Land Use Cover Classification Research Articles

Agricultural land use and morphometry explain substantial variation in nutrient and ion concentrations in lakes across Canada

Declines in freshwater quality resulting from anthropogenic nutrient input remain a persistent issue worldwide. Yet, we still have a limited understanding of the magnitude and scale at which most lakes have been affected by human activities, namely Land Use/Land Cover (LULC) alterations. In response, the NSERC Canadian Lake Pulse Network has compiled the first nationwide systematic database of lake quality metrics by surveying 664 lakes across 12 ecozones over 3 years. To assess the influence of catchment development on water quality and its spatial variation, we built generally additive models and multivariate regressions to quantify the association between watershed LULC and lake temperature, Secchi depth, as well as chlorophyll- a, limiting nutrient, and ion concentrations. We found that agricultural and urban land use explained the greatest proportion of variation in water quality among LULC categories ( R2 = 0.20–0.29). Overall, our study highlights that drivers of water quality are similar across regions; however, baseline conditions vary, so freshwater ecosystem management strategies must consider their geographic context to better predict where water quality thresholds will be surpassed.

Urban land use land cover classification based on GF-6 satellite imagery and multi-feature optimization

Urban land use/land cover (LULC) classification has long been a hotspot for remote sensing applications. With high spatio-temporal resolution and multispectral, the recently launched GF-6 satellite provides ideal open imagery for LULC mapping. In this study, we utilized multitemporal GF-6 images to generate six types of land features, including spectral bands, texture features, built-up, waterbody, vegetation, and red-edge indices. The minimum Redundancy Maximum Relevance (mRMR) algorithm was employed to optimize feature selection. Subsequently, Random Forest (RF) and Extreme Gradient Boosting (XGBT) were assessed using different feature selections. Besides, various feature configurations were designed for LULC classification and comparison. The results indicate that the mRMR-based RF method achieved the highest overall accuracy of 91.37%. The temporal red-edge indices were important features for urban LULC classification and contributed mainly to grassland and cropland. These results supplement existing classification methods and assist in improving LULC mapping in urban areas with complex landscapes.

Implications of future land-use/cover pattern change on landslide susceptibility at a national level: A scenario-based analysis in Romania

Land-use/cover (LUC) change is a global change forcing with important impacts on landslides, which highlights the need to evaluate the potential future evolution of these hazards under future LUC scenarios. Accordingly, the current paper provides a first national-level exploration on the possible effects of LUC pattern change on landslide susceptibility (LS) in Romania, shedding light on the evolution of one of Europe’s landslide hotspot countries as well as one of the territories significantly affected by recent LUC change, especially after the fall of the communist regime. Two countrywide future LS maps were modelled for 2075, by applying a previously developed national scale-adapted LS empirical procedure, integrating therein two existing long-term LUC pattern forecasts and assuming no change for the other landslide controlling factors. The estimated future LS maps were then compared to the current LS zonation using a pixel-based analysis to estimate the contribution of LUC class transitions to the LS variation in time. The quantitative examination of LS changes suggests an overall net slight decrease at the national level under both scenarios (on an average of 7.3% of the LUC change area), with more significant regional variations. The location and extent of the estimated potential LUC transitions reflect themselves into a clear influence on the future LS level manifesting itself, on average, over 7.1% of the country’s area. The analysis demonstrates that the expected increase in the forest-covered area brings the most important contribution to the decrease of the LS level. An average of 44.2% of the total LS decrease area was mainly predicted by the transitions from pastures and natural grasslands/scrubs to forests, and 11.4% by that from heterogeneous agricultural areas to forests. Conversely, transitions to pastures and natural grasslands resulted in the most important contribution to the increase of the LS level (52.2%). Transitions among other agricultural lands resulted in slight susceptibility variations or in no change. Through the analysed LUC scenarios, differing in terms of the implementation of environmental management regulations inside protected areas, the study proves that a more appropriate land management could have an important influence on decreasing LS, a finding of particular relevance for encouraging and extending a conscious application of protective land measures, especially on slopes which are degraded or prone to instability, in a country otherwise characterised by an overall relative implementation of environmental policies. In addition, the analysis reveals that a significant proportion (24.2%) of the projected new built-up areas is expected to spread by 2075 over lands with moderate, high and very high LS, resulting in potentially new landslide exposure hotspots in the future in both rural and urban areas. The obtained results could be used in directing landslide risk reduction actions and decisions primarily towards the management of LUC and towards prioritizing areas for such measures, while, from a theoretical perspective, this approach could be extended to future investigations related to LUC change implications for landslide hazard and risk in Romania, but could also be followed in case of other national territories.

