Cover Classification Research Articles

Mapping Pinus spp. Forestry and Land Cover Classes Using High-resolution PlanetScope Satellite Data: Experimenting Images from Different Seasons and Machine Learning Methods

The Remote Sensing and machine learning techniques are cost-effective ways of mapping land use and cover, especially forestry areas. This is essential for the management and planning of such resources. The purpose of this study was to identify which classifier (Random Forest or Support Vector Machine) reach the best accuracy in land use and cover classification and determine which is the best season of year for Pinus spp. forest mapping. PlanetScope multispectral image was used with 3.7 m of spatial resolution, collected over the coastal region of Rio Grande do Sul state. The input variables for the classifiers were the four spectral bands: RGB and NIR, and the NDVI vegetation index. In both classifiers, high accuracy values were obtained, as well as for all seasons of the year. The Random Forest classifier obtained better results in the spring and summer seasons, while in the autumn and winter seasons there was no significant difference between the classifiers for the classification of Pinus spp. forests. The results reached an adequate precision to be used for the management and monitoring of the land use and cover in the municipality of São José do Norte, RS.

Land Use Change Modelling Using Logistic Regression, Random Forest and Additive Logistic Regression in Kubu Raya Regency, West Kalimantan

Kubu Raya Regency is a regency in the province of West Kalimantan which has a wetland ecosystem including a high-density swamp or peatland ecosystem along with an extensive area of mangroves. The function of wetland ecosystems is essential for fauna, as a source of livelihood for the surrounding community and as storage reservoir for carbon stocks. Most of the land in Kubu Raya Regency is peatland. As a consequence, peat has long been used for agriculture and as a source of livelihood for the community. Along with the vast area of peat, the regency also has a potential high risk of peat fires. This study aims to predict land use changes in Kubu Raya Regency using three statistical machine learning models, specifically Logistic Regression (LR), Random Forest (RF) and Additive Logistic Regression (ALR). Land cover map data were acquired from the Ministry of Environment and Forestry and subsequently reclassified into six types of land cover at a resolution of 100 m. The land cover data were employed to classify land use or land cover class for the Kubu Raya regency, for the years 2009, 2015 and 2020. Based on model performance, RF provides greater accuracy and F1 score as opposed to LR and ALR. The outcome of this study is expected to provide knowledge and recommendations that may aid in developing future sustainable development planning and management for Kubu Raya Regency.

Identification of Deforestation in Protected Forest Areas Using Land Cover mapping (A Case Study of Bukit Suligi Protected Forest)

In order to identify deforestation in Protected Forest Areas, land cover mapping has been used as a relevant method. This research aims to identify and map deforestation rates around the Bukit Suligi Protected Forest Area, as well as changes in land cover classes that occurred during the 2015-2022 period. The results showed that the deforestation rate reached 1,362.55 hectares per year or about 10.45% per year, a very alarming figure. This indicates a serious threat to forest sustainability and biodiversity in the area.

A novel multi-class land use/land cover classification using deep kernel attention transformer for hyperspectral images

A novel multi-class land use/land cover classification using deep kernel attention transformer for hyperspectral images

Impact of Land Use Change on the Water Environment of a Key Marsh Area in Vientiane Capital, Laos

The water environment is critical to maintaining ecosystem balance and human well-being globally. It is essential to comprehend the effects of land use change on water quantity and quality for sustainable development of the urban environment. Expansion of urban areas leads to intensified human activity and increased pollution loads in natural waterbodies. This study aimed to monitor changes in land use over a span of two decades to evaluate the condition of the water environment in That Luang Marsh (TLM). The land use and land cover (LULC) classes, including agricultural land, bare land, built-up land, vegetation, waterbody, and wetland, were categorized via Landsat images utilizing the maximum likelihood algorithm. A digital elevation model was used to estimate the water surface area and volume, and the nutrient delivery ratio model was employed to analyze nutrient distribution across the LULC classes. The results showed that from 2001 to 2020, the bare land, built-up, waterbody, and wetland areas increased by 29.92, 18.64, 0.87, and 0.16 times, respectively, while the agricultural and vegetation land decreased by 0.80 and 0.76 times, respectively. A binary logistic regression model for influential factors implies that road network expansion within the special economic zone in TLM could result in an increase in residential areas, thereby impacting the LULC classes. The increase in water volume showed a robust correlation with the expansion of built-up land, bare land, and waterbody. TLM had an average nitrate-nitrogen export of 317 tons/year with a 95% confidence interval of (56, 770) tons/year in 2020. The distribution over LULC classes affected the export, which varied dynamically. Vegetation land had the highest nitrate-nitrogen load of 0.57 tons/ha/year, probably due to poorly managed use of fertilizers. The developed land surface for an artificial lake could lead to an increase in the water volume, which could be involved in the dilution of nutrient concentration. Therefore, it is crucial to prioritize policies that aim to establish sustainable urban water environments through rational urban planning and by making LULC management a primary consideration, especially for developing countries undergoing similar processes of urbanization along the Mekong River in Southeast Asia.

