Performance Evaluation of Downscaling Sentinel-2 Imagery for Land Use and Land Cover Classification by Spectral-Spatial Features

Anthropogenic activities dominated tropical forest carbon balance in two contrary ways over the Greater Mekong Subregion in the 21st century

The tropical forest carbon (C) balance threatened by extensive socio-economic development in the Greater Mekong Subregion (GMS) in Asia is a notable data gap and remains contentious. Here we generated a long-term spatially quantified assessment of changes in forests and C stocks from 1999 to 2019 at a spatial resolution of 30 m, based on multiple streams of state-of-the-art high-resolution satellite imagery and in situ observations. Our results show that (i) about 0.54 million square kilometers (21.0% of the region) experienced forest cover transitions with a net increase in forest cover by 4.3% (0.11 million square kilometers, equivalent to 0.31 petagram of C [Pg C] stocks); (ii) forest losses mainly in Cambodia, Thailand, and in the south of Vietnam, were also counteracted by forest gains in China due mainly to afforestation; and (iii) at the national level during the study period an increase in both C stocks and C sequestration (net C gain of 0.087 Pg C) in China from new plantation, offset anthropogenetic emissions (net C loss of 0.074 Pg C) mainly in Cambodia and Thailand from deforestation. Political, social, and economic factors significantly influenced forest cover change and C sequestration in the GMS, positively in China while negatively in other countries, especially in Cambodia and Thailand. These findings have implications on national strategies for climate change mitigation and adaptation in other hotspots of tropical forests.

Land-Use and Land-Cover Classification in Semi-Arid Areas from Medium-Resolution Remote-Sensing Imagery: A Deep Learning Approach

Detailed Land-Use and Land-Cover (LULC) information is of pivotal importance in, e.g., urban/rural planning, disaster management, and climate change adaptation. Recently, Deep Learning (DL) has emerged as a paradigm shift for LULC classification. To date, little research has focused on using DL methods for LULC mapping in semi-arid regions, and none that we are aware of have compared the use of different Sentinel-2 image band combinations for mapping LULC in semi-arid landscapes with deep Convolutional Neural Network (CNN) models. Sentinel-2 multispectral image bands have varying spatial resolutions, and there is often high spectral similarity of different LULC features in semi-arid regions; therefore, selection of suitable Sentinel-2 bands could be an important factor for LULC mapping in these areas. Our study contributes to the remote sensing literature by testing different Sentinel-2 bands, as well as the transferability of well-optimized CNNs, for semi-arid LULC classification in semi-arid regions. We first trained a CNN model in one semi-arid study site (Gujranwala city, Gujranwala Saddar and Wazirabadtownships, Pakistan), and then applied the pre-trained model to map LULC in two additional semi-arid study sites (Lahore and Faisalabad city, Pakistan). Two different composite images were compared: (i) a four-band composite with 10 m spatial resolution image bands (Near-Infrared (NIR), green, blue, and red bands), and (ii) a ten-band composite made by adding two Short Wave Infrared (SWIR) bands and four vegetation red-edge bands to the four-band composite. Experimental results corroborate the validity of the proposed CNN architecture. Notably, the four-band CNN model has shown robustness in semi-arid regions, where spatially and spectrally confusing land-covers are present.

Genus-Level Mapping of Invasive Floating Aquatic Vegetation Using Sentinel-2 Satellite Remote Sensing

Invasive floating aquatic vegetation negatively impacts wetland ecosystems and mapping this vegetation through space and time can aid in designing and assessing effective control strategies. Current remote sensing methods for mapping floating aquatic vegetation at the genus level relies on airborne imaging spectroscopy, resulting in temporal gaps because routine hyperspectral satellite coverage is not yet available. Here we achieved genus level and species level discrimination between two invasive aquatic vegetation species using Sentinel 2 multispectral satellite data and machine-learning classifiers in summer and fall. The species of concern were water hyacinth (Eichornia crassipes) and water primrose (Ludwigia spp.). Our classifiers also identified submerged and emergent aquatic vegetation at the community level. Random forest models using Sentinel-2 data achieved an average overall accuracy of 90%, and class accuracies of 79–91% and 85–95% for water hyacinth and water primrose, respectively. To our knowledge, this is the first study that has mapped water primrose to the genus level using satellite remote sensing. Sentinel-2 derived maps compared well to those derived from airborne imaging spectroscopy and we also identified misclassifications that can be attributed to the coarser Sentinel-2 spectral and spatial resolutions. Our results demonstrate that the intra-annual temporal gaps between airborne imaging spectroscopy observations can be supplemented with Sentinel-2 satellite data and thus, rapidly growing/expanding vegetation can be tracked in real time. Such improvements have potential management benefits by improving the understanding of the phenology, spread, competitive advantages, and vulnerabilities of these aquatic plants.