Spatio-temporal Variation Analysis of Thermal Climate in Response to Urban Cover Changes: A Case Study of Delhi Using Landsat Data

ABSTRACT Urbanization and changes in Land Use Land Cover (LULC) influence the formation of urban microclimate (UMC). The current study analysed the urbanization and microclimatic changes in the National Capital Territory (NCT) of Delhi utilizing the Landsat-derived LULC map, Land Surface Temperature (LST), and a variety of spectral indices viz., Heat Effect Contribution Index (HECI), Surface Urban Heat Island (SUHI), Thermal Index (TI) and Environment Criticality Index (ECI). The results indicate that Delhi has undergone a substantial increase in urbanization and concomitant urban warming (1999–2021). The mean LST had risen across all LULC classes, with vegetation exhibiting the most significant variance in mean LST. Largely, water bodies and vegetation lands converted to other LULC classes yielded the highest rise in LST. Notably, the contribution of the build-up to the heat effect rose by more than 40%. A rise in LST and its association with different LULC classes led to the widespread development and spatiotemporal shifting of SUHI and a subsequent decline in thermal comfort. This research uniquely quantifies LST changes associated with shifts in LULC categories and utilizes multiple heat indices to study the UMC of Delhi, offering crucial insights for future planning and policy formulation.

The effects of habitat loss and fragmentation on the relative abundance and conservation of Ludwig’s Bustard Neotis ludwigii in South Africa

Ludwig’s Bustard Neotis ludwigii is near-endemic to the semi-arid shrublands of southwestern southern Africa and is listed as Endangered. The primary threat to this species is collisions with overhead powerlines; however, loss and fragmentation of its habitat may contribute to its population decline. The 1990 and 2020 land-use land-cover (LULC) maps prepared for South Africa were used to determine the LULC categories that best describe suitable habitat for Ludwig’s Bustard using beta regressions and data on exact localities, as recorded in the real-time BirdLasser logging app. Optimized Hot Spot Analysis was used to determine the priority areas for conserving Ludwig’s Bustard and its habitat. Beta regression models were compiled using the reporting rate for this bird and the total surface area of suitable habitat, and three other landscape metrics, calculated using the per pentad LULC categories considered suitable habitat for the bird. These analyses revealed that, between the first and second Southern African Bird Atlas Projects, the distribution range of Ludwig’s Bustard increased by 153 quarter-degree grid cells, whereas the reporting rate for the species declined by 5.6%, along with a small net decline in its quarterdegree-grid-cell-based area of occupancy, its pentad-based area of occupancy and habitat. The habitat of Ludwig’s Bustard is dominated by shrubland with limited grassland, in primarily the Nama Karoo and Succulent Karoo biomes, and the Grassy Karoo Biome in the eastern part of its distribution. In the relative abundance hot spot of Ludwig’s Bustard, there does not currently appear to be any significant loss or fragmentation (distance between patches) of its habitat. A landscape-based approach to conserving Ludwig’s Bustard can be achieved by expanding the current network of protected areas and conserving critical biodiversity areas and ecosystem support areas, especially in the bird’s relative abundance hot spots in the Nama Karoo and Succulent Karoo biomes.

