates artificial neural networks to capture the relationships between the land use type occurrence probability and driving factors. It incorporates common points, common edges, distance, and aggregation parameters alongside a cellular automata model. Taking the Yellow River Basin as an example for analysis, this study compares the model's simulation performance before and after enhancement, focusing on the land use types with the most and least improvement. The results indicate that the refined model achieves a superior fitness function value in simulating land use within the Yellow River Basin. HIGHLIGHTS The Enhanced Markov-FLUS model has been meticulously constructed through the strategic design of a comprehensive set of shape control parameters. The refined model demonstrates significantly elevated fitness values, particularly within a 3 × 3 Moore neighborhood framework. This model stands out in accurately forecasting land use types characterized by enhanced contiguity and more regularized shapes. The improved model significantly diminishes the miss rate, notably along the edges of land type boundaries.

Understanding encroachment typologies through remote sensing and socio-economic analysis: enhancing national park management in Kerinci Seblat National Park, Indonesia.

Encroachment remains a persistent challenge for Kerinci Seblat National Park (KSNP), despite its designation as a protected area. While agricultural use by local communities continues, limited understanding of the actors and their motivations hampers effective policy responses. This study addresses this gap by integrating remote sensing with socio-economic analysis to examine encroacher typologies and their spatial dynamics. Land use and land cover (LULC) changes were detected using Landsat 5, 7, and 8 imagery across six time periods (1988-2022), analyzed through maximum likelihood classification. Field surveys were conducted with 206 households, alongside in-depth interviews with customary leaders, village heads, and KSNP. We found an increasing trend of encroachment since the initial identification of the KSNP to the present, which correlates with the expansion of agricultural land. We grouped the typologies of encroachers into indigenous landless (23%), indigenous people with economic opportunities (29%), sly opportunists (2%), indigenous people as investors (3%), workers/profit-sharing partners (42%), and local migrants (1%). The dominant typology was workers/profit-sharing partners, which indicates that this partnership has a broad influence and wide coverage. Grouping actors supports the implementation of programs according to their motives and characteristics. The solution to encroachment should include a livelihood improvement program for indigenous people without land ownership, the establishment of utilization (e.g., agroforestry areas) and buffer zones, and enhanced law enforcement for other typologies.

HRMS-SCD:A High-Resolution Multi-Scene Satellite Imagery Dataset for Comprehensive Land-Cover Semantic Change Detection

Abstract. Semantic change detection (SCD) focuses on identifying changes in surface coverage while simultaneously classifying the types of changes. This approach provides detailed information valuable for urban planning, environmental monitoring, and other applications, making it a key area of interest in remote sensing research. Despite recent advances, existing SCD studies are hindered by the lack of high-resolution satellite imagery datasets and insufficiently comprehensive semantic label coverage in publicly available datasets. To address these limitations, we have developed a large-scale high-resolution remote sensing dataset consisting of 11,587 satellite image pairs, each with 1-meter spatial resolution and a size of 512 × 512 pixels, representing land cover changes across Beijing between 2017 and 2018. This dataset encompasses diverse land surface scenes with comprehensive semantic annotations. Furthermore, it includes full-coverage semantic segmentation labels from pre-change phases and a larger sample size of 2048 × 2048 pixels to support future research on multi-class and large-format change detection. We benchmark eight state-of-the-art SCD algorithms using this dataset, providing critical performance metrics that serve as valuable references for subsequent research. This dataset not only addresses existing gaps but also establishes a robust foundation for advancing deep learning-based semantic change detection, enabling more accurate and comprehensive analysis of complex and diverse land cover changes. More information about the project can be found at https://github.com/17x-osborn/HRMS-SCD.

Seasonal variability in hydrologic responses due to climate and land-cover changes in the Pothohar plateau, Pakistan

ABSTRACT The research uses hydrological modeling to evaluate the impacts of climate and land-use changes on the Soan Basin, crucial for water resources management. The HEC-HMS model predicted the SRB's hydrological response to current and future rainfall. Post-calibration, the model analyzed the effects of climate and landcover changes on hydrological response using data from six meteorological gauges. The basin was divided into four subbasins with distinct characteristics. Model's calibration, from 2010 to 2015, and validation, from 2016 to 2018, showed an ideal fit between expected and actual outflows. The MPI-ESM1-2-HR GCM was selected to estimate future rainfall and temperature under the SSP2 and SSP5 scenarios. Projections indicated increased rainfall, with SSP2 forecasting a 17% increase and SSP5 a 25% increase by the century's end. The projected streamflow (2019–2100) is also expected to increase under SSP2 and SSP5 scenarios compared to the baseline period (1990–2018). This study can aid adaptive water management in the Soan Basin by addressing projected increases in climate and land-use variability. It can also support policymakers in crafting strategies for climate resilience and guiding sustainable land development planning.