Downscaling Switzerland Land Use/Land Cover Data Using Nearest Neighbors and an Expert System

High spatial and thematic resolution of Land Use/Cover (LU/LC) maps are central for accurate watershed analyses, improved species, and habitat distribution modeling as well as ecosystem services assessment, robust assessments of LU/LC changes, and calculation of indices. Downscaled LU/LC maps for Switzerland were obtained for three time periods by blending two inputs: the Swiss topographic base map at a 1:25,000 scale and the national LU/LC statistics obtained from aerial photointerpretation on a 100 m regular lattice of points. The spatial resolution of the resulting LU/LC map was improved by a factor of 16 to reach a resolution of 25 m, while the thematic resolution was increased from 29 (in the base map) to 62 land use categories. The method combines a simple inverse distance spatial weighting of 36 nearest neighbors’ information and an expert system of correspondence between input base map categories and possible output LU/LC types. The developed algorithm, written in Python, reads and writes gridded layers of more than 64 million pixels. Given the size of the analyzed area, a High-Performance Computing (HPC) cluster was used to parallelize the data and the analysis and to obtain results more efficiently. The method presented in this study is a generalizable approach that can be used to downscale different types of geographic information.

A wildfire growth prediction and evaluation approach using Landsat and MODIS data

The increasing wildfire occurrence due to global climate changes urged the improvement of present wildfire growth prediction and evaluation methods. This study aimed to propose novel solutions to their two primary limitations, including the lack of robust fuel classification method and the low spatial resolution of wildfire growth accuracy assessment while ensuring wide applicability using open data satellite missions and software. The first objective was to create a robust two-step fuel model classification method consisted of the supervised machine learning classification of generalized land cover classes in the 1st level and their individual unsupervised classification to vegetation subtypes in the 2nd level. The second objective was creating a wildfire prediction accuracy assessment method using MODIS 250 m images, which overcome the limitations of low spatial resolution while preserving sub-daily temporal resolution. The wildfire on the Korčula island in Croatia was analyzed in the study, being specific for its long duration from 18 to 24 July 2015. The wildfire ignition occurred in the isolated area, which prolonged the response time from emergency agencies. Random Forest (RF) with input Landsat 8 spectral bands and indices resulted in the highest classification accuracy in the 1st classification level with an overall agreement of 83.6%. The vegetation subclasses from the 2nd classification level were matched to the 13 standard fuel models for the input in FARSITE software. The predicted wildfire evaluation showed the highest mean accuracy of 0.906 for the first two days, which decreased to 0.722 in the latter stages of the active wildfire caused by overprediction. The proposed two-step fuel model classification presented a cost-efficient solution to the fuel map creation in any part of the world, with a disadvantage of no in-situ ground truth identification and accuracy assessment for 2nd classification level. The evaluation of wildfire growth prediction with 250 m images enabled high spatial and temporal resolution of the assessment, while its limitations of wildfire overprediction and the negative effects of wildfire smoke in MODIS images should be addressed in future research.

Mapping bare ground in New Zealand hill-country agriculture and forestry for soil erosion risk assessment: An automated satellite remote-sensing method

Removing vegetation cover from hill-slope land increases risk for soil erosion and delivery of sediment to waterways. In New Zealand's productive landscapes, clear-fell harvesting of forestry blocks and winter forage grazing by agricultural livestock are two significant causes of vegetation removal. Bare ground exposed by these activities varies annually and seasonally in location and spatial extent. Modelling soil erosion therefore requires temporally and spatially explicit mapping of this bare ground. We have developed an automated mapping method using time-series satellite imagery, thereby enabling wide-area coverage and ease of updating. The temporal analysis identifies land use along with the period of vegetation removal. It produces results per land parcel (in vector format) for use in a Geographic Information System. We present a description of our method, national maps and statistics of bare ground extent in New Zealand's hill-country forestry and winter forage grazing land in 2018, and an assessment of accuracy. The attributes of the mapped land parcels are designed for input into a soil erosion estimation model such as the New Zealand Universal Soil Loss Equation.