Spatial and temporal changes of land use land cover distribution in selected sites of the southern coastal zone of Sri Lanka

The southern coastal zone of Sri Lanka has been subjected to a variety of natural and anthropogenic influences over the past three decades. Understanding impacts of such influences on Land Use Land Cover (LULC) is vital for proper management of the coastal zone. This study, therefore, focused on mapping the change/s in the distribution of selected LULC in the selected sites, Bundala, Galle, Kalametiya, and Hambantota of the southern coastal zone, over last 24 years using satellite imagery. LULC maps of nine classes (mangrove, inland vegetation, marsh and grass, sand, bare land, built-up, cultivation, water, and unclassified) were created by employing an on-screen digitization technique achieving an accuracy of >70%. Trend analysis and two-sample proportion tests were used for statistical analyses, whereas geometric calculations were used for descriptive analyses. The results showed the highest overall change in LULC in Kalametiya, followed by Hambantota, Galle, and Bundala. The changes in the LULC classes are mainly attributed to the conversion of water to mangroves, marsh and grass areas in Kalametiya, water to built-up areas in Hambantota, cultivations to built-up areas in Galle, and marsh and grass areas to bare lands in Bundala. The causes of LULC changes were site specific. Trend analyses indicate the least LULC changes in Bundala possibly by 2025. The study highlights the significance of taking into account geographical dislocations when considering and anticipating the potential impacts of development projects over broader extents.

Quantifying uncertainty in land-use land-cover classification using conformal statistics

Land-use land-cover (LULC) change is one of the most important anthropogenic threats to biodiversity and ecosystems integrity. As a result, the systematic generation of annual regional, national, and global LULC map products derived from the classification of satellite imagery data have become critical inputs for multiple scientific disciplines. The importance of quantifying pixel-level uncertainty to improve the robustness of downstream analyses has long been acknowledged but this practice is still not widely adopted in the generation of these LULC products. The lack of uncertainty quantification is likely due to the fact that most approaches that have been put forward for this task are too computationally intensive for large-scale analysis (e.g., bootstrapping). In this article, we describe how conformal statistics can be used to quantify pixel-level uncertainty in a way that is not computationally intensive, is statistically rigorous despite relying on few assumptions, and can be used together with any classification algorithm that produces class probabilities. Our simulation results show how the size of the predictive sets created by conformal statistics can be used as an indicator of classification uncertainty at the pixel level. Our analysis based on data from the Brazilian Amazon reveals that both forest and water have high certainty whereas pasture and the “natural (other)” category have substantial uncertainty. This information can guide additional ground-truth data collection and the resulting raster combining the LULC classification with the uncertainty results can be used to communicate in a transparent way to downstream users which classified pixels have high or low uncertainty. Given the importance of systematic LULC maps and uncertainty quantification, we believe that this approach will find wide use in the remote sensing community.

Spatial Simulation and Prediction of Land Use/Land Cover in the Transnational Ili-Balkhash Basin

Exploring the future trends of land use/land cover (LULC) changes is significant for the sustainable development of a region. The simulation and prediction of LULC in a large-scale basin in an arid zone can help the future land management planning and rational allocation of resources in this ecologically fragile region. Using the whole Ili-Balkhash Basin as the study area, the patch-generating land use simulation (PLUS) model and a combination of PLUS and Markov predictions (PLUS–Markov) were used to simulate and predict land use in 2020 based on the assessment of the accuracy of LULC classification in the global dataset. The accuracy of simulations and predictions using the model were measured for LULC data covering different time periods. Model settings with better simulation results were selected for simulating and predicting possible future land use conditions in the basin. The future predictions for 2025 and 2030, which are based on historical land change characteristics, indicate that the overall future spatial pattern of LULC in the basin remains relatively stable in general without the influence of other external factors. Over the time scale of the future five years, the expansion of croplands and barren areas in the basin primarily stems from the loss of grasslands. Approximately 48% of the converted grassland areas are transformed into croplands, while around 40% are converted into barren areas. In the longer time scale of the future decade, the conversion of grasslands to croplands in the basin is also evident. However, the expansion phenomenon of urban and built-up lands at the expense of croplands is more significant, with approximately 774.2 km2 of croplands developing into urban and built-up lands. This work provides an effective new approach for simulating and predicting LULC in data-deficient basins at a large scale in arid regions, thereby establishing a foundation for future research on the impact of human activities on basin hydrology and related studies.