Impact of Land Use Changes on Ecosystem Services Supply: A Meta Analysis of the Italian Context

Changes in land use and land cover (LULC) are caused by several factors, including climate change, socio-demographic dynamics, human pressures and urban sprawl. These factors alter the structure and functionality of ecosystems and their capacity to provide ecosystem goods and services to society. The study of LULC changes is important for understanding the dynamics of relationships between environmental, social and economic components and for analyzing the factors affecting natural capital. Including ecosystem services (ES) in spatial planning tools and sectoral policies is useful for improving governance. In this paper, the impact of LULC changes on ES provision has been estimated. To this end, we carried out a literature review (Step 1) to select the biophysical and economic coefficients of ES supply by land cover classes and collect them in a database (Step 2). We subsequently aggregated the economic and biophysical coefficients by macro classes (Step 3) and, using the benefit transfer approach, we estimated the change in the supply of ESs concerning permanence and transition phenomena in Italy from 1990 to 2018 (Step 4). The transition phenomena analysis also allowed us to evaluate the consequences of urbanization and urban green space governance on ES supply. Indeed, these urban green spaces can help reduce risks to people’s health and safety and mitigate the effects induced by climate change. In total, approximately 800 coefficients (biophysical and economic) of ESs supplied by Corine Land Cover classes were acquired. The results show a reduction in the annual supply of ecosystem services of EUR 927 million (2022) caused by LULC changes between 1990 and 2018. This research proposes a methodology to improve knowledge of ESs concerning anthropogenic impacts and to support land-use planning policies regarding Agenda 2030 for Sustainable Development Goals.

SPATIO-TEMPORAL DYNAMICS OF URBAN THERMAL ENVIRONMENT IN UDAIPUR CITY, RAJASTHAN, INDIA

Abstract. The Udaipur urban agglomeration was selected to analyse the urban heat island (UHI) effect in the area from 2005 to 2017. Landsat 7 ETM+ sensor data was used to derive land surface temperatures (LST) and map landscape characteristics. The agglomeration was classified into seven land use land cover classes including agriculture cropped, agriculture fallow, barren, built up, scrub, vegetation and water using unsupervised classification method. The LST results were obtained using NDVI method. Urban heat island intensities were mapped using z-score method. Regions with UHI values of above 2 standard deviations were considered to reflect UHI effect. Results show an overall decrease of 10.2 percent in agricultural cropped and fallow lands. The scrub class also shows a moderate decline of 2.3 percent. Increase in area were observed for built up, barren and vegetation classes by 8.8, 2.1 and 1.9 percent respectively. The dominant LULC change was the transformation of agricultural lands to built-up class. The class-wise mean LST observations increase in order of water, vegetation, cropped, fallow, built up, scrub and barren land. The mean LST records an increase of 1.7° C from 30.1° C to 31.8° C over the study period. Results suggest that UHI effects are more prominent in urban fringe area corresponding with the built up and barren land cover classes contrary to its core area. Built up areas surrounded by barren lands left vacant for city’s sprawl in the future exhibit highest UHI intensities.