What Happens in the City When Long-Term Urban Expansion and (Un)Sustainable Fringe Development Occur: The Case Study of Rome

This study investigates long-term landscape transformations (1949–2016) in urban Rome, Central Italy, through a spatial distribution of seven metrics (core, islet, perforation, edge, loop, bridge, branch) derived from a Morphological Spatial Pattern Analysis (MSPA) analyzed separately for seven land-use classes (built-up areas, arable land, crop mosaic, vineyards, olive groves, forests, pastures). A Principal Component Analysis (PCA) has been finally adopted to characterize landscape structure at 1949 and 2016. Results of the MSPA demonstrate how both natural and agricultural land-uses have decreased following urban expansion. Moreover, the percent ‘core’ area of each class declined substantially, although with different intensity. These results clearly indicate ‘winners’ and ‘losers’ after long-term landscape transformations: urban settlements and forests belong to the former category, the remaining land-use classes (mostly agricultural) belong to the latter category. Descriptive statistics and multivariate exploratory techniques finally documented the intrinsic complexity characteristic of actual landscapes. The findings of this study also demonstrate how settlements have expanded chaotically over the study area, reflecting a progressive ‘fractalization’ and inhomogeneity of fringe landscapes, with negative implications for metropolitan sustainability at large. These transformations were unable to leverage processes of settlement and economic re-agglomeration around sub-centers typical of polycentric development in the most advanced socioeconomic contexts.

Machine Learning-Based Processing Proof-of-Concept Pipeline for Semi-Automatic Sentinel-2 Imagery Download, Cloudiness Filtering, Classifications, and Updates of Open Land Use/Land Cover Datasets

Land use and land cover are continuously changing in today’s world. Both domains, therefore, have to rely on updates of external information sources from which the relevant land use/land cover (classification) is extracted. Satellite images are frequent candidates due to their temporal and spatial resolution. On the contrary, the extraction of relevant land use/land cover information is demanding in terms of knowledge base and time. The presented approach offers a proof-of-concept machine-learning pipeline that takes care of the entire complex process in the following manner. The relevant Sentinel-2 images are obtained through the pipeline. Later, cloud masking is performed, including the linear interpolation of merged-feature time frames. Subsequently, four-dimensional arrays are created with all potential training data to become a basis for estimators from the scikit-learn library; the LightGBM estimator is then used. Finally, the classified content is applied to the open land use and open land cover databases. The verification of the provided experiment was conducted against detailed cadastral data, to which Shannon’s entropy was applied since the number of cadaster information classes was naturally consistent. The experiment showed a good overall accuracy (OA) of 85.9%. It yielded a classified land use/land cover map of the study area consisting of 7188 km2 in the southern part of the South Moravian Region in the Czech Republic. The developed proof-of-concept machine-learning pipeline is replicable to any other area of interest so far as the requirements for input data are met.

Spectral Unmixing for Mapping a Hydrothermal Field in a Volcanic Environment Applied on ASTER, Landsat-8/OLI, and Sentinel-2 MSI Satellite Multispectral Data: The Nisyros (Greece) Case Study

The aim of this study was to propose a methodology that provides a detailed description of the argillic zone of a hydrothermal field, based on satellite multispectral data. More specifically, we developed a method based on spectral unmixing where hydroxyl-bearing alteration is represented by a single endmember (representing clays) and the three (nearly) non-altered primary volcanic lithologies, namely, two types of lava flows (basic and acidic compositions) and the loose materials (alluvial/beach deposits, scree, pyroclastic deposits, etc.), are represented by three endmembers. We also used one endmember representing elemental sulfur that is present in fumarolic vents hosted by active hydrothermal craters. The methodology was applied in the south part of Lakki plain inside the Nisyros volcano caldera (Greece), using Sentinel-2, Landsat-8/OLI, and ASTER satellite multispectral datasets. Specifically, it was applied separately to each one of the three datasets. The spectral unmixing results, combined with the relative geological map, provide quantitative estimations of the primary volcanic and loose material areas affected by alteration. In addition, pixels with high abundance values of hydroxyl-bearing alteration corresponded to mapped areas with strong hydrothermal alteration. The developed methodology is superior to conventional approaches (e.g., alteration spectral index) in terms of its ability to describe the overall pattern of the hydrothermal field. The most accurate results were taken when applied to ASTER or Sentinel-2 MSI data.