An Assessment of Land Use and Land Cover Change for Lubovane Reservoir Sub-Catchment in Eswatini

The increasing population in rural areas, increased livestock densities and extensive deforestation have been reported as the main drivers of land degradation in Eswatini. Land degradation, along with biodiversity loss and climate change presents serious challenges to the environment, economy and the country’s development agenda.  This study was conducted to assess the land use land cover (LULC) changes within the Lubovane reservoir catchment. Landsat 4-5 TM images were used for mapping LULC changes for 1995, 2000, 2005, 2009 and a Landsat 8 image was used for mapping 2015 LULC. A supervised LULC classification was conducted using 6 classes (water, settlements, irrigation, cultivation, shrubs and forests, as well as bare land) in ArcGIS version 10.3.1. The classification was validated using a confusion matrix and the results reflected that water, irrigation, cultivation, forests and shrubs were well classified. The LULC assessment results indicated that there was low coverage of water bodies observed from 1996 – 2005, while a 3% increase was observed in 2009. Water coverage decreased to 1.9% in 2015 due to the El-Niño induced drought that hit Southern Africa, resulting in low inflow to the dam. A reduction of shrubs and forest cover was experienced in 2000 due to conversion of forested areas into settlements for resettled households. However, a slight increment of shrubs and forest was observed from 2009 to 2015. A reduction in the concentration of forests cover around the reservoir, an increase of settlements and bare land were also observed.

Polarimetric Decomposition and Machine Learning‐Based Classification of L‐ and S‐Band Airborne SAR (LS‐ASAR) Data

AbstractThe polarimetric Synthetic Aperture Radar (SAR) data sets have been widely exploited for land use land cover (LULC) classification due to their sensitivity to the structural and dielectric properties of the imaging target. In this study, the potential of fully polarimetric L‐ and S‐band Airborne SAR (LS‐ASAR) data sets were explored for the machine‐learning‐based classification of Urban, Vegetation, Waterbody, and Open Ground. This work was done by utilizing dual‐frequency L‐ and S‐band airborne data of Santa Barbara, California, USA, acquired under the Airborne SAR (LS ASAR) campaign, a precursor airborne mission to the space‐borne NASA‐ISRO (NISAR) mission. The LS‐ASAR polarimetric information was utilized for LULC classification using the SVM classifier. The roll‐invariant Barnes, and eigenvalue/eigenvector‐based Cloude and H/A/Alpha decomposition were implemented to retrieve the scattering parameters. The backscatter response of classes was studied, and separability analysis was done to reduce the misclassification error between six class pairs‐ Vegetation—Urban, Vegetation—Waterbody, Vegetation—Open Ground, Urban—Waterbody, Urban—Open Ground, and Water—Open Ground. The decomposition models failed to achieve the desirable separability index for all six class pairs; consequently, the classification of Barnes, Cloude, H/A/Alpha decomposition showed misclassification between vegetation‐urban class, and waterbody‐open ground class for both L‐ and S‐band data sets. The effort was made toward improving the classification accuracy by integrating the roll‐invariant and eigenvalue‐eigenvector scattering parameters of the multifrequency L‐ and S‐band data set. This method presented the desirable separability index for all class‐pair; eventually highest classification accuracy was achieved i.e. 93.35% ( = 0.91) by significantly reducing the misclassification error between class pairs.

Four-decade (1977-2017) landscape tale of tourist reservoir hotspot El Piñol-Guatapé, Colombia.