A CORRELATION ANALYSIS OF LAND SURFACE TEMPERATURE AND EVAPOTRANSPIRATION IN AN URBAN SETTING

Abstract. Urbanisation is found to increase the Land Surface Temperature (LST), which results in the formation of Urban Heat Islands (UHI). This effect is a result of the natural land cover being replaced with impermeable surfaces like buildings, roads, and pavements, which absorb and hold onto heat more. As a result, metropolitan regions frequently experience substantially hotter temperatures than their nearby rural counterparts, especially during heat waves. This arises partly due to reduction in Actual Evapotranspiration (AET). AET and LST have a close relationship. AET rates are greatly influenced by the amount of vegetation and their transpiration capacity, which in turn can have an influence on LST in urban regions. The type of Land Use and Land Cover (LULC) has a significant impact on how LST affects the interchange of energy between the earth's surface and the atmosphere. Vegetation AET has a huge potential to reduce both temperatures. The temperature of a water body is lower than that of the nearby urban. The objective of this research is to determine the variation in AET and LST with regards to varying types of land cover by finding correlation between them. The spatio-temporal analysis is done for the summer and winter season of 2020. The correlation is studied between LST and AET for each land cover class and for both seasons for 2020 and results are plotted. The green areas and lakes of the city showed a higher AET and lower LST, whereas the built environment behaved in just the opposite manner. It can be inferred that it is essential to comprehend the local context to effectively determine the relationship between LST and AET in a particular scenario.

Exploring the interplay of landscape changes and ecosystem services maximization in man-managed lagoon areas

Coastal lagoons have long been subject to continuous changes caused by mutual interactions with human activities. Monitoring such changes becomes critical, particularly when modifications in landscape and land cover classes can affect their capacity to ensure Ecosystem Services (ESs). In the Venice lagoon, some confined areas called “valli da pesca” supply provisioning ESs, namely aquaculture and hunting, but also other ESs important for the entire lagoon, such as regulating and cultural ones. Being heavily modified ecosystems under human control, valli da pesca underwent considerable morphological evolution depending on the maximized ES and the applied management. Using remote sensing data from different sources, we reconstructed changes in land cover and landscape elements in valli da pesca over the last century. By calculating landscape indicators related to land, saltmarshes, and water, we found that landscape features were initially similar for all the valli da pesca. Then, a process began between 1975 and 1987, in which management devoted to maximizing different ESs shaped the land cover in specific patterns. This study confirms the importance of these areas in the context of the entire lagoon and suggests the need to monitor their land cover changes to avoid the depletion of their capacity to conserve landscape elements and the related ESs. In this task, remote sensing data represents an important source of historical data that can deepen the knowledge about human-Nature interactions, capable of tracing the landscape evolution and the dynamics in the ESs supply as responses to human interventions.

Land Cover Modelling with Sentinel 2 in Water Balance Calculations of Urban Sites

This study investigates how a land cover map is produced using Sentinel 2 satellite images and how the resulting map is integrated with the calculation of hydrological parameters. Due to its capacity to deliver high-resolution spatial data, satellite imagery is increasingly being used for land cover mapping (1). The process involves several stages, including pre-processing of the Sentinel 2 imagery, image classification using machine learning algorithms, and post-processing of the classified image to generate a land cover map. The resulting land cover map is then integrated with hydrological parameters such as precipitation, evapotranspiration, and soil characteristics to calculate various hydrological parameters. Debrecen, north-east Hungary is the study area. This study's method is suitable for usage in other cities. Understanding the relationship between land cover and water availability requires the integration of the land cover map with hydrological parameters (2). The study demonstrates the potential of using remote sensing data for hydrological studies and highlights the importance of integrating various data sources for accurate estimation of hydrological parameters. The results show that the land cover classes have a significant impact on the water balance of urban sites. This study outlines the key steps involved in creating a land cover map using Sentinel 2 satellite imagery and integrating it with hydrological parameters calculation. The application of this strategy supports sustainable water management techniques and offers insightful information about a region's hydrological processes.

A hybrid modelling approach for detecting seasonal variations in inland Green-Blue Ecosystems