Assessment of the Capability of Sentinel-2 Imagery for Iron-Bearing Minerals Mapping: A Case Study in the Cuprite Area, Nevada

With several bands covering iron-bearing mineral spectral features, Sentinel-2 has advantages for iron mapping. However, due to the inconsistent spatial resolution, the sensitivity of Sentinel-2 data to detect iron-bearing minerals may be decreased by excluding the 60 m bands and neglecting the 20 m vegetation red-edge bands. Hence, the capability of Sentinel-2 for iron-bearing minerals mapping were assessed by applying a multivariate (MV) method to pansharpen Sentinel-2 data. Firstly, the Sentinel-2 bands with spatial resolution 20 m and 60 m (except band 10) were pansharpened to 10 m. Then, extraction of iron-bearing minerals from the MV-fused image was explored in the Cuprite area, Nevada, USA. With the complete set of 12 bands with a fine spatial resolution, three band ratios (6/1, 6/8A and (6 + 7)/8A) of the fused image were proposed for the extraction of hematite + goethite, hematite + jarosite and the mixture of iron-bearing minerals, respectively. Additionally, band ratios of Sentinel-2 data for iron-bearing minerals in previous studies were modified with substitution of narrow near infrared band 8A for band 8. Results demonstrated that the capability for detection of iron-bearing minerals using Sentinel-2 data was improved by consideration of two extra bands and the unified fine spatial resolution.

Sentinel-2 Data for Land Cover/Use Mapping: A Review

The advancement in satellite remote sensing technology has revolutionised the approaches to monitoring the Earth’s surface. The development of the Copernicus Programme by the European Space Agency (ESA) and the European Union (EU) has contributed to the effective monitoring of the Earth’s surface by producing the Sentinel-2 multispectral products. Sentinel-2 satellites are the second constellation of the ESA Sentinel missions and carry onboard multispectral scanners. The primary objective of the Sentinel-2 mission is to provide high resolution satellite data for land cover/use monitoring, climate change and disaster monitoring, as well as complementing the other satellite missions such as Landsat. Since the launch of Sentinel-2 multispectral instruments in 2015, there have been many studies on land cover/use classification which use Sentinel-2 images. However, no review studies have been dedicated to the application of ESA Sentinel-2 land cover/use monitoring. Therefore, this review focuses on two aspects: (1) assessing the contribution of ESA Sentinel-2 to land cover/use classification, and (2) exploring the performance of Sentinel-2 data in different applications (e.g., forest, urban area and natural hazard monitoring). The present review shows that Sentinel-2 has a positive impact on land cover/use monitoring, specifically in monitoring of crop, forests, urban areas, and water resources. The contemporary high adoption and application of Sentinel-2 can be attributed to the higher spatial resolution (10 m) than other medium spatial resolution images, the high temporal resolution of 5 days and the availability of the red-edge bands with multiple applications. The ability to integrate Sentinel-2 data with other remotely sensed data, as part of data analysis, improves the overall accuracy (OA) when working with Sentinel-2 images. The free access policy drives the increasing use of Sentinel-2 data, especially in developing countries where financial resources for the acquisition of remotely sensed data are limited. The literature also shows that the use of Sentinel-2 data produces high accuracies (>80%) with machine-learning classifiers such as support vector machine (SVM) and Random forest (RF). However, other classifiers such as maximum likelihood analysis are also common. Although Sentinel-2 offers many opportunities for land cover/use classification, there are challenges which include mismatching with Landsat OLI-8 data, a lack of thermal bands, and the differences in spatial resolution among the bands of Sentinel-2. Sentinel-2 data show promise and have the potential to contribute significantly towards land cover/use monitoring.