Efficient management of land use/land cover (LULC) features is vital for a balanced sustainable ecosystem. Thus, this work aimed to document the LULC changes in the less studied El Peñol-Guatapé reservoir, Antioquia, Colombia, especially in the reservoir area due to the construction of a hydro-electric power plant. For this study, Landsat images of 1977, 1986, 1997, and 2017 were used and the results indicated an increase in the settlement area and road networks by 0.10 and 0.60%, respectively, while during 1986 to 2017, cropland, plantation, dense forest, and open forest areas presented an increase of 0.52, 1.06, 2.87 and 2.61%, respectively. However, the marshy vegetation, scrub forest and fallow land decreased to - 0.51, - 3.79 and - 4.29%, respectively, in the same period. The water body before and after the completion of reservoir project denoted an increase from 13.1 km2 in 1977 to 45.7 km2 in 1986. This study provides a first-hand report on LULC dynamics in this tourism dominated municipalities that will serve as a reference for ecosystem management to reconcile the conflicts between different LULC classes in ecologically enriched regions.

Quantifying the impacts of spatiotemporal land use and land cover changes on soil loss across agroecologies and slope categories using GIS and RUSLE model in Zoa watershed, southwest Ethiopia

BackgroundSoil erosion in Ethiopian highlands is highly consistent with land use/land cover (LULC) changes that are associated with deforestation and a decline in biodiversity. However, identifying soil erosion-prone areas and quantifying soil loss in rugged terrains and various agroecologies due to LULC changes have not been dedicated to scientific studies. Therefore, we quantified the impacts of spatiotemporal LULC changes on soil loss across agroecologies and slope categories using GIS and RUSLE model from 1985 to 2021 in Zoa watershed, southwest Ethiopia. Moreover, prioritizing erosion severity risks at sub-watersheds and quantifying temporal sediment yield is essential for better conservation planning. Landsat images, rainfall, Digital Elevation Model, and soil data were obtained from field observations and secondary sources.ResultsBareland and farmland have been expanding at the expense of other land use types. The annual soil loss in the watershed ranged from 0 to 113.21 t ha−1 year−1, 0 to 163.16 t ha−1 year−1, and 0 to 194.58 mathrm{t},{mathrm{ha}}^{-1},{mathrm{year}}^{-1} with a mean annual soil loss of 21.07, 29.35 and 40.93 t ha−1 year−1 in 1985, 2000, and 2021, respectively. Among LULC classes, the highest soil loss was generated from bareland (31.73 mathrm{t },{mathrm{ha}}^{-1},{mathrm{year}}^{-1}) and farmland (27.08 t ha−1 year−1 ) in 1985 later upsurged to 35.52 mathrm{t },{mathrm{ha}}^{-1},{mathrm{year}}^{-1} and 59.91 t ha−1 year−1 in 2021, respectively, due to the maximum susceptibility of soil erosion risks from unprotected surfaces. The results also revealed that the lowland agroecology generated the highest mean soil loss of 24.05 mathrm{t },{mathrm{ha}}^{-1},{mathrm{year}}^{-1} in 1985, 39.74 t ha−1 year−1 in 2000, and increased to 57.55 mathrm{t },{mathrm{ha}}^{-1},{mathrm{year}}^{-1} in 2021. Considering the slope categories, the highest and most excruciating soil loss was engendered from steep (35.55–60.78 t ha–1 year–1) and very steep (52.48–72.69 mathrm{t },{mathrm{ha}}^{-1},{mathrm{year}}^{-1}) slope terrains during 1985–2021. The northwestern part of the watershed is the most erosion-prone area which is now expanding to the central and western parts of the watershed. The sediment yield increased at the fastest rate at the watershed outlet, from 39.3% in 1985 to 94.26% in 2021.ConclusionsThe results of this study indicated that the conversion of other LULC categories into farmland was the most detrimental to a watershed in terms of soil loss, which necessitates the implementation of appropriate soil and water conservation measures with effective design by considering spatial variability to reduce soil erosion hazards.