Community based forest management enhances forest condition of village forests within the eastern Afromontane biodiversity hotspot

Despite the advent of Community Based Forest Management (CBFM) more than three decades ago our understanding of its impacts on forest condition is limited. Here we test whether CBFM enhances forest condition of two (Litwang’ata and Intake) Village Land Forest Reserves (VLFRs) within the eastern Afromontane biodiversity hotspot. Remote Sensing and GIS methods triangulated with Key Informant Interviews (KIIs) and ground truthing were employed to assess the spatial–temporal land cover changes before (i.e. 2019) and after (i.e. 2024) the introduction of CBFM. Results show that agricultural land decreased in Intake from 26 hectares (ha) to 1 ha corresponding to the decrease of 95.2% whereas it decreased from 3 to 2 ha equivalent to the decrease of − 55.5% in Litwang’ata forest post CBFM introduction. Open woodland decreased for the two VLFRs at the expense of closed woodland for Intake and at the cost of grass/shrub land for Litwang’ata. Closed woodland increased by 1318 ha equivalent to the cover change of 39.8% for Intake forest whereas it slightly decreased by 38 ha corresponding to the cover change of − 3.7% five years post CBFM inception. Overall, agricultural land, grass/shrub land and open woodland decreased with absolute land covers of 26 ha, 199 ha, 1054 ha, respectively. On the other hand, closed woodland overall increased with absolute land cover of 1280 ha. Change in forest governance by devolving ownership and management to local communities has the potential to enhance the recovery and protection of village forests within the eastern Afromontane biodiversity hotspot.

Accounting for alternation in temporal quality analysis in MapBiomas Brazil

ABSTRACT Land use and land cover maps are an important resource for understanding interactions between humans and their environment across space and time with current mapping efforts often spanning upwards of 20 years. We present here a new method for a robust assessment of land cover transitions over time and apply this methodology to the yearly MapBiomas land use land cover maps of Brazil spanning 1985–2022. Based on a reference sample of 85,152 points, we find MapBiomas to have limited accuracy as an indicator of yearly land use change, but consistent over the full mapping period. Alternation, a newly defined error component, captures the number of land use transitions a location experiences throughout time. It is the primary reason for differences in estimates of annual change and is 4.6 times more frequent in the MapBiomas product than reference data. Differences in alternations are particularly prevalent in transitions from pasture to savanna and forest classes. The total land use changes detected over the 37 year study period are consistent between the reference data and the MapBiomas classification with 232 million hectares and 252 million hectares, or 27% and 29% of the Brazilian territory respectively. HIGHLIGHTS 1. We present a new method, Alternation, to measure quality in land cover transitions over time. 2. The methodology is applied to MapBiomas land cover maps of Brazil from 1985 to 2022. 3. Land cover change is consistent across the full time series, but inconsistent at the annual scale. 4. MapBiomas has 4.6 times more annual transitions than the reference data.

Soybean, a Commodity without Borders: Socio-Environmental Impacts in Protected Areas and Indigenous Communities in Mato Grosso do Sul/Brazil

The occupation of the cone-south region of the state of Mato Grosso do Sul is not exclusively territorial, it is successively temporal and depends on public and private social actors, who build, elaborate, and provide conditions for the existence of the dynamics that are established there. The economic spatial expansion, especially for agricultural and cattle raising activities, attracts capital to change land use, intensifying the pressure on more vulnerable areas, such as conservation units and indigenous lands. The main change in land use and cover observed in the last 30 years was the loss of native vegetation inside conservation units (-29%) and indigenous lands (-34%) and the conversion of pasture areas, which had a reduction of 76% and 38%, to plant soy, which increased 536% and 98% inside indigenous lands and conservation units, respectively. Thus, it is clear that delimiting a conservation area or circumscribing traditional populations in a small area without resources prevents them from reproducing their way of life and, paradoxically, induces them to develop predatory practices against the environment or to lease their lands to large landowners, as the only means of guaranteeing their subsistence and not falling into poverty. This study highlights the urgent need to rethink territorial policies to ensure sustainable and socially just land use models.

