Radiometric Indices Research Articles

Beyond traditional methods: Innovative integration of LISS IV and Sentinel 2A imagery for unparalleled insight into Himalayan ibex habitat suitability.

The utilization of satellite images in conservation research is becoming more prevalent due to advancements in remote sensing technologies. To achieve accurate classification of wildlife habitats, it is important to consider the different capabilities of spectral and spatial resolution. Our study aimed to develop a method for accurately classifying habitat types of the Himalayan ibex (Capra sibirica) using satellite data. We used LISS IV and Sentinel 2A data to address both spectral and spatial issues. Furthermore, we integrated the LISS IV data with the Sentinel 2A data, considering their individual geometric information. The Random Forest approach outperformed other algorithms in supervised classification techniques. The integrated image had the highest level of accuracy, with an overall accuracy of 86.17% and a Kappa coefficient of 0.84. Furthermore, to delineate the suitable habitat for the Himalayan ibex, we employed ensemble modelling techniques that incorporated Land Cover Land Use data from LISS IV, Sentinel 2A, and Integrated image, separately. Additionally, we incorporated other predictors including topographical features, soil and water radiometric indices. The integrated image demonstrated superior accuracy in predicting the suitable habitat for the species. The identification of suitable habitats was found to be contingent upon the consideration of two key factors: the Soil Adjusted Vegetation Index and elevation. The study findings are important for advancing conservation measures. Using accurate classification methods helps identify important landscape components. This study offers a novel and important approach to conservation planning by accurately categorising Land Cover Land Use and identifying critical habitats for the species.

A New Approach for Prediction of Foliar Dust in a Coal Mining Region and Its Impacts on Vegetation Physiological Processes Using Multi‐Source Satellite Data Sets

AbstractEstimating foliar dust (FD) is essential in understanding the complex interaction between FD, vegetation, and the environment. The elevated FD has a significant impacts on vegetation physiological processes. The present study aims to explore the potential of multi‐sensor optical satellite data sets (e.g., Landsat‐8, 9; Sentinel‐2B, and PlanetScope) in conjunction with in situ data sets for FD estimation over the Jharsuguda coal mining region in Eastern India. The efficacy of different spectral bands and various radiometric indices (RIs) was tested using linear regression models for FD estimation. Furthermore, the study attempts to quantify the impacts of FD on vegetation's physiological processes (e.g., carbon uptake, transpiration, water use efficiency, leaf temperature) through proxy data sets. The key findings of the study uncovered sensor‐specific and common trends in vegetation spectral profiles under varying FD concentrations. A saturation threshold was observed around 50 g/m2 of FD concentration, beyond which additional FD concentration exhibited limited impact on spectral reflectance. On the other hand, the assessment of FD estimation models revealed distinct performances and shared trends across various satellite sensors. Notably, near‐infrared and shortwave infrared‐1 bands, along with certain RIs, such as the Global Environmental Monitoring Index and the Non‐Linear Index, emerged as pivotal for accurate FD estimation. Besides, the study results revealed that vegetation‐associated carbon uptake experienced a ∼2 to 3 gC reduction for every additional gram of FD per square meter. Moreover, the vegetation transpiration reduction per unit of FD ranged from approximately 0.0005 to 0.0006 mm/m2/day, highlighting a moderate impact on transpiration levels. These findings aid a significant evidence base to our understanding of FD's impact on vegetation physiological processes.

Development of a Pre-Automatized Processing Chain for Agricultural Monitoring Using a Multi-Sensor and Multi-Temporal Approach

Understanding crop types and their annual cycles is key to managing natural resources, especially when the pressures on these resources are attributable to climate change and social, environmental, and economic policies. In recent years, the space sector’s development, with programs such as Copernicus, has enabled a greater availability of satellite data. This study uses a multi-sensor approach to retrieve crop information by developing a Proof of Concept for the integration of high-resolution SAR imagery and optical data. The main goal is to develop a pre-automatized processing chain that explores the temporal dimension of different crop. Results are related to the advantage of using a multi-sensor approach to retrieve vegetation biomass and vertical structure for the identification of phenological stages and different crops. The novelty consists of investigating the multi-temporal pattern of radiometric indices and radar backscatter to detect the different phenological stages of each crop, identifying the Day of the Year (DoY) in which the classes showed greater separability. The current study could be considered a benchmark for the exploitation of future multi-sensor missions in downstream services for the agricultural sector, strengthening the evolution of Copernicus services.