Determination of carbon storage amounts in above-ground biomass using NDVI based on land use/land cover classes in Bolu

In this study, the land use/land cover (LULC) class of Bolu province in 1990, 2006, and 2018 and LULC class transitions between periods were examined. In addition, the amount of carbon storage in LULC classes of above-ground biomass (AGB) for the years 1990-2018 was determined through NDVI. CORINE land classification was used to determine the change of LULC classes. Landsat 5 TM and Landsat 8 OLI satellite images were used to determine the carbon storage amounts for the selected periods. There has been an increase of approximately 81.56% (4797.1 ha) in the LULC class of artificial surfaces in 28 years. A total increase of 2.87% (5539.8 ha) has been experienced in agricultural areas in the 28-year period. Forest and semi-natural areas decreased by 1.83% (11745.5 ha) and water bodies have increased approximately 3.4 times in 28 years. In terms of carbon amounts in AGB of the LULC classes, there has been approximately 59.5% carbon accumulation over 28 years. While a total of 1583467.3 tons of carbon was stored in 1990, this amount increased to 2664549.5 tons of carbon in 2018. Of the total carbon stored in 1990, 82.5% was stored in the forest and semi-natural areas, and 17% was stored in agricultural areas. While in 2018, 76.3% of the total carbon was stored in forest and semi-natural areas, and 23% in agricultural areas. In the 28-year period, a total of 1000151.7 tons of carbon has accumulated in the AGBs of the LULC classes that have not been converted to a different class. On the other hand, a total of 80930.6 tons of carbon was transferred from one class to another over the transformed lands. In total, 1081082.3 tons of carbon accumulation occurred between 1990 and 2018.

A Comparison of Deep Transfer Learning Methods for Land Use and Land Cover Classification

The pace of Land Use/Land Cover (LULC) change has accelerated due to population growth, industrialization, and economic development. To understand and analyze this transformation, it is essential to examine changes in LULC meticulously. LULC classification is a fundamental and complex task that plays a significant role in farming decision making and urban planning for long-term development in the earth observation system. Recent advances in deep learning, transfer learning, and remote sensing technology have simplified the LULC classification problem. Deep transfer learning is particularly useful for addressing the issue of insufficient training data because it reduces the need for equally distributed data. In this study, thirty-nine deep transfer learning models were systematically evaluated alongside multiple deep transfer learning models for LULC classification using a consistent set of criteria. Our experiments will be conducted under controlled conditions to provide valuable insights for future research on LULC classification using deep transfer learning models. Among our models, ResNet50, EfficientNetV2B0, and ResNet152 were the top performers in terms of kappa and accuracy scores. ResNet152 required three times longer training time than EfficientNetV2B0 on our test computer, while ResNet50 took roughly twice as long. ResNet50 achieved an overall f1-score of 0.967 on the test set, with the Highway class having the lowest score and the Sea Lake class having the highest.

Quantitative Assessment of the Relationship between Land Use/Land Cover Changes and Wildfires in Southern Europe

Wildfires are key drivers of land use/land cover (LULC) dynamics by burning vegetation and affecting human infrastructure. On the contrary, LULC changes (LULCCs) may affect the fire regime by influencing vegetation type, burnable areas, fuel loads and continuity. This study investigates the relationship between LULCC and wildfires. We developed a methodology based on different indicators, which allowed us to quantitatively assess and better understand the transitions between LULC classes and burnt area (BA) in Europe in the last two decades (2000–2019). The assessment was performed for the entire European continent and, independently, for each of the five European countries most affected by wildfires: Portugal, Spain, France, Italy and Greece. The main results are the following: (i) LULCC analysis revealed a net loss in forests and arable land and a net gain in shrubs; (ii) most of the BA occurred in forests (42% for the whole of Europe), especially in coniferous forests; (iii) transitions from BA generally were to transitional woodland/shrub or, again, to BA. Overall, our results confirm the existence of a strong relationship between wildfires and LULCCs in Europe, which was quantified in the present study. These findings are of paramount importance in fire and environmental system management and ecology.