Spatial tools for river management: capacity building through GIS

ABSTRACT The state of river ecosystems around the world is a serious cause for concern. This is primarily due to anthropogenic influences from water use, artificial structures, and land use or land cover changes in catchments. At the same time, climate change leads to gradual but notable shifts and intensification in all aspects of the water cycle, a trend which is projected to be continued in the future. Accordingly, river management comprises both preservative or reactive and adaptive or proactive measures to deal with climate-related issues, human-caused conflicts and their contemporaneity as well as the general framework of digitization and personnel. This study highlights, how a ‘spatial view’ on these challenges contributes to a broader understanding on different scales, e.g. of overlapping interests, mutually exclusive objectives, functional connectivity or cascading effects and, thus makes these challenges manageable. Therefore, a holistic water management approach requires spatial data and tools as part of the geoinformation system (GIS). In the second section of the study, examples of GIS tools, techniques, and applications are presented, explaining their potential to enhance efficient river management. We propose further training in this field as a measure for capacity building for the involved stakeholders.

Forecasting high-risk areas for dengue outbreaks in China: A trend analysis of Aedes albopictus and Aedes aegypti distributions from 2014 to 2030

BackgroundGlobal warming, urbanization and the resumption of global population movements post-COVID-19 have made the prevention and control of dengue and its vector transmission more challenging. To tackle this issue, this study evaluated and predicted the suitable habitats of Aedes albopictus and Aedes aegypti in China. This study aims to identify the key influencing factors, analyze patterns of habitat expansion and contraction, and explore regions in China at risk of dengue fever outbreaks in the future.Methodology/principal findingsThis study utilized mosquito distribution data from 2010 to 2023 and employed Maxent to map the distribution of suitable habitats. The key influencing factors and response curves were further analysed, and patterns of habitat expansion and contraction were investigated. The findings reveal that the main variables affecting the distributions of Ae. albopictus and Ae. aegypti are annual precipitation, annual mean temperature, and land use and land cover changes. The suitable habitat for Ae. albopictus shows a significant northward expansion trend, reaching large areas in Shandong and Henan province by 2030. The suitable habitat area for Ae. albopictus is increasing annually and can reach approximately 2.38 million square kilometers by 2030. Compared to the outbreak year of dengue fever in China in 2019, the suitable habitat area for Ae. albopictus in 2030 will increase by approximately 17.06%, with a growth of 2.57% in the sum of high-risk and medium-risk suitable habitat areas. In contrast, the suitable habitats of Ae. aegypti are primarily concentrated in Guangdong, Hainan and Yunnan Provinces.Conclusion/significanceThis study compared the potential changes in the distributions of suitable habitats for Ae. albopictus and Ae. aegypti in 2014, 2019, and 2023 and predicted suitable habitats for 2030, as well as contraction and expansion trends in the suitable habitats of Ae. albopictus. The findings aim to identify regions at risk of future dengue fever outbreaks in China, providing a scientific basis for public health authorities to develop effective dengue prevention and control strategies.

Modeling forest and rangeland ecosystem responses to drought across Hyrcanian bioclimatic zones of Iran using GLM and LAI analysis

Drought substantially affects ecosystem structure and function, shaping vegetation dynamics and influencing long-term environmental sustainability. This study examines drought effects on forest and rangeland ecosystems across three bioclimatic zones in the Hyrcanian region of Iran. MODIS-derived Leaf Area Index (LAI) data (2001–2022) and Standardized Precipitation Index (SPI) were used with a generalized linear model (GLM) to assess vegetation responses. The findings indicate that rangeland ecosystems, especially in Zones II and III, are susceptible to drought, with SPI accounting for over 80% of the observed LAI variability in these regions. Forests better withstand dry conditions, with SPI explaining about half of the changes in LAI Zone III, with high elevation and snow-dominated precipitation, is drought-sensitive. Zone I near the Caspian Sea has higher humidity and more stable conditions. Zone II, with a semi-humid cold climate, exhibits the largest LAI fluctuations due to its strong dependence on moisture. Elevation, vegetation type, and climate critically influence drought responses. Targeted land management, including water optimization and conservation, is essential. Future research should integrate additional factors such as soil moisture, land cover change, and anthropogenic pressures such as deforestation, overgrazing, and environmental degradation alongside predictive modeling to enhance ecological sustainability.