Comparison of Recent Remote Sensing Data Using an Artificial Neural Network to Predict Soil Moisture by Focusing on Radiometric Indices

Remote sensing data is widely used as a common variable for digital soil mapping estimating models. The aim of this study, quite recently made available to researchers Operational Land Imager 2 (OLI–2) have structure Landsat 9 and Landsat 8 (OLI) and Sentinel 2A (MSI) to compare the performance of soil moisture estimation in multi-layer perceptron network (MLP) artificial intelligence algorithm of image data. The working area is 886.78 km2 and soil sampling was performed at 66 points for gravimetric soil moisture determination. In addition, after the satellite images were pre-processed, Soil Adjusted Vegetation Index (SAVI) and Normalized Difference Moisture Index (NDMI) were calculated. Landsat 9 (OLI-2) based SAVI and NDMI showed a moderately significant positive correlation relationship with gravimetric soil moisture (rSAVI-SM=0.62, rNMDI-SM=0.44). The relationship between Landsat 8 (OLI) (rSAVI-SM=0.57, rNDMI-SM=0.11) and Sentinel 2A (MSI) (rSAVI-SM=0.42, rNDMI-SM=0.27) based radiometric indices and soil moisture was lower than Landsat 9 (OLI-2). RMSE values of MLP models were found to be respectively 0.79, 1.16 and 1.17 for Landsat 9 (OLI-2), Landsat 8 (OLI) and Sentinel 2A (MSI). Our results showed that with an Operational Land Imager (OLI-2) and near and short-wave infrared wavelengths improvements to multispectral imaging have improved soil moisture estimation success.

Analysis of the Taquari Megafan through radiometric indices

The complex landscapes of tropical monsoonal lowlands make water resource estimates difficult, given sparsely distributed monitoring networks. Satellite radiometry has facilitated increased access to environmental data. Here, we use radiometric indices to analyze the Taquari megafan and identify features of the Pantanal subregions that characterize it. Recent periods of strong droughts and floods were selected in the region for statistical analysis using the following radiometric indices: NDWI-Gao, NDWI-McFeeters and MNDWI. The behavior of the NDWI-Gao showed that the Paiaguás and Nhecolândia subregions had similar vegetation moisture values in 2011 and 2012, but surprisingly, the two subregions did not attain maximum values of vegetation moisture higher than values in the Taquari subregion during severe flooding in 2011. The Paiaguás and Nhecolândia subregions were more correlated in terms of vegetation moisture than open water features. The MNDWI analysis of the Taquari subregion, however, revealed the highest levels of liquid water in the vegetation and the highest values for water bodies/courses. The permanently flooded state of the Taquari subregion may explain why the vegetation appears to be moister in the NDWI-Gao. The NDWI-Gao data were consistent for dry vegetation in 2020, recording values of −0.7, while other dry periods such as 2010 and 2013 recorded values of −0.4. The vegetation responded quickly to the alternation of drought/flood cycles, registering high values of liquid water content in the period 2011–2012 and reduced values in 2012–2013.