Land Use and Land Cover Mapping with VHR and Multi-Temporal Sentinel-2 Imagery

Land Use/Land Cover (LULC) mapping is the first step in monitoring urban sprawl and its environmental, economic and societal impacts. While satellite imagery and vegetation indices are commonly used for LULC mapping, the limited resolution of these images can hamper object recognition for Geographic Object-Based Image Analysis (GEOBIA). In this study, we utilize very high-resolution (VHR) optical imagery with a resolution of 50 cm to improve object recognition for GEOBIA LULC classification. We focused on the city of Nice, France, and identified ten LULC classes using a Random Forest classifier in Google Earth Engine. We investigate the impact of adding Gray-Level Co-Occurrence Matrix (GLCM) texture information and spectral indices with their temporal components, such as maximum value, standard deviation, phase and amplitude from the multi-spectral and multi-temporal Sentinel-2 imagery. This work focuses on identifying which input features result in the highest increase in accuracy. The results show that adding a single VHR image improves the classification accuracy from 62.62% to 67.05%, especially when the spectral indices and temporal analysis are not included. The impact of the GLCM is similar but smaller than the VHR image. Overall, the inclusion of temporal analysis improves the classification accuracy to 74.30%. The blue band of the VHR image had the largest impact on the classification, followed by the amplitude of the green-red vegetation index and the phase of the normalized multi-band drought index.

Spatiotemporal Patterns in Land Use/Land Cover Observed by Fusion of Multi-Source Fine-Resolution Data in West Africa

Land Use/Land Cover (LULC) change is a major global concern and a topic of scientific debate. In West Africa, the key trend among the changes of the past few years is the loss of natural vegetation related to changes in different LULC categories, e.g., water bodies, wetland, and bare soil. However, not all detected changes in these LULC categories are relevant for LULC change management intervention in a resource-constrained continent, as a massive change in the dominant LULC types may be due to errors in the LULC maps. Previous LULC change analysis detected large discrepancies in the existing LULC maps in Africa. Here, we applied an open and synergistic framework to update and improve the existing LULC maps for West Africa at five-year intervals from 1990 to 2020—updating them to a finer spatial resolution of 30 m. Next, we detected spatial–temporal patterns in past and present LULC changes with the intensity analysis framework, focusing on the following periods: 1990–2000, 2000–2010, and 2010–2020. A faster annual rate of overall transition was detected in 1990–2000 and 2010–2020 than in 2000–2010. We observed consistent increases in shrubland and grassland in all of the periods, which confirms the observed re-greening of rangeland in West Africa. By contrast, forestland areas experienced consistent decreases over the entire period, indicating deforestation and degradation. We observed a net loss for cropland in the drought period and net gains in the subsequent periods. The settlement category also gained actively in all periods. Net losses of wetland and bare land categories were also observed in all of the periods. We observed net gains in water bodies in the 1990–2000 period and net losses in the 2010–2020 period. We highlighted the active forestland losses as systematic and issued a clarion call for an intervention. The simultaneous active gross loss and gain intensity of cropland raises food security concerns and should act as an early warning sign to policy makers that the food security of marginal geographic locations is under threat, despite the massive expansion of cropland observed in this study area. Instead of focusing on the dynamics of all the LULC categories that may be irrelevant, the intensity analysis framework was vital in identifying the settlement category relevant for LULC change management intervention in West Africa, as well as a cost-effective LULC change management approach.

Land use/land cover classification using hyperspectral soil reflectance features in the Eastern Himalayas, India

Information on periodic land use/land cover (LULC) changes are imperative for regional agricultural planning and policymaking. In this study, 332 soil samples were collected across five LULC types (coal mine degraded land, grassland, shifting cultivation or Jhum, plantation, and upland agriculture) of Eastern Indian Himalayas and subsequently characterized via traditional laboratory protocols and hyperspectral sensor. The soil physicochemical data and spectral data were separately used to classify six LULC types using three machine learning (ML) algorithms [support vector machine (SVM), random forest (RF), and K-nearest neighbors (KNN)]. Results indicated that the change in LULC directly impacted the dynamics of soil properties. The principal component analysis highlighted the interrelationships between the suite of soil physicochemical properties in classifying soil samples from different LULC types. All three ML algorithms exhibited that the physicochemical properties of the soil can perfectly classify LULC types. Spectral data from the hyperspectral sensor also demonstrated good classification accuracy. Overall, SVM performed better than RF and KNN, producing 93% and 78% classification accuracy using soil properties and spectral data, respectively. Moreover, the RF algorithm could select the influential soil and spectral variables for classifying LULC types. In the future, the approach tested herein can be used to classify several LULC types into the archived national soil spectral database. More research is needed to include a wide range of soil data and more LULC classes for a comprehensive classification using advanced deep learning tools.