ALST-W integrated index for enhanced surface temperature mapping of water bodies and vegetation using Landsat 8/9 satellite bands

Researchers are developing new methods to analyze changes in satellite data across various locations using remote sensing and geographic information systems (GIS). Land Surface Temperature (LST) maps are important indices for understanding changes in global land use and land cover (LU/LC). This study introduces the ALST-W (Adaptive Land Surface Temperature of Water Bodies) index to investigate the impact of water bodies on the LST map of the non-forest-covered Javadi Hills region, India, using Landsat 9/8 images for 2020, 2022, and 2024. The ALST-W results were compared with reference maps from Google Earth Engine (GEE), and the findings showed a good average accuracy of 95.06%. This study introduces the new index of the ALST-W, which displays the temperature data for high and low vegetation, along with the water bodies in a single raster map. The information from this work helps communities and policymakers understand environmental changes and take informed actions to protect vegetation and water bodies from significant future loss.

Correction: Predicting changes in land use and land cover using remote sensing and land change modeler

Correction: Predicting changes in land use and land cover using remote sensing and land change modeler

Bi-decadal trends in land cover and ecological dynamics across the Korean Peninsula with implications for sustainable development

This study analyzes land cover and ecological changes on the Korean Peninsula over two decades (2001–2021), emphasizing the relationship between ecological diversity and key drivers such as population dynamics, land use changes, and climatic factors. Leveraging Earth observation data on the Google Earth Engine (GEE) platform and machine learning-based analysis enables a broader range of trend analyses. Key results include a 4% increase in urban areas, a 1.1% rise in forest coverage, and a 3.1% decrease in cropland. Population trends show a 7.44% growth in urban regions and a 1.22% decline in rural areas, while SDG indices reveal a 10.56% increase in the Urban Green Space Index in some areas. The innovative Ecological Characteristic Map (ECM) provides significant insights, highlighting the variability in ecological conditions and the need for region-specific conservation strategies. This work contributes to the discourse on sustainable development in the Korean Peninsula, fostering collaboration while laying the groundwork for peace-building initiatives and science diplomacy.

Indian leopards (Panthera pardus fusca) facing space struggles in a Metropolitan district of Northeast India.

Urbanization affects wildlife species, particularly in wildland-urban interfaces. In the rapidly urbanizing landscape of Assam in India, we explored the relationship between urban expansion and wildlife conservation, focusing on the Indian leopard (Panthera pardus fusca). We assessed the land use and land cover changes, human-leopard interactions, and habitat suitability analysis of leopard. We also estimated the relative leopard abundance based on suitable patches within the free space in Guwahati using human building dataset. Our analysis shows that built-up areas have expanded from 6.23% in 1989 to 16.99% in 2019, while natural habitats vital for leopard have substantially reduced from 11.39 to 0.33%, respectively. Our survey in buffer zones around key forest areas indicates a growing trend in interactions, leading to a decrease in free space vital for leopard abundance. We estimated that Guwahati city can currently support 14 leopards (38-8 SD) within the existing landscape, but increased urban development could reduce this number to 4 (10-2 SD). Our study highlights the challenges of wildlife conservation in urban landscapes and the need for strategies that balance urban development with biodiversity protection. Our findings align with Target 4 of the Kunming-Montreal Global Biodiversity Framework, which emphasizes managing human-wildlife interactions to reduce conflicts and foster coexistence by 2030.

Impact of Major Road Network on Landscape and Growth Pattern in Tehri Region: A Geospatial Technique

ABSTRACTThis study examines land use and land cover (LULC) changes in the Tehri region from 2000 to 2024, utilizing the Shannon Entropy Index (H) as an indicator of sustainable urban growth by analyzing spatial dispersion patterns of built‐up areas. Landsat imagery from 2000, 2010, 2020, and 2024 was classified into barren land, built‐up areas, forest areas, and water bodies using the Maximum Likelihood Classifier. In 2000, barren land dominated with 53.23%, followed by forests at 42.54%, built‐up areas at 3.75%, and water bodies at 0.48%. By 2024, forests expanded to 51.67%, barren land decreased to 33.16%, built‐up areas grew significantly to 14.55%, and water bodies increased slightly to 0.63%. The Shannon Entropy Index was calculated based on proximity to the city center and major road networks, dividing the study area into zones. Results reveal a rising normalized Shannon Entropy Index for the city center (HC′) and main road (HR′), with values of 0.978 and 0.58 in 2024, reflecting increased urban fragmentation. Zones 8 and 16 exhibited the highest spatial dispersion (0.96), indicating irregular development patterns and outward expansion. In 2000, the Tehri region was classified as type D. Over the years, it has mostly shifted to type B, showing more dispersed and possibly unsustainable growth. We advocate for a comprehensive space utilization control strategy that integrates the city center and the peri‐urban area. This research demonstrates the utility of the Shannon Entropy Index in identifying potential zones of unsustainable development in hilly regions and offers a replicable framework for other areas with similar topographical and developmental dynamics.