Deforestation, environmental pollution, and the overexploitation of resources, in addition to the Earth's natural cycles, are scaling up the impacts of climate change in the provision of Ecosystem Services (ES). Green-Blue Ecosystems (GBE) are impacted by climatic conditions, topography, and water presence. Data-driven modelling techniques may effectively capture the effects of seasonal variations while modelling natural ecosystems. This research proposes a hybrid modelling approach that combines Deep Learning and traditional Machine Learning, Sensitivity Analysis and Feature Importance Evaluation (FIE) to investigate seasonality effects on mapping GBE. The models, built using satellite imagery from the Spring and Summer seasons of the Mediterranean climate zone, included spectral indices, topography (DEM), and groundwater depth (GD). The model that best suited the analysis was selected using sensitivity tests and hyperparameter optimization. The study shows that land cover classes of transitional woodland shrubs, inland marshes, cultivated land parcels, and watercourses are better classified in the Spring, with an accuracy of 0.814. FIE indicates that spectral indices are the most important predictors for detecting green ecosystems in both seasons. Additionally, DEM and GD are the most relevant predictors to classify watercourses in the Summer. An analytical examination of the input data and hyperparameter settings facilitates understanding of models' behaviour while improving models' prediction. This research provides an advanced understanding of the effects of seasonal variations on the status of GBE and enhances understanding of modelling ES in areas with a growing need for changes in land use and high water supply demand.

A global land cover training dataset from 1984 to 2020

State-of-the-art cloud computing platforms such as Google Earth Engine (GEE) enable regional-to-global land cover and land cover change mapping with machine learning algorithms. However, collection of high-quality training data, which is necessary for accurate land cover mapping, remains costly and labor-intensive. To address this need, we created a global database of nearly 2 million training units spanning the period from 1984 to 2020 for seven primary and nine secondary land cover classes. Our training data collection approach leveraged GEE and machine learning algorithms to ensure data quality and biogeographic representation. We sampled the spectral-temporal feature space from Landsat imagery to efficiently allocate training data across global ecoregions and incorporated publicly available and collaborator-provided datasets to our database. To reflect the underlying regional class distribution and post-disturbance landscapes, we strategically augmented the database. We used a machine learning-based cross-validation procedure to remove potentially mis-labeled training units. Our training database is relevant for a wide array of studies such as land cover change, agriculture, forestry, hydrology, urban development, among many others.

Earthworm functional group assemblages vary with land cover/use in human-impacted landscapes of Sudano-Guinean area, Cameroon

The semiarid zones of Cameroon are experiencing rapid land degradation and desertification due to climate change and human activities. Earthworms play a crucial role in soil functioning and are important indicators of soil quality. Their role as soil process regulators and soil ecosystem engineers has been widely demonstrated. Despite their influence in maintaining ecosystem functioning in tropical soils and considering the number of ecosystems present in Cameroon, few data are available on their distribution in Cameroonian soils in general and those of the northern part (semiarid zones) particularly. This study aims to assess the abundance and biomass of earthworm functional groups in different land uses (woody savannah, grass savannah, gallery forest, and open land) impacted by human activities in the Sudano-Guinean area of Cameroon. During two rainy seasons in 2015 and 2016, we compared earthworm functional group densities and biomasses between three herbaceous land cover (HLC) classes and identified soil parameters and land use history that affect their occurrence. Results showed that earthworm functional group densities differed between land use types and herbaceous land cover. In woody savannahs, HLC over 75 % and 90 % greatly promoted the occurrence of endogeic and anecic earthworms, while in grass savannahs the densities of the two functional groups were higher in HLC over 75 %. Moreover, species from the epigeic group occurred abundantly in the gallery forests and represented about 70 % of earthworm species recorded in our study. Some environmental factors such as soil texture, pH, altitude and soil organic matter affected earthworm abundance, while previous human activities such as pasture/grazing and fallow on land uses contributed to the occurrence of at least one of the earthworm functional groups. The study highlights the importance of understanding earthworm community dynamics in the area for effective land use management and soil restoration practices in the face of land degradation and climate change.

Assessment of the stand structure of protective forest monitoring based on statistical models in Irano-Turanian phytogeographical regions of Iran.

One of the most important data that forest management planners need for effective decisions is the data related to the forest structure. The aim of the study is to investigate and analyse the structure of protective forests in Irano-Turanian phytogeographical regions, Iran. Since there are many shrubs in this phytogeographical region, it is very difficult to measure the stand diameter at any height (breast or root height). For this reason, it is necessary to analyse the parameters of height and crown cover to investigate and analyse forest structure. For that purpose, two study plots were selected, and basic data were analysed by using statistical distributions, scatter plots and R2 coefficients. With EasyFit software and Anderson‒Darling test, it was found that the Weibull (3P) and Pearson 6 (4P) distributions for the crown cover factor and the Gen-Pareto and Pert distributions for the height factor have the best goodness-of-fit for the distribution of the different crown cover classes and heights in the studied forest. Moreover, the results confirm that there is a very weak R2 coefficient between crown cover and root collar diameter, with R2 = 0.513 and 0.369 in plots 1 and 2, respectively. Therefore, the combination of crown cover and height parameters is more suitable for use when analysing stand structure in such forests, although the values of R2 are still low (0.673 and 0.524 in plots 1 and 2, respectively). The results of this study show that in protective forests with many shrubs, it is better to focus on the height and crown cover of ​trees and of shrubs rather than on parameters related to stand/tree diameter when analysing stand structure.