Multitemporal 2016-2018 Sentinel-2 Data Enhancement for Landscape Archaeology: The Case Study of the Foggia Province, Southern Italy

This paper is focused on the use of satellite Sentinel-2 data for assessing their capability in the identification of archaeological buried remains. We selected the “Tavoliere delle Puglie” (Foggia, Italy) as a test area because it is characterized by a long human frequentation and is very rich in archaeological remains. The investigations were performed using multi-temporal Sentinel-2 data and spectral indices, commonly used in satellite-based archaeology, and herein analyzed in known archaeological areas to capture the spectral signatures of soil and crop marks and characterize their temporal behavior using Time Series Analysis and Spectral Un-mixing. Tasseled Cap Transformation and Principal Component Analysis have been also adopted to enhance archaeological features. Results from investigations were compared with independent data sources and enabled us to (i) characterize the spectral signatures of soil and crop marks, (ii) assess the performance of the diverse spectral channels and indices, and (iii) identify the best period of the year to capture the archaeological proxy indicators. Additional very important results of our investigations were (i) the discovery of unknown archaeological areas and (ii) the setup of a database of archaeological features devised ad hoc to characterize and categorize the diverse typologies of archaeological remains detected using Sentinel-2 Data.

Mapping Maize Fields by Using Multi-Temporal Sentinel-1A and Sentinel-2A Images in Makarfi, Northern Nigeria, Africa

A timely and accurate crop type mapping is very significant, and a prerequisite for agricultural regions and ensuring global food security. The combination of remotely sensed optical and radar datasets presents an opportunity for acquiring crop information at relative spatial resolution and temporal resolution adequately to capture the growth profiles of various crop species. In this paper, we employed Sentinel-1A (S-1) and Sentinel-2A (S-2) data acquired between the end of June and early September 2016, on a semi-arid area in northern Nigeria. A different set of (VV and VH) SAR and optical (SI and SB) images, illustrating crop phenological development stage, were employed as inputs to the two machines learning Random Forest (RF) and Support Vector Machine (SVM) algorithms to automatically map maize fields. Significant increases in overall classification were shown when the multi-temporal spectral indices (SI) and spectral band (SB) datasets were added with the different integration of SAR datasets (i.e., VV and VH). The best overall accuracy (OA) for maize (96.93%) was derived by using RF classification algorithms with SI-SB-SAR datasets, although the SI datasets for RF and SB datasets for SVM also produced high overall maize classification accuracies, of 97.04% and 97.44%. The outcomes indicate the robustness of the RF or SVM methods to produce high-resolution maps of maize for subsequent application from agronomists, policy planners, and the government, because such information is lacking in our study area.

Preventive Archaeology Based on Open Remote Sensing Data and Tools: The Cases of Sant’Arsenio (SA) and Foggia (FG), Italy

Sentinel-2 data have been used in various fields of human activity. In cultural heritage, their potential is still to be fully explored. This paper aims to illustrate how remote sensing and open source tools are useful for archaeological investigations. The whole issue revolves around the application of satellite (Sentinel-2) and accessory tools for the identification, knowledge and protection of the cultural heritage of two areas of southern Italy: Sant’Arsenio (SA) and Foggia (FG). Both study cases were selected for a specific reason: to demonstrate the usefulness of open data and software for research and preservation of cultural heritage, as in the case of urban sprawl, development of public works (gas- and oil-pipelines, etc.) or intensive use of land for agricultural purposes. The results obtained are relevant for the knowledge improvement and very useful to operate in the field of preventive archaeology, for the evaluation and management of risk, the planning of city-expansion or infrastructures that could damage the buried heritage.