Deep Insight on Land Use/Land Cover Geospatial Assessment through Internet-Based Validation Tool in Upper Karkheh River Basin (KRB), South-West Iran

Recently, the demand for high-quality land use/land cover (LULC) information for near-real-time crop type mapping, in particular for multi-relief landscapes, has increased. While the LULC classes are inherently imbalanced, the statistics generally overestimate the majority classes and underestimate the minority ones. Therefore, the aim of this study was to assess the classes of the 10 m European Satellite Agency (ESA) WorldCover 2020 land use/land cover product with the support of the Google Earth Engine (GEE) in the Honam sub-basin, south-west Iran, using the LACOVAL (validation tool for regional-scale land cover and land cover change) online platform. The effect of imbalanced ground truth has also been explored. Four sampling schemes were employed on a total of 720 collected ground truth points over approximately 14,100 ha. The grassland and cropland totally canopied 94% of the study area, while barren land, shrubland, trees and built-up covered the rest. The results of the validation accuracy showed that the equalized sampling scheme was more realistically successful than the others in terms of roughly the same overall accuracy (91.6%), mean user’s accuracy (91.6%), mean producers’ accuracy (91.9%), mean partial portmanteau (91.9%) and kappa (0.9). The product was statistically improved to 93.5% ± 0.04 by the assembling approach and segmented with the help of supplementary datasets and visual interpretation. The findings confirmed that, in mapping LULC, data of classes should be balanced before accuracy assessment. It is concluded that the product is a reliable dataset for environmental modeling at the regional scale but needs some modifications for barren land and grassland classes in mountainous semi-arid regions of the globe.

A multi-layer perceptron–Markov chain based LULC change analysis and prediction using remote sensing data in Prayagraj district, India

Land use land cover (LULC) mapping and temporal observations are indispensable drivers for sustainable development. This research showed the growth trends and land use transition for the Prayagraj district in the last three decades. Supervised classification of Landsat images was performed on 5-year temporal intervals using a maximum likelihood classifier. All satellite images were organized into six major LULC feature classes viz agriculture/open land, barren land, built-up, forest, sand, and water. The overall accuracy of LULC classification was achieved by more than 89% in all seven temporal points. Furthermore, the accuracy of the classified maps was estimated through area-based error matrix. The Land Change Modeler tool of TerrSet 2020 software was used to analyze the transition of classes and to incorporate the multi-layer perceptron-Markov chain (MLP-MC) technique. The transition potentials were included in MLP-MC with the help of sensitive explanatory variables and significant transitions of classes. Furthermore, these transition potentials and the Markov chain transition matrix were used to predict the future LULC dynamics and vulnerability. The change analysis revealed that a significant portion of the agriculture/open land gradually decreased and got converted to built-up land. The results depicted that agriculture/open land was reduced by 8.03% in the last three decades while the built-up region was grown by 199.61%. Forest area was continuously decreasing while the sand area increased due to river meandering. Overall, more than 75% of accuracy was achieved in MLP. The prediction model was first validated with observed data, and then the LULC scenario of 2035 and 2050 was simulated. LULC of 2050 showed that the built-up area would likely reach 13.90% of district area whereas the forest area would remain only 0.79%. The prediction model has given the output in the form of future LULC map along with projected potential transition maps. This would be useful for sustainable urban planning to deal with the alarming rate of built-up growthand agriculture/open landshrinkage.