Automatic wide area land cover mapping using Sentinel-1 multitemporal data

This study introduces a methodology for land cover mapping across extensive areas, utilizing multitemporal Sentinel-1 Synthetic Aperture Radar (SAR) data. The objective is to effectively process SAR data to extract spatio-temporal features that encapsulate temporal patterns within various land cover classes. The paper outlines the approach for processing multitemporal SAR data and presents an innovative technique for the selection of training points from an existing Medium Resolution Land Cover (MRLC) map. The methodology was tested across four distinct regions of interest, each spanning 100 × 100 km2, located in Siberia, Italy, Brazil, and Africa. These regions were chosen to evaluate the methodology’s applicability in diverse climate environments. The study reports both qualitative and quantitative results, showcasing the validity of the proposed procedure and the potential of SAR data for land cover mapping. The experimental outcomes demonstrate an average increase of 16% in overall accuracy compared to existing global products. The results suggest that the presented approach holds promise for enhancing land cover mapping accuracy, particularly when applied to extensive areas with varying land cover classes and environmental conditions. The ability to leverage multitemporal SAR data for this purpose opens new possibilities for improving global land cover maps and their applications.

Emissions of pesticides in the European Union: a new regional-level dataset

We present a European Union (EU)-wide dataset of estimated quantities of active substances of plant protection product applied on crops (also called “emissions”). Our estimates are derived from data reported by eight EU countries and extrapolated to encompass all EU regions using regression models. These models consider both climate and agricultural land use data. This allows us to spatially represent pesticide use at NUTS Level 3 of the European statistical mapping units, and within various agricultural land cover classes in each region. We compare our estimates with aggregated data provided by EUROSTAT and with independent, detailed data for the United Kingdom, highlighting an error typically within one order of magnitude. Our estimates can provide insights into the distribution and patterns of pesticide use in the EU around the year 2015. The estimate is most reliable for Western and Southern Europe. Outside these regions, data scarcity makes extrapolation more uncertain, potentially limiting the ability to accurate depict regional variations in pesticide use.

A SHALLOW NEURAL NETWORK MODEL FOR URBAN LAND COVER CLASSIFICATION USING VHR SATELLITE IMAGE FEATURES

Abstract. Recently, image classification techniques using neural networks have received considerable attention in sustainable urban development, since their applications have an extreme effect on building distribution, infrastructural networks, and water resource management. In this research, a back-propagation shallow neural network model is presented for very high resolution satellite image classification in urban environments. Workflow procedures consider selecting and collecting data, preparing required study areas, extracting distinctive features, and applying the classification process. Visual interpretation is performed to identify observed land cover classes and detect distinctive features in the urban environment. Pre-processing techniques are implemented to present the used images in a more suited form for the classification techniques. A shallow neural network model (supported by MathWorks MATLAB environment) is successfully applied and results are evaluated. The proposed model is tested for classifying both WorldView-2 and WorldView-3 multispectral images with different spatial and spectral characteristics to check the model’s applicability to various kinds of satellite imagery and different study areas. Model outcomes are compared to two well-known classification methods; the Nearest Neighbour object-based method and the Maximum Likelihood pixel-based classifier, to validate and check the model stability. The overall accuracy achieved by the proposed model is 86.25% and 83.25%, while the nearest neighbour approach has obtained 84.50% and 82.75%, and the maximum likelihood classifier has accomplished 82.50% and 80.25% for study area 1 and study area 2 respectively. Obtained results indicate that the developed shallow neural network model achieves a promising accuracy for urban land cover classification in comparison with the standard techniques.