Automatic Updating of Land Cover Maps in Rapidly Urbanizing Regions by Relational Knowledge Transferring from GlobeLand30

Land-cover map is the basis of research and application related to urban planning, environmental management and ecological protection. Land-cover updating is an essential task especially in a rapidly urbanizing region, where fast development makes it necessary to monitor land-cover change in a timely manner. However, conventional approaches always have the limitations of large amounts of sample collection and exploitation of relational knowledge between multi-modality remote sensing datasets. With some global land-cover products being available, it is important to produce new land-cover maps based on the existing land-cover products and time series images. To this end, a novel transfer learning based automatic approach was proposed for updating land cover maps of rapidly urbanizing regions. In detail, the proposed method is composed of the following three steps. The first is to design a strategy to extract reliable land-cover information from the historical land-cover map for one of the images (source domain). Then, a novel relational knowledge transfer technique is applied to transfer label information. Finally, classifiers are trained on the transferred samples with spatio-spectral features. The experimental results show that aforementioned steps can select sufficient effective samples for target images, and for the main land-cover classes in a rapidly urbanizing region; the results of an updated map show good performance in both precision and vision. Therefore, the proposed approach provides an automatic solution for urban land-cover mapping with a high degree of accuracy.

Comparing Sentinel-2 MSI and Landsat 8 OLI Imagery for Monitoring Selective Logging in the Brazilian Amazon

Imagery from medium resolution satellites, such as Landsat, have long been used to map forest disturbances in the tropics. However, the Landsat spatial resolution (30 m) has often been considered too coarse for reliably mapping small-scale selective logging. Imagery from the recently launched Sentinel-2 sensor, with a resampled 10 m spatial resolution, may improve the detection of forest disturbances. This study compared the performance of Landsat 8 and Sentinel-2 data for the detection of selective logging in an area located in the Brazilian Amazon. Logging impacts in seven areas, which had governmental authorization for harvesting timber, were mapped by calculating the difference of a self-referenced normalized burn ratio (ΔrNBR) index over corresponding time periods (2016–2017) for imagery of both satellite sensors. A robust reference dataset was built using both high- and very-high-resolution imagery. It was used to define optimum ΔrNBR thresholds for forest disturbance maps, via a bootstrapping procedure, and for estimating accuracies and areas. A further assessment of our approach was also performed in three unlogged areas. Additionally, field data regarding logging infrastructure were collected in the seven study sites where logging occurred. Both satellites showed the same performance in terms of accuracy, with area-adjusted overall accuracies of 96.7% and 95.7% for Sentinel-2 and Landsat 8, respectively. However, Landsat 8 mapped 36.9% more area of selective logging compared to Sentinel-2 data. Logging infrastructure was better detected from Sentinel-2 (43.2%) than Landsat 8 (35.5%) data, confirming its potential for mapping small-scale logging. We assessed the impacted area by selective logging with a regular 300 m × 300 m grid over the pixel-based results, leading to 1143 ha and 1197 ha of disturbed forest on Sentinel-2 and Landsat 8 data, respectively. No substantial differences in terms of accuracy were found by adding three unlogged areas to the original seven study sites.

Separating Built-Up Areas from Bare Land in Mediterranean Cities Using Sentinel-2A Imagery

In this research work, a multi-index-based support vector machine (SVM) classification approach has been proposed to determine the complex and morphologically heterogeneous land cover/use (LCU) patterns of cities, with a special focus on separating bare lands and built-up regions, using Istanbul, Turkey as the main study region, and Ankara and Konya (in Turkey) as the independent test regions. The multi-index approach was constructed using three-band combinations of spectral indices, where each index represents one of the three major land cover categories, green areas, water bodies, and built-up regions. Additionally, a shortwave infrared-based index, the Normalized Difference Tillage Index (NDTI), was proposed as an alternative to existing built-up indices. All possible index combinations and the original ten-band Sentinel-2A image were classified with the SVM algorithm, to map seven LCU classes, and an accuracy assessment was performed to determine the multi-index combination that provided the highest performance. The SVM classification results revealed that the multi-index combination of the normalized difference tillage index (NDTI), the red-edge-based normalized vegetation index (NDVIre), and the modified normalized difference water index (MNDWI) improved the mapping accuracy of the heterogeneous urban areas and provided an effective separation of bare land from built-up areas. This combination showed an outstanding overall performance with a 93% accuracy and a 0.91 kappa value for all LCU classes. The results of the test regions provided similar findings and the same index combination clearly outperformed the other approaches, with 92% accuracy and a 0.90 kappa value for Ankara, and an 84% accuracy and a 0.79 kappa value for Konya. The multi-index combination of the normalized difference built-up index (NDBI), the NDVIre, and the MNDWI, ranked second in the assessment, with similar accuracies to that of the ten-band image classification.