LAYING THE FOUNDATION FOR AN ARTIFICIAL NEURAL NETWORK FOR PHOTOGRAMMETRIC RIVERINE BATHYMETRY

Abstract. This work aims to test the effectiveness of artificial intelligence for correcting water refraction in shallow inland water using very high-resolution images collected by Unmanned Aerial Systems (UAS) and processed through a total FOSS workflow. The tests focus on using synthetic information extracted from the visible component of the electromagnetic spectrum. An artificial neural network is created using data of three morphologically similar alpine rivers. The RGB information, the SfM depth and seven radiometric indices are calculated and stacked in an 11-bands raster (input dataset). The depths are calculated as the difference between the Up component of the bathymetry cross-sections and the water surface quotas and constitute the dependent variable of the regression. The dataset is then scaled. The observations of one of the analyzed case studies are used as the unseen dataset to test the generalization capability of the model. The remaining observations are divided into test (20%) and training (80%) datasets. The generated NN is a 3-layer MLP model with one hidden layer and the Rectified Linear Unit (ReLU) and sigmoid activation functions. The weights are initialized to small Gaussian random values, and kernel regularizers, L1 and L2, are added to reduce the overfitting. Weights are updated with the Adam search technique, and the mean squared error is the loss function. The importance and significance of 11 variables are assessed. The model has a 0.70 r-squared score on the test dataset and 0.77 on the training dataset. The MAE is 0.06 and the RMSE 0.08, similar results obtained from the unseen dataset. Although the good metrics, the model shows some difficulties generalizing swallow depths.

An Ultra-Resolution Features Extraction Suite for Community-Level Vegetation Differentiation and Mapping at a Sub-Meter Resolution

This paper presents two categories of features extraction and mapping suite, a very high-resolution suite and an ultra-resolution suite at 2 m and 0.5 m resolutions, respectively, for the differentiation and mapping of land cover and community-level vegetation types. The features extraction flow of the ultra-resolution suite involves pan-sharpening of the multispectral image, color-transformation of the pan-sharpened image, and the generation of panchromatic textural features. The performance of the ultra-resolution features extraction suite was compared with the very high-resolution features extraction suite that involves the calculation of radiometric indices and color-transformation of the multi-spectral image. This research was implemented in three mountainous ecosystems located in a cool temperate region. Three machine learning classifiers, Random Forests, XGBoost, and SoftVoting, were employed with a 10-fold cross-validation method for quantitatively evaluating the performance of the two suites. The ultra-resolution suite provided 5.3% more accuracy than the very high-resolution suite using single-date autumn images. Addition of summer images gained 12.8% accuracy for the ultra-resolution suite and 13.2% accuracy for the very high-resolution suite across all sites, while the ultra-resolution suite showed 4.9% more accuracy than the very high-resolution suite. The features extraction and mapping suites presented in this research are expected to meet the growing need for differentiating land cover and community-level vegetation types at a large scale.

Spatial evolution of land use and land cover for the Tatu-Bola Wildlife Refuge - Pernambuco, using orbital images

This article aims the analyze NDVI (Normalized Difference Vegetation Index), NDWI (Normalized Difference Water Index), and NDBI (Normalized Difference Built-up Index) radiometric indices for TM and OLI sensor scenes aboard the LANDSAT satellite. The study area is the Tatu-Bola Wildlife Refuge, created in March 2015 through State Decree number 41,546, and its surroundings. The study aimed to monitor the spatial and temporal evolution of land use and land cover in a period before and after the implementation of the Armadillo-Bola Wildlife Refuge, using indices of vegetation (NDVI), moisture (NDWI), and exposed soil (NDBI) and its relationship with the surroundings. TM and OLI images (orbit and point 217-066) were used on board the LANDSAT satellite 5 (23-05-1987, 15-05-1996, and 01-02-2005) and 8 (28-01-2015 and 17- 03-2021), radiometrically and geometrically corrected and obtained by accessing EarthExplorer, with the calculation of radiance and reflectance being performed, and the indices applied. In order to better identify the area with vegetation and exposed soil, the RGB composition was performed, with NDBI, NDVI, and NDWI indices for Landsat images and fusion of multispectral and panchromatic bands in WPM sensor images from the CBERS-4A satellite from 23- 08-2020 (orbit and point 198-124) and 18-09-2020 (199-124). Digital numbers of cloud and shadow pixels were excluded through supervised classification. Between 1987 and 2021, it was possible to observe a decrease in vegetation cover, especially around the RVS Armadillo-balla, and an increase in area, with NDVI values lower than 0.40. The Refuge area in Petrolina has the largest amount of deforested area. The use of RGB composition of the indices and the merged images from the CBERS-4A satellite, with a spatial resolution of 2 m, present a significant improvement in identifying areas of exposed soil and dry vegetation.

U Migration during Weathering of Uraniferous Syenite on Tropical Climate: Case of Ngombas in the Southern Cameroon

The lolodorf syenite axis is known for its radiometric indices. In the Ngombas area the migration and retention of uranium in the regolith developed on syenite is studied using a combination of mineralogy and geochemistry in a bid to understand the processes involved in the dissolution and redistribution of uranium in the secondary environment. A trench dug in the area shows three horizons from the bottom to the top. They include the saprolite, B and Ah horizons. Petrographic and XRD investigations of the syenites reveals minerals such as plagioclase, potassic feldspar, amphibole, pyroxene, biotite, quartz, hematite, zircon coupled with uraninite, U-monazite. The regolith developed on the syenite shows relics of plagioclase, amphibole, quartz, hematite, goethite, chlorite, vermiculite, kaolinite. This is associated with U-bearing minerals such as uranothorite, U-monazite, U-zircon, U-florencite and U-rhadophane. The chemical alteration index (CIA), and gain and loss diagrams indicate that the horizons are more weathered from the top to the bottom. The presence of uranium bearing phases such as uranothorite indicates that U-minerals were dissolved, migrated and sorbed on thorite. The occurrences of U-florencite and U-rhabdophane in the weathering blanket indicate that uranyl is stabilized by phosphate minerals. Under oxidizing conditions the stability of hexavalent uranium is favored by the presence of clay minerals and Fe/Mn-oxyhydroxides. Thus, the migration of uranium in Ngombas is sequestrated by clay blended on Fe-oxides through the process of sorption. The U-bearing phases in the regolith that survived weathering include monazite and zircon. The presence of accessories minerals (U-zircon, U-monazite), sorption of uranium by phosphates, by Fe/Mn-oxyhydroxides, and clays minerals play important roles to reduce the U migration in environmental impact of Ngombas region.

A new method for pixel classification for rice variety identification using spectral and time series data from Sentinel-2 satellite imagery

In the agriculture sector food productivity, security, and sustainability, imposed challenges on farmers, regulatory bodies, and policymakers due to increasing demand and depleting natural resources and environmental concerns. Rice crop holds a prominent place in Pakistan’s agriculture sector, it is not only consumed locally but also exported to many countries including China. Gathering crop information such as variety maps, yield estimation, or etc. can help farmers, regulatory bodies, policymakers, and rice mills in decision-making. In Pakistan, crop information is collected through manual field surveys that require a lot of human labor, are costly, and are time-consuming. One cannot ignore human error and bias in the process. A new framework for pixel classification is proposed that uses both spectral and time-series data of Sentinal-2 satellite for mapping two rice varieties “Basmati” and “IRRI, grown in Pakistan. The data were collected from twelve rice fields (approx. 307 acres) of different geographical locations at 16-time instances to cover the complete rice-growing season (May–October) in 2019. A linear spectral unmixing model is used to determine sub-pixel information of water, soil, and vegetation content, which is used for labeling each pixel for supervised learning. The input to our classifier is a 16 × 15 image formed using 15 spectral features (12 spectral bands and 3 radiometric indices) of 16 carefully selected different time instances for each pixel. The output is a pixel-level classification (semantic segmentation) of each pixel into Basmati, IRRI, and others (soil, water, etc.). Experimental results have exhibited an excellent overall accuracy of 98.6% with the proposed approach. The Basmati rice obtained higher accuracy of 99.7% as compared to IRRI rice with an accuracy of 95.2%.

Combined Radiometric Analysis Related to Guava Leaf Phenology in Response to Soil Application of Paclobutrazol (PBZ)

The leaf-response to three-soil applied treatments of Paclobutrazol (PBZ; 1000, 2000 and 0 ppm-control) was studied in a high-density plantation of eight guava (Psidium guajava) genotypes trees. All materials were prunned in vase form, with two to three major branches, yearly prunning for triggering the annual production cycle, and average height of 2.0 m. The dataset comprises fourth radiometric indices highly related to plant physiological activities. The dataset model took into account data collection dates, guava genotypes, and the positional effect of sun radiation on leaves based on their proximity to the canopy level and downward to the base of the woody seasonal-branch. Unexpectedly, there were no significant differences (NS) in PBZ treatments for genotypes, leaf position and radiometric indices. Analysis of the radiometric indices data revealed that anthocyanin (ARI index) and chlorophyll (PRI index) have a strong inverse relationship. Significant differences (P ≤ 0.05) were found between guava genotypes, and anthocyanin content; these results show that guava genotypes have varied responses, which could derive in their classification based-on drought resistance or low water requirements, however, it is important to note that additional research is required to determine the scope of these indications.

Understanding the role of the radiometric indices in temporal evapotranspiration estimation in arid environments

Understanding the role of the radiometric indices in temporal evapotranspiration estimation in arid environments

Species Distribution Modelling performance and its implication for Sentinel-2-based prediction of invasive Prosopis juliflora in lower Awash River basin, Ethiopia

BackgroundSpecies Distribution Modelling (SDM) coupled with freely available multispectral imagery from Sentinel-2 (S2) satellite provides an immense contribution in monitoring invasive species. However, attempts to evaluate the performances of SDMs using S2 spectral bands and S2 Radiometric Indices (S2-RIs) and biophysical variables, in particular, were limited. Hence, this study aimed at evaluating the performance of six commonly used SDMs and one ensemble model for S2-based variables in modelling the current distribution of Prosopis juliflora in the lower Awash River basin, Ethiopia. Thirty-five variables were computed from Sentinel-2B level-2A, and out of the variables, twelve significant variables were selected using Variable Inflation Factor (VIF). A total of 680 presence and absence data were collected to train and validate variables using the tenfold bootstrap replication approach in the R software “sdm” package. The performance of the models was evaluated using sensitivity, specificity, True Skill Statistics (TSS), kappa coefficient, area under the curve (AUC), and correlation.ResultsOur findings demonstrated that except bioclim all machine learning and regression models provided successful prediction. Among the tested models, Random Forest (RF) performed better with 93% TSS and 99% AUC followed by Boosted Regression Trees (BRT), ensemble, Generalized Additive Model (GAM), Support Vector Machine (SVM), and Generalized Linear Model (GLM) in decreasing order. The relative influence of vegetation indices was the highest followed by soil indices, biophysical variables, and water indices in decreasing order. According to RF prediction, 16.14% (1553.5 km2) of the study area was invaded by the alien species.ConclusionsOur results highlighted that S2-RIs and biophysical variables combined with machine learning and regression models have a higher capacity to model invasive species distribution. Besides, the use of machine learning algorithms such as RF algorithm is highly essential for remote sensing-based invasive SDM.

Spatial and transient modelling of land use/land cover (LULC) dynamics in a Sahelian landscape under semi-arid climate in northern Burkina Faso

Sahelian landscapes have gone through rapid changes over the past decades under high pressure from an ever-growing population, climate hazards and land degradation. Understanding drivers of change and forecasting LULC dynamics in time and space are critical for sustainable land management. This research aimed at analyzing LULC patterns and forecasting future changes in the landscape at the watershed level in semi-arid climate in northern Burkina Faso. Three LULC categories (natural vegetation, degradation soils, and cultivated areas) were considered. Seasonal vegetation dynamics were assessed through the monitoring of radiometric indices over the years 2015, 2016 and 2017. The post-harvest period (October-November) was found to be optimal for single-date LULC mapping. Furthermore, LULC maps for 1986, 1999, 2009 and 2017 were produced through a supervised classification of Landsat satellite images. For land change modelling and prediction, a Multilayer Perceptron (MLP) neural network was calibrated between 1999 and 2009 and validated against 2017 reference map, then used to simulate prediction maps for 2030 and 2050, in conjunction with population growth forecasts. Analysis of LULC changes revealed degradation soils were the major contributor to cultivated areas expansion between 1986 and 2017. Land use maps predicted for 2030 and 2050 indicated an expected increase in cultivated areas of 10.4 % and 22.7 % respectively. These results highlight the strength of the human-environment relationship in the watershed and call for careful planning and conservation actions for a sustainable balance between sustainable agricultural production and natural resources conservation.

Automatic Extraction and Filtering of OpenStreetMap Data to Generate Training Datasets for Land Use Land Cover Classification

This paper tests an automated methodology for generating training data from OpenStreetMap (OSM) to classify Sentinel-2 imagery into Land Use/Land Cover (LULC) classes. Different sets of training data were generated and used as inputs for the image classification. Firstly, OSM data was converted into LULC maps using the OSM2LULC_4T software package. The Random Forest classifier was then trained to classify a time-series of Sentinel-2 imagery into 8 LULC classes with samples extracted from: (1) The LULC maps produced by OSM2LULC_4T (TD0); (2) the TD1 dataset, obtained after removing mixed pixels from TD0; (3) the TD2 dataset, obtained by filtering TD1 using radiometric indices. The classification results were generalized using a majority filter and hybrid maps were created by merging the classification results with the OSM2LULC outputs. The accuracy of all generated maps was assessed using the 2018 official “Carta de Ocupação do Solo” (COS). The methodology was applied to two study areas with different characteristics. The results show that in some cases the filtering procedures improve the training data and the classification results. This automated methodology allowed the production of maps with overall accuracy between 55% and 78% greater than that of COS, even though the used nomenclature includes classes that can be easily confused by the classifiers.

Spatial analysis of Hyperion hyperspectral indices to map the vegetation state in the coastal oases of Tunisia

An elevated human presence due to the involvement of the coastal oases of Tunisia in the global petrochemical industry and population pressure in the 1970s has resulted in major changes in the oases’ agro–ecosystem environment. The consequences of this have been urbanisation and rural exodus, priority to the industrial sectors and services at the expense of agriculture, high mobility and rise of trade. The coastal oases of Gabes located in the South-East of Tunisia are considered in this study. This has been affected by sharp degradation, mainly of anthropogenic origins such as demographic growth, extension of the urban areas and creation of a highly contaminating chemical zone amplifying their environmental vulnerability. Satellite data is an essential tool in the study and mapping of these types of environment and for that, we started with the mapping of the vegetative land use using the vegetation indices derived from the hyperspectral scene of the Hyperion sensor (25 April 2010) and field data. This has allowed us to better characterise the most vulnerable areas and to identify the socio–environmental risks. The analysis of the radiometric indices leads to the definition of the spatial extension of vegetation cover in the oases. This study has permitted us to outline the oases’ typologies in Gabes and to discuss their dynamics in the short term.

Radiometric indices-based spectro-spatial approach for hyperspectral image classification

For the Remote Sensing (RS) community, research on image classification remains a challenging issue aiming towards continuously enhancing understanding images content. Quite recently, Deep Learning (DL) based approaches have shown promising performances on various datasets. Nevertheless, for Hyperspectral images (HSI), using DL approaches is hampered due to the limited number of labeled data and the high dimensionality of the feature space, leading to the curse of dimensionality problem. In this paper, we propose a new classification approach of HSI images based on Low-Dimensional Feature Vector (LDFV) representation consisting of specific radiometric indices. This handcrafted image representation significantly reduces the feature space while keeping relevant spectral information. Conventional classifiers such as SVM or KNN can be used thereafter on the bases of the computed LDFV. Next, we consider the spatial information in order to revise the classification results. Our approach is tested on multiple HSI datasets including the ‘Indian Pines’ and the ‘Salinas Scene’ datasets, showing interesting results compared to the state-of-the-art methods while significantly reducing the processing time.

Assessing Worldview-3 multispectral imaging abilities to map the tree diversity in semi-arid parklands

Semi-arid parkland agrosystems are strongly sensitive to climate change and anthropic pressure. In the context of sustainability research, trees are considered critical for various ecosystem services covering environment quality as well as food security and health. But their actual ecological impact on both cropland and natural vegetation is not well understood yet, and collecting spatial and structural information around agroforestry systems is becoming an important issue. Tree mapping in semi-arid parklands could be one of these prerequisites. While for obtaining an exhaustive inventory of individual trees and for analysing their spatial distribution, remote sensing is the ideal tool. However, it has been noted that depending on the spatial resolution and sensor spectral characteristics, tree species cannot be distinguished clearly, even in the sparsely vegetated semi-arid ecosystems of West Africa. Thus, this work focuses on assessing the capabilities of Worldview-3 imagery, acquired in 8 spectral bands, to detect, delineate, and identify certain key tree species in the Faidherbia albida parkland in Bambey, Senegal, based on a ground-truth database corresponding to 5000 trees. The tree crowns are delineated through NDVI thresholding and consecutive filtering to provide object-based radiometric signatures, radiometric indices, and textural information. A factorial discriminant analysis is then performed, which indicates that only four out of the seven most abundant species in the study area can be discriminated: “Faidherbia albida”,” Azadirachta indica”, “Balanites aegyptiaca” and “Tamarindus indica”. Next, random forest and support vector machine classifiers are employed to identify the optimal combination of classifier parameters to discriminate these classes with a high accuracy, robustness, and stability. The linear support vector machine with cost=1 and gamma=0.01 provides the optimal results with a global accuracy of 88 % and kappa of 0.71. This classifier is applied to the whole study area to map all the trees with crowns larger than 2 m, sorted in four identified species and a fifth common group of unidentified species. This map thus enables analysing the variability in tree density and the spatial distribution of different species. Such information can afterwards be correlated to the ecological functioning of the parkland and local practices, and offers promising opportunities to help future sustainability initiatives in different socio-ecological contexts.

Impact of Various Atmospheric Corrections on Sentinel-2 Land Cover Classification Accuracy Using Machine Learning Classifiers

Atmospheric correction is one of the key parts of remote sensing preprocessing because it can influence and change the final classification result. This research examines the impact of five different atmospheric correction processing on land cover classification accuracy using Sentinel-2 satellite imagery. Those are surface reflectance (SREF), standardized surface reflectance (STDSREF), Sentinel-2 atmospheric correction (S2AC), image correction for atmospheric effects (iCOR), dark object subtraction (DOS) and top of the atmosphere (TOA) reflectance without any atmospheric correction. Sentinel-2 images corrected with stated atmospheric corrections were classified using four different machine learning classification techniques namely extreme gradient boosting (XGB), random forests (RF), support vector machine (SVM) and catboost (CB). For classification, five different classes were used: bare land, low vegetation, high vegetation, water and built-up area. SVM classification provided the best overall result for twelve dates, for all atmospheric corrections. It was the best method for both cases: when using Sentinel-2 bands and radiometric indices and when using just spectral bands. The best atmospheric correction for classification with SVM using radiometric indices is S2AC with the median value of 96.54% and the best correction without radiometric indices is STDSREF with the median value of 96.83%.

Classification and mapping of saltmarsh vegetation combining multispectral images with field data

Salt marshes are areas of high conservation value encompassing diverse ecological gradients responsible for creating unique vegetation structure and composition. In complement to the large body of studies developing vegetation mapping methods through the use of remote sensing data, we tested for the possibility of developing a cost-effective method to map salt marsh vegetation as a basis for monitoring a French Nature Reserve. Using classical multivariate ordination and cluster analyses, accurate and ecologically relevant vegetation groups matching existing typologies were determined from a vegetation database collected for management and conservation rather than mapping purposes. This resulted in six distinct vegetation groups, which were mapped through the combination of the NIR spectral band and radiometric indices (NDVI and NDWI) derived from multispectral 2 m-resolution satellite images (Pleiades images). The addition of a LiDAR-derived digital elevation model (DEM) layer was also tested. The final classified map of the reserve based only on optical layers had an overall accuracy of 75.5% (Kappa coefficient of 0.71), with varying success between the different vegetation groups. The addition of the DEM slightly decreased map accuracy to 73.6% (Kappa of 0.68). Decreasing the number of records used for map training had detectable negative effects on map accuracy. This study demonstrated that using already existing field observations combined with only a few spectral bands and radiometric indices from multi-temporal multispectral images with a high spatial resolution can be used as a basis to aid in vegetation classification and mapping of saltmarsh habitats, with the goal of monitoring their dynamics.