Landsat sub-pixel land cover dynamics in the Brazilian Amazon

The Brazilian Amazon land cover changes rapidly due to anthropogenic and climate drivers. Deforestation and forest disturbances associated with logging and fires, combined with extreme droughts, warmer air, and surface temperatures, have led to high tree mortality and harmful net carbon emissions in this region. Regional attempts to characterize land cover dynamics in this region focused on one or two anthropogenic drivers (i.e., deforestation and forest degradation). Land cover studies have also used a limited temporal scale (i.e., 10–15 years), focusing mainly on global and country-scale forest change. In this study, we propose a novel approach to characterize and measure land cover dynamics in the Amazon biome. First, we defined 10 fundamental land cover classes: forest, flooded forest, shrubland, natural grassland, pastureland, cropland, outcrop, bare and impervious, wetland, and water. Second, we mapped the land cover based on the compositional abundance of Landsat sub-pixel information that makes up these land cover classes: green vegetation (GV), non-photosynthetic vegetation, soil, and shade. Third, we processed all Landsat scenes with <50% cloud cover. Then, we applied a step-wise random forest machine learning algorithm and empirical decision rules to classify intra-annual and annual land cover classes between 1985 and 2022. Finally, we estimated the yearly land cover changes in forested and non-forested ecosystems and characterized the major change drivers. In 2022, forest covered 78.6% (331.9 Mha) of the Amazon biome, with 1.4% of secondary regrowth in more than 5 years. Total herbaceous covered 15.6% of the area, with the majority of pastureland (13.5%) and the remaining natural grassland. Water was the third largest land cover class with 2.4%, followed by cropland (1.2%) and shrubland (0.4%), with 89% overall accuracy. Most of the forest changes were driven by pasture and cropland conversion, and there are signs that climate change is the primary driver of the loss of aquatic ecosystems. Existing carbon emission models disregard the types of land cover changes presented in the studies. The twenty first century requires a more encompassing and integrated approach to monitoring anthropogenic and climate changes in the Amazon biome for better mitigation, adaptation, and conservation policies.

Determination of change in the land use and land cover of the Samsun Bafra Delta Plain from 1990 to 2020 using GIS and Remote Sensing Techniques

Land use and land cover changes can have detrimental effects on the ecology, if they are not properly aligned with the characteristics of the land. This study aims to evaluate the temporal changes in land use and land cover of Bafra Delta plain, situated in the east of Samsun province. The region is one of the most significant plains within the Black Sea area. Remote sensing technique was utilized in this research which made use of Landsat images from 1990, 2000, 2010, and 2020. Supervised classification was applied in ENVI 5.3v software to perform calculations, resulting in six main classes. Field work was applied to classify the unclassified classes. The resulting six land use-land cover classes were agriculture lands, forest, dune, marshy, water surface, and artificial areas. To determine land use efficiency, analogue data was digitised and transferred to a GIS database. The agricultural areas occupy the largest portion of the plain, followed by hazelnut and artificial areas. The changes over the last decade, notably the growth of artificial areas and water surfaces, and the reduction of arable lands, highlight significant variations in size across the areas. Furthermore, the study indicated that remote sensing and geographic information system techniques play a crucial role in identifying and monitoring land cover and land use trends on a large-scale to produce accurate and timely data. Poorly adapted land use changes can cause major ecological damage. The aim of this study is to identify the changes over time in land use and land cover of Bafra Delta plain, located to the east of Samsun city and one of the most significant plains in the Black Sea region, using remote sensing techniques. To this end, Landsat images from 1990, 2000, 2010 and 2020 are utilized. To perform the calculations, ENVI 5.3v software was employed, applying a supervised classification technique that resulted in forming six main classes. Fieldwork was conducted to classify the unclassified classes. The resulting land-use and land-cover classes were agricultural land, forest, dunes, marshland, water surface, and artificial areas. To evaluate land-use efficiency, analogue data were digitalised and imported into a GIS database. The plain's most extensive land-use areas consist of agricultural lands, followed by hazelnut and artificial areas. In the last decade, the rise in artificial and water surfaces and the decline in agricultural areas highlights significant changes in the region's size. This study also emphasises the crucial role of remote sensing and geographic information system techniques in generating fast and consistent data for monitoring large-scale land cover and land use trends.

Design-based mapping of land use/land cover classes with bootstrap estimation of precision by nearest-neighbour interpolation

Design-based mapping of land use/land cover classes with bootstrap estimation of precision by nearest-neighbour